Báo cáo khoa học: Characterization of natural vasostatin-containing peptides in rat heart docx

To answer this question and explore the potential autocrine⁄ paracrine role of these peptides, we investigated the presence of endogenous N-terminal CGA-derived fragments in the rat hear

in rat heart

Elise Glattard1, Tommaso Angelone2, Jean-Marc Strub3, Angelo Corti4, Dominique Aunis1,

Bruno Tota2, Marie-He´le`ne Metz-Boutigue1and Yannick Goumon1

1 Inserm U575, Physiopathologie du Syste`me Nerveux, Strasbourg, France

2 Laboratory of Cardiovascular Physiology, Department of Cell Biology, University of Calabria, Arcavacata di Rende, Italy

3 CNRS, UMR 7512, Laboratoire de Spectrome´trie de Masse BioOrganique, Strasbourg, France

4 Department of Biological and Technological Research, San Raffaele H Scientific Institute, Milan, Italy

Chromogranin A (CGA) is the major member of a

family of acidic glycoproteins, named

chromogra-nins⁄ secretogranins, originally described in chromaffin

cell granules of bovine adrenal medulla [1,2] It is

pre-sent in numerous tissues [3], and is stored and released

together with neurotransmitters and hormones in the

nervous, endocrine and diffuse neuroendocrine systems

[2,4–7]

CGA is a well-conserved protein that is widely

distributed, from paramecia to mammals [2,8] It is

characterized by a high abundance (17%) of glutamic

acid residues and several dibasic sites in its sequence,

representing cleavage sites for endopeptidases (i.e

prohormone convertases) and carboxypeptidase E⁄ H [9–12], giving rise to various derived peptides display-ing numerous biological effects [7,13]

Post-translational modifications (phosphorylation and O-glycosylation) of bovine and human CGA have been identified [14–17] Phosphorylations are present all along the sequence, whereas glycosylation sites are mainly located in the median part of CGA

CGA and its derived fragments are known to be released into blood in response to stress, reaching sev-eral nanomolar concentrations in the periphsev-eral circu-lation of man [18,19] Of these fragments, bovine CGA(4–16) and CGA(47–60) have been shown to

Keywords

chromogranin A; heart; post-translational

modifications; rat; vasostatin

Correspondence

Y Goumon, Inserm U575, 5, rue Blaise

Pascal, F-67084 Strasbourg Cedex, France

Fax: +33 3 88 60 08 06

Tel: +33 3 88 45 67 24

E-mail: goumon@neurochem.u-strasbg.fr

(Received 17 February 2006, revised 18

May 2006, accepted 24 May 2006)

doi:10.1111/j.1742-4658.2006.05334.x

Chromogranin A (CGA) is a protein that is stored and released together with neurotransmitters and hormones in the nervous, endocrine and diffuse neuroendocrine systems As human vasostatins I and II [CGA(1–76) and CGA(1–113), respectively] have been reported to affect vessel motility and exert concentration-dependent cardiosuppressive effects on isolated whole heart preparations of eel, frog and rat (i.e negative inotropism and anti-adrenergic activity), we investigated the presence of vasostatin-containing peptides in rat heart Rat heart extracts were purified by RP-HPLC, and the resulting fractions analyzed for the presence of CGA N-terminal frag-ments using dot-blot analysis CGA-immunoreactive fractions were submit-ted to western blot and MS analysis using the TOF⁄ TOF technique Four endogenous N-terminal CGA-derived peptides [CGA(4–113), CGA(1–124), CGA(1–135) and CGA(1–199)] containing the vasostatin sequence were characterized The following post-translational modifications of these frag-ments were identified: phosphorylation at Ser96, O-glycosylation (trisaccha-ride, NAcGal-Gal-NeuAc) at Thr126, and oxidation at three methionine residues This first identification of CGA-derived peptides containing the vasostatin motif in rat heart supports their role in cardiac physiology by

an autocrine⁄ paracrine mechanism

Abbreviation

CGA, chromogranin A.

affect microbial growth and nociception [5,20–23].

Vasostatin-I [human CGA(1–76)] and its N-terminal

domain [CGA(1–40)] have also been reported to affect

the dilatation of vessels [24–28] probably via an

endo-thelium-independent vasoinhibitory action [26] In

addition, circulating CGA concentrations are increased

in cardiac pathology [29] In patients with chronic

heart failure, these high concentrations are correlated

with the severity of the disease, representing a

prognos-tic indicator of mortality [29] The recent finding that

CGA-knockout mice show hypertension and cardiac

enlargement [30] further stresses the importance of

CGA in cardiac physiology Cardiosuppressive effects

(i.e negative inotropy) of N-terminal CGA-derived

peptides have been well documented in vertebrate heart

[31] In particular, in the isolated working heart of the

frog Rana esculenta, the concentration-dependent

neg-ative inotropism and inhibition of the

b-adrenergic-dependent positive inotropy elicited by the N-terminal

domain of CGA [i.e human recombinant vasostatin I

and bovine CGA(7–57)], as well as by the synthetic

peptides corresponding to frog and bovine CGA(4–16),

CGA(47–66), and bovine CGA(1–40), have been

ana-lysed in details [32] More recently, using the

Lange-ndorff-perfused rat heart model, we have reported that

the two recombinant human vasostatins STA-CGA(1–

76) (vasostatin I) and STA-CGA(1–113) (vasostatin II)

display similar negative inotropic effects and

antagon-ize the b-adrenergic-dependent positive inotropism, the

latter being counteracted by vasostatin I in a

noncom-petitive type of antagonism [33]

An important question arising from these results is

whether cardiac tissues produce vasostatin-containing

peptides To answer this question and explore the

potential autocrine⁄ paracrine role of these peptides, we

investigated the presence of endogenous N-terminal

CGA-derived fragments in the rat heart Our studies

using biochemical techniques (RP-HPLC, dot-blot,

western blot) and MS analysis (TOF-TOF MS)

allowed us to investigate the N-terminal processing of

CGA in rat heart extracts Our data reveal the

pres-ence of several vasostatin I-containing and

vasosta-tin II-containing peptides in the CGA(1–199) domain,

strongly supporting their role as intracardiac regulators

of cardiac contractile performance

Results

Western blot analysis of N-terminal CGA-derived

peptides present in rat heart extract

Experiments were carried out to determine whether

acidic or heat treatment of rat heart was able to enrich

and⁄ or generate artificial CGA fragments (Experimen-tal procedures) Boiled and unboiled rat heart tissue extracts were prepared in 40% acetic acid or 0.1% (v⁄ v) trifluoroacetic acid and loaded on an SDS ⁄ 15% polyacrylamide gel to compare the fragmentation pat-tern of CGA A rat heart control extracted in the pres-ence of antiprotease cocktail and not treated with acid

or heat was also used Gels were electrotransferred to polyvinylidene difluoride membrane and submitted

to western blot analysis using a rabbit antibody to CGA(4–16) The electrophoretic profiles of the treated CGA fragments were identical with that of the non-treated extract, indicating that no additional fragment was generated by the treatments used here and that CGA fragments were enriched using these treatments (Fig 1) These results also indicate that 0.1% (v⁄ v) tri-fluoroacetic acid in water used for HPLC purification did not induce further hydrolysis (Fig 1)

CGA maturation in rat heart was compared with that in rat adrenal to determine whether CGA

Fig 1 Western blot analysis of CGA fragments present in different rat heart preparations The CGA-immunodetection pattern is shown for rat heart tissue prepared under five different conditions: untreated (no acid and no heating but addition of antiprotease cock-tail); in 40% acetic acid alone; in 40% acetic acid followed by heat-ing at 100
C for 3 min; in 0.1% trifluoroacetic acid; in 0.1% trifluoroacetic acid followed by heating at 100
C for 3 min Samples (20 lg) were separated by SDS ⁄ PAGE (15% gel) and electrotransferred to a polyvinylidene difluoride membrane Immuno-detection was carried out with the rabbit polyclonal antibody to bovine CGA(4–16).

maturation is tissue-dependent as reported in other

species [20] Rat heart extract, treated with acetic acid

and heated, was subjected to SDS⁄ PAGE (15% gel)

(Fig 2) Rat adrenal gland extract also treated with

acetic acid and heated was used as control In rat

heart, three groups of bands ranging from 17 to

80 kDa (Fig 2, lane 2 and labels) were identified: (a)

containing strongly labeled high-molecular-mass bands

(a, b); (b) containing other major bands (c, d, g); (c)

containing weakly labeled bands (e, f, h, i, j, k) Thus,

it appears that the a and b immunoreactive bands,

ran-ging from 80 to 50 kDa, correspond to the apparent

molecular mass of intact rat CGA (with and without

post-translational modifications), as well as to several

large N-terminal-derived fragments Such a broad

pat-tern is due to the presence of glutamic acid clusters

[e.g CGA(216–228)], known to modify the

electropho-retic migration of GGA [CGA(1–448), 48 kDa] with

the apparent molecular mass of 80 kDa (Fig 2, lane 2,

a label) [9] The second group containing bands c, d

and g at 45, 37 and 27 kDa, respectively, corresponds

to shorter N-terminal CGA-derived fragments The

third group is composed of weakly labeled bands (e, f,

h, i, j and k labels) which correspond to 32, 30, 24, 23,

18 and 17 kDa apparent molecular mass, respectively Interestingly, the 24-kDa and 23-kDa bands (h and i, respectively) probably correspond to vasostatin II [rat CGA(1–128)] previously reported to migrate with an apparent molecular mass of 21–23 kDa [34]

Purification and immunodetection of N-terminal CGA-derived fragments

Rat heart extract (2 mg) was purified using the RP-HPLC technique (Fig 3A) Each peak was col-lected and dotted on to nitrocellulose sheet before immunodetection with a rabbit polyclonal antibody to CGA(4–16) [35] Immunolabels were observed for frac-tions 1, 6 and 10–14 (Fig 3A)

To evaluate the apparent molecular mass of CGA N-terminal fragments present in these fractions, an ali-quot (2 : 5, v⁄ v) of each fraction (1–14) was subjected

to SDS⁄ PAGE (15% gel), electrotransferred to a poly(vinylidene difluoride) membrane, and immunode-tected with the CGA(4–16) antibody (Fig 3B) High-molecular-mass CGA-immunoreactive fragments (80–

70 kDa corresponding to the whole CGA with and without post-translational modifications and long C-terminal truncated protein) were recovered in the more hydrophobic fractions as reported for bovine CGA [9] Low-molecular-mass fragments (22–25 kDa

in Fig 3B) representing CGA(1–124⁄ 128 ⁄ 130) were recovered in fractions 1 and 14

CGA sequence comparison from several species The rat sequence contains three highly conserved amino-acid domains on comparison with bovine and human CGA (Fig 4, underlined) Thus, the N-ter-minal rat CGA(1–76) displays 85% similarity and 93% homology with the bovine CGA sequence, as well as, respectively, 86% and 94% with the human sequence (clustalw software [36]) The CGA(94–106) domain possesses 84% identity and 92% homology with bovine and human CGA sequences The C-terminal rat CGA(334–448) also displays 88% identity and 98% homology with bovine CGA as well as 91% iden-tity and 96% homology with human CGA

As CGA is known to have post-translational modifi-cations, we investigated whether the N-terminal domain of rat CGA could be modified The presence

of putative phosphorylations was examined using net-phos 2.0 software [37] Phosphorylations could poten-tially be present on several serine residues located at position 35, 50, 96, 133, 168, 182 and 185 Among these, Ser96 in rat CGA corresponds to a previously

Mr

Fig 2 Comparison of CGA fragments present in rat adrenal gland

and heart Samples were separated by SDS⁄ PAGE (15% gel) and

electrotransferred to a poly(vinylidene difluoride) membrane Lane

1, whole rat adrenal protein extract (50 lg); lane 2, rat heart extract

(20 lg) treated with 40% acetic acid + heating at 100
C for 3 min.

Western blot analysis, using polyclonal antibody to bovine CGA(4–

16) was performed to detect N-terminal CGA fragments.

described phosphorylated residue (Ser81) in the bovine

CGA sequence (Fig 5A) [17,38]

In addition, the netoglyc 3.1 software [39] allowed

us to predict potential O-glycosylation on residues

Ser119, Thr126, Thr173, Thr177, Thr181, Thr193 and

Thr194 Consensus N-glycosylation sites (N-X-T⁄ S)

[40] were also present on Asn107 and Asn175

(netn-glyc1.0; Fig 5A)

Characterization of N-terminal CGA-derived

fragments present in rat heart extract

TOF⁄ TOF MS was used because it is a direct, highly

sensitive and precise technique validated for the

prote-omic approach of peptides [41] By direct MS analysis

of the immunoreactive fractions, we determined the

experimental molecular masses of N-terminal

CGA-derived fragments, compared them with the theoretical

molecular mass, and evaluated possible

post-transla-tional modifications (Table 1) Phosphorylations,

gly-cosylations and oxidations were predicted using the

netphos 2.0 and netoglyc 3.1 software, as well as from previously described post-translational modifica-tions on CGA from other species (Fig 5A) Using this technique, phosphorylation was detected at fragments CGA(4–113), CGA(1–124), CGA(1–135) and CGA(1– 199), whereas O-glycosylation (trisaccharide, NAcGal-Gal-NeuAc) was only found at CGA(1–135) (Table 1) Oxidation could be expected at methionine residues 7,

15, 32, 162 and 163 (Figs 4 and 5A)

With this approach, it has been possible to charac-terize several fragments starting from residue 1 or 4 of CGA and ending at residues 113, 124, 135 and 199, corresponding to potential cleavage sites located at basic residues (Table 1 and Fig 5) The cleavage site Val3⁄ Asn4 had previously been reported by us [9]

Discussion

CGA and its derived fragments are present in large, dense core secretory vesicles of all endocrine and neuroendocrine tissues In chromaffin cells, CGA

A

B

M r

Fig 3 HPLC purification of CGA fragments from rat heart extract (A) 2 mg rat heart extract was fractionated on a Macherey– Nagel RP Nucleosil 300–5C-18 column (4 · 250 mm) Aliquots of each HPLC frac-tion (1–14) were dotted on to a nitrocellu-lose sheet and immunodetected with the polyclonal antibody to bovine CGA(4–16) Immunoreactive fractions are underlined on the chromatogram (B) Eluted fractions 1–14 were separated by SDS ⁄ PAGE (15% gel) and electrotransferred to a poly(vinylidene difluoride) membrane before immunodetec-tion with the polyclonal antibody to bovine CGA(4–16).

fragments are co-released by exocytosis with

catechol-amines in response to secretagogues [7] The natural

occurrence of CGA N-terminal peptides in rat

endo-crine tissues is supported by the demonstration of the

presence in the pituitary gland of both CGA and

vas-ostatin II (apparent molecular mass 21–23 kDa) using

gel filtration and western blot analysis with specific

N-terminal CGA-directed antibody [rat CGA(98–106)]

[34] Previous studies have shown, with no ambiguity,

the presence of CGA in secretory granules of rat atrial

myoendocrine cells [42] In fact, this

immunohisto-chemical study using electron microscopy has shown

colocalization of CGA and atrial natriuretic peptides

in rat heart cardiomyocytes In addition, a western

blot analysis performed on the secretory granules

isola-ted from rat atrial myoendocrine cells has revealed the

presence of different CGA-immunoreactive bands

CGA has also been detected in rat Purkinje fiber cells

of the conducting system, in both rat atrium and

vent-ricle, as well as in H9c2 rat cardiomyocytes [43] The

possibility that CGA-derived fragments could also

ori-ginate from nerve termini innervating the heart cannot

be excluded [44] However, direct demonstration of the

intracardiac processing of CGA has so far been

lack-ing Our data represent the first clear evidence that the

heart produces and processes vasostatin-containing

peptides This strongly suggests that these fragments

play a role in the autocrine⁄ paracrine regulation of

cardiac function, being directly involved in the ino-tropic modulation [33] CGA-derived peptides are stored with atrial natriuretic peptides in the secretory vesicles of rat cardiomyocytes [42] The actions of CGA-derived peptides and atrial natriuretic peptides may be closely integrated under normal [45] and stress-ful or pathophysiological conditions [46], but this remains to be clarified

Our results indicate that, among the CGA frag-ments, a major broad immunoreactive band at an apparent molecular mass of 80–50 kDa is present in rat heart extract (Fig 2, a and b) This high-molecu-lar-mass immunoreactive band results from both the whole CGA protein (with and without post-transla-tional modifications) and various long C-terminal trun-cated CGA fragments This suggests that, compared with the rat adrenal gland where almost no intact CGA is found, the maturation process looks incom-plete and specific to the heart (Fig 2, lane 1) It appears also that other low-molecular-mass fragments differ from those observed in adrenal extract except for the 27-kDa fragment (Fig 2, g label) Shorter frag-ments (Fig 2, h, i, j, k labels) exist in the heart and include the cardioactive motif (i.e the vasostatin I sequence or a portion of it) Under normal or abnor-mal (e.g stressful or pathophysiological) conditions in response to a specific stimulus-induced proteolytic acti-vation, an increase in lower-molecular-mass fragments,

Fig 4 CGA primary structure analysis Sequence comparison of CGA sequences from rat (r), bovine (b) and human (h) Underlined sequences correspond to highly conserved regions Basic residues are indicated in bold.

as well as production of new fragments, could be

induced Some of these fragments may also result from

extracellular processing, as reported for CGA in the

adrenal gland [9] In addition, the proteomic approach

using RP-HPLC (Fig 3) and MS techniques allowed

us to characterize three peptides starting at the first

N-terminal residue and one at the fourth residue of the

protein (Table 1 and Fig 5B) All these peptides

(Table 1, Figs 4 and 5) ended at a monobasic potential cleavage site that could be recognized by prohormone convertases and carboxypeptidases present in rat heart [47,48] The difference in the theoretical and experi-mental molecular mass showed that some of these fragments are phosphorylated, oxidized, and⁄ or glycos-ylated (Table 1) The presence of phosphorylation, pre-viously reported on Ser81 of bovine CGA, could be

Fig 5 Characterization of N-terminal CGA fragments present in rat heart extract (A) Sequences alignment of rat (r), bovine (b) and human (h) N-terminal CGA sequence Phosphorylations are marked in bold under-lined letters and O-glycosylations as a black round shape Localization of phosphorylation and glycosylation sites for bovine and human CGA are those reported in literature Cleavage sites are indicated by arrows (B) Schematic representation of the N-terminal rat CGA fragments, , phosphorylation;

O, O-glycosylation.

assigned to Ser96 as this site is well conserved

(Fig 5A) [17,38] and was predicted to be potentially

phosphorylated (netphos 2.0 software) In addition,

we observed molecular mass differences attributable to

O-glycosylation (NAcGal-Gal-NeuAc), probably at

Thr126 (Fig 5A) according to O-glycosylation site

pre-dictions (netoglyc 3.1 software) This modification

was only observed on CGA(1–135) and not shorter

fragments [i.e CGA(1–124) and CGA(4–113)], thus

protecting the long fragment from proteolytic

degrada-tion CGA(1–135) is of particular interest because it is

found in various forms, i.e unmodified, oxidized,

phosphorylated and⁄ or O-glycosylated

(NAcGal-Gal-NeuAc; Table 1)

Like other chromogranin family members, CGA is a

protein precursor that is actively processed to

low-molecular-mass bioactive peptides [49] Rat CGA is a

448-amino-acid protein with a pI of 4.5–5.0 containing a

disulfide bridge and displaying numerous monobasic

(40) and dibasic (nine) residues (Fig 4) that could

repre-sent potential cleavage sites for proteolytic subtilisin-like

and trypsin-like enzymes including PC1⁄ 3, PC2, PC4,

PC5⁄ 6, PC7 ⁄ 8, furine ⁄ SPC1 ⁄ PACE [50], as well as

carboxypeptidases [51] In rat heart, the detected

PC1⁄ 3, PC2 and carboxypeptidase H ⁄ E [47,48,52] might

be involved in the CGA maturation process We have

previously characterized intragranular and extracellular

processing of CGA in chromaffin granules from bovine

adrenal medulla, showing that the processing starts

at both the N-terminus and the C-terminus of the

pro-tein [9] Among the fragments generated, CGA(1–76)

represents the major product of proteolytic

process-ing in bovine adrenal medulla [9,53], whereas the first

N-terminal cleavage product of rat CGA is b-granin

[rCGA(1–128)], corresponding to vasostatin II, due to

the lack of the first dibasic site [34,54,55] (Fig 4)

As CGA fragments are likely to be secreted by car-diomyocytes, extracellular processing can be expected

to occur in the secreted medium, as shown for CGA and proenkephalin-A secreted from chromaffin cells [9,56] The presence of extracellular proteases both on cardiomyocyte cell membranes and in the extracellular matrix suggests that extracellular processing does occur, as proposed for angiotensin II In the heart, an-giotensin converting enzyme and⁄ or renin, which are present on cell membranes, are involved in the conver-sion of angiotensinogen into angiotensin II [57] Because of the presence of extracellular proteases dur-ing tissue homogenization (for review, see [58]), we cannot exclude the possibility that CGA-derived frag-ments are extracellulary processed Such a mechanism may be involved in the production of additional N-ter-minal CGA-derived peptides displaying biological activity

Glycosylation of CGA has been reported to be involved in its 3D folding, its protection against prote-ase activity [9,15], as well as in trafficking [59] Glyco-sylation of the bovine CGA sequence has also been reported to modulate the antimicrobial activity of the chromacin [CGA(173–194)] fragment [60] The present data reveal that several rat heart CGA-derived pep-tides possess O-glycosylation sites The presence of multiple forms of the CGA(1–135) fragment is intrigu-ing and suggests an important role for the post-trans-lational modifications of this peptide, as previously reported for chromacin [61]

Phosphorylation of bovine and human CGA has been extensively studied In the former, phosphoryla-tion has been found at serine residues 81, 124, 297,

307, 372 and 376, as well as at Tyr173 (Fig 5B) [17,60,61] In the human protein, phosphorylation has been reported at serine residues 200, 252 and 315

Table 1 MS analysis of HPLC fractions CGA-derived peptides were identified by comparing the experimental mass obtained by TOF ⁄ TOF

MS analysis with the calculated one Proteolytic cleavage sites are indicated O, Oxidation; Na + , sodium adducts; O-Gly, O-glycosylation;

P, phosphorylation.

HPLC

fraction

Molecular mass (Da)

Location

Cleavage

(Fig 5B) [14] Phosphorylation of CGA is likely to be

related not only to the type of organ and the

physiolo-gical state, but also to peptide activity For example,

phosphorylation is known to modulate the

antimicro-bial activity of chromacin [60] and other intragranular

peptides such as enkelytin derived from

proenkephalin-A [bisphosphorylated PEproenkephalin-A(209–237)] [62]

Our results also indicate the presence of oxidized

fragments Oxidation of CGA residues has previously

been described [61] In chromaffin granules, for

instance, MS analysis has revealed oxidation at

methi-onine residues, which may be related to a redox

mech-anism inside the granules [63]

Taken together, our data clearly indicate that, in rat

heart, CGA is a precursor to numerous peptides that

may control different physiological functions The

con-cept of the heart as an endocrine organ was firmly

established almost 25 years ago with the discovery of

atrial natriuretic peptides present in the ‘atrial specific

secretory granules’ and identified as homeostatic agents

for protection against plasma volume overload ([64,65]

and references therein) The present study points to the

vasostatin-derived peptides as novel intracardiac

mod-ulators exerting a counter-regulatory inotropic role

particularly against b-adrenergic-elicited positive

ino-tropism [27,32,33,66] through an autocrine and⁄ or

paracrine mechanism

Experimental procedures

Rat heart extract

Male Wistar rats (Charles River Laboratories, Les Oncins,

Italy S.p.A) weighing 250–350 g were housed three per cage

in a ventilated cage rack system under standard conditions

Animals had food and water access ad libitum Animal care,

killing, and experiments were carried out according to the

European Community guiding principles in the care and

use of animals, and the projects were supervised by the

local ethics committee Whole hearts (atria and ventricles)

of male Wistar rats were removed just after death and

washed with NaCl⁄ Pi buffer to remove blood cells They

were then homogenized in the presence of 2 mL 40% (v⁄ v)

acetic acid in water with a Polytron mixer (Richmond

Agencies, Wigan, UK) to give 0.5 g fresh tissue per mL

The protein extract was centrifuged at 8000 g for 30 min at

4
C, and the supernatant containing the acid-stable

pro-teins was collected and boiled for 3 min to isolate the

ther-mostable chromogranins The extract was then centrifuged

at 12 000 g for 30 min to enrich the supernatant in

heat-stable proteins In some experiments, protease inhibitors

were included, but their presence did not modify the

pat-tern of proteolytic fragments Control experiments, using

different extraction conditions, were performed to deter-mine if the acetic acid (40%, v⁄ v; Arcos, Fairlaw, NJ, USA), trifluoroacetic acid (0.1% v⁄ v; HPLC conditions; Sigma Aldrich, Steinheim, Germany) or heat treatment (boiling for 3 min) were responsible for the generation of new CGA fragments An extraction performed in the pres-ence of an antiprotease cocktail that inhibits most of the proteases (except metalloproteases and aspartic proteases; Complete mini cocktail, Roche Diagnostics, Mannheim, Germany), in the absence of acid and without heating was performed as a control

Rat adrenal glands were homogenized in the presence of

2 mL 40% (v⁄ v) acetic acid The protein extract was centri-fuged at 8000 g for 30 min at 4
C The resulting superna-tant was boiled for 3 min and centrifuged at 12 000 g (30 min) The supernatant was used as control

After protein quantification using the Bradford technique (Protein assay; Bio-Rad, Marnes la Coquette, France), the extract was diluted in an adequate volume of ultrapure water, and aliquots of 20 lg protein were stored at)20
C

Western blot analysis Rat heart extract or HPLC fractions were separated by SDS⁄ PAGE (15% acrylamide; Euromedex, Souffelweyers-heim, France) Proteins were electrotransferred (45 min,

75 V) to polyvinylidene difluoride membrane (Amersham Biosciences, Uppsala, Sweden) and immunodetected with specific rabbit polyclonal antibodies raised against bovine CGA(4–16) fragment (1 : 1000 dilution) [53] A goat anti-rabbit IgG conjugated to horseradish peroxidase was used

as a secondary antibody (1 : 10 000 dilution; Sigma-Ald-rich) The poly(vinylidene difluoride) membrane was treated

as previously described [67], and immunodetection was per-formed with the SuperSignal West Dura Extended Dur-ation Substrate kit (Pierce, Perbio Science, Brebie`res, France) according to the manufacturer’s instructions

Dot-blot analysis

To detect the presence of CGA fragments in RP-HPLC fractions, aliquots (1 : 5, v⁄ v) were applied to nitrocellulose membrane (Amersham Biosciences) for dot-blot analysis The membrane was then treated as described above for western blot analysis

Purification of CGA-derived peptides by RP-HPLC CGA-derived peptides present in rat heart extracts were purified using an A¨kta Purifier HPLC system (Amersham Biosciences) and a Nucleosil reverse-phase 300–5C18 col-umn (4· 250 mm; particle size 5 lm, porosity 300 A˚; Macherey-Nagel, Hoerdt, France) [53] Absorbance was monitored at 214 nm, and the solvent system consisted of

0.1% (v⁄ v) trifluoroacetic acid in water (solvent A) and

0.09% (v⁄ v) trifluoroacetic acid in 70% acetonitrile (Carlo

Erba, Rodano Mi, Italy) in water (solvent B) Elutions were

performed at a flow rate of 700 lLÆmin)1using the gradient

indicated on the chromatograms

MS analysis

Mass spectra were acquired using an UltraflexTM TOF⁄

TOF mass spectrometer (Bruker Daltonik GmbH, Bremen,

Germany) with gridless ion optics under the control of

Flexcontrol 2.0 [41] This instrument, equipped with

SCOUTTM high-resolution optics with X-Y multisample

probe and gridless reflector, was used at a maximum

accel-erating potential of 25 kV and operated in reflector mode

for MS analysis Ionization was accomplished with a

337-nm beam from a nitrogen laser with a repetition rate of

20 Hz The output signal from the detector was digitized

at a sampling rate of 2 GHz The samples were prepared

by standard dried-droplet preparation on stainless-steel

MALDI targets using 2,5-dihydroxybenzoic acid as matrix

The external calibration of MALDI mass spectra was

per-formed using singly charged monoisotopic peaks of a

mix-ture of bradykinin(1–7) (m⁄ z 757.400), human angiotensin

II (m⁄ z 1046.542), human angiotensin I (m ⁄ z 1296.685),

substance P (m⁄ z 1347.735), bombesin (m ⁄ z 1619.822),

renin (m⁄ z 1758.933), ACTH(1–17) (adrenocorticotropic

hormone; m⁄ z 2093.087) and ACTH(18–39) (m⁄ z

2465.199)

Acknowledgements

This work was funded by Inserm, the University

Louis-Pasteur of Strasbourg, the French Ministe`re

De´le´gue´ a` la Recherche et a` l’Enseignement Supe´rieur

(Ph.D grant to EG), the Fondation pour la Recherche

Me´dicale (to MHMB), the Ligue Contre le Cancer (to

DA), the Programme Hospitalier de Recherche

Cli-nique 3150 (to MHMB), the Egide Program (Galile´e,

to MHMB) and Vinci Program (to TA) We thank N

Aslan and R Lang for technical assistance

References

1 Helle KB (1968) The chromogranin of the adrenal

medulla: a high-density lipoprotein Biochem J 109,

43P–44P

2 Simon JP & Aunis D (1989) Biochemistry of the

chro-mogranin A protein family Biochem J 262, 1–13

3 Iacangelo A, Okayama H & Eiden LE (1988) Primary

structure of rat chromogranin A and distribution of its

mRNA FEBS Lett 227, 115–121

4 Aunis D (1998) Exocytosis in chromaffin cells of the

adrenal medulla Int Rev Cytol 181, 213–320

5 Aunis D & Metz-Boutigue MH (2000) Chromogranins: current concepts Structural and functional aspects Adv Exp Med Biol 482, 21–38

6 Winkler H & Fischer-Colbrie R (1992) The chromogra-nins A and B: the first 25 years and future perspectives Neuroscience 49, 497–528

7 Helle KB (2004) The granin family of uniquely acidic proteins of the diffuse neuroendocrine system: compar-ative and functional aspects Biol Rev Camb Philos Soc

79, 769–794

8 Rieker S, Fischer-Colbrie R, Eiden L & Winkler H (1988) Phylogenetic distribution of peptides related to chromogranins A and B J Neurochem 50, 1066–1073

9 Metz-Boutigue MH, Garcia-Sablone P, Hogue-Angeletti

R & Aunis D (1993) Intracellular and extracellular pro-cessing of chromogranin A Determination of cleavage sites Eur J Biochem 217, 247–257

10 Mains RE, Zhou A & Parkinson D (1996) The biosyn-thetic processing and secretion of endogenous carboxy-peptidase H in mouse pituitary cells Ann N Y Acad Sci

805, 10–18

11 Doblinger A, Becker A, Seidah NG & Laslop A (2003) Proteolytic processing of chromogranin A by the pro-hormone convertase PC2 Regul Pept 111, 111–116

12 Eskeland NL, Zhou A, Dinh TQ, Wu H, Parmer RJ, Mains RE & O’Connor DT (1996) Chromogranin A processing and secretion: specific role of endogenous and exogenous prohormone convertases in the regulated secretory pathway J Clin Invest 98, 148–156

13 Iacangelo AL & Eiden LE (1995) Chromogranin A: cur-rent status as a precursor for bioactive peptides and a granulogenic⁄ sorting factor in the regulated secretory pathway Regul Pept 58, 65–88

14 Gadroy P, Stridsberg M, Capon C, Michalski JC, Strub

JM, Van Dorsselaer A, Aunis D & Metz-Boutigue MH (1998) Phosphorylation and O-glycosylation sites of human chromogranin A (CGA79-439) from urine of patients with carcinoid tumors J Biol Chem 273, 34087–34097

15 Strub JM, Garcia-Sablone P, Lonning K, Taupenot L, Hubert P, Van Dorsselaer A, Aunis D & Metz-Boutigue

MH (1995) Processing of chromogranin B in bovine adrenal medulla Identification of secretolytin, the endo-genous C-terminal fragment of residues 614–626 with antibacterial activity Eur J Biochem 229, 356–368

16 Bauer SH, Zhang XY, Liang F, De Potter WP, Claeys

M & Przybylski M (1997) Isolation and identification of intact chromogranin A and two N-terminal processing products, vasostatin I and II, from bovine adrenal medulla chromaffin granules by chromatographic and mass spectrometric methods Neuropeptides 31, 273–280

17 Bauer SH, Zhang XY, Van Dongen W, Claeys M & Przybylski M (1999) Chromogranin A from bovine adrenal medulla: molecular characterization of

glycosylations, phosphorylations, and sequence

hetero-geneities by mass spectrometry Anal Biochem 274, 69–

80

18 Nobels FR, Kwekkeboom DJ, Bouillon R & Lamberts

SW (1998) Chromogranin A: its clinical value as marker

of neuroendocrine tumours Eur J Clin Invest 28, 431–

440

19 Spadaro A, Ajello A, Morace C, Zirilli A, D’Arrigo G,

Luigiano C, Martino F, Bene A, Migliorato D, Turiano

S, et al (2005) Serum chromogranin-A in hepatocellular

carcinoma: diagnostic utility and limits World J

Gast-roenterol 11, 1987–1990

20 Dillen L, Miserez B, Claeys M, Aunis D & De Potter

W (1993) Posttranslational processing of proenkephalins

and chromogranins⁄ secretogranins Neurochem Int 22,

315–352

21 Ghia JE, Crenner F, Metz-Boutigue MH, Aunis D &

Angel F (2004) The effect of a chromogranin A-derived

peptide (CgA4–16) in the writhing nociceptive response

induced by acetic acid in rats Life Sci 75, 1787–1799

22 Ghia JE, Crenner F, Metz-Boutigue MH, Aunis D &

Angel F (2004) Effects of a chromogranin-derived

pep-tide (CgA 47–66) in the writhing nociceptive response

induced by acetic acid in rats Regul Pept 119, 199–207

23 Lugardon K, Chasserot-Golaz S, Kieffer AE,

Maget-Dana R, Nullans G, Kieffer B, Aunis D &

Metz-Bouti-gue MH (2002) Structural and biological

characteriza-tion of chromofungin, the antifungal chromogranin A

(47–66)-derived peptide Ann N Y Acad Sci 971, 359–

361

24 Aardal S, Galindo E, Aunis D & Helle KB (1993)

Human chromostatin inhibits endothelin-1-induced

con-tractures in human blood vessels Regul Pept 47, 25–32

25 Angeletti RH, Aardal S, Serck-Hanssen G, Gee P &

Helle KB (1994) Vasoinhibitory activity of synthetic

peptides from the amino terminus of chromogranin A

Acta Physiol Scand 152, 11–19

26 Brekke JF, Osol GJ & Helle KB (2002) N-terminal

chromogranin-derived peptides as dilators of bovine

coronary resistance arteries Regul Pept 105, 93–100

27 Corti A, Mannarino C, Mazza R, Angelone T, Longhi

R & Tota B (2004) Chromogranin A N-terminal

frag-ments vasostatin-1 and the synthetic CGA 7–57 peptide

act as cardiostatins on the isolated working frog heart

Gen Comp Endocrinol 136, 217–224

28 Mandala M, Brekke JF, Serck-Hanssen G,

Metz-Bou-tigue MH & Helle KB (2005) Chromogranin A-derived

peptides: interaction with the rat posterior cerebral

artery Regul Pept 124, 73–80

29 Ceconi C, Ferrari R, Bachetti T, Opasich C, Volterrani

M, Colombo B, Parrinello G & Corti A (2002)

Chromo-granin A in heart failure; a novel neurohumoral factor

and a predictor for mortality Eur Heart J 23, 967–974

30 Mahapatra NR, O’Connor DT, Vaingankar SM, Hikim

AP, Mahata M, Ray S, Staite E, Wu H, Gu Y, Dalton

N, et al (2005) Hypertension from targeted ablation of chromogranin A can be rescued by the human ortholog

J Clin Invest 115, 1942–1952

31 Tota B, Imbrogno S, Mannarino C & Mazza R (2004) Vasostatins and negative inotropy in vertebrate hearts Curr Med Chem 4, 195–201

32 Tota B, Mazza R, Angelone T, Nullans G, Metz-Boutigue MH, Aunis D & Helle KB (2003) Peptides from the N-terminal domain of chromogranin A (vasostatins) exert negative inotropic effects in the isolated frog heart Regul Pept 114, 123–130

33 Cerra MC, De Iuri L, Angelone T, Corti A & Tota B (2006) Recombinant N-terminal fragments of chromo-granin-A modulate cardiac function of the Langendorff-perfused rat heart Basic Res Cardiol 101, 43–52

34 McVicar CM, Cunningham RT, Harriott P, Johnston

CF, Buchanan KD & Curry WJ (2001) Analysis of the post-translational processing of chromogranin A in rat neuroendocrine tissue employing an N-terminal site-specific antiserum J Neuroendocrinol 13, 588–595

35 Lugardon K, Chasserot-Golaz S, Kieffer AE, Maget-Dana R, Nullans G, Kieffer B, Aunis D & Metz-Boutigue MH (2001) Structural and biological characterization of chromofungin, the antifungal chro-mogranin A-(47–66)-derived peptide J Biol Chem 276, 35875–35882

36 Higgins DG (1994) CLUSTAL V: multiple alignment of DNA and protein sequences Methods Mol Biol 25, 307–318

37 Blom N, Gammeltoft S & Brunak S (1999) Sequence and structure-based prediction of eukaryotic protein phosphorylation sites J Mol Biol 294, 1351–1362

38 Zhang X, Dillen L, Bauer SH, Van Dongen W, Liang

F, Przybylski M, Esmans E, De Potter WP & Claeys M (1997) Mass spectrometric identification of phosphory-lated vasostatin II, a chromogranin A-derived protein fragment (1–113) Biochim Biophys Acta 1343, 287–298

39 Hansen JE, Lund O, Tolstrup N, Gooley AA, Williams

KL & Brunak S (1998) NetOglyc: prediction of mucin type O-glycosylation sites based on sequence context and surface accessibility Glycoconj J 15, 115–130

40 Kornfeld R & Kornfeld S (1985) Assembly of aspara-gine-linked oligosaccharides Annu Rev Biochem 54, 631–664

41 Morelle W, Slomianny MC, Diemer H, Schaeffer C, van Dorsselaer A & Michalski JC (2004) Fragmentation characteristics of permethylated oligosaccharides using a matrix-assisted laser desorption⁄ ionization two-stage time-of-flight (TOF⁄ TOF) tandem mass spectrometer Rapid Commun Mass Spectrom 18, 2637–2649

42 Steiner HJ, Weiler R, Ludescher C, Schmid KW & Winkler H (1990) Chromogranins A and B are co-localized with atrial natriuretic peptides in secretory granules of rat heart J Histochem Cytochem 38, 845– 850