The repellency and toxicity effects of essential oils from the Libyan plants Salvadora persica and Rosmarinus officinalis against nymphs of Ixodes ricinus

Essential oils extracted from the leaves of Libyan Rosemary (Rosmarinus officinalis L.), and Miswak (Salvadora persica L.) were evaluated for their acaricidal and repellent effects on Ixodes ricinus L. nymphs (Acari: Ixodidae) using a bioassay based on an ‘open filter paper method’. Rosmarinus officinalis leaf essential oil diluted to 0.5 and 1 µl/cm2 in acetone exhibited, respectively, 20 and 100% tick mortality after about 5 h of exposure. A total of 50 and 95% of I. ricinus nymphs were killed by direct contact with the oil when exposed to lethal concentrations (LC) of 0.7 µl/cm2 (LC50) and 0.95 µl/cm2 (LC95), respectively. The LC50 (0.5 µl/cm2) was reached before the end of the first 24 h of exposure time (ET), as tick mortality at 24 h was 60%. Salvadora persica leaf essential oil at 1 µl/cm2 showed a significant repellency effect against I. ricinus nymphs at 1.5 h ET. A 95% repellency was observed at a repellent concentration (RC95) of 1 µl/cm2 of S. persica, but no significant mortality was recorded at this dose of S. persica oil. Gas chromatography–mass spectrometry analyses showed that the main monoterpenes in both oils were 1,8-cineol, α-pinene, and β-pinene, although in markedly different proportions. These results suggest that essential oils have substantial potential as alternative approaches for I. ricinus tick control.


Introduction
Many plant essential oils show a broad spectrum of effects against arthropods pests, including anti-feedant, repellent, and growth regulatory and anti-vector activity. Previous investigations indicate that some chemical constituents of these oils interfere with the octopaminergic nervous system of the insects (Bischof and Enan 2004;Kostyukovsky et al. 2002), which has the advantage of being a target site not shared with mammals. In fact, most essential oils have been shown to be relatively non-toxic to mammals and fish, and thus meet the criteria for "reduced risk" pesticides (Stroh et al. 1998). Part of the explanation for this success is the unique repertoire of diverse chemical structures that can be found in nature. Examples include azadirachtin, isolated from the neem tree, and other bioactive compounds that have been used successfully for pest control (Isman et al. 1990;Park et al. 2003;Wang et al. 2010). In Libya, many plants have medicinal or economic importance and have had a deep impact on the life and culture of the Libyan people. For example, Salvadora persica, commonly known as Miswak, is used for its antibacterial effect on several oral bacteria (Alali and Al-Lafi 2003;Sofrata 2010).
Rosmarinus officinalis is an aromatic herbal remedy that can relieve cramps and pain, and improve the circulation to the head. Thus its common name in Arabic translates to "herb of remembrance" (Begum et al. 2013). It also stimulates the liver and gallbladder, and improves digestion by controlling pathogenic organisms. The plant parts used include leaves and flowers, which contain essential oils in several species of Rosemary (Rosmarinus spp.). This special combined with the wide availability of essential oils from the flavor and fragrance industries, has made it possible to fast track commercialization of essential oil-based pesticides or acaricides.
Ticks (Acari: Ixodoidea) and tick-borne diseases are distributed worldwide (Ghosh et al. 2007). In recent years these have received increasingly serious attention because of their expanding territories (Bueno-Marí 2013;Bueno-Marí et al. 2015) and impact on public health (Estrada-Peña and Jongejan 1999;Jongejan and Uilenberg 2004). Tick-borne diseases such as the bacterial Lyme disease and the viral tick-borne encephalitis (TBE) have received greater attention as the number of cases increases every year (Centers for disease control and prevention (cdc) 2017). Thus, the discovery of effective and environmentally safe tick-repellents and pesticides (acaricides) has become more urgent. At the same time, public concerns about the environmental impact and safety of chemical applications are driving research into alternative, sustainable methods for pest control, such as plant-based acaricides.
One of the most common tick species is Ixodes ricinus L. (Acari: Ixodidae), which occurs over a wide geographical range, from Ireland in the west to the Ural Mountains (Russia) in the east, and from Scandinavia in the north to Morocco and Egypt in the south (Estrada-Peña et al. 2018). Ixodes ricinus (L.) can transmit Lyme borreliosis (LB), caused by the spirochete Borreliella burgdorferi and affecting 100,000-200,000 people every year in Europe (Pålsson et al. 2008). Pathogens known to be carried by this species are tickborne encephalitis virus (TBEV) (Pritt et al. 2016), ehrlichiosis, tularaemia, rickettsiosis, and babesiosis (Aberer 2009). Ixodes ricinus has increased its known geographical distribution and density-activity in many areas of Europe in the last decades, possibly due at least partly to the changing climate (Lindgren et al. 2000;Jaenson et al. 2012;Medlock et al. 2013). This is reflected in an increasing number of reported cases of tick-borne diseases (Michelet et al. 2014). Examples from Scandinavia include a correlation between changes in tick distribution or abundance (Mannelli et al. 2012) and the incidence of LB 1 3 and TBE (Lindgren and Gustafson 2001). Further, studies from the Czech Republic show the occurrence of I. ricinus and TBE cases has shifted in recent decades to include higher altitudes, probably due to a prolonged vegetation period, in particular milder autumns (Mannelli et al. 2012).
The nymphal stage of I. ricinus is the most medically relevant. Nymphs can transmit the same diseases as adults (Heyman et al. 2010), and they are more abundant. The much smaller size of nymphs compared to adult ticks also means that the nymphs are much harder to detect and therefore less likely to be removed before transmission occurs. Thus nymphs probably pose more of a risk than adult ticks (Randolph 1998). Nymphs of I. ricinus feed for 5-10 days during spring-summer before dropping to the ground and moulting to adults. The multi-host behaviour of I. ricinus also gives it considerable opportunities to spread pathogens among numerous animal species including humans (Bischof and Enan 2004). Since the human risk of acquiring tick borne diseases transmitted from biting by l. ricinus is broadly linked to tick nymph density (Lindgren et al. 2000), we focused on the nymphal stage in this study. This is considered the most important stage of ticks as a vector of disease in humans.

Tick collection and preservation
Unfed nymphs of I. ricinus were collected in a woodland area 6-8 km south of Uppsala (59.85862°N, 17.64373°E) in east-central Sweden, during April-September 2016 by dragging a 1 m 2 light-coloured flannel cloth over the ground vegetation (Mejlon and Jaenson 1993). The cloth was inspected at each 10 m step, at which time all nymphs adhering to the cloth were collected. Nymphs were maintained at 85-95% relative humidity (RH) and ≈ 4 °C in complete darkness for 2 months. Before testing, the nymphs were allowed to adapt to the test environment (21-23 °C, 85-95% RH) for 24 h.

Plant material
Plant materials were collected as wild grown material in the Green Mountain (Al-Jabaal Al-Akhder), a heavily forested fertile upland area in northeaster Libya. The collection consisted of two plants, S. persica (Salvadoraceae) and R. officinalis (Labiatae/Lamiaceae). Plant leaves were stored in polyethylene bags in a freezer at − 20 °C, before steam distillation within 30 days. The yield from 100 g of fresh leaves of R. officinalis and S. persica was 2.05 and 0.07 g of essential oil, respectively (Table 1).

Essential oils
Oils were prepared by steam distillation from leaves and stems, which were placed in a glass vessel. Steam produced in a separate container at elevated pressure was then introduced into the vessel. After about 4 h exposure time, the steam was collected and cooled in a Liebig cooler, and the oil/water mixture was collected in glass vials. The oil was then extracted with n-pentane 99)%) and dehydrated with MgSO 4 , after which the mixture was filtered and weighed (Table 1).

GC-MS analysis
The essential oils were analysed by a gas chromatograph coupled with a mass spectrometer (GC-MS). The oil was diluted with hexane and around 300 ng/1 µl was injected syringe into a Hewlett Packard GC 6890 N (Agilent Technologies, USA) equipped with a DB-5 column (30 m length, 0.25 mm internal diameter and 0.25 µm stationary phase film thicknesses, Agilent Technologies) for separation of the volatiles. The GC injector temperature was set up at 250 °C and held isothermally throughout the analysis. The GC oven temperature was held isothermally at 40 °C for 2 min, followed by an increase of 4 °C min −1 up to 200 °C, followed by an increase of 10 °C min −1 up to 280 °C, and then held isothermally at 280 °C for 10 min. High purity helium gas (5.0 Lab Line, Strandmøllen, Denmark) was used as the mobile phase with a constant flow of 1 ml/min. The ion source of the Hewlett Packard 5973 mass spectrometer (Agilent Technologies) was operated at 230 °C, with an electron ionization energy of 70 eV and a solvent delay of 5 min. All compounds were identified by comparing their mass spectra and retention indexes with those presented in the NIST-2008 MS library, published retention index values and available authentic standards using MSD Productivity ChemStation (v.02.01.1177). The quantitative composition of essential oil was reported as a relative percentage of the total peak area.

Laboratory bioassays
Acaricidal and repellency effects were evaluated using an 'open filter paper method' (WHO 1996). For S. persica, 64 µl oil was diluted in 1 ml of acetone, and tested in three replicates at a concentration of 1 µl oil per cm 2 of filter paper. For R. officinalis 64 and 32 µl of oil were diluted in 1 ml acetone to give concentrations of 1 and 0.5 µl/cm 2 . Each oil acetone mix was uniformly spread with a micropipette onto a round Whatman filter paper no.1 (64 cm) and air dried for 20 min. Filter paper impregnated with solvent (acetone) was used as control. Each replicate consisted of 10 nymphs selected randomly and introduced into a plastic cup with a single piece of filter paper sitting on the bottom. The plastic cups were covered by fine-meshed cloth with rubber bands around the rim to prevent the ticks from escaping. Each cup was placed separately in a closed container with wet toilet paper at the bottom to maintain an appropriate humidity. Repellency was determined by recording nymphal behaviour. Non-repelled nymphs were those that moved freely in the cup throughout the trial period, while nymphs that moved to the top edges of the cup were considered 'repelled'. Nymphs were checked every 30 min for 6 h. This is the minimum amount of time observed for transmission of Borrellia while most studies indicate that Borrellia are required at least 24 h for transmission (Cook 2015;Hynote et al. 2012;Strle et al. 1996).
Nymph mortality was assessed every 30 min up to 5 h, and then checked every 24 h after (Table 3). Prior to mortality ticks tended to show toxicity effects, particularly changes in their movement and behaviour such as trembling, knockdown and paralysis. Mortality was assessed by visual inspection and light microscopy and by prodding with a pin. All tests were repeated three times and were performed in the laboratory at 20-21 °C and 89-97% relative humidity.
Tick mortality and repellency were calculated as follows: Dose-response curves were obtained by plotting the percentage mortality for each exposure time versus oil concentration, with the slope of the line calculated. LC 50 and LC 95 were then determined from these curves. Time-response curves were obtained by plotting percentage mortality versus exposure time for each concentration. LT 50 and LT 95 were then determined from the resulting curves (Oris and Bailer 1997).

Statistical analyses
Groups were compared using a one-way ANOVA with multiple regression test. Variances were compared using an F test, and t test was used for determining equality of means. Shared variance (R 2 ) was also estimated to account for the effect of the different oils. A value of p < 0.05 was considered significant. All statistical analyses were carried out using Prism 7 for Windows (Graph Pad Software, USA).

Extraction of essential oils
Two plants widely used in traditional medicine in Libya were studied, R. officinalis and S. persica. Extracts from both planted were assayed for their possible repellency and acaricidal effects against nymphs of the widespread common tick, I. ricinus. A total of 100 g of fresh leaves were collected in Libya for each plant. After extraction yielded essential oils in the amounts of 2.05 g from R. officinalis and 0.07 g from S. persica (Table 1). 1 3 most abundant of these were (2E)-Hexenal (32.7%), 1.8-cineol (20.8%), β-pinene (16.8%), and α-pinene (5.9%). Together these account for 75.5% of the total ion current (TIC). For R. officinalis, 28 compounds were identified, corresponding to 98.23% of the TIC. The majority of the detected compounds were oxygenated monoterpenes and monoterpene hydrocarbons in different proportions. The most abundant of these were 1,8-cineole (26.76%), α-pinene (13.03%), verbenone (11.51%) and borneol (7.08%), constituting 58.3% of the oil. The main volatile compounds of S. persica and R. officinalis essential oils and their relative percentage based on the TIC chromatogram are shown in Table 2.

GC-MS analysis of plant oils
Overall, the volatile compound composition of R. officinalis was more diverse than that of S. persica. Nonetheless, both plants showed a large component of 1,8-cineole, which was the major constituent in R. officinalis (26.76% of TIC) and the second most abundant constituent in S. persica (20.8% of TIC). Both plant oil extracts also showed a substantial amount of α-and/or β-pinene (16.8% α-and 5.9% β-pinene in S. persica; 13.03% α-pinene in R. officinalis. However, the major constituent in S. persica oil was (2E)-Hexenal, which was absent in R. officinalis oil ( Table 2).

Bioassays of oil toxicity against ticks
To assay possible toxicity effects of the oils against I. ricinus, nymphs were subjected to various length exposures to the oils and their survival recorded. For R. officinalis oil, mortality was assayed at doses of 1.0 and 0.5 µl/cm 2 of. At 1.0 µl/cm 2 , the ticks remained alive during the first hour but had already stopped all movement and appeared paralyzed but trembling when prodded with a pinpoint tool. Mortality started to appear in the second hour and 50% mortality (LC 50 ) was reached between the third and the fourth hour of exposure (Fig. 1). To get a more precise estimate of optimal exposure time and concentration, mortality was plotted against oil concentration for various exposure times, and LC and LT values were calculated from the curves (Fig. 2). The LC 50 and LC 95 are estimated to be 0.7 and 0.95 µl/cm 2 , respectively, and the LT 50 and LT 95 for 1.0 µl/cm 2 R. officinalis oil are estimated to be 3.75 and 4.5 h, respectively (Table 3). Statistical testing confirmed that R. officinalis at 1 µl/cm 2 has a significant toxic effect within 24 h (F = 1559, DF = 20, t = 4.346, R 2 = 0.4857, p = 0.0003) (Fig. 1).
Although the toxicity effect of the lower dose of 0.5 µl/cm 2 R. officinalis was not within the framework of what can be considered acceptable in terms of lethal time (LT 50 = 19 h, LT 95 < 24 h Table 3), this concentration of oil does show a repellency effect. When exposed to this concentration of the oil, most of the ticks immediately began to climb the cup wall and then stayed at the top, which is the maximum distance from the oil-impregnated filter paper. This repellency effect decreased gradually over the first three hours, at which point only 40% of the ticks remained at the top of the cup (Fig. 3). Statistical tests confirmed that the oil has a significant repellency effect against I. ricinus ticks (t = 6.793, DF = 10, R 2 = 0.8219, p < 0.0001) (Fig. 3). However, the fact that the number of ticks at the top of the cup decreased with increasing exposure time shows that this is not a long-lasting repellency effect as it declines quite sharply after 4 h of exposure (Fig. 3).
No tick mortality was observed for 1 µl/cm 2 of S. persica oil (t = 1, df = 10, R 2 = 0.09091, p = 0.34). Thus, S. persica oil does not have a significant toxic effect against I. ricinus nymphs. Nonetheless, S. persica oil showed a repellency effect against the ticks. At 1 µl/cm 2 oil, the ticks started to climb the cup's wall immediately. This repellency effect remained at nearly 100% for at least five hours (Fig. 4), although it had fully disappeared when next checked at 24 h. Statistical testing confirmed that S. persica oil at 1 µl/cm 2 has a significant repellency effect against I. ricinus tick (t = 9.977, df = 10, R 2 = 0.9087, p < 0.0001; Fig. 4). At 1.5 h, the 95% repellency concentration (RC 95 ) was estimated at 0.95 µl/cm 2 or roughly 1 µl/cm 2 of S. persica oil. Meanwhile the RT 95 for 1 µl/cm 2 of S.  . 1 Mortality of Ixodes ricinus nymphs exposed to Rosmarinus officinalis oil over 24 h. Ixodes ricinus nymphs were exposed to 0 (control), 0.5 and 1.0 µl/cm 2 of R. officinalis oil. Nymph mortality was recorded every 0.5 h over a 5 h period and then again at 24 h, LT 50 and LT 95 were calculated from this graph (Table 3) persica oil was estimated at 5 h (Fig. 4). Thus, S. persica oil appears to have nearly complete repellency against I. ricinus nymphs for around 5 h.

Discussion
We have studied the toxicity and repellency of R. officinalis and S. persica essential oils against nymphs of the common tick, I. ricinus. Our results show that R. officinalis has both significant toxicity (Fig. 1, Fig. 2) and significant repellency (Fig. 3) effects against the ticks (Table 3), while S. persica exhibits significant repellency (Fig. 4).  GC-MS analysis indicates that the oils have major compounds in common, especially the main monoterpene hydrocarbons 1,8-cineol, α-pinene, and β-pinene, although these are present in different proportions in the two oils (Table 2). Both oils also have a small amount of β-caryophyllene oxide which previous work has shown to have acaricidal activity against Dermatophagoides farinae and D. pteronyssinus (house dust mites) (Oh et al. 2014). In Libya, in the past, people built their tents beside S. persica plants to avoid insects and Acari (El-Mogrebi 2003). The fresh leaves are also used in traditional medicine for treating coughs, asthma, scurvy, rheumatism, piles and other diseases. The flowers are small and fragrant, are used as a stimulant, and are mildly purgative. Extracts of the leaves have been shown to have a considerable anti-bacterial effect on several different aerobic oral bacteria (Alali and Al-Lafi 2003).
The essential oil of S. persica possesses highly effective repellency against I. ricinus nymphs. In fact, this repellency was nearly 100% for at least 5 h, although it had reached zero by 24 h. Several of the monoterpenes present in S. persica essential oil (Table 2) are known to be effective tick repellent compounds. These are also more stable compared to the highly volatile α-and β -pinene found in R. officinalis and can thus explain the more long lasting effect of the S. persica oil. This is also consistent with previous work showing a repellent effect of S. persica against I. ricinus at higher doses (Garboui et al. 2009). In fact, the present study shows repellency activity of S. persica with a ten times lower dose than the previous work. Fig. 3 Repellency effect of 0.5 µl/cm 2 Rosmarinus officinalis oil against Ixodes ricinus nymph exposed for 24 h. Repellency (%) effect of 0.5 µl/cm 2 concentration of R. officinalis oil during different exposure times (hours) against I. ricinus nymphs 1 3 The major constituents of S. persica leaf essential oil are (2E)-Hexenal (32.7%) and 1,8-cineole (20.8%) followed by β-pinene (16.8%) and α-pinene (5.9). Repellency effects of 1,8-cineole have been previously reported, but this work found the oil to be too volatile to be used in ten times lower dose (Pålsson et al. 2008). This compound (1,8-cineole) has also been shown to act as a repellent against several species of insects, particularly the nymphs of the German cockroach (Liu et al. 2011), stored product insects (Yoon et al. 2007) and the mosquito, Culex pipiens (Traboulsi et al. 2002). Previous work has also shown that α-pinene isolated from the essential oil of Dianthus caryophyllum exhibits a strong repellent effect against ticks (I. ricinus) (Tunón et al. 2006).
In North African countries, R. officinalis is highly prized for its essential oil and its biological properties. It is the most abundantly exported herb, in the region constituting approximately 12.70% of all dried herb and medicinal plant exports (Hufbauer and Brunel 2009). In Libya, wild rosemary is very common and is used locally especially in the high Green Mountain (Al Jabil Al Akhder). Since ancient times it has been used for medicinal purposes and is known for its antibacterial (Okoh et al. 2010), antifungal (Sertkaya et al. 2010) and insecticidal (Zoubiri and Baaliouamer 2011) properties. Our study shows that R. officinalis oil has significant acaricide activity in the laboratory. This is consistent with previous work showing that Rosemary oil produces > 85% larval mortality in the Asian blue tick, Rhipicephalus (Boophilus) microplus (Martinez-Velazquez et al. 2011). The aromatic vapour of rosemary oil has also been shown to have ovicidal and larvicidal effects against several stored product pests (Papachristos and Stamopoulos 2004) (Tunc et al. 2000) and to Fig. 4 Repellency effect of 1 µl/cm 2 Salvadora persica oil against Ixodes ricinus nymphs exposed for 24 h. Repellency (%) effect of 1 µl/cm 2 concentration of S. persica oil during different exposure times (hours) against I. ricinus nymphs function as fumigant against the two-spotted spider mite (Tetranychus urticae) Koch (Cho et al. 2004). The oil can have a sublethal effect as well, for example acting as a repellent to onion thrips, Thrips tabaci Lind (Koschier and Sedy 2003) and as well as nymphs of I. ricinus (Fig. 3) and (El-Seedi et al. 2012).
Most of the compounds were identified by GC-MS and are monoterpenes e.g. 1,8-cineole, α-pinene, camphor and borneol, all of which are known to have repellent activity and which together are the major constituents of R. officinalis essential oil (Table 2). This result is in agreement with previous work (El-Seedi et al. 2012;Martinez-Velazquez et al. 2011;Schubert et al. 2017). The acaricidal activity of essential oils has previously been attributed to the presence of constituents such as 1.8-cineole (Cavalcanti et al. 2010) (Hüe et al. 2015). Essential oils from Eucalyptus globulus and its major monoterpene 1,8-cineole showed toxicity against human head lice (Yang et al. 2004). However, despite the abundance of this compound in S. persica oil (20.8%, Table 2), this oil did not show toxicity to the ticks in our study in comparison to R. officinalis. This may be due to that the presence of oxygenated monoterpenes (such as Verbenone and Borneol) is much higher in the content of R. officinalis. Another prominent constituent of both oils was α-pinene, which has also been shown to produce strong acute toxicity against Pear psylla (Nyabayo et al. 2015). This compound has also been extracted from Laurus novocanariensis essential oil and demonstrated to have potent acaricidal activity against Psoroptes cuniculi mites with 100% mortality (Macchioni et al. 2006). Additionally, acaricide doses close to the LC 50 with lower dose 0.5 µl/cm 2 , it could induce rather delayed toxification and survival of toxemic ticks for several days (syndrome of a slow death according to Uspenskiy (Uspenskiy 1982).

Conclusion
We found that both of the oils used in this study have effects against I. ricinus ticks. S. persica oil had a stronger and longer-lasting repellency effect than R. officinalis oil; while only R. officinalis oil had a significant toxic effect against I. ricinus ticks. GC-MS shows that both oils have large concentrations of similar terpenes. Thus, the different ratios of these terpenes in the different oils may account for the differences in their effects against ticks.