Upload
others
View
14
Download
0
Embed Size (px)
Citation preview
1
Ivermectin binding sites in human and invertebrate Cys-loop receptors 1
Timothy Lynagh1, Joseph W. Lynch2 2
3 1 Neurosensory Systems Group, Technical University of Darmstadt, Schnittspahnstrasse 3, 64287 4
Darmstadt, Germany 5 2 Queensland Brain Institute and School of Biomedical Sciences, The University of Queensland, 6
Brisbane, QLD 4072, Australia 7
8
Corresponding author: Lynch, J.W. ([email protected]). Phone: +617 33466375 9
10
11
2
Abstract 12
Ivermectin is a gold-standard antiparasitic drug that has been used successfully to treat billions of 13
humans, livestock and pets. Until recently, the binding site on its Cys-loop receptor target had 14
been a mystery. Recent protein crystal structures, site-directed mutagenesis data and molecular 15
modelling now explain how ivermectin binds to these receptors and reveal why it is selective for 16
invertebrate members of the Cys-loop receptor family. Combining this with emerging genomic 17
information, we are now in a position to predict species sensitivity to ivermectin and better 18
understand the molecular basis of ivermectin resistance. An understanding of the molecular 19
structure of the ivermectin binding site, which is formed at the interface of two adjacent subunits 20
in the transmembrane domain of the receptor, should also aid the development of new lead 21
compounds as both anthelmintics and as therapies for a wide variety of human neurological 22
disorders. 23
24
New insight into a wonder drug 25
Orally- or topically-administered ivermectin is used to eliminate parasitic nematodes and 26
arthropods from animals and humans[1]. It is effective yet well-tolerated due to the potency with 27
which it activates a uniquely invertebrate member of the Cys-loop receptor family of ligand-gated 28
ion channels[2]. Although the activation of Cys-loop receptors by neurotransmitter agonists has 29
been studied in molecular detail for 20 years, the ivermectin interaction with these receptors has 30
received relatively little attention. This interaction requires attention, given the evidence that 31
ivermectin resistance can occur via mutations that affect the ivermectin sensitivity of Cys-loop 32
receptors [3,4]. In the last year, the binding sites of ivermectin in nematode and human Cys-loop 33
receptors have been identified, revealing key molecular determinants of ivermectin sensitivity. 34
Coinciding with these discoveries, a growing number of complete genomes have revealed the 35
Cys-loop receptor complements of species of known ivermectin susceptibility. Combining this 36
knowledge should allow the prediction of ivermectin sensitivity, help understand resistance 37
mechanisms and inform the development of broader spectrum antiparasitic drugs. Recent insights 38
in this area are reviewed here, following a brief introduction concerning the use of ivermectin and 39
the structural basis of Cys-loop receptor function. 40
3
41
Origins of ivermectin 42
Intensive screening of soil-dwelling microbes in the 1970s led to the isolation from the 43
bacterium, Streptomyces avermectinius, of a group of natural macrocyclic lactones with potent 44
nematicidal activity, named avermectins [5]. Saturation of the avermectin 22-23 double bond 45
produced 22,23-dihydroavermectin B1, or ivermectin. Ivermectin consists of a mixture of B1a 46
and B1b components in a 80:20 ratio. These compounds vary in structure only at C25, a moiety 47
that is not involved in binding. As the B1a component is more lethal to nematode, insect and 48
acarid parasites [1], it is conventional to refer to either the mixture or the B1a component as 49
‘ivermectin’. Much attention has been paid to the use of ivermectin against Onchocerca volvulus, 50
a mosquito-transmitted parasitic nematode that causes river blindness, but its use now extends to 51
other human diseases including lymphatic filariasis, strongyloidiasis, scabies and head lice [1]. 52
Despite its broad antiparasitic spectrum, ivermectin is ineffective against platyhelminths [6,7], 53
including the blood fluke Schistosoma mansoni which causes bilharzia [8]. 54
55
Ivermectin targets Cys-loop receptors 56
Early investigations into the lethal effect of ivermectin on arthropods and nematodes showed that 57
ivermectin increased chloride conductance in neurons and thus decreased responses of neurons to 58
excitatory input [9,10]. The search for the identity of this conductance led to the cloning of two 59
novel subunits from the free-living nematode Caenorhabditis elegans that formed a recombinant 60
glutamate-gated chloride channel (GluClR) that was activated irreversibly by nanomolar 61
ivermectin concentrations [2]. Numerous GluClR subunits have since been cloned in various 62
ivermectin-susceptible invertebrates [11,12]. The identification of GluClRs as the target of 63
ivermectin was confirmed by their high ivermectin sensitivity when expressed recombinantly, 64
their expression in tissues sensitive to low doses of ivermectin, and the development of 65
ivermectin resistance upon their mutation [3,4,12-15]. GluClRs belong to the Cys-loop receptor 66
family of ligand-gated channels, which includes human excitatory nicotinic acetylcholine and 67
type-3 5-hydroxytryptamine receptors (nAChRs and 5-HT3Rs) and inhibitory GABA type-A and 68
glycine receptors (GABAARs and GlyRs). 69
4
The Cys-loop receptor family is large, with 102 subunits in C. elegans and 45 in humans [16,17]. 70
Numerous inhibitory receptors in this family are activated by ivermectin, including arthropod 71
histamine-gated chloride channels (HisCls) and pH-gated chloride channels (pHCls) and, perhaps 72
unexpectedly, human GABAARs and GlyRs. A deciding factor in the combined lethality in 73
nematode/arthropod parasites and safety in mammalian hosts is the differential potency with 74
which ivermectin activates these receptors. α GluClRs are typically activated by low (< 20) 75
nanomolar concentrations [2,14,18,19], whereas GABAARs and GlyRs require concentrations in 76
the low (1-10) micromolar range [20-23]. HisCls and pHCls respond to micromolar ivermectin, 77
but the dose dependency has not been established [24-26]. At several other receptors, ivermectin 78
activates no current but modulates (inhibits or potentiates) the responses to the neurotransmitter 79
agonist. These include some invertebrate GABAARs and the human α7 nAChR (below).. 80
The sensitivity of human Cys-loop receptors to micromolar ivermectin, along with the 81
preliminary findings of anti-spasticity efficacy in humans [28], begs the question as to what 82
concentration reaches in the brain. Therapeutic doses in mice, dogs and humans produce plasma 83
levels of 5-10 nM that persist for days, with ~2 nM in the brain [29-31], which is insufficient to 84
activate GABAARs and GlyRs. When α GluClRs are exogenously expressed in the mouse brain, 85
animals respond to systemic ivermectin [32], indicating that ivermectin safety rests on the 86
reduced sensitivity of mammalian Cys-loop receptors. P-glycoprotein, which transports 87
avermectins out of the nervous system, is also crucial to ivermectin potency in mammals and 88
nematodes, evidenced by the increased ivermectin toxicity that results upon its mutation [33,34]. 89
90
Structure-activity relationships in Cys-loop receptors and ivermectin 91
Before describing in detail the structural features that determine ivermectin sensitivity, we briefly 92
outline the structural basis of function in Cys-loop receptors and the determinants of efficacy 93
within the ivermectin molecule. Functional Cys-loop receptor pentamers are formed by five of 94
the same or, as is usually the case for native receptors, two to three different subunits (Figure 95
1a,b). A single subunit consists of: a large extracellular N-terminal domain (ECD) and four 96
membrane-bound helices (M1-M4) that constitute the transmembrane domain (TMD). Two 97
highly conserved ECD Cys residues are crosslinked within a functional receptor, forming the 98
5
eponymous Cys-loop, and the apposition of five M2 helices forms the central channel pore 99
(Figure 1b,c). Neurotransmitter ligands bind at the extracellular interface of two adjacent subunits 100
(red segments in Figure 1b), resulting in the closure of binding domain loop C around the agonist. 101
This triggers structural rearrangements at the ECD-TMD interface (purple segments in Figure 102
1b), and ultimately the opening of the channel gate (Figure 1c,d), allowing ions to flow through 103
the channel [35]. This chemoelectric signalling mechanism underlies most rapid synaptic 104
transmission in the brain. 105
The structural basis of activity of avermectins and milbemycins (Streptomyces spp-derived 106
compounds with similar structures and antiparasitic spectra to avermectins) has been extensively 107
reviewed[36]. Briefly, all compounds share a macrocyclic backbone and differ at the 108
disaccharide, spiroketal and benzofuran moieties, respectively located at the three “corners” of 109
the molecule (Figure 2). Evidence indicates that the benzofuran is the most important potency 110
determinant at GluClRs. For example, substitution of the hydroxyl at the C5 position on the 111
benzofuran ring ("C5-OH") for oxime or ketone in several avermectins reduces nematicidal 112
activity 100- and 10000-fold, respectively [37], implying that either an H-bond donating or 113
generally hydrophilic C5-substituent is required for high potency. In contrast, changes to the 114
spiroketal group have little effect on nematicidal potency [36,37]. Similarly, hydrogenation of 115
double bonds C10-C11 (in the macrocyclic backbone) and 22-23 (in the spiroketal) does not 116
reduce acaricidal potency, whereas hydrogenation of double bonds C3-C4 and C8-C9 (in or 117
adjacent to the benzofuran) reduces both acaricidal potency and activity at mouse GABAARs 118
[38,39]. The sugar moiety is not essential for nematicidal potency [36,37]. However, given that 119
epimerisation of the dissacharide decreases activity at GluClRs [40,41], polar C13 substitutions 120
yield less biologically active avermectins than lipophillic C13 substitutions [42], and size of the 121
sugar moiety correlates with affinity for P-glycoprotein [43], it is evident that the sugar may 122
modulate avermectin potency in vivo. 123
124
The binding site in anionic Cys-loop receptors 125
Hibbs and Gouaux recently solved the crystal structure of the C. elegans α GluClR complexed 126
with ivermectin [44]. This gives a clear picture of the ivermectin binding site and sheds light on 127
previous functional studies. Furthermore, with over 30 % amino acid sequence identity with the 128
6
α1 GlyR, it provides a high resolution template for the modelling of vertebrate GABAARs and 129
GlyRs, which had previously relied on nAChR and prokaryotic receptor structures with much 130
lower homologies [45-47]. Their structure shows ivermectin binding between the M3 of one 131
subunit (the principal or (+) face) and the M1 of an adjacent subunit (the complementary or (-) 132
face), with the benzofuran moiety contacting the M2 helix of the (+) subunit (Figure 3a). The 133
structure implies the existence of three H-bonds: (i) between the benzofuran C5-OH and the 134
hydroxyl sidechain of (+)M2-Ser15' (Ser260 in the α GluClR), (ii) between a spiroketal oxygen 135
and the hydroxyl sidechain of (+)M3-Thr285, and (iii) between the benzofuran C7-OH and the 136
backbone carbonyl oxygen of (-)M1-Leu218 (Figure 3b). Leu218, Ser260 and Thr285 are 137
highlighted in yellow in Figure 3c, as are residues at the equivalent positions in various other 138
Cys-loop receptors. Twelve other sidechains, highlighted in orange or red in Figure 3c, contribute 139
potential Van der Waals interactions. One of these is (+)M3-Gly281, a position in anionic Cys-140
loop receptors we will henceforth refer to as "M3-Gly". M3-Gly is entirely conserved in GluClRs 141
that show nanomolar ivermectin sensitivity (Figure 3c). In the crystal structure, the α carbon of 142
M3-Gly is approximately 4.5 Å from each corner of the ivermectin macrocycle, such that any 143
larger sidechain at this position - that is, any amino acid other than Gly - would hinder the access 144
of ivermectin to its binding site or alter the binding conformation of ivermectin (Figure 3b). 145
Hence, in the Haemonchus contortus α3B GluClR, the mutation of M3-Gly to Ala (the human α1 146
GlyR equivalent) shifts the ivermectin EC50 for activation from 39 nM up to 1.2 µM - exactly that 147
of the human α1 GlyR [48]. Indeed, the α3B GluClR and the α1 GlyR can be tuned to 148
remarkably similar ivermectin sensitivities by introducing a common residue at the M3-Gly 149
position [48]. Consistent with this, the Gly323Asp mutation (at the M3-Gly position) in the 150
Tetranychus urticae (two-spotted spider mite) GluClR was shown to increase by 18-fold the 151
avermectin dose required for half-maximal lethality [4]. Indeed, the equivalent Gly-Asp 152
substitution abolished binding of a tritiated milbemycin to the α3B GluClR [49]. We have not 153
found in the literature any example of a highly ivermectin-sensitive receptor without a Gly at the 154
M3-Gly position. 155
7
The binding site at the α1 GlyR has also been investigated using mutagenesis and molecular 156
modelling. Ivermectin was modeled as wedging into the same TMD interface, with the 157
benzofuran oriented towards the (+)M2 [50]. Surprisingly, the docking of ivermectin to the 158
Ala288Gly α1 GlyR (this is the M3-Gly position) did not differ greatly from the unmutated 159
receptor, suggesting that the on-rate of ivermectin, rather than the ultimate binding conformation 160
of ivermectin, is changed by this substitution [50]. Alternately, the glycine substitution may 161
increase the gating equilibrium constant, thus facilitating receptor activation by ivermectin. Bulky 162
substitutions at only two positions decreased both agonist and modulatory effects of ivermectin, 163
indicative of a decisive role in binding. These positions are (+)Ala288 and (-)Pro230. Both of 164
these positions are highlighted in red in Figure 3c, revealing the apparent requirement of a small 165
residue at the M3-Gly position and the conservation of the "M1-Pro". M1-Pro is in close 166
proximity to the crucial benzofuran moiety of ivermectin (Figure 3b), providing a possible steric 167
explanation for decreased binding at this mutant. In addition, this highly conserved residue 168
disrupts M1 helical structure thereby exposing a backbone carbonyl that is thought to form a H-169
bond with the ivermectin C7-OH [44]. However, the existence of this bond requires confirmation, 170
given that C7-OH is equidistant from both the M1 backbone carbonyl and the endogenous 171
ivermectin C1 carbonyl. 172
A significant role for the third putative H-bond (between O14 and T285 – see Figure 3b) can be 173
eliminated on the basis of site-directed mutagenesis experiments in the Ala288Gly α1 GlyR, and 174
the fact that the three GluClRs with nanomolar ivermectin affinity do not contain H-bonding 175
sidechains at the corresponding M3 position [50]. Moving deeper into the pore, substitution of α1 176
GlyR (+)M3-Leu291, (+)M2-Thr264 or (-)M1-Leu233 for bulky Trp sidechains does not reduce 177
ivermectin sensitivity but converts ivermectin from an agonist to an antagonist of glycine-178
activated currents [50]. Still further removed from the binding site, mutations in the conserved 179
Cys-loop, the pre-M1 domain and the M2-M3 linker (Figure 1b) decrease ivermectin potency 180
most likely by impairing its gating efficacy [14,18,49-51]. 181
Native Cys-loop receptors are often heteromeric and thus do not necessarily possess the five 182
binding sites available to ivermectin in homomers. The following evidence at the GlyR, however, 183
suggests that the binding of two molecules of ivermectin is sufficient for either direct activation 184
8
or potentiation. First, ivermectin activates α1β heteromeric and α1 homomeric GlyRs with equal 185
potency [21], even though heteromers possess a β-α1-β-α1-β arrangement [52] and ivermectin 186
only binds to the (+)α1/(-)β interface [48]. Second, the β-Ile312Gly substitution (at the M3-Gly 187
position) confers sensitivity to potentiation on the otherwise ivermectin-insensitive α1-188
Ala288Phe/β heteromer, presumably by introducing a (+)β/(-)α1 ivermectin binding site [48]. 189
190
M2 H-bonds 191
The putative H-bond between the ivermectin C5-OH and the M2-15' Ser of the GluClR and GlyR 192
seems a logical ivermectin binding interaction, as (i) it links the ivermectin molecule to the pore-193
lining M2 helix[44] and (ii) a H-bond donor at the C5-OH is crucial to avermectin nematicidal 194
potency (above). However, because various GluClRs with nanomolar ivermectin sensitivities 195
contain either an Ala or Ile at this position (Figure 3c), it is evident this H-bond is not required for 196
high ivermectin potency [4,18,53]. In support of this, the mutant M2-Ser15'Ile substitution in the 197
α1 GlyR causes no change to the agonist or modulatory effects of ivermectin [50]. However, 198
most anionic Cys-loop receptor subunits possess several nearby residues that could act as 199
potential H-bond acceptors in the absence of one at the M2-15' position. These include (+)M2-200
12', (-)M2-14', (+)M2-19' and also an (-)M1 position (Figure 4). The proximity of these 201
sidechains to the ivermectin molecule is suggested by site-directed mutagenesis data from the α1 202
GlyR: the Thr12'Trp receptor is not activated by ivermectin, the Gln226Cys receptor has 203
decreased ivermectin sensitivity and the Gln266Trp receptor has increased sensitivity [50]. The 204
conservation of polarity in this domain of ivermectin-sensitive receptors is illustrated by the 205
boxed residues in Figure 3c. 206
207
A clear pattern? 208
We now group the above ivermectin sensitivity determinants together and consider their chemical 209
properties in several receptors with known high ivermectin sensitivity. Dividing M2-12', M2-14', 210
M2-15', M2-19' and the M1-Gln residue into their (+) and (-) components gives a "(+)Thr-Ser-211
Arg/(-)Gln-Gln" arrangement, in the case of the α1 GlyR (Figure 4). The following five 212
9
homomeric Cys-loop receptors are particularly sensitive to ivermectin activation (i.e., activated 213
by 1-20 nM ivermectin): Ala288Gly α1 GlyRs, C. elegans α, H. contortus α and α3B and 214
Drosophila α GluClRs. Respectively, these subunits have (+)Thr-Ser-Arg/(-)Gln-Gln, (+)Thr-215
Ser-Asp/(-)Gln-Gln, (+)Thr-Ala-Asn/(-)Gln-Gln, (+)Thr-Ala-Asn/(-)Gln-Gln and (+)Ala-Thr-216
Asn/(-)Gln-Gln arrangements, and thus present an overwhelmingly polar environment to the 217
ivermectin C5-OH moiety. Given that the α1 GlyR possesses a similar polarity in this region to 218
GluClRs, we infer that its large M3-Gly sidechain is solely responsible for its reduced ivermectin 219
sensitivity. UNC-49B/UNC-49C heteromeric GABAARs possess an M3-Gly at the (+) faces of 220
UNC-49C, yet are not activated by ivermectin [54,55]. However, with a (+)Thr-Met-Asn/(-)Gln-221
Leu arrangement, they are distinct from the GluClRs and GlyRs in that both the M2-14' and M2-222
15' positions are occupied by larger hydrophobic residues (Figure 3c), perhaps limiting H-bonds 223
with and channel activation by ivermectin [44]. Similarly, the Dermacentor variabilis (American 224
dog tick) homomeric RDL receptor possesses an M3-Gly but is not sensitive to ivermectin 225
(Figure 3), and its M2-14' and M2-15' positions are occupied by hydrophobic residues [56]. By 226
contrast, MOD-1 is insensitive to ivermectin [57], despite its ostensibly suitable (+)Thr-Ser-227
Arg/(-)Gln-Ile arrangement; MOD-1 insensitivity is likely due to a Val residue at the M3-Gly 228
position preventing access to the site [50]. Curiously, although mammalian GABAAR π subunits 229
possess an M3-Gly, incorporation of the π subunit actually decreased the ivermectin sensitivity of 230
α1β2 GABAARs [48]. Again we suggest that the ivermectin-insensitivity results from 231
hydrophobic residues at both M2-14' and M2-15' positions in the π subunit [58]. Finally, 232
functional homomeric β GluClRs from C. elegans and C. onchophora are also completely 233
insensitive to ivermectin [59], and studies with radiolabelled ivermectin show that it does not 234
even bind to the H. contortus β GluClR [60]. These receptors contain a (+)Thr-Gln-Asn/(-)Gln-235
Met arrangement and an M3-Gly. We thus conclude that the (-)M1-Met is responsible for 236
disrupting ivermectin sensitivity. In the above analysis, we have used the GluClR crystal 237
structure to infer that the five polar residues are essential for high ivermectin affinity. However, it 238
is also possible that the overwhelmingly polar environment facilitates receptor gating by 239
ivermectin. 240
10
An alternate possibility in GABAARs is that ivermectin binds to a different site. For example, 241
GAB-1/HG1A heteromers and UNC-49B homomers possess Val, Leu or Asn residues at the M3-242
Gly position, and these receptors are potentiated but not activated by ivermectin [54,55,61], 243
perhaps reflecting the exclusion of ivermectin from the interfacial site and the existence of an 244
alternate site. Moreover, HG1A and HG1E differ by only four amino acids - in the ECD and in 245
M4 - and, whereas GAB1/HG1A heteromers are potentiated by ivermectin, GAB1/HG1E 246
heteromers are inhibited [61]. An alternative site, consistent with HG1A/E structural differences, 247
might be similar to that proposed for the mammalian α7 nAChR (see below). The possibility of 248
distinct ivermectin binding sites in various GABAARs, together with the large diversity in 249
GABAAR subunits, might underlie the differential effects of ivermectin on various native 250
GABAAR preparations [62]. 251
252
An alternative site for potentiation of nAChRs 253
Due to the differing M2-M3 domain lengths in anionic and cationic Cys-loop receptors, it is 254
difficult to determine which amino acid occupies the M3-Gly position in the mammalian α7 255
nAChR [63]. Most alignments suggest that Ala275 in the α7 nAChR and Leu279 in the Torpedo 256
marmorata (marbled electric ray) nAChR α subunit occupy the M3-Gly position. Although the 4 257
Å resolution structure of the T. marmorata nAChR places this Leu279 in the (+)M3 helix, it faces 258
the outside of the membrane, 1-2 helical turns above the level of (-)M1Pro (Figure 5a). In the T. 259
marmorata nAChR structure, the (+)M3 residue physically opposite (-)M1-Pro is either Met282, 260
Ile283 or Ile286, which equate with little ambiguity to Met278, Ile279 and Gly282 in the α7 261
nAChR (Figure 5a,b). Thus, the α7 nAchR structural equivalent to the M3-Gly might Gly282, 262
and the selectivity of ivermectin for α7 nAChRs over insensitive cationic Cys-loop receptors may 263
be due to Val, Ile and Thr sidechains at this position (e.g., in nematode ACR-16 receptors and 264
mammalian 5-HT3Rs [64,65]). Accordingly, the lack of H-bond propensity at M2-14', M2-15' and 265
M2-19' positions in the α7 nAChR might limit the direct agonist efficacy of ivermectin. 266
However, if the structural equivalent in α7 nAChR is Met278 or Ile279, access of ivermectin to 267
the interfacial site is likely precluded, and ivermectin might bind to a distinct site. 268
11
Computer docking simulations from two separate laboratories have shown that the lowest energy 269
docking of ivermectin to the α7 nAChR was in an intrasubunit cavity, as opposed to the 270
interfacial site in anionic receptors [64,66]. In both studies, the ivermectin molecule wedges 271
between M1 and M4 from the same subunit, partly in contact with the surrounding lipids, with 272
the benzofuran moiety oriented towards either M2 or M3 of the same subunit. Substitution of α7 273
nAChR TMD residues for the non-conserved equivalents from the 5-HT3AR resulted in decreased 274
potentation by ivermectin or conversion of potentiation to inhibition [64]. Three of the four 275
mutations that simply decrease potentiation line a space between M1 and M4 of the one subunit, 276
albeit two helical turns below the docked ivermectin molecule. Consistent with putative existence 277
of this site in the α7 nAChR and not in GluClRs and GlyRs, the latter possess large sidechains 278
that might obscure this cavity (Figure 5). Finally, the effects of ivermectin at nAChRs are 279
reversible and are decreased by ECD-acting competitive antagonists [67], which differs from 280
results at most anionic receptors and supports a different site or mechanism of ivermectin at 281
nAChRs. 282
283
Structural bases of ivermectin resistance 284
There are several reports that ivermectin resistance occurs nematodes in natural settings [1], in 285
some cases due to mutations in the coding sequences or changes in expression levels of GluClRs 286
or GABAARs [3,68,69]. In laboratory-induced mutant strains, single mutations in GluClRs have 287
been clearly shown to cause avermectin resistance. For example, mutation of the M3-Gly position 288
in the T. urticae GluClR causes an 18-fold increase in abamectin LD50 [4]. (A C22-C23 double 289
bond in the spiroketal of abamectin is the only difference from ivermectin.) It is also possible that 290
ivermectin insensitivity may be caused by mutations that allosterically disrupt the efficacy with 291
which ivermectin activates the receptors, without necessarily affecting ivermectin affinity. This 292
was first suggested by an M2-M3 linker mutation in a Drosophila GluClR that caused a four-fold 293
increase in ivermectin LD100 [14]. Disruptions to the efficacy of GluClR gating by ivermectin 294
most likely account for the effects of mutations identified in several resistant H. contortus strains 295
[18,49]. Of course, if a GluClR mutation converts ivermectin to an inhibitor, as demonstrated at 296
recombinant GlyRs and nAChRs (above), the usual CNS depression elicited by ivermectin could 297
12
be converted to excitation. This may underlie the resistance seen in nematodes with an up-298
regulation of HG1E GABAAR subunits that are inhibited by ivermectin [61,68]. 299
300
Predicting species susceptibility to ivermectin 301
The discovery of molecular determinants of ivermectin sensitivity in Cys-loop receptors provides 302
an opportunity to investigate parasites of known ivermectin susceptibility to establish the 303
molecular basis for this susceptibility. The complete genome of the trematode S. mansoni reveals 304
13 Cys-loop receptor subunits [70], much fewer than the 102 found in C. elegans and 45 in 305
humans [16,17]. It is therefore likely that ivermectin has fewer potential Cys-loop receptor targets 306
in S. mansoni than in C .elegans which might help explain why S. mansoni is insensitive to 307
ivermectin. On the other hand, arthropods also possess few GluClRs but are ivermectin-308
susceptible, suggesting that there must be some feature of S. mansoni GluClRs (other than a small 309
number in the genome) that renders them insensitive to ivermectin. An amino acid sequence 310
alignment of recently available transcripts in several trematodes suggests that they indeed possess 311
GluClRs (Figure 6). Although their functional properties remain to be investigated, these 312
transcripts possess ~30 % homology with α GluClRs and α1 GlyRs, suggesting a functional 313
relationship. Furthermore, they possess a Gly residue on the (+) face of the ECD neurotransmitter 314
binding domain (Figure 6), at a position where small residues are correlated with Glu binding and 315
larger residues are correlated with Gly, GABA or His binding[71]. These putative GluClRs 316
include largely polar (+)Ile-Ser-Ala/(-)Ser-Gln arrangements (as per the scheme above), but large 317
residues at the M3-Gly position undoubtably preclude ivermectin binding (Figure 6). Thus, large 318
sidechains at the M3-Gly position, as opposed to a lack of GluClRs, may well underlie the 319
ivermectin-insensitivity of trematodes. The free-living trematode Schmidtea mediterranea also 320
possesses GluClR-like transcripts with large sidechains at the M3-Gly position (Figure 6). The 321
recent sequencing of Ascaris suum (pig roundworm), Anopheles gambiae (mosquito) and 322
Daphnia pulex (water flea), by contrast, reveal one or more M3-Gly GluClR transcripts, as 323
expected for ivermectin-susceptible phyla [72-74]. This reiterates the importance of the M3-Gly 324
position. Of course, species susceptibility also depends on if and where the relevant subunits are 325
expressed, and the functionality with which an M3-Gly subunit heteromerises with other 326
subunits. 327
13
328
Concluding remarks 329
Ivermectin binds to nematode GluClRs and mammalian GlyRs at the interface of adjacent (+) and 330
(-) subunits in the TMD of the receptor. High affinity for this site is controlled by an M3 Gly 331
residue on the (+)M3 helix that allows rapid access to the site. The C5 hydroxyl of the ivermectin 332
benzofuran likely forms an H-bond within the subunit interface cleft, leading to potent channel 333
activation. Potent activity requires polar sidechains at either M2-14', M2-15' or M2-19' (and 334
perhaps other) positions. Scanning Cys-loop receptor amino acid sequences for these 335
determinants predicts ivermectin sensitivity reasonably well, and scanning genomes for these 336
transcripts may enable the prediction of species susceptibility to ivermectin. However, several 337
questions are yet to be answered, including the structural basis for modulation (not activation) by 338
ivermectin of certain GABAARs and nAChRs, and the inability of ivermectin to activate β 339
GluClRs. Attempts to answer these questions will be aided by the structural template provided by 340
the GluClR crystal structure. Given the emergence of ivermectin resistance in parasitic nematode 341
pests [75,76], the information presented here may help in the design of novel anthelmintics 342
targeting GluClRs or other nematode Cys-loop receptors. Similarly, because ivermectin binding 343
sites also exist in human Cys-loop receptors, understanding the ivermectin binding site structure 344
may help identify new therapeutic pharmacophores for a wide variety of human neurological 345
disorders. 346
347
14
Figure legends 348
Figure 1. Molecular architecture of Cys-loop receptors. (a) The C.elegans α GluClR crystal 349
structure, viewed from the extracellular space (top panel) and from within the lipid membrane 350
(lower panel) [44]. (b) Single subunit from (a). Red indicates segments that contribute to the 351
neurotransmitter binding domain, which is formed between the (+) face of one subunit and the (-) 352
face of an adjacent subunit; magenta indicates Loop 2, the conserved Cys-loop, the pre-M1 353
domain, and the M2-M3 linker, which couple neurotransmitter binding to channel movement; 354
helices M1-M4 are indicated; M2-Leu9' sidechains that contribute to the channel gate are shown 355
as blue spheres [35]. (c) Crystal structure of the closed bacterial channel ELIC [46] and (d) 356
crystal structure of the open bacterial channel GLIC [45], both viewed from the extracellular 357
space with ECDs removed for clarity; the hydrophobic M2-9' sidechain is moved from the 358
channel axis in models of channel activation. 359
360
Figure 2. The ivermectin molecule. (a) Natta projection of ivermectin, including the C atom 361
numbering referred to in the text. (b) 3D structure of ivermectin, as used for docking to the α1 362
GlyR model[50]. Certain atoms/substituents are labelled, for comparison with (a) or for their 363
significance in the text. Image in (b) prepared with PyMOL Molecular Graphics System 364
(Schrödinger, LLC). 365
366
Figure 3. Ivermectin binding sites and sensitivity determinants in anionic Cys-loop receptors. (a) 367
The ivermectin binding site in the C. elegans α GluClR [44]. Adjacent (+) and (-) subunits are 368
shown in grey and black, respectively, with M2 domains from the remaining three subunits 369
shown in blue and all ECDs removed for clarity. Yellow lines indicate H-bonds with indicated 370
residues. Orange positions indicate residues that form Van der Waals interactions with the 371
ivermectin molecule. (b) The same site as in (a), but viewed from (+)M2-Ser260 (15'). Red lines 372
indicate proximity of Pro230 (3.4 Å) and Gly281 (4.0 and 4.5 Å) to ivermectin; yellow lines 373
indicate H-bonds with (-)Leu218 and (+)Thr285. (c) Amino acid sequence alignment of TMD 374
portions of Cys-loop receptor subunits of known ivermectin sensitivity. Orange and red (Van der 375
Waals interactions) and yellow (H-bonds) highlighting indicates residues that line the site, as in 376
15
(a) and (b); boxed residues constitute the polar binding site for the ivermectin benzofuran in 377
ivermectin-sensitive receptors (see text and Figure 4). Cel, Caenorhabditis elegans; Hco, 378
Haemonchus contortus; Dme, Drosophila melanogaster; Tur, Tetranychus urticae; Hsa, Homo 379
sapiens; Ssc, Sacrcoptes scabiei; Dme, Dermacentor variabilis; Mmu, Mus musculus. "nM" and 380
"µM" (nanomolar and micromolar) indicate the concentration range in which receptors of that 381
subunit are robustly activated by ivermectin. "mod" (modulatory) means that the receptor is not 382
activated but is potentiated or inhibited by micromolar concentrations of ivermectin. *The T. 383
urticae α GluClR confers ivermectin susceptibility on the organism [4]; it is therefore considered 384
highly ivermectin-sensitive although the recombinant receptor has not been tested. **The 385
recombinant D. variabilis RDL has only been exposed to 100 nM ivermectin, which had no 386
agonist or modulatory effect [56]. Lines joining subunits indicate that these form heteromeric 387
receptors. References for all subunits listed are provided in the main text, except for Cel GluClR 388
α2B [77] and Cel GLC-3 [78]. 389
390
Figure 4. Possible H-bond partners of the ivermectin C5-hydroxyl. The ivermectin binding site in 391
an α1 GlyR homology model [50], viewed from the (+)M3 helix, with the (+)M3 helix removed. 392
Distances (yellow lines) are 8.1, 2.9, 4.0, 3.3 and 10.1 Å, clockwise from M2-12'. 393
394
Figure 5. Comparison of an intrasubunit site in nAChRs and anionic Cys-loop receptors. (a) 395
TMD portions of adjacent (+)α and (-)γ subunits of the Torpedo marmorata ("Tma") nAChR 396
(left; electron microscopy structure [47]) and of adjacent subunits of the C. elegans α GluClR 397
(right, [44]), seen from within the membrane plane. For each receptor, the (+) subunit is shown in 398
light grey, the (-) subunit in dark. nAChR α Leu279 (green) aligns with the GluClR M3-Gly 399
position in most amino acid sequence alignments, but Met282, Ile283 (yellow) and Ile286(red) 400
are structurally more equivalent to GluClR M3-Gly, indicated by the proximity to M1-Pro (red in 401
both receptors). Shown in magenta are several residues that line a space between M1 and M4 of 402
each subunit (these are equivalent in amino acid alignments but were selected simply due to their 403
apparent structural equivalence). Shown in cyan are residues equivalent to those whose 404
substitution in the rat α7 nAChR decreases potentiation by ivermectin [64]; only one of these is 405
16
indicated in the GluClR structure, as it is one of the aligned residues in (b). (b) Amino acid 406
sequence alignment. Colours indicate the residues shown in (a). Circled residues are smaller in 407
the α7 nAChR than in other cationic receptors, providing a possible explanation for selectivity of 408
ivermectin for the α7 nAChR over other cationic receptors. Boxed residues are smaller in the α7 409
nAChR than in anionic Cy-loop receptors, providing a possible explanation for the selectivity of 410
ivermectin for the M1-M4 site in the former. 411
412
Figure 6. Putative ivermectin-insensitive receptor subunits from trematodes. An amino acid 413
sequence alignment of characterised GluClR, GlyR, HisCl and GABAAR (UNC-49B) subunits 414
from vertebrates, insects and nematodes (black) and uncharacterised subunits from trematodes 415
(purple). Sma, Schistosoma mansoni; Sja, Schistosoma japonicum; Sha, Schistosoma 416
haematobium; Csi, Clonorchis sinensis; Sme, Schmidtea mediterranea. Loop B is part of the 417
neurotransmitter binding domain (see Figure 1). A Glu residue at the cyan position is important to 418
the binding of the agonists glycine, histamine and GABA [71]. Although the putative trematode 419
subunits possess several features of the ivermectin binding site, such as the M1-Pro (first red 420
residue) and several polar sidechains (boxed residues), they possess large hydrophobic sidechains 421
at the M3-Gly position (second red residue), indicative of ivermectin insensitivity. Smp_104890 422
and Smp_096480 [70] and Sjp_0079260 [79] were retrieved at GeneDB. Sha_300716 was 423
retrieved from the original publicaton [80]. GAA36399.1 [81] was retrieved from GenBank. 424
mk4.001914 [82] was retrieved from SmedGD. 425
426
17
References 427
1 Crump, A. and Omura, S. (2011) Ivermectin, 'wonder drug' from Japan: the human use 428
perspective. Proc Jpn Acad Ser B Phys Biol Sci 87, 13-28 429
2 Cully, D.F., et al. (1994) Cloning of an avermectin-sensitive glutamate-gated chloride channel 430
from Caenorhabditis elegans. Nature 371, 707-711 431
3 Ghosh, R., et al. (2012) Natural variation in a chloride channel subunit confers avermectin 432
resistance in C. elegans. Science 335, 574-578 433
4 Kwon, D.H., et al. (2010) A point mutation in a glutamate-gated chloride channel confers 434
abamectin resistance in the two-spotted spider mite, Tetranychus urticae Koch. Insect Mol 435
Biol 19, 583-591 436
5 Omura, S. and Crump, A. (2004) The life and times of ivermectin - a success story. Nat Rev 437
Microbiol 2, 984-989 438
6 Campbell, W.C., et al. (1983) Ivermectin: a potent new antiparasitic agent. Science 221, 823-439
828 440
7 Shoop, W.L., et al. (1995) Avermectins and milbemycins against Fasciola hepatica: in vivo 441
drug efficacy and in vitro receptor binding. Int J Parasitol 25, 923-927 442
8 Gryseels, B., et al. (2006) Human schistosomiasis. Lancet 368, 1106-1118 443
9 Fritz, L.C., et al. (1979) Avermectin B1a irreversibly blocks postsynaptic potentials at the 444
lobster neuromuscular junction by reducing muscle membrane resistance. Proc Natl Acad Sci 445
U S A 76, 2062-2066 446
10 Kass, I.S., et al. (1980) Avermectin B1a, a paralyzing anthelmintic that affects interneurons 447
and inhibitory motoneurons in Ascaris. Proc Natl Acad Sci U S A 77, 6211-6215 448
11 Kehoe, J., et al. (2009) Aplysia cys-loop glutamate-gated chloride channels reveal convergent 449
evolution of ligand specificity. J Mol Evol 69, 125-141 450
12 Wolstenholme, A.J. and Rogers, A.T. (2005) Glutamate-gated chloride channels and the 451
mode of action of the avermectin/milbemycin anthelmintics. Parasitology 131 Suppl, S85-95 452
13 Dent, J.A., et al. (2000) The genetics of ivermectin resistance in Caenorhabditis elegans. 453
Proc Natl Acad Sci U S A 97, 2674-2679 454
18
14 Kane, N.S., et al. (2000) Drug-resistant Drosophila indicate glutamate-gated chloride 455
channels are targets for the antiparasitics nodulisporic acid and ivermectin. Proc Natl Acad 456
Sci U S A 97, 13949-13954 457
15 Ludmerer, S.W., et al. (2002) Ivermectin and nodulisporic acid receptors in Drosophila 458
melanogaster contain both gamma-aminobutyric acid-gated Rdl and glutamate-gated GluCl 459
alpha chloride channel subunits. Biochemistry 41, 6548-6560 460
16 Collingridge, G.L., et al. (2009) A nomenclature for ligand-gated ion channels. 461
Neuropharmacology 56, 2-5 462
17 Jones, A.K. and Sattelle, D.B. (2008) The cys-loop ligand-gated ion channel gene 463
superfamily of the nematode, Caenorhabditis elegans. Invert Neurosci 8, 41-47 464
18 McCavera, S., et al. (2009) An ivermectin-sensitive glutamate-gated chloride channel from 465
the parasitic nematode Haemonchus contortus. Molecular pharmacology 75, 1347-1355 466
19 Martin, R.J. and Pennington, A.J. (1989) A patch-clamp study of effects of 467
dihydroavermectin on Ascaris muscle. Br J Pharmacol 98, 747-756 468
20 Adelsberger, H., et al. (2000) Activation of rat recombinant alpha(1)beta(2)gamma(2S) 469
GABA(A) receptor by the insecticide ivermectin. Eur J Pharmacol 394, 163-170 470
21 Shan, Q., et al. (2001) Ivermectin, an unconventional agonist of the glycine receptor chloride 471
channel. J Biol Chem 276, 12556-12564 472
22 Robertson, B. (1989) Actions of anaesthetics and avermectin on GABAA chloride channels in 473
mammalian dorsal root ganglion neurones. Br J Pharmacol 98, 167-176 474
23 Schonrock, B. and Bormann, J. (1993) Activation of Cl- channels by avermectin in rat 475
cultured hippocampal neurons. Naunyn Schmiedebergs Arch Pharmacol 348, 628-632 476
24 Mounsey, K.E., et al. (2007) Molecular characterisation of a pH-gated chloride channel from 477
Sarcoptes scabiei. Invert Neurosci 7, 149-156 478
25 Schnizler, K., et al. (2005) A novel chloride channel in Drosophila melanogaster is inhibited 479
by protons. J Biol Chem 280, 16254-16262 480
26 Zheng, Y., et al. (2002) Identification of two novel Drosophila melanogaster histamine-gated 481
chloride channel subunits expressed in the eye. J Biol Chem 277, 2000-2005 482
27 Khakh, B.S., et al. (1999) Allosteric control of gating and kinetics at P2X(4) receptor 483
channels. J Neurosci 19, 7289-7299 484
19
28 Costa, J.L. and Diazgranados, J.A. (1994) Ivermectin for spasticity in spinal-cord injury. 485
Lancet 343, 739 486
29 Al-Azzam, S.I., et al. (2007) Comparison of the pharmacokinetics of moxidectin and 487
ivermectin after oral administration to beagle dogs. Biopharm Drug Dispos 28, 431-438 488
30 Geyer, J., et al. (2009) Brain penetration of ivermectin and selamectin in mdr1a,b P-489
glycoprotein- and bcrp- deficient knockout mice. J Vet Pharmacol Ther 32, 87-96 490
31 Njoo, F.L., et al. (1995) Ivermectin detection in serum of onchocerciasis patients: relationship 491
to adverse reactions. Am J Trop Med Hyg 52, 94-97 492
32 Lerchner, W., et al. (2007) Reversible silencing of neuronal excitability in behaving mice by 493
a genetically targeted, ivermectin-gated Cl- channel. Neuron 54, 35-49 494
33 Schinkel, A.H., et al. (1994) Disruption of the mouse mdr1a P-glycoprotein gene leads to a 495
deficiency in the blood-brain barrier and to increased sensitivity to drugs. Cell 77, 491-502 496
34 Xu, M., et al. (1998) Ivermectin resistance in nematodes may be caused by alteration of P-497
glycoprotein homolog. Molecular and biochemical parasitology 91, 327-335 498
35 Miller, P.S. and Smart, T.G. (2010) Binding, activation and modulation of Cys-loop 499
receptors. Trends Pharmacol Sci 31, 161-174 500
36 Shoop, W.L., et al. (1995) Structure and activity of avermectins and milbemycins in animal 501
health. Vet Parasitol 59, 139-156 502
37 Michael, B., et al. (2001) Comparison of ivermectin, doramectin, selamectin, and eleven 503
intermediates in a nematode larval development assay. J Parasitol 87, 692-696 504
38 Fisher, M.H. (1990) Advances in avermectin research. Pure and Appl. Chem. 62, 1231-1240 505
39 Payne, D.T. and Soderlund, D.M. (1993) Actions of avermectin analogues on gamma-506
aminobutyric acid (GABA)-sensitive and GABA-insensitive chloride channels in mouse 507
brain. Pesticide Biochemistry and Physiology 47, 178-184 508
40 Arena, J.P., et al. (1995) The mechanism of action of avermectins in Caenorhabditis elegans: 509
correlation between activation of glutamate-sensitive chloride current, membrane binding, 510
and biological activity. J Parasitol 81, 286-294 511
41 Blizzard, T.A., et al. (1992) Synthesis and biological activity of 13-epi-avermectins: potent 512
anthelmintic agents with an increased margin of safety. J Med Chem 35, 3873-3878 513
20
42 Mrozik, H., et al. (1989) Syntheses and biological activities of 13-substituted avermectin 514
aglycons. J Med Chem 32, 375-381 515
43 Lespine, A., et al. (2007) Interaction of macrocyclic lactones with P-glycoprotein: structure-516
affinity relationship. Eur J Pharm Sci 30, 84-94 517
44 Hibbs, R.E. and Gouaux, E. (2011) Principles of activation and permeation in an anion-518
selective Cys-loop receptor. Nature 474, 54-60 519
45 Bocquet, N., et al. (2009) X-ray structure of a pentameric ligand-gated ion channel in an 520
apparently open conformation. Nature 457, 111-114 521
46 Hilf, R.J. and Dutzler, R. (2008) X-ray structure of a prokaryotic pentameric ligand-gated ion 522
channel. Nature 452, 375-379 523
47 Unwin, N. (2005) Refined structure of the nicotinic acetylcholine receptor at 4A resolution. J 524
Mol Biol 346, 967-989 525
48 Lynagh, T. and Lynch, J.W. (2010) A glycine residue essential for high ivermectin sensitivity 526
in Cys-loop ion channel receptors. Int J Parasitol 40, 1477-1481 527
49 Yamaguchi, M., et al. (2012) Amino acid residues of both the extracellular and 528
transmembrane domains influence binding of the antiparasitic agent milbemycin to 529
Haemonchus contortus AVR-14B glutamate-gated chloride channels. Biochem Biophys Res 530
Commun 16, 562-566 531
50 Lynagh, T., et al. (2011) Molecular determinants of ivermectin sensitivity at the glycine 532
receptor chloride channel. J Biol Chem 286, 43913-43924 533
51 Njue, A.I., et al. (2004) Mutations in the extracellular domains of glutamate-gated chloride 534
channel alpha3 and beta subunits from ivermectin-resistant Cooperia oncophora affect 535
agonist sensitivity. J Neurochem 89, 1137-1147 536
52 Grudzinska, J., et al. (2005) The [beta] Subunit Determines the Ligand Binding Properties of 537
Synaptic Glycine Receptors. Neuron 45, 727-739 538
53 Forrester, S.G., et al. (2003) Haemonchus contortus: HcGluCla expressed in Xenopus oocytes 539
forms a glutamate-gated ion channel that is activated by ibotenate and the antiparasitic drug 540
ivermectin. Molecular and biochemical parasitology 129, 115-121 541
21
54 Brown, D.D., et al. (2011) Pharmacological characterization of the Haemonchus contortus 542
GABA-gated chloride channel, Hco-UNC-49: Modulation by macrocyclic lactone 543
anthelmintics and a receptor for piperazine. Vet Parasitol 185, 201-209 544
55 Siddiqui, S.Z., et al. (2010) An UNC-49 GABA receptor subunit from the parasitic nematode 545
Haemonchus contortus is associated with enhanced GABA sensitivity in nematode 546
heteromeric channels. J Neurochem 113, 1113-1122 547
56 Zheng, Y., et al. (2003) RdlDv, a novel GABA-gated chloride channel gene from the 548
American dog tick Dermacentor variabilis. Insect Biochem Mol Biol 33, 595-599 549
57 Ranganathan, R., et al. (2000) MOD-1 is a serotonin-gated chloride channel that modulates 550
locomotory behaviour in C. elegans. Nature 408, 470-475 551
58 Hedblom, E. and Kirkness, E.F. (1997) A novel class of GABAA receptor subunit in tissues 552
of the reproductive system. J Biol Chem 272, 15346-15350 553
59 Etter, A., et al. (1996) An amino acid substitution in the pore region of a glutamate-gated 554
chloride channel enables the coupling of ligand binding to channel gating. J Biol Chem 271, 555
16035-16039 556
60 Cheeseman, C.L., et al. (2001) High-affinity ivermectin binding to recombinant subunits of 557
the Haemonchus contortus glutamate-gated chloride channel. Molecular and biochemical 558
parasitology 114, 161-168 559
61 Feng, X.P., et al. (2002) Study of the nematode putative GABA type-A receptor subunits: 560
evidence for modulation by ivermectin. J Neurochem 83, 870-878 561
62 Huang, J. and Casida, J.E. (1997) Avermectin B1a binds to high- and low-affinity sites with 562
dual effects on the gamma-aminobutyric acid-gated chloride channel of cultured cerebellar 563
granule neurons. J Pharmacol Exp Ther 281, 261-266 564
63 Jansen, M. and Akabas, M.H. (2006) State-dependent cross-linking of the M2 and M3 565
segments: functional basis for the alignment of GABAA and acetylcholine receptor M3 566
segments. J Neurosci 26, 4492-4499 567
64 Collins, T. and Millar, N.S. (2010) Nicotinic acetylcholine receptor transmembrane mutations 568
convert ivermectin from a positive to a negative allosteric modulator. Molecular 569
pharmacology 78, 198-204 570
22
65 Raymond, V., et al. (2000) Anthelmintic actions on homomer-forming nicotinic acetylcholine 571
receptor subunits: chicken alpha7 and ACR-16 from the nematode Caenorhabditis elegans. 572
Neuroscience 101, 785-791 573
66 Sattelle, D.B., et al. (2009) Comparative pharmacology and computational modelling yield 574
insights into allosteric modulation of human alpha7 nicotinic acetylcholine receptors. 575
Biochem Pharmacol 78, 836-843 576
67 Krause, R.M., et al. (1998) Ivermectin: a positive allosteric effector of the alpha7 neuronal 577
nicotinic acetylcholine receptor. Molecular pharmacology 53, 283-294 578
68 Blackhall, W.J., et al. (2003) Selection at a gamma-aminobutyric acid receptor gene in 579
Haemonchus contortus resistant to avermectins/milbemycins. Molecular and biochemical 580
parasitology 131, 137-145 581
69 El-Abdellati, A., et al. (2011) Altered avr-14B gene transcription patterns in ivermectin-582
resistant isolates of the cattle parasites, Cooperia oncophora and Ostertagia ostertagi. Int J 583
Parasitol 41, 951-957 584
70 Berriman, M., et al. (2009) The genome of the blood fluke Schistosoma mansoni. Nature 460, 585
352-358 586
71 Cromer, B.A., et al. (2002) Anxiety over GABA(A) receptor structure relieved by AChBP. 587
Trends Biochem Sci 27, 280-287 588
72 Colbourne, J.K., et al. (2011) The ecoresponsive genome of Daphnia pulex. Science 331, 555-589
561 590
73 Holt, R.A., et al. (2002) The genome sequence of the malaria mosquito Anopheles gambiae. 591
Science 298, 129-149 592
74 Jex, A.R., et al. (2011) Ascaris suum draft genome. Nature 479, 529-533 593
75 McCavera, S., et al. (2007) Nematode ligand-gated chloride channels: an appraisal of their 594
involvement in macrocyclic lactone resistance and prospects for developing molecular 595
markers. Parasitology 134, 1111-1121 596
76 Osei-Atweneboana, M.Y., et al. (2011) Phenotypic evidence of emerging ivermectin 597
resistance in Onchocerca volvulus. PLoS neglected tropical diseases 5, e998 598
23
77 Vassilatis, D.K., et al. (1997) Genetic and biochemical evidence for a novel avermectin-599
sensitive chloride channel in Caenorhabditis elegans. Isolation and characterization. J Biol 600
Chem 272, 33167-33174 601
78 Horoszok, L., et al. (2001) GLC-3: a novel fipronil and BIDN-sensitive, but picrotoxinin-602
insensitive, L-glutamate-gated chloride channel subunit from Caenorhabditis elegans. Br J 603
Pharmacol 132, 1247-1254 604
79 Consortium, S.j.G.S.a.F.A. (2009) The Schistosoma japonicum genome reveals features of 605
host-parasite interplay. Nature 460, 345-351 606
80 Young, N.D., et al. (2012) Whole-genome sequence of Schistosoma haematobium. Nat Genet 607
44, 221-225 608
81 Wang, X., et al. (2011) The draft genome of the carcinogenic human liver fluke Clonorchis 609
sinensis. Genome biology 12, R107 610
82 Robb, S.M., et al. (2008) SmedGD: the Schmidtea mediterranea genome database. Nucleic 611
acids research 36, D599-606 612
613
614
M3M1 M215'
M1-Pro
M3-Gly
12'19'
ASYAYESFGYESLSHESYCYGSYGYGSYGYGSYGYGSYGYGSYGYGSYGY
DVWIGACMTFIFDIWMAVCLLFVFDVWMSSCSVFVFDVFLNFCFVMVFDIWLFVCLAFVVDIWTIACITFNSDIWTIACITFNSDIWTIACITFNSDVWTIACIAFNSDIWIIACIIFVI
CelHsaDmeHcoSmaSmaSjaShaCloSme
GluClR aGlyR aHisCl1
UNC-49BSmp_104890Smp_096480Sjp_0079260Sha_300716
GAA36399.1mk4.001914
LLQLYIPSCMLIQMYIPSLLLFHTYIPSALSMNIVIPSMLLIYAYLPSLLLASTYIPNILLASTYIPNILLASTYIPNILLASTYIPNTLLASTYIPNVL
LTMTAQSAGINLTMTTQSSGSRLTLATQNTQSQLTMTTLITTTNLALITQSAAILLGVITQSTSYALGVITQSTSYALGVITQSTSYALGVITQTTSYALGILTQATGMS
Loop B M1 M2 M3