Báo cáo khoa học: New insights into evolution of crustacean hyperglycaemic hormone in decapods – first characterization in Anomura pdf

strigosa are composed of signal peptides of 29 and 31 amino acids, respectively, and CHH precursor-related peptides CPRPs of 50 and 40 amino acids, respectively, followed by a mature hor

hormone in decapods – first characterization in Anomura Nicolas Montagne´, Daniel Soyez, Dominique Gallois, Ce´line Ollivaux and Jean-Yves Toullec

Universite´ Pierre et Marie Curie – Paris 6, FRE 2852 CNRS – Prote´ines: Biochimie Structurale et Fonctionnelle, Equipe Biogene`se des Peptides Isome`res, Paris, France

The neurohormones of the crustacean hyperglycaemic

hormone (CHH) family are structurally related

pep-tides encoded by a multigene family that is specific to

arthropods In decapods, these neurohormones are

mainly produced in the major neuroendocrine organ,

situated in the eyestalk: the X-organ⁄ sinus gland

(XO⁄ SG) system They play important roles in

metab-olism, reproduction and development of the animals

Their size ranges from 72 to 83 amino acid residues,

and their main structural signature is the conserved spacing of six cysteyl residues, arranged in three disul-fide bridges [1] CHH family peptides are also present

in hexapods, as the ion transport peptide (ITP), a neu-ropeptide characterized in several insect species [2–4] shares the same structural signature

With regard to crustaceans, two subtypes may be distinguished when amino acid sequences of the various CHH family peptides are aligned [5] Type I

Keywords

Anomura; CHH; CPRP; molecular evolution;

neuropeptide

Correspondence

N Montagne´, Equipe Biogene`se des

Peptides Isome`res – FRE 2852 CNRS,

Universite´ Pierre et Marie Curie, 7 Quai

Saint-Bernard, 75252 Paris Cedex 05,

France

Fax: +33 1 44 27 23 61

Tel: +33 1 44 27 22 53

E-mail: nicolas.montagne@snv.jussieu.fr

Database

Sequences for P bernhardus CHH and

G strigosa CHH have been submitted to

the GenBank database under the accession

numbers DQ450960 and EF492145,

respec-tively

(Received 11 October 2007, revised 4

December 2007, accepted 17 December

2007)

doi:10.1111/j.1742-4658.2007.06245.x

The neuropeptides of the crustacean hyperglycaemic hormone (CHH) family are encoded by a multigene family and are involved in a wide spec-trum of essential functions In order to characterize CHH family peptides

in one of the last groups of decapods not yet investigated, CHH was stud-ied in two anomurans: the hermit crab Pagurus bernhardus and the squat lobster Galathea strigosa Using RT-PCR and 3¢ and 5¢ RACE methods, a preproCHH cDNA was cloned from the major neuroendocrine organs (X-organs) of these two species Hormone precursors deduced from these cDNAs in P bernhardus and G strigosa are composed of signal peptides of

29 and 31 amino acids, respectively, and CHH precursor-related peptides (CPRPs) of 50 and 40 amino acids, respectively, followed by a mature hor-mone of 72 amino acids The presence of these predicted CHHs and their related CPRPs was confirmed by performing MALDI-TOF mass spectro-metry on sinus glands, the main neurohaemal organs of decapods These analyses also suggest the presence, in sinus glands of both species, of a pep-tide related to the moult-inhibiting hormone (MIH), another member of the CHH family Accordingly, immunostaining of the X-organ⁄ sinus gland complex of P bernhardus with heterologous anti-CHH and anti-MIH sera showed the presence of distinct cells producing CHH and MIH-like pro-teins A phylogenetic analysis of CHHs, including anomuran sequences, based on maximum-likelihood methods, was performed The phylogenetic position of this taxon, as a sister group to Brachyura, is in agreement with previously reported results, and confirms the utility of CHH as a molecular model for understanding inter-taxa relationships Finally, the paraphyly of penaeid CHHs and the structural diversity of CPRPs are discussed

Abbreviations

CHH, crustacean hyperglycemic hormone; CPRP, CHH precursor-related peptide; IPRP, ITP precursor-related peptide; ITP, ion transport peptide; MIH, moult-inhibiting hormone; MOIH, mandibular organ-inhibiting hormone; SG, sinus gland; VIH, vitellogenesis-inhibiting hormone;

XO, X-organ.

peptides, the CHHs sensu stricto, are typically

72 amino acid residues long, and their protein

precur-sors contain, between the signal peptide and the CHH

progenitor sequence, a cryptic peptide called a CHH

precursor-related peptide (CPRP), which is removed

during precursor post-translational processing CPRP

is co-released with CHH within the haemolymph, from

the SG nerve endings [6], but, to date, no function has

been assigned to it In every decapod species

investi-gated, at least one CHH form was found, but the

pres-ence of several isoforms in a single species has

frequently been reported: these isoforms may arise

from expression of different genes or from various

post-translational modifications such as N-terminal

cyclization or l⁄ d isomerization of a specific residue

[7] In addition, CHH-producing sites are located

outside eyestalks, and synthesize either an

eyestalk-like CHH [8] or a CHH-eyestalk-like peptide arising from the

same gene by tissue-specific alternative splicing, with

unknown function [9]

Historically, CHH was named so because of its most

prominent bioactivity upon injection within the

ani-mal, namely rapid and sustained hyperglycaemia

However, a number of experimental studies have since

demonstrated that CHH is a pleiotropic hormone, but

its precise physiological roles are far from clear and

vary greatly according to species For example, CHH

is involved in the control of female reproduction,

either positively in the lobster Homarus americanus [10]

or negatively in the green tiger prawn Penaeus

semi-sulcatus [11] CHH inhibits the synthesis of methyl

farnesoate – a juvenile hormone-like compound that

mainly acts on gonad growth – in the spider crab

Libi-nia emarginata, acts on lipid metabolism and is

impli-cated in osmoregulation in several decapod species

[12] In the shore crab Carcinus maenas, CHH may be

involved in the control of moulting, together with a

neuropeptide from the type II subfamily: the

moult-inhibiting hormone (MIH) [13]

Type II subfamily peptides include moult-inhibiting

hormone (MIH), vitellogenesis-inhibiting hormone

(VIH) and mandibular organ-inhibiting hormone

(MOIH) At the preprohormone level, no cryptic

pep-tide such as CPRP is associated with type II peppep-tides

A paradigm in crustacean endocrinology is that,

dur-ing the intermoult stage, MIH inhibits ecdysteroid

bio-synthesis by the moulting glands, i.e the Y-organs

[14] MIHs have been described in all decapod groups

investigated, except the Homarida, in which a CHH

isoform has a moult-inhibiting function [15] In this

taxon, another type II peptide has been described: the

VIH, which controls ovarian development by

inhibit-ing vitellogenesis in female lobsters [10] This function

seems to be performed by CHH in other decapods, as mentioned above The last type II peptide known to date is MOIH It inhibits methyl farnesoate synthesis

by the mandibular organ, and has only been character-ized in the crab Cancer pagurus [16] It may have arisen from an MIH gene duplication in the genus Canceronly [17]

Since the first elucidation of a CHH sequence, in the shore crab Carcinus maenas [18], about 100 CHH family peptides have been characterized in 35 decapod species, but some taxa remain relatively poorly investi-gated For example, over 70% of all CHHs known have been described in brachyurans and penaeids, and

a few decapod groups remain unexplored In order to better comprehend the diversity of this peptide family,

we decided to focus our studies on one of these unin-vestigated groups: Anomura Two species found in the French littoral were selected for this work: the hermit crab Pagurus bernhardus and the squat lobster Gala-thea strigosa We cloned full-length CHH precursor cDNAs from X-organs of the two species, and analy-sed the peptide content of the sinus glands by MALDI-TOF mass spectrometry to check for the presence of the peptides predicted from molecular cloning Furthermore, immunochemistry was per-formed on P bernhardus eyestalks using heterologous anti-CHH and anti-MIH sera The CPRP and CHH sequences identified in this work were included in sepa-rate alignments, and a phylogenetic tree of decapod CHHs was estimated by maximum-likelihood recon-struction

Results

Molecular cloning of CHH precursor cDNAs from

P bernhardus and G strigosa After RT-PCR and 3¢ and 5¢ RACE steps, a complete preproCHH cDNA sequence was obtained from the total RNA extract of XO cells of P bernhardus This

796 bp sequence contains an open reading frame of

468 bp, encoding a 155 amino acid prepropeptide (Fig 1) The most likely signal peptide cleavage site, based on the neural network and the hidden Markov model prediction methods, is between residues Ser29 and Arg30 The 126 amino acid residue propeptide resulting from signal peptide excision contains a typi-cal di-basic processing site at residues Lys80–Arg81, which is the cleavage site between the 50 amino acid CPRP and the 74 amino acid CHH The propeptide ends with a C-terminal Gly154–Lys155, a typical amidation site This results in the production of a

72 residue amidated mature hormone In the 3¢ UTR,

a putative polyadenylation signal (AATAAA) is

pres-ent 12 bp upstream of the poly(A) tail

As for the hermit crab, a complete preproCHH

cDNA was also sequenced from G strigosa XO

mate-rial This cDNA is 889 bp in length and contains an

open reading frame of 444 bp, corresponding to a

preprohormone of 147 amino acid residues (Fig 2)

This precursor can be divided as follows: a signal

pep-tide of 31 residues (the most probable signal peppep-tide

cleavage site predicted by both methods is between

Ala31 and Arg32), a CPRP of 40 residues, a di-basic

cleavage site for potential prohormone convertase

mat-uration, and a 74 amino acid residue CHH, with two

final residues (Gly146–Lys147) that may be removed

during C-terminal amidation of the peptide The only

putative polyadenylation signal found (ATTAAA) is

69 bp upstream of the poly(A) tail

Mass spectrometry analyses on P bernhardus and G strigosa sinus gland extracts

Mass spectra obtained by analysis of a small amount

of SG extract (0.03 SG equivalent) are presented in Fig 3 In the spectrum from P bernhardus SG extract (Fig 3A), several ions were observed in the 2000–

12 000 m⁄ z range The ion with an m ⁄ z at 8345 very likely corresponds to CHH, as this mass value is very close to the M + H+value of 8345.6 Da calculated from the cDNA sequence and taking into account the predicted post-translational maturation steps (forma-tion of three disulfide bridges, cycliza(forma-tion of the N-ter-minal glutaminyl residue and C-terN-ter-minal amidation)

In addition, the spectrum shows an ion at m⁄ z 9392 and three prominent ions with m⁄ z values at

4787, 4844 and 5001, respectively The latter probably

Fig 1 Nucleotide sequence of the

P bernhardus CHH precursor cDNA, with

the complete open reading frame in capital

letters The deduced amino acid sequence

is indicated below (the asterisk indicates the

stop codon) Both nucleotide and amino acid

numbers are indicated at the end of the

lines The putative polyadenylation signal

(AATAAA) is indicated in bold italic letters.

The locations of upstream (PabCHH-U1 and

PabCHH-U2) and downstream (PabCHH-D1

and PabCHH-D2) primers, used for 5¢ RACE

and 3¢ RACE, respectively, are indicated by

grey arrows.

corresponds to the CPRP, as calculation of the mass

of the putative CPRP present in the CHH precursor

gives a theoretical M + H+value of 5002.6 Da In

addition, several ions with masses below 3000 were

observed Calculations were performed to check

whether these ions could correspond to multi-charged

ions of other observed peaks, with no result

Similar conclusions may be drawn upon examination

of the mass spectrum from G strigosa SG extract

(Fig 3B): an ion with an m⁄ z at 8315 was present,

which may correspond to CHH, as the calculated

M + H+value from cloning data is 8316.5 Da Also,

as with P bernhardus spectrum, an ion with

high-er m⁄ z (9175) was present in the spectrum, and the ion

with an m⁄ z at 4197 very likely corresponds to the

CPRP, the theoretical mass deduced from the cDNA

being 4198.7 Da The major ion mass observed in

G strigosa SG extract was at m⁄ z 7605: this value

perfectly fits with the predicted mass of the CHH trun-cated by seven amino acid residues on the C-terminus side, which may result in an ion with a calculated mass

of 7604.6 Da

Anti-CHH and anti-MIH immunoreactivities in

P bernhardus eyestalks Immunocytochemistry experiments were conducted on whole mounts of eyestalks of P bernhardus using anti-Homarus americanus CHH and anti-Cancer pagurus MIH sera Confocal micrographs revealed that both the anti-CHH and anti-MIH antisera produced intense and homogenous labelling all along the neurons of the

XO⁄ SG system (Fig 4A) Classically, the X-organ (the grouping of perikarya in which the neuropeptides are synthesized) is located inside the medulla terminalis of the eyestalk, whereas neuronal endings constituting the

Fig 2 Nucleotide sequence of the

G strigosa CHH precursor cDNA, with the complete open reading frame in capital letters The deduced amino acid sequence

is indicated below (the asterisk indicates the stop codon) Both nucleotide and amino acid numbers are indicated at the end of the lines The putative polyadenylation signal (ATTAAA) is indicated in bold italic letters The locations of upstream (GasCHH-U1 and GasCHH-U2) and downstream (GasCHH-D1 and GasCHH-D2) primers, used for 5¢ RACE and 3¢ RACE, respectively, are indicated by grey arrows.

sinus gland are located on the periphery of the upper

medulla interna About 30 CHH-immunoreactive

(CHH-IR) and 10 or so MIH-immunoreactive

(MIH-IR) perikarya were observed in the XO (27 CHH-IR

and nine MIH-IR cells in Fig 4B) Labelling was

cytoplasmic and granular, and co-localization of the

two labels (which should result in an orange coloration

on merged images) was never observed With regard to

morphological criteria, the two types of perikarya were

indistinguishable: they displayed an ovoid shape a

mean size of 30· 35 lm, with the exception of one

larger CHH-IR cell body that was found in every

preparation examined (arrow on Fig 4B) CHH-IR

and MIH-IR cells were not located in separate areas

of the XO, as a cluster of five MIH-IR cell bodies was

observed near the start of the axonal tract, whereas

the others were dispersed among CHH-IR structures

at the periphery of the XO Axons were grouped

in a tract of approximately 1.5 mm long and

40 lm wide, in which MIH-IR and CHH-IR axons

were well separated (Fig 4C) More varicosities were

seen in the MIH-IR axons than in the CHH-IR

axons The SG measured around 600 lm long by

300 lm wide MIH-IR neuronal endings were grouped

in a central area of the SG, whereas CHH-IR endings

were distributed all over the neurohaemal structure (Fig 4D)

Discussion

Early studies on CHH in Anomura mainly focused on the group specificity of these peptides, based on cross-injection of eyestalk extracts [19,20] These experiments revealed a weak cross-reactivity between Anomura and closely related taxa (Brachyura and Astacida): eyestalk extracts of P bernhardus caused hyperglycaemia in the crab Carcinus maenas, but not in the crayfish Asta-cus leptodactylus or Orconectes limosus, and extracts from any of these species failed to increase glycaemia

in the hermit crab More intriguing is the fact that, in the squat lobster Munida rugosa, even injection of its own eyestalk extract did not trigger hyperglycaemia Similarly, in a more recent study, the effect of lipo-polysaccharide injection was examined in various crus-tacean groups [21] Strong hyperglycaemia was elicited

in most of the species studied (in intact but not in eyestalk-ablated animals), including the hermit crab Paguristes oculatus, but not in M rugosa [21] There-fore, the question of the presence or not of CHH in anomurans, especially in squat lobsters, had remained open until now Given these data, we chose one species each of the Paguridae and Galatheidae families to search for anomuran CHHs

In the present study, two complete CHH precursor cDNA sequences were obtained from the X-organs

of P bernhardus and G strigosa All the structural fea-tures of the CHHs (type I peptides) are present in the deduced amino acid sequences: the presence of a CPRP, the di-basic cleavage site (Lys–Arg) between the CPRP and the mature hormone sequence, the posi-tion of the six cysteyl residues in the sequence, the size

of the putative mature peptide (72 amino acid resi-dues), and presence of the amidation signal (Gly–Lys)

at the C-terminal end Occurrence of the predicted CHHs in the sinus glands of the two species was con-firmed by comparison of the calculated molecular masses and those measured by mass spectrometry anal-yses performed on crude SG extracts For each species, MALDI-TOF mass spectrometry generated an ion with an average m⁄ z value that was in agreement with the masses deduced from predicted sequences In

G strigosa SG extract, the major ion at 7605 m⁄ z cor-responds to a truncated form of CHH in which the seven last residues have been removed by a proteolytic process Such degradation at the C-terminal side of the CHH has been noted previously [15,22], and this may

be due to C-terminal proteolysis during preparation of

SG extracts

Fig 3 MALDI-TOF mass spectra of sinus gland extracts (A)

Analy-sis of tissue equivalent to 0.03 SG from P bernhardus (B) AnalyAnaly-sis

of 0.03 SG equivalents from G strigosa.

The presence in the mass spectra of a single ion

between m⁄ z 8000 and 9000 (which is the mass range

for all CHHs characterized so far) suggests that a

sin-gle CHH form is present in the sinus glands of

P bernhardus and G strigosa However, the presence

of a single ion does not preclude the existence of

ste-reoisomers, which are not distinguishable by mass

spectrometry, as seen in various astacidean species

[23,24] On the other hand, a single peptide may

origi-nate from several CHH genes that differ at the level of

the CPRP, the signal peptide or the untranslated

region (UTR) but encode identical mature hormones

Such a situation has been described in Brachyura

[9,25], Homarida [26] and Astacida [27] Indeed, in

the mass spectrum of P bernhardus SG extract, in

addition to an ion corresponding to the mass of the deduced CPRP (5001 Da), two others with close

m⁄ z values (4844 and 4787) were detected that could correspond to CPRPs from different CHH precursors The fact that the corresponding cDNAs were not found in our study may be explained by a paucity of their transcripts in the XO cells relative to the major one that has been cloned In G strigosa SG, occur-rence of the CPRP deduced from the nucleotide sequence was also confirmed by mass spectroscopy, but, unlike P bernhardus, no other putative CPRPs were detected

In all decapod taxa, in addition to CHH, type II hormones (MIH, VIH, MOIH) are present in the SG Our results strongly suggest that this is also true for

A

B

D

C

Fig 4 Confocal micrographs showing the distribution of CHH-immunoreactive (green) and MIH-immunoreactive (red) structures in the eyestalk of P bernhardus (A) Anti-CHH and anti-MIH labelling of the whole X-organ ⁄ sinus gland system (assembled from three projections, B, C and D, each consisting of a series of confocal sections) (B) Twenty-seven CHH-IR and nine MIH-IR perikarya observed in the X-organ, with no co-localization (C) CHH-IR and MIH-IR axons in the axonal tract (D) CHH-IR and MIH-IR neuronal endings in the sinus gland.

Anomura A first indication is given by mass

spec-trometry analysis: ions with m⁄ z values of 9392 and

9175 were found in SG extracts of P bernhardus and

G strigosa, respectively, which fit with the mass range

of the type II peptides characterized so far As for the

CHHs, only a single ion was detected in this mass

range, suggesting that only one mature type II peptide

is present in these species, as is the case in most

deca-pods A second indication of the presence of type II

peptide(s) in Anomura is seen in our

immunohisto-chemical study Immunostaining performed on hermit

crab eyestalks using both anti-Homarus americanus

CHH and anti-Cancer pagurus MIH sera revealed the

existence of two distinct groups of neurons: the

major-ity of them were stained only by the anti-CHH serum,

and therefore probably represent the CHH secretory

cells, whereas other neurons were only reactive to the

anti-MIH serum This result indicates that the XO⁄ SG

system of P bernhardus synthesizes a peptide that

shares structural similarities with brachyuran MIHs

However, it is not yet known whether this peptide is a

functional MIH or not The organization of the

dis-tinct CHH and MIH-like production systems observed

in the hermit crab is similar in brachyurans [28], but

the number of immunoreactive perikarya in the

XO⁄ SG system (about 40) is significantly lower in

P bernhardus This is especially true for MIH-IR cells,

which are three times less numerous than CHH-IR

ones in the hermit crab, compared with two times less

in crabs During this study, we attempted to clone an

MIH-like peptide using degenerate primers deduced

from consensus sequences in the MIH subfamily, but

these attempts were not successful

The tree presented in Fig 5 was estimated by

maxi-mum likelihood from the updated CHH data set that

includes anomuran CHHs (Table 1) In this phylogeny,

the Anomura appear to be clearly monophyletic, as

P bernhardusCHH and G strigosa CHH form a clade

supported by a bootstrap value of 98 Monophyly is

also well supported for Brachyura, Astacida and

Homarida, thus enlightening the high group specificity

of CHHs The Anomura are a sister group to

Brachy-ura, which is consistent with their currently recognized

phylogenetic position, i.e grouping in a clade named

Meiura Taking this analysis further, relationships

between the various groups of Reptantia deduced from

our results are identical to those proposed in a recent

phylogeny established on the basis of both molecular

and morphological data [29], but it should be noted

that these inter-group relationships are not significantly

supported by the bootstrap scores (e.g 45 for Meiura,

or 25 for the entire Reptantia) These low bootstrap

values could be explained by the short length of CHH

sequences compared to the number of taxa included in the data set (72 residues and 29 taxa), and also by the relatively fast evolution of CHH among crustaceans Unlike the taxa cited above, the Penaeidea are para-phyletic, and the fact that two types of genes (contain-ing either three or four exons) can encode penaeid CHHs may explain this paraphyly Indeed, all CHH genes formerly described in Penaeidea contained three exons, whereas those described in other decapods (Pleocyemata) contained four exons [30] However, recently, two genes with four exons have been described in the white shrimp Litopenaeus vannamei (one of which encodes CHH2) in addition to a three-exon one (encoding the so-called CHH) [31,32] Addi-tionally, we have deduced a CHH sequence from an EST of Marsupenaeus japonicus eyestalk (see Experi-mental procedures for detail), and, based on the structure of the mRNA, this CHH also arises from transcription of a four-exon gene In Fig 5,

L vannamei CHH2 and Ma japonicus CHH cluster together with Pleocyemata CHHs – also arising from four-exon genes – whereas the other penaeid sequences form a separate branch, in which only peptides arising from three-exon genes are present This separation at the base of the tree between the ‘three-exon’ and ‘four-exon’ clades, which is well supported by the bootstrap values, probably represents a duplication event associ-ated with an exon deletion that may have occurred only in Penaeidea, leading to the presence of two types

of CHH gene in this taxon According to this hypothe-sis, the ancestral CHH gene would exhibit four exons, which agrees with the presence of a similar four-exon pattern in genes encoding insect ITPs [30] To be definitively established, such a scheme requires eluci-dation of CHH gene structure in taxa other than decapods

Among decapods, the size of CPRPs is not as well conserved as that of CHHs Until the present study, the number of amino acid residues in CPRPs was known to range from 4, in the giant tiger prawn Penaeus monodon [33], to 43, in the euryhaline crab Pachygrapsus marmoratus [25] With 50 residues, the CPRP deduced from the preproCHH sequence of

P bernhardus is the longest ever reported Associated with the fact that G strigosa CPRP is ten residues shorter, this illustrates well the variability of these cryptic peptides, even within the same decapod group As seen in the alignment of CPRP sequences (Fig 6A), the first 15 residues at the N-terminal end are relatively well conserved in Pleocyemata The only CPRP that does not exhibit this 15-residue stretch is that from G strigosa in which the three residues at positions 9–11 are missing Consensus

sequences of this domain in each taxon are shown in

Fig 6B They are represented separately for each

taxon rather than for all decapods to avoid the bias

generated by over-representation of brachyuran

CPRPs compared with other taxa Indeed, there are

only two available CPRP sequences in Anomura

(present study) and in Caridea, for which both

sequences are from Macrobrachium sp only With

regard to Penaeidea, the CPRPs deduced from

L vannamei CHH2 [32] and Ma japonicus CHH precursors (arising from four-exon genes) are similar

to those of Pleocyemata and do possess the 15-resi-due N-terminal domain (Table 1) On the other hand, the other penaeid CPRPs, in which this region is either reduced to the first 4–6 residues or is entirely absent, seem to be related to CHH precursors encoded by three-exon genes, as demonstrated for Metapenaeus ensis CHHs A and B [34,35] The CPRP

Fig 5 Phylogeny of decapod CHHs based on maximum-likelihood analysis of the CHH amino acid data set (29 taxa, 72 residues) using a JTT + G model of protein evolution Schistocerca gregaria ITP was assigned as the out-group Sequence accession numbers or references are given in Table 1 Numbers at nodes are bootstrap values based on 100 replicates The Anomuran sequences that were determined in this study are shown in bold For taxa in which the CHH genes have been sequenced, the number of exons (three or four) is indicated in a black circle after the name of the species A four-exon pattern was also assigned for taxa in which two peptides arising by alternative splic-ing have been described.

C-terminal sequence seems to be much less conserved,

except for a histidine or a glutamine located four

res-idues before the C-terminal end However, the

vari-ability is less significant when comparisons are made

within each taxon, indicating the possibility of an

intra-taxon signature (Fig 6B) In penaeid CPRPs

obtained from three-exon genes, the lack of such a

signature emphasizes the high variability of these

genes compared with four-exon ones Interestingly,

an histidyl residue is also present at an identical posi-tion in the precursors of the only hexapod peptide known to belong to the CHH family, the ion trans-port peptide (ITP), in which a short sequence (7–

10 amino acid residues) is present between the signal peptide and the mature ITP (Fig 6A) To date, these ITP precursor-related peptides (IPRPs) are only

Table 1 Amino acid sequences used in this study CHH, crustacean hyperglycemic hormone; CPRP, CHH precursor-related peptide; IPRP, ITP precursor-related peptide; ITP, ion transport peptide; MOIH, mandibular organ-inhibiting hormone; SG, sinus gland.

Sequence

Mature peptide included in the CHH data set

Precursor-related peptide included in the CPRP data set

UniProt acc number or reference

a Not labelled as CHH in the database but undoubtedly a type I peptide based on its structure b Deduced from the sequence of the alterna-tively spliced product submitted to the database.

B

Fig 6 (A) Multiple-sequence alignment of 32 CHH precursor-related peptide (CPRP) sequences from 26 decapod species and three ITP pre-cursor-related peptides (IPRP) sequences from three hexapod species, with the amino acid number indicated at the end of the line (see Table 1 for sequence accession numbers or references) (B) Consensus sequences of N- and C-terminal domains of each taxon The one-let-ter code and standard colouring are used (http://biomodel.uah.es/Jmol/colors/jmol_colors.en.htm) The major amino acids are shown by large letters; small fonts indicate minor amino acids in the sequences Two letters of the same font size indicate equivalent occurrence.