Callicarpa Species from Central Vietnam: Essential Oil Compositions and Mosquito Larvicidal Activities

There are around 140 species in the genus Callicarpa, with 23 species occurring in Vietnam. The Vietnamese Callicarpa species have been poorly studied. In this work, the leaf essential oils of C. bodinieri, C. candicans, C. formosana, C. longifolia, C. nudiflora, C. petelotii, C. rubella, and C. sinuata, have been obtained from plants growing in central Vietnam. The chemical compositions of the essential oils were determined using gas chromatography – mass spectrometry. Mosquito larvicidal activities of the essential oils were carried out against Aedes aegypti. All of the Callicarpa leaf essential oils showed larvicidal activity, but two samples of C. candicans were particularly active with 48-h LC50 values of 2.1 and 3.8 μg/mL. Callicarpa candicans essential oil should be considered as a potential alternative mosquito control agent.


Introduction
There are around 140 species of Callicarpa L. distributed in tropical and subtropical locations [1]. The genus has been placed in either the Verbenaceae or the Lamiaceae, but is currently placed in Lamiaceae [2,3]. Members of the genus have been used as fish poisons and in herbal medicine [1,2]. In this work, we present the essential oil compositions of several Callicarpa species growing wild in central Vietnam. In addition, some of the essential oils were screened for mosquito larvicidal activity.
Callicarpa bodinieri H. Lév. is native to western and central China [3], Vietnam, Laos, Cambodia, and Thailand [4]. The plant is used in traditional Chinese medicine to treat hematemesis (oral decoction of the leaves) and to treat wounds and bruises (fresh leaves externally) [5]. Flavonoids, sterols, triterpenoids [2], and diterpenoids [6] have been characterized in the leaves of C. bodinieri.
Callicarpa candicans (Burm. f.) Hochr. is native to southeast Asia, including China (Quangdong, Hainan), Burma, Cambodia, India, Laos, the Philippines, Thailand, and Vietnam [3]. The plant has been used as a fish poison in the Philippines [7], India [8], and Thailand [9]. In Vietnamese traditional reported for the first time. As far as we are aware, none of the Callicarpa essential oils presented in this work has been previously investigated in terms of mosquito larvicidal activity.

Callicarpa longifolia
Leaf essential oils of C. longifolia were obtained from Son Tra Peninsula (Da Nang City) and from Nghia Dan district (Nghe An province). Sesquiterpene hydrocarbons and oxygenated sesquiterpenoids dominated both essential oils (Table 5). There were, however, notable differences in the chemical profiles. For example, β-selinene was relatively abundant in the Nghia Dan sample (13.2%), but much less in the sample from Da Nang (3.2%). Conversely, trans-β-guaiene was abundant in the Da Nang sample (22.2%), but much lower in the Nghia Dan sample (0.4%). To our knowledge, there are no previous reports on the essential oil of C. longifolia.

Callicarpa nudiflora
Unlike the essential oils of other Callicarpa species in this investigation, the leaf essential oil of C. nudiflora was dominated by the monoterpenes α-pinene (8.1%) and β-pinene (34.2%). Caryophyllene oxide (20.1%) was also an abundant component ( Table 6). The chemical composition of Vietnamese C. nudiflora is markedly different from the leaf essential oil from China [46]. The Chinese sample showed only small quantities of αand β-pinene (0.1% and 1.6%, respectively) and caryophyllene oxide was not observed. Conversely, humulene epoxide II was abundant in the sample from China (17.3%), but relatively minor in the sample from Vietnam (0.5%). Bisabolene oxide was abundant in the Chinese essential oil (10.5%), but was not detected in the sample from Vietnam.

Mosquito Larvicidal Activity
The 24-h and 48-h mosquito larvicidal activities of the Callicarpa leaf essential oil are summarized in Tables 11 and 12. As far as we are aware, there have been no previous larvicidal investigations on these Callicarpa essential oils. Due to limited supply of some of the essential oils and limited supplies of mosquito larvae, not all essential oils could be screened against both mosquito species. Dias and Moraes have concluded that plant essential oils are considered larvicidal against Ae. aegypti if the LC 50 values are less than 100 µg/mL [48]. Based on these guidelines, all of the Callicarpa essential oils showed good larvicidal activity. However, the leaf essential oils of C. candicans, from Nghia Dan district, Nghe An province and from Dai Loc district, Quang Nam province were particularly active with 48-h LC 50 values of 3.8 and 2.1 µg/mL, respectively, against Ae. aegypti. The leaf essential oils of C. candicans were also effective larvicidal agents against Cx. quinquefasciatus. Table 11. Twenty-four-hour mosquito larvicidal activity (µg/mL) of Callicarpa leaf essential oils from central Vietnam. The leaf essential oils of C. candicans were rich in (E)-caryophyllene, caryophyllene oxide, β-selinene and atractylone. Both (E)-caryophyllene and caryophyllene oxide have shown only weak larvicidal activity against Ae. aegypti [48]. However, atractylone may be contributing to the larvicidal activity; the compound has shown insecticidal [49] as well as acaricidal activity [50]. β-Selinene has also shown insecticidal activity [49]. In addition to the insecticidal properties of atractylone and β-selinene, there may be synergistic effects between these components and (E)-caryophyllene, caryophyllene oxide, or other minor components. Scalerandi and co-workers have shown that Musca domestica preferentially oxidize major essential oil components in a mixture while the components in lesser concentrations act as toxicants [51]. In addition, there were several unidentified components, particularly in the Dai Loc sample, that may be contributing to the larvicidal effects.
The larvicidal activity of C. bodinieri leaf essential oil was the weakest of the Callicarpa species tested with a 24-h LC 50 of 54 µg/mL. Limonene was one of the major components (8.0%), and this compound had shown larvicidal activity against Ae. aegypti of around 30 µg/mL [59]. Caryophyllene oxide, another major component (9.8%) is inactive against Ae. aegypti [39]. Although apparently not tested against mosquito larvae, β-selinene (8.9% in C. bodinieri leaf essential oil) is known to be insecticidal against Drosophila melanogaster adults [49].
The leaf essential oils of C. formosana, C. rubella (Nam Giai), C. rubella (Tay Giang), and C. sinuata showed comparable larvicidal activities with 24-h LC 50 ranging from 24.2 to 31.9 µg/mL. However, the chemical compositions of the essential oils were very different.
In order to evaluate potential correlation between constituents and larvicidal activities, multivariate analyses (hierarchical cluster analysis, HCA, and principal component analysis, PCA, were undertaken. The hierarchical cluster analysis (Figure 1) showed four groupings. Group 1 is made up of the two C. candicans samples and represents a very larvicidal group (average 24-h LC 50 and LC 90 = 4.02 and 9.34 µg/mL). The major components in this group are atractylone (average 20.9%) and caryophyllene oxide (average 8.1%). Group 2 is a single sample, C. petelotii is somewhat active with 24-h LC 50 and LC 90 of 19.1 and 37.9 µg/mL and α-humulene, α-selinene, and humulene epoxide II as the major components. Group 3 (C. sinuata, C. formosana, and both C. rubella samples) had average 24-h larvicidal LC 50   The principal component analysis ( Figure 2) does not reveal any clear associations between chemical components and larvicidal activity. (E)-Caryophyllene, caryophyllene oxide, and α-humulene were found in all of the samples and therefore correlate with the essential oil samples and not necessarily with the larvicidal activities. Apparently the synergistic and antagonistic interactions of the components in these essential oils are too subtle to be parsed out with so few data.

Plant Material
Plant material (leaves and/or stem bark) from Callicarpa species was collected from several locations in central Vietnam (Table 13). The plant material from several individuals from each site were combined in order to provide enough plant material for each species. The plants were identified by Dr. Do Ngoc Dai, and voucher specimens (Table 13) have been deposited in the School of Natural Science Education, Vinh University. The fresh plant materials (2.0 kg each) were shredded and hydrodistilled for 4 h using a Clevenger type apparatus (Witeg Labortechnik, Wertheim, Germany). The yields of essential oils are summarized in Table 13.

Gas Chromatography-Mass Spectrometry
Each of the Callicarpa essential oils was analyzed by GC-MS using a Shimadzu GCMS-QP2010 Ultra (Shimadzu Scientific Instruments, Columbia, MD, USA) operated in the electron impact (EI) mode (electron energy = 70 eV), scan range = 40-400 atomic mass units, scan rate = 3.0 scans/s, and GC-MS solution software. The GC column was a ZB-5 fused silica capillary column (Phenomenex, Torrance, CA, USA) (30 m length × 0.25 mm internal diameter) with a (5% phenyl)-polymethylsiloxane stationary phase and a film thickness of 0.25 µm. The carrier gas was helium with a column head pressure of 552 kPa and flow rate of 1.37 mL/min. Injector temperature was 250 • C and the ion source temperature was 200 • C. The GC oven temperature program was programmed for 50 • C initial temperature, temperature increased at a rate of 2 • C/min to 260 • C. A 5% w/v solution of the sample in CH 2 Cl 2 was prepared and 0.1 µL was injected with a splitting mode (30:1). Identification of the oil components was based on their retention indices determined by reference to a homologous series of n-alkanes (C 8 -C 40 ), and by comparison of their mass spectral fragmentation patterns with those reported in the databases [60][61][62][63]. The percentages of each component in the essential oils are reported as raw percentages based on total ion current without standardization.

Mosquito Larvicidal Assay
Eggs of Ae. aegypti were purchased from Institute of Biotechnology, Vietnam Academy of Science and Technology, and maintained at the Laboratory of Department of Pharmacy of Duy Tan University, Da Nang, Vietnam. For the assay, aliquots of the essential oils of Callicarpa species, dissolved in DMSO (1% stock solution), was placed in a 500-mL beaker and added to water that contained 20 larvae (third and early fourth instar). With each experiment, a set of controls using DMSO was also run for comparison. Mortality was recorded after 24 h and again after 48 h of exposure during which no nutritional supplement was added. The experiments were carried out 25 ± 2 • C. Each test was conducted with four replicates with several concentrations (100, 50, 25, 12.5, 6.0, 3.0, 1.5, and 0.75 µg/mL). Larvicidal activity against Culex quinquefasciatus (The larvae were fed on Koi fish food: Adults were provided with a 10% sucrose solution and a 1-week-old chick for blood feeding.) were determined similarly with concentrations of 150, 100, 50, 25, 6.0, 3.0, 1.5, and 0.75 µg/mL. Permethrin was used as a positive control. The acute larvicidal effects on Ae. aegypti, and Cx. quinquefasciatus were recorded 24 h and 48 h after treatment. The data obtained were subjected to log-probit analysis [64] to obtain LC 50 values, LC 90 values, and 95% confidence limits using XLSTAT v. 2018.5 (Addinsoft, Paris, France).

Statistical Analysis
Mosquito larvicidal activities (LC 50 and LC 90 ) against Ae. aegypti and Cx. quinquefasciatus were determined by log-probit analysis using XLSTAT v. 2018.5 (Addinsoft, Paris, France). The more abundant chemical components of the Callicarpa essential oils were used in the multivariate analyses. The essential oil compositions were treated as operational taxonomic units (OTUs), and the concentrations (percentages) of 26 major essential oil components and the 24-h LC 50 and LC 90 larvicidal activity data were used to determine the associations between the Callicarpa essential oils using agglomerative hierarchical cluster (AHC) analysis using XLSTAT Premium, version 2018.5 (Addinsoft, Paris, France). Dissimilarity was determined using Euclidean distance, and clustering was defined using Ward's method. For the principal component analysis (PCA), the 26 major components and the larvicidal data were taken as variables using a Pearson correlation matrix using XLSTAT Premium, version 2018.5 (Addinsoft, Paris, France). A total of 280 data (28 variables × 10 samples) were used for the PCA.

Conclusions
There are profound chemical variations in the leaf essential oils of Callicarpa species, both between species and within species. All of the Callicarpa leaf essential oils showed larvicidal activity against Ae. aegypti. However, C. candicans showed excellent mosquito larvicidal activity against Ae. aegypti as well as Cx. quinquefasciatus, which can be attributed to atractylone and/or to unidentified components. This essential oil, therefore, may represent a low-cost and environmentally friendly mosquito control agent. Nevertheless, although the larvicidal activities of Callicarpa leaf essential oils are promising, additional screening on non-target organisms is needed [41,42].