Receptor variability-driven evolution of snake toxins

2018·2.2 MB·

by  Ji, Xian-Hong

ZOOLOGICAL RESEARCH
Receptor variability-driven evolution of snake toxins
Xian-HongJi1,2,#,Shang-FeiZhang1,2,#,BinGao1,Shun-YiZhu1,*
1GroupofPeptideBiologyandEvolution,StateKeyLaboratoryofIntegratedManagementofPestInsectsandRodents,InstituteofZoology,
ChineseAcademyofSciences,Beijing100101,China
2UniversityofChineseAcademyofSciences,Beijing100049,China
ABSTRACT threefingers(Fryetal.,2003).
Three-finger toxins (TFTs) are well-recognized non- Thelong-termevolutionaryprocessthathasresultedinthe
high diversity of TFTs conforms to the birth-and-death model
enzymaticvenomproteinsfoundinsnakes. However,
ofmultigenefamilyevolution(Neietal.,1997). Thismodelof
although TFTs exhibit accelerated evolution, the
evolutionproducesaseriesoftoxins,allowingsnakestoadapt
drivers of this evolution remain poorly understood. to a variety of prey and predators (Nei et al., 1997). Snakes
The structural complexes between long-chain employ their venom as a weapon to disable prey (primary
α-neurotoxins,asubfamilyofTFTs,andtheirnicotinic function)andasadefensivetoolagainstpredators(secondary
function) (Kang et al., 2011). Snake toxins and prey are
acetylcholinereceptortargetshavebeendetermined
likelyinvolvedinaco-evolutionaryarmsrace,wherebyevolving
in previous research, providing an opportunity to
toxin resistance in prey species and novel toxin evolution in
address such questions. In the current study, we
snake species exert mutual selective effects (Arbuckle et al.,
observed several previously identified positively 2017; Calvete, 2017; Jansa & Voss, 2011; Voss & Jansa,
selected sites (PSSs) and the highly variable 2012). However, despite evidence of accelerated evolution
C-terminal loop of these toxins at the toxin/receptor in the TFT family (Sunagar et al., 2013), the cause of this
evolutionremainsvague.
interface. Of interest, analysis of the molecular
The TFT family members can target various receptors and
adaptation of the toxin-recognition regions in the
ion channels with high affinity and specificity (Doley et al.,
corresponding receptors provided no statistical
2009; Kini, 2011). The α-neurotoxins (α-NTXs) of TFTs
evidence for positive selection. However, these can interact with nicotinic acetylcholine receptors (nAChRs),
regionsaccumulatedabundantaminoacidvariations inhibiting postsynaptic membrane ion flow and leading to
inthereceptorsfromthepreyofsnakes,suggesting flaccidparalysis(Barberetal.,2013;Chiappinellietal.,1996).
Based on the chain length and number of disulphide bridges,
that accelerated substitution of TFTs could be a
α-NTXs are usually divided into short or long α-NTXs. Short
consequenceofadaptationtothesevariations. Tothe
α-NTXscontain60–62aminoacidresidueswithfourdisulfide
bestofourknowledge,thisatypicalevolution,initially bonds,whereaslongα-NTXscontain66–75residuesandfive
discoveredinscorpions,isreportedinsnaketoxinsfor disulfidebonds(Dajas-Bailadoretal.,1998).
thefirsttimeandmaybeapplicablefortheevolution nAChRs are pentameric transmembrane proteins of
ligand-gated ion channels and are formed by different
oftoxinsfromothervenomousanimals.
Keywords:Three-fingertoxins;Nicotinicacetylcholine
receptor;Driver
Received: 17March2018; Accepted: 13July2018; Online: 25July
2018
INTRODUCTION
Foundation items: This work was supported by the National Natural
Three-finger toxins (TFTs) are among the most abundantly Science Foundation of China (Grant No. 31570773) and State Key
secretedandeffectivecomponentsofsnakevenom. Theyare
Laboratory of Integrated Management of Pest Insects and Rodents
encoded by a large multigene family and show a diversity of
(ChineseIPM1707)
functionalactivities(Fryetal.,2003; Kini,2011). Membersin
thisfamilycontain57–82residuesandfourconserveddisulfide #Authorscontributedequallytothiswork
bridges (Endo & Tamiya, 1987), and are identified by three *Correspondingauthor,E-mail:[email protected]
loopsextendingfromacentralcoreresemblingtheirnamesake DOI:10.24272/j.issn.2095-8137.2018.063
SciencePress ZoologicalResearch39(6):431–436,2018 431
combinationsoffivesubunits,thatis,α,β,γ,δandε(Corringer toxins were used to create sequence logos to identify the
etal.,2000;Karlin,2002).Eachsubunitiscomposedofalarge conservation of each position (Supplementary Figure S1).
N-terminal extracellular domain, which serves as the major Each logo comprises stacks of letters, one stack for each
binding site for the toxins, followed by four transmembrane position in the sequence (Crooks et al., 2004). The overall
helicesandasmallC-terminalextracellulardomain(Dellisanti heightofeachstackshowsthesequenceconservationofthat
etal.,2007;Karlin,2002). nAChRscanbefurtherdividedinto position, whereas the height of the symbols within the stack
muscular or neuronal types (Corringer et al., 2000; Wang et indicates the comparative frequency of the amino acid at the
al.,2002). α-NTXscanpotentlyantagonizetheα1subunitof position (Crooks et al., 2004). Generally, sequence logos
heteropentameric muscle nAChRs ((α1) β1γδ in fetal muscle provide a richer and more precise description of conserved
2
and (α1) β1εδ in adult muscle) and the α7, α8, α9 or α10 andvariableregionswithinsequences. WeusedtheConSurf
2
subunits of homopentameric neuronal nAChRs (Karlin, 2002; program (http://consurf.tau.ac.il/) under default parameters to
Sineetal.,2013).Crystalstructuresinthecomplexesbetween calculateconservationscoresoftheaminoacidsequencesof
long α-NTXs and related receptors have been identified and therelatedreceptors. ConSurfnotonlydependsonsequence
offeropportunitiestoexplorethemolecularmechanismdriving alignmentsbutalsoonphylogenetictreestoidentifyconserved
theacceleratedevolutionofthesetoxins(Bourneetal.,2005; and variable regions (Landau et al., 2005). A Bayesian tree
Dellisantietal.,2007;Huangetal.,2013). was constructed using the corresponding alignments with the
Inthiswork, wefoundseveralsitespreviouslyidentifiedas JTTevolutionarysubstitutionmodel. Oneoftheadvantagesof
positively selected sites (PSSs) as well as the highly variable ConSurfcomparedwithothermethodsisthatthecomputation
C-terminalloopofthetoxinstobelocatedatthetoxin/receptor of the evolutionary rate is more precise when employing the
binding interface (Bourne et al., 2005; Sunagar et al., 2013). empiricalBayesianormaximum-likelihoodmethods.
Structural and evolutionary analyses of the toxin-recognition
Positiveselectionanalysis
region from the receptors (α1, α7, α9 and α10 subunits from
Excess nonsynonymous substitutions compared with
nAChRs)uncoveredanatypicalevolutionbetweensnakesand
synonymous substitutions (ω=d /d >1) is an important
their prey, in which the amino acid diversity of the nAChR N S
sign of positive selection at the molecular level (Yang, 1998;
toxin-bindingregionsappearedtodrivetheadaptiveevolution
Zhu & Gao, 2016). To perform such analysis, we compared
of the TFT family. This study showed good agreement with
two pairs of site models (M1a (neutral)/M2a (selection) and
ourpreviousresearchonscorpiontoxinsandsodiumchannels
M7 (beta)/M8 (beta & ω)) to measure the selective pressure
fromtheircompetitors(Zhangetal.,2015). Furthermore, this
of the receptors to which the long α-NTXs bind (α1, α7, α9
paper provides a broader vision into the evolution between
and α10 subunits from nAChRs). Model M2a and M8 add a
venomousanimalsandtheirprey/predators.
site class to M1a and M7, respectively, with the free ω ratio
calculatedfromthedataandusedtodeterminetheprobability
MATERIALSANDMETHODS
of positive selection (Anisimova et al., 2001; Anisimova et al.,
Sequenceanalysis 2003; Wong et al., 2004; Yang & Swanson, 2002). As M1a
ClustalX (http://www.clustal.org) was used to align all and M7 are nested within their respective alternative models
nucleotide and amino acid sequences. A total of 130 long (M2aandM8)andhavetwomoreparameters, χ2 distribution
α-NTX sequences were aligned and used for sequence logo can be used for the likelihood ratio test to compare the fit
analysis with the WebLogo program (Supplementary Figure of the two competing models. We used the Bayes Empirical
S1). For the receptors, 76 sequences of muscle-type α1 Bayes method to calculate the posterior possibility that each
nAChRs(threefromReptilia,threefromAmphibian,fourfrom codon is from the site class of positive selection. The Bayes
Aves, 31 from small mammals, and 35 from fishes), 59 Empirical Bayes method is an improvement of the previous
neuronal-type α7 nAChRs (one from Gastropoda, two from Naïve Empirical Bayes method and accounts for sampling
Arthropoda, three from Reptilia, three from Amphibian, three errors in the maximum-likelihood estimates of parameters in
from Aves, 18 from small mammals, and 29 from fishes), themodel(Nielsen&Yang,1998). Siteswithahighpossibility
68 neuronal-type α9 nAChRs (three from Reptilia, three from (≥95%) of coming from the class with ω>1 are likely under
Amphibian, four from Aves, 30 from small mammals, and 28 positiveselectionandcanbeanalyzedfurther(Yang,1998).
from fishes), and 52 neuronal-type α10 nAChRs (two from
RESULTS
Reptilia, two from Amphibian, four from Aves, 24 from small
mammals,and20fromfishes)(SupplementaryFiguresS2–S5) PSSs of toxins locating on the toxin-receptor complex
were also aligned. For positive selection analyses, aligned interface
nucleotidesequencesofthereceptorswereusedtoconstruct TheN-terminalextracellulardomainsofnAChRs,whichconsist
neighbor-joiningtrees. of a 10 stranded β-sandwich and an N-terminal α-helix, act
as the major binding sites for long α-NTXs (Changeux et al.,
WebLogoandConSurfanalyses
1970). Thethreeregionsoflongα-NTXscomprisingfingersI
WebLogo can be used to generate sequence logos, which andIIandtheC-terminusareinvolvedininteractionswiththe
are graphical representations of the patterns within multiple receptors,withfingerIIbeing the main stabilizing interaction
sequence alignments. The alignments of the aforementioned
432 www.zoores.ac.cn
center (Bourne et al., 2005). The α211 structure (mouse et al. (2013) on the complex structure of α-bungarotoxin and
nAChR α1 subunit (PDB entry 2qc1)) can be seen as a its receptor (Figure 1A). The toxin-receptor complex model
representative of the N-terminal extracellular domain of the showed that some PSSs of the long α-NTXs were located
nAChR subunit (Dellisanti et al., 2007), in which loops β4-β5 on the toxin binding surface with the receptors (Bourne et al.,
(loop A), β7-β8 (loop B) and β9-β10 (loop C) serve as the 2005; Dellisanti et al., 2007; Huang et al., 2013). Almost
principal ligand-binding interfaces (Brejc et al., 2001; Unwin, all PSSs in finger II (Ala31, Phe32, Ser34, Ser35 and
2005)andloopCisthemostimportantregionforhighaffinity Val39) of α-bungarotoxin are involved in the interactions with
with long α-NTXs, as revealed by site-directed mutations the receptors (Dellisanti et al., 2007; Dimitropoulos et al.,
(Fruchart-Gaillardetal.,2002;Levandoskietal.,1999).LoopC 2011; Huang et al., 2013). The same situation appears in
ofα211isenvelopedbyfingersIandIIandtheC-terminalloop α-cobratoxin(longα-NTX)sincetheirfastevolvedsites(Ala28,
of the toxin, with finger II inserted into the ligand-binding site Phe29, Ser31, Ile32 and Arg36) in finger II also overlap with
wrapped by loops A, B and C of α211 (Figure 1A) (Dellisanti the toxin binding sites of α-cobratoxins (Bourne et al., 2005;
et al., 2007). In addiTtihoe nev,olufitinongarey rdriIverisof ssnaakne dtoxwinisc hed by loop C, Dimitropoulosetal.,2011).
whereasfingerIIIweaklycontributestoα211binding(Dellisanti The C-terminal loop of the long α-NTXs also contributes
etal.,2007). to the interaction with related receptors. Multiple sequence
FIGURE CAPTIONS
alignmentsoflongα-NTXs,generatedbyClustalX,wereused
to create sequence logos (Figure 1B, Supplementary Figure
S1) (Crooks et al., 2004). The C-terminal loop of the long
α-NTXs, indicated in the red box in Figure 1B, was highly
variablebasedontheWebLogoanalyses,andthewholeC-tail
involves extensive insertions and deletions. Thus, this loop
mightundergopositiveselectiondespitethetechnicaldifficulty
indetectingPSSsfromindel-containingsequences.
Noevidenceforpositiveselectioninthetoxin-recognition
regionsofreceptors
Snakes employ their venom to immobilize various prey,
includingsnails,insects,fishes,toads,lizards,chickens,small
mammals and even other snakes (Kang et al., 2011). The
B
maximum-likelihoodmodelsofcodonsubstitutionswereused
to identify selective pressure in the toxin-recognition regions
Figure1α-bungarotoxininteractswithα211andthesequence
Figure 1 α-bungarotoxin interacts with α211 and the sequence logos of long α-NTXs of the receptors from the prey of snakes. However, no
logosoflongα-NTXs positiveselectionsignalsweredetectedinthereceptorsoflong
A: Interaction between α-bungarotoxin (long α-NTXs) and α211 (mouse nAChR α1 subunit)
A: Interaction between α-bungarotoxin (long α-NTXs) and α211. α-NTXs(α1,α7,α9andα10subunits)(Table1,Supplementary
(αP-DbuBn egnatrryo t2oqxcin1). isα-bsuhnogwarnotoinxinw ihs esahotwann idn wαh2e1a1t ainnd αlig21h1t ibnl ulieg.ht bTluhee. TPheS PSSsSs of theTables S1–S4). The maximum-likelihood estimates under M0
showed that the average ω ratios for all receptor sequence
loofntgh eα-lNonTgXsα -aNreT iXn spainrke ainndp tinhek .CT-theermthinraele lolooopp iss coifrcαle2d1.1 T,hme athjorreet olxoionp-sb ionfd αin2g11, major
pairs ranged from 0.02 to 0.07. M2a and M8 detected no
regions,arehighlightedinblue. B:Sequencelogosoflongα-NTXs. The
toxin-binding regions, are highlighted in blue. B: Sequence logos of long α-NTXs. C-terminalevolution-accelerating sites. The ωs under M8 of the α1 and
C-terminalloopofthelongα-NTXsinvolvedinreceptorbindingisindicated
loop of the long α-NTXs is indicated in a red box. α10 subunits of nAChRs from snake prey equaled 1 (Table
inaredbox. 1). Under the M8 model, the α7 and α9 subunits of nAChRs
showed ωs>1 (Table 1), but their proportions (p1) equaled 0,
WemappedthePSSsoflongα-NTXsidentifiedbySunagar provingthatnoPSSsexisted(SupplementaryTablesS1–S4).
Table1Parameterestimatesandlikelihoodratiostatistics(2∆l)fordifferentsubunittypesofnAChRs
Type l(M1a) l(M2a) ω2(M2a) 2∆l l(M7) l(M8) ωs(M8) 2∆l
α1subunit –12874.0 –12874.0 1.00 0 –12617.4 –12607.4 1.00 20
α7subunit –13163.9 –13163.9 1.00 0 –12785.7 –12785.7 34.66 0
α9subunit –11856.3 –11856.3 1.00 0 –11626.9 –11626.9 1.42 0
α10subunit 12 –9954.2 –9954.2 1.00 0 –9746.8 –9746.8 1.00 0
listheloglikelihood;2∆lisbetweennullmodelsandtheiralternativemodels: M1a/M2aandM7/M8. TwoωvaluesinM2aandM8are
boldfaced.
ZoologicalResearch39(6):431–436,2018 433
Snaketoxinsbindtovariableregionsoftheirreceptors We analyzed the interaction pairs in the α-bungarotoxin
AlthoughPSSsweredetectedinthetoxins,nonewerefoundin and α1 nAChR subunit complex, which were Ile11-Phe189/
thetoxin-recognitionregionsoftheinvolvedreceptors. Based Pro194, Val31-Tyr93/Asp99/Phe100, Phe32-Tyr93/Phe100/
onthecomplexmodelsbetweenthelongα-NTXsandrelated Trp149/Tyr190, Val39-Val188/Tyr190, His68-Pro194, Pro69-
receptors, we further analyzed the evolutionary conservation Ser191, Lys70-Pro194 and Gln71-Pro194, with the toxin sites
oftheN-terminalextracellulardomainregionsofthenAChRα1 correspondingtothePSSsinfingersIandIIandtheC-terminal
andα7subunitsfromsnakepreyusingConSurf(Dellisantiet loop (Dellisanti et al., 2007; Dimitropoulos et al., 2011). The
al.,2007;Huangetal.,2013)(Figure2). Ourresultsindicated loopA(Tyr93, Asp99andPhe100), loopB(Trp149)andloop
that loop C demonstrated the greatest variation among the C (Val188, Phe189, Tyr190, Ser191, Pro194 and Pro197)
threereceptorloops(Figure2). were involved in the interaction with the PSSs of the toxin
The evolutionary driver of snake toxins
bindingsurface.AlthoughTyr93inloopAandTrp149inloopB
wereconserved,Asp99andPhe100inloopAexhibitedsome
variability (99: Asp/Asn and 100: Phe/Tyr). In contrast, the
binding sites in loop C were more variable (188: Val/Arg/Lys,
189: Phe/Tyr/Thr, 191: Ala/Ser/Thr/Pro and 194: Pro/Gln)
(Figure2A).
For the α-bungarotoxin and α7 nAChR subunit complex,
the interaction pairs include Ile11-Phe183/Lys188, Phe32-
Tyr91/Trp145/Arg182/Tyr184, Val39-Arg182/Tyr184, His68-
Lys188, Pro69-Glu185, Lys70-Cys186/Lys188 and Gln71-
Lys188, with the sites of the toxins also corresponding to
the PSSs in fingers I and II and the C-terminal loop (Huang
et al., 2013). Compared with the conserved residues in loop
A and loop B (Tyr91 and Trp145), the binding sites in loop
C were diverse (182: Lys/Arg/Leu/Ser/Asn, 183: Phe/Tyr/Ile,
185: Glu/Asp/Asn/Gly and 188: Lys/Asp/Glu/Pro/Gln) (Figure
2B). The additional PSSs (Val31, Ser34 and Ser35) in finger
II contacted residues of the complementary subunit, and
includedSer32,Ser34,Leu36,Trp53,Gln55,Gln114,Leu116

and Asp160 (Huang et al., 2013). These residues in the
complementary subunit may provide the driving force for the
additionalsitesinfingerII.
Taken together, our results suggest that the variable
toxin-recognition region in the receptors might drive the
acceleratedevolutionofthetoxin-bindingresidues.
DISCUSSION
BymappingthePSSsofthelongα-NTXsonthetoxin-receptor
complex, we found that several of them are located on the
toxin-binding surface of the receptor (Figure 1A). In addition
to these PSSs, high sequence diversity was also observed
in the C-terminal loop of the long α-NTXs. Thus, given its
importanceintheinteractionwithreceptors,wesurmisedthat
itcouldbeanacceleratedsubstitutedregionforadaptationto
receptor variability (Figure 1B). Several PSSs were detected
infingerIIIofthelongα-NTXs, althoughfingerIIIcontributed

little to the binding. This may be due to its high flexibility
Figure2ConSurfresultsofα-bungarotoxinwiththereceptors
in the complex. Other mechanisms may be involved in the
Figure 2 ConSurf results of α-bungarotoxin with the receptors
ConSurfresultsofα-bungarotoxinwithα1(PDBentry2qc1)andα7(PDB accelerated substitutions of this finger, which requires further
entry4hqp)nAChRsareshowninAandB,respectively. ConSurfprovides investigation.
ConSurf results of α-bungarotoxin with α1 (PDB entry 2qc1) and α7 (PDB entry 4hqp)
evolutionaryconservationscoresfornAChRs. PSSsoffingerIandfingerII We further observed the amino acid variability of the
andtheresiduesoftheC-terminalloopinα-bungarotoxinarehighlightedin principaltoxin-recognitionregions(mainlyinloopC)inrelated
nAChRs are shown in A and B, respectively. ConSurfn pArCohvRidseusb uenvitosl(uFtiigounrear2y) .cCoonmsepravreadtiownith loopAandloopB,
red(PSSsoffingerIIarepartlyhidden).Toxin-bindingregions(loopA,loop
loopCexhibitedthegreatestvariationduetoitspredominant
BandloopC)oftheα1andα7nAChRsarespherized.Bindingsitesvariable
scores for nAChRs. PSSs of finger I and finger II and throel ereinsitdhueetso xoinf- rtehcee pCto-rteinrmteriancatlio lnoso.pT hiuns ,weconcludedthat
inloopCareindicated. the evolutionary variability of the toxin-recognition regions of
α-bungarotoxin are highlighted in red (PSSs of finger II are partly hidden). Toxin-binding
434rewgwiown.zso (olroeos.pac A.cn, loop B and loop C) of the α1 and α7 nAChRs are spherized. Binding sites
variable in loop C are indicated.

13
thereceptorsisapossibledriverfortheacceleratedevolution resistanceinanimals.Toxicon,140:118–131.
ofthetoxins. BarberCM,IsbisterGK,HodgsonWC.2013.Alphaneurotoxins.Toxicon,66:
Short-chain α-NTXs can bind to nAChRs with high affinity 47–58.
(Trémeauetal.,1995). LoopIIofshort-chainα-NTXspushes
Bourne Y, Talley TT, Hansen SB, Taylor P, Marchot P. 2005. Crystal
into the ligand-binding pocket of nAChRs, whereas the tips
structureofaCbtx-AChBPcomplexrevealsessentialinteractionsbetween
of loops I and III contact nAChRs only in a ‘surface-touch’
snakealpha-neurotoxinsandnicotinicreceptors.TheEMBOJournal,24(8):
way (Mordvintsev et al., 2005). Previous study on Torpedo
1512–1522.
californica showed that the C-loop is vital for the binding of
BrejcK,VanDijkWJ,KlaassenRV,SchuurmansM,VanDerOostJ,Smit
shortα-NTXstonAChRs(Bourneetal.,2005;Mordvintsevet
al.,2005).However,astherearenospecificcoordinatefilesof AB,SixmaTK.2001.CrystalstructureofanACh-bindingproteinrevealsthe
complexesbetweenshortα-NTXsandtheirtargets,hindering ligand-bindingdomainofnicotinicreceptors.Nature,411(6835):269–276.
furtherstudy. CalveteJJ.2017.Venomics:Integrativevenomproteomicsandbeyond.The
Our previous study on the evolution of scorpion toxins and BiochemicalJournal,474(5):611–634.
voltage-gated sodium (Nav) channels indicated that variability Changeux JP, Kasai M, Lee CY. 1970. Use of a snake venom toxin to
of the toxin-recognition regions in the Nav channels from characterize thecholinergic receptor protein. Proceedings of theNational
scorpion predators and prey is a putative driver of the
AcademyofSciencesoftheUnitedStatesofAmerica,67(3):1241–1247.
accelerated evolution of the functional regions of scorpion
ChiappinelliVA,WeaverWR,MclaneKE,Conti-FineBM,FiordalisiJJ,Grant
toxinsfollowinggeneduplication(Zhangetal.,2015).Similarly,
GA. 1996. Binding of native κ-neurotoxins and site-directed mutants to
PSSs exist in the binding interface of the long α-NTXs,
nicotinicacetylcholinereceptors.Toxicon,34(11-12):1243–1256.
though only amino acid variability was detected in the
Corringer PJ, Le Novère N, Changeux JP. 2000. Nicotinic receptors at
principal toxin-recognition regions of the related nAChR
subunits, suggesting that amino-acid substitutions on the theaminoacidlevel.AnnualReviewofPharmacologyandToxicology,40:
toxin-recognition surface in related nAChR subunits could 431–458.
provide a force driving the accelerated evolution of the toxins. Crooks GE, Hon G, Chandonia JM, Brenner SE. 2004. WebLogo: A
Thus, the atypical co-evolutionary manner between snake sequencelogogenerator.GenomeResearch,14(6):1188–1190.
toxins and their receptors is similar to our previous research Dajas-Bailador F, Costa G, Dajas F, Emmett S. 1998. Effects of
on scorpion toxins and their targets (Zhang et al., 2015). We alpha-erabutoxin, alpha-bungarotoxin, alpha-cobratoxin and fasciculin on
supposed that accelerated evolution of the receptor-bound the nicotine-evoked release of dopamine in the rat striatum in vivo.
regions of the snake toxins is a consequence of adaptation
NeurochemistryInternational,33(4):307–312.
to variable receptors of their prey. From the viewpoint of
DellisantiCD,YaoY,StroudJC,WangZZ,ChenL.2007.Crystalstructure
a co-evolutionary arms race between predators and prey
oftheextracellulardomainofnAChRalpha1boundtoalpha-bungarotoxinat
(Arbuckle et al., 2017; Calvete, 2017; Jansa & Voss, 2011;
1.94Aresolution.NatureNeuroscience,10(8):953–962.
Voss&Jansa,2012), itappearsthatpreymightexertgreater
Dimitropoulos N, Papakyriakou A, Dalkas GA, Chasapis CT, Poulas
selective pressure on their predators, as described in our
current study. As this evolutionary manner has been shown K, Spyroulias GA. 2011. A computational investigation on the role of
tooccurintwodistantspecies,webelieveitwillberevealedin glycosylationinthebindingofalpha1nicotinicacetylcholinereceptorwith
moretoxinsfromdiversevenomousanimals. twoalpha-neurotoxins.Proteins,79(1):142–152.
DoleyR,MackessySP,KiniRM.2009.Roleofacceleratedsegmentswitchin
COMPETINGINTERESTS exonstoaltertargeting(ASSET)inthemolecularevolutionofsnakevenom
proteins.BMCEvolutionaryBiology,9:146.
Theauthorsdeclarethattheyhavenocompetinginterests.
Endo T, Tamiya N. 1987. Current view on the structure-function
AUTHORS’CONTRIBUTIONS relationshipofpostsynapticneurotoxinsfromsnakevenoms.Pharmacology
&Therapeutics,34(3):403–451.
S.Y.Z.conceivedanddesignedthestudy. X.H.J.andS.F.Z.analyzedthe Fruchart-Gaillard C, Gilquin B, Antil-Delbeke S, Le Novere N, Tamiya T,
data.X.H.J.,S.F.Z.,B.G.andS.Y.Z.jointlywrotethemanuscript.
Corringer PJ, Changeux JP, Menez A, Servent D. 2002. Experimentally
basedmodelofacomplexbetweenasnaketoxinandthealpha7nicotinic
REFERENCES
receptor.ProceedingsoftheNationalAcademyofSciencesoftheUnited
Anisimova M, Bielawski JP, Yang ZH. 2001. Accuracy and power of the StatesofAmerica,99(5):3216–3221.
likelihood ratio test in detecting adaptive molecular evolution. Molecular FryBG,WusterW,KiniRM,BrusicV,KhanA,VenkataramanD,RooneyAP.
BiologyandEvolution,18(8):1585–1592. 2003.Molecularevolutionandphylogenyofelapidsnakevenomthree-finger
Anisimova M, Nielsen R, Yang ZH. 2003. Effect of recombination on the toxins.JournalofMolecularEvolution,57(1):110–129.
accuracyofthelikelihoodmethodfordetectingpositiveselectionatamino Huang S, Li SX, Bren N, Cheng K, Gomoto R, Chen L, Sine SM.
acidsites.Genetics,164(3):1229–1236. 2013. Complex between α-bungarotoxin and an α7 nicotinic receptor
Arbuckle K, Rodriguez De La Vega RC, Casewell NR. 2017. Coevolution ligand-binding domain chimaera. The Biochemical Journal, 454(2):
takes the sting out of it: Evolutionary biology and mechanisms of toxin 303–310.
ZoologicalResearch39(6):431–436,2018 435
JansaSA,VossRS.2011.Adaptiveevolutionofthevenom-targetedvWF coupling contributes to high-affinity subtype-selective binding of
proteininopossumsthateatpitvipers.PloSOne,6(6):e20997. alpha-bungarotoxintonicotinicreceptors.TheBiochemicalJournal,454(2):
KangTS,GeorgievaD,GenovN,MurakamiMT,SinhaM,KumarRP,Kaur 311–321.
P, Kumar S, Dey S, Sharma S, Vrielink A, Betzel C, Takeda S, Arni RK, SunagarK,JacksonTNW,UndheimEAB,AliSA,AntunesA,FryBG.2013.
SinghTP,KiniRM.2011.Enzymatictoxinsfromsnakevenom: structural Three-fingered RAVERs: Rapid Accumulation of Variations in Exposed
characterizationandmechanismofcatalysis.TheFEBSJournal, 278(23): Residuesofsnakevenomtoxins.Toxins,5(11):2172–2208.
4544–4576. Trémeau O, Lemaire C, Drevet P, Pinkasfeld S, Ducancel F, Boulain JC,
KarlinA.2002.Emergingstructureofthenicotinicacetylcholinereceptors. MénezA.1995.GeneticEngineeringofSnakeToxins.JournalofBiological
NatureReviews.Neuroscience,3(2):102–114. Chemistry,270(16):9362–9369.
Kini RM. 2011. Evolution of three-finger toxins - a versatile mini protein UnwinN.2005.Refinedstructureofthenicotinicacetylcholinereceptorat
scaffold.ActaChimicaSlovenica,58(4):693–701. 4Aresolution.JournalofMolecularBiology,346(4):967–989.
LandauM,MayroseI,RosenbergY,GlaserF,MartzE,PupkoT,Ben-Tal VossRS,JansaSA.2012.Snake-venomresistanceasamammaliantrophic
N.2005.ConSurf2005:Theprojectionofevolutionaryconservationscores adaptation: lessons from didelphid marsupials. Biological Reviews of the
ofresiduesonproteinstructures.NucleicAcidsResearch,33(WebServer CambridgePhilosophicalSociety,87(4):822–837.
issue):W299–302. WangN,Orr-UrtregerA,KorczynAD.2002.Theroleofneuronalnicotinic
LevandoskiMM,LinY,MoiseL,MclaughlinJT,CooperE,HawrotE.1999. acetylcholinereceptorsubunitsinautonomicganglia:lessonsfromknockout
Chimeric analysis of a neuronal nicotinic acetylcholine receptor reveals mice.ProgressinNeurobiology,68(5):341–360.
amino acids conferring sensitivity to alpha-bungarotoxin. The Journal of Wong WS, Yang Z, Goldman N, Nielsen R. 2004. Accuracy and power
BiologicalChemistry,274(37):26113–26119. of statistical methods for detecting adaptive evolution in protein coding
Mordvintsev DY, Polyak YL, Levtsova OV, Tourleigh YV, Kasheverov IE, sequences and for identifying positively selected sites. Genetics, 168(2):
ShaitanKV,UtkinYN,TsetlinVI.2005.Amodelforshortalpha-neurotoxin 1041–1051.
bound to nicotinic acetylcholine receptor from Torpedo californica: Yang Z. 1998. Likelihood ratio tests for detecting positive selection and
comparison with long-chain alpha-neurotoxins and alpha-conotoxins. applicationtoprimatelysozymeevolution.MolecularBiologyandEvolution,
ComputationalBiologyandChemistry,29(6):398–411. 15(5):568–573.
NeiM,GuX,SitnikovaT.1997.Evolutionbythebirth-and-deathprocess YangZ,SwansonWJ.2002.Codon-substitutionmodelstodetectadaptive
inmultigenefamiliesofthevertebrateimmunesystem.Proceedingsofthe evolution that account for heterogeneous selective pressures among site
National Academy of Sciences of the United States of America, 94(15): classes.MolecularBiologyandEvolution,19(1):49–57.
7799–7806. ZhangS,GaoB,ZhuS.2015.Target-DrivenEvolutionofScorpionToxins.
NielsenR,YangZ.1998.Likelihoodmodelsfordetectingpositivelyselected ScientificReports,5:14973.
amino acid sites and applications to the HIV-1 envelope gene. Genetics, ZhuS,GaoB.2016.Positiveselectionincathelicidinhostdefensepeptides:
148(3):929–936. adaptation to exogenous pathogens or endogenous receptors?. Heredity,
Sine SM, Huang S, Li SX, daCosta CJB, Chen L. 2013. Inter-residue 118(5):453–465.
436 www.zoores.ac.cn