B. xylophilus was collected from the Jinju in Korea by the method described in previous study [17] and identified by real-time species-specific PCR [18]. Identified nematode was reared on a lawn of Botrytis cinerea cultured on PDA plates (media-grown propagative mixed stage, MGPS) at 28°C for up to a week. Fresh nematode washed by M9 buffer [19] was used immediately after separation from plates.

Recombinant BxACEs were expressed by baculovirus system described in previous studies and their activity was verified by kinetics [17]. Immunogens for polyclonal antibody production were expressed using a bacterial expression system. cDNA fragments encoding 100 amino acids from the N-terminus of each AChE but excluding the signal peptide sequence were inserted into the pET28a(+) expression vector (Merck, Darmstadt, Germany) and then cloned into Escherichia coli BL21(DE3). Immunogens were expressed by IPTG induction, and then purified using a His-tag column. The purified antigens were injected into a rabbit three times, and BxACEPabs were obtained (Ab Frontier, Seoul, Korea). BxACEPabs were purified by an affinity chromatography column using the respective antigens.

MGPS of B. xylophilus was used for immunohistochemistry of BxACEs. A whole-body immunohistochemistry procedure was conducted using the tube-fixation protocol according to Wormbook [20]. The nematodes were rinsed with M9 buffer more than three times and fixed with 4% paraformaldehyde after freeze-fracturing with liquid nitrogen. Subsequently, β-mercaptoethanol and collagenase (type VII, Sigma-Aldrich, St. Louis, MO) were added to increase the permeability of the antibody. The collagenase-treated nematodes were blocked in 10% goat serum albumin (Jackson ImmunoResearch, West Grove, PA) in antibody buffer (pH 7.2). The BxACEPabs and anti-rabbit Alexa568 (Molecular probes, Eugene, OR) were added successively. A mixture of BxACEPab and the target recombinant BxACE (1 5 w/w) was used as a negative control, whereas mixtures with BxACEPab and the other recombinant BxACEs (1 5 w/w) were used for a positive control. The mixtures were pre-absorbed for 6 hr at room temperature. The nematodes treated with primary and secondary antibodies were mixed with Vectashield (Vector, Burlingame, CA) and mounted on glass slides. The whole-mount samples were photographed on a Zeiss LSM510 (Carl Zeiss, Oberkochen, Germany) and IX71 inverted optical microscope (Olympus, Tokyo, Japan). Digital images were processed using an LSM image browser (Carl Zeiss) and Adobe Photoshop (Adobe Inc, San Jose, CA). The nervous system anatomy of B. xylophilus was based on other model nematodes, including C. elegans and Ascaris suum, because there is a remarkable conservation of neuronal morphology in nematodes despite large differences in size and habitat [21].

To determine whether BxACEs were associated with the membrane, proteins from B. xylophilus were extracted using an ultrasonicator, Sonifier 450 (Branson Ultrasonics, Danbury, CT) in 0.1 M Tris-HCl (pH 7.8) containing 20 mM NaCl in the presence or absence of 0.5% Triton-X. Native PAGE was conducted at 120 V for 90 min at 4°C, and the gel was electro-blotted to a nitrocellulose membrane (GE Healthcare, Uppsala, Sweden). Each BxACE was detected using BxACEPabs.

Acetylthiocholine iodide (ASChI), 5,5′-dithiobis(2-nitro-benzonic acid) (DTNB), bovine serum albumin (BSA), α-pinene and limonene were purchased from Sigma-Aldrich. Insecticides, including dichlorvos, chlorpyrifos-oxon, and carbofuran, were purchased from ChemService (West Chester, PA).

To investigate the effects of BxACE-3 on the inhibition of BxACE-1 and BxACE-2, BxACE-1 and BxACE-2 were mixed in a ratio of 6:1, based on their transcription levels in the pine-grown propagative stage (PGPS) [17] to simulate the in vivo state of BxACEs, and used in an inhibition assay. Five concentrations of dichlorvos, chlorpyrifos-oxon and carbofuran were directly added to the mixture of BxACE-1 and BxACE-2. Otherwise, the inhibitors were first pre-incubated with BxACE-3 for 10 min and then added to the mixture of BxACE-1 and BxACE-2. In the final BxACE mixture, the molar ratio of BxACE-1:BxACE-2:BxACE-3 was 6:1:18, on the basis of their respective transcript levels in the PGPS [17]. A mixture containing all three BxACEs but no inhibitors was used as control. In another control, the same molar amount of BSA instead of BxACE-3 was added to the mixture of BxACE-1 and BxACE-2 to measure the nonspecific interaction of BSA with inhibitors. The final concentrations of ASChI and DTNB were 1 mM and 0.4 mM, respectively. The remaining activity was measured after 5-min incubation with inhibitor as described above, and IC₅₀ was calculated using the Polo Plus program (Ver. 2.0, Leora software, Petaluma, CA). An equality analysis between different treatments was conducted using GLM repeated measures ANOVA in the SPSS program (Version 12.0 K for Windows, SPSS Inc., Chicago, IL).

To determine the anti-AChE activity of α-pinene and limonene, each BxACE sample was incubated with six serially diluted concentrations of the test chemical in 0.1 M Tris-HCl buffer (20 mM NaCl and 1% acetone, pH 7.8) containing ASChI (final 1 mM) and DTNB (final 0.4 mM). The remaining AChE activity was measured as described above.

For the amplification of DNA templates used for dsRNA synthesis, primers containing T7 polymerase promoter sequence (Table S1) were designed using the BLOCK-iT^TM RNAi designer (https://rnaidesigner.invitrogen.com/rnaiexpress, Invitrogen, Carlsbad, CA). The pQE30 plasmid sequence, which does not exist in B. xylophilus, was used as the negative control. A PCR product was amplified using Ex Taq polymerase (Takara, Shiga, Japan), from which double-strand RNA (dsRNA) was prepared using MEGAscript RNAi Kit (Ambion, Austin, Texas). Approximately 300–500 mixed-stage nematodes were soaked in 20 µl RNAi solution (0.25x Mg²⁺-free M9 buffer) containing 3 mM spermidine (Sigma-Aldrich), 0.05% gelatin (Sigma-Aldrich), lipofectin (Invitrogen) and dsRNA (4 µg/μl).

Chlorpyrifos-methyl and fenamiphos were purchased from ChemService. After incubation in dsRNA solution for 1 day at 30°C, approximately 50–100 nematodes were treated with a diagnostic dose of chlorpyrifos-methyl, dichlorvos and fenamiphos (LC₅₀ value, unpublished) by soaking at RT and mortality was determined at 24 hr post-treatment. Total RNA was extracted from remaining nematodes using TRI reagent (MRC, Cincinnati, OH) and used for qPCR to determine the extent of target gene knockdown. The qPCR was conducted by the procedure described in previous studies [17]. Bioassay and qPCR were conducted in three biological repeats.

The specificity of three BxACEPabs for immunohistochemistry was determined by Western blot using recombinant BxAChEs. Little cross-reactivity was observed among the three BxACEPabs (Fig. S1). BxACE-1 was observed in several histological regions of B. xylophilus (Fig. 1A). BxACE-1-specific fluorescent signals were detected in the nerve ring (NR) around the pharynx, where most sensory integration takes place, the putative ventral nerve cord (VNC), the preanal ganglia region (PGA) and the dorsal nerve cord (DNC). BxACE-2 was detected in limited regions that were different from those expressing BxACE-1 (Fig. 1B). BxACE-2 was mostly observed around the NR, particularly at the post-NR, which appears to be the retrovesicular ganglia (RVG), a cluster of 13 neuronal cell bodies in the VNC [21], and at the vulva of females, but not at the spiculus of males. BxACE-1 and BxACE-2 signals were mostly non-overlapping. BxACE-3 was observed around the putative hypodermal cells (HC) at the end of the head (Fig. 1C) and the tip of the tail, which cover the major opening of the body. In addition, BxACE-3 was observed at the putative intestinal regions and the canal-associated neuron (CAN), where BxACE-1 and BxACE-2 were not detected. BxACE-3 was not observed around the NR.

Native PAGE using the crude proteins extracted with buffer with or without Triton-X 100 followed by Western blot revealed that a large fraction of BxACE-3 was successfully extracted with the buffer without detergent. In contrast, relatively much reduced amounts of BxACE-1 or BxACE-2 were extracted without detergent. These findings suggest that a large part of BxACE-3 is not associated with a lipid anchor and exists mostly in the soluble fraction unlike the membrane-anchored BxACE-1 and BxACE-2, further suggesting the distribution of BxACE-3 in extracellular space (Fig. 1D).

In fact, BxACE-3 does not have a GPI-anchoring sequence or an H-type hydrophobic sequence [17], suggesting that it is likely secreted to the extracellular space or to the hemocoel. Presence of soluble form AChE3 has not been confirmed in other nematodes, including the free-living nematode Caenorhabditis elegans and some phytopathogenic nematodes such as H. glycines, M. arenaria and M. incognita assumed membrane-associated protein [7], [15], [16], [22], [23].

Based on the assumption that a large fraction of BxACE-3 is present in a soluble form in the extracellular space or/and body fluid, we tested the hypothesis that BxACE-3 plays a role in chemical defense against xenobiotics as in the case of vertebrate BuChE [24]. The inhibition rate of three anti-cholinesterase inhibitors (dichlorvos, chlorpyrifos-oxon and carbofuran) was measured by in vitro assay using a pooled preparation of purified BxACE-1 and BxACE-2 with or without the addition of BxACE-3 (Fig. 2). When the mixture of BxACE-1 and BxACE-2 was inhibited with diagnostic doses of the inhibitors, the overall inhibition was always significantly reduced (1.9- to 2.3-fold based on IC₅₀) in the presence of BxACE-3 as determined by GLM repeated measures ANOVA (p<0.05) (Table S2). In particular, the lowest inhibition rate was observed when inhibitors were pre-incubated with BxACE-3 for 10 min and then added to the mixture of BxACE-1 and BxACE-2. The addition of BSA to the mixture of BxACEs also reduced the rate of inhibition, though the rates of reduction were lower than those obtained by the addition of BxACE-3. Because BxACE-3 activity itself was negligible, the addition of BxACE-3 to the mixture of BxACE-1 and BxACE-2 did not likely contribute to the remaining AChE activity measured after inhibition. In fact, overall control AChE activity was not changed by the addition of BxACE-3 to the mixture of BxACE-1 and BxACE-2. Therefore, the significant reduction in AChE inhibition observed in the presence of BxACE-3 indicates that BxACE-3 protects BxACE-1 and BxACE-2 from inhibition, perhaps by sequestering inhibitors.

BxACE-3 exhibited the highest transcription level in all of the life stages [17], which is contrast to C. elegans in which the transcript level of orthologous AChE was similar to other AChEs as estimated by Northern blotting [7]. BxACE-3 also showed the lowest catalytic efficiency toward several substrates in spite of its high affinity to them [17], implying that it is not likely involved in the hydrolysis of neurotransmitter. These findings further suggest that BxACE-3 function as a ubiquitous sequestration protein, rather than as a specific hydrolase, by binding nonspecifically to a variety of xenobiotics. Taken together, BxACE-3 appears to play little role in postsynaptic transmission but rather has non-classical functions, such as protection of the nervous system from toxic xenobiotics, probably by sequestering them.

In this sense, BxACE-3 resembles the vertebrate butyrylcholinesterase (BuChE, EC 3.1.1.8), which serves as an effective sequestering agent against toxins or toxicants before they arrive at their target sites [25], [26]. The BuChE is known to have several functions, including a backup for AChE, an auxiliary role in development and a protective function in detoxifying orally ingested toxic compounds [27], [28], [29]. This soluble enzyme is observed at high concentrations in plasma, liver, lung and intestine [30] and expressed ten times more than AChE in humans [31]. It is generally agreed that BuChE removes the artificial or natural anti-cholinesterase inhibitors by being irreversibly inhibited by them [32], [33]. In fact, intravenous administration of human BuChE to figs or monkeys increases LD₅₀ against nerve agents, demonstrating its protective function [25], [26], [34]. Although BuChE appears to be diverged from AChE in early chordate lineage [35] and to belong to a completely separate lineage from BxACE-3 [17], it is interesting that both vertebrate BuChE and nematode BxACE-3 share a similar physiological function as a bioscavenger providing chemical defense against xenobiotics.

BxACE-3 transcription in the nematodes treated with Bxace-3 dsRNA (Bxace-3^–) was 48% suppressed compared to the nematodes treated with pQE30 dsRNA (pQE30^–) (Fig. 3A). Nevertheless, no abnormal phenotype or mobility was observed in the Bxace-3^– nematodes (data not shown). In bioassays using chlorpyrifos-methyl, dichlorvos and fenamiphos, however, the mortality of Bxace-3^– nematodes (49.5, 41.2 and 38.0%) was 1.3–2.0-fold higher than that of pQE30^– nematodes (27.2, 20.4 and 28.4%), demonstrating that knowndown of Bxace-3 increased the mortality of B. xylophilus (Fig. 3B). These lines of evidence confirm the role of BxACE-3 in chemical defense against various xenobiotics, including nematicides.

Constant exposure of the monophagous B. xylophilus to the toxic secondary compounds of pine tree hosts may have driven BxACE-3 to adapt to such a chemical defense function. B. xylophilus spends most of its life inside pine tree except for migration via insect vectors and must counteract against various physical and chemical defenses by the host pine tree, particularly in the propagative stage. Pine resin is a major defense compound against various pests, including insects, nematodes and fungi [36], [37] and is composed of several terpenes, toxic primary metabolites. The α-pinene (bicyclic terpene) and limonene (monocyclic terpene), which are the most common constituents of pine resin, are known to be AChE inhibitors [38], [39], [40], [41], [42]. Therefore, it can be postulated that B. xylophilus has evolved the chemical defense system based on BxACE-3 against the anti-AChE terpene compounds.

To test this hypothesis, AChE inhibition assays using α-pinene and limonene were conducted. α-pinene inhibited BxACE-1 in a dose-dependent manner (Fig. 4). BxACE-1 activity was inhibited approximately 25% by 2 mM α-pinene but not by limonene. BxACE-2 was not inhibited by these compounds but BxACE-3 activity was inhibited ∼10% by 2 mM α-pinene and limonene. Inhibition of BxACE-3 by both α-pinene and limonene indicates that BxACE-3 interacts with these secondary toxins, thereby protecting the main postsynaptic BxACE-1 from inhibition. The higher level of expression in the pine-grown stage and the specific distribution around the end of head, the tip of tail and the canal, the first barriers against xenobiotics, further imply a chemical defense function of BxACE-3 against the secondary pine tree toxins. Considering that the free-living C. elegans ACE-3 shows different expression patterns from BxACE-3 [9], the chemical defense function of BxACE-3 is likely unique to phytophagous nematodes such as B. xylophilus.

AChE is presumed to be generated before cnidarians, but ancient AChE appears to have non-neuronal functions because AChE of cnidarians shows a low catalytic activity and is mostly expressed in non-nervous tissues [43], [44]. It is interesting that AChE in cnidarians is mainly associated with non-neuronal functions although they have the postsynaptic nervous system [45], [46]. These facts suggest that early AChEs did not play roles related to neuronal function and the association of AChE to postsynaptic transmission was an event after the evolution of cnidarian. In fact, the generation of proteins having the activity in nervous tissues and the structural features of the active classical AChE appears in the ancient bilaterian, such as some flatworms and nematodes [47], [48]. Based on this, it is highly probable that the Nematoda is an intermediate phylum in the early bilaterian having both neuronal and non-neuronal AChEs. In case of B. xylophilus, both BxACE-1 and BxACE-2, which have relatively high catalytic activities, are involved in the postsynaptic transmission, whereas BxACE-3 appears to adapt to a non-neuronal role of chemical defense against xenobiotics. The role of BxACE-3 as a bioscavenger provides important insights into the functional diversification of AChE.