CKD (chronic kidney disease) is a life-threatening pathology, often requiring HD (haemodialysis) and characterized by high OS (oxidative stress), inflammation and perturbation of vascular endothelium. HD patients have increased levels of vWF (von Willebrand factor), a large protein (~240 kDa) released as UL-vWF (ultra large-vWF polymers, molecular mass ~20000–50000 kDa) from vascular endothelial cells and megakaryocytes, and responsible for the initiation of primary haemostasis. The pro-haemostatic potential of vWF increases with its length, which is proteolytically regulated by ADAMTS-13 (a disintegrin and metalloproteinase with thrombospondin motifs 13), a zinc-protease cleaving vWF at the single Tyr1605–Met1606 bond, and by LSPs (leucocyte serine proteases), released by activated PMNs (polymorphonuclear cells) during bacterial infections. Previous studies have shown that in vitro oxidation of Met1606 hinders vWF cleavage by ADAMTS-13, resulting in the accumulation of UL-vWF that are not only more pro-thrombotic than shorter vWF oligomers, but also more efficient in binding to bacterial adhesins during sepsis. Notably, HD patients have increased risk of developing dramatic cardiovascular and septic complications, whose underlying mechanisms are largely unknown. In the present study, we first purified vWF from HD patients and then chemically characterized its oxidative state. Interestingly, HD-vWF contains high carbonyl levels and increased proportion of UL-vWF polymers that are also more resistant to ADAMTS-13. Using TMS (targeted MS) techniques, we estimated that HD-vWF contains >10% of Met1606 in the sulfoxide form. We conclude that oxidation of Met1606, impairing ADAMTS-13 cleavage, results in the accumulation of UL-vWF polymers, which recruit and activate platelets more efficiently and bind more tightly to bacterial adhesins, thus contributing to the development of thrombotic and septic complications in CKD.

INTRODUCTION

CKD (chronic kidney disease), often requiring HD (haemodialysis) treatment, is a life-threatening disease of outstanding clinical relevance and dramatic social and economic burden [1,2]. Available data conclusively demonstrate that CKD is characterized by high OS (oxidative stress) [35], with increased production of ROS (reactive oxygen species) and RNS (reactive nitrogen species), including H2O2 (hydrogen peroxide), superoxide radical (O2) and PN (peroxinitrite; ONOO) [6,7]. CKD is also associated with a pro-inflammatory state [8] and perturbation of vascular endothelium [9], whereby CKD can be ultimately regarded as a vasculopathic state [10]. Markedly increased expression of NADPH oxidase (a key enzyme in ROS generation) has been observed in the vascular endothelium of HD patients, together with high levels of the pro-inflammatory cytokine IL-6 (interleukin 6) and acute-phase CRP (C-reactive protein) [3]. Extracellular reduced thiols (e.g. free cysteine and homocysteine and albumin-bound cysteine), constituting an important component of the natural antioxidant defence [11,12], are depleted in HD patients, whereas the corresponding oxidized thiols accumulate, with a resulting pro-thrombotic effect [3,13]. Many other oxidation end-products of lipids, carbohydrates, and proteins accumulate in the plasma of HD patients primarily as a result of their concomitant increased production and diminished renal clearance in CKD [3], but also due to the enhanced release of myeloperoxidase from phagocytes that are activated during HD treatment [14]. Noteworthy, increased plasma levels of the haemostatic vWF (von Willebrand factor) have been found in uremic patients and represents a key marker of vascular endothelium perturbation occurring in CKD [15,16].

vWF is a large (molecular mass of monomers ~240 kDa) multidomain protein [17,18] released as UL-vWF [ultra large-vWF polymers, (20–50)×103 kDa] from endothelial cells and platelets upon stimulated and basal secretion [19]. Under conditions of normal blood flow, UL-vWF is in a globular/collapsed state and exerts only poor haemostatic activity [20]. Conversely, under the high shear forces generated at sites of vascular injury, UL-vWF multimers elongate [21] and initiate primary haemostasis by exposing specific regions responsible for the interaction with collagen in the damaged subendothelium and for platelet recruitment and activation [22,23]. Notably, the platelet-aggregating potential of vWF is crucially dependent on its length [23]. Indeed, longer UL-vWF multimers are intrinsically more sensitive to shear-induced unfolding [24] and therefore are more pro-thrombotic than shorter vWF species [22,23]. The length of UL-vWF is regulated proteolytically by ADAMTS-13 (a disintegrin and metalloproteinase with thrombospondin motifs 13), a zinc-dependent plasma protease that exclusively cleaves vWF at the single peptide bond Tyr1605–Met1606 [25] that is recognized by the protease only after shear-induced partial unfolding of vWF [26]. Remarkably, apart from its key role in haemostasis, UL-vWF also participates in bacterial infection [27]. Many bacterial pathogens express on their surface adhesin molecules, such as staphylococcal Protein A, that preferentially engage UL-vWF polymers to adhere to vascular vessels and transmigrate into tissues, thus expanding infection [27]. Bacterial adhesion also triggers recruitment, activation and extravasation of PMNs (polymorphonuclear cells) [27,28], secreting ROS and LSPs (leucocyte serine proteases: elastase, proteinase-3 and cathepsin-G) [29] that efficiently cleave UL-vWF multimers at or near the same peptide bond hydrolysed by ADAMTS-13 [30,31]. Hence, the haemostatic potential of vWF is the result of a dynamic equilibrium existing between the concentration of bioactive UL-vWF released from vascular endothelial cells and the proteolytic efficiency of circulating ADAMTS-13 and LSPs in different clinical settings.

Recently, we have shown that vWF purified from the plasma of patients with T2D (Type 2 diabetes), a severe metabolic disease characterized by OS and complicated by dramatic thrombotic microangiopathies [32], contains higher carbonyl levels, taken as reliable markers of protein oxidative damage [33], than those found in vWF from healthy subjects and higher proportion of UL-vWF polymers that were more resistant to proteolysis by ADAMTS-13 and retained the capacity of interacting with platelets [34]. Interestingly, similar results were obtained after in vitro oxidation of normal vWF with physiological concentrations of PN. Hence we proposed that impairment of ADAMTS-13-mediated proteolytic shortening of vWF by oxidation of the vulnerable Tyr1605–Met1606 bond might represent a novel pro-thrombotic mechanism contributing to the pathogenesis of microangiopathic complications in T2D [34]. Due to the exceedingly low amounts of T2D-vWF available and the complexity of the vWF molecule, containing many oxidant-sensitive amino acids, our hypothesis was tested on the synthetic vWF peptide vWF74 (i.e. the minimal peptide sequence, residues 1596–1669, of vWF that can be efficiently cleaved by ADAMTS-13) and its analogues vWF74-NT (3-nitrotyrosine) and vWF74-MetSO (methionine sulfoxide), where Tyr1605 and Met1606 were replaced by their corresponding nitro-oxidation products NT and MetSO respectively. Remarkably, vWF74-MetSO was almost fully resistant to proteolysis by ADAMTS-13, whereas vWF74-NT was hydrolysed as efficiently as the unmodified peptide vWF74 [34], indicating that Met1606 oxidation (but not Tyr1605 nitration) hinders vWF cleavage. This finding is unique to ADAMTS-13 and opposes the general trend whereby oxidative modifications destabilize proteins and enhance their susceptibility to proteolytic attack [29]. The molecular basis of this behaviour has been recently revealed by modelling and docking simulations, showing that the resistance of the Tyr1605–MetSO1606 bond to ADAMTS-13 hydrolysis is caused by loss of hydrophobic interactions and steric clashes introduced in the protease-active site upon methionine oxidation [35].

Notably, HD patients have a greatly increased risk of CVD (cardiovascular disease) that accounts for up to 60% of deaths in CKD [1,36]. The excess incidence of CVD has been rationalized taking into account ‘non-traditional’ risk factors for CVD, including OS, inflammation and endothelial dysfunction [9]. Notwithstanding, the biochemical root linking high OS to CVD in HD patients is largely unknown [3,4]. Another major clinical issue in CKD entails the high frequency with which these patients develop sepsis (i.e. in 56% of cases) that may even double the mortality rate [37]. Again, the underlying molecular mechanism leading to septic complications is still elusive [37]. These considerations and the concomitant presence in HD patients of high OS and elevated vWF concentrations, prompted us to chemically characterize the oxidative state of vWF purified from HD patients. Our data show that HD-vWF contains high carbonyl levels and increased proportion of UL-vWF polymers that are also more resistant to ADAMTS-13. Using TMS (targeted MS) techniques, we estimate that >10% of Met1606 in vWF from HD patients is oxidatively modified to MetSO. Hence we conclude that accumulation of UL-vWF, ultimately caused by oxidation of Met1606 occurring under high OS, may contribute to the development of cardiovascular and septic complications in CKD.

EXPERIMENTAL

Human blood samples

Blood samples were supplied by the institutional blood bank of the ‘A. Gemelli’ Hospital (Catholic University School of Medicine, Rome, Italy) and of the University Hospital (University of Padua, Padua, Italy), while the protocols for all clinical studies were scrutinized and approved by the corresponding Institutional Review Boards. All subjects gave their informed consent to the present study. Healthy subjects (n=39) were blood donors from the institutional blood bank. They were between 38 and 58 years of age, were in good health, non-smokers and had no risk factors for CVD. Patients (n=39) with ESRD (end-stage renal disease), treated with HD three times/week for 1–18 years, without malignant neoplasms, were age- and sex-matched with controls. Considering the known effect of blood group on the level of circulating vWF, the control subjects were also blood group-matched with the HD patients [38]. Blood samples were collected in 3.8% citrate, then centrifuged at 5000 rev./min for 20 min at 20°C. The plasma samples were then stored at −80°C. Frozen blood samples from healthy or HD patients were thawed at 37°C.

Purification of vWF from HD patients

vWF from the plasma pool of healthy subjects and of HD patients with CKD was purified, essentially following the cryoprecipitation method [39], as described previously [34,40]. EDTA-anti-coagulated blood samples were collected from 30 patients of the Padova University Hospital, under chronic dialysis treatment (i.e. 210–240 min/dialysis session, three times a week, using polysulfone dialysers) for at least 1 year (1–15 years range). Five samples (200 ml) of pooled plasma, obtained after blood centrifugation (5000 rev./min for 20 min at 20°C), were stored at −80°C until use. Patients were selected on the basis of lack of malignancy, heart failure, chronic pulmonary diseases and hospitalization in the preceding 6 months.

Briefly, the frozen plasma pool (200 ml) was thawed at room temperature, added with 2 g of PEG [poly(ethylene glycol)]-6000 (Sigma) to a final 1% (w/v) concentration, and gently stirred for 15 min. This solution was added with 1 mM PMSF (2 ml, 0.1 M) and 10 mM EDTA (4 ml, 0.5 M), as protease inhibitors and left overnight at 4°C under gentle magnetic stirring. The suspension was then centrifuged in 50 ml Falcon tubes at 3000 rev./min for 1 h at 2°C. In each Falcon tube, the supernatant was discarded and the pellet resuspended under gentle stirring with 2 ml of 110 mM sodium citrate buffer, pH 7.4, and 6.5 ml of 25 mM Tris/HCl, pH 6.8, containing 0.35 M NaCl and 2.6 M glycine. The suspension was centrifuged at 3000 rev./min for 45 min at 25°C. The pellet was discarded, while the supernatant was added with solid NaCl to a final concentration of 1.55 M. The suspension was stirred for 30 min and then centrifuged at 6000 rev./min for 30 min at 25°C. The supernatant was discarded, and the pellet (derived from the 200 ml plasma pool) was dissolved in citrate buffer (1 ml), divided into aliquots and stored at −80°C. Finally, the solution was fractionated by gel-filtration chromatography. The citrate solution was thawed, centrifuged at 13000 rev./min for 5 min, filtered at 0.45 μm and then loaded on to an in-house packed (2×50 cm) Sephacryl S-400 column eluted with 10 mM Hepes, pH 7.5, and 150 mM NaCl at a flow rate of 0.5 ml/min. Fractions (0.5 ml) were collected and analysed by SDS/PAGE on a Bio-Rad (Hercules) TGX 4–15% gradient gel, under reducing conditions. The gel was run at 12 mA constant current and Coomassie Blue stained. The high-molecular-mass standards (Sigma, catalogue number 17-0615-01) were used as protein markers. Fractions containing a single band at molecular mass compatible with that of intact vWF were collected, pooled and the concentration of the final solution was determined spectrophotometrically by measuring the absorbance value at 280 nm, using a molar absorption coefficient (ϵ) of 0.846 mg−1·cm2, calculated on the amino acid sequence of vWF monomer. The quality of vWF preparations was assessed by measuring vWF concentration as antigen (vWF:Ag) and RiCof (ristocetin cofactor) (vWF:RiCof), according to the immunoturbidometric assays ‘vWF antigen’ and ‘vWF activity’ (Instrumentation Laboratory), as detailed previously [34,40].

Determination of the carbonyl content of vWF purified from HD patients

vWF from HD patients of the ‘A. Gemelli’ University Hospital (Rome, Italy) was micropurified by immunoaffinity chromatography, as detailed previously [34]. Briefly, rabbit polyclonal anti-vWF antibody (2 mg) (Dako) was covalently coupled to Affi-Gel-10 agarose beads (1 ml) (Bio-Rad Laboratories). For each HD patient, the plasma sample (1 ml) was added to the anti-vWF antibody-conjugated beads (0.1 ml). After 1 h of incubation, the suspension was centrifuged (1 min at 1000 rev./min), the supernatant discarded and the gel washed three times with 1 ml of 10 mM Hepes, pH 7.4, 0.15 M NaCl. vWF was recovered by incubating the settled gel beads with a 6 M guanidine hydrochloride solution (0.1 ml) for 15 min under mild agitation. The total protein content was measured by the BCA (bicinchoninic acid) method, using the Bio-Rad Laboratories protein assay, and the carbonyl content of purified vWF samples was measured using the OxiSelect™ Protein Carbonyl ELISA Kit (Cell Biolabs), as reported previously [34]. In this assay, carbonyls were quantified by reaction with DNP (dinitrophenylhydrazine), to form phenylhydrazone derivatives [33], and subsequent incubation with biotinylated anti-DNP antibody followed by strepdavidin-linked HRP (horseradish peroxidase). A calibration curve was obtained with BSA at known content of carbonyl groups. The assay allowed us to obtain a reproducible sensitivity down to 10 pmol carbonyl/mg of protein, with an inter-assay variation of 13%.

Hydrolysis by ADAMTS-13 of vWF purified from HD patients

Normal vWF and vWF samples (100 μl and 20 μg/ml) from HD patients were incubated with recombinant ADAMTS-13 (5 nM) at 37°C in 5 mM Tris/HCl, pH 8.0, and 3 mM CaCl2 in the presence of sulfate-free ristocetin (1.5 mg/ml). After 0, 1 and 2 h, aliquots (50 μl) were sampled and the reaction stopped by adding EDTA (10 mM final concentration). The samples were analysed by SDS-agarose electrophoresis on a 1.2% agarose gel, followed by immunoblotting with rabbit anti-human vWF polyclonal antibody and HRP-conjugated secondary anti-rabbit antibody (Dako), as detailed previously [31,34].

In vitro oxidation of vWF from healthy subjects

The concentration of sodium PN (Cayman Chemical), was determined spectrophotometrically by measuring the absorbance of the solution at 302 nm, using a molar absorptivity value of 1670 M−1·cm−1 [34]. The stock solution was stored at −80°C and remained stable for at least 4 weeks. Immediately before use, the stock solution was diluted in 0.1 M NaOH and, during the experiments, was maintained in an ice bath. This vWF preparation (200 μl, 0.4 mg/ml) was subjected to ‘pulsed oxidation’ reaction with PN (10 μM) by multiple additions of oxidant solution (2 μl, 0.1 M stock solution in 0.1 M NaOH) for 60 min at a rate of one addition per minute. As a control, the reference vWF solution was treated with NaOH alone under identical experimental conditions.

Production of PTPs (proteotypic peptides)

PTPs of vWF are peptides representative of vWF containing Met1606 in the unmodified (Met-PTP) or sulfoxide (MetSO-PTP) form, were produced by proteolysis of the synthetic peptides vWF74 or vWF74-MetSO, with Glu-C protease from Staphylococcus aureus (Calbiochem). The pseudo-wild-type peptide vWF74 (D1596REQAPNLVYMVTGNPASDEIKRLPGDIQVVPIGVGPNANVQELERIGWPNAPILIQDFETLPREAPDLVLQRA1669), in which the C-terminal Cys1669 was replaced by alanine, and its MetSO derivative (vWF74-MetSO) were obtained by stepwise solid-phase synthesis, as detailed previously [34]. Proteolysis of vWF74 or vWF74-MetSO (220 μl, 0.34 mg/ml) in 0.1 M Hepes buffer, pH 7.4, containing 0.15 M NaCl, was conducted for 23 h at 37°C at a protease/substrate ratio of 1:20 (w/w). The reaction was stopped by the addition of TFA (trifluoroacetic acid) and fractionated on a Grace-Vydac (Hesperia) C18 analytical column (0.46×25 cm; 5 μm granulometry) equilibrated in 0.1% aqueous TFA and eluted with a linear acetonitrile–0.078% TFA gradient at a flow rate of 0.8 ml/min. The peptide material eluted in correspondence with the chromatographic peaks was collected and analysed by MS on a model Mariner ESI–TOF (electrospray ionization–time-of-flight) spectrometer (PerSeptive BioSystems). The two PTPs Met-PTP and MetSO-PTP, identified by MS analysis, were micropurified (~100 μg) by RP-HPLC (reverse-phase-HPLC) and subsequently used for MS analysis.

TMS analysis

vWF (125 μl, 0.4 mg/ml) in 0.1 M Tris/HCl buffer, pH 7.5, and 0.15 M NaCl was first reduced for 20 min at 80°C with 4 mM DTT (dithiothreitol) (5 μl, 120 mM stock solution) and then alkylated with 10 mM iodoacetamide (5 μl, 280 mM stock solution) for 15 min at 37°C in the dark, keeping the solution pH constant at 7.5 by adding Tris base (1 M). The reaction mixture (135 μl), containing the RCM (reduced and carboxamidomethylated) vWF, was treated with 250 μM PN in a ‘single-shot’ addition (25 μl of a 1.6 mM stock solution), to yield the corresponding oxidized species, RCM-vWF-Ox (oxidized vWF). Samples of RCM-vWF-Ox were also obtained by ‘multiple-shot’ addition of PN (10 μM final concentration) for 60 min at a rate of one shot per min, as detailed above. Samples of RCM-vWF and RCM-vWF-Ox were digested in parallel for 24 h at 37°C by adding to the reaction mixture Glu-C endoprotease (11 μl, 0.234 mg/ml stock solution) at a protease/RCM-vWF ratio of 1:20 (w/w). Alternatively, vWF was first oxidized with PN, according to the single- or multiple-shot procedure, under continuous vortexing, subjected to reduction and carboxamidomethylation reaction and finally to proteolysis with Glu-C protease.

Proteolysis mixtures were directly analysed by LC-MS (liquid chromatography-MS) using a micropump-200 HPLC system from PerkinElmer connected to a Mariner mass spectrometer. Samples (25 μl) of digested RCM-vWF and RCM-vWF-Ox (8 μg≈32 pmols) were added with an acidic solution (60 μl) of 6 M GdmCl (guanidinium chloride) in 1% (v/v) aqueous HCOOH (formic acid) and loaded on to a Grace-Vydac C18 microbore column (1×100 mm, 5 μm granulometry) equilibrated for 20 min with 1% aqueous HCOOH, containing 1% acetonitrile, and eluted with an exponential acetonitrile–1% HCOOH gradient from 1 to 80% in 50 min at a flow rate of 10 μl/min. Spray tip potential was set at 4.0 kV, while the nozzle potential and temperature were set at 200 V and 140°C respectively. The resulting chromatographic trace was obtained by recording the TIC (total ion current) produced at the microchannel plate detector as a function of time. The reference PTPs Met-PTP and MetSO-PTP were analysed by LC-MS under identical conditions.

MS/MS (tandem MS) measurements were performed on a Micro Q-TOF mass spectrometer from Micromass equipped with a Z-spray nanoflow electrospray ionization interface and connected to a model CapLC capillary HPLC system from Waters. Mass spectra of the peptide digests of RCM-vWF and RCM-vWF-Ox were acquired using the nanoelectrospray source operating at capillary, cone and extractor voltages of 2700, 35 and 1 V respectively (positive ion mode). Digested vWF samples (7 μl≈2 μg) were loaded on to a C18 (75 μm×150 mm, 3.5 μm granulometry) NanoEase column (Waters), eluted with a linear acetonitrile–0.1% HCOOH gradient from 5 to 70% in 42 min at a flow rate of 0.2 μl/min. MS/MS analyses were performed selectively for the proteolytic peptides having monoisotopic m/z values identical, within the error of mass determination (i.e. <20 ppm), with those of the reference PTPs Met-PTP [i.e. m/z 903.49(2+)] and MetSO-PTP [i.e. m/z 911.48(2+)].

RESULTS

Purification of vWF from healthy subjects and from HD patients

Blood samples from healthy subjects or from patients with CKD and a history of HD treatment were pooled, centrifuged and the resulting plasma samples were stored at −80°C. After thawing, vWF was purified from normal and CKD plasma following essentially the cryoprecipitation method (see the Experimental section) [39]. The cryoprecipitate was finally fractionated on a Sephacryl S-400 gel-filtration column (Figure 1). The purity and integrity of our vWF preparations were assessed by reducing SDS/PAGE (Figure 1, inset), showing a single band at approximately 250 kDa for vWF purified from either healthy subjects or HD patients. From 200 ml of plasma, approximately 200 μg of highly homogeneous and intact vWF were obtained. The concentration of the final purified vWF solutions was determined spectrophotometrically, and the quality of vWF preparations was assessed by immunoturbidometric assays that measure vWF:Ag and vWF:RiCof [34,40]. Our vWF preparations displayed good protein quality and function, with a vWF:RiCof/vWF:Ag ratio of 0.81±0.15.

Purification and analysis of HD-vWF by gel-filtration chromatography and SDS/PAGE

Figure 1
Purification and analysis of HD-vWF by gel-filtration chromatography and SDS/PAGE

An aliquot (500 μl) of the citrate solution deriving from the cryoprecipitated plasma of HD patients was loaded on to a Sephacryl S-400 column (see the Experimental section). vWF multimers were eluted with the void volume, as indicated. Inset: an aliquot (10 μg) of purified HD-vWF was analysed by SDS/PAGE on a gradient 4–15% gel, under reducing conditions. As a control, normal vWF, purified by the same procedure, was also loaded. STD, high-molecular mass protein standard; vWF, normal von Willebrand factor purified from healthy subjects; HD-vWF, von Willebrand factor purified from HD patients.

Figure 1
Purification and analysis of HD-vWF by gel-filtration chromatography and SDS/PAGE

An aliquot (500 μl) of the citrate solution deriving from the cryoprecipitated plasma of HD patients was loaded on to a Sephacryl S-400 column (see the Experimental section). vWF multimers were eluted with the void volume, as indicated. Inset: an aliquot (10 μg) of purified HD-vWF was analysed by SDS/PAGE on a gradient 4–15% gel, under reducing conditions. As a control, normal vWF, purified by the same procedure, was also loaded. STD, high-molecular mass protein standard; vWF, normal von Willebrand factor purified from healthy subjects; HD-vWF, von Willebrand factor purified from HD patients.

Determination of the carbonyl content and proteolysis by ADAMTS-13 of vWF purified from HD patients

vWF samples from patients (n=39) with ESRD were micropurified by immunoaffinity chromatography (see the Experimental section) and the carbonyl content was determined as a reliable marker of oxidative modifications in proteins [33]. Under oxidizing conditions, sensitive amino acid side chains (i.e. serine, threonine and tyrosine) are converted into their corresponding formyl/ketonic derivatives, with a resulting increase in the protein carbonyl groups. Notably, the carbonyl content (Figure 2A) of HD-vWF was appreciably higher than in normal vWF (i.e. 1.17±0.97 compared with 0.12±0.10 nmol/mg of protein respectively; P<0.0001). SDS-agarose gel electrophoresis, followed by Western-blot analysis, showed that HD-vWF contains a higher proportion of ultra-large multimers (UL-vWF) than normal vWF (Figure 2B). Furthermore, Figure 2(B) showed also that HD-vWF is significantly more resistant to proteolysis by ADAMTS-13 than normal vWF after 1 and 2 h of incubation.

Analysis of vWF from patients with ESRD

Figure 2
Analysis of vWF from patients with ESRD

(A) Determination of the carbonyl content in vWF from patients with ESRD and from healthy subjects. The horizontal lines represent the means of the corresponding data set. (B) Western blot electrophoretic analysis of vWF multimers from a patient with ESRD and a pool of 10 normal subjects. The samples were analysed by SDS-agarose electrophoresis on a 1.2% agarose gel, followed by immunoblotting with anti-human vWF polyclonal antibody (see the Experimental section). The horizontal line delimits the form with very high-molecular mass (up), which is more represented in the ESRD patients. The same immunoelectrophoretic run was used to investigate the sensitivity of purified vWF to cleavage by ADAMTS-13. vWF samples (100 μl, 20 μg/ml) were incubated for 1 and 2 h at 37°C with ADAMTS-13 (5 nM) in the presence of ristocetin (1.5 mg/ml) and then analysed by SDS-agarose electrophoresis and Western blotting.

Figure 2
Analysis of vWF from patients with ESRD

(A) Determination of the carbonyl content in vWF from patients with ESRD and from healthy subjects. The horizontal lines represent the means of the corresponding data set. (B) Western blot electrophoretic analysis of vWF multimers from a patient with ESRD and a pool of 10 normal subjects. The samples were analysed by SDS-agarose electrophoresis on a 1.2% agarose gel, followed by immunoblotting with anti-human vWF polyclonal antibody (see the Experimental section). The horizontal line delimits the form with very high-molecular mass (up), which is more represented in the ESRD patients. The same immunoelectrophoretic run was used to investigate the sensitivity of purified vWF to cleavage by ADAMTS-13. vWF samples (100 μl, 20 μg/ml) were incubated for 1 and 2 h at 37°C with ADAMTS-13 (5 nM) in the presence of ristocetin (1.5 mg/ml) and then analysed by SDS-agarose electrophoresis and Western blotting.

Mass spectrometric identification of MetSO at position 1606 of normal vWF oxidized in vitro with PN

Oxidation of plasma purified UL-vWF was carried out in vitro with PN (10 μM) for 60 min using the ‘pulsed oxidation’ procedure (see the Experimental section) developed previously for the synthetic vWF74 peptide [34]. This procedure aimed to possibly mimic the continuous flow of oxidant (1–10 μM) that may be produced, even for hours, at inflammatory sites in vivo [41]. The strategy we used in the present study for identifying MetSO at position 1606 of vWF-Ox is based on the TMS approach [42] and involves the following steps: (i) selection and production of PTPs with known physico-chemical properties (e.g. retention time in RP-HPLC, m/z ratio and fragmentation pattern in MS) that are unique to vWF with Met1606 in the unmodified (Met-PTP) or sulfoxide (MetSO-PTP) forms, to be used as reference peptides in targeted LC-MS experiments; (ii) production of vWF-Ox; (iii) enzymatic fragmentation of vWF and vWF-Ox with Glu-C protease; (iv) LC-MS analysis of the proteolysis mixture and identification of PTPs by TMS; and (v) determination of the amino acid sequence of PTPs in the proteolysis mixture by MS/MS analysis.

From the analysis of the theoretical fragmentation pattern of vWF with Glu-C endoprotease, the peptide sequence Q1599APNLVYMVTGNPASDE1615 was selected as a suitable PTP, representative of vWF with unmodified or oxidized Met1606. This peptide, in fact, apart from tyrosine and methionine, does not contain any other oxidizable amino acid that might otherwise complicate the interpretation of the resulting MS spectra. Using this approach, we showed that only Met1606 (but not Tyr1605) was modified. To prepare Met-PTP and MetSO-PTP, the synthetic peptides vWF74 and vWF74-MetSO, produced as detailed previously [34], were treated with Glu-C protease and the corresponding reaction mixtures fractionated by RP-HPLC (Figures 3A and 3B). All fragments were identified by MS analysis (see Figure 3C and Supplementary Table S1 at http://www.BiochemJ.org/bj/442/bj4420423add.htm), whereas the PTPs, Met-PTP [m/z 903.49(2+)] and MetSO-PTP [m/z 911.48(2+)], were purified on a microgram scale and then used as reference peptides in TMS analysis (Figure 4). Proteolysis reactions of unmodified and PN-treated vWF were analysed by LC-MS on a Mariner ESI–TOF spectrometer or on a Micro Q-TOF mass spectrometer (Figures 4A and 4B). The presence of the monoisotopic peptide species at m/z values 903.12(2+) and 911.51(2+) (Figures 4C and 4D), eluting in correspondence with the reference PTPs, allowed us to unequivocally identify Met-PTP and MetSO-PTP in the TIC trace of the proteolysis reaction of untreated and PN-treated UL-vWF. The mass difference between MetSO-PTP [1821.18 a.m.u. (atomic mass unit)] and Met-PTP (1805.10 a.m.u.) is 16 a.m.u., compatible with Met→MetSO oxidation. Notably, we could not identify any other peptide matching the mass of PTP nitrated at Tyr1605, in agreement with our previous results obtained with vWF74, showing that Tyr1605 remains unmodified [34]. In the final step, the chemical identity of Met-PTP and MetSO-PTP in the proteolysis reactions was confirmed by MS/MS sequence analysis (Figures 4E and 4F), whereby only peptide species with m/z values corresponding to those of Met-PTP and MetSO-PTP were selected (Figures 4C and 4D). The mass values of the peptides belonging to the b- and y-series and in particular the identification of eight fragment ion pairs (Supplementary Table S2 at http://www.BiochemJ.org/bj/442/bj4420423add.htm), differ on average by 16.00±0.01 a.m.u., provided clear-cut evidence for the presence of MetSO at position 1606 of vWF-Ox.

Production of the PTPs Met-PTP and MetSO-PTP

Figure 3
Production of the PTPs Met-PTP and MetSO-PTP

RP-HPLC analysis of the proteolysis reaction of vWF74 (A) and vWF74-MetSO (B) with Glu-C protease. The chemical identity of the proteolytic fragments, deriving from vWF74 (p1, p2, p3, etc.) or vWF74-MetSO (p1*, p2*, p3*, etc.), was established by MS analysis and is reported in Supplementary Table S1 at http://www.BiochemJ.org/bj/442/bj4420423add.htm. The peptides p4 and p2* were identified as the PTPs Met-PTP and MetSO-PTP respectively. As expected from the polar nature of MetSO, MetSO-PTP elutes at retention times much shorter than Met-PTP. (C) Amino acid sequence of the synthetic peptide vWF74. The identified fragments are underlined, while the PTP (i.e. Met-PTP or MetSO-PTP) is in bold. For clarity, the scissile tyrosine–methionine bond is in grey.

Figure 3
Production of the PTPs Met-PTP and MetSO-PTP

RP-HPLC analysis of the proteolysis reaction of vWF74 (A) and vWF74-MetSO (B) with Glu-C protease. The chemical identity of the proteolytic fragments, deriving from vWF74 (p1, p2, p3, etc.) or vWF74-MetSO (p1*, p2*, p3*, etc.), was established by MS analysis and is reported in Supplementary Table S1 at http://www.BiochemJ.org/bj/442/bj4420423add.htm. The peptides p4 and p2* were identified as the PTPs Met-PTP and MetSO-PTP respectively. As expected from the polar nature of MetSO, MetSO-PTP elutes at retention times much shorter than Met-PTP. (C) Amino acid sequence of the synthetic peptide vWF74. The identified fragments are underlined, while the PTP (i.e. Met-PTP or MetSO-PTP) is in bold. For clarity, the scissile tyrosine–methionine bond is in grey.

LC-MS/MS identification of MetSO at position 1606 of normal vWF oxidized in vitro with PN (Ox-vWF)

Figure 4
LC-MS/MS identification of MetSO at position 1606 of normal vWF oxidized in vitro with PN (Ox-vWF)

(A and B) LC-MS analysis of the proteolysis reaction of vWF (A) and Ox-vWF (B) with Glu-C protease. Intact (RCM-vWF) and oxidized (RCM-Ox-vWF) vWF samples (8 μg) with cysteines reduced and carboxamidomethylated were digested with Glu-C protease and analysed by LC-MS on a Mariner ESI–TOF spectrometer, using a microbore C18 column (see the Experimental section). The TIC traces of the reference peptides Met-PTP and MetSO-PTP (shaded peaks) are superimposed to those of the corresponding proteolytic mixtures of untreated and PN-oxidized vWF. (CF) MS and MS/MS spectra of the PTPs Met-PTP and MetSO-PTP selected in the proteolysis mixture of vWF (C and E) and Ox-vWF (D and F). Reaction samples (2 μg) were analysed by LC-MS on a Micro Q-TOF mass spectrometer, using a capillary C18 column. The peptide species with m/z values identical with those of Met-PTP and MetSO-PTP were selected and subjected to sequence analysis. The mass values and the amino acid sequence of the fragment ions identified in the b- and y-series are reported in Supplementary Table S2 at http://www.BiochemJ.org/bj/442/bj4420423add.htm.

Figure 4
LC-MS/MS identification of MetSO at position 1606 of normal vWF oxidized in vitro with PN (Ox-vWF)

(A and B) LC-MS analysis of the proteolysis reaction of vWF (A) and Ox-vWF (B) with Glu-C protease. Intact (RCM-vWF) and oxidized (RCM-Ox-vWF) vWF samples (8 μg) with cysteines reduced and carboxamidomethylated were digested with Glu-C protease and analysed by LC-MS on a Mariner ESI–TOF spectrometer, using a microbore C18 column (see the Experimental section). The TIC traces of the reference peptides Met-PTP and MetSO-PTP (shaded peaks) are superimposed to those of the corresponding proteolytic mixtures of untreated and PN-oxidized vWF. (CF) MS and MS/MS spectra of the PTPs Met-PTP and MetSO-PTP selected in the proteolysis mixture of vWF (C and E) and Ox-vWF (D and F). Reaction samples (2 μg) were analysed by LC-MS on a Micro Q-TOF mass spectrometer, using a capillary C18 column. The peptide species with m/z values identical with those of Met-PTP and MetSO-PTP were selected and subjected to sequence analysis. The mass values and the amino acid sequence of the fragment ions identified in the b- and y-series are reported in Supplementary Table S2 at http://www.BiochemJ.org/bj/442/bj4420423add.htm.

Noteworthy, at variance with the conventional ‘shotgun’ MS approach recently used for analysing HClO (hypochlorous acid)-mediated vWF oxidation [43], our TMS method only requires identification of the PTP MetSO-PTP in the LC-MS trace of proteolysed vWF, neglecting all other oxidative modifications that might occur at a variable extent in vWF sequence at sensitive amino acids (i.e. there are 169 cysteine, 56 phenylalanine, 53 histidine, 49 tyrosine, 41 methionine and 18 tryptophan residues in the vWF sequence) in different oxidant milieu (e.g. H2O2, O2·−, ONOO and HClO).

Mass spectrometric identification of MetSO at position 1606 of vWF purified from HD patients

vWF purified from the plasma of HD patients was treated according to the TMS procedure set up for normal vWF samples oxidized in vitro with PN (see above). The LC-MS analysis of the proteolysis reaction of HD-vWF with Glu-C endoprotease is shown in Figure 5(A). At retention times compatible with those of the reference PTPs Met-PTP and MetSO-PTP, two doubly charged monoisotopic species could be detected at m/z values of 903.49(2+) (Figure 5B) and 911.49(2+) (Figure 5C). These values are identical with those of the reference PTPs Met-PTP [m/z 903.47(2+)] and MetSO-PTP [m/z 911.50(2+)] and account for a mass difference of 16 a.m.u. Peptide sequencing by MS/MS measurements allowed us to confirm the presence of Met-PTP and MetSO-PTP in the proteolysis mixture of HD-vWF (Figures 5D and 5E). Even in this case, eight fragment ion pairs of the b- and y-series were identified in Met-PTP and MetSO-PTP (i.e. b8/b8*, b9/b9*, y10/y10*, etc.) (Table 1), with an average mass difference of 15.79±0.05 a.m.u. This value is compatible with the presence of MetSO in the fragment ions generated from oxidized vWF molecules that are present in HD-vWF sample.

LC-MS/MS identification of MetSO at position 1606 in vWF purified from HD patients

Figure 5
LC-MS/MS identification of MetSO at position 1606 in vWF purified from HD patients

(A) LC-MS analysis of the proteolysis reaction of HD-vWF with Glu-C protease. A sample (10 μg) of purified HD-vWF with cysteines reduced and carboxamidomethylated was digested with Glu-C protease and analysed by LC-MS on a Mariner ESI–TOF spectrometer, using a microbore C18 column (see the Experimental section). The arrows on the TIC trace indicate where the PTPs Met-PTP and MetSO-PTP elute. (BE) MS (B and C) and MS/MS (D and E) spectra of the PTPs selected in the proteolysis mixture of HD-vWF (see A). Reaction samples (2 μg) were analysed by LC-MS on a Micro Q-TOF mass spectrometer, using a capillary C18 column. The peptide species with m/z values identical with those of Met-PTP and MetSO-PTP were selected and subjected to sequence analysis. The mass values and the amino acid sequence of the fragment ions identified in the b- and y-series are reported in Table 1.

Figure 5
LC-MS/MS identification of MetSO at position 1606 in vWF purified from HD patients

(A) LC-MS analysis of the proteolysis reaction of HD-vWF with Glu-C protease. A sample (10 μg) of purified HD-vWF with cysteines reduced and carboxamidomethylated was digested with Glu-C protease and analysed by LC-MS on a Mariner ESI–TOF spectrometer, using a microbore C18 column (see the Experimental section). The arrows on the TIC trace indicate where the PTPs Met-PTP and MetSO-PTP elute. (BE) MS (B and C) and MS/MS (D and E) spectra of the PTPs selected in the proteolysis mixture of HD-vWF (see A). Reaction samples (2 μg) were analysed by LC-MS on a Micro Q-TOF mass spectrometer, using a capillary C18 column. The peptide species with m/z values identical with those of Met-PTP and MetSO-PTP were selected and subjected to sequence analysis. The mass values and the amino acid sequence of the fragment ions identified in the b- and y-series are reported in Table 1.

Table 1
MS/MS data of the PTPs Met-PTP and MetSO-PTP identified in HD-vWF

The PTPs Met-PTP and MetSO-PTP, selected during the LC-MS analysis of the proteolysis reaction of HD-vWF with Glu-C protease (see Figure 5A), were subjected to MS/MS analysis. Ion fragments are listed in order of increasing mass values. Fragment ions containing Met1606 in the unmodified or sulfoxide form are shown in bold.

Met-PTP MetSO-PTP 
Fragment ion (MH)+Amino acid sequence Fragment ion† (MH)+Amino acid sequence 
y148.08 (148.06) Glu1615 y148.06 (148.06) Glu1615 
y263.14 (263.09) Asp1614–Glu1615 y263.06 (263.09) Asp1614–Glu1615 
y350.19 (350.12) Ser1613–Glu1615 y350.13 (350.12) Ser1613–Glu1615 
y518.32 (518.21) Pro1611–Glu1615 y518.22 (518.21) Pro1611–Glu1615 
b623.48 (623.35) Gln1599–Val1604 b524.30 (524.28) Gln1599–Leu1603 
y689.45 (689.27) Gly1609–Glu1615 b623.37 (623.35) Gln1599–Val1604 
b786.60 (786.41) Gln1599–Tyr1605 y632.27 (632.25) Asn1610–Glu1615 
y790.46 (790.32) Thr1608–Glu1615 y689.29 (689.27) Gly1609–Glu1615 
y889.62 (889.39) Val1607–Glu1615 b786.45 (786.41) Gln1599–Tyr1605 
b8 917.68 (917.46) Gln1599–Met1606 y790.33 (790.32) Thr1608–Glu1615 
b9 1016.77 (1016.52) Gln1599–Val1607 y889.42 (889.39) Val1607–Glu1615 
y10 1020.61 (1020.43) Met1606–Glu1615 b8 933.52 (933.46) Gln1599–MetSO1606 
b10 1117.81 (1117.57) Gln1599–Thr1608 b9 1032.53 (1032.52) Gln1599–V1607 
y11 1183.73 (1183.49) Tyr1605–Glu1615 y10 1036.46 (1036.43) MetSO1606–Glu1615 
b12 1288.90 (1288.64) Gln1599–Asn1610 b10 1133.59 (1133.57) Gln1599–Thr1608 
b14 1457.08 (1456.73) Gln1599–Ala1612 b11 1190.58 (1190.59) Gln1599–Gly1609 
   y11 1199.55 (1199.49) Tyr1605–Glu1615 
   b12 1304.70 (1304.64) Gln1599–Asn1610 
   b14 1472.78 (1472.73) Gln1599–Ala1612 
Met-PTP MetSO-PTP 
Fragment ion (MH)+Amino acid sequence Fragment ion† (MH)+Amino acid sequence 
y148.08 (148.06) Glu1615 y148.06 (148.06) Glu1615 
y263.14 (263.09) Asp1614–Glu1615 y263.06 (263.09) Asp1614–Glu1615 
y350.19 (350.12) Ser1613–Glu1615 y350.13 (350.12) Ser1613–Glu1615 
y518.32 (518.21) Pro1611–Glu1615 y518.22 (518.21) Pro1611–Glu1615 
b623.48 (623.35) Gln1599–Val1604 b524.30 (524.28) Gln1599–Leu1603 
y689.45 (689.27) Gly1609–Glu1615 b623.37 (623.35) Gln1599–Val1604 
b786.60 (786.41) Gln1599–Tyr1605 y632.27 (632.25) Asn1610–Glu1615 
y790.46 (790.32) Thr1608–Glu1615 y689.29 (689.27) Gly1609–Glu1615 
y889.62 (889.39) Val1607–Glu1615 b786.45 (786.41) Gln1599–Tyr1605 
b8 917.68 (917.46) Gln1599–Met1606 y790.33 (790.32) Thr1608–Glu1615 
b9 1016.77 (1016.52) Gln1599–Val1607 y889.42 (889.39) Val1607–Glu1615 
y10 1020.61 (1020.43) Met1606–Glu1615 b8 933.52 (933.46) Gln1599–MetSO1606 
b10 1117.81 (1117.57) Gln1599–Thr1608 b9 1032.53 (1032.52) Gln1599–V1607 
y11 1183.73 (1183.49) Tyr1605–Glu1615 y10 1036.46 (1036.43) MetSO1606–Glu1615 
b12 1288.90 (1288.64) Gln1599–Asn1610 b10 1133.59 (1133.57) Gln1599–Thr1608 
b14 1457.08 (1456.73) Gln1599–Ala1612 b11 1190.58 (1190.59) Gln1599–Gly1609 
   y11 1199.55 (1199.49) Tyr1605–Glu1615 
   b12 1304.70 (1304.64) Gln1599–Asn1610 
   b14 1472.78 (1472.73) Gln1599–Ala1612 
*

Experimental and theoretical (in parentheses) mass values of singly charged fragment ions.

Fragment ions derived from MetSO-PTP.

Notably, from the intensity of the monoisotopic species of Met-PTP and MetSO-PTP in the free and Na+-bound forms (Figures 5B and 5C), a relative abundance higher than 10% was estimated for the oxidatively modified MetSO-PTP. To rule out the possibility that oxidized Met1606 (at least in part) could be also present in normal vWF, as a result of the basal oxidative potential in the plasma of healthy subjects or even artefactually generated in HD-vWF during purification and handling, we attempted to identify MetSO-PTP in the LC-MS trace of the Glu-C proteolysis reaction of normal vWF purified from healthy subjects. The data shown in Supplementary Figure S1 (at http://www.BiochemJ.org/bj/442/bj4420423add.htm) unequivocally indicate that, within the limitations of the analytical techniques used, the percentage of MetSO-PTP in normal vWF is lower than 1%, compared with the unmodified Met-PTP. These results allow us to reasonably interpret the increase of MetSO-PTP in HD-vWF as arising from the oxidative conditions unique to the plasma of HD patients. At this stage of the work, however, either limited availability of plasma samples and low yield of vWF purification impair quantification of Met1606 in vWF from individual patients.

DISCUSSION

OS is a generic term used for defining the imbalance of the equilibrium existing between the generation of ROS and their scavenging in vivo [44]. Shifting this equilibrium in favour of ROS production can lead to oxidation of biomacromolecules (e.g. proteins, lipids and nucleic acids), inducing cell death and tissue injury [45,46]. Biochemical and clinical data accumulated so far indicate that OS is involved in many clinical settings, also associated with pro-inflammatory states and dysfunction of vascular endothelium (i.e. CVD and neurodegenerative diseases, atherosclerosis, cancer, rheumatoid arthritis, inflammatory bowel disease, sepsis, CKD and T2D) [45].

PN and HClO are thought to be the major effectors of OS in vivo [41,47,48]. However, a realistic understanding of the relative importance of these two oxidants in health and disease should start from the knowledge of their different reactivity, cellular localization and in vivo function. PN is produced in vivo from NO (nitric oxide) and superoxide radical (O2) [46] mainly by vascular endothelial cells, where vWF is also stored and released; it easily crosses biological membranes and its concentration increases from 0.3 μM/s, under basal conditions, up to 30–300 μM/s at inflammatory sites in diseases related to chronic inflammatory states [41,45]. Furthermore, PN reacts directly with cysteine, methionine and tryptophan generating sulfenic acid, MetSO and 2-oxindolylalanine respectively, whereas it reacts indirectly with phenylalanine, histidine and tyrosine [47]. Compared with PN, HClO is a stronger oxidant, mainly involved in microbial killing and is produced in vivo (30–400 μM/h) by the H2O2/Cl myeloperoxidase system of activated PMNs, together with a burst of LSPs, as part of the mammalian immune defence system [29,48].

Recently, we have investigated the effects of vWF oxidation by PN or HClO on the susceptibility of vWF to proteolysis by physiologically relevant proteases, such as ADAMTS-13 and LSPs, and then related these effects to alterations of the haemostatic process observed in different clinical settings (i.e. T2D and sepsis) [31,34]. In particular, using the minimal vWF substrate for ADAMTS-13 (i.e. vWF74), we showed that oxidation of Met1606 in vWF to MetSO hinders cleavage by ADAMTS-13, with a resulting accumulation of the more pro-haemostatic vWF polymers, UL-vWF [34]. Conversely, in vitro treatment of vWF with activated PMNs, simultaneously releasing LSPs and ROS, does not reduce but enhances proteolysis of UL-vWF by LSPs [31]. Hence it seems that the same chemical modification (i.e. oxidation of Met1606) has opposite effects on the susceptibility of vWF to the proteases responsible for the regulation of its length (i.e. ADAMTS-13 and LSPs) and, ultimately, of its haemostatic potential in different clinical settings. Altogether, these considerations emphasize the different physiopathological meaning of vWF oxidation by PN and HClO and highlight the existence of two different pathways of coupled oxidative/proteolytic reactions involving the ADAMTS-13-vWF and LSPs-vWF systems. Chronic OS by PN, mainly released by perturbed endothelial cells in chronic inflammatory diseases (e.g. vascular aging and diabetes), has a negative effect on the ADAMTS-13/vWF pathway [34], with a resulting accumulation of pro-thrombotic UL-vWF polymers. Conversely, acute OS caused by HClO, rapidly released by activated PMNs during bacterial infection, has a positive effect on the LSPs/vWF pathway [31], thus (partially) compensating the inhibition that Met1606 oxidation exerts on ADAMTS-13 proteolysis. Furthermore, the enhanced proteolysis of oxidized vWF by LSPs, reducing the concentration of UL-vWF polymers amenable to interact with bacterial adhesins (see the Introduction), may be regarded as a mechanism aimed at limiting the vWF-mediated adhesion and invasion of bacteria in a highly oxidant milieu (i.e. sepsis), where ADAMTS-13 activity is heavily compromised [31].

Considering the concomitant increase of OS and vWF levels in CKD, in the present study, we chemically characterized the oxidative state of Met1606, as the major site of proteolytic regulation of vWF length. Hence we purified to homogeneity enough quantities of vWF from a pool of plasma from HD patients (Figure 1). As already found with vWF isolated from T2D patients [34], the average carbonyl content of HD-vWF was at least 10-fold higher than in normal vWF (Figure 2A) and favourably correlated with an increased proportion of UL-vWF polymers that were also more resistant to ADAMTS-13 cleavage (Figure 2B). A major achievement of this work relate to the development of a sensitive (2–10 μg of vWF) and reliable TMS method for detecting MetSO at position 1606 of UL-vWF samples. The experimental procedure was first set up and validated with normal vWF, oxidized in vitro with physiological concentrations of PN (Figures 3 and 4), and then used for analysing HD-vWF (Figure 5 and Table 1). MetSO was unequivocally detected at position 1606 of HD-vWF, even though quantitative MS analysis (Figure 5B) indicated that the proportion of vWF molecules containing MetSO at position 1606 was relatively small (i.e. >10%). However, this value represents the average extent of Met1606 oxidation in a vWF sample obtained from a pool of plasma derived from 30 patients undergoing HD therapy since a variable time period (i.e. from 1 to 15 years; see the Experimental section). Further work is required to correlate the extent of Met1606 modification in individual patient plasmas with the severity of CKD and/or with the time period of HD treatment. Moreover, it should be considered that haemostasis is a dynamic process resulting from the delicate equilibrium between finely regulated pro-coagulant and anti-coagulant systems, such that even small perturbation of this equilibrium can cause thrombotic or haemorrhagic effects [49]. For instance, in normal individuals only approximately 10% of circulating fibrinogen molecules contain the elongated γ-chain variant (termed γ′), derived from alternative splicing of mRNA and inserting 20 amino acids at the C-terminus of the γ chain. The γ′ chain binds to thrombin exosite 2, inhibits platelet activation by thrombin and its decreased expression is positively correlated with higher risk of thrombosis [50]. Likewise, in vivo nitration of Tyr292 and Tyr422 in the fibrinogen β-chain of smokers occurs with yields as low as 0.6–6%. Nevertheless, the kinetics of fibrin formation and degradation in these patients is markedly altered and tyrosine nitration has been proposed as an important risk factor for thrombosis [51]. In the case of vWF, tensile force acting on vWF molecules increases with the square of multimer length [24]. Therefore UL-vWF polymers are more easily stretched in shear flow [24] and start to form a protein network that efficiently adheres to platelet GpIb receptors, thus increasing the risk of thrombosis [23]. Under physiological conditions, this effect is opposed by shear-induced unfolding of the A2 domain and subsequent cleavage of the newly exposed Tyr1605–Met1606 bond by ADAMTS-13 [24]. Therefore it is not surprising that under high OS, even relatively small amounts of oxidized UL-vWF multimers, no longer sensitive to proteolysis by ADAMTS-13, may function as nucleators for the generation of networks of vWF fibres, thus amplifying platelet adhesion and thrombus formation. Likewise, UL-vWF polymers might increase the probability of vWF-mediated adhesion of bacteria to vascular vessels and their migration into tissues [27,28], thus expanding infection. Accordingly, previous studies showed that experimental group C Streptococcus-induced endocarditis failed to develop in pigs with von Willebrand disease, an inherited bleeding disorder characterized by loss of UL-vWF polymers [52].

In conclusion, in the present study, we have characterized the oxidative state of vWF purified from patients with CKD and developed a convenient MS method of general applicability for detecting MetSO at position 1606 as a ‘functional marker’ of the susceptibility of vWF to proteolysis in diseases associated with high OS. These results are unprecedented and may contribute to elucidate the molecular root leading to either thrombotic or septic complications in CKD.

Abbreviations

     
  • ADAMTS-13

    a disintegrin and metalloproteinase with thrombospondin motifs 13

  •  
  • a.m.u.

    atomic mass unit

  •  
  • CKD

    chronic kidney disease

  •  
  • CVD

    cardiovascular disease

  •  
  • DNP

    dinitrophenylhydrazine

  •  
  • ESI–TOF

    electrospray ionization–time-of-flight

  •  
  • ESRD

    end-stage renal disease

  •  
  • HD

    haemodialysis

  •  
  • HRP

    horseradish peroxidase

  •  
  • LC-MS

    liquid chromatography-MS

  •  
  • LSP

    leucocyte serine protease

  •  
  • MetSO

    methionine sulfoxide

  •  
  • MS/MS

    tandem MS

  •  
  • NT

    3-nitrotyrosine

  •  
  • OS

    oxidative stress

  •  
  • PMN

    polymorphonuclear cell

  •  
  • PN

    peroxynitrite

  •  
  • PTP

    proteotypic peptide

  •  
  • RCM

    reduced and carboxamidomethylated

  •  
  • RiCof

    ristocetin cofactor

  •  
  • ROS

    reactive oxygen species

  •  
  • RP-HPLC

    reverse-phase-HPLC

  •  
  • TFA

    trifluoroacetic acid

  •  
  • TIC

    total ion current

  •  
  • TMS

    targeted MS

  •  
  • T2D

    Type 2 diabetes

  •  
  • vWF

    von Willebrand factor

  •  
  • UL-vWF

    ultra large-vWF

  •  
  • vWF-Ox

    oxidized vWF

AUTHOR CONTRIBUTION

Vincenzo De Filippis designed the experiments, analysed the results and wrote the paper; Stefano Lancellotti, Fabio Maset, Barbara Spolaore, Nicola Pozzi and Laura Oggianu performed the experiments; Giovanni Gambaro and Lorenzo A. Calò supplied the plasma from HD patients and revised the paper; Raimondo De Cristofaro designed the experiments, analysed the results and revised the paper.

V.D.F. thanks Professor Angelo Fontana for providing accessibility to instrumentation for MS/MS analyses, Dr Chiara D'Orlando for helping us with the purification of HD-vWF and Dr Daniele Dalzoppo for critically reading the paper before submission.

FUNDING

This work was financially supported by the Ministero dell'Istruzione, dell'Università e della Ricerca Scientifica (MIUR) [PRIN2007 grant numbers 2007T9HTFB_002 (to V.D.F.) and 2007T9HTFB_003 (to R.D.C.)].

References

References
1
Go
A. S.
Chertow
G. M.
Fan
D.
McCulloch
C. E.
Hsu
C. Y.
Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization
N. Engl. J. Med.
2004
, vol. 
351
 (pg. 
1296
-
1305
)
2
United States Renal Data System
USRDS 2010 Annual Data Report: Atlas of Chronic Kidney Diseases and End-Stage Renal Disease in The United States
2010
Bethesda
National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases
3
Himmelfarb
J.
Hakim
R. M.
Oxidative stress in uremia
Curr. Opin. Nephrol. Hypertens.
2003
, vol. 
12
 (pg. 
593
-
598
)
4
Himmelfarb
J.
McMonagle
E.
Freedman
S.
Klenzak
J.
McMenamin
E.
Le
P.
Pupim
L. B.
Ikizler
T. A.
Oxidative stress is increased in critically ill patients with acute renal failure
J. Am. Soc. Nephrol.
2004
, vol. 
15
 (pg. 
2449
-
2456
)
5
Massy
Z. A.
Stenvinkel
P.
Drueke
T. B.
The role of oxidative stress in chronic kidney disease
Semin. Dial.
2009
, vol. 
22
 (pg. 
405
-
408
)
6
Walker
L. M.
Walker
P. D.
Imam
S. Z.
Ali
S. F.
Muldrew
K. L.
Mayeux
P. R.
Oxidative stress and reactive nitrogen species generation during renal ischemia
Toxicol. Sci.
2001
, vol. 
63
 (pg. 
143
-
148
)
7
Vinas
J. L.
Sola
A.
Hotter
G.
Mitochondrial NOS upregulation during renal I/R causes apoptosis in a peroxynitrite-dependent manner
Kidney Int.
2006
, vol. 
69
 (pg. 
1403
-
1409
)
8
Zimmermann
J.
Herrlinger
S.
Pruy
A.
Metzger
T.
Wanner
C.
Inflammation enhances cardiovascular risk and mortality in hemodialysis patients
Kidney Int.
1999
, vol. 
55
 (pg. 
648
-
658
)
9
Annuk
M.
Zilmer
M.
Lind
L.
Linde
T.
Fellström
B.
Oxidative stress and endothelial function in chronic renal failure
J. Am. Soc. Nephrol.
2001
, vol. 
12
 (pg. 
2747
-
2752
)
10
Luke
R. G
Chronic renal failure: a vasculopathic state
N. Engl. J. Med.
1998
, vol. 
339
 (pg. 
841
-
843
)
11
Ueland
P. M.
Mansoor
M. A.
Guttormsen
A. B.
Müller
F.
Aukrust
P.
Refsum
H.
Svardal
A. M.
Reduced, oxidized and protein-bound forms of homocysteine and other aminothiols in plasma comprise the redox thiol status: a possible element of the extracellular antioxidant defense system
J. Nutr.
1996
, vol. 
126
 
4 Suppl.
(pg. 
1281S
-
1284S
)
12
Halliwell
B.
Albumin: an important extracellular antioxidant?
Biochem. Pharmacol.
1998
, vol. 
37
 (pg. 
569
-
571
)
13
Bostom
A. G.
Lathrop
L.
Hyperhomocysteinemia in end-stage renal disease: prevalence, etiology and potential relationship to arteriosclerotic outcomes
Kidney Int.
1997
, vol. 
52
 (pg. 
10
-
20
)
14
Ward
R.
McLeish
K.
Hemodialysis with cellulose membranes primes the neutrophil oxidative burst
Artif. Organs
1995
, vol. 
19
 (pg. 
801
-
807
)
15
Małyszko
J. S.
Małyszko
J.
Hryszko
T.
Koźminski
P.
Pawlak
K.
Mysliwiec
M.
Markers of endothelial damage in patients on hemodialysis and hemodiafiltration
J. Nephrol.
2006
, vol. 
19
 (pg. 
150
-
154
)
16
Kirmizis
D.
Tsiandoulas
A.
Pangalou
M.
Koutoupa
E.
Rozi
P.
Protopappa
M.
Barboutis
K.
Validity of plasma fibrinogen, D-dimer, and the von Willebrand factor as markers of cardiovascular morbidity in patients on chronic hemodialysis
Med. Sci. Monit.
2006
, vol. 
12
 (pg. 
CR55
-
CR62
)
17
Wagner
D. D.
Cell biology of von Willebrand factor
Annu. Rev. Cell Biol.
1990
, vol. 
6
 (pg. 
217
-
246
)
18
Sadler
J. E.
Biochemistry and genetics of von Willebrand factor
Annu. Rev. Biochem.
1998
, vol. 
67
 (pg. 
395
-
424
)
19
Sadler
J. E.
von Willebrand factor assembly and secretion
J. Thromb. Haemostasis
2009
, vol. 
7
 
Suppl. 1
(pg. 
24
-
27
)
20
Schneider
S. W.
Nuschele
S.
Wixforth
A.
Gorzelanny
C.
Alexander-Katz
A.
Netz
R. R.
Schneider
M. F.
Shear-induced unfolding triggers adhesion of von Willebrand factor fibers
Proc. Natl. Acad. Sci. U.S.A.
2007
, vol. 
104
 (pg. 
7899
-
903
)
21
Siedlecki
C. A.
Lestini
B. J.
Kottke-Marchant
K. K.
Eppell
S. J.
Wilson
D. L.
Marchant
R. E.
Shear-dependent changes in the three-dimensional structure of human von Willebrand factor
Blood
1996
, vol. 
88
 (pg. 
2939
-
2950
)
22
Sadler
J. E.
von Willebrand factor: two sides of a coin
J. Thromb. Haemostasis
2005
, vol. 
3
 (pg. 
1702
-
1709
)
23
Ruggeri
Z. M.
The role of von Willebrand factor in thrombus formation
Thromb. Res.
2007
, vol. 
120
 
Suppl. 1
(pg. 
S5
-
S9
)
24
Zhang
X.
Halvorsen
K.
Zhang
C. Z.
Wong
W. P.
Springer
T. A.
Mechanoenzymatic cleavage of the ultralarge vascular protein von Willebrand factor
Science
2009
, vol. 
324
 (pg. 
1330
-
1334
)
25
Furlan
M.
Robles
R.
Lamie
B.
Partial purification and characterization of a protease from human plasma cleaving von Willebrand factor to fragments produced by in vivo proteolysis
Blood
1996
, vol. 
87
 (pg. 
4223
-
4234
)
26
Chung
D. W.
Fujikawa
K.
Processing of von Willebrand factor by ADAMTS-13
Biochemistry
2002
, vol. 
41
 (pg. 
11065
-
11070
)
27
Hartleib
J.
Kohler
N.
Dickinson
R. B.
Chhatwal
G. S.
Sixma
J. J.
Hartford
O. M.
Foster
T. J.
Peters
G.
Kehrel
B. E.
Herrmann
M.
Protein A is the von Willebrand factor binding protein on Staphylococcus aureus
Blood
2000
, vol. 
96
 (pg. 
2149
-
2156
)
28
Petri
B.
Broermann
A.
Li
H.
Khandoga
A. G.
Zarbock
A.
Krombach
F.
Goerge
T.
Schneider
S. W.
Jones
C.
Nieswandt
B.
, et al. 
Von Willebrand factor promotes leukocyte extravasation
Blood
2010
, vol. 
116
 (pg. 
4712
-
4719
)
29
Klebanoff
S. J.
Myeloperoxidase: friend and foe
J. Leukoc. Biol.
2005
, vol. 
77
 (pg. 
598
-
625
)
30
Raife
T. J.
Cao
W.
Atkinson
B. S.
Bedell
B.
Montgomery
R. R.
Lentz
S. R.
Johnson
G. F.
Zheng
X. L.
Leukocyte proteases cleave von Willebrand factor at or near the ADAMTS13 cleavage site
Blood
2009
, vol. 
114
 (pg. 
1666
-
1674
)
31
Lancellotti
S.
De Filippis
V.
Pozzi
N.
Oggianu
L.
Rutella
S.
Scaglione
G. L.
Maset
F.
Peyvandi
F.
Mannucci
P. M.
De Cristofaro
R.
Oxidized von Willebrand factor is efficiently cleaved by serine proteases from primary granules of leukocytes: divergence from ADAMTS-13, J
Thromb. Haemostasis
2011
, vol. 
9
 (pg. 
1620
-
1627
)
32
Brownlee
M.
Biochemistry and molecular cell biology of diabetic complications
Nature
2001
, vol. 
414
 (pg. 
813
-
820
)
33
Reznick
A. Z.
Packer
L.
Oxidative damage to proteins: spectrophotometric method for carbonyl assay
Methods Enzymol.
1994
, vol. 
233
 (pg. 
357
-
363
)
34
Lancellotti
S.
De Filippis
V.
Pozzi
N.
Peyvandi
F.
Palla
R.
Rocca
B.
Rutella
S.
Pitocco
D.
Mannucci
P. M.
De Cristofaro
R.
Formation of methionine sulfoxide by peroxynitrite at position 1606 of von Willebrand factor inhibits its cleavage by ADAMTS-13: A new prothrombotic mechanism in diseases associated with oxidative stress
Free Radical Biol. Med.
2010
, vol. 
48
 (pg. 
446
-
456
)
35
Pozzi
N.
Lancellotti
S.
De Cristofaro
R.
De Filippis
V.
Modeling ADAMTS13–vWF interaction: implications for oxidative stress-related cardiovascular diseases and type-2A von Willebrand disease
Biophys. Chem.
2012
, vol. 
160
 (pg. 
1
-
11
)
36
Foley
R.
Parfrey
P. S.
Sarnak
M. J.
Epidemiology of cardiovascular disease in chronic renal disease
J. Am. Soc. Nephrol.
1998
, vol. 
32
 
12 Suppl.
(pg. 
S16
-
S23
)
37
Mehta
R. L.
Bouchard
J.
Soroko
S. B.
Ikizler
T. A.
Paganini
E. P.
Chertow
G. M.
Himmelfarb
J.
Sepsis as a cause and consequence of acute kidney injury: program to improve care in acute renal disease
Intensive Care Med.
2011
, vol. 
37
 (pg. 
241
-
248
)
38
van Schooten
C. J.
Denis
C. V.
Lisman
T.
Eikenboom
J. C.
Leebeek
F. W.
Goudemand
J.
Fressinaud
E.
van den Berg
H. M.
de Groot
P. G.
Lenting
P. J.
Variations in glycosylation of von Willebrand factor with O-linked sialylated T antigen are associated with its plasma levels
Blood
2007
, vol. 
109
 (pg. 
2430
-
2437
)
39
Thorell
L.
Blomback
B.
Purification of the factor VIII complex
Thromb. Res.
1984
, vol. 
35
 (pg. 
431
-
450
)
40
De Cristofaro
R.
Peyvandi
F.
Palla
R.
Lavoretano
S.
Lombardi
R.
Merati
G.
Romitelli
F.
Di Stasio
E.
Mannucci
P. M.
Role of chloride ions in modulation of the interaction between von Willebrand factor and ADAMTS-13
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
23295
-
13302
)
41
Ferrer-Sueta
G.
Radi
R.
Chemical biology of peroxynitrite: kinetics, diffusion, and radicals
ACS Chem. Biol.
2009
, vol. 
4
 (pg. 
161
-
177
)
42
Hüttenhain
R.
Malmström
J.
Picotti
P.
Aebersold
R.
Perspectives of targeted mass spectrometry for protein biomarker verification
Curr. Opin. Chem. Biol.
2009
, vol. 
13
 (pg. 
518
-
525
)
43
Chen
J.
Fu
X.
Wang
Y.
Ling
M.
McMullen
B.
Kulman
J.
Chung
D. W.
Lopez
J. A.
Oxidative modification of von Willebrand factor by neutrophil oxidants inhibits its cleavage by ADAMTS13
Blood
2010
, vol. 
115
 (pg. 
706
-
712
)
44
Forman
H. J.
Maiorino
M.
Ursini
F.
Signaling functions of reactive oxygen species
Biochemistry
2010
, vol. 
49
 (pg. 
835
-
842
)
45
Pacher
P.
Beckman
J. S.
Liaudet
L.
Nitric oxide and peroxynitrite in health and disease
Physiol. Rev.
2007
, vol. 
87
 (pg. 
315
-
424
)
46
Liaudet
L.
Vassalli
G.
Pacher
P.
Role of peroxynitrite in the redox regulation of cell signal transduction pathways
Front. Biosci.
2009
, vol. 
14
 (pg. 
4809
-
4814
)
47
Alvarez
B.
Radi
R.
Peroxynitrite reactivity with amino acids and proteins
Amino Acids
2003
, vol. 
25
 (pg. 
295
-
311
)
48
Hawkins
C. L.
Pattison
D. I.
Davies
M. J.
Hypochlorite-induced oxidation of amino acids, peptides and proteins
Amino Acids
2003
, vol. 
25
 (pg. 
259
-
274
)
49
Dahlbäck
B.
Blood coagulation
Lancet
2000
, vol. 
355
 (pg. 
1627
-
1632
)
50
Lancellotti
S.
Rutella
S.
De Filippis
V.
Pozzi
N.
Rocca
B.
De Cristofaro
R.
Fibrinogen elongated γ-chain inhibits thrombin-induced platelet response, hindering the interaction with different receptors
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
30193
-
30204
)
51
Parastatidis
I.
Thomson
L.
Burke
A.
Chernysh
I.
Nagaswami
C.
Visser
J.
Stamer
S.
Liebler
D. C.
Koliakos
G.
Heijnen
H. F.
, et al. 
Fibrinogen beta-chain tyrosine nitration is a prothrombotic risk factor
J Biol. Chem.
2008
, vol. 
283
 (pg. 
33846
-
33853
)
52
Johnson
C. M.
Bowie
E. J.
Pigs with von Willebrand disease may be resistant to experimental infective endocarditis
J. Lab. Clin. Med.
1992
, vol. 
120
 (pg. 
553
-
558
)

Author notes

2

Present address: Department of Biochemistry and Molecular Biology, Doisy Research Center, Saint Louis University, 1100 South Grand Blvd, St. Louis, MO, 63104, U.S.A.

Supplementary data