GC-MS and LC-MS/MS pilot studies on the guanidine (NG)-dimethylation in native, asymmetrically and symmetrically NG-dimethylated arginine-vasopressin peptides and proteins in human red blood cells

https://doi.org/10.1016/j.jchromb.2020.122024Get rights and content

Highlights

  • Asymmetric Arg-dimethylation in peptides and proteins was studied by GC-MS and LC-MS/MS.

  • Several Arg-dimethylated structure proteins were identified in human red blood cells.

  • GC-MS but not LC-MS/MS discriminates between asymmetrically and symmetrically dimethylated Arg-vasopressin.

  • Asymmetrical and symmetrical dimethylation of Arg-vasopressin changes is platelet aggregatory potency.

Abstract

Protein-arginine methyltransferases catalyze the methylation of the guanidine (NG) group of proteinic L-arginine (Arg) to produce monomethyl and dimethylarginine proteins. Their proteolysis releases the free amino acids monomethylarginine (MMA), symmetric dimethylarginine (SDMA) and asymmetric dimethylarginine (ADMA), respectively. MMA, SDMA and ADMA are inhibitors of the nitric oxide synthase (NOS) activity. High circulating and low urinary concentrations of ADMA and SDMA are considered risk factors in the cardiovascular and renal systems, mainly due to their inhibitory action on NOS activity. Identity, biological activity and concentration of NG-methylated proteins are largely unknown. The present study addressed these issues by using GC-MS and LC-MS/MS approaches. GC-MS was used to quantify free ADMA released by classical HCl-catalyzed hydrolysis of three synthetic Arg-vasopressin (V) peptides and of unknown endogenous NG-dimethylated proteins. The cyclic (c) disulfide forms of Arg-vasopressin analogs, i.e., Arg-vasopressin (cV-Arg-Gly-NH2), asymmetrically NG-dimethylated vasopressin (cV-ADMA-Gly-NH2) and symmetrically NG-dimethylated vasopressin (cV-SDMA-Gly-NH2) were used as model peptides in quantitative GC-MS analyses of ADMA, SDMA and other expected amino acids from the hydrolyzed Arg-vasopressin analogs. cV-ADMA-Gly-NH2 and cV-SDMA-Gly-NH2 were discriminated from cV-Arg-Gly-NH2 by LC-MS and LC-MS/MS, yet they were indistinguishable from each other. The same applies to the respective open (o) reduced and di-S-acetamide forms of oV-ADMA-Gly-NH2, oV-SDMA-Gly-NH2 and oV-Arg-Gly-NH2. Our LC-MS and LC-MS/MS studies suggest that the Arg-vasopressin analogs form [(M−H)]+ and [(M−H)+H]+ in the positive ESI mode and undergo in part conversion of their terminal Gly-NH2 (NH2, 16 Da) group to Gly-OH (OH, 17 Da). The product ion mass spectra of the di-S-acetamide forms are complex and contain several intense mass fragments differing by 1 Da. cV-ADMA-Gly-NH2 and cV-SDMA-Gly-NH2 induced platelet aggregation in platelet-rich human plasma with moderately different initial velocity and maximal aggregation rates compared to cV-Arg-Gly-NH2. Previous studies showed that human red blood cells are rich in large (>50 kDa) ADMA-containing proteins of unknown identity. Our LC-MS/MS proteomic study identified several membrane and cytosolic erythrocytic NG-dimethylated proteins, including spectrin-α (280 kDa), spectrin-β (247 kDa) and protein 4.1 (80 kDa). Being responsible for the stability of the erythrocyte membrane, the newly identified main targets for NG-dimethylation in human erythrocytes should be given a closer look in erythrocytic diseases like hereditary spherocytosis.

Introduction

Free L-arginine is the substrate of all known nitric oxide synthase (NOS; EC 1.14.13.39) isoforms [1]. The imine group of the terminal guanidine (NG) group of L-arginine is oxidized by NOS to nitric oxide (NO). L-Citrulline is the second product of this reaction. The free amino acids monomethylarginine (MMA, NG-monomethyl-L-arginine), asymmetric dimethylarginine (ADMA, NG,NG-dimethyl-L-arginine) and symmetric dimethylarginine (SDMA, NG,N′G-dimethyl-L-arginine) are endogenous inhibitors of NOS activity [2]. The relative inhibitory potency against the activity of neuronal NOS is MMA > ADMA > SDMA [2]. High concentrations of circulating and low concentrations of urinary ADMA and SDMA are considered cardiovascular risk factors/markers [3]. SDMA is generally considered to be biologically inactive with respect to NOS activity, although SDMA has been reported to inhibit the activity of recombinant neuronal NOS (nNOS) [2]. The cardiovascular risk arising from ADMA is generally attributed to its inhibitory action on endothelial NOS (eNOS). However, the inhibitory potency of ADMA towards eNOS is very weak (e.g., IC50 ≈ 12 µM ADMA) [4]. It is therefore unlikely that the poor inhibitory potency of ADMA and SDMA against eNOS may explain their cardiovascular risks. We hypothesized that these NG-dimethylated Arg analogs may exert additional not yet recognized atherosclerotic effects (reviewed in Ref. [5]). Recently, SDMA, but not ADMA, has been reported to modify the structure of high-density lipoprotein (HDL) and to induce endothelial dysfunction by activating the Toll-like receptor-2 [6], [7], [8]. These newer findings may explain the atherosclerotic effect of SDMA.

Amino acid residues in proteins undergo many different post-translational modifications (PTM) [9], resulting in altered biological functions of those proteins. Two abundant PTM are the citrullination and the methylation of Arg residues (Fig. 1A). The citrullination is catalyzed by the Ca2+-dependent peptidyl-arginine deiminase (PAD; EC 3.5.3.15) [10]. It is assumed that the citrullination is associated with autoimmune diseases, notably rheumatoid arthritis [11]. This is because the newly generated L-citrulline moiety in proteins gives rise to the generation of new autoantibodies. The methylation of certain Arg moieties in proteins is catalyzed by the large family of protein-arginine methyltransferases (PRMT; EC 2.1.1.125), which uses the universal methyl-group donor S-adenosylmethionine (SAM) as the cofactor [12]; PRMT-catalyzed NG-methylation produces proteins that bear MMA, SDMA or ADMA moieties (Fig. 1A). Analogous to citrullination, NG-methylation of proteins may also induce autoimmunity, yet the pathophysiology of this PTM is much less understood. Recent studies indicated that Arg residues in certain transporters and ion channels are NG-dimethylated and that this PTM changes their activity [13], [14], [15].

Identity, biological functions and concentration of NG-methylated Arg proteins are largely unknown. Their proteolysis releases the free amino acids MMA, ADMA and SDMA. They circulate in blood and are distinctly differently metabolized and eliminated by the human body. Thus, SDMA is excreted almost unchanged in the urine. MMA and ADMA are excreted in part unchanged (by about 10% for ADMA) and in part as monomethylamine and dimethylamine (DMA), respectively, upon hydrolysis by dimethylarginine dimethylaminohydrolase (DDAH; EC 3.5.3.18) [1], [16]. The concentration of the free amino acids ADMA and SDMA in blood and urine are several times higher than that of MMA.

The concentration of asymmetrically NG-dimethylated Arg proteins in human serum is about 5 times lower than that of free ADMA [17]. These ADMA-bearing proteins have not been identified thus far. It is solely known that human serum albumin is essentially free of asymmetrically NG-dimethylated Arg moieties [18]. In serum of healthy elderly subjects, we found an inverse correlation between asymmetric NG-dimethylation and citrullination [17], suggesting potential interaction between these PTM. Human erythrocytes are rich in asymmetrically NG-dimethylated proteins. In proteolysates (6 M HCl, 20 h, 110 °C) of lyzed human erythrocytes, we measured by GC-MS about 15 µM of asymmetrically NG-dimethylated large (>50 kDa) proteins of unknown identity [19]. It is worth mentioning that neither human serum albumin [18] nor human erythrocytic hemoglobin are asymmetrically NG-dimethylated to a considerable degree [20].

The aim of the present study was to contribute to two important aspects of the NG-dimethylation of proteins and peptides: 1) to identify and quantify asymmetrically NG-dimethylated proteins in human red blood cells; and 2) to test the biological activity of NG-dimethylated peptides in comparison to the native non-modified peptides. To reach these goals, we used mass spectrometry-based analytical techniques, i.e., LC-MS/MS and GC-MS approaches, and well-characterized synthetic L-arginine-containing peptides (Fig. 1B). As a model peptide, we chose vasopressin (V), the endogenous anti-diuretic hormone (ADH). The L-arginine-vasopressin (peptide sequence: CYFQNCPRG-NH2) is a nona-peptide and contains a single Arg moiety and two Cys moieties. The biologically and pharmacologically active L-arginine-vasopressin is cyclic (c) due to the Cys-Cys disulfide bridge and the terminal Gly moiety is amidated: cV-Arg-Gly-NH2. In the present work, we used commercially available native arginine-vasopressin (cV-Arg-Gly-NH2) and two custom-made NG-dimethylated arginine-vasopressin analogs: asymmetrically NG-dimethylated arginine-vasopressin (cV-ADMA-Gly-NH2) and the symmetrically NG-dimethylated arginine-vasopressin (cV-SDMA-Gly-NH2). Table 1 and the supplementary Fig. S1 inform of the structures, abbreviations and explanations of the arginine-vasopressin analogs discussed in the present study.

Section snippets

Chemicals and materials

NG,N′G-Di-[2H3]methyl-L-arginine (d6-SDMA, 99% isotopic purity at 2H) and bovine liver catalase were obtained from Sigma-Aldrich (Steinheim, Germany). Sodium azide, acrylamide and dithiothreitol (DTT) were from Merck (Darmstadt, Germany). Arg8-vasopressin (V-Arg) acetate salt (purity, ≥95% by HPLC), angiotensin I and II acetate salt hydrates (declared chemical purity, ≥93% by HPLC), and S-(5′-adenosyl)-L-methionine dihydrochloride (declared chemical purity, ≥75% by HPLC) were purchased from

LC-MS spectra of arginine-vasopressin and analogs

The LC-MS spectrum obtained by positive ESI (+ESI) of the native L-arginine vasopressin form, cV-Arg-Gly-NH2, contained two major mass fragments at m/z 542.7 and 1084.4 in the m/z range 300 – 1600 (Fig. S2A). Very similar LC-MS have been reported for cV-Arg-Gly-NH2 by others, who assigned these ions to the protonated doubly and singly charged molecular cations [M+2H]2+ and [M+H]+, respectively [26], [27], [28]. However, because the molecular mass of the charge-free cV-Arg-Gly-NH2 is 1084.2 (

Discussion

In this work, we performed GC-MS and LC-MS/MS pilot studies on the dimethylation of the guanidine (NG) group of L-arginine in peptides and proteins (Fig. 1). Arginine-vasopressin and custom-made asymmetrically and symmetrically NG-dimethylated vasopressin analogs were used as model peptides. Arginine-vasopressin (V-Arg) is the endogenous nona-peptide Cys-Tyr-Phe-Gln-Asn-Cys-Pro-Arg-Gly-NH2. V-Arg contains a single Arg residue on position 8, a disulfide bridge built between two Cys moieties

Conclusion

GC-MS is useful to quantify free ADMA released by classical HCl-catalyzed hydrolysis of synthetic peptides and endogenous NG-dimethylated proteins. LC-MS/MS in the +ESI mode allow the identification of native NG-dimethylated peptides such as cV-ADMA-Gly-NH2 and cV-SDMA-Gly-NH2 upon complete chromatographic separation. The type of NG-dimethylation cannot be determined by CID of [M−H]+/[(M−H)+H]2+, but requires GC-MS analysis of ADMA and SDMA released from the arginine-vasopressin peptides by

CRediT authorship contribution statement

Alexander Bollenbach: Methodology, Investigation, Software, Writing - review & editing, Data curation. Stepan Gambaryan: Methodology, Investigation, Writing - review & editing. Igor Mindukshev: Methodology, Investigation. Andreas Pich: Methodology, Software, Writing - review & editing, Data curation. Dimitrios Tsikas: Conceptualization, Writing - review & editing, Data curation, Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

We thank Dr. Ute Gravemann (German Red Cross Blood Service NSTOB, Springe, Germany) for her advice in human platelet aggregation measurements on the aggregometer model PAP-8 and for careful proofreading of our manuscript.

References (43)

  • A. Böhmer et al.

    Human hemoglobin does not contain asymmetric dimethylarginine (ADMA)

    Nitric Oxide

    (2012)
  • A. Bollenbach et al.

    GC-MS quantification of urinary symmetric dimethylarginine (SDMA), a whole-body symmetric L-arginine methylation index

    Anal. Biochem.

    (2018)
  • A. Thomas et al.

    Determination of vasopressin and desmopressin in urine by means of liquid chromatography coupled to quadrupole time-of-flight mass spectrometry for doping control purpo

    Anal. Chim. Acta

    (2011)
  • D. Zhang et al.

    Development and validation of a highly sensitive LC-MS/MS assay for the quantification of arginine vasopressin in human plasma and urine: Application in preterm neonates and child

    J. Pharm. Biomed. Anal.

    (2014)
  • M. Hashimoto et al.

    Severe Hypomyelination and Developmental Defects Are Caused in Mice Lacking Protein Arginine Methyltransferase 1 (PRMT1) in the Central Nervous System

    J. Biol. Chem.

    (2016)
  • M. Davids et al.

    Plasma concentrations of arginine and asymmetric dimethylarginine do not reflect their intracellular concentrations in peripheral blood mononuclear cells

    Metabolism

    (2013)
  • J. Filep et al.

    Mechanism of vasopressin-induced platelet aggregation

    Thromb. Res.

    (1987)
  • I. Zvereva et al.

    Comparison of various in vitro model systems of the metabolism of synthetic doping peptides: Proteolytic enzymes, human blood serum, liver and kidney microsomes and liver S9 fraction

    J. Proteomics

    (2016)
  • M.B. Dillon et al.

    Automethylation of protein arginine methyltransferase 8 (PRMT8) regulates activity by impeding S-adenosylmethionine sensitivity

    J. Biol. Chem.

    (2013)
  • A. Böhmer et al.

    Doubts concerning functional endothelial nitric oxide synthase in human erythrocytes

    Blood

    (2012)
  • S. Schlesinger et al.

    Asymmetric and symmetric dimethylarginine as risk markers for total mortality and cardiovascular outcomes: A systematic review and meta-analysis of prospective studies

    PLoS ONE

    (2016)
  • View full text