The mammary gland-specific marsupial ELP and eutherian CTI share a common ancestral gene

21 Pages·2012·2.49 MB·English

by  Elizabeth A Pharo

Pharoetal.BMCEvolutionaryBiology2012,12:80
http://www.biomedcentral.com/1471-2148/12/80
RESEARCH ARTICLE Open Access
The mammary gland-specific marsupial ELP and
eutherian CTI share a common ancestral gene
Elizabeth A Pharo1,2*, Alison A De Leo1,2, Marilyn B Renfree1,3, Peter C Thomson2,4, Christophe M Lefèvre1,2,5 and
Kevin R Nicholas1,2,5
Abstract
Background: The marsupial early lactation protein(ELP) gene is expressed in the mammary gland and the protein is
secreted intomilk during earlylactation (Phase 2A). MatureELP shares approximately 55.4% similarity withthe
colostrum-specificbovinecolostrum trypsin inhibitor (CTI)protein. Although ELP and CTI bothhave a single bovine
pancreatic trypsin inhibitor (BPTI)-Kunitz domainand aresecreted only during the early lactation phases,their
evolutionary history is yet tobe investigated.
Results:Tammar ELPwas isolated from agenomic library and the fat-taileddunnart and Southern koala ELPgenes
cloned from genomic DNA. The tammarELP genewas expressedonly in the mammary gland during late
pregnancy (Phase 1) and early lactation (Phase 2A). The opossum and fat-tailed dunnart ELP and cow CTI transcripts
were cloned from RNA isolated from the mammary gland and dog CTI from cells in colostrum. The putativemature
ELP and CTI peptides shared 44.6%-62.2%similarity. Insilico analyses identified the ELPand CTI genes inthe other
species examined and provided compelling evidence that theyevolved from acommon ancestral gene. In addition,
whilst the eutherianCTI genewas conserved in the Laurasiatherian orders Carnivoraand Cetartiodactyla, it had
become a pseudogene inothers. These data suggest that bovineCTI may bethe ancestral gene of theArtiodactyla-
specific, rapidlyevolving chromosome13 pancreatic trypsin inhibitor (PTI),spleentrypsin inhibitor (STI) and the five
placenta-specific trophoblast Kunitzdomainprotein(TKDP1-5)genes.
Conclusions: MarsupialELP and eutherianCTI evolved from an ancestral therian mammal gene beforethe
divergence of marsupialsand eutheriansbetween 130 and 160 million years ago.The retention of theELP gene in
marsupialssuggests that this earlylactation-specific milk protein may have an importantrole in the immunologically
naïve young of these species.
Background milk production, and there are dramatic changes in milk
Marsupials and eutherians diverged between 130 and 160 composition [5,9-13]. In contrast, eutherians have a long
million years ago [1-3] and evolved very different pregnancy during which maternal investment is high
reproductive strategies [4-6]. Marsupials have an ultra- [14,15]. During eutherian lactation, milk composition
shortgestationranging from10.7daysforthestripe-faced remains relatively constant apart from the initial produc-
dunnart(Smithopsismacroura)[7]to38daysforthelong- tionofcolostrum 24–36hrpostpartum (pp)[16].
nosedpotoroo(Potoroustridactylus)[8]anddeliveranal- Thetammar wallaby (Macropus eugenii)hasa26.5-day
tricialyoung[5]. pregnancy after embryonic diapause [17]. After giving
Organogenesis is completed after birth supported by a birth, the tammar produces milk for ~300 days until the
long and physiologically complex lactation, during which young is weaned. Phase 1 of lactation is comprised of
thereis an increasein maternalmammary gland sizeand mammary development during pregnancy and lactogen-
esis around parturition. At birth, the altricial young
(~400 mg) attaches to one of the four teats [5,9,13,18].
*Correspondence:[email protected] Lactation proceeds only in the sucked gland, whilst the
1DepartmentofZoology,TheUniversityofMelbourne,Melbourne,Victoria
remaining three glands regress [5,9]. The young remains
3010,Australia.
2CooperativeResearchCentreforInnovativeDairyProducts permanently attached to the teat from the day of birth
Fulllistofauthorinformationisavailableattheendofthearticle
©2012Pharoetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative
CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedtheoriginalworkisproperlycited.
Pharoetal.BMCEvolutionaryBiology2012,12:80 Page2of21
http://www.biomedcentral.com/1471-2148/12/80
until day 100 pp (Phase 2A) followed by detachment immunoglobulins against proteolysis during the crucial
from the teat and a period of intermittent sucking while period of immunoglobulin transfer from cow to calf via
confined in the pouch between days 100–200 pp (Phase colostrum. However, its function is yet to be determined.
2B) [5,13,18]. The final phase is from day 200 to at least Although CTI and ELP are expressed in early milk, bo-
day 300 when the young suckles variably and begins to vine CTI secretion is brief (~1-2 days) [31,37], but mar-
graze as well as maintaining a milk intake (Phase 3) [18]. supial ELP expression is prolonged (up to 100 days pp)
These phases are highly correlated with changes in milk [20,21,25,28]. However, their secretion in milk is corre-
composition and mammary gland gene expression lated with the period of immuno-incompetence in the
[10,13,19]. Milk protein genes such as α-lactalbumin, β- young[29,31].
lactoglobulin (LGB), α-casein, β-casein and κ-casein are The Kunitz domain was thought to have evolved over
induced at parturition and expressed throughout lacta- 500 million years ago [38] and is now ubiquitous in
tion, whilst others are expressed and secreted in a phase- mammals, reptiles, birds, plants, insects, nematodes,
specific manner [13]. Early lactation protein (ELP) is venoms from snakes, spiders, cone snails and sea ane-
expressed during Phase 2A only [13,20,21], whey acidic mones and in viruses and bacteria [39-42]. The arche-
protein (WAP) is Phase 2B-specific [22] and late lacta- typal protein of the Kunitz domain and the BPTI-Kunitz
tion protein A and B are characteristic to late Phase 2B/ family I2, clan IB of serine endopeptidase inhibitors in
Phase3andPhase3respectively[23,24]. the MEROPS database [43,44] is the much studied bo-
TheELPgenewasfirstidentified inanAustralianmar- vine pancreatic trypsin inhibitor, also known as aprotinin
supial, the brushtail possum (Trichosurus vulpecula) (reviewed in [45]). The Kunitz domain is characterised
[25]. ELP encodesa smallprecursor protein witha single by six conserved cysteine residues which form three di-
bovine pancreatic trypsin inhibitor (BPTI)-Kunitz do- sulphide bonds, producing a compact, globular protein
main characteristic to serine protease inhibitors. ELP is of α+β folds [43,46,47]. Serine endopeptidase inhibition
secreted in milk in multiple isoforms, which include an occurs through the bindingof the P reactive site residue
1
~8 kDa peptide and a heavily N-glycosylated protein within the ‘binding loop’ of the Kunitz domain to a
(~16 kDa) [25]. ELP was later identified in the tammar serine residue within the catalytic cleft of the protease
[13,20,21,26], the stripe-faced and fat-tailed dunnarts [47,48]. This is a reversible, tight-binding, 1:1 interaction
(Sminthopsis macroura and Sminthopsis crassicaudata [44,48]. Furthermore, the Kunitz domain P residue
1
respectively) and the South American grey short-tailed determinesprotease-specificity[39,47].
opossum (Monodelphis domestica) [27] (Refer to Add- Since its evolution, the Kunitz domain has been incor-
itional file 1: Table S1 for the species in which the puta- porated into many different genes [43,44]. In general,
tive functional ELP/CTI gene, transcript and protein each domain is encoded by a single exon [43,49]. Some
have been identified). Marsupial ELP expression is lim- genes encode proteins with a single Kunitz domain, e.g.
ited to the early phase of lactation [13,20,21,27,28] at the ELP, CTI, PTI, spleen trypsin inhibitor (STI), the five
time the mother produces milk for an immunologically trophoblast Kunitz domain protein genes (TKDP1-5) and
naïve young [29,30]. During this period, the tammar serine protease inhibitor Kunitz-type-3 (SPINT3) and
young is permanently attached to the teat and protected SPINT4. These genes, apart from the TKDPs, have 3
by humoral (passive) immunity acquired from its exons. The first exon encodes the signal- and pro-pep-
mother’smilkanditsowninnateimmunity[18,30]. tide, the second, a single Kunitz domain and the third, a
Whilst an ELP orthologue is yet to be identified in short C-terminus. However, the TKDPs have a variable
eutherians,tammarandpossumELPshare~37%similar- number of unique N domains inserted between the sig-
ity with bovine colostrum trypsin inhibitor (CTI) [20,25]. nal peptide and the Kunitz domain-encoding exon
CTI was discovered by chance in bovine colostrum over [50,51]. Genes that encode multiple Kunitz domains in-
60 years ago [31]. Putative CTI proteins with trypsin in- clude: hepatocyte growth factor activator inhibitor 1 and
hibitor activity were subsequently isolated from colos- 2, also known as SPINT1 and SPINT2 respectively (two
trum of the pig [32], cat, sheep, goat, dog, reindeer, domains), tissue factor pathway inhibitor 1 and 2 (three
ferret and Blue fox [33], but were not found in equine domains); with up to 12 domains in the Ac-KPI-1 I
colostrum [34]. These glycosylated proteins inhibited nematode (Ancylostoma caninum) protein [38,43,44]. In
serine endopeptidases such as trypsin, pepsin and addition, the Kunitz domain has been integrated into
chymotrypsin [31,32,35]. However, of these putative CTI multi-domain proteins, some of which include: the colla-
proteins, only bovine CTI has been sequenced (Add- gen α3(VI), α1(VII) and α1(XXVIII) chains, WFDC6 and
itionalfile1:TableS1)andfoundtocontainaKunitzdo- WFDC8, amyloid beta A4 protein, α1-microglobulin/
main which generally indicates serine protease inhi- bikunin precursor (AMBP), SPINLW1 [serine peptidase
bitor activity (see below) [36]. Laskowski and Laskowski inhibitor-like, with Kunitz and WAP domains 1 (eppin)]
[31] hypothesised that bovine CTI protected and the WAP, follistatin/kazal, immunoglobulin, Kunitz
Pharoetal.BMCEvolutionaryBiology2012,12:80 Page3of21
http://www.biomedcentral.com/1471-2148/12/80
and netrin domain containing (WFIKKN)1 and 2 pro- identified within these genomes with future improve-
teins [39]. Furthermore, each domain within a multi- mentsinsequence coverageandassemblies.
Kunitz domain protein, may exhibit different protease TheCTI genewaspresentinthe Laurasiatherianorders
activity, such asforthe threetandemly repeated domains Cetartiodactyla (cow, pig, common bottle-nosed dolphin)
within both tissue factor pathway inhibitor 1 and 2 and Carnivora (dog, cat, Giant panda). However, based
[43,44,52]. upon current genome assemblies, it is a pseudogene in
The early lactation/colostrum-specific expression of Afrotheria, Xenarthra, Euarchontoglires and the Laura-
ELP/CTI suggests these Kunitz domain-encoding genes siatherianordersChiropteraandPerissodactyla.
may play an important role in the neonate. The sequen- The mammalian ELP/CTI gene was composed of 3
cingof the tammar genome[53],in additionto the avail- exons and 2 introns (Figure 1). The marsupial ELP gene
ability of numerous vertebrate genomes including one rangedfrom~1.4kbforthekoalato~4.8kbforthestripe
other marsupial, the opossum, a monotreme, the platy- faceddunnart,whilsteutherianELPspannedfrom~2.5kb
pus, many eutherians, birds (chicken, Zebra finch), fish for the panda to ~3.8 kb for the pig. ELP exon 1 and 2
(Zebrafish, Japanese medaka, Three-spine stickleback, sizes respectively were highly conserved across all mam-
Tiger and Green spotted puffers), amphibian (African mals (Figure 1). Exon 1 encoded the putative signal pep-
clawed frog) andreptile (Green anole lizard), provides an tide and the first four amino acids at the N-terminus of
invaluable resource with which to investigate the evolu- the protein.The216bpexon 2(with the exception ofthe
tionofthesegenes.Weusedacomparativegenomicsap- koala, 210 bp) encoded the remainder of the N-terminal
proach based upon bioinformatics and PCR-based region, plus a single BPTI-Kunitz domain towards its 3'-
cloning of cDNA and genomic DNA to characterise the end. ELP/CTI exon 3 differed most and encoded a
marsupial ELP and eutherian CTI genes and investigate maximum of seven amino acids. The ELP/CTI transcripts
theirevolutionaryhistory. (putative translation start site to the polyadenylation sig-
nal, inclusive) were short. Marsupial ELP and eutherian
CTI transcripts ranged from 425–447 bp and 416–428 bp
Results respectively and shared 56.1%-63.6% similarity at the nu-
ELP/CTIevolvedfromacommonancestralgene cleotide level (Additional file 2: Figure S1; Additional file
To determine whether the marsupial ELP gene was 3:TablesS2A,S2B).Ahighlyconservedmarsupial-specific
present in other species, we used multiple approaches. region (87%-100%) was also identified within the ELP 3'-
We clonedtheELPgenesofthekoalaandfat-taileddun- UTR (nt 420–475, Additional file 2: Figure S1; Additional
nart and isolated tammar ELP from a genomic library. file3:TableS2C).
ELP/CTI transcripts were cloned from the mammary Based upon signal peptide analysis [56], the putative
gland of the cow, opossum and fat-tailed dunnart and ELP/CTI peptides identified in this study were predicted
the dog CTI transcript was cloned from epithelial cells to be secreted in milk, as for tammar and possum ELP
isolated from canine colostrum. We performed BLAST and bovine CTI [20,25,26,31]. The mature ELP and CTI
searches of genomic databases (Ensembl, Release 62, peptides shared 44.6%-62.2% similarity (Table 1; Add-
April 2011 [49], NCBI GenBank nr and WGS [54] and itional file 4: Table S3A). In addition, the conservation of
UCSC [55]), using a cut-off of E-value≤1e-8 (nucleo- the two Kunitz domain motifs in all species suggested
tides) and E-value≤1e-17 (proteins). To further refine they may inhibit the S1 family of serine endopeptidases
the identification of ELP/CTI orthologues based upon like many other members of the BPTI-Kunitz family
protein sequence, we also compared gene structures [43,44]. The BPTI KUNITZ 2 motif [C1-C6, C2-C4 and
(where possible) to identify genes with a similar three- C3-C5, Prosite: PS00280] indicates the 3 disulphide
exon structure to ELP/CTI. Based upon these methods, bonds whichdetermine the structure of thedomain (Fig-
no genes orthologous to marsupial ELP/eutherian CTI ure 2). This motif spanned the entire 51 amino acid
were present in fish (Zebrafish, Tiger and green spotted Kunitz domain (aa 23–73, C23-C73, C32-C56 and C48-
puffers, Three-spined stickleback), birds (chicken, zebra C69, Figure 2). The second shorter motif BPTI KUNITZ
finch), amphibian (African clawed frog), reptile (Green 1 [F-x(2)-{I}-G-C-x(6)-[FY]-x(5)-C; where x represents
anolelizard),monotreme(platypus), norsea squirts,fruit any residue, those within square brackets are permitted,
fly, nematode (Caenorhabditis elegans) or yeast. How- but those within curly brackets are not, Prosite:
ever, many of the current genomes available provide only PS00280]waslocatedwithinBPTIKUNITZ2(aa51–69,
low sequence coverage (e.g. anole lizard, 2x; green spot- Figure 2). A putative trypsin interaction site within the
ted pufferfish, 2.5x; chicken, zebra finch and platypus, Kunitz domain (from KU NCBI cd00109) [57], is also
6x; elephant, 7x). Many assemblies are also incomplete depicted(aa30–34,36,Figure2).
(contain gaps) and may contain incorrect assemblies. Conserved amino acid residues within a protein pro-
Hence it is possible that ELP/CTI orthologues may be vide an indication of sites essential for its structure and
Pharoetal.BMCEvolutionaryBiology2012,12:80 Page4of21
http://www.biomedcentral.com/1471-2148/12/80
ELP/CTI
Ex1 Ex2 Ex3
Dunnart FT 73 216 154
(4,726 bp)
3,960 323
Dunnart SF 73 216 154
(4,818 bp)
4,069 306
Koala 73 210 142
(1,387 bp)
718 314
Opossum 73 216 158
(4,497 bp)
2,219 1,831
Tammar 73 216 146
(4,268 bp)
3,522 311
Cat 76 216 134
(3,283 bp)
2,346 511
Cow 76 216 133
(2,907 bp)
1,788 694
Dog 76 216 124
(2,578 bp)
1,671 491
Dolphin 76 216 136
(2,720 bp)
1,785 507
Panda 76 216 134
(2,561 bp)
1,645 490
Pig 76 216 136
(3,771 bp)
2,849 494
0 1 2 3 4 5kb
Figure1StructureofthemarsupialELPandeutherianCTIgenes.TheELP/CTIgenesofthestripe-faced(SF)dunnart(Sminthopsismacroura)
[GenBank:AC186006],fat-tailed(FT)dunnart(Sminthopsiscrassicaudata)[GenBank:JN191336],koala(Phascolarctoscinereus)[GenBank:JN191337],
opossum(Monodelphisdomestica)[GenBank:BK008085],tammar(Macropuseugenii)[GenBank:JN191335],cat(Feliscatus,Abyssiniandomesticcat)
[GenBank:BK008083],cow(BosTaurus,HerefordBreed)[Ensembl:ENSBTAG00000016127],dog(Canisfamiliaris,Boxerbreed)[GenBank:BK008082],
dolphin(Tursiopstruncatus)[GenBank:BK008086],pig(Susscrofadomestica)[Ensembl:F1SD34_PIG(ENSSSCG00000007398)]andGiantpanda
(Ailuropodamelanoleuca)[GenBank:BK008084]have3exonsand2introns.Genesizeisindicatedwithinbracketsandreferstothenumberof
nucleotidesfromtheputativetranslationstart(ATG,exon1)tothepolyadenylationsignal(AATAAA,inclusive,exon3).Exonsarecolour-coded:
exon1(greenrectangle),theKunitzdomain-encodingexon2(blue)andexon3(red)andexonsizeisindicatedinboldtext.Intronsizesare
italicised.ThehorizontalscalebarindicatestherelativesizesoftheELP/CTIgenes(kb),withtheputativetranslationstartsite(ATG)ofall
sequencesalignedwiththeorigin(0kb).Genesaredrawnapproximatelytoscale.
biological function. Comparison of the marsupial ELP across divergent species. The opossum and dunnart ELP
and eutherian CTI precursor proteins showed that the proteins were truncated at the end of exon 2, with the
signal peptide (57.1%-81.0% similarity), the 51 aa BPTI stop codon encoded by one nucleotide in exon 2 and
KUNITZ 2 motif (54.9%-68.6%), plus the shorter 19 aa two in exon 3 (nt 323–325 inclusive; Additional file 2:
BPTIKUNITZ1motifwithinit(63.2%-73.7%)were con- Figure S1). For all other species, two different stop
served. However, the 20–22 residue linear chain of the codons within exon 3 were used. For the panda, cat and
mature ELP/CTI N-terminus had marsupial-specific and dog, the TAA stop codon (nt 333–335) was used. How-
eutherian-specific homology (59.1%-100%, Table 1; Add- ever, for the pig, cow, dolphin and the remainder of the
itional file 4: Tables S3B, S3C, S3D, S3E). Conservation marsupials, the equivalent TGA stop codon (nt 344–346
of the short (3–10 residue) C-terminus was variable inclusive)wasused.
(Additional file 4: Table S3F). Thiswasin partdue to the Surprisingly, there was little conservation of the amino
use of different stop codons in ELP/CTI transcripts acid residue type (physiochemical properties) at the P
1
Pharoetal.BMCEvolutionaryBiology2012,12:80 Page5of21
http://www.biomedcentral.com/1471-2148/12/80
Table1HomologybetweenandwithinthemarsupialELPandeutherianCTIpeptides1
Speciescomparisons Signalpeptide Maturepeptide N-terminus Kunitzmotif2(51aa) Kunitzmotif1(19aa) C-terminus
MarsupialELP 85-95% 67.5-100% 59.1-100% 76.5-100% 84.2-100% 20-100%
EutherianCTI 57.1-90.5% 70.7-88.6% 59.1-90.9% 76.5-94.1% 84.2-100% 40-83.3%
MarsupialELPvsEutherianCTI 57.1-81.0% 44.6-62.2% 18.2-59.1% 54.9-68.6% 63.2-73.7% 10-60%
PairwiseaminoacidssimilaritieswerecalculatedusingMatGAT2.01(BLOSUM62matrix).
1RefertoAdditionalfile4:TablesS3forindividualspeciescomparisons.
reactive site within the Kunitz domain (residue 33, SelectivepressureactinguponmarsupialELPand
Figure2).AlthoughtheP residuetype(basicaminoacid eutherianCTI
1
with a positively charged side chain) was conserved The evolutionaryselectionpressureacting upondifferent
amongst eutherians: K (lysine) for the pig, cow and dol- regions of the protein-coding marsupial ELP and euther-
phin and R (arginine) for the cat, dog and panda, this ian CTI transcripts was determined by dN/dS analysis
was not so for marsupials. The opossum and possum (Table 2). ThedN/dS ratio measures the number of non-
ELP P residue was acidic with a negatively charged side synonymous changes per non-synonymous site (those
1
chain (D, aspartate). However, the P residue for tammar which produce amino acid substitutions) compared to
1
(S, serine) and the koala and dunnarts (N, asparagine) thenumberofsynonymouschanges persynonymoussite
waspolarwithunchargedside chains. (no amino acid change) [59,60]. A ratio of dN/dS=1
AlthoughP residuesdiffered,allELP/CTIpeptideswere suggests a neutral condition, with nucleotide changes ac-
1
predictedtobeN-glycosylatedatasparagine-42,consistent cumulating in the absence of selection pressure, i.e. both
for bovine CTI [58] and therefore should be larger than dN and dS occur at the same rates. dN/dS<1 indicates
theirpredictedmasses(8.6to9.6kDa,datanotshown). purifyingselection, withamino acidchanges not tolerated.
Trypsin interaction site BPTIKUNITZ1(PS00280)
Exon1 Exon2 Exon3
from KU (NCBI cd00109) F-x(2)-{I}-G-C-x(6)-[FY]-x(5)-C
Signal peptide 1 11 21 31P-P' 41 51 61 71 81
11
Pig MKLSLSLALCLTLCLPGMASSGKTLASLKQEASQELFQTPPALCQLPPVGGPCKASLRRYFYNSTSAECELFMYGGCQGNANNFETTAICRRVCNPPDTKVKNG
Dolphin MKLSRLLALCLTLCLVGLASSGKTSANLQQEASQELLQTPPALCQLPAVRGPCKASLHRYFYNSTSIECEPFTYGGCQGNANNFETTEICVRVCKPPETKVKSS
Cow MKLSCLLALCLTPCLVGLASSGETSDNLKQEASQDLFQTPPDLCQLPQARGPCKAALLRYFYNSTSSACEPFTYGGCQGNDNNFETTEMCLRICQPPETEDKS-
Dog MKFSLFLALCFLLGLVGITSLEKASAHLRQEAFQELSQTLPVLCQLPPGKGPCRGRFYRYFYNSTSSECEHFIYGGCQGNANNFETTEICLKICKPPETR----
Panda MKFSPFLALCFLLCLVGISSSEKASAHLKHEAPRELSQALPAMCQLRPAKGPCRGLFYRYFSNSTSSECEHFTYGGCQGNANDFETTEICSRICKPPETG----
Cat MKFSLFLALCFPFCLVGIASSEKTSAHLEREAPQELLQTLPALCRLPPVEGPCRGRFYRYFYNSTAHECEHFTYGGCRGNANNFETTEMCLKVCKPPGTR----
Tammar MKFT-IVALYFALSLAGMTSSEKCLDQIQVNSLENLSLLVPSLCLLPPVRGNCSSQILHYFYNTTSRTCETFIYSGCNGNRNNFNSEEYCLKTCRRNKNRNNNN
Possum MKFT-IIALCLALSLVGMTSSEKLLDRIRANSLENLSRLVPSLCLLPSGRGNCDSQILRYFYNATSHTCEVFLYSGCNGNGNNFDSLECCLKTCRLNKYRNNN-
Koala MKFT-IVALCFALSLAGLTSSEKLSD--HVNSLENPYQLVPSLCLLSPARGNCNSQTLRYFYNTTSRTCEAFIYSGCHGNGNNFDSLQCCLKTCRPNKNRNDNN
DunnartSFMKFT-IIALCFAFSLAGMTSSEKLLDQIPVNSLENPSRLVPALCQLSPQRGNCNDNIRRYYYNTTSRICEEFIYTGCNGNGNNFDSVECCLKTCKLN-------
DunnartFTMKFT-IIALFFAFSLAGMTSSEKLLDQIPMNSLENPSRLVPALCQLSPQRGNCNDNIRRYYYNTTSRICEEFIYTGCNGNGNNFDSVECCLKTCKLN-------
Opossum MKFT-IVALCFALGLAGITSSEEVLEQNPLNTQENPVPLVLPLCLLPPERGNCDSLNLRYFYNSTSRLCEAFIYSGCNGNGNNFDTVECCLKTCRPN-------
**:: :** : * *::* : :: .: :* * * *# :*: *:*: ** * * **.** *:*:: * : *.
+-------------------------------------------------+
-- +-----------------------+ --
----C----------C#--------------C---------C------------C----- C--
--+----------------------+
51 residues
BPTI KUNITZ 2 (PS50279)
Figure2AlignmentofthemarsupialELPandeutherianCTIprecursorproteins.ThenucleotidesequencesoftheELP/CTImRNAtranscripts
ofthefollowingspecieswereconceptuallytranslatedandalignedwithClustalW2:tammar[GenBank:JN191338;UniProtKB/Swiss-Prot:O62845
(matureprotein)],brushtailpossum[GenBank:U34208],fat-taileddunnart(FT)[GenBank:JN191339],opossum[GenBank:JN191340],cow(Holstein-
Friesianbreed)[GenBank:JN191341]anddog[GenBank:JN191342].Thestripe-faceddunnart(SF)[GenBank:AC186006],koala[GenBank:
JN191337],cat[GenBank:BK008083],pig[Ensembl:F1SD34_PIG(ENSSSCT00000008098)],dolphin[GenBank:BK008086],andpanda[GenBank:
BK008084]ELP/CTIgeneswereconceptuallysplicedbaseduponconservedsplicesitesandtranslated.Aminoacidresiduesarenumberedbased
uponthestart(N-terminus)ofthematureELP/CTIpeptides.Blackshadingindicatesnucleotideresiduescommontoatleast10ofthespeciesand
grey,theremainderthatdiffer.Thesixconservedcysteineresidues(C1-C6,C2-C4andC3-C5),whichformthethreedisulphidebondsand
produceaglobularproteinareshadedred.Tealshadingindicatesaminoacidscommontomarsupialsandblue,thosecommontoeutherians.
Thelocationofexonsisindicatedbyarrows.Thepredictedsignalpeptidesareboxed(blue).TheBPTIKUNITZ1and2motifsareindicated(green
andredbarsrespectively)andtheputativetrypsininteractionsitefromtheKUmotif(NCBIcd00109)isdepictedbyorangetriangles.Theputative
P andP'reactivesiteresiduesareshadedyellowandpurplerespectively.Italicisedasparagine(N)residuesindicatepredictedsitesofpost-
1 1
translationalN-glycosylation.Conservationbetweengroupsofaminoacidswithstronglysimilarproperties,i.e.,scoring>0.5intheGonnetPAM
250matrixisindicated(:).Conservationbetweengroupsofaminoacidswithweaklysimilarproperties(scoring<0.5intheGonnetPAM250
matrix)isalsonoted(.).Gapswithinthealignmentareindicated(−).
Pharoetal.BMCEvolutionaryBiology2012,12:80 Page6of21
http://www.biomedcentral.com/1471-2148/12/80
In contrast, dN/dS>1 is indicative of positive Darwinian to chromosome 1q (Figure 3). The ELP/CTI gene was
selectionforaminoacidchanges[59,61]. locatedonasyntenicsegmentinthemarsupial(stripe-faced
The protein-coding marsupial ELP and eutherian CTI dunnart [27] and opossum) and eutherian genomes [49,55]
transcripts and regions within them generally exhibited a andwasgenerallyflankedbyoneorbothofthesingle-copy
trend towards purifying selection, with a dN/dS ratio <1 genesphosphatidylinositolglycan,classT(PIGT)andWAP
(Table 2). However, basedupon codon-based Z-tests, only four disulphide core domain 2 (WFDC2), confirming they
theeutherianCTIBPTIKUNITZ1motif(57ntencoding weretrueorthologues(Figure4).
19amino acids) was foundto beundergoing purifying se- The PIGT-WFDC2 region of bovine chromosome 13
lection(p<0.05).Althoughtheregionsencodingthemar- (~74.51-75.14Mb)wasunique.BovineCTIwasadjacentto
supial BPTI KUNITZ 1 motif (p=0.103) and the PIGT, but there was an insertion of ~602 kb between the
marsupial and eutherian BPTI KUNITZ 2 motifs CTI and WFDC2 genes [49,55] (data not shown). This re-
(p=0.101 and p=0.105 respectively) exhibited a strong gion included 7 Artiodactyla-specific Kunitz domain-en-
trendtowardspurifyingselection,thetestvalues(dN<dS) coding genes including PTI, STI, plus the five placenta-
werenotsignificant.Thistendencywasalsoconsistentfor specific TKDP1-TKDP5 genes inclusive [50,63]. Further-
the putative trypsin interaction site. In contrast, three more, the SPINLW1 gene which contains both a Kunitz
regionsoftheELP/CTItranscriptsshowedatrendtowards and a WAP domain and the eutherian-specific SPINT4
positive selection (dN/dS>1). These included the regions genewerelocatedafurther~38kband~90kbrespectively
encodingtheELP/CTIN-terminusandthe eutherian CTI downstream from WFDC2 [49,55] (data not shown). As
signal peptide. However, based upon codon-based Z-tests mentioned previously, these genes, with the exception of
(dN>dS),onlytheeutherianCTIsignalpeptide(p<0.05) SPINLW1andtheTKDPs,shareasimilar3-exonstructure.
wasundergoingpositiveselection. However, the TKDPs differ due to the likely “exonisation”
of an intron and its subsequent duplication to produce a
MarsupialELPandeutherianCTIsharecommonflanking variablenumber of tripartite N-domains between the exon
genes encodingthesignalpeptideandtheKunitzdomain[50,51].
In order to confirm that the marsupial ELP and eutherian
CTI genes were orthologous, we characterised the location CTIhasbeenlostinsomeeutherians
and arrangement of ELP/CTI and its flanking genes. We Using the canine sequence as the basis for mVISTA
usedfluorescenceinsituhybridisationtomaptammarELP comparative analysis [64], the region between the PIGT
Table2Averageratesofsynonymous(dS)andnon-synonymous(dN)substitutionsoccurringinmarsupialELPand
eutherianCTI
ELP/CTIprotein-coding dN SE dS SE dN/dSRatio (a)Neutral (b)Purifying (c)Positive
region selectiontest selectiontest selectiontest
(dN6¼dS)+* (dN<dS)+* (dN>dS)+*
Precursorprotein Marsupials 0.145 0.022 0.190 0.033 0.763 0.256(NS{) 0.117(NS) 1.000(NS)
Eutherians 0.194 0.026 0.225 0.033 0.862 0.232(NS) 0.472(NS) 1.000(NS)
Matureprotein Marsupials 0.166 0.026 0.185 0.036 0.897 0.653(NS) 0.334(NS) 1.000(NS)
Eutherians 0.186 0.028 0.242 0.039 0.786 0.273(NS) 0.130(NS) 1.000(NS)
Signalpeptide Marsupials 0.071 0.029 0.226 0.094 0.314 0.133(NS) 0.064(NS) 1.000(NS)
Eutherians 0.225 0.072 0.165 0.069 1.36 0.451(NS) 1.000(NS) 0.224(NS)
N-terminus Marsupials 0.240 0.064 0.116 0.048 2.07 0.064(NS) 1.000(NS) 0.041*
Eutherians 0.242 0.050 0.224 0.065 1.08 0.842(NS) 1.000(NS) 0.424(NS)
BTPIKUNITZ2# Marsupials 0.146 0.031 0.224 0.052 0.651 0.215(NS) 0.101(NS) 1.000(NS)
Eutherians 0.162 0.035 0.243 0.054 0.667 0.200(NS) 0.105(NS) 1.000(NS)
BPTIKUNITZ1~ Marsupials 0.095 0.030 0.223 0.098 0.426 0.212(NS) 0.103(NS) 1.000(NS)
Eutherians 0.066 0.026 0.264 0.110 0.250 0.122(NS) 0.046* 1.000(NS)
Trypsininteractionsite^ Marsupials 0.230 0.136 0.323 0.181 0.712 0.740(NS) 0.363(NS) 1.000(NS)
Eutherians 0.175 0.093 0.228 0.131 0.768 0.689(NS) 0.345(NS) 1.000(NS)
#PS50279153nt,51aa.
~PS0028057nt,19aa.
^18nt,6aasitefromKU(NCBIcd00109).
+CodonbasedZ-testsinMEGA5.
*p<0.05.
{NSnotsignificant.
Pharoetal.BMCEvolutionaryBiology2012,12:80 Page7of21
http://www.biomedcentral.com/1471-2148/12/80
frames). CTI exon 2 of the mouse, rat, large flying fox
(Pteropusvampyrus)andhorsealsoappearedtohavebeen
disruptedbydeletionsresultinginframe-shiftswhencom-
pared to the functional/protein-coding dog CTI exon 2.
Thedisruptionoftheprotein-codingregionofequineCTI
exons 1 and 2 by at least one mutation and one deletion
respectively would produce a frame-shift, suggested these
1q
were a recent occurrence (Additional file 5: Figure S2B
(ii)).
1q
TransposableelementswithintheELP/CTIgenes
Transposable elements integrate randomly into the gen-
ome, so the probability of the same element(s) integrat-
ing independently into orthologous positions in different
species is extremely low. They therefore act as genetic
markers and can be used to determine the phylogenetic
relationship between genes and species [65]. Further evi-
dence that marsupial ELP and eutherian CTI evolved
from a common ancestral gene was provided by CEN-
SOR retrotransposon analysis [66] (Additional file 6: Fig-
ure S3). Retroelements of conserved fragment size and
orientation were located within the PIGT-ELP/CTI re-
Figure3LocalisationofthetammarELPgenetoMacropus gion. However, the elephant and human which appear to
eugeniichromosome1qusingFISH. have lost CTI exons 2 and 3, had also lost retrotranspo-
sons in the corresponding region, but gained a MER5A
and WFDC2 genes was examined using the available element.
genome assemblies - which have variable sequence
coverage, contain gaps and may contain misassembled BovineCTI,PTI,STIandtheTKDPsshareacommon
sequences. Whilst the ELP/CTI gene was present in ancestralgene
some mammals, it appeared to have become a disrupted The location of the 8 Kunitz-domain encoding genes (in-
pseudogene in others such as the African Savanna ele- cludingCTI)onbovine chromosome13betweenthePIGT
phant and human (Figure 5). Exon 1 of the elephant and and WFDC2 genes and the Artiodactyla-specific distribu-
human CTI genes (signal- and pro-peptide) was present, tionofPTI,STIandTKDP1-5(cowandsheep[51,63])sug-
but exon 2 (Kunitz domain) and exon 3 (C-terminus) gested they may have evolved from CTI. This hypothesis
were absent (red boxes, Figure 5), suggesting they had was supported by phylogenetic analysis of the protein-cod-
been excised or transposed, whilst the horse and mouse ing regions of the mammalian ELP/CTI, bovine PTI, STI
CTIgenesinitiallyappearedintact. andTKDP1-5transcripts,withbovineSLPIusedasanout-
A closer examination of the nucleotide sequence be- grouproot(SLPIomitted,Figure6).Severaldifferentmeth-
tween PIGTand WFDC2 in these and other species using ods in PHYLIP were used to determine the evolutionary
the Ensembl and UCSC genome databases revealed that relationships. These included the character-based max-
differentmutationshadmostlikelydisruptedtheCTIgene. imum-likelihood (with/without a molecular clock) and
Exon 1 was disrupted in the elephant, Hoffmann's two- maximum parsimony, as well as distance-based analysis
toed sloth (Choloepus hoffmanni), armadillo (Dasypus (Fitch-Margoliash tree method using the Kimura distance
novemcinctus), human and other primates and horse, with model of nucleotide substitution). Trees were evaluated
exon 2 (Kunitz domain) also excised for these species, using the bootstrap method (100 replicates). Of the algo-
apart from the horse. Additional file 5: Figure S2A (i) rithms used, the maximum likelihood method using a mo-
depicts a nucleotide alignment of the functional/protein- lecular clock assumption, which assumes a constant
coding dog CTI exon 1 compared with the putative dis- evolutionary rate for all species, produced a tree with the
rupted CTI exon 1 of the elephant, sloth, human and highest bootstrap values. Huttley and colleagues [67] have
horse. Additional file 5: Figure S2A (ii) shows the trans- shown that the eutherian nucleotide substitution rates are
lated sequences to highlight mutations and/or deletions ~30% slower than for marsupials. However, all methods
within the signal peptide region of CTI. The deletion of produced consensus trees which consistently separated the
twonucleotides within human CTI exon 1 would produce 19 sequences into the two groups depicted (Figure 6). The
a frame-shift (as depicted by the +1 and +2 reading hypothesis that bovine CTI was the ancestral gene for
Pharoetal.BMCEvolutionaryBiology2012,12:80 Page8of21
http://www.biomedcentral.com/1471-2148/12/80
Opossum
Chr. 1 SYS1 TP53TG5 DBNDD2 PIGT ELP WFDC2 AEBP1 POLD2 MYL7 GCK YKT6
(501.25 to
501.67Mb)
Human
Chr. 7 AEBP1 POLD2 MYL7 GCK YKT6
p13-p15.1
HCuhmr. a2n0 SYS1-DBNDD2 DLBONCD1D020 PIGT CTI L13O0C115070 WFDC2 RPL5P2 SPINT3 WFDC6 H1PS2P1D SPINLW1
q12-q13.12 SYS1 TP53TG5 128997
Mouse
Chr. 2H3 Sys1 Tp53tg5 Dbndd2 PigT Cti Gm14317 Wfdc2 Spint3 Wfdc6a Spinlw1
Dog
Chr. 24 LOC LOC
(35.64 to SYS1 TP53TG5 DBNDD2 PIGT CTI WFDC2 611069 WFDC8 CE10 611096
35.89 kb)
Cow ~602 kbp
Chr. 13
(74.48 to SYS1 TP53TG5 DBNDD2 PIGT CTI WFDC2 SPINLW1
75.20 kb)
Chicken
C(5h.0r.8 2 t0o (-) SYS1 LO1C84219 TP53TG5 DBNDD2 PIGT SRC BLCAP MANBAL GHRH EMILIN3
4.93 Mb)
Zebrafish
Chr. 23 si:zfos-
(12.27 to ppplr3da sycp2 phactr3a PIGT wfdc2 snx21 haus8 452g4.1
12.72 Mb)
Figure4ChromosomallocationoftheELP/CTIgeneindifferentspecies.TheELP/CTIgenewaslocatedwithinasyntenicblockonopossum
Chr.1(~501.34Mb),humanChr.20q12-13.12,mouseChr.2H3,dogchr.24(~35.7Mb)andcowChr.13(~74.5Mb)[49,55].However,ELP/CTIwas
reducedtoapseudogeneinthehumanandmouse(redarrow,whitediagonalstripes)andwasabsentinthechickenandzebrafish.TheELP/CTI
genewaslocatedonthereversestrandandwasgenerallyflankedbyone,orbothofthesingle-copygenesPIGTandWFDC2.Theregion
upstreamofPIGTwasconservedinmammalsandthechickenandincludedtheSYS1[Golgi-localizedintegralmembraneproteinhomolog(S.
cerevisiae)],TP53TG5(TP53-targetgene5protein),andDBNDD2[dysbindin(dystrobrevinbindingprotein1)domaincontaining2]genes.However,a
chromosomalbreakpointwaslocateddownstreamfromtheeutherianWFDC2gene.Opossumchromosome1containedtheAEBP1(Adipocyte
enhancerbindingprotein1),POLD2[polymerase(DNAdirected),delta2,regulatorysubunit50kDa],MYL7(myosin,lightchain7,regulatory)andYKT6
[YKT6v-SNAREhomolog(S.cerevisiae)]genesandwasorthologoustohumanchromosome7p13-p15.1.Incontrast,theeutherianchromosomes
containedanumberofgeneswhichencodedKunitzand/orWAPdomains.TheseincludedSPINT3,SPINLW1,WFDC8andWFDC6,whichwere
likelytohavearisenbygeneanddomainduplications[62].Notably,therewasaninsertof~602kbbetweenbovineCTIandWFDC2.Arrows
indicatethearrangementandorientationofgenesandarenotdrawntoscale.
\bovine PTI, STI and TKDP1-5 was supported by both [28](Figure7A).ELPtranscriptsweredetectedinthetam-
an alignment of precursor proteins and phylogenetic marmammaryglandfrom~day17ofpregnancyonwards,
analysis of CTI, PTI, STI, TKDP1-5 and the SPINT4 throughoutearlylactation(Phase2A)until~day87oflac-
protein-coding transcripts (Additional file 7: Figure S4; tation. ELP was then down-regulated to minimal levels for
Additional file 8: Figure S5). Interestingly, the size of the remainder of lactation. This wasconsistent with a pre-
the Kunitz domain-encoding exon varied. Whilst the vious study of late Phase 2A/Phase 2B mammary tissues,
bovine CTI exon was 216 bp, those of the TKDPs were but the precise timing of ELP gene induction was not
196 bp, with 192 bp for PTI and STI and 175 bp for investigated [13,20,21]. Neither ELP, nor LGB was
SPINT4. Furthermore, apart from CTI and SPINT4, expressed in the virgin mammary gland and both genes
none of the Kunitz domains were predicted to be N- weredown-regulatedpostpartuminthenon-suckedglands
glycosylated. Additional evidence of the evolutionary (Figure7A),asinthebrushtailpossum[28].
history of the CTI, PTI, STI and TKDP1-5 genes was LGB expression peaked in the mammary gland during
provided by mVISTA (Additional file 9: Figures S6A Phase3,consistentwith[68].
and S5B (i-viii) and CENSOR analysis (Additional file AlthoughcDNAmicroarrayanalysisofthetammarmam-
10: Figure S7; Additional file 11: Table S4). mary gland (Figure 7B; Additional file 12: Table S5) was
baseduponcomparativeexpressionlevelsratherthanactual
TammarELPexpressionisup-regulatedatparturitionand transcript levels, the data was consistent with quantitative
ismammary-specific analysis of the Northern blot (data not shown) and micro-
Northern analysis showed that tammar ELP was up-regu- array data reported by [69]. Lastly, Northern analysis of
lated at parturition, consistent with brushtail possum ELP assortedtammartissuesamplesindicatedthatexpressionof
Pharoetal.BMCEvolutionaryBiology2012,12:80 Page9of21
http://www.biomedcentral.com/1471-2148/12/80
PIGT ELP/CTI
Dog
~602 kbp 100%
Cow 10%
100%
Elephant 10%
100%
Horse 10%
100%
Human 10%
100%
Mouse 10%
100%
Opossum 10%
35,682 35,684 35,686 35,688 35,690 35,692 35,694 35,696 kbp
gene exon CNS UTR WFDC2
Dog
100%
Cow 10%
100%
Elephant 10%
100%
Horse 10%
100%
Human 10%
100%
Mouse 10%
100%
Opossum 10%
35,698 35,700 35,702 35,704 35,706 35,708 35,710 35,712 kbp
Figure5VISTAplotofpairwisealignmentsforselectedmammalsintheregioncontainingthePIGT,ELP/CTIandWFDC2genes.
SequencehomologywithinthePIGT-ELP/CTI-WFDC2regionofthedog,cow,elephant,horse,human,mouseandopossumgenomeswas
determinedwithmVISTA[64].Thedogsequencewasusedasthereferencesequence(horizontalaxis,dogchromosome24numbering).Grey
horizontalarrowsindicategenelocationanddirectionoftranscription.Bluerectanglesindicatecodingexonsanduntranslatedregions(UTRs)of
thegenearedepictedbylightgreenrectangles.Exon1ofcanineWFDC2wasmissing(gapinthecurrentassembly)fromthedoggenomeandis
indicatedbyabluerectanglewithdiagonalwhitestripes.Therightaxisindicatesthepercentageidentitywithina100bpwindowforeach
pairwisecomparison,rangingfrom10%to100%.Regionssharinggreaterthan25%identityareshadedandtheblackhorizontallineindicates
70%identity.TheregioncontainingtheKunitzdomain-encodingELP/CTIexon2wasconservedinthecow,horse,mouseandopossum,butwas
absentintheelephantandhumanCTIgenes(redboxes).
ELP andLGBwasmammary gland-specific (Figure 8), un- ancestor before the divergence of marsupials and euther-
like the ubiquitously expressed cystatin C (CST3) gene iansatleast130millionyearsago[1,2,70].
(datanotshown).
MammalianELP/CTIandtheevolutionofbovinePTI,STI
Discussion andtheTKDPs
ELPwasoriginallythoughttobeamarsupial-specificgene The Kunitz-type inhibitor domain has been duplicated
[19].However,wehaveshownthatthemarsupialELPand many times throughout evolutionary history [38]. This
eutherian CTI genes evolved from a common therian an- was no more evident than for the region of bovine
cestralgene(Figure9).MammalianELP/CTIwasgenerally chromosome 13 on which CTI and the 7 CTI-like genes
flanked by one or both of the single copy PIGT and were located. The PTI, STI and TKDP1-5 genes were
WFDC2 genes in a region that was syntenic to that of specific to the order Cetartiodactyla, sub-order Rumi-
other mammals. The conserved genomic structure of 3 nantia [50,51,63,72], strong evidence they evolved from
exonsand2intronsandhomologoustransposableelement CTI after the divergence of the Ruminantia ~25-35 MYA
fragmentsconfirmedthatELPandCTIweretrueortholo- [1]. The CTI, PTI and STI genes had a similar 3-exon
gues.CTIwasalsoidentifiedastheputativeancestralgene structure and conserved regions within both coding and
of the ruminant-specific PTI, STI and TKDP1-5 genes. non-coding segments. The PTI and STI genes and pro-
Based upon current genome sequencing and assemblies, teins were homologous and almost certainly arose by
ELP/CTI wasnotfoundinbirds,fish,reptiles,noramphi- gene duplication [73]. However, the TKDP1-5 genes
bians, suggesting the gene was present in the therian had one or more additional exons inserted between
Pharoetal.BMCEvolutionaryBiology2012,12:80 Page10of21
http://www.biomedcentral.com/1471-2148/12/80
6 (TKDP2, 3 and 4) and 12 exons (TKDP1) [50,51,72].
Tammar
These added exons encode tripartite N-domains which
74
Possum had no similarity to database sequences or motifs and
61
84 evolved recently due to the “exonization” of an intron
Koala
within an active MER retrotransposon and its subse-
100 100 Dunnart FT quent duplication [50,63]. These elements have been
Dunnart SF associated with genetic rearrangements and deletions
[74]. This may explain the excision of CTI exons 2
Opossum
ELP/CTI (Kunitz domain) and 3 (C terminus) for the elephant
100 89 Dog and primates, based upon current genome sequencing
100 Panda and assemblies.
Cat
100 LackofconservationoftheELP/CTIputativeP reactive
1
Dolphin
82 siteresidue
100 Cow AllputativeELP/CTIpeptideswerepredictedtobesecreted
and shared a conserved single 51 amino acid Kunitz do-
Pig
main. The conserved location of the 6 cysteine residues
100 STI which form three disulphide bonds suggested ELP/CTI
PTI would, like bovine CTI [75] and PTI [46] form a globular
protein.However,neithertheidentity,physiochemicalprop-
TKDP2
100 40 erties of the ELP/CTI P reactive site residue, the trypsin
Bovine PTI, 1
35 TKDP1 STI and interaction site, nor the N- and C-terminus of the proteins
the TKDPs wereconserved.TheP “warhead”residueplaysanessential
TKDP3 1
91 68 role in the interaction of a Kunitz inhibitor domain with a
TKDP4
serine protease and a P mutation may alter the protease
1
TKDP5 specificity of the Kunitz domain to a particular substrate
Figure6AphylogenetictreeofELP/CTIandtheCTI-likebovine and the reaction kinetics [48,76]. Kunitz inhibitors with a
PTI,STIandTKDP1,2,3,4and5family.Theevolutionary basic residue, K (Cetartiodactyla) or R (Carnivora) at P
1
relationshipbetweentheprotein-codingregionsofthemarsupialELP, generally inhibit trypsin or trypsin-like serine endopepti-
eutherianCTIandbovineTKDP1-5,PTIandSTItranscriptswas
dases such as chymotrypsin, pepsin, plasmin and kallikrein
determinedbymaximumlikelihoodanalysisusingamolecularclock
in vitro (e.g. bovine CTI and PTI) [31,38,77]. However,
assumption.ThebovineSLPItranscriptwasusedasanoutgroup(data
notshown).Twomaingroupswereformed:1.mammalianELP/CTIand Kunitz domains with smaller, uncharged residues at P1,
2.bovineCTI,PTIandtheTKDPs.Numbersatbranchpointsindicate such as serine, generally inhibit elastase-like proteases (eg.
confidencelevelsasdeterminedbybootstrapvalues(100replicates). neutrophil elastase) [43,47,76]. In contrast, Kunitz domains
PhylogenetictreeswereproducedwithPhylipsoftwareversion3.69.
with an acidic, negatively-charged P residue (e.g. TKDP2)
TranscriptswerealignedwithMUSCLEandboostrappedvalues 1
exhibitminimalantiproteaseactivityinvitro[72].Compari-
generatedwithSEQBOOT.Maximumlikelihoodtreesweregenerated
withDNAMLKusingatransition/transversionratioof1.34,aGamma son of BPTI Kunitz domains suggested that the marsupial
distributionshapeof1.39with5HiddenMarkovModelcategories, ELPP aminoacidswerequiterare[43,49,55].Furthermore,
1
globalrearrangementsandwitharandomisedinputorderjumbled the absence of purifying selection within the putative ELP/
once.Theprotein-codingregionsofthefollowingtranscriptswereused
CTI trypsin interaction site and the lack of conservation of
intheanalysis:ELP/CTI,tammar[GenBank:JN191338],fat-taileddunnart
P residuesprovidesintriguingquestionsastotherole(s)of
[GenBank:JN191339],stripe-faceddunnart[GenBank:AC186006],koala 1
[GenBank:JN191337]opossum[GenBank:JN191340],brushtailpossum, themarsupialELPandeutherianCTIproteinsinvivo.
cow[GenBank:JN191341],dog[GenBank:JN191342],cat[GenBank: Not all Kunitz domains act as protease inhibitors [43].
BK008083],pig[Ensembl:F1SD34_PIG(ENSSSCT00000008098)],Giant As mentioned previously, snake and spider venoms con-
panda[GenBank:BK008084],andCommonbottlenosedolphin
tainproteinswithKunitzdomains[40].Somedomainsin-
[GenBank:BK008086],andthefollowingbovinetranscripts:PTI
hibit trypsin or chymotrypsin via P , whilst others lack
[GenBank:NM_001001554],STI[GenBank:NM_205786],TKDP1[GenBank: 1
NM_205776],TKDP2[GenBank:NM_001012683],TKDP3[GenBank: anti-proteaseactivitybuthaveneurotoxiceffectsbyacting
XM_584746],TKDP4[GenBank:NM_205775],andTKDP5[GenBank: as potassium channel blockers [41]. Peigneur and collea-
XM_614808]andSLPI[GenBank:NM_001098865]. gues [78]recently reported a sea anemoneKunitzdomain
protein, APEKTx1 (Anthopleura elegantissima potassium
channel toxin 1) which had dual functions. It exhibited
the signal- and pro-peptide-encoding and Kunitz do- both trypsin-inhibitor activity and selectively blocked the
main-encoding exons (equivalent to intron 1 of CTI, Kv1.1 type of voltage-gated potassium channels. Further-
PTI and STI) resulting in an expansion to 4 (TKDP5), more, not all Kunitz protease inhibitors act via the P
1