Proteolytic processing pattern of the endothelin-1 precursor in vivo

Peptides 26 (2005) 2482–2486

Proteolytic processing pattern of the endothelin-1 precursor in vivo Joachim Struck ∗ , Nils G. Morgenthaler, Andreas Bergmann Department of Research, BRAHMS Aktiengesellschaft, Neuendorfstr. 25, D-16761 Hennigsdorf, Germany Received 19 April 2005; received in revised form 12 May 2005; accepted 12 May 2005 Available online 22 June 2005

Abstract Endothelin-1 (ET-1) is a potent vasoconstrictor, which has been implicated in diseases involving dysfunctions of the cardiovascular system. For the biogenesis of ET-1, a larger precursor peptide (proET-1) is cleaved at two sites to give rise to bigET-1, which is subsequently cleaved to generate mature ET-1. In the present study, we investigated, which other peptides are derived from proET-1 in vivo. Six sandwich immunoassays covering various regions of proET-1 were developed and used to detect circulating proET-1 immunoreactivities in plasma of healthy subjects and septic patients. With this approach we could (a) demonstrate that, in addition to bigET-1/ET-1, three stable proET-1 fragments are generated, (b) exclude two previously discussed regions as sites for prohormone conversion and (c) show that the proteolytic processing pattern of proET-1 is unchanged under pathological conditions, which are associated with elevated levels of proET-1 fragments. The high stability and similarity in concentration of the proET-1 fragments suggest that these might be non-functional in the circulation. Stable proET-1 fragments maybe used in the future as reliable diagnostic targets to indirectly assess the release of ET-1, which might help to more selectively direct therapeutic measures. © 2005 Elsevier Inc. All rights reserved. Keywords: Sepsis; Shock; Prohormone; Biomarker; Cardiovascular dysfunction; Endothelium

1. Introduction Endothelin-1 (ET-1), a 21 amino acid peptide, is the most abundant member of a family of endothelins (ET-1, ET-2 and ET-3). It is derived mainly from vascular endothelial cells and acts in an autocrine or paracrine manner as a vasoconstrictor on the vasculature [13]. ET-1 is involved in a number of physiological and pathological processes, including the vascular changes associated with pulmonary hypertension and sepsis [2,8,9,14]. ET-1 originates from a larger precursor peptide (pre-proET-1; 212 amino acids), after removal of the signal sequence it is further proteolytically processed at specific basic residues leading to the production of bigET-1 (38 amino acids). ET-1 is finally excised from bigET-1 by the action of endothelin-coverting enzyme (ECE-1) [26]. A function of the peptide moieties flanking bigET-1 within proET-1 is not known. In vitro evidence suggests that furin ∗

Corresponding author. Tel.: +49 3302 883 797; fax: +49 3302 883 621. E-mail address: [email protected] (J. Struck).

0196-9781/$ – see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2005.05.010

as well as PC7 can cleave proET-1 at various sites [4,5], but it is not known, whether this also occurs in vivo, and whether additionally other proteases might act on these peptide moieties. Here, we describe the generation of antibodies with epitopes scattered throughout the entire sequence of proET-1 and the use of several antibody combinations in sandwich immunoassays to characterize the nature and stability of fragments of proET-1 in the circulation of healthy subjects and sepsis patients.

2. Materials and methods 2.1. Chemicals If not stated otherwise, chemicals were obtained at p.a. grade from Merck, Darmstadt, Germany. Bovine Serum Albumin and unspecific sheep and rabbit IgG were from SIGMA, Deisenhofen, Germany.

J. Struck et al. / Peptides 26 (2005) 2482–2486

2.2. Peptides Six peptides related to proET-1 were chemically synthesized, purified and qualitiy-controlled by standard procedures (JERINI AG, Berlin, Germany). The peptides were: PAV15 (sequence CAPETAVLGAELSAV, representing positions 18–31 of pre-proET-1, plus an N-terminal cystein residue), MPGC22 (sequence GENGGEKPTPSPPWRLRRSKRC, representing positions 32–53 of pre-proET-1), PLC16 (sequence LENLLPTKATDRENRC, representing positions 94–109 of pre-proET-1), PEC13 (sequence EKDWNNHKKGKDC, representing positions 136–148 of preproET-1), PCT15 (sequence CRSSEEHLRQTRSET, representing positions 168–181 of pre-proET-1, plus an N-terminal cystein residue) and PCW14 (sequence CSRERYVTHNRAHW, representing positions 200–212 of pre-proET-1, plus an N-terminal cystein residue). Recombinant full-length proET-1 (representing positions 18–212 of pre-proET-1) in form of a crude extract of total soluble protein from a transformed E. coli strain was obtained from InVivo GmbH, Hennigsdorf, Germany.

2483

serum (SIGMA, Deisenhofen, Germany) served as standards, and arbitrary units were assigned. The immunoassays were performed by incubating 50 ␮l of samples/standards and 200 ␮l labeled tracer antibody (106 relative light units per determination in each assay) in coated tubes under agitation for 2 h at 22 ◦ C. Tubes were washed four times with LUMItest wash solution (BRAHMS Aktiengesellschaft, Hennigsdorf, Germany), and bound chemiluminescence was measured with a LB952T luminometer (Berthold, Wildbad, Germany). The lower detection limits of the assays were determined as the value calculated for two standard deviations above the mean of a zero-standard 10-fold replicate. 2.5. Plasma samples EDTA-plasma samples were collected from apparently healthy control subjects as well as patients with severe sepsis or septic shock (as defined by the ACCP/SCCM consensus conference [1]) following ethical guidelines, and were stored at −20 ◦ C until further use.

2.3. Antibodies 3. Results Rabbit antisera against peptides PAV15, MPGC22, PLC16 and PEC13, and sheep antisera directed against peptides PCT15 and PCW14 were generated, and antibodies were purified according to previously applied protocols [15]. Rabbit antisera were obtained from InVivo GmbH, Hennigsdorf, Germany and sheep antisera from Micropharm Ltd., Carmarthenshire, UK. Briefly, peptides were conjugated via m-maleimidobenzoyl-N-hydroxysuccinimid ester (MBS) to keyhole limpet hemocyanin (KLH), and animals were immunized with the conjugates. Peptide-specific antibodies were purified from antisera by affinity chromatography employing SulfoLink gel (PIERCE, Rockford, IL, USA) derivatized with peptides PAV15, MPGC22, PLC16, PEC13, PCT15 and PCW14, respectively. 2.4. Immunoassays Six chemiluminescence label-coated tube-based sandwich immunoassays were set up similarly to a previously developed immunoassay [15]. The assays were named according to the positions of the N- and C-terminal amino acid within pre-proET-1, which are contained in the peptides used for the generation of the antibodies: assay 18–53 (anti-PAV15 antibody: solid phase and anti-MPGC22 antibody: tracer), assay 32–109 (anti-MPGC22 antibody: tracer and anti-PLC16 antibody: solid phase), assay 32–181 (anti-MPGC22 antibody: tracer and anti-PCT15 antibody: solid phase), assay 94–148 (anti-PLC16 antibody: solid phase and anti-PEC13 antibody: tracer), assay 136–181 (anti-PEC13 antibody: tracer and anti-PCT15 antibody: solid phase) and assay 168–212 (antiPCT15 antibody: solid phase and anti-PCW14 antibody: tracer). Dilutions of recombinant proET-1 in normal horse

Polyclonal antibodies against six epitopes of proET-1 were generated, purified and used to develop sandwich immunoassays covering the following regions of pre-proET1: positions 18–53, 32–109, 32–181, 94–148, 136–181 and 168–212 (Fig. 1). Using recombinant full-length proET-1 as standard material, we first demonstrated that all assays were technically able to detect proET-1 immunoreactivity (Table 1). The assays exhibited similar lower detection limits, ranging from 1 to 8 arb. U/l (Table 1). When plasma samples of healthy subjects (n = 10) and of septic patients (n = 20) were assessed for proET-1 immunoreactivity, three assays (assays 32–109, 32–181 and 136–181) failed to detect any proET-1 immunoreactivity, whereas three other assays (assays 18–53, 94–148 and 168–212) detected moderately elevated concentrations for normal subjects and strongly elevated levels for septic patients (Table 1). The median concentrations determined by the three assays were: assay 18–53: 22 arb. U/l (range 5–59 arb. U/l) for normal subjects and 148 arb. U/l (range 51–655 arb. U/l) for septic patients, assay 94–148: 21 arb. U/l (range 8–48 arb. U/l) for normal subjects and 94 arb. U/l (range 21–214 arb. U/l) for septic patients and assay 168–212: 32 arb. U/l (range 4–48 arb. U/l) for normal subjects and 199 arb. U/l (range 130–426 arb. U/l) for septic patients (Fig. 2). Finally, using the three assays, we investigated the stability of the proET-1 moieties in septic plasma samples (n = 20) after storage for 24 h at 4 and 22 ◦ C, and for 4 weeks at −20 ◦ C (Table 2). For most conditions proET-1 immunoreactivities were completely stable; a slight but statistically significant decay was observed for assays 18–53 (24 h at 4 and 22 ◦ C) and 168–212 (24 h at 22 ◦ C).

J. Struck et al. / Peptides 26 (2005) 2482–2486

2484

Fig. 1. Structural features of pre-proET-1. In the schematic representation of the molecule various moieties are shown (drawn to scale). The signal sequence (position 1–17), bigET-1 (position 53–90) and ET-1 (position 53–73), peptides used for the generation of antibodies. Putative recognition motifs for prohormone convertases (Rx; corresponding amino acid sequence given below), as proposed by Blais et al. [4], are indicated. Antibody pairs used in six sandwich immunoassays are shown, and positions of pre-proET-1 covered by the immunoassays are denoted. In the schematic representation of the sandwich immunoassays, a continous line between the antibodies indicates, that immunoreactivity was detected with the assay in patient samples, whereas a dotted line indicates that immunoreactivity was not detectable in such samples. Table 1 Characteristics of sandwich immunoassays covering various regions of proET-1 Standard

Assay 18–53

Assay 32–109

Assay 32–181

Assay 94–148

Assay 136–181

Assay 168–212

B [RLU]

B [RLU]

B [RLU]

B [RLU]

B [RLU]

B [RLU]

0 arb. U/l 30 arb. U/l 90 arb. U/l 270 arb. U/l 810 arb. U/l 2430 arb. U/l

186 319 674 1626 4450 12963

184 413 907 2242 5884 17289

176 400 812 2100 5480 15067

252 596 1183 3048 9071 24212

300 597 1256 2812 8132 23750

116 1004 2524 7606 18884 39832

Detection limit [arb. U/l]

8

4

3

5

6

1

Controls

+



+


+

Sepsis

+++



+++


+++

The names of the assays used are related to the terminal positions within pre-proET-1 covered by the antibodies used (for details see Section 2). Signals detected with the assays for various concentrations of recombinant proET-1 (standard, [arb. U/l]) are given in relative light units [RLU]. The lower detection limits determined for the assays are indicated (for details see Section 2). The lowest two lines qualitatively describe mean values of proET-1 immunoreactivity detected with the assays for plasma samples of healthy control subjects (n = 10) and of septic patients (n = 20) (
Assay 18–53

Assay 94–148

Assay 168–212

96.0 ± 1.3%* 89.5 ± 2.2%* 101.0 ± 0.7%

101.2 ± 1.3% 105.0 ± 2.6% 98.8 ± 0.8%

97.9 ± 1.7% 93.8 ± 2.8%* 99.9 ± 0.6%

Shown are mean recoveries with standard errors of initially detected immunoreactivities of 20 samples analyzed in three assays after storage as indicated. * p < 0.05.

4. Discussion Work on the elucidation of the proteolytic processing pathway of proET-1 has mainly focussed on the formation ET-1 so far, but not on other regions of the precursor. In proET1, six putative recognition sites for prohormone convertases have been identified previously, termed R1a/R1b, R2, R3,

R4 and R5 [4]. In vitro, the prohormone convertases furin and PC7 have been shown to cleave with similar specificities at most of these sites, and it has been suggested that these convertases are involved in the maturation of proET-1 in endothelial cells also in vivo [4,5]. We have sought to identify the pattern of proET-1 processing that actually occurs in vivo. As tools for assessing this pattern, six sandwich immunoassays covering various portions of proET-1 were developed and shown to be technically suitable. To investigate the processing pattern in vivo, we selected healthy subjects, but also severe septic patients as model system, because for this condition elevated plasma concentrations of ET-1 and bigET-1 have been reported [6,17,20,21,25]. Since severe sepsis is a highly complex disorder [7], it is beyond the scope of this study to infer from the limited number of samples analyzed here to a correlation of measured proET-1 immunoreactivities with particular clinical conditions.

J. Struck et al. / Peptides 26 (2005) 2482–2486

Fig. 2. Plasma concentrations of proET-1 fragments. The names of the sandwich immunoassays used are related to the terminal positions within pre-proET-1 covered by the antibodies used (for details see Section 2). Samples of healthy control subjects (n = 10) and of septic patients (n = 20) were measured and are represented as box-and-whiskers plots.

For any sandwich immunoassay covering prohormone convertase recognition sites R1, R2 or R4 (assays 32–109, 32–181 and 136–181), no proET-1 immunoreactivity was detectable in plasma, neither for healthy subjects nor for septic patients. On the contrary, assays 18–53, 94–148 (covering R3) and 168–212 (covering R5) detected moderately elevated proET-1 immunoreactivity in plasma from normal subjects, and strongly elevated proET-1 immunoreactivity in plasma from septic patients. Thus, we conclude that sites R3 and R5 are not used for conversion of proET-1 in vivo, whereas proteolytic processing apparently occurs at sites R1, R2 and R4, both under normal and pathological conditions. The resolution of our analytical approach was not high enough to distinguish, whether R1-processing occurs at subsites R1a or R1b, since these adjoin directly to each other. Our data show that besides bigET-1, which is later processed to ET-1, there are three other peptides derived from proET-1 present in the circulation, which are detectable by assays 18–53, 94–148 and 168–212, respectively. A limitation of our study is, that the precise termini of the fragments remain elusive, since we have not determined their identity by mass spectrometry. All three fragments exhibit a similar high stability in isolated plasma. The plasma concentrations of the three fragments were fairly similar among healthy subjects, and among septic patients, respectively. Both the high stability ex vivo and the similarity in plasma concentration indicate that the three fragments are not subject to rapid turnover, and suggest that none of the fragments has a function, which is tightly regulated, or maybe even be without function in the circulation. This is in sharp contrast to mature ET-1, which has a plasma half-life time of only about 1–2 min [24] and acts in

2485

an autocrine or paracrine manner [22]. It is these features of ET-1, which have hampered so far its reliable quantification in plasma samples [18]. ET-1 is considered a pathology-driving factor in cardiovascular diseases like pulmonary hypertension, chronic heart failure and others, and also in septic complications [3,8,12,14,23]. Consequently, quantification of released amounts of ET-1 could help to more selectively direct therapies, such as application of ET-1 receptor antagonists and other measures [11]. Since the proET-1 fragments we describe in this study, which stem from the same precursor as ET-1, are apparently much more stable than mature ET-1, they could be used as alternative diagnostic targets to indirectly assess the release of ET-1. The measurement of stable precursor fragments as a mirror of mature peptide release has been successfully applied previously for natriuretic peptides of the A- and the B-type and for adrenomedullin [15,16,19]. In conclusion, the use of various sandwich immunoassays enabled us to delineate the proteolytic processing pattern of proET-1 in vivo, both under normal and septic conditions, and we identified stable proET-1 fragments, which should have the potential to be used in the future as diagnostic targets for the indirect quantification of ET-1 release to assist therapy of cardiovascular diseases and sepsis. Acknowledgements We wish to thank Chen Tao, Christine Alonso, Marko Talke, Detlef Hintzen (BRAHMS Aktiengesellschaft) and Dr. Martina Strebelow and Sonja Tietz (Unicus Karlsburg OHG) for excellent technical assistance. References [1] American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992;20:864–74. [2] Agapitov AV, Haynes WG. Role of endothelin in cardiovascular disease. J Renin Angiotensin Aldosterone Syst 2002;3:1–15. [3] Battistini B, Forget MA, Laight D. Potential roles for endothelins in systemic inflammatory response syndrome with a particular relationship to cytokines. Shock 1996;5:167–83. [4] Blais V, Fugere M, Denault JB, Klarskov K, Day R, Leduc R. Processing of proendothelin-1 by members of the subtilisin-like proprotein convertase family. FEBS Lett 2002;524:43–8. [5] Denault JB, Claing A, D’Orleans-Juste P, Sawamura T, Kido T, Masaki T, et al. Processing of proendothelin-1 by human furin convertase. FEBS Lett 1995;362:276–80. [6] Hirata Y, Mitaka C, Emori T, Amaha K, Marumo F. Plasma endothelins in sepsis syndrome. JAMA 1993;270:2182. [7] Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med 2003;348:138–50. [8] Humbert M, Morrell NW, Archer SL, Stenmark KR, MacLean MR, Lang IM, et al. Cellular and molecular pathobiology of pulmonary arterial hypertension. J Am Coll Cardiol 2004;43:13S–24S. [9] Kedzierski RM, Yanagisawa M. Endothelin system: the doubleedged sword in health and disease. Annu Rev Pharmacol Toxicol 2001;41:851–76.

2486

J. Struck et al. / Peptides 26 (2005) 2482–2486

[11] Luscher TF, Barton M. Endothelins and endothelin receptor antagonists: therapeutic considerations for a novel class of cardiovascular drugs. Circulation 2000;102:2434–40. [12] Luscher TF, Enseleit F, Pacher R, Mitrovic V, Schulze MR, Willenbrock R, et al. Hemodynamic and neurohumoral effects of selective endothelin A (ET(A)) receptor blockade in chronic heart failure: the Heart Failure ET(A) Receptor Blockade Trial (HEAT). Circulation 2002;106:2666–72. [13] Masaki T. Historical review: endothelin. Trends Pharmacol Sci 2004;25:219–24. [14] Maurer M, Wedemeyer J, Metz M, Piliponsky AM, Weller K, Chatterjea D, et al. Mast cells promote homeostasis by limiting endothelin-1-induced toxicity. Nature 2004;432:512–6. [15] Morgenthaler NG, Struck J, Thomas B, Bergmann A. Immunoluminometric assay for the midregion of pro-atrial natriuretic peptide in human plasma. Clin Chem 2004;50:234–6. [16] Mueller T, Gegenhuber A, Poelz W, Haltmayer M. Head-to-head comparison of the diagnostic utility of BNP and NT-proBNP in symptomatic and asymptomatic structural heart disease. Clin Chim Acta 2004;341:41–8. [17] Pittet JF, Morel DR, Hemsen A, Gunning K, Lacroix JS, Suter PM, et al. Elevated plasma endothelin-1 concentrations are associated with the severity of illness in patients with sepsis. Ann Surg 1991;213:261–4. [18] Rossi GP, Seccia TM, Albertin G, Pessina AC. Measurement of endothelin: clinical and research use. Ann Clin Biochem 2000;37(Pt 5):608–26.

[19] Struck J, Tao C, Morgenthaler NG, Bergmann A. Identification of an adrenomedullin precursor fragment in plasma of sepsis patients. Peptides 2004;25:1369–72. [20] Tschaikowsky K, Sagner S, Lehnert N, Kaul M, Ritter J. Endothelin in septic patients: effects on cardiovascular and renal function and its relationship to proinflammatory cytokines. Crit Care Med 2000;28:1854–60. [21] Voerman HJ, Stehouwer CD, van Kamp GJ, Strack van Schijndel RJ, Groeneveld AB, Thijs LG. Plasma endothelin levels are increased during septic shock. Crit Care Med 1992;20:1097– 101. [22] Wagner OF, Christ G, Wojta J, Vierhapper H, Parzer S, Nowotny PJ, et al. Polar secretion of endothelin-1 by cultured endothelial cells. J Biol Chem 1992;267:16066–8. [23] Wanecek M, Weitzberg E, Rudehill A, Oldner A. The endothelin system in septic and endotoxin shock. Eur J Pharmacol 2000;407: 1–15. [24] Weitzberg E, Ahlborg G, Lundberg JM. Long-lasting vasoconstriction and efficient regional extraction of endothelin-1 in human splanchnic and renal tissues. Biochem Biophys Res Commun 1991;180:1298–303. [25] Weitzberg E, Lundberg JM, Rudehill A. Elevated plasma levels of endothelin in patients with sepsis syndrome. Circ Shock 1991;33:222–7. [26] Xu D, Emoto N, Giaid A, Slaughter C, Kaw S, deWit D, et al. ECE1: a membrane-bound metalloprotease that catalyzes the proteolytic activation of big endothelin-1. Cell 1994;78:473–85.