Spinosad is a macrocyclic lactone, isolated from the microorganism Saccharopolyspora spinosa (Sparks et al. 1998; Thompson et al. 2000). It is a naturally occurring mixture of two components, spinosyn A and spinosyn D (Fig. 1) and was introduced as a commercial insecticide in 1997 (Thompson et al. 2000). Spinosad is used extensively in crop protection to control a wide range of insect pests, including lepidoptera and thysanoptera, but it is also used in animal health applications and to control head lice in humans. Insect toxicity is associated with widespread neuronal excitation in insects (Salgado et al. 1998), because of its action on nicotinic acetylcholine receptors (nAChRs) (Salgado and Saar 2004).
Figure 1. Chemical structure of spinosad and ivermectin. Spinosad is a mixture of spinosin A (in which R = H) and spinosin D (in which R = CH3). Also shown is ivermectin, another macrocyclic lactone pesticide. Ivermectin is a mixture of dihydroavermectin B1a (in which R = CH2CH3) and dihydroavermectin B1b (in which R = CH3).
Download figure to PowerPoint
Nicotinic receptors are members of the Cys-loop family of ligand-gated ion channels (Lester et al. 2004) and are important neurotransmitter receptor subtypes in both vertebrate and invertebrate species (Millar and Denholm 2007; Millar and Gotti 2009; Jones and Sattelle 2010). The Cys-loop family includes both excitatory (cation-selective) receptors, such as nAChRs, and also inhibitory (anion-selective) receptors (Lester et al. 2004). The inhibitory glutamate-gated chloride channel (GluCl), found in several invertebrate species, has close structural similarity to nAChRs and is the target site for ivermectin (Fig. 1), another macrocyclic lactone pesticide (Wolstenholme 2010).
In common with all Cys-loop receptors, nAChRs and GluCls are transmembrane proteins in which five subunits are arranged around a central ion channel pore. Each of the five subunits contains four α-helical transmembrane domains (TM1–TM4), with the second of these domains lining the ion channel pore. The conventional orthosteric agonist binding site is located within the extracellular domain of Cys-loop receptors at the interface between two adjacent subunits (Sine, 2002 #1539). However, several allosteric modulatory sites have also been identified in Cys-loop receptors. In the case of ivermectin, there is clear evidence that it interacts with an allosteric site in the transmembrane domain of GluCls (Hibbs and Gouaux 2011). In addition, ivermectin is an allosteric modulator of nAChRs, and there is evidence that it interacts with nAChRs via the receptor transmembrane region (Krause et al. 1998; Collins and Millar 2010). The binding site of spinosad on nAChRs is less well defined, but there is evidence that it also acts an allosteric ligand (Salgado and Saar 2004) at a site that is distinct from the conventional extracellular agonist binding site (Orr et al. 2009).
In common with most other pesticides, resistance to macrocyclic lactones such as spinosad and ivermectin is an established problem and one that is increasing as a result of intensive pesticide use (Wolstenholme and Kaplan 2012). Resistance to spinosad has been reported in several insect species (Wolstenholme and Kaplan 2012). For example, there have been reports of resistance in Colorado potato beetle Leptinotarsa decemlineata (Mota-Sanchez et al. 2006), house fly Musca domestica (Shono and Scott 2003) and tobacco budworm, Heliothis virescens (Young et al. 2003). In such species, there is evidence of resistance being a result of either enhanced metabolism (Markussen and Kristensen 2011) or a consequence of changes in the target site (Roe et al. 2010). Studies conducted with the model insect species Drosophila melanogaster have implicated the nAChR Dα6 subunit in determining target-site resistance to spinosad (Perry et al. 2007; Watson et al. 2010). For example, a Dα6 knockout strain of D. melanogaster has been shown to confer high levels of resistance to spinosad (Perry et al. 2007). In addition, a variety of chemically-induced mutations within Dα6 (generating either truncated proteins or mis-sense mutations) have been found to confer resistance to spinosad (Watson et al. 2010). Further evidence indicating that resistance to spinosad can arise through changes to its target-site (the nAChR α6 subunit) is provided by studies with the diamondback moth, Plutella xylostella (Baxter et al. 2010; Rinkevich et al. 2010). Resistance to spinosad in P. xylostella has been linked to mis-spliced transcripts of the nAChR α6 subunit resulting in expression of a truncated subunit protein (Baxter et al. 2010) and to point mutations generating premature stop codons (Rinkevich et al. 2010).
High levels of resistance to the insecticide spinosad have been reported in western flower thrips (Frankliniella occidentalis), particularly in areas such as southern Spain, where spinosad has been used intensively to protect greenhouse crops (Bielza et al. 2007a, b; Bielza 2008). In this study, we describe work conducted with a previously reported laboratory-selected strain of F. occidentalis (R1S) displaying high levels of resistance (resistance ratio > 350 000) to spinosad (Bielza et al. 2007b). R1S was selected from a field population of F. occidentalis collected in 2003 (in Almeria, Spain), from greenhouses that had been subjected to intensive treatment with spinosad (Bielza et al. 2007a, b). Resistance to spinosad in strain R1S has been reported to be autosomal, almost completely recessive and controlled by a single locus (Bielza et al. 2007b).
Initial studies of spinosad-resistant F. occidentalis indicated that resistance might be associated with target-site changes, rather than enhanced metabolism (Bielza et al. 2007a). These findings have prompted us to employ molecular biological techniques to examine the nAChR α6 subunit in F. occidentalis. A nicotinic acetylcholine receptor point mutation (G275E), located in the transmembrane region of the receptor has been identified in spinosad-resistant F. occidentalis. In addition to its identification in a laboratory-selected strain (R1S), we have also identified this resistance-associated mutation in a recently isolated field population of F. occidentalis (Guillén and Bielza 2012). As well as providing evidence for target-site resistance to spinosad in F. occidentalis, work described in this article also provides support for the proposal that spinosad acts as a nAChR allosteric modulator via a transmembrane binding site.
- Top of page
- Materials and methods
There is extensive evidence to indicate that insecticide resistance can arise as a consequence of enhanced metabolic detoxification (Scott 1999). However, in recent years, there has been increasing evidence that resistance can also occur as a consequence of mutations in the insecticide target site. The phenomenon of target-site resistance is well established for several insecticides, including organophosphates, acting on acetylcholine esterase and pyrethroids, acting on voltage-gated sodium channels. In contrast, it is only relatively recently that target-site resistance has been reported for insecticides acting on nAChRs (Millar and Denholm 2007; Wolstenholme and Kaplan 2012). For example, point mutations altering single amino acids in nAChR α or β subunits have been described that are associated with resistance to neonicotinoid insecticides (Liu et al. 2005; Bass et al. 2011).
Evidence has also accumulated in recent years to indicate that resistance to spinosad can occur as a result of changes in its target site, the nAChR (Wolstenholme and Kaplan 2012). Disruption of the Dα6 subunit in Drosophila has been reported to confer resistance to spinosad (Perry et al. 2007). In addition, mis-spliced α6 transcripts and truncated α6 subunits in Plutella xyostella are associated with resistance to spinosad (Baxter et al. 2010; Rinkevich et al. 2010). There is also evidence for resistance to spinosad in Drosophila because of chemically-induced mutations resulting in truncated or non-functional Dα6 subunits (Watson et al. 2010). In contrast, this study has identified a resistance-associated mutation located at a position close to a plausible binding site for spinosad. In addition, when this mutation is introduced into a closely related vertebrate nAChR, it generated a functional receptor with reduced sensitivity to spinosad, but no apparent effect on the potency of the endogenous agonist acetylcholine. In this respect, the spinosad resistance-associated mutation identified in Foα6 resembles a previously characterized nAChR mutation that is associated with resistance to neonicotinoid insecticides (Liu et al. 2005), which has a profound effect on agonist activation by neonicotinoids but only minimal effects on agonist activation by acetylcholine (Liu et al. 2006).
The best structural data available for the transmembrane region of a native nAChR are that generated by electron diffraction studies conducted with receptors purified from Torpedo electric organ (Unwin 2005). However, on the basis of higher-resolution X-ray diffraction data from bacterial ligand-gated ion channels and GluCl, it has been proposed that the assignment of amino acids in the TM3 domain of the Torpedo nAChR is of register by four residues, equivalent to about one turn of the α helix (Corringer et al. 2010; Hibbs and Gouaux 2011). On the basis of this information, we have assigned the position of the amino acid that is mutated in spinosad-resistant F. occidentalis as being the fourth amino acid from the top of the TM3 helix.
As is illustrated in Fig. 2, the position of the G275E mutation in Foα6 is at a position analogous to an aspartic acid (D) in GluCl. Not only is this aspartic acid residue in very close proximity (4.4 Å) to bound ivermectin in the GluCl crystal structure, it is also one of the amino acids that is involved in forming a van der Waals interaction with ivermectin (Hibbs and Gouaux 2011). Given the known location of the ivermectin biding site in GluCl, it seems plausible that the G275E mutation might be in close proximity to the spinosad binding site on nAChRs. This is consistent with our data indicating that the A272E mutation in human α7 nAChRs has an effect on the modulation of agonist responses by both spinosad and ivermectin (Fig. 5). The finding that spinosad does not modulate agonist responses in a subunit chimera containing the extracellular domain of the nAChR α7 subunit but the transmembrane domain of the 5-HT3A subunit (Fig. 5d), is also consistent with spinosad binding at a transmembrane location, similar to the known binding site of ivermectin on GluCl (Hibbs and Gouaux 2011). Furthermore, competition binding data (Fig. 6) provide further support for the conclusion that spinosad binds at a site other than the conventional orthosteric nicotinic binding site and is in agreement with previous evidence indicating that spinosad modulates nAChRs by interacting with a site distinct from the conventional agonist binding site (Orr et al. 2009). There are reports that spinosad acts as an agonist on some insect nAChRs (Salgado and Saar 2004; Watson et al. 2010). This is entirely consistent with spinosad acting via an allosteric transmembrane site, given the recent evidence indicating that nAChRs can be activated by allosteric agonists binding to a transmembrane site (Gill et al. 2011, 2012).
Further evidence that the TM3 domain of Cys-loop receptors is important in the binding of macrocyclic lactones comes from studies conducted with an insect GluCl channel and a vertebrate glycine receptor (GlyR). Both studies have examined mutations influencing ivermectin and both have identified an amino acid in TM3 that is predicted to lie four amino acids below that of the G275E mutation in Foα6. Significantly, as four amino acids corresponds to one turn of an α-helix, the residue identified in this study and that identified in the insect GluCl and vertebrate GlyR are predicted to have side chains pointing in the same approximate orientation. A study investigating resistance to ivermectin identified a resistance-associated point mutation (G323D) in the TM3 domain of a GluCl subunit from the two-spotted spider mite Tetranychus urticae (Kwon et al. 2010). Interestingly, like the mutation that we have identified in Foα6, the mutation identified in the GluCl also corresponds to a change from a glycine to an acidic residue (Kwon et al. 2010). In addition, studies with the vertebrate GlyR have demonstrated that the amino acid in GlyR α1 subunit equivalent to G323 the GluCl in (A228) can confer either enhanced sensitivity (A288G) or reduced sensitivity (A288T) to ivermectin (Lynagh and Lynch 2010; Lynagh et al. 2011).
Considerable problems have been encountered in expressing insect nAChRs in heterologous expression systems (Millar 1999; Millar and Lansdell 2010). Indeed, such problems have been reported in connection with α6 subunits cloned from other insect species (Lansdell and Millar 2004). In situations where functional expression has been achieved with α6-containing nAChRs, it has been reported to be inconsistent and often unsuccessful (Watson et al. 2010). Attempts were made to express the cloned Foα6 subunit in Xenopus oocytes, but these were unsuccessful. Because of the relative ease with which the vertebrate nAChR α7 subunit can be expressed as a functional homomeric receptor, it has been used extensively as a model for investigating mutations affecting neonicotinoid insecticides (Matsuda et al. 2000; Shimomura et al. 2002, 2003; Amiri et al. 2008). By comparing the functional properties of the vertebrate α7 nAChR containing a TM3 A272E mutation, it has been possible to demonstrate that this mutation has no significant effect on acetylcholine potency, as might be expected for a mutation located far from the extracellular binding site for acetylcholine. The absence of an effect on acetylcholine agonist potency is similar to the effects that have been described previously for a target-site mutation associated with resistance to neonicotinoid insecticides (Liu et al. 2006). Significantly, the resistance-associated mutation identified in Foα6 has also been found to abolish modulation of human α7 nAChRs by spinosad. Although spinosad appears to act as an agonist of insect nAChRs (Salgado and Saar 2004; Watson et al. 2010), with features similar to that of an allosteric agonist (Gill et al. 2011), we have found that spinosad is an antagonist of human nAChRs. This difference in the influence of spinosad on two different nAChRs is not unexpected. The chemically related macrocyclic lactone ivermectin is a positive allosteric modulator of human α7 nAChRs (Krause et al. 1998), but a single point mutation in the transmembrane region can convert it from a positive to a negative allosteric modulator (Collins and Millar 2010). Consequently, it is plausible that spinosad might interact at a similar transmembrane site in insect and human nAChRs but have opposing effects. What is significant is that a A272E mutation introduced into the human α7 nAChR abolishes the modulatory effects of spinosad, perhaps through a direct action on its binding and, consequently, this mutation might reasonably be expected to have a similar effect on the interaction of spinosad with insect nAChRs. It seems likely that both spinosad and ivermectin modulate Cys-loop receptors by interacting with an allosteric transmembrane site. Furthermore, it appears that both of these macrocyclic lactones, depending on the receptor they are acting upon, can exert a range of allosteric modulatory effects. These include positive allosteric modulation (potentiation), negative allosteric modulation (non-competitive antagonism) and allosteric agonist activation (activation in the absence of a conventional orthosteric agonist). As has been demonstrated recently for allosteric modulators of vertebrate nAChRs, it appears that all of these effects can potentially occur through transmembrane allosteric binding sites (Young et al. 2008; Collins et al. 2011; Gill et al. 2011).
In summary, we have identified a resistance-associated point mutation in the transmembrane domain of the Foα6 subunit, in a position analogous to the known binding site for ivermectin in a related Cys-loop receptor. Studies with the vertebrate nAChR α7 subunit provide evidence to suggest that the TM3 G275E mutation identified in Foα6 may be responsible for conferring target-site resistance to spinosad by exerting a selective effect on modulation by spinosad at its presumed allosteric binding site, together with a negligible effect on acetylcholine acting at its extracellular orthosteric binding site.