Introduction

Plant volatiles (essential oils, EOs) are among the first natural products that humankind exploited for its benefit. This choice among the numerous plant-deriving extracts can be rationalized considering their simple and efficient sensory screening by the primitive societies. Civilization evolvement depended on its early stages by the development of those skills of the fundamental human senses: vision, taste, smell, hear, and touch. A significant advantage available to modern empirical scientists rests upon the knowledge accumulation through the centuries of civilization.

A typical example of traditional knowledge exploitation can be considered the case of Juniperus EOs. Based on such piece of information, we conducted a screening research on the bioactivity of EOs isolated from various parts of the plant against mosquitoes (Giatropoulos et al. 2013). Our results have confirmed the ancient references highlighting among the plants of Juniperus species those of Juniperus phoenicea and Juniperus drupacea as the more promising for exploitation in mosquitoes control (Vourlioti-Arapi et al. 2012). Although the relevant results have been so far fruitful enough, the exploitation and incorporation of these EOs in the contemporary agrochemicals industry require further documentation, concerning the determination of the best available combination of their phytochemicals.

Since the aforementioned activity screening has already determined the potential plant species, a second step referring the implementation of a second, more in-depth, screening has to follow it in order to assess the most appropriate harvesting season and plant part, both in respect the EO yield and bioactivity. In this regard, previous studies have showed that the EO content exhibits significant quantitative fluctuation, depending on the plant material harvested season. In particular, thorough study of the seasonal variation of EOs obtained from the leaves of Tunisian J. phoenicea (Medini et al. 2009; Ennajar et al. 2011) revealed different yields and suggested intriguing quantitative differentiation in EOs composition. Additionally, this quantitative fluctuation of EOs ingredients, when compared against their exhibited bioactivities, has been proven adequate to indicate the cause–effect relationship between compounds and/or combinations of them with the expressed bioactivity (Ennajar et al. 2011).

The orientation of the present study towards the commercial exploitation of Juniperus EOs in question also dictated the inclusion of the screening of their repellency properties in addition to their larvicidal bioassays. The repellent properties constitute a major factor for the successful application of commercial biocides, mainly under the following two perspectives:

  1. a.

    The utilization of larvicidal biocides precludes that mosquitoes will approach and lay their eggs in the breeding area undisturbed. Thus, in the case of a strong repellent, regardless of its larvicidal efficacy, it is useless since it does not allow mosquitoes to approach the area of application.

  2. b.

    Since a widely used method of choice of general population for the protection from mosquitoes is the utilization of repellent nets, an essential oil with significant repellent properties, which are useless as biocide, may find a significant industrial application.

Within this conceptual framework, a study herein focuses on the determination of the most active EO content within the seasonal variation of the investigated species, namely the two Juniperus species populations that were selected for sampling. In order to reveal the subsequent cause–effect relations between their EOs and their respective repellent and larvicidal properties against the Asian tiger mosquito (Aedes albopictus), all tests have been performed in a uniform way and the results obtained are thoroughly discussed.

Materials and methods

Plant material

Two indigenous in Greece species of Juniperus—J. phoenicea L. and J. drupacea Labill—were collected in three different vegetative stages, during dormancy (winter), sprouting–flowering (spring), and early fruit development (summer). The sampling included three different samples from each vegetative stage: leaves, immature, and mature fruits (with respect to their availability). Full collection details are included in Table 1. A voucher specimen of each plant is deposited in the herbarium of the Agricultural University of Athens, Athens, Greece.

Table 1 Collection dates, plant part distilled and process applied, and essential oil yield of the herbal material
Full size table

Isolation of the essential oils

The freshly collected plant materials were stored in −18 °C and, before their distillation, were separated into two batches. The first was subjected into steam distillation per se, while the second batch was processed before distillation as follows: Leaves were finely chopped and berries were exhaustibly crushed. All distillations were performed with 3 L of H2O in a Clevenger-type apparatus for 3 h to afford the respective EOs. The resulting oils were dried over anhydrous sodium sulphate and stored at 4 °C. Their yields are included in Table 1.

Gas chromatography–mass spectrometry (GC/MS)

All GC analyses were carried out on an Agilent Technologies 7890A gas chromatograph, fitted with a HP 5MS 30 m × 0.25 mm × 0.25 μm film thickness capillary column and FID. The column temperature was programmed from 60 to 280 °C at an initial rate of 3 °C/min. The injector and detector temperatures were programmed at 230 and 300 °C, respectively. Hydrogen was used as the carrier gas at the flow rate of 1 mL/min.

The GC/MS analyses were performed on the same instrument connected with an Agilent 5957C, VL MS Detector with Triple-Axis Detector system operating in EI mode (equipped with a HP 5MS 30 m × 0.25 mm × 0.25 μm film thickness capillary column) and He as the carrier gas (1 mL/min). The initial column temperature was 60 °C and heated gradually to 280 °C with a 3 °C/min rate. The identification of the compounds was based on comparison of their retention indices (RI) obtained using various n-alkanes (C9–C24). Also, their EI-mass spectra was compared with the NIST/NBS, Wiley libraries spectra, and literature (Adams 1995; Massada 1976). Additionally, the identity of the indicated phytochemicals was confirmed by comparison with available authentic samples.

Mosquito rearing

The laboratory colony of A. albopictus has been maintained in the Benaki Phytopathological Institute, as described previously (Giatropoulos et al. 2012). Adult mosquitoes were kept in wooden framed cage (33 × 33 × 33 cm) with a 32 × 32 mesh, with easy access to 10 % sucrose solution on a cotton wick. Females were chicken blood fed by using Hemotek membrane feeding system (Hemotek). Larvae were fed ad libitum with powdered fish food (JBL Novo Tom 10 % Artemia) in each pan until the adults emerged. The eggs were kept wet for few days and then placed in the pans for hatching.

Larvicidal bioassays

Stock solutions of EOs tested were prepared in ethanol and maintained in a freezer as 10 % w/v stock solution in dimethyl sulfoxide (DMSO).

The larvicidal bioassays were carried out according to the test method for larval susceptibility proposed by the World Health Organization (WHO 1981). Twenty 3rd to 4th instar larvae of A. albopictus were placed in 2 % v/v aqueous solution of DMSO (98 mL of tap water plus 2 mL of DMSO), followed by addition of the tested material solution with gentle shaking to ensure a homogeneous test solution. For each tested material, one screening dose was tested (29 mg L−1, ≈29 μL of 10 % w/v stock solution). Five replicates were used at this discriminating dose, and a control treatment with water and DMSO was also included. Beakers with larvae were placed at 25 ± 2 °C, 80 ± 2 % relative humidity, and photoperiod of 14:10 h (L/D).

Repellency bioassays

For the evaluation of the repellent activity of the EOs tested against adults of A. albopictus, the method of human landing counts was used, as described by Giatropoulos et al. (2013). Concisely, the repellency bioassays were conducted in similar wooden cages containing 100 adult mosquitoes (♀:♂, 1:1), 5 to 10 days old, starving for 12 h at the same temperature and photoperiodic conditions as the laboratory-kept colony. A plastic glove bearing a square opening (5 × 5 cm) and a 24 cm2 Whatman chromatography paper saturated with 50 μL of 100 μg μL−1 EO stock solution in dichloromethane (DCM) (≈0.2 mg/cm2), which was placed around the opening, were inserted for 5 min into the cage, serving as a replicate. At this dose, no landings were counted when DEET was applied. In total, eight replicates were used per treatment, including also control treatments with the solvent DCM as control and N,N-diethyl-meta-toluamide (DEET) as positive control and four human volunteers participated.

Data analysis

The larvicidal effect of EOs was recorded 24 h after treatment. Results are presented as mean percentage of dead larvae (±s.e.), and EOs were classified to low, moderate, or very good larvicides, if the mortality rates ranged between 0–50, 50–80, and 80–100 %, respectively. Data concerning the adult repellency (mosquito landings) and larval toxicity (percentage larval mortality) for the essential oils of each plant were analyzed using Kruskal–Wallis test (Sokal and Rohlf 1995). When significant differences were detected, Mann–Whitney U test (Sokal and Rohlf 1995) was carried out for pair-wise comparison (SPSS 21.0).

Results

Phytochemical analysis

The detailed qualitative and quantitative analytical data of the main constituents (and their respective retention indices) of steam volatiles have been summarized in Tables 2 and 3. In total, 52 phytochemicals were identified, 44 in the EOs of J. phoenicea, and 20 in the EOs of J. drupacea (12 were present in both EOs). Since in a previous study we have extensively studied and discussed the qualitative composition of these EOs (Vourlioti-Arapi et al. 2012), we focused herein on the investigation of their yields and composition, in respect to the applied process, the plant part utilized for the EO retrieval, all in respect with the observed seasonal variations.

Table 2 Juniperus phoenicea essential oils analysis (a = RI comparison, b = MS comparison, c = Authentic samples comparison)
Full size table
Table 3 Juniperus drupacea essential oils analysis (a = RI comparison, b = MS comparison, c = Authentic samples comparison)
Full size table

In respect to the EOs yield, we noticed a significant variation depending on the applied process, the plant part, and the collection period. More specifically, the study of the influence of the pre-distillation process was performed against the average EO yields from the same plant part in between seasons and taxa. Those average yields, presented in Fig. 1, indicate as the most prosperous plant part—in respect with their EOs yield—the crashed ripe berries, which produced EO exceeding by 615.9 % the yield of the corresponding chopped leaves and 378.88 % the crashed unripe berries. The applied process was found to induce always an increase to the EO yield, 7.3 % for leaves, 292.6 % for unripe berries, and 830.7 % for ripe berries.

Fig. 1
figure 1

Fluctuation of the average essential oil yield (in ml/kg) derived from three different Juniperus sp. plant parts and two pre-distillation processes

Full size image

Similarly, the study of the influence of the collection seasons was performed against the average EO yields from the same plant part including all taxa and pre-distillation process. Those average yields, presented in Fig. 2, indicate as the most prosperous plant part in—in respect with their EOs yield—the ripe berries, which comprise the richest EO source, in respect to their yield, followed by the winter-collected unripe berries. Leaves produced EOs with almost uniform content over seasons. Similar findings are also obtained for the spring and summer collection of unripe berries, which subtract the EO yield of the unripe berries winter collection by 155.8 %.

Fig. 2
figure 2

Fluctuation of the average essential oil yield (in ml/kg) derived from three different Juniperus sp. plant parts and during three collection seasons

Full size image

Larvicidal bioassays

In Table 4, the mean percentage (±s.e.) of dead larvae presents and indicates results 24 h after the implementation of the testing material. Only three EOs out of 21 tested proved to have high toxicity. It seems that J14 had 100 % efficacy as larvicidal. J14 is derived from the leaves of J. phoenicea through conventional distillation, during summertime, being plenty of a-pinene and β-phellandrene. By contrast, J01, J04, and J09 deriving from J. phoenicea as well as J31, deriving from J. drupacea, had no larvicidal activity.

Table 4 Adult repellency and larvicidal toxicity of essential oils derived from leaves and berries of J. phoenicea. A discriminated dose of 29 mg/ml was employed for each tested material including also control treatments (for repellant bioassays with the solvent DCM as control and DEET as positive control and for larvicidal bioassays DMSO as control)
Full size table

Repellency bioassays

Table 5 summarizes the results over the mean number of landings on the uncovered area of the glove for 5 min. Twelve out of 24 EOs proved to be “DEET-like.” Among them, J02, J05, J16, J18, and J30 had full repellent activity against mosquitoes (zero landings). These EOs were equally distributed between the two Juniperus species tested. All of these EOs tested with full repellent activity had a-pinene and limonene as a major compound for J. phoenicea and J. drupacea EO, respectively. Based on the obtained results, no significant relationship between toxicity and phytochemical content was detected.

Table 5 Adult repellency and larvicidal toxicity of essential oils derived from leaves and berries of J. drupacea. A discriminated dose of 29 mg/ml was employed for each tested material including also control treatments (for repellant bioassays with the solvent DCM as control and DEET as positive control and for larvicidal bioassays DMSO as control)
Full size table

Discussion

Phytochemical analysis

The seasonal variation was also studied for both species in respect to the main constituents of EOs from leaves and unripe, reporting for the first time the seasonal variation of the EOs of J. drupacea and providing a uniform framework for the combined study of previous relative research endeavors on the EOs of J. phoenicea. The most significant difference observed between the EOs of the two species is related to the substitution of the predominating EOs of J. phoenicea phellandrene isomers with limonene that constitutes the major component of all J. drupacea EOs.

The seasonal variation of EOs derived from the leaves of J. phoenicea has been investigated previously for the respective natural populations of Tunisia (Medini et al. 2009; Ennajar et al. 2011). Although these studies comprised a solid background for the present endeavor, they do not report on the chemical pattern of leaves EOs seasonal variation. In this endeavor, we determined a reduction of α-pinene depending on the increase of temperature and photoperiod, as seasons proceed from winter to summer. This pattern was reversed for both phellandrene isomer concentrations, while the remaining major components displayed a rather stable concentration pattern over seasons. With respect to the seasonal variation of EOs of the unripe berries—presented herein for the first time—similar patterns were obtained for α-pinene and β-phellandrene as well as for the rest of the major compounds.

The pattern of the seasonal decrease of α-pinene of J. phoenicea EOs was also detected for the EOs of leaves and berries of J. drupacea. Similar fluctuation with the EOs of J. phoenicea and consistent within the samples of J. drupacea was detected for myrcene. Additionally, an increasing pattern for the EOs of leaves, similar to those of phellandrene in J. phoenicea EOs, was detected for the molecules of limonene, germacrene D and (+)-3-carene. On the contrary, the fluctuation of the major compounds of unripe berries presents a unique pattern among the compounds and EOs studied. In particular, limonene and germacrene D displayed two peaks in their fluctuation over seasons. Limonene presented its maximum in the spring sample exceeding by 66.8 % the average concentration of winter and summer samples. Germacrene D presented a reverse pattern presenting, also in the spring sample, its minimum, accounting for a 77.5 % decrease in relation with the average concentration of winter and summer samples. All aforementioned results and observations have been depicted in Figs. 3, 4, 5, and 6.

Fig. 3
figure 3

Seasonal fluctuation of leaf essential oil major compounds (>5 %) from J. phoenicea (in % of EO)

Full size image
Fig. 4
figure 4

Seasonal fluctuation of unripe berry essential oil major compounds (>5 %) from J. phoenicea (in % of EO)

Full size image
Fig. 5
figure 5

Seasonal fluctuation of leaf essential oil major compounds (>5 %) from J. drupacea (in % of EO)

Full size image
Fig. 6
figure 6

Seasonal fluctuation of unripe berry essential oil major compounds (>5 %) from J. drupacea (in % of EO)

Full size image

Larvicidal bioassays

Generally, the EOS from plants belonging to Cupressaceae family have already suggested that they have insect toxicity, so the EOs of several Cupressacea species were employing against other arthropod pests, such as ticks and stored product beetles, and the results were promising (Lebouvier et al. 2013; Hashemi and Rostaefar 2014).

Previous studies for the larvicidal effect of EOs derived from Juniperus species have already demonstrated good results against mosquito larvae (Amer and Mehlhorn 2006a; Vourlioti-Arapi et al. 2012; Giatropoulos et al. 2013; Dias and Moraes 2014). Amer and Mehlhorn (2006a) found that two Juniperus species were included in the most toxic among the EOs tested. Further investigation revealed that Juniperus virginiana had better toxicological profile (LD50 < 10 ppm) compared with Juniperus communis against Aedes and Anopheles species. The EO of J. virginiana was also examined after different storage conditions and found to be effective even 2 weeks in the light (Amer and Mehlhorn 2006b).

Our results are, generally, in agreement with previous data, although different chemotypes seems to exist due to seasonal variation of the chemical compounds. The combined study of larvicidal results and the fluctuation of the major compounds of the EOs from J. phoenicea may suggest that the high amount of phellandrenes, α- and β-, that was detected in J14 and J15 may well explain the high toxicity pattern of the two aforementioned EOs (Evergetis et al. 2013). On the other hand, the major constituents of J01 included the same substances with those found in J14; however, no larvicidal activity was detected in the case of J01. According to Giatropoulos et al. (2013), some commonly found substances in EOs with significant larvicidal activity are a-pinene, 3-carene, (R)-(+)-limonene, myrcene, and terpinen-4-ol. Similar studies also stressed that limonenes had a strong larvicidal effect against A. albopictus; by contrast, pinenes had a lower one (Giatropoulos et al. 2012).

Repellency bioassays

The EOs derived fro Juniperus species have been previously tested as potential repellant agents and found to be more effective against Culex and Anopheles mosquito species rather than Aedes (Amer and Mehlhorn 2006c). The same EOs were also tested for their protection period, and the results showed that both had longer protection period against Anopheles and Culex mosquito species. When another EO from Juniperus tested against the Afro-tropical malarial vector (Anopheles arabiensis), results indicated a strong repellent activity (Karunamoorthi et al. 2014). The authors did not provide the chemical analysis, and consequently, they failed to identify the bio-active compounds.

Previous studies showed that most hydrocarbons (e.g., a-pinene) had a lower repellent efficacy against adult mosquitoes compared to aldehydes, oxides, or alcohols (Weldon et al. 2011). However, Giatropoulos et al. (2013) pointed out that the investigation of the repellent activity of EOs derived from some Cupressaceae species has given varying results. The same phenomenon detected when EOs from two Juniperus and one Cupressus species were evaluated against Aedes aegypti and only the EO for J. communis had repellent activity (Carroll et al. 2011). Due to the fact that chemical analysis failed to identify the total number of the compound assumptions, regarding the bio-active compounds, it cannot be supported.

Conclusions

In conclusion, the current work documented the efficacy of the pre-distillation process with respect to the EO yield and indicated the winter as the most prosperous season for the maximization of the Juniperus EO yield. Additionally, the berries proved to be the plant part of choice considering the EO yield, which increased its content through the maturing process. The EO constituent’s study in between the two Juniperus taxa revealed a fundamental discriminating character for all EO in the form of the major compound, which was α-pinene for the J. phoenicea EOs and limonene for the J. drupacea EOs. Among the EOs studied, several presented promising results both as potential larvicidal and repellant agents. In more details, J14, J15, J19, and J25 had strong larvicidal activity and a DEET-like pattern as repellant. Essential oils are a mixture of different components, mostly terpenes. In order to investigate the contribution of bio-active compounds (major and minors) to EOs total activity, it must be isolated directly from crude essential oils and tested against mosquitoes.