Bioinformatic discovery and initial characterisation of nine novel antimicrobial peptide genes in the chicken

Immunogene ics (2004) 56: 170–177 DOI 10.1007/s00251-004-0675-0 ORIGINAL PAPER Da id J. Lynn .Rowan Higgs .Susan Gaines . Joanna Tie ney .Tha appel James .And ew T. Lloyd . Ma io A. Fa es .G ace Mulcahy .Cliona O’Fa elly Bioin o ma ic disco e y and ini ial cha ac e isa ion o nine no el an imic obial pep ide genes in he chicken Recei ed: 30 Janua y 2004 / Re ised: 22 Ma ch 2004 / Accep ed: 22 Ma ch 2004 / Published online: 18 May 2004 #Sp inge -Ve lag 2004 Abs ac An imic obial pep ides (AMPs) a e essen ial componen s o inna e immuni y in a ange o species omD osophila o humans and a e gene ally hough o ac by dis up ing he memb ane in eg i y o mic obes. In o de o disco e no el AMPs in he chicken, we ha e implemen ed a bioin o ma ic app oach ha in ol es he clus e ing o mo e han 420,000 chicken exp essed sequence ags (ESTs). Simila i y sea ching o p o eins— p edic ed o be encoded by hese EST clus e s— o homology o known AMPs has esul ed in he in silico iden i ica ion o ull-leng h sequences o se en no el gallinacins (Gal-4 o Gal-10), a no el ca helicidin and a no el li e -exp essed an imic obial pep ide 2 (LEAP-2) in he chicken. Di e en ial gene exp ession o hese no el genes has been demons a ed ac oss a panel o chicken issues. An e olu iona y analysis o he gallinacin amily has de ec ed si es—p ima ily in he ma u e AMP— ha a e unde posi i e selec ion in hese molecules. The unc ional implica ions o hese esul s a e discussed. Keywo ds Chicken .De ensins .An imic obial pep ide . Inna e immune sys em In oduc ion An imic obial pep ides (AMPs), essen ial componen s o inna e hos de ence in species as di e se as plan s, lies and mammals a e gene ally hough o ac by dis up ing he memb ane in eg i y o mic obes (Kagan e al. 1990; Sa chell e al. 2003). In an age when an ibio ic esis ance is an inc easing p oblem, hese pep ides a e o in e es as po en ial no el pha maceu ical agen s. In e eb a es, he e a e wo majo amilies o AMPs: de ensins and ca he- licidins. In mammals, α-de ensins and β-de ensins a e wo s uc u ally dis inc ca ionic, cys eine- ich AMPs, which di e in size and in he spacing o a six-cys eine s uc u al mo i (Liu e al. 1997). α-de ensins a e unique o mammals, bu β-de ensins a e much mo e widely dis ibu ed and, in he chicken, ou β-de ensins, known as gallinacins, ha e been desc ibed o da e (E ans e al. 1994; Ha wig e al. 1994; Zhao e al. 2001). Homologous pep ides ha e also been desc ibed in he u key (E ans e al. 1994; Zhao e al. 2001). Gal-1, Gal-1αand Gal-2 we e isola ed om chicken he e ophils, while Gal-3 was shown o be cons i u i ely exp essed in he epi helia o a ange o issues and o be inducible in he achea ollowing in ec ion (Zhao e al. 2001). Gallinacins exhibi a iable ac i i y agains a numbe o G am-posi i e and G am- nega i e bac e ia, and Gal-1 and Gal-1αa e ac i e agains he yeas Candida albicans (E ans e al. 1995; Ha mon 1998). Ca helicidins a e a amily o highly di e se AMPs bu a e all encoded by p ep opep ides con aining highly D. J. Lynn .R. Higgs .S. Gaines .T. James .A. T. Lloyd . C. O’Fa elly (*) Educa ion and Resea ch Cen e, S . Vincen ’s Uni e si y Hospi al, 4 Dublin, I eland e-mail: [email p o ec ed] Tel.: +353-1-2094940 Fax: +353-1-2838123 D. J. Lynn .R. Higgs Depa men o Medicine, Uni e si y College Dublin, 4 Bel ield, Dublin, I eland T. James Moyne Ins i u e o P e en i e Medicine, T ini y College Dublin, 2 Dublin, I eland A. T. Lloyd Depa men o Gene ics, T ini y College Dublin, 2 Dublin, I eland J. Tie ney .G. Mulcahy Depa men o Mic obiology and Pa asi ology, Facul y o Ve e ina y Medicine, Uni e si y College Dublin, 4 Bel ield, Dublin, I eland M. A. Fa es Biology Depa men , Na ional Uni e si y o I eland, Maynoo h, I eland Conway Ins i u e, Uni e si y College Dublin, 4 Bel ield, Dublin, I eland conse ed ca helin domains. To da e, ca helicidins ha e only been desc ibed in mammals, including humans, monkeys, ho ses, ca le, sheep, goa s, pigs, abbi s, mice and guinea pig ( o e iew, see Ramana han e al. 2002). Each species has a iable numbe s o ca helicidin genes; a iodac yls in pa icula ha e high copy numbe s (Scocchi e al. 1997), whe eas humans and mice ha e only one gene copy each (Age be h e al. 1995; Cowland e al. 1995; Gallo e al. 1997; La ick e al. 1995). Neu ophils a e a pa icula ly ich sou ce o ca helicidins in a a ie y o species. In humans, ca helicidins ha e also been ound o be exp essed in se e al o he issues, including he es is (Age be h e al. 1995; Malm e al. 2000), squamous epi helia (F ohm Nilsson e al. 1999; Nize e al. 2001), ai way epi helia (Bals e al. 1998), swea glands (Mu akami e al. 2002b), sali a y glands (Mu akami e al. 2002a) and colon (Hase e al. 2002). Ca helicidins ha e a wide spec um o an imic obial ac i i y and ha e been shown o be ac i e agains G am-nega i e and G am- posi i e bac e ia (T a is e al. 2000), ungi (Shin e al. 2000) and en eloped i uses (Tamamu a e al. 1995). Ac i i y agains a pa icula mic obe depends on he ype o ma u e pep ide and he species om which i o igina es. In his s udy, we desc ibe a bioin o ma ics app oach o he iden i ica ion o no el AMPs in he chicken. This me hod in ol es homology sea ching o clus e ed chicken exp essed sequences ags (ESTs) by BLAST (Al schul e al. 1997) and by he mo e sensi i e hidden Ma ko model (HMM) p o ile sea ching (Eddy 1998). Ou app oach has iden i ied eigh no el an imic obial pep ides, se en gallinacins and one ca helicidin in he chicken. We show ha all hese pep ides as well as li e -exp essed an imi- c obial pep ide 2 (LEAP-2)—an AMP we ha e p e iously iden i ied in chicken (Lynn e al. 2003)—a e exp essed a he mRNA le el in a panel o chicken issues. Fu he - mo e, an e olu iona y analysis o he gallinacin amily has p o ided e idence ha ce ain amino acid si es in he ac i e pep ide a e subjec o posi i e selec ion. Ma e ials and me hods F om he dbEST Web si e (h p://www.ncbi.nlm.nih.go /dbEST/), 422,426 chicken ESTs we e downloaded. P io o he clus e ing s ep, epea sequences in he ESTs we e masked using Repea Mas- ke (Smi and G een, unpublished), and con aminan sequences we e emo ed using SeqClean (h p://www. ig .o g/ db/ gi/so wa e/) o minimize alse clus e ing due o spu ious EST simila i y. The ESTs we e hen clus e ed using The Ins i u e o Genomic Resea ch Gene Indices clus e ing ools (TGICL) (Pe ea e al. 2003) (a ailable om h p://www. ig .o g/ db/ gi/so wa e/). ESTs we e clus e ed i hey sha ed mo e han 30 bp o a leas 95% iden i y. The clus e ed con ig sequences a e a ailable a h p://www.bin .o g/ immunogene ics/con igs. a. The sequences o each clus e we e pos -p ocessed wi h ESTScan (Iseli e al. 1999). ESTScan de ec s and econs uc s po en ial coding egions in ESTs, using a no el HMM me hod ha can au oma ically co ec o ame-shi e o s. F om he coding egions, he p edic ed p o ein can be de e mined (a ailable a h p:// www.bin .o g/immunogene ics/es scanp edic ions. a). A da abase o known AMPs, as iden i ied by an SRS sea ch (h p://s s.ebi.ac.uk/) o he Swissp o -T embl p o ein da abase was cons uc ed and is a ailable a h p://e cbin o1.ucd.ie/APPDb/. All AMPs om his da abase we e sea ched agains he clus e ed EST con igs and he da abase o p o eins p edic ed by ESTScan om he EST clus e s using he BLAST p og ams, wi h an E- alue cu -o o 0.001 (Al schul e al. 1997). Molecula weigh and ne cha ge o he comple e chicken p o eins we e p edic ed using he P o Pa am ool (h p://www.expasy.o g/ ools/p o pa am.h ml). To sea ch o no el an imic obial p o eins ha could ha e been missed by BLAST, we cons uc ed HMMs o a numbe o AMP amilies. All he sequences anno a ed as ei he α-de ensins, ca helicidins o hepcidins in he P am da abase (Ba eman e al. 2002)—a manually cu a ed and anno a ed collec ion o p o ein amilies (h p://www.sange .ac.uk/So wa e/P am)—we e ex ac ed. The gallinacin amily is no ep esen ed in he P am da abase, so all known gallinacin sequences we e ex ac ed om he Na ional Cen e o Bio echnology In o ma ion GenPep p o ein da abase. These sequences included Gal-1 (P46156), Gal-1α(Q9DG59), Gal- 2 (P46158), Gal-3 (Q9DG58), u key he e ophil pep ide-1 (THP1) (P80391), THP2 (P80392) and u key β-de ensin (TBD) (Q9DG57). To c ea e an HMM p o ile o each amily, all cons i uen sequences o AMPs we e aligned using he T-Co ee p og am (No edame e al. 2000), and hese alignmen s we e used as inpu o he HMMER, e sion 2.1.1, sui e o p og ams (Eddy 1998) (h p://hmme .wus l. edu/). The HMM o each amily was cons uc ed using he Table 1 PCR p ime sequences and p edic ed p oduc leng hs Ta ge mRNA a 5′P ime 3′P ime P oduc size (bp) Gal-1 5′-GAAATGCTCAAGATTTCACCTCTG-3′5′-CCTTTATTCAGCAGAGAAAAGCAG-3′231 Gal-2 5′-GCATAAACACTTCATGAGTCCATC-3′5′-GAAGAAAGGCAGTGCAGAAGATA-3′166 Gal-3 5′-CCTTCTTCCTCTTGTTTCTCCAG-3′5′-ATCAACCTCATATGCTCTTCCAC-3′158 Gal-4 5′-GATCCTTTACCTGCTGCTGTCT-3′5′-TCCTCACACAGCAAGATTTTAGTC-3′185 Gal-5 5′-GATCCTTTACCTGCTGCTGTCT-3′5′-AGCAAGAGCCTATTCCATTGTTAC-3′176 Gal-6 5′-ATGAGAATCCTTTTCTTCCTTGTTGC-3′5′-TTAGGAGCTAGGTGCCCATTTGCAGC-3′201 Gal-7 5′-ATCGTGCTCCTCTTTGTGGCAGTTCA-3′5′-CTACAACCATCTACAGCAAGAATACT-3′171 Gal-8 5′-CTGTTCTCCTCTTCCTCTTCCAG-3′5′-AATCTTGGCACAGCAGTTTAACA-3′170 Gal-9 5′-ATGCAGATCCTGCCTCTCCTCTTTGCT-3′5′-TCAGGAATACCATCGGCTCCGGCAGCAGAA-3′201 Gal-10 5′-ATGAGGAACCTTTGTTTCGTGT-3′5′-TCAGGTCTTGGTGGGAGTTGGTG-3′198 LEAP-2 5′-CACCATGCACTGTTTGAAAATTATGGCA-3′5′-TCACTCGGAGGCCGTTCTAAGGAA-3′235 Ca helicidin 5′-CACCATGCTGAGCTGCTGGGTGCTGCTG-3′5′-TCACTTCTTCTTGATCGCCCGGTA-3′451 β-ac in 5′-GCGCTCGTTGTTGACA-3′5′-TCATCCCAGTTGGTGACA-3′206 a Gal-Gallinacin, LEAP-2 li e -exp essed an imic obial pep ide 2 171 hmmbuild p og am, and hmmcalib a e was used o calib a e E- alue sco es. The HMM p o iles we e hen used o sea ch agains he da abase o p o eins p edic ed by ESTScan om he EST clus e s using he hmmsea ch p og am. E olu iona y analysis o he gallinacins A mul iple-sequence alignmen o he gallinacin amily o AMPs (including he no el sequences and he homologous u key sequences) was cons uc ed using he T-Co ee p og am (No edame e al. 2000). A neighbo -joining phylogene ic ee was in e ed om he p o ein alignmen using MEGA, e sion 2.1, wi h he Poisson co ec ed model implemen ed (Kuma e al. 2001). One housand boo s ap eplica es we e ca ied ou o es he signi icance o each node in he ee. To cons uc an alignmen o he coding sequences, he p o ein alignmen was used as a empla e and a ‘copygaps’Pe l sc ip was used o align he DNA, main aining he gaps ha we e p esen in he p o ein alignmen . Any columns in he DNA alignmen ha had mo e han h ee gap cha ac e s we e emo ed. The opology o he neighbo -joining ee and he DNA alignmen we e used as inpu o he CODEML and CODEMLSITES p og ams om he PAML package, e sion 3.12 (Yang 1997), o es o e idence o posi i e selec ion du ing he e olu ion o he gallinacins. The p inciple in ol ed in such es s is o compa e he a es o synonymous (d S ) and non-synonymous (amino acid changing: d N ) changes among he DNA sequences. I amino acid changes a e selec i ely neu al (i.e. mu a ions ha a e nei he ad an ageous o dele e ious), hey will be ixed a he same a e as synonymous mu a ions and ω a io (d N /d S )=1. ω alues >1 a e aken o indica e ha amino acid changes a e accumula ing a a as e a e han is accep able unde a neu al mu a ion model. Tha is o say, he a e o amino acid changes (d N ) signi ican ly exceeds he a e o synonymous changes (d S ) a he DNA le el. The CODEML p og am es s o a iable selec i e p essu es among lineages in he phylogeny by looking o signi ican di e ences in ω a ios. To es o a iable selec i e p essu es among phylogene ic lineages, he one- a io model, which assumes an equal ω a io o all b anches in he phylogeny, was compa ed o he ee- a ios model, which allows an independen ω a io o each b anch (Yang 1998; Yang and Nielsen 1998). The esul o his p og am is a log-likelihood alue o each model. To es which is he a ou ed model, he log- likelihood alues o each model a e compa ed by a likelihood a io es (LRT). Twice he log-likelihood di e ence be ween he wo models is compa ed o a χ 2 dis ibu ion wi h n−1d , whe e nis he numbe o b anches o he phylogeny. I a signi ican P- alue is ob ained, i can be concluded ha he ee- a ios model is he a ou ed model, and b anches on he phylogeny wi h ω alues >1 a e subjec o posi i e selec ion. Posi i e selec ion in amino acid si es Ano he way o looking o posi i e selec ion is o look o signi ican a iabili y in ω a ios among amino acid si es in he mul iple-sequence alignmen (Nielsen and Yang 1998). The CODEMLSITES p og am de e mines whe he any o six p og es- si ely mo e complex models o e olu ion a e signi ican ly be e a explaining he obse ed a ia ion in he da ase (Yang e al. 2000). The i s es compa es he models M0 and M3. Model M0 is an e olu iona y model whe eby all he amino acid si es ha e a single ω alue. This model is compa ed o M3, which classi ies he amino acid si es in o one o h ee classes, wi h he p opo ion o si es belonging o a pa icula class and he ω alues o each class o si e es ima ed by CODEMLSITES om he da a. M3 is a es o amino acid si es subjec o posi i e selec ion, as i allows o he p esence o si es wi h ω>1. The second es compa es he models M1 and M2. M1 is a model o neu al e olu ion whe e amino acid si es can be conse ed (ω=0) o neu ally e ol ing (ω=1). Model M2 is a es o selec ion, as i allows o he p esence o si es whe e ωis a ee pa ame e and as such can ha e a alue >1. The inal es , which compa es he models M7 and M8, is he mos s ingen es . M7 allows o si es wi h ω alues ha ollow a βdis ibu ion o alues be ween ω= 0 and ω=1. Model M8 is he same as M7 bu allows o he p esence o si es wi h ω>1, and compa ing hese wo models is a es o selec ion. As wi h he es o posi i e selec ion among lineages, CODEML- SITES es ima es a log-likelihood alue o each model. To es which a e he a ou ed models, he log-likelihood alues o M0 e sus M3, M1 e sus M2 and M7 e sus M8 we e compa ed by LRTs. Pos e io Bayesian p obabili ies we e calcula ed o de e mine which amino si es belong o which si e classes (Nielsen and Yang 1998). I signi ican a iabili y is e ealed, hen hose si es, which ha e ω>1 and high pos e io p obabili ies, a e likely o be unde posi i e, di e si ying selec ion. Exp ession o AMPs in chicken issue One-day-old male chickens (Cobb 500 b oile ) we e pu chased om he Knocknaga m Ha che y, hen housed in a loo pen in he Biomedical Facili y, Uni e si y College Dublin, Bel ield, Dublin, I eland. En i onmen al empe a u e was kep a a cons an 25°C. Animals we e ed comme cial coccidios a - ee s a e /g owe a ion and wa e ad lib. One bi d was sac i iced a 3 weeks o age by in a enous pen oba bi one sodium inocula ion. The issues we e quickly dissec ed, squeezed be ween Wha man il e pape o emo e excess blood, insed in saline and snap ozen in liquid ni ogen. Tissues we e s o ed a −80°C un il p ocessed u he . Following pul e isa ion o he issues using a Mik o-Dismemb a o U (B. B aun Bio ech In e na ional), o al cellula RNA was pu i ied using he RNeasy Ki (Qiagen, Wes Sussex, UK) acco ding o he manu ac u e ’s ecommenda ions. Spec ophome ic analysis was pe o med in o de o assess he quan i y and quali y o o al RNA. Single-s anded cDNA was syn hesised om 1 μg RNA using oligo- dT p ime (P omega, Madison, Wis.) and Omnisc ip (Qiagen). The AMP-speci ic cDNAs we e ampli ied by PCR using Taq polyme ase (Qiagen) and p ime s designed in e nally om he coding sequence o Gal-1 o Gal-10, LEAP-2, ca helicidin and β-ac in. Thi y cycles (94°C o 30 s, 55°C o 30 s and 72°C o 30 s) we e used o ampli ica ion. PCR p oduc s we e sepa a ed by elec opho esis on e hidium b omide-s ained 2% aga ose gels and isualised using Eagle Eye (S a agene, La Jolla, Cali .). A lis o PCR p ime sequences and p oduc leng hs a e shown in Table 1. cDNA cloning Posi i e issues om he abo e exp ession panel we e chosen as sou ce ma e ial o each speci ic gene. The cDNA was ampli ied as desc ibed abo e bu using P u DNA polyme ase (P omega) and gene-speci ic p ime s wi h CACC o e hangs ups eam o he s a codon, hus p o iding he complemen a y sequence necessa y o di ec ional cloning. The ampli ied cDNA was pu i ied (GenElu e PCR Clean-up Ki , Sigma, S . Louis, Mo.), and liga ed in o he pcDNA 3.1 cloning ec o (In i ogen, G oningen, The Ne he - lands). Cloned plasmids we e sequenced (Ad anced Bio echnology Cen e, London, UK), using ec o -speci ic p ime s and compa ed o he EST consensus sequence. Resul s In he absence o comple e genome sequence, ESTs a e a ich sou ce o no el sequence in o ma ion. By de ini ion, ESTs a e sho , e o -p one sequences. Clus e ing o ESTs ha a e likely o be encoded by he same mRNA educes he edundancy in he EST da abase, imp o es he 172 sequence quali y and inc eases he sequence co e age o a pa icula clus e . By implemen ing a bioin o ma ic app oach ha in- ol es he clus e ing o mo e han 420,000 chicken ESTs, we ha e iden i ied eigh no el AMPs in he chicken. Clus e ing o hese ESTs esul ed in he gene a ion o 34,819 chicken con igs and om his, 29,344 coding sequences we e p edic ed. Gi en es ima es o he gene numbe in human o 30,000 (Pennisi 2003), we expec o ha e a leas pa ial sequence in o ma ion o mos chicken genes. TBLASTN sea ches (which sea ches p o ein que ies agains a nucleo ide da abase) o known AMPs agains he clus e ed EST con igs iden i ied i e con igs wi h homology o he β-de ensins, which we ha e named Gal-4 o Gal-8. These sea ches also iden i ied a no el chicken ca helicidin. BLASTP sea ches (which sea ches p o ein que ies agains a p o ein da abase ansla ed in all six eading ames) o p o eins, p edic ed by ESTScan o be encoded by hese EST clus e s, ailed o iden i y any o he gallinacins o ca helicidins. Due o hei small size and poo sequence conse a ion, sea ching o no el AMPs by con en ional homology sea ch ools such as BLAST (Al schul e al. 1997) may mean ha signi ican hi s a e missed. Howe e , he p esence o conse ed mo i s makes hese pep ides good candida es o HMM p o ile sea ching. An HMM p o ile is a p obabilis ic model o a p o ein amily mul iple- sequence alignmen , which uses posi ion-speci ic sco es o indica e he likelihood o each amino acid occu ing in each posi ion in he alignmen (Eddy 1998). To sea ch o no el AMPs ha could be missed by BLAST, we cons uc ed HMM p o iles o a numbe o AMP amilies, including α-de ensins, gallinacins, ca helicidins and hepcidins. The HMM p o iles we e hen used o sea ch agains he da abase o p o eins p edic ed by ESTScan om he EST clus e s. This me hod iden i ied wo addi ional gallinacins (Gal-9 and Gal-10). Fu he mo e, his app oach also led o he iden i ica ion o a gallinacin- like sequence ha has unusual cys eine spacing. The e is e idence om o he species ha β-de ensins wi h al e na i e cys eine mo i s a e s ill ac i e as AMPs (Maxwell e al. 2003). We ha e cloned and sequenced all he no el AMPs iden i ied in his s udy and submi ed he sequences o GenBank. The accession numbe s and p ope ies o he p edic ed encoded p o eins a e summa ised in Table 2. All he sequences we e in ag eemen wi h he EST p edic ions, excep o Gal-10, which had a single synonymous change a base posi ion 159 om C o T. We ha e examined he exp ession o he known gallinacins (Gal-1–Gal-3), he no el gallinacins (Gal-4–Gal-10), ca helicidin and LEAP- 2 in a panel o 21 di e en issues om a heal hy 3-week- old chicken (Fig. 4). These issues co e he diges i e sys em, he espi a o y sys em, he geni o-u ina y sys em and se e al o he a eas o he chicken ana omy. The known gallinacins Gal-1 and Gal-2 a e exp essed s ongly in he bone ma ow and he lung, as has been p e iously shown (Zhao e al. 2001). Howe e , we ha e also shown s ong exp ession o Gal-1 and Gal-2 in he es is, mode a e exp ession in he bu sa and in es ine and low exp ession in he cloaca, gall bladde , b ain and panc eas. Gal-2 is also exp essed a low le els in he achea, ai sacs and spleen. The hi d known gallinacin, Gal-3, was exp essed in he ongue and bone ma ow, as has p e iously been shown (Zhao e al. 2001), howe e , we ound no Gal-3 exp ession in o he issues. The no el gallinacins (Gal-4–Gal-10) exhibi a iable exp ession ac oss mos o he issues examined, wi h di e en gallinacins being exp essed in di e en issues. The phylogene ically ela ed Gal-4 and Gal-5 show a simila pa e n o exp ession, as does Gal-7, wi h all h ee being highly exp essed in he bone ma ow and es is. Gal- 8 is also s ongly exp essed in he es is, and along wi h Gal-6 shows e y s ong exp ession in he li e , gall bladde and kidneys. Gal-9 is he only no el gallinacin o be exp essed in he ongue and also shows low exp ession in he oesophagus, achea, b ain and bone ma ow whils Gal-10 shows low exp ession in he la ge in es ine, kidneys and es is. In addi ion o LEAP-2 being highly exp essed in he li e , simila le els o exp ession we e ound in he Fig. 1 Neighbo -joined ee o he gallinacin amily o AMPs. Cons uc ed using MEGA, e sion 2.1 (Poisson co ec ed model, 1,000 boo s ap eplica es). B anches wi h less han 50% boo s ap suppo ha e been collapsed. GAL1–10 Gallinacins (Gal-) 1–10, GAL1A Gal-1α,THP u key he e ophil pep ide, TBD u key β- de ensin, BD07 MOUSE mouse β-de ensin 7 (Q91V70) Table 2 GenBank accession numbe s and p ope ies o no el AMPs Name Accession numbe Leng h (aa) M Ne cha ge Gal-4 AY534892 67 7.5 +8 Gal-5 AY534893 67 7.6 +7 Gal-6 AY534894 67 7.3 +3 Gal-7 AY534895 63 7.2 +8 Gal-8 AY534896 68 7.1 +2 Gal-9 AY534897 66 7.4 +4 Gal-10 AY534898 65 7.2 +2 LEAP-2 AY534899 76 8.8 +8 Ca helicidin AY534900 148 16.1 +2 173 in es ine, gall bladde and kidneys. These esul s a e consis en wi h s udies o human LEAP-2, which is also exp essed in he li e , kidney and colon (K ause e al. 2003). Chicken ca helicidin is exp essed ac oss a wide a ie y o issues, bu shows pa icula ly high le els o exp ession in he bu sa, es is and bone ma ow, and is he only no el AMP o show exp ession in he gizza d. E olu iona y analysis o he gallinacins We ha e pe o med an e olu iona y analysis o he gallinacin amily and ha e de ec ed si es ha a e unde posi i e selec ion in hese molecules. A neighbo -joining phylogene ic ee was econs uc ed om he amino acid alignmen (Fig. 1). This ee opology was used in he subsequen analyses o de ec adap i e e olu ion. To es o a iable ω a ios among phylogene ic lineages, he one- a io model (Goldman and Yang 1994)—which assumes he same ω a io o all lineages—was compa ed using he LRT o he ee- a io model (Yang 1998), which assumes an independen ω a io o each b anch. The ee- a io model is no signi ican ly be e han he one- a io model (P>0.5) bu does, howe e , p edic a iable ω alues among lineages, some o which a e g ea e han 1 (Fig. 2). Since he LRT did no e eal a signi ican di e ence, we canno conclude ha he e is e idence o posi i e selec ion among he gallinacin lineages. To es o posi i e selec ion a indi idual amino acid si es, LRTs we e ca ied ou be ween model M0 and M3, M1 and M2, and M7 and M8. All 3 models (M2, M3 and M8) which allow o selec ion (Table 3) a e signi ican ly a ou ed o e he o he models (P<0.001) in all cases (Table 4). Gallinacins a e encoded as p ep opep ides ha a e p o eoly ically clea ed o elease he C- e minal AMP. All o he si es p edic ed o be subjec o posi i e selec ion a e loca ed in he ma u e AMP and no in he p ep opep- ide egion (Fig. 3a), sugges ing unc ional signi icance. The si es p edic ed o be subjec o posi i e selec ion ha e been displayed supe imposed on he h ee-dimensional s uc u e o mouse β-de ensin 7, a ela ed molecule o which a h ee-dimensional s uc u e is a ailable (Fig. 3b). The si es unde posi i e selec ion occu h oughou he molecule and no pa icula pa e n o clus e ing o si es is disce nable. Discussion We ha e applied a bioin o ma ics app oach ha in ol es he clus e ing o mo e han 420,000 ESTs o he iden i- ica ion o no el AMPs in he chicken. This app oach has Table 3 E idence o adap i e e olu ion among si es in chicken gallinacins. ℓLog-likelihood alue o model, d N /d S Ra io o a es o synonymous (d S ) and non-synonymous (amino acid changing: d N ) changes among he DNA sequences Model ℓd N /d S Posi i ely Selec ed Si es M0—one- a io −2,662.94 0.5834 M1—neu al −2,581.48 0.8923 M2—selec ion −2,541.32 2.5815 25,29,30,33,34,39,42,45,47,49,52,54,55,57,58,63,65 M3—disc e e −2,516.42 1.2094 21,23,24,25,26,27,28,29,30,32,33,34,35,39,42,45,46, 47,48,49,52,54,55,56,57,58,61,62,63,64,65 M7—β−2,531.96 0.6170 M8—βand ω−2,516.42 1.2987 25,29,30,42,45,52,57,63,65 Fig. 2 Phylogeny o gallinacins. B anch leng hs we e es ima ed by maximum likelihood unde he ee- a io model, which assumes an independen ω alue o each b anch. B anches wi h no ω alues shown had alues=∞.ω alues >1 a e shown in bold ace Table 4 Likelihood a io es o de ec adap i e e olu ion Models 2Δℓχ 2 alue d P- alue M1 e sus M2 2(−2581.48–2541.32) 80.32 2 <0.001 M0 e sus M3 2(−2662.94–2541.32) 243.24 4 <0.001 M7 e sus M8 2(−2531.96–2516.42) 31.08 2 <0.001 174 Fig. 3 a Si es p edic ed o be unde posi i e selec ion in he gallinacins. Si es p edic ed o be unde posi i e selec ion a e highligh ed in he mul iple-sequence alignmen . Si es shown in ed a e hose si es p edic ed o be unde posi i e selec ion (model M8). Pos e io p obabili ies o hese si es a e all g ea e han 0.95. Si es shown in blue a e he si es ha 100% conse ed ac oss all ope a ional axonomic uni s (OTUs). The ma u e AMP o Gal-1 is highligh ed. bThe s uc u e o he ma u e mouse β-de ensin 7 (PDB en y = 1E4T) was displayed using RasMol, e sion 2.7.2.1 (h p://www.open asmol.o g/so wa e/ asmol/). Si es shown in ed a e hose si es p edic ed o be unde posi i e selec ion in he gallinacins. Si es shown in blue a e he si es ha 100% conse ed ac oss all OTUs Fig. 4 Exp ession o Gal-1– Gal-10, ca helicidin (Ca h) and li e -exp essed an imic obial pep ide 2 (LEAP-2) in a panel o issues om a heal hy chicken. 1 ongue, 2oesophagus, 3p o- en iculus, 4c op, 5gizza d, 6 li e , 7small in es ine, 8 LARGE in es ine, 9cloaca, 10 bu sa o Fab icius, 11 gall blad- de , 12 achea,13 lung, 14 ai sacs, 15 b ain, 16 skin, 17 kidney, 18 spleen, 19 panc eas, 20 es is, 21 bone ma ow 175 iden i ied nine no el AMPs, se en o which a e gallinacins, one a ca helicidin and one a LEAP-2 (Lynn e al. 2003). We ha e shown he di e en ial exp ession o hese genes in a panel o issues om a single chicken and ha e cloned and sequenced he mRNAs encoding hese no el AMPs. Iden i ica ion o hese AMPs in he chicken will aid in he s udy o he inna e immune esponse o he chicken, economically an impo an species. Mo eo e , hese no el AMPs may be exploi ed o he de elopmen o new he apeu ic agen s o economically signi ican chicken diseases such as coccidiosis, which esul s in a loss o he wo ld poul y indus y ha is es ima ed a $700 million annually. These AMPs could be po en ially de eloped as na u al al e na i es o he a i icial an ibio ics ha a e commonly ed o chickens and which a e o g owing public conce n. In his s udy, we ha e also de ec ed posi i e selec ion a se e al amino acid si es loca ed in he ac i e an imic obial pep ide egion o he gallinacin amily o an imic obial pep ides. I is likely ha as bi ds e ol ed o occupy new niches, hey we e aced wi h new anges o mic obial pa hogens. E olu ion o an imic obial pep ides wi h new sensi i i ies capable o a ge ing no el in ec ious agen s would con e a selec i e ad an age. The e is expe imen al e idence ha gallinacin pep ides a e di e se in hei po ency agains di e en pa hogens. Fo example, Gal-1 and Gal-1αa e ac i e agains he yeas Candida albicans, whe eas he o he s es ed so a do no show ac i i y (E ans e al. 1995; Ha mon 1998). Ou esul s indica e ha gallinacins ha e been subjec o adap i e e olu ion o inc ease he s uc u al and unc ional di e si y o his p o ein amily. This is an e ec i e esponse in an a ms ace agains an inc easing di e si y o mic obial pa ho- gens. Acknowledgemen s This esea ch was suppo ed by he Food Ins i u ional Resea ch Measu e g an no. 01/R&D/D/135 om he I ish Depa men o Ag icul u e, Food and Ru al De elopmen . All expe imen s desc ibed in his manusc ip comply wi h he cu en laws o he Republic o I eland. Au ho s Da id Lynn and Rowan Higgs con ibu ed equally o his wo k. Re e ences Age be h B, Gunne H, Odebe g J, Kogne P, Boman HG, Gudmundsson GH (1995) FALL-39, a pu a i e human pep ide an ibio ic, is cys eine- ee and exp essed in bone ma ow and es is. P oc Na l Acad Sci USA 92:195–199 Al schul SF, Madden TL, Scha e AA, Zhang J, Zhang Z, Mille W, Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new gene a ion o p o ein da abase sea ch p og ams. Nucleic Acids Res 25:3389–3402 Bals R, Wang X, Zaslo M, Wilson JM (1998) The pep ide an ibio ic LL-37/hCAP-18 is exp essed in epi helia o he human lung whe e i has b oad an imic obial ac i i y a he ai way su ace. P oc Na l Acad Sci USA 95:9541–9546 Ba eman A, Bi ney E, Ce u i L, Du bin R, E wille L, Eddy SR, G i i hs-Jones S, Howe KL, Ma shall M, Sonnhamme EL (2002) The P am p o ein amilies da abase. Nucleic Acids Res 30:276–280 Cowland JB, Johnsen AH, Bo egaa d N (1995) hCAP-18, a ca helin/p o-bac enecin-like p o ein o human neu ophil spe- ci ic g anules. FEBS Le 368:173–176 Eddy SR (1998) P o ile hidden Ma ko models. Bioin o ma ics 14:755–763 E ans EW, Beach GG, Wunde lich J, Ha mon BG (1994) Isola ion o an imic obial pep ides om a ian he e ophils. J Leukoc Biol 56:661–665 E ans EW, Beach FG, Moo e KM, Jackwood MW, Glisson JR, Ha mon BG (1995) An imic obial ac i i y o chicken and u key he e ophil pep ides CHP1, CHP2, THP1, and THP3. Ve Mic obiol 47:295–303 F ohm Nilsson M, Sands ed B, So ensen O, Webe G, Bo egaa d N, S ahle-Backdahl M (1999) The human ca ionic an imic o- bial p o ein (hCAP18), a pep ide an ibio ic, is widely exp essed in human squamous epi helia and colocalizes wi h in e leukin- 6. In ec Immun 67:2561–2566 Gallo RL, Kim KJ, Be n ield M, Kozak CA, Zane i M, Me luzzi L, Genna o R (1997) Iden i ica ion o CRAMP, a ca helin- ela ed an imic obial pep ide exp essed in he emb yonic and adul mouse. J Biol Chem 272:13088–93 Goldman N, Yang Z (1994) A codon-based model o nucleo ide subs i u ion o p o ein-coding DNA sequences. Mol Biol E ol 11:725–736 Ha mon BG (1998) A ian he e ophils in in lamma ion and disease esis ance. Poul y Sci 77:972–977 Ha wig SS, Swide ek KM, Kok yako VN, Tan L, Lee TD, Panyu ich EA, Aleshina GM, Shamo a OV, Leh e RI (1994) Gallinacins: cys eine- ich an imic obial pep ides o chicken leukocy es. FEBS Le 342:281–285 Hase K, Eckmann L, Leopa d JD, Va ki N, Kagno MF (2002) Cell di e en ia ion is a key de e minan o ca helicidin LL-37/ human ca ionic an imic obial p o ein 18 exp ession by human colon epi helium. In ec Immun 70:953–963 Iseli C, Jongeneel CV, Buche P (1999) ESTScan: a p og am o de ec ing, e alua ing, and econs uc ing po en ial coding egions in EST sequences. P oceedings o he in e na ional con e ence on in elligen sys ems molecula biology, pp 138– 148 Kagan BL, Sels ed ME, Ganz T, Leh e RI (1990) An imic obial de ensin pep ides o m ol age-dependen ion-pe meable chan- nels in plana lipid bilaye memb anes. P oc Na l Acad Sci USA 87:210–214 K ause A, Silla d R, Kleemeie B, Klu e E, Ma onde E, Conejo- Ga cia JR, Fo ssmann WG, Schulz-Knappe P, Nehls MC, Wa le F, Wa le S, Ade mann K (2003) Isola ion and biochemical cha ac e iza ion o LEAP-2, a no el blood pep ide exp essed in he li e . P o ein Sci 12:143–152 Kuma S, Tamu a K, Jakobsen IB, Nei M (2001) MEGA2: molecula e olu iona y gene ics analysis so wa e. Bioin o - ma ics 17:1244–1245 La ick JW, Hi a a M, Zhong J, W igh SC (1995) An i-mic obial ac i i y o human CAP18 pep ides. Immuno echnology 1:65– 72 Liu L, Zhao C, Heng HH, Ganz T (1997) The human β-de ensin-1 and α-de ensins a e encoded by adjacen genes: wo pep ide amilies wi h di e ing disul ide opology sha e a common ances y. Genomics 43:316–320 Lynn DJ, Lloyd AT, O’Fa elly C (2003) In silico iden i ica ion o componen s o he Toll-like ecep o (TLR) signaling pa hway in clus e ed chicken exp essed sequence ags (ESTs). Ve Immunol Immunopa hol 93:177–184 Malm J, So ensen O, Pe sson T, F ohm-Nilsson M, Johansson B, Bja ell A, Lilja H, S ahle-Backdahl M, Bo egaa d N, Eges en A (2000) The human ca ionic an imic obial p o ein (hCAP-18) is exp essed in he epi helium o human epididymis, is p esen in seminal plasma a high concen a ions, and is a ached o spe ma ozoa. In ec Immun 68:4297–4302 Maxwell AI, Mo ison GM, Do in JR (2003) Rapid sequence di e gence in mammalian β-de ensins by adap i e e olu ion. Mol Immunol 40:413–421 176 Mu akami M, Oh ake T, Do schne RA, Gallo RL (2002a) Ca helicidin an imic obial pep ides a e exp essed in sali a y glands and sali a. J Den Res 81:845–850 Mu akami M, Oh ake T, Do schne RA, Schi ek B, Ga be C, Gallo RL (2002b) Ca helicidin an i-mic obial pep ide exp ession in swea , an inna e de ense sys em o he skin. J In es De ma ol 119:1090–1095 Nielsen R, Yang Z (1998) Likelihood models o de ec ing posi i ely selec ed amino acid si es and applica ions o he HIV-1 en elope gene. Gene ics 148:929–936 Nize V, Oh ake T, Lau h X, T owb idge J, Rudisill J, Do schne RA, Pes onjamasp V, Pi aino J, Hu ne K, Gallo RL (2001) Inna e an imic obial pep ide p o ec s he skin om in asi e bac e ial in ec ion. Na u e 414:454–457 No edame C, Higgins DG, He inga J (2000) T-Co ee: a no el me hod o as and accu a e mul iple-sequence alignmen . J Mol Biol 302:205–217 Pennisi E (2003) Human genome. A low numbe wins he GeneSweep Pool. Science 300:1484 Pe ea G, Huang X, Liang F, An onescu V, Sul ana R, Ka amyche a S, Lee Y, Whi e J, Cheung F, Pa izi B, Tsai J, Quackenbush J (2003) TIGR Gene Indices clus e ing ools (TGICL): a so wa e sys em o as clus e ing o la ge EST da ase s. Bioin o ma ics 19:651–652 Ramana han B, Da is EG, Ross CR, Blecha F (2002) Ca helicidins: mic obicidal ac i i y, mechanisms o ac ion, and oles in inna e immuni y. Mic obes In ec 4:361–372 Sa chell DP, Sheynis T, Shi a uji Y, Kolushe a S, Ouelle e AJ, Jelinek R (2003) In e ac ions o mouse Pane h cell α-de ensins and α-de ensin p ecu so s wi h memb anes. P osegmen inhi- bi ion o pep ide associa ion wi h biomime ic memb anes. J Biol Chem 278:13838–13846 Scocchi M, Wang S, Zane i M (1997) S uc u al o ganiza ion o he bo ine ca helicidin gene amily and iden i ica ion o a no el membe . FEBS Le 417:311–315 Shin SY, Kang SW, Lee DG, Eom SH, Song WK, Kim JI (2000) CRAMP analogues ha ing po en an ibio ic ac i i y agains bac e ial, ungal, and umo cells wi hou hemoly ic ac i i y. Biochem Biophys Res Commun 275:904–909 Tamamu a H, Mu akami T, Ho iuchi S, Sugiha a K, O aka A, Takada W, Ibuka T, Waki M, Yamamo o N, Fujii N (1995) Syn hesis o p o eg in- ela ed pep ides and hei an ibac e ial and an i-human immunode iciency i us ac i i y. Chem Pha m Bull (Tokyo) 43:853–858 T a is SM, Ande son NN, Fo sy h WR, Espi i u C, Conway BD, G eenbe g EP, McC ay PB J , Leh e RI, Welsh MJ, Tack BF (2000) Bac e icidal ac i i y o mammalian ca helicidin-de i ed pep ides. In ec Immun 68:2748–2755 Yang Z (1997) PAML: a p og am package o phylogene ic analysis by maximum likelihood. Compu Appl Biosci 13:555–556 Yang Z (1998) Likelihood a io es s o de ec ing posi i e selec ion and applica ion o p ima e lysozyme e olu ion. Mol Biol E ol 15:568–573 Yang Z, Nielsen R (1998) Synonymous and nonsynonymous a e a ia ion in nuclea genes o mammals. J Mol E ol 46:409–418 Yang Z, Nielsen R, Goldman N, Pede sen AM (2000) Codon- subs i u ion models o he e ogeneous selec ion p essu e a amino acid si es. Gene ics 155:431–449 Zhao C, Nguyen T, Liu L, Sacco RE, B ogden KA, Leh e RI (2001) Gallinacin-3, an inducible epi helial β-de ensin in he chicken. In ec Immun 69:2684–2691 177