In such a complex system as this, where l- and d- epimers of different neuropeptides are present in the neuroendocrine organ, a puzzling question remains unanswered: how are these peptid
Trang 1aminoacyl isomerization in the crustacean hyperglycaemic hormone family
Ce´line Ollivaux1,2,3, Dominique Gallois4, Mohamed Amiche4, Maryse Boscame´ric4
and Daniel Soyez4
1 Universite´ Pierre et Marie Curie – Paris 06, UMR 7150 Mer et Sante´, E´quipe Physiologie Compare´e des Erythrocytes, Station Biologique
de Roscoff, France
2 Centre National de la Recherche Scientifique, UMR 7150, Station Biologique de Roscoff, France
3 Universite´ Europe´enne de Bretagne, UEB, Rennes, France
4 Equipe Biogene`se des Signaux Peptidiques, ER3, Universite´ Paris, France
Introduction
Modification of the chirality of a single aminoacyl
resi-due within a peptide chain is a subtle and intriguing
mechanism that remains poorly known to date, and
which leads to structural and functional diversification
of eukaryotic bioactive peptides Subsequent to the
study by Montecucchi et al [1] describing the presence
of a d-alanyl residue at the second position of
dermor-phin, a powerful opioid peptide isolated from skin
secretions of the tree frog Phyllomedusa sauvagei, such
a phenomenon has been reported in vertebrates, in different opioid peptides [2], in antibacterial and haemolytic peptides from frog skin [3], as well as in venom from the mammal Platypus [4] (Table 1)
In invertebrates, d-amino acid containing peptides (DAACPs) were isolated from the nervous system of molluscs and crustaceans [5], and from venom of a
Keywords
aminoacyl isomerization; confocal laser
scanning microscopy; crustacean
hyperglycaemic hormone; MALDI-TOF MS;
vitellogenesis inhibiting hormone
Correspondence
C Ollivaux, Universite´ Pierre et Marie Curie
– Paris 06, UMR 7150 Mer et Sante´, E´quipe
Physiologie Compare´e des Erythrocytes,
Station Biologique de Roscoff, 29682
Roscoff, Cedex, France
Fax: +33 1 44 27 23 61
Tel.: +33 1 44 27 22 62
E-mail: celine.ollivaux@upmc.fr
(Received 29 March 2009, revised 23 June
2009, accepted 26 June 2009)
doi:10.1111/j.1742-4658.2009.07180.x
d-aminoacyl residues have been detected in various animal peptides from several taxa, especially vertebrates and arthropods This unusual polymor-phism was shown to occur in isoforms of the crustacean hyperglycaemic hormone (CHH) of the American lobster because a d-phenylalanyl residue was found in position 3 of the sequence (CHH and d-Phe3 CHH) In the present study, we report the detailed strategy used to characterize, in the lobster neuroendocrine system, isomers of another member of the CHH family, vitellogenesis inhibiting hormone (VIH) We have demonstrated that the fourth residue is either an l- or a d- tryptophanyl residue (VIH and d-Trp4 VIH) Furthermore, use of antisera specifically recognizing the epimers of CHH and VIH reveals that aminoacyl isomerization occurs in specialized cells of the X organ–sinus gland neurosecretory system and that the d-forms of the two neuropeptides are not only present in the same cells, but, importantly, also are co-packaged within the same secretory vesicles
Abbreviations
CHH, crustacean hyperglycaemic hormone; DAACP, D -amino acid containing peptide; gp, guinea pig; PTM, post-translational modification;
r, rat; rb, rabbit; VIH, vitellogenesis inhibiting hormone; XO, X organ.
Trang 2spider [6] and cone snails [7] l- to d-aminoacyl
isomerization, or epimerization, comprises a true
post-translational modification (PTM) of peptides [8], as are
glycosylation or phosphorylation, where an l-residue,
which is always the same in a given peptide, is
con-verted into its d-counterpart [9] This PTM occurs at a
late step of peptide precursor processing, after the
cleavage of the propeptide [10] The nature of the
d-res-idue varies according to the peptides, although it has
consistently been found at the extremities of the
sequence at the second or third position, mostly at the
N-terminus and more rarely at the C-terminal end,
except in several conotoxins (e.g the contryphan, an
octapeptide with a d-tryptophan in the fourth position)
[11] (Table 1) This intriguing modification is likely to
involve a specific enzyme, designated as
peptidylaminoa-cyl-l⁄ d-isomerase or, more simply, isomerase [12] or
epimerase [13] Amino acid sequences of putative
enzymes have been obtained from venom of the spider
Agenelopsis aperta[14], and more recently from the skin
secretion of Bombina frogs [15] Unexpectedly, these
enzymes appear to be totally unrelated with regard to
their structures and their target residues In addition,
such an isomerase has been isolated from platypus
venom, but remains unsequenced [12,16]
Although d-amino acids have been mostly found in
the sequence of small (i.e 3–40 residues) peptides,
remarkable exceptions include peptides belonging to
the crustacean hyperglycaemic hormone (CHH) family,
which are 72–83 residues long, elaborated in the major
crustacean neurosecretory system, the X organ–sinus
gland complex In several crustacean species, two
epimers of CHH can be purified, which differ in the
configuration of the phenylalanyl residue at position 3
(i.e either l or d) To date, the presence of both CHHs (named CHH and d-Phe3 CHH) has been dem-onstrated only in Astacoidea (crayfish and lobsters), where CHH displays the same N-terminal aminoacyl sequence, at least up to the tenth residue [17] Phe3 isomerization has major physiological consequences on CHH biological activity because the all-l-peptide is strictly hyperglycaemic, whereas d-Phe3 CHH may exhibit, in addition to higher hyperglycaemic potency, other functions, such as moult inhibition [18] or osmo-regulation [19] At the present, it is unclear whether functional differences between CHH isomers are related to binding to specific receptors or to differences
in haemolymphatic clearance rate Indeed, DAACPs are known to be more stable because they are less sus-ceptible to protease degradation As noted above, CHH constitutes the archetype of an original peptide family mostly found in crustaceans, the CHH family, which contains other essential neurohormones, such as moult-inhibiting hormone, mandibular organ inhibiting hormone and vitellogenesis inhibiting hormone (VIH; also called gonad inhibiting hormone) [20]
In Homarus americanus, where d-Phe3 CHHs were first identified [21], VIH is present in the neurohaemal organs, the sinus glands, as two isoforms (VIH I and II) with an identical 77-amino acid sequence, molecu-lar mass (9135 Da) and pI (6.8) [22] Only one cDNA was cloned, encoding a precursor with a signal peptide directly flanking the progenitor VIH sequence [23] With regard to its biological functions, VIH may inhi-bit vitellogenesis synthesis in ovary or at extra-ovarian sites and also may inhibit vitellogenin uptake in oocytes [24] When American lobster VIHs were tested
in a heterologous in vivo assay, only VIH I, the hydro-philic form, demonstrated significant inhibitory activity with respect to repressing oocyte growth that had been induced by eyestalk removal in grass shrimps [25] To date, no function has been assigned to VIH II, the hydrophobic form
VIH has also been detected in male American lobster sinus gland [26], in the Norway lobster Nephrops nor-vegicus[27] and in the woodlouse Armadillidium vulgare [28] In the latter species, sinus gland grafting experi-ments have suggested that VIH could be involved in androgenic gland growth [29]
In the present study, we describe an experimental strategy that was applied to determine the difference between VIH I and II from the X organ–sinus gland system of the lobster H americanus We demonstrate that VIHs differ in the chirality of the tryptophan at position 4 This result has been exploited to develop specific antisera recognizing specifically the N-terminal end of VIH and d-Trp4 VIH, which has allowed
Table 1 D -amino acid containing peptides in animals Bold and
underlined letters indicate the D -residues CHH, crustacean
hyper-glycemic hormone; VIH, vitellogenesis inhibiting hormone; OvCNP,
Ornithorhyncus venom C-type natriuretic peptide; DPL,
defensin-like peptide.
Frog Dermal gland Dermorphin YAFGTPSNH2 [1]
Platypus Venom gland OvCNP LLHDHPN [32]
Mussel Muscle FFRF amide ALAGDHFFRFNH2 [52]
Cone snail Venom duct Contryphan GChPWEPWC [11]
Spider Venom gland Xagatoxin MEGLSF A [50]
Lobster Sinus gland CHH pQEVFDQAC [21]
Study
Trang 3identification of VIHs from the European lobster
Homarus gammarussinus gland extracts [30]
In such a complex system as this, where l- and
d- epimers of different neuropeptides are present in the
neuroendocrine organ, a puzzling question remains
unanswered: how are these peptides distributed in the
different cells of the X organ–sinus gland system? In
other words, what is the cellular specificity of the
isom-erization process? To answer this question, the
distri-bution of VIH and CHH isomers within the lobster
X organ–sinus gland complex was studied by
immuno-fluorescence⁄ confocal microscopy This technique was
supplemented by an immunogold-electron microscopy
study of the neuronal endings in the sinus gland to
determine the subcellular localization of the different
epimers Taken together, these results allow a
discus-sion of the existence of enzyme(s) converting l- to
d- residues in the peptide chain in terms of substrate
specificity and availability
Results
Purification and characterization of VIH I and II
VIHs were purified from H americanus sinus gland
extract by RP-HPLC Figure 1 shows a typical elution
pattern resulting from the fractionation of an extract
of 30 lobster sinus glands The major peptides were
eluted between 38% and 40% acetonitrile, and were
identified as CHH B, d-Phe3 CHH B, VIH I, VIH II,
CHH A and d-Phe3 CHH A, respectively, according
to their elution order, by reference to a previous study
[21] These assumptions were confirmed by a direct
ELISA performed on aliquots of the different
frac-tions, using antisera anti-4 (recognizing both VIHs) and the two antisera discriminating CHH and
d-Phe3 CHH, guinea pig (gp)-anti-pQl and rabbit (rb)-anti-pQd, respectively (not shown) Examination
of chromatograms from different sinus gland batches shows a constant abundance ratio of the different pep-tides, with the VIH I peak being half the size of the VIH II peak (Fig 1)
The rationale of our experimental approach for identifying a putative d-residue in VIH was reliant on: (a) DAACPs being generally more hydrophobic than their l-counterparts in most DAACPs from eukaryotes known to date, including CHH, and (b) the d-residue being found near the N-terminal end, between the sec-ond and the fourth position of the sequence Conse-quently, we hypothesized that the d-residue was present in the N-terminal heptapeptide of VIH II, the hydrophobic form In a first attempt to identify the putative d-residue in VIH, we considered that, in CHH A and B, the d-amino acid is the phenylalanine
at position 3 [21] We assessed whether the nature of the modified residue (Phe3 in CHH and Phe5 in VIH)
or its position (Phe3 in CHH and Ala3 in VIH) may
be conserved in CHH and VIH Thus, heptapeptides corresponding to the N-terminal sequence of VIH were synthesized with different configurations (i.e Hep-l, Hep-dA3 and Hep-dF5) Samples of VIHs (ten sinus gland equivalents, i.e 50 pmol VIH I and
100 pmol VIH II) were cleaved with endoproteinase Asp-N and the resulting fragments were separated by RP-HPLC After fractionation of VIH I hydrolysate, one fragment was eluted with the same retention time
as the synthetic peptide Hep-l (Fig 2A) The mass of this fragment was determined by MALDI-TOF MS to
be 796 Da, which is a value identical to the mass of Hep-l Because no other predicted fragment from endoproteinase-Asp-N digestion of VIH displayed a similar mass, this fragment could be identified as the N-terminal heptapeptide of VIH I After similar cleav-age of VIH II by endo-Asp-N and RP-HPLC, MS analysis indicated that a peptide with a molecular mass
of 795.9 Da was eluted at 31.5 min (i.e later than Hep-l, Hep-dA3 and Hep-dF5; not shown)
In a next step, to determine which residue may be in the d-configuration in the fragments resulting from VIH II digestion, we considered that numerous DAACPs display a d-residue at position 2 (in snail excitatory peptides [31], in frog opioid peptides [1] and
in platypus venom [32]) In addition, the tryptophan residue at position 4 may be a good candidate because contryphan, a conotoxin isolated from gastropod venom has a d-Trp4 [11] To test these hypotheses, the synthetic peptides Hep-dS2 and Hep-dW4 were used
39
39 5
40 0.1 u AU
220 nm
CHH B
CHH A
D -Phe 3 CHH B
D -Phe 3 CHH A
49 53 Retention time (min)
VIH I VIH II
45
41
Fig 1 RP-HPLC profile of an acetic acid extract of 30 lobster sinus
glands Only the part of the chromatogram where CHHs and VIHs
are eluted is shown The nature of the ultraviolet absorbance peaks
was assessed by ELISA as well as by comparison with previously
published similar analyses [26].
Trang 4as standards Figure 2B shows that the elution time of
Hep-dW4 is identical to that of the VIH II fragment
By MALDI-TOF MS, the mass of this fragment was
found to be 795.8 Da, similar to the synthetic peptide
Moreover, co-injection on RP-HPLC of the VIH II
fragment and Hep-dW4 resulted in a single
symmetri-cal peak (not shown) Therefore, the results obtained
indicated that the two VIH isomers differed by the
configuration of the tryptophan at the fourth position
of the sequence and, consequently, they were named VIH and d-Trp4 VIH
This assumption was further confirmed by immuno-assays realized with the antisera directed against the N-terminal decapeptides of VIH with all l-residues [rat (r)-anti-l] or with the tryptophan residue in d-configu-ration (gp-anti-dW4) Indeed, ELISA using affinity-purified antisera with the different heptapeptides demonstrated a strict specificity of these antisera for the antigen peptide, with cross-reactivity for the other heptapeptides being lower than 1% (not shown) Accordingly, when ELISA was performed on fraction aliquots from RP-HPLC of sinus gland extract (as in Fig 1), no signal was obtained using these antisera with fractions corresponding to CHHs A and CHHs
B, whereas r-anti-l gave a strong signal with VIH I only and gp-anti-dW4 with VIH II exclusively, which confirmed unambiguously the presence of the d-Trp in VIH II (not shown)
Characterization of cell types in relation to CHH and VIH isomers
To study the distribution of CHHs and VIHs in the
X organ cells, immunohistochemical labelling of whole-mounts of X organ–sinus gland complexes was realized using different sets of antibodies In addition, immunogold labelling was performed on sinus gland sections to localize these peptides at the subcellular level, within secretory granules
Localization of VIH and D-Trp4 VIH in X organ neuroendocrine cells
Confocal analyses of whole-mounts of X organ–sinus gland complexes of the lobster H americanus were per-formed after double immunofluorescent labelling using purified antisera r-anti-l and gp-anti-dW4 Different cell types were observed: the larger neuroendocrine cell bodies (70 ± 7 lm diameter soma) were strongly labelled with gp-anti-dW4 (green cells; Fig 3A), with the labelling being cytoplasmic and granular; only some of these cells, with a smaller diameter (56 ± 7 lm) were also stained with r-anti-l, the yel-low⁄ orange colour, variable in a same organ, attesting
to labelling with both antisera For the sake of clarity, both types are subsequently referred to as d-VIH cells Smaller VIH-producing cells (31 ± 7 lm diameter soma) were grouped in a distinct region Their peri-karya were immunoreactive with r-anti-l exclusively (red cells called l-VIH cells) A total of 14 d-VIH cells (nine green and five yellow cell bodies) and 19 l-VIH cells (red soma) were counted per X organ Most
Hep- L
Hep- D A3
0.01AU
220 nm
Hep- D F5
34 36
32
Retention time (min)
Hep- D W4
36
Hep- L
Hep- D F5 Hep- D A3 Hep- D S2
0.01AU
220 nm 32 Acetonitrile (%)
34
Retention time (min)
26 30 34
A
B
Fig 2 (A) RP-HPLC profile of VIH I digest (ten sinus gland
equiva-lents) Only the part of the chromatogram where fragments elute is
shown The nature of the ultraviolet absorbance peaks was
assessed by comparison with retention times of standards (arrows)
(i.e heptapeptides Hep- L , Hep- D A3 and Hep- D F5) coupled with
MALDI-TOF mass analysis (B) RP-HPLC profile of VIH II digest
(ten sinus gland equivalents) Compared with the previous analysis
shown in Fig 2A, the synthetic heptapeptides Hep- D S2 and
Hep-D W4 were added to the standard mixture.
Trang 550 µm
A
50 µm
B
50 µm
R
LG
SG XO
ME MI MT H
20 µm
D
20 µm
E
50 µm
F
20 µm
G
50 µm
Fig 3 Confocal micrographs of double immunolabelled whole mounts of lobster X organ–sinus gland Images were collected as a focal series and processed to create 2D projections (single composite images) Central drawing: schematic representation of the lobster eyestalk nervous structures and the neuroendocrine complex (X organ–sinus gland) R, retina; LG, lamina ganglionaris; ME, medulla externa; MI, medulla interna; MT, medulla terminalis (A) General view of X organ labelled with r-anti- L and gp-anti- D W4 to visualize small L -VIH cells (red) and larger D -VIH cells (green or yellow) (B) Axonal arborizations of both cell types observed in the sinus gland (red for L -VIH varicosities and green for D -VIH ones) (C) Distribution of CHH cells in X organ showing green cell bodies ( L -CHH cells) labelled only with gp-anti-pQ L and orange somata ( D -CHH cells) corresponding to labelling with gp-anti-pQ L and rb-anti-pQ D antisera (D) Enlargement of axon terminals in the sinus gland showing both types of secretory granules corresponding to L -CHH cells (green) and D -CHH cells (red) (E) Immunolocalization of
D -Trp4 VIH and D -Phe3 CHH in the X organ where three cell types were observed : D -CHH cells (red), D -VIH cells (green) and D -cells produc-ing both D -isomers (orange) (F) Enlarged view of the three cell types showing the variations of coloration for D -cells: D -CHH cells (red, thin arrow), D -VIH cells (green, short arrow) and D -cells producing both D -isomers (orange, long arrows) (G) Sinus gland axonal arborizations containing D -Trp4 VIH (green) and D -Phe3 CHH (red) (H) Enlargement of axon terminals in the sinus gland with labelling as in (G).
Trang 6d- and l-VIH cells appeared segregated, whereas
several somas of the two types were dispersed in the
X organ (Fig 3A), most likely as a result of
arte-factual displacement of the cell bodies during the
prep-aration of the organ In the sinus gland, different types
of axonal arborizations containing VIH, d-Trp4 VIH,
or very rarely both, were observed (Fig 3B)
Ultrastructural observations after immunogold
label-ling of sinus gland sections reveal the presence of
sev-eral types of terminals differing by the morphology,
size and electron density of the secretory granules
After double labelling with r-anti-l and gp-anti-dW4,
particles were observed on two categories of axon
terminals: those exclusively labelled with r-anti-l
(l-VIH terminals; Fig 4A) and a few d-VIH terminals
labelled with gp-anti-dW4 (Fig 4B) No terminals with
double labelling were found In each terminal, the
secretory vesicles were densely packed and, although
scarce, the gold labelling was strictly restricted to them
Localization of CHH and D-Phe3 CHH in X organ
neuroendocrine cells
Whole-mounts of H americanus eyestalks were
incubated with antisera gp-anti-pQl and rb-anti-pQd,
recognizing CHH and d-Phe3 CHH, respectively
Confocal micrographs revealed two distinct cell types:
either green, labelled only with gp-anti-pQl (l-CHH
cells) or orange⁄ red, labelled with both gp-anti-pQl
and rb-anti-pQd antisera in variable proportions,
(d-CHH cells; Fig 3C) In a cluster of 34 cells, 13
d-CHH cells and 21 l-CHH cells were localized in
dis-tinct regions of the X organ The diameter of their cell
bodies was 64 ± 8 lm and 44 ± 7 lm, respectively
Intensively stained axons could be traced to the sinus
gland where two types of axonal arborizations were
observed, either green or red Rarely, orange
colour-ation attested the presence of both CHH types
(Fig 3D)
Accordingly, when sinus gland ultrathin sections were
observed after double labelling by gp-anti-pQl and
rb-anti-pQd, two different categories of terminals were
revealed: l-CHH terminals, labelled with gp-anti-pQl
(Fig 4C) and d-CHH terminals with a strong
rb-anti-pQd labelling (Fig 4D) Those terminals were also
rarely labelled with gp-anti-pQl (Fig 4E)
Distribution of D-Trp4 VIH and D-Phe3 CHH in X
organ neuroendocrine cells
To address the question whether d-epimers of CHH and
VIH are expressed in the same cells or not, confocal
analysis of whole-mounts of X organ–sinus gland
complexes was performed after double immunofluo-rescent labelling using specific antisera gp-anti-dW4 and rb-anti-pQd, recognizing d-Trp4 VIH and
d-Phe3 CHH, respectively Three types of cells could
be distinguished: seven green cells strongly labelled with the gp-anti-dW4 (56 ± 7 lm diameter soma; pre-viously called d-VIH cells), ten red cells immunoreac-tive with rb-anti-pQd antiserum (60 ± 5 lm diameter soma; previously called d-CHH cells) and five yel-low⁄ orange cells stained with both antisera, simply called d-cells (65 ± 7 lm diameter soma; Fig 3E) Among these latter cells, large differences in colour-ation were observed as a result of varicolour-ations in the relative amounts of both d-isomers (Fig 3F) Immunohistochemical staining of axonal arborization
in the neurohemal organ showed the three cell types with clustered granules immunoreactive either for one antiserum or for both antisera (Fig 3G,H)
To test the hypothesis of vesicular co-packaging of
d-epimers of CHH and VIH, double immunogold labelling with various associations of antibodies against the different forms was performed on ultrathin sections for examination by electron microscopy Using rb-anti-pQd and gp-anti-dW4, specific d-VIH and d-CHH terminals were observed (Fig 4F–I) Mixed terminals were also detected in other parts of the sinus gland, demonstrating that d-Trp4 VIH and
d-Phe3 CHH were not only colocalized in the same terminals (Fig 4J), but also in same secretory vesicles (Fig 4K) These three categories of terminals were usually found in different regions of the sinus gland,
as described above, but close juxtapositions of differ-ent terminals were sometimes observed (Fig 4G)
Discussion Although the existence of two VIH isoforms with iden-tical sequence, molecular mass and isoelectric point has been known for more than 15 years [22], the nat-ure of the difference between the two peptides had not yet been elucidated The demonstration of the presence
of a d-Phe3 residue in CHH A and B from the
H americanussome years ago [21] opened the possibil-ity that a d-residue may be present in one of the VIH isoforms as well Furthermore, in a previous study, ELISA experiments using specific antibodies have suggested the presence of a d-residue in VIH from
H gammarus [30]
In the present study, we have demonstrated, using a combination of RP-HPLC, MALDI-TOF MS (peptide mapping) and immunoassays, that the most hydropho-bic VIH isoform contains a d-tryptophanyl residue
at position 4, whereas a l-Trp4 is present in the
Trang 7500 nm
500 nm
500 nm
100 nm
500 nm
200 nm
C
F D
I
K
J E
Fig 4 Sections of double immunogold labelling of axon terminals in the lobster sinus gland (A, B) Double labelling with r-anti- L (10 nm gold particles) and gp-anti- D W4 (20 nm gold particles) antisera (A) Axon terminal containing secretion granules with only VIH (B) Axon terminal containing secretion with D -Trp4 VIH (C–E) Double labelling with gp-anti-pQ L (20 nm gold particles) and rb-anti-pQ D (10 nm gold particles) antisera (C) Two neighbouring CHH axon terminals (D) Axon terminal with only D -Phe3 CHH (E) Axon terminal showing some secretion granules labelled with both antisera (F–K) Double labelling with rb-anti-pQ D (10 nm gold particles) and gp-anti- D W4 (20 nm gold particles) (F) Axon terminal containing D -Trp4 VIH (G) Two neighbouring axon terminals, one containing D -Trp4 VIH and the other D -Phe3 CHH (H, I) Higher magnifications of D -Trp4 VIH and D -Phe3 CHH containing granules, respectively (J) General view of an axon terminal containing both
D -isomers (K) Enlarged view of an ending with secretion granules labelled with both rb-anti- pQ D and gp-anti- D W4 antisera.
Trang 8hydrophilic form The results obtained therefore
estab-lish that the lobster neuroendocrine system elaborates a
mixture of epimers of two different neurohormones:
CHH, d-Phe3 CHH, VIH and d-Trp4 VIH The
coex-istence, in a same organism, of several peptides with a
d-residue of different natures and positions has been
documented in venoms of cone snails [33] and of
platy-pus [32], and also in frog skin secretions [34]
Neverthe-less, no general consensus for the isomerization site,
nor for the nature of the surrounding residues, could
be demonstrated to date, with the prediction of single
d-amino acid presence being relevant exclusively within
a well-defined and restricted peptide family, as
illus-trated by a recent study of the I1-conotoxin superfamily
[35] In the case of the CHH family, the occurrence of
DAACPs appears to be problematic to predict because
CHH variants with a d-residue have been detected only
in Astacidea (rocky lobsters and crayfish), where the
N-terminal sequence is pGlu-Val-Phe-Asp-Glu-Ala
(with the d-residue always being the Phe3), and not in
CHH from other species with close overall sequence
similarity, such as the peneid shrimp Penaeus vannamei
with the sequence Ser-Leu-Phe-Asp-Pro-Ser [36],
nor even in those species with an identical sequence of
four residues, such as the spiny lobster Jasus lalandii:
Ala-Val-Phe-Asp-Glu-Ser [37] However, it is worth
noting that the presence of a d-amino acid in CHHs
from these last species may be overlooked because no
specific investigation procedure, such as amino acid
chiral analysis, was ever performed
Even if the presence of DAACPs begins to be well
documented in various animal groups, this
phenome-non has been rarely studied at the cellular level
Indeed, cellular aspects have been investigated in the
serous dermal gland of frogs However, owing to the
peculiar syncytial structure of this organ, the
localiza-tion of dermorphin and its all-l counterpart in the
dif-ferent cell compartments was difficult to analyze [34]
By contrast, in crustaceans, DAACPs are elaborated in
discrete, well-identified neuroendocrine cells located in
an easily accessible organ, with these characteristics
having already allowed detailed studies of their
bio-genesis [10,38–41]
In previous studies, the expression of CHH and VIH
in H americanus X organ–sinus gland complex was
investigated at the peptide level, but without
distinc-tion of epimers [42] It was observed by
immuno-histochemistry that some cells were labelled only by
anti-CHH or anti-VIH antisera, whereas several others
were reactive with both antisera, as an indication of
the presence of both hormones in these cells The
existence of mixed CHH⁄ VIH cells was later confirmed
at the mRNA level [26] When, at an early stage of
our study, we considered the cellular distribution of CHH and VIH isomers in the lobster X organ, one attractive simple hypothesis was that the mixed (double labelled) cells observed previously could con-stitute the synthesis site of the d-epimer of the two hormones (d-Phe3 CHH and d-Trp4 VIH), whereas the cells labelled exclusively by CHH or by anti-VIH would produce only the l-counterpart of each peptide To test this initial hypothesis, the organization
of CHH- and VIH- expressing cells, in relation to iso-mers, was investigated by immunohistochemistry and immunocytochemistry, using antisera specific to each epimer
Confocal examination of in toto preparations has shown that approximately 33 cells were immunostained with gp-anti-dW4 or r-anti-l antiserum, which agrees with the results obtained in a previously study [26], but diverges from those of other studies [42,43] report-ing a smaller figure (approximately 20 cells) It may be that the number of cells expressing VIH at a given time varies according to the reproductive stage of the animals, such as the haemolymphatic VIH level [44] Regarding the CHH-producing cells, our observa-tions indicate that their number (n = 34 soma were counted) is close to the number of VIH cells (n = 33), with a similar ratio between l-CHH cells (n = 21) and
d-CHH cells (n = 13) of approximately 1.6 versus 1.3 for VIH cells This figure fits well with the ratio of CHH⁄ d-Phe3 CHH quantified in the lobster sinus glands [45; present study], which is not the case of VIH isomers, as noted earlier On confocal mountings, subtle variations in colouration between d-CHH peri-karya were observed, although this much less pro-nounced than for VIH cells This agrees with the results obtained in the crayfish Orconectes limosus, showing that l-isomer of CHH is always present in the different parts of the d-CHH cells, in decreasing amounts from the cell body to the axon terminal in the sinus gland, as a result of late and progressive isomerization of the Phe3 of the CHH during the migration of the secretion vesicles along the axonal tract [39,41]
The immunohistochemical and immunocytochemical results obtained in the present study invalidate our starting hypothesis proposing that the cells coexpress-ing VIH and CHH in the lobster X organ, as observed
in previous studies [26,42], were actually producing the
d-isomers of both hormones Indeed, the lobster neu-roendocrine system is more complex than expected because, in addition to cells containing exclusively
l-isomers of CHH or VIH, three types of cells were found, producing: (a) d-Phe3 CHH (d-CHH cells), (b)
d-Trp4 VIH (d-VIH cells) or (c) a mixture of both
Trang 9peptides (d-cells) Overall, five cell types producing
CHH and VIH could be identified in the lobster X
organ (Fig 5) Nevertheless, it should be noted that
d-CHH cells, d-VIH cells and d-cells may correspond
to one single cell type producing a mixture of
d-Trp4 VIH and d-Phe3 CHH in different proportions
depending upon its physiological stage Indeed, in situ
hybridization in combination with
immunohistochem-istry revealed that strong immunostaining of CHH and
VIH may coincide with a weak or null mRNA
label-ling and vice versa [26] The existence of a sixth type
of cell, producing a mixture of l-epimers of CHH and
VIH had to been considered It was researched by
double labellings of successive ultrathin sinus glands
sections (not shown), although this proved to be in
vain Similarly, no terminals exhibited simultaneous
labelling for CHH and d-Trp4 VIH or VIH and
d-Phe3 CHH (not shown)
At present, it is not possible to assign a functional
significance to the colocalization of the CHH and VIH
epimers, especially because cellular colocalization of
different neurohormones of the CHH family is not a
general rule in the crustaceans By contrast to the
situ-ation in the lobster, VIH and CHH are synthesized
and released by different cellular types in woodlouse
[46] and in Norway lobster [47] Similarly,
moult-inhibiting hormone and CHH are localized in distinct
neurosecretory pathways in several crabs [48], whereas
colocalization of both peptides has been reported in
prawns [49]
The presence, in the same cells and even in the same
secretory granules, of peptides displaying a d-residue
of different nature (Phe or Trp) and at different
posi-tion (third or fourth) raises the quesposi-tion of the nature and characteristics of the enzyme(s) responsible for this modification Indeed, only a few peptide isomerases, exhibiting very different substrate specificity, have been isolated so far, from the funnel-web spider A aperta [50], the frog P sauvagei [15] and from platypus venom [12,16] Only structures of the first two are known, which appear to be totally unrelated because the frog enzyme presents similarities with the N-termi-nal H-domain of human IgG-Fc binding protein and the spider isomerase appears to belong to the serine protease family Obviously, the large structural and functional divergences between these enzymes impede any speculation about the characteristics of the puta-tive lobster isomerase(s) Nevertheless, the occurrence,
in lobster X organ cells, of two distinct enzymes with different substrate specificity and a variable expression pattern appears as a likely working hypothesis
Major challenges remaining for the future are the identification of the putative peptide isomerase(s) in crustaceans from these specialized cells from the X organ–sinus gland complex of the American lobster as well as the characterization of the receptors of CHH and VIH epimers, aiming to provide insights on the functional significance of the intriguing PTM that is
l- to d-aminoacyl isomerization
Experimental procedures
Animals and peptide purification
H americanus, weighing 300–500 g, were obtained from a commercial supplier (Metro, Bobigny, France) To reduce
Fig 5 General diagram of precursor pro-cessing of VIH and CHH isomers in relation
to the different cell types in X organ–sinus gland complex CPRP, CHH precursor-related peptide.aAmidation can be pre-, co- or post-cleavage of CPRP b Cylization of CHH N-terminus is optional (N-terminus unblocked CHH can be released) and, simi-lar to isomerization, it occurs after CPRP cleavage c By contrast to CHH, VIH is not N-terminal cyclized L -CHH and L -VIH cells secrete exclusively CHH and VIH, respec-tively, whereas D -CHH and D -VIH cells release mainly the D -isomer of the respec-tive hormone, in addition to a variable amount of L -isomer D cells secrete mainly the D -form of both CHH and VIH Besides isomerization, the same PTMs occur in every type of CHH or VIH cell.
Trang 10their stress, lobsters were maintained in the laboratory for
2 weeks before experimentation, in filtered and
re-circulat-ing artificial seawater at 13C In the aquarium, hollow
bricks provided shelters for the animals, which were fed
weekly with mussels The physiological status of the donors
(sex, moulting and reproductive stage) was not recorded
The animals were anaesthetized on crushed ice for 30 min
before dissection
VIH I and II were extracted from 30 H americanus sinus
glands (sinus gland equivalents, i.e 1.5 lg or 160 pmol
for VIH I and 2.7 lg or 280 pmol for VIH II) Peptides
were extracted and purified as described previously [10]
HPLC fractions from the elution zone of CHHs and VIHs
were collected manually to achieve optimal resolution
between CHH and VIH isomers, respectively HPLC
frac-tions containing VIH I and II were dried under vacuum
USA)
Enzymatic cleavage and digest fractionation
Purified and dried VIHs (30 sinus gland equivalents) were
redissolved in 5 lL of acetonitrile⁄ water (1 : 1, v ⁄ v) and
mixed with 0.4 lg of enzyme (Endoproteinase Asp-N
sequencing grade, EC 3.4.24.33; Roche Diagnostics, Baˆle,
Switzerland) in 50 lL of phosphate buffer (50 mm, pH 8)
After 22 h at 37C under stirring, the reaction was stopped
by adding 5 lL of acetic acid (2 m) Then, the digests of
each VIH were fractionated on a Nucleosil C18 (5 lm,
con-nected to the pump system and spectrophotometer Peptides
were eluted from the column by a gradient of acetonitrile
in water at a flow-rate of 0.2 mLÆmin)1 Both solvents
con-tained trifluoroacetic acid (0.1% in water and 0.08% in
ace-tonitrile) HPLC fractions from the elution zone of
synthetic peptides (see below) were collected manually and
dried under vacuum
Solid-phase peptide synthesis and antisera
production
Heptapeptides (Hep-l, Hep-dS2, Hep-dA3, Hep-dW4 and
Hep-dF5) with the sequence corresponding to the
N-termi-nus of lobster VIH and with all l-residues or a d-residue at
different positions (Table 2) were synthesized using
solid-phase FastMoc chemistry with a 433A Automated Peptide
Synthesizer (Applied Biosystems, Foster City, CA, USA) as
described previously [51] The homogeneity of the synthetic
peptides preparation was assessed by MALDI-TOF MS
and analytical RP-HPLC
Decapeptides named Dec-l and Dec-dW4 (Table 2),
cor-responding to the N-terminus of VIH I and II, respectively,
France) and injected, after coupling to keyhole limpet
haemocyanin, into rats (for Dec-l) or guinea pigs (for
Dec-dW4) to generate polyclonal antisera When necessary, specific IgG was purified according to a batch procedure that has been described previously [41]
MS
Positive-ion mass spectra were recorded in reflectron mode with a single stage reflectron matrix-assisted laser
spec-trometer (Voyager DE RP; Perseptive Biosystems Inc., Framingham, MA, USA) as described previously [51]
ELISA Specificity assays
Direct ELISA was performed to determine the specificity of antisera The wells of a plastic microtitre plate (Nunc, Ros-kilde, Denmark) were coated in triplicate with 100 ng of synthetic hepta- or decapeptides In addition to the antisera r-anti-l (made in rat) and gp-anti-dW4 (produced in guinea pig), two antisera discriminating CHH isomers (gp-anti-pQl made in guinea pig and rb-anti- pQd in rabbit) [38] and an antiserum raised against purified VIH II (i.e anti-4 pro-duced in guinea pig and recognizing both VIHs) [45] were used as primary antisera (Table 3) Secondary antibodies (anti-rat IgG, anti-guinea pig IgG and anti-rabbit IgG; all raised in goat and conjugated to alkaline-phosphatase; Sigma, Saint Louis, MO, USA) were used at 1 : 2000 dilu-tion (Table 3) Cross-reactivity of antisera between l- and
d-peptides was calculated as the ratio between absorbance values obtained with the cross-reacting and the immunogen peptides
Analysis of native VIH
Direct ELISA on RP-HPLC fractions from the elution zone
of CHHs and VIHs from lobster sinus glands was per-formed: 10 lL aliquots of each fraction were pipetted in triplicate into the wells of a microtitre plate and dried
Table 2 N-terminal amino acid sequence of vitellogenesis inhibit-ing hormone (VIH) and the synthetic peptides used in the present study D -residues are indicated by bold and underlined letters.