Introduction

The pinewood nematode, Bursaphelenchus xylophilus, is the causal agent of pine wilt disease (Kiyohara and Tokushige 1971; Mamiya 1983). The nematode is transmitted by oviposition of vector insects, such as Monochamus alternatus, in dead or dying pine trees (Kobayashi et al. 1984). Control of the disease depends primarily on fumigation of disease–infected trees, aerial application of synthetic pesticides against M. alternatus or injection of synthetic nematicides against B. xylophilus (Lee et al. 2003; Yoshida 2006). However, the discovery and development of new nematicides is getting more expensive (Chitwood 2002). For this reason, several new chemical products are in the process of being registered, and only a few commercial nematicides remain in use. Among these pesticides, organophosphates (OPs) and carbamates (CBs) pesticides are mainly used in the B. xylophilus control programme. Acetylcholinesterase (AChE), the target for the action of OPs and CBs pesticides that terminates nerve impulses by hydrolyzing the neurotransmitter acetylcholine (ACh) to acetic acid and choline at the synapses and neuromuscular junction in most vertebrates, insects and nematodes (Kang et al. 2011; Massoulie et al. 1993; Opperman and Chang 1990; Selkirk et al. 2005). Thus, the inhibition of AChE leads to the dysfunction of the nervous system and death (Kang et al. 2012). However, OPs and CBs pesticides are known to have negative environmental side effects, including high toxicity (Kang et al. 2013). To avoid hazards to the environment and humans caused by the use of traditional synthetic nematicides, it is necessary to find safer alternatives, that are environmentally benign, cheap and effective with a design based on naturally occurring compounds in plants.

Harmine is a beta–carboline alkaloid of Peganum harmala that possesses various types of pharmaceutical properties in vitro and in vivo (Cao et al. 2007; Lin et al. 2002), and extensive spectrum of biological activities such as antitumoral (Chen et al. 2005), antiprotozoal (Mirzaie et al. 2007), antimicrobial (Behidj-Benyounes et al. 2014; Benzekri et al. 2016), insecticidal (Chermenskaya et al. 2010; Nenaah 2011; Rharrabe et al. 2007), and nematicidal (Jakobsen et al. 2013; Ntalli and Caboni 2012) activities. A variety of mechanisms were proposed for harmine, including intercalation into DNA (Sharma et al. 2016), inhibition of cyclooxygenase (Hamsa and Kuttan 2010), monoamine oxidase (Herraiz et al. 2010), AChE (He et al. 2015; Zhang et al. 2013; Zheng et al. 2009), and interaction with L–type Ca2+ channel, opioid receptor, dopamine receptor, γ–aminobutyric acid receptor, 5–hydroxytryptamine receptor, benzodiazepine receptor, imidazoline receptor, and peroxisome proliferator–activated receptor gamma (PPARγ) in the nervous system (Glennon et al. 2000; Khorana et al. 2003; Moloudizargari et al. 2013; Waki et al. 2007).

In our earlier work, we have evaluated the antitumor activity of a series of harmine derivatives by substitution of 1–, 2–, 7– and 9–position of beta–carboline nucleus through C1, N2, N9–alkylation, and C7–oxyalkylation (Guo et al. 2014; Wu et al. 2014). However, 7–position as alkoxy group substituted harmine derivatives showed more potent neurotoxicity than harmine in our experiment (Cao et al. 2013). Here we investigate the nematicidal effect against B. xylophilus and inhibitory activity of these derivatives on AChE in vitro and in vivo.

Materials and methods

Harmine derivatives structure

Harmine derivatives were provided by Rihui Cao (Sun Yat–sen University). The structures of compounds are shown in Table 1

Table 1 Structure of harmine derivatives
Full size table

Chemicals and instruments

Harmine hydrochloride (>98%) was obtained from AMRESCO. The following chemicals were obtained from Sigma–Aldrich (USA): dimethyl sulfoxide, AChE from electric eel (Electrophorus electricus), acetylthiocholine iodide, 5,5–dithiobisbis–nitrobenzoic acid (DTNB), L–glutathione (L–GSH), chloroform (AR), isopropanol (AR), alcohol absolute (AR).

Instruments used in the study were microplate reader (Bio–Tek Instruments, USA), Precise pH instrument (Lab850, SCHOTT insruments, Germany), 5804R Refrigerated centrifuge (Eppendorf, Germany), and ND1000 Nucleic acid protein quantitative instrument (Eppendorf, Germany).

Nematodes

Pinewood nematodes, B. xylophilus, were provided by the Research Center of Nematodes of Plant Quarantine of South China Agricultural University and cultured on fungal mats of Pestalotia sp. grown on potato dextrose agar (PDA) plates at 25 °C for about ten days. Nematodes were extracted from fungal cultures with sterile distilled water in shallow pans. They were collected after 4–8 h and concentrated. The nematodes were rinsed from the filter disks with sterile distilled water and collected (Park et al. 2005).

Nematicidal effect

Solutions of compounds were prepared by serial dilution with distilled water containing 2% DMSO (dimethyl sulfoxide). The nematicidal effect assay was conducted, according to the method described by Kim et al. (2008) with minor modifications. Test solutions were introduced into wells of 24–well plates. In each well, the concentration of nematodes was between 150 and 200 specimens (mixtures of juveniles and adults) per 500 μl of water. Controls received distilled water containing 2% DMSO solution. The active compounds were diluted successively to a series of concentrations for nematicidal assay. Treated and control nematodes were kept at 25 °C. Dead and active nematodes in each well were recorded after incubation for 48 h under a stereomicroscope. Nematodes were defined as dead if their bodies were motionless and straightened. The mortality in the assays was corrected by eliminating natural death in the water containing 2% DMSO solution according to the formula: corrected mortality (%) = (mortality of treatment- mortality of control) / (1- mortality of control) × 100. Each treatment repeated three times.

Inhibitory effect on acetylcholinesterase in vitro

The inhibitory effect on AChE were measured using the spectrophotometric method, according to the method described previously (Choi et al. 2015; Ellman et al. 1961). The median inhibitory concentration (IC50) was calculated using an enzyme inhibition dose response curve, with harmine hydrochloride as a positive control. The assays were conducted in triplicate.

Inhibitory effect on acetylcholinesterase in vivo

The inhibitory effect on AChE in vivo was measured using the spectrophotometric method as described previously (Revathi et al. 2013). B. xylophilus was treated by 20 μg/mL active compounds for 48 h. The treated and control nematodes were washed with double distilled water and centrifuged to remove compounds. The nematodes were separately homogenized in centrifuge tubes (1.5 mL) using a TIANGEN hand homogenizer in 500 μL of ice–cold sodium phosphate buffer (20 mM, pH 7.2) to estimate the AChE enzyme activity. The homogenates were centrifuged (10,000 g at 4 °C) for 20 min and supernatants were used for the further analyses. The final homogenates were held in −20 °C until used in assays. Two hundred microlitre supernatants were mixed with 100 μL sodium phosphate buffer (20 mM, pH 7.2) and 100 μL 5 mM substrate solution incubated for 30 min at 25 °C. Then 100 μL 5 mM DTNB−ethanol reagent was added and the absorption was read immediately at 412 nm on a microplate reader. The median inhibitory concentration (IC50) was calculated using an enzyme inhibition dose response curve, with harmine hydrochloride as a positive control. All experiments were performed in triplicate.

Docking study

Docking studies were performed using BioSolveIT LeadIT Version: 2.1.8 (Germany). For this purpose, crystal structure of AChE (PDB codes: 1C2O) of electric eel (Electrophorus electricus) was obtained from the Protein Data Bank in order to prepare the protein for docking studies (Bourne et al. 1999). The docking procedure was followed using the standard protocol implemented in LeadIT. The amino acid sequence of the subunit was edited to remove the extracellular region and residues. The initial structures of these compounds were built and energetically minimized. The active site of the receptor for docking studies was identified as the amino acid residues in 6.5 Å pocket.

Statistical analysis

All data were expressed as means ± standard deviations of triplicate measurements. Standard deviations (SD) did not exceed 5% for the majority of the values obtained. The treatment means were subjected to a one–way ANOVA and Tukey’s multiple. The median lethal concentration (LC50) against B. xylophilus and IC50 on AChE were obtained according to probit analysis.

Results

Nematicidal effect of harmine derivatives

Nematicidal effects of harmine hydrochloride and harmine derivatives are shown in Table 2. Compounds 10, 11, 12, 13 and 26 showed 100% nematicidal effect against B. xylophilus; compounds 20, 23, 24, 25 and 35 showed 40–90% nematicidal effect, and slightly higher than the parent compound, harmine hydrochloride (39%), 15 compounds had an effect ranging between 3 and 40%, the other 18 compounds have no activity, at a concentration of 20 μg/mL. In our experiment, the dead nematodes treated with harmine derivatives 10, 11, 12 and 13 usually had straight bodies, whereas nematodes treated with harmine derivatives 20, 23, 24, 25, 26 and 35, usually had semicircular and coiling shapes.

Table 2 Preliminary nematicidal effects of harmine hydrochloride and harmine derivatives against Bursaphelenchus xylophilus at a concentration of 20 μg/mL
Full size table

To gain further insight into potency of these compounds, harmine hydrochloride and the compounds 10, 11, 12, 13, 20, 23, 24, 25, 26 and 35 were investigated further in a serial concentration gradient to determine their LC50 values (Table 3). The LC50 values of harmine hydrochloride and the compounds 10, 11, 12, 13, 20, 23, 24, 25, 26 and 35 were 27.50, 1.63, 1.63, 1.75, 1.44, 25.59, 18.31, 10.22, 12.72, 5.84 and 13.32 μg/mL, respectively.

Table 3 Nematicidal effects of harmine hydrochloride and harmine derivatives against Bursaphelenchus xylophilus
Full size table

Inhibitory effect of harmine derivatives against AChE in vitro

The inhibitory effects of harmine hydrochloride and harmine derivatives on AChE from electric eel in vitro were evaluated (Table 4). Compound 13 showed the greatest potential to inhibit AChE activity. Compounds 10, 11, 12 and 13 had a prominent inhibitory effect on AChE, with IC50 values of 0.92, 0.90, 0.82 and 0.07 μg/mL, respectively. Compounds 20, 23, 24, 25 and 35 had low to moderate effect on AChE activity with IC50 values of 11.65, 28.26, 13.26, 14.40 and 5.00 μg/mL, respectively. The IC50 values of harmine hydrochloride was 5.50 μg/mL. However, compound 26 showed no inhibitory effect.

Table 4 Inhibition of acetylcholinesterase by harmine hydrochloride and harmine derivatives in vitro
Full size table

Inhibitory effect of harmine derivatives against AChE in vivo

Because these compounds showed good inhibitory effect on AChE in vitro, we further evaluated the inhibitory effect of harmine hydrochloride and harmine derivatives on AChE of B. xylophilus in vivo. Similar results were obtained (Table 5). Compound 13 proved to be the most active. IC50 values of the compounds 10, 11, 12 and 13 were 17.16, 14.56, 13.63 and 3.06 μg/mL, respectively, significantly lower than that of harmine hydrochloride, 578.44 μg/mL. Compound 20, 23, 24, 25, 26 and 35 had no effect.

Table 5 Inhibition of acetylcholinesterase by harmine hydrochloride and harmine derivatives in vivo
Full size table

Study on structure–activity relationship (SAR)

In our study, 2,7,9–trisubstituted harmine derivative 10, 11, 12, 13, 20, 23, 24, 25, 26 and 35 showed good nematicidal effect (Table 1). Among the 2,9–positions together as the electron–donating substituted and the 7–position as electron–withdrawing substituted harmine derivatives 10, 11, 12 and 13, showed 100% mortality at a concentration of 20 μg/mL, with LC50 values of 1.63, 1.63, 1.75 and 1.44 μg/mL, respectively (Tables 2 and 3), and had also a remarkable inhibitory effect on AChE activity, with IC50 values of 0.92, 0.90, 0.82 and 0.07 μg/mL in vitro and 17.16, 14.56, 13.63 and 3.06 μg/mL in vivo, respectively (Tables 4 and 5). In the case of compound 20 with 2,9–positions together as the electron–donating substituent groups, compounds 23 and 25 with 6–position as bromine, compound 24 with 8–position as bromine and compound 35 with 7–position as electron–withdrawing substituent groups, showed 40–87% mortality at a concentration of 20 μg/mL, with LC50 values of 25.59, 18.31, 12.72, 10.22 and 13.32 μg/mL, respectively (Tables 2 and 3). These compounds had a moderate inhibitory effect on AChE activity in vitro, with IC50 values of 11.65, 28.26, 14.40, 13.26 and 5.00 μg/mL, respectively (Table 4), but had no inhibitory effect in vivo (Table 5). In addition, the harmine derivative 26, with 9–position as the electron–donating and 6–position as bromine substituted, showed also 100% mortality at a concentration of 20 μg/mL, with LC50 values of 5.84 μg/mL (Tables 2 and 3); but showed no inhibitory effect on AChE activity in vitro nor in vivo in our experiment (Table 5).

Docking study

To clarify the binding mode of our synthesized compounds, compound 13 with high AChE inhibitory activity was docked into the active site of AChE by FlexX™ Docking within BioSolveIT LeadIT Version: 2.1.8 software package. The result revealed that compound 13 was flexibly docked automatically in the active site of AChE. Firstly, the beta–carboline ring and the ethyl group in 9- position of harmine derivatives are oriented into substrate access AChE in the middle and forms hydrophobic and Van der Waals interactions with surrounding hydrophobic residues (yellow). Secondly, the phenyl group attached to the beta–carboline of the side chain interacts with the phenyl group of aromatic amino acids through the formation of π-π face-to-edge interaction and N+ cation in 2–position interacts with the anionic site in the peripheral anionic site (PAS) (green). Furthermore, the phenethoxy group in 7- position of beta–carboline can bind to the catalytic active site (CAS) (orange) of AChE through the former binding mode (Fig. 1).

Fig. 1
figure 1

Simulated binding site interactions of compound 13 with the acetylcholinesterase (AChE) structure of electric eel (Electrophorus electricus) depicted as spectrum-colored cartoon (containing α-helix, β-sheet and flexible loop) and the arrow shows the AChE core (yellow) colored in spectrum with the N-terminus beginning in blue and the C terminus finishing in red. The carbon atoms of compound 13 are green. Simulated docking shows docking compound 13 in the same site formed π − π stacking interactions and hydrogen bonding

Full size image

Discussion

The naturally and synthetic beta–carboline derivatives provide a valuable scaffold in medicinal chemistry as well as agrochemical applications with herbicidal, fungicidal, insecticidal and nematicidal activities (Bloomquist et al. 1997; El Hassan et al. 2013; Larson et al. 1988; Li et al. 2015; Liu et al. 2014; Saeed and Shawkat 2014; Sodaeizadeh et al. 2010; Song et al. 2014). In the current study, we report the 2,7,9–trisubstituted harmine derivatives against B. xylophilus and their inhibitory effects on AChE in vitro and in vivo and detailed studies of SAR. This work clearly shows that compounds 10, 11, 12 and 13 with 2,9–positions together as electron–donating substituent groups and the 7–position as electron–withdrawing substituent group caused 100% mortality at a concentration of 20 μg/mL and had remarkable inhibitory effect on AChE activity in vitro and in vivo. This result agrees with previous findings that the substituent at the 7–position of the beta–carboline alkaloids and 3,9–positions of quinoline alkaloids played a critical role in AChE or BChE inhibition (Yang et al. 2015; Zhao et al. 2013; Zheng et al. 2009). This also agrees with previous findings that 2,9–positions together as the electron–donating substituted and 7–position as electron–withdrawing substituted harmine derivatives had remarkable neurotoxic effects including tremor, twitch and jumping in experimental animal models and cytotoxic properties in vitro (Shi et al. 2013). Due to the different residues in the PAS, the three AChEs (BxACE-1, BxACE-2, BxACE-3) of pinewood nematode showed differential inhibition properties by the plant essential oils, organophosphates (OPs) and carbamates (CBs) (Kang et al. 2011, 2012, 2013). The differential inhibition activities against AChE in vitro and in vivo of harmine derivatives can be helpful in developing an efficient anti-AChE nematicidal agent for control of pinewood nematode.

On the other hand, compound 26 with 9–position as electron–donating substituent groups and 6–position as bromine, compound 20 with 2,9–positions together as the electron–donating substituent groups, compounds 23 and 25 with 6–position as bromine, compound 24 with 8–position as bromine and compound 35 with 7–position as electron–withdrawing substituent groups caused 40–100% mortality at a concentration of 20 μg/mL, but proved to be ineffective in inhibiting AChE in vitro and in vivo, and so their exact mode of action against pinewood nematode is unclear. Kong et al. (2006) reported that pinewood nematode bodies treated with the muscle activity blockers levamisole hydrochloride and morantal tatrate usually exhibited semicircular and coiling shapes, respectively. Because the symptoms resemble nerve intoxication, we speculate that the compounds might act on the nervous system.

These results suggest that the nematicidal mode of action among the different substituent groups of harmine derivatives might be different. As compared with previous work, the binding conformation of compound 13 in this docking simulation showed the similar binding mode of alkylene-linked tacrine dimers (Chen et al. 2014; Lu et al. 2011; Rydberg et al. 2006). It helps us to optimize these harmine derivatives to inhibit AChE. To some extent, these findings, combined with the SARs results, could explain the novel mechanism of the nematicidal effect by these new derivatives.

Overall, harmine derivatives 10, 11, 12 and 13, appear to be useful as natural product–based nematicides for B. xylophilus through inhibition of AChE. In addition, the docking results revealed that the structural uniqueness of our newly designed harmine derivatives led to a unique molecular recognition and binding mode and were in good agreement with their high nematicidal potential, which also explained the SARs observed in vitro and in vivo. These results suggest that beta–carboline nucleus is the primary pharmacophore of the nematicidal activity of beta–carboline derivatives and may be helpful in structure–guided design and development of novel nematicidal harmine derivatives based on AChE inhibitors. Since harmine derivatives 20, 23, 24, 25, 26 and 35, have potential as natural product–based nematicides for B. xylophilus, the mechanism of their nematicidal activity deserves further study. The active derivatives could be considered as potential alternatives to currently used pinewood nematode control agents or as lead compounds for the development of synthetic environment friendly nematicides with enhanced activity.