BLOOD COMPONENTS Increasing susceptibility of nitric oxide–mediated inhibitory platelet signaling during storage of apheresis-derived platelet concentrates Anna Kobsar, Erdwine Klinker, Sabine Kuhn, Angela Koessler, Pinar Yilmaz, Markus Boeck, and Juergen Koessler

BACKGROUND: Storage of platelets (PLTs) affects PLT integrity and functionality, a process named the PLT storage lesion. Normal PLT function essentially depends on the balanced interaction of activating and inhibitory signaling pathways. As there are poor data on the alterations of inhibitory signaling during storage of PLT concentrates, this study investigates the modulation capability of the cyclic nucleotide–mediated inhibitory pathways by use of the nitric oxide donor diethylamine diazenium diolate (DEA/NO). STUDY DESIGN AND METHODS: PLTs were obtained from whole blood (WB) and from apheresis-derived PLT concentrates (APCs) stored for 0, 2, and 5 days. Vasodilator-stimulated phosphoprotein (VASP) phosphorylation, cyclic nucleotide concentrations, fibrinogen binding, and agonist-induced aggregation were measured without or after stimulation with DEA/NO. RESULTS: DEA/NO-induced VASP phosphorylation was significantly higher in PLTs from APCs on Days 2 and 5 compared to WB, conditioned by a stronger increase of cyclic guanosine monophosphate (cGMP), but not cyclic adenosine monophosphate (cAMP), in stored PLTs. A quantity of 5 nmol/L DEA/NO neither influenced thrombin receptor activator peptide 6 and collagen-induced aggregation nor fibrinogen binding in freshly collected PLTs, whereas it significantly inhibited both in stored PLTs. CONCLUSION: Stored PLTs showed an impairment of intracellular cGMP regulation, resulting in exceeding inhibition of agonist-induced aggregation and fibrinogen binding in the course of storage. The observed effects could be an important mechanism contributing to the storage lesion with reduced activating potential of PLTs.

1782

TRANSFUSION Volume 54, July 2014

P

latelets (PLTs) in apheresis-derived PLT concentrates (APCs) undergo changes during storage resulting in the reduction of their integrity and functionality, a process named the PLT storage lesion.1 Previous investigations of stored APCs revealed an increased expression of CD62P, indicating PLT preactivation and decreased expression of glycoprotein GPIIb/ IIIa and GPIb/IX complex on the PLT surface.2 Furthermore, both the number of high-affinity thrombin receptors and thrombin-induced fibrinogen binding are reduced throughout storage of APCs.3 In addition, a progressive functional impairment of stored PLTs was described, associated with diminished adhesive capacities4 or reduced agonist-induced PLT aggregation.5-7 Normal PLT function is also conditioned by the inhibitory regulation of activating signal cascades mediated by cyclic adenosine monophosphate (cAMP)- and cyclic guanosine monophosphate (cGMP)-dependent

ABBREVIATIONS: APC(s) = apheresis-derived platelet concentrate(s); cAMP = cyclic adenosine monophosphate; cGMP = cyclic guanosine monophosphate; DEA/NO = nitric oxide donor diethylamine diazenium diolate; NO = nitric oxide; ODQ = 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one; PDE5A = phosphodiesterase 5A; PRP = platelet-rich plasma; sGC = soluble guanylate cyclase; TRAP-6 = thrombin receptor activator peptide 6; VASP = vasodilator-stimulated phosphoprotein; WB = whole blood. From the Institute of Transfusion Medicine and Haemotherapy, University of Wuerzburg, Wuerzburg, Germany. Address correspondence to: Juergen Koessler, MD, Institute of Transfusion Medicine and Haemotherapy, University of Wuerzburg, Oberduerrbacher Straße 6, D-97080 Wuerzburg, Germany; e-mail: [email protected]. Received for publication July 31, 2013; revision received October 31, 2013, and accepted November 13, 2013. doi: 10.1111/trf.12584 © 2014 AABB TRANSFUSION 2014;54:1782-1789.

NO-MEDIATED SIGNALING IN STORED PLTs

phosphorylation of target proteins.8 The vasodilatorstimulated phosphoprotein (VASP) is an actin-binding protein and the main substrate for both cAMP- and cGMPdependent protein kinases (PKA and PKG) that are able to phosphorylate this protein at Ser157 and Ser239 with different affinities in PLTs.8-11 Once phosphorylated, it retains PLT GPIIb/IIIa in the resting conformation leading to inhibition of fibrinogen binding, adhesion, and aggregation.12,13 Nitric oxide (NO) is a potent vasodilator and inhibitor of PLT function. In vivo it is produced by endothelial cells. In PLTs, NO stimulates soluble guanylate cyclase (sGC), resulting in increased cGMP production followed by PKG activation and VASP phosphorylation.8,14 Profound data on the alterations of the inhibitory signaling pathways in PLTs under storage are lacking. Recently, we were able to demonstrate that phosphodiesterase 5A (PDE5A) activity decreases during storage of APCs, accompanied by increasing cGMP levels and enhanced basal VASP phosphorylation.15 The aim of this study was to investigate the modulation capacity of the inhibitory cyclic nucleotide–mediated signaling in the course of PLT storage. For this purpose, the NO donor diethylamine diazenium diolate (DEA/ NO)16,17 was used to stimulate the inhibitory PLT pathways.18 DEA/NO was chosen because of its short half-life (approx. 2-4 min) compared to substances such as spermine NONOate (approx. 40 min).19 DEA/NO was applied in low concentrations (2 and/or 5 nmol/L, with submaximal inhibitory effects) to avoid permanent or excessive NO release that could potentially result in a nonphysiologic overstimulation of inhibitory signaling cascades. In addition, the functional effects of NO release in PLTs of stored APCs was measured by agonist-induced light transmission aggregometry and fibrinogen binding.

MATERIALS AND METHODS Materials DEA/NO was obtained from Enzo Life Sciences GmbH (Loerrach, Germany). 1H-[1,2,4]Oxadiazolo[4,3a]quinoxalin-1-one (ODQ) was from Tocris Bioscience (Bristol, UK). Thrombin receptor activator peptide 6 (TRAP-6) was obtained from BACHEM (Weil am Rhein, Germany). Collagen reagent Horm was from NYCOMED (Axis-Shield, Liverpool, UK). Mouse monoclonal fluorescein isothiocyanate (FITC)-conjugated anti-fibrinogen antibody, Clone 9F9, immunoglobulin (Ig)G1-FITC and corresponding FITC-conjugated isotype control IgG1, Clone 2DNP-2H11/2H12, were from STAGO Germany (Duesseldorf, Germany). Rabbit monoclonal anti-Panactin antibody was from New England Biolabs GmbH (Frankfurt am Main, Germany). Phospho-VASP Ser239 and phospho-VASP Ser157 antibodies were from Nanotools (Teningen, Germany). Horseradish peroxidase (HRP)-

conjugated goat anti-rabbit and anti-mouse antibodies were from Bio-Rad Laboratories, Inc. (Muenchen, Germany).

Blood collection and APCs Venous blood and APCs were obtained from seven informed healthy voluntary donors (aged 21 to 49 years, three male, four female). Peripheral blood was collected in polystyrene tubes containing 3.2% citrate buffer (106 mmol/L trisodium citrate, Sarstedt, Nuembrecht, Germany). APCs (2.5 × 1011 PLTs in 250 mL of plasma) were collected using automated blood collection devices with software (Trima Accel and LRS PLT, Plasma Set 5.1, CaridianBCT, now Terumo BCT, Lakewood, CO) according to current guidelines and the approval of authorities. The ratio of inlet blood volume to anticoagulant (ACD-A) was 10:1. The APCs were stored for 5 days under continuous agitation on a flatbed shaker at 22 ± 2°C. On Days 0, 2, and 5 samples from APCs were taken for analysis under sterile conditions. Our studies with human PLTs were approved by our local ethics committee of the University of Wuerzburg. The study was performed according to our institutional guidelines and to the Declaration of Helsinki.

Preparation and stimulation of washed human PLTs Washed PLTs were prepared as described.10 Briefly, 3 mmol/L ethylene glycol tetraacetic acid (EGTA) was added to whole blood (WB) to prevent PLT activation. PLTrich plasma (PRP) was obtained by centrifugation at 280 × g for 5 minutes. Subsequently, samples of PRP (or APCs after addition of 3 mmol/L EGTA) were centrifuged at 430 × g for 10 minutes. The pelleted PLTs were then washed once in CGS buffer (120 mmol/L sodium chloride, 12.9 mmol/L trisodium citrate, 30 mmol/L d-glucose, pH 6.5) and resuspended in HEPES buffer (150 mmol/L NaCl, 5 mmol/L KCl, 1 mmol/L MgCl2, 10 mmol/L d-glucose, 10 mmol/L HEPES, pH 7.4) to a final concentration of 3 × 108 PLTs/mL for cyclic nucleotide samples or 1 × 108 PLTs/mL for Western blot samples. After resting for 15 minutes in a 37°C water bath, washed PLTs were incubated for 5 minutes with buffer or 10 μmol/L ODQ followed by 2 minutes of stimulation with 2 or 5 nmol/L DEA/NO. The reaction was either stopped with sodium dodecyl sulfate (SDS) loading buffer (200 mmol/L TrisHCl, pH 6.7, 10% 2-mercaptoethanol, 6% SDS, 15% glycerol, and 0.03% bromophenol blue) for Western blot or with 5% trichloroacetic acid for cyclic nucleotide samples.

Western blot analysis Washed PLTs were separated by SDS-polyacrylamide gel electrophoresis and then transferred onto nitrocellulose Volume 54, July 2014

TRANSFUSION

1783

KOBSAR ET AL.

experiments, each made with different blood donors. Differences between groups were analyzed by Wilcoxon matched-pairs test. The t test was used when appropriate. p values of less than 0.05 were considered significant.

membranes. The membranes were incubated with appropriate primary antibodies overnight at 4°C. For visualization of the signal, goat anti-rabbit or anti-mouse IgGs conjugated with HRP were used as secondary antibodies, followed by enhanced chemiluminescence detection (GE Healthcare, Piscataway, NJ). Blots were analyzed densitometrically using computer software (NIH Image J, http://rsbweb.nih.gov/ij/) for uncalibrated optical density.

RESULTS Basal and DEA/NO-induced PLT VASP phosphorylation levels increase during APC storage

Flow cytometry analysis of human PLTs

The basal levels of VASP phosphorylation at Ser239 were comparable in freshly collected PLTs (from WB) and in APCs on Day 0, directly after apheresis, and increased significantly during storage, twofold on Day 2 and almost fivefold on Day 5 (Fig. 1).15 A quantity of 2 nmol/L DEA/NO induced a twofold higher VASP phosphorylation at Ser239 in APCs on Day 0 compared to WB. On Days 2 and 5, the stimulated levels were four- and fivefold higher (Fig. 1). At all measuring points, stimulation with 5 nmol/L DEA/NO resulted in a stronger VASP phosphorylation at Ser239 than stimulation with 2 nmol/L. However, the differences became less pronounced in the course of APC

A quantity of 10 μL of PRP or samples from APCs, diluted with plasma to PRP PLT concentration, were incubated with 10 μL of mouse anti-fibrinogen FITC-conjugated antibody or mouse FITC-conjugated isotype control and then incubated for 2 minutes with buffer or 5 nmol/L DEA/NO followed by 2 minutes of stimulation with 5 μmol/L TRAP-6. Samples were stopped with 0.1% formaldehyde, diluted with 400 μL of phosphate-buffered saline/5 mmol/L glucose/0.5% bovine serum albumin, and subsequently analyzed on a flow cytometer (FACSCalibur, Becton Dickinson, Franklin Lakes, NJ) using computer software (CELLQuest, Version 6.0, Becton Dickinson). The PLT population was identified by its forward and side scatter distribution and 20,000 events were analyzed for mean fluorescence.

+ + + + + +

cAMP and cGMP measurement cAMP and cGMP in washed PLTs were detected by cAMP enzyme-linked immunoassay (EIA) and GMP EIA kits, respectively, following the manufacturer’s instructions (Cayman Chemical, Hamburg, Germany).

Statistical analysis All experiments were performed at least in triplicate and data are presented as mean ± one standard deviation (SD). The n values refer to the number of 1784

TRANSFUSION Volume 54, July 2014

+

4

*

3

1

*

*

* 2

*

*

+

* *

+10 mmol/L ODQ

+10 mmol/L ODQ

+10 mmol/L ODQ

+10 mmol/L ODQ

0

2n m 5n ol C m /L on ol DE tro /L A l D /N 2n EA O m 5n ol C /N m /L on O ol DE tro /L A l 2n D /N E O m 5n ol C A/N / m L on O ol D tr /L EA ol D /N 2n EA O m 5n ol C /N / m L on O ol D tr /L EA ol 2n D /N EA O m 5n ol C /N m /L on O ol DE tro /L A l D /N 2n EA O m 5n ol C /N m /L on O ol DE tro /L A l 2n D /N E O m 5n ol C A/N / m L on O ol D tr /L EA ol D /N 2n EA O m 5n ol C /N / m L on O ol D tr /L EA ol D /N EA O /N O

PLT aggregation was measured using an aggregometer (APACT 4004, LabiTec, Ahrensburg, Germany). PRP or samples from APCs, diluted with plasma to PRP PLT concentration, were incubated with buffer or 5 nmol/L DEA/NO for 2 minutes. PLT aggregation was induced by addition of 5 μmol/L TRAP-6 or 5 or 10 μg/mL collagen. Aggregation was measured for 5 minutes under continuous stirring at 1000 rpm and 37°C.

Phospho-VASP Ser 239 (arbitrary units)

PLT aggregation

WB/PRP, Day 0

APCs, Day 0

APCs, Day 2

APCs, Day 5

Fig. 1. VASP Ser239 phosphorylation in DEA/NO-stimulated washed PLTs from WB and from APCs on Days 0, 2, and 5 is mediated by the activation of sGC. Washed human PLTs from WB or APCs (1 × 108/mL) were incubated with buffer or 10 μmol/L sGC inhibitor ODQ for 5 minutes, followed by stimulation with indicated concentrations of DEA/NO for 2 minutes. Samples were stopped with SDS loading buffer and analyzed on Western blot for VASP Ser239 phosphorylation. After scanning, bands were quantified by the Image J program. VASP phosphorylation in the samples was normalized to actin control. Blots are representative of six independent experiments. (Results are given as mean ± SD; n = 6; *p < 0.05 compared to Day 0; +p < 0.05 compared as indicated.)

NO-MEDIATED SIGNALING IN STORED PLTs

tion with 5 nmol/L DEA/NO, the accumulation of cGMP was signifi+ 20 cantly stronger in PLTs from APCs on * * Day 2 or Day 5 than from WB or APCs 15 on Day 0. The induced increase was * * 7.1 ± 1.4-fold in PLTs from WB, 10 * * 13.2 ± 1.9-fold in APCs on Day 2, and +10 mmol/L ODQ * +10 mmol/L ODQ +10 mmol/L ODQ +10 mmol/L ODQ 5 * 14.3 ± 2.2-fold in APCs on Day 5. A similar trend was found for the stimu0 lation with 2 nmol/L DEA/NO in APCs on Days 2 and 5, but not yet significant. ODQ effectively prevented DEA/NO-induced cGMP accumulation (Fig. 2). There were no changes of basal WB/PRP, Day 0 APCs, Day 0 APCs, Day 2 APCs, Day 5 cAMP concentrations under storage as described previously.15 After Fig. 2. DEA/NO-induced changes of intracellular cGMP concentrations in washed PLTs stimulation with DEA/NO, cAMP from WB and from APCs on Day 0, Day 2, or Day 5 are mediated by the activation of 8 levels increased only weakly, up to a sGC. Washed human PLTs from WB or APCs (1 × 10 /mL) were incubated with buffer or maximum of approximately twofold 10 μmol/L sGC inhibitor ODQ for 5 minutes, followed by stimulation with indicated with 5 nmol/L DEA/NO (data not concentrations of DEA/NO for 2 minutes and stopped with 5% trichloroacetic acid. shown). A quantity of 2 nmol/L After extraction with ether, the contents of cGMP were measured with the cGMP EIA DEA/NO induced a significant elevakit according to the manufacturer’s instructions. Results are presented as fold accution of the intracellular cAMP concenmulation compared to cyclic nucleotide levels in PLTs from WB and APCs on Day 0 tration (1.75-fold) in freshly collected (mean ± SD; n = 7; *p < 0.05 compared to Day 0; +p < 0.05 compared as indicated). PLTs from WB, but not in PLTs from stored APCs. In contrast, stimulation with 5 nmol/L storage. Compared to WB, an increase of 22% in APCs on DEA/NO resulted in a similar and significant cAMP accuDay 0, of 74% on Day 2, and of approximately twofold on mulation in all investigated samples: 2.1 ± 0.3-fold in WB, Day 5 was observed after stimulation with 5 nmol/L 1.7 ± 0.1-fold in APCs on Day 0, 1.7 ± 0.1-fold in APCs on DEA/NO (Fig. 1). Day 2, and 1.7 ± 0.3-fold in APCs on Day 5. ODQ comStimulated VASP phosphorylation levels at Ser157 were pletely prevented DEA/NO-induced changes of the cAMP comparable in WB and in APCs on Day 0 after apheresis content in PLTs (data not shown). (data not shown). Incubation with 5 nmol/L DEA/NO resulted in stronger VASP phosphorylation at Ser157 than incubation with 2 nmol/L again. However, under APC DEA/NO inhibits TRAP-6–induced aggregation and storage, the excess of VASP phosphorylation became relafibrinogen binding in stored PLTs tively less emphasized with 5 nmol/L DEA/NO. On Day 2 of APC storage, 2 nmol/L DEA/NO-stimulated PLT VASP The functional effect of DEA/NO on fresh and stored phosphorylation at Ser157 was approximately 2.4-fold and, PLTs was investigated by TRAP-6–induced aggregation and fibrinogen-binding studies. A quantity of 5 nmol/L on Day 5, almost threefold stronger compared to WB. DEA/NO influenced 5 μmol/L TRAP-induced PLT aggreUnder the same conditions, 5 nmol/L DEA/NO resulted in gation neither in PRP from WB nor in PLTs derived from an increase of 43% on Day 2 and of 67% on Day 5. In all freshly prepared APCs on Day 0 (Fig. 3). Mean aggregation experiments, ODQ, a selective, irreversible inhibitor of values were 84.6 ± 2.8% in PRP without DEA/NO treatsGC, blocked DEA/NO-stimulated VASP phosphorylation ment and 84.8 ± 2.3% after DEA/NO treatment. PLTs from (data not shown). APCs on Day 0 aggregated slightly, but not significantly, stronger than in PRP from WB, with mean aggregations DEA/NO-induced increase of cGMP, but not of of 86.5 ± 4.0% (without DEA/NO) and 87.7 ± 3.2% (after cAMP, is higher in stored PLTs than in WB DEA/NO; Fig. 3). On Day 2 of storage, TRAP-6–induced PLT aggregation According to previous data15 basal PLT cGMP levels was somewhat decreased (69.3 ± 9.5%) compared to Day 0 increased during storage, by almost 60% in APCs on Day 5 or to PRP from WB. Besides, a weak, but significant, inhicompared to Day 0. Both investigated DEA/NO concenbition of TRAP-6–induced PLT aggregation (63.3 ± 11.2%) trations (2 and 5 nmol/L) induced a significant increase of in the presence of 5 nmol/L DEA/NO was observed the cGMP content in PLTs from WB and stored APCs com(Fig. 3). pared to unstimulated controls (Fig. 2). After the stimula-

2n cGMP (fold accumulation) m 5n ol C m /L on ol DE tro /L A l D /N 2n EA O m o 5n l C /N m /L on O ol DE tro /L A l 2n D /N E O m 5n ol C A/N / m L on O ol D tr /L EA ol D /N 2n EA O m 5n ol/ C /NO m L D on ol E tro /L A l D /N 2n EA O m 5n ol/ Co /NO m L D nt ol E ro /L A l 2n D /N EA O m o 5n l C /N m /L on O ol DE tro /L A l 2n D /N EA O m o 5n l/ C /N m L D on O ol E tro /L A l D /NO 2n EA m 5n ol C /N m /L D on O ol E tro /L A l D /N EA O /N O

+

Volume 54, July 2014

TRANSFUSION

1785

KOBSAR ET AL.

120

WB/PRP, Day 0

APCs, Day 0

APCs, Day 2

APCs, Day 5

80 60 40 20

0 DEA/NO -

WB/PRP, Day 0

APCs, Day 0

APCs, Day 2

-

-

APCs, Day 5

100

5 mg/mL collagen-induced PLT aggregation, %

5 µmol/L TRAP-6–induced PLT aggregation, %

100

120

80 60 40 20

+

-

+

-

+

-

+

0 DEA/NO -

+

+

+

-

+

Fig. 3. TRAP-6–induced aggregation is reduced by DEA/NO stimulation in PLTs from APCs on Day 2 and Day 5. The box-

Fig. 4. A quantity of 5 μg/mL collagen-induced aggregation is reduced by DEA/NO stimulation in PLTs from APCs on Day 2

and-whisker plots show the distribution of 5 μmol/L TRAP-6– induced aggregation values in WB/PRP and from APCs on

and Day 5. The box-and-whisker plots show the distribution of the 5 μg/mL collagen-induced aggregation values of WB/PRP

Days 0, 2, or 5 without (●) and after 2 minutes of incubation with 5 nmol/L DEA/NO (○). Results are presented in percent aggregation.

and of APCs on Days 0, 2, or 5 without (●) and after 2 minutes of incubation with 5 nmol/L DEA/NO (○). Results are presented in percent aggregation.

TABLE 1. DEA/NO decreases TRAP-6–induced fibrinogen binding in PLTs from APCs stored for 5 days* PLT stimulation 5 μmol/L TRAP-6 5 nmol/L DEA/NO + 5 μmol/L TRAP-6 Isotype control, 5 μmol/L TRAP-6 Isotype control, 5 nmol/L DEA/NO + 5 μmol/L TRAP-6

Fibrinogen binding 152.27 ± 28.08† 142.22 ± 24.18† 2.27 ± 0.14‡ 2.23 ± 0.25‡

* PLTs were stimulated as described and fibrinogen binding was determined by flow cytometry. Fluorescence intensity is presented as mean ± SD; n = 6. † p = 0.0047. ‡ p = 0.6619.

On Day 5 of APC storage, the capability of PLTs to aggregate in response to 5 μmol/L TRAP-6 (44.6 ± 10.1%) significantly decreased. Furthermore, 5 nmol/L DEA/NO induced a more pronounced inhibition of TRAP-6– induced aggregation to a mean value of 36.0 ± 10.1% (Fig. 3). TRAP-6–induced fibrinogen binding decreased in stored PLTs compared to PLTs from WB or from freshly prepared APCs.15 Like in aggregation experiments, 5 nmol/L DEA/NO did not have an influence on TRAP-6– induced fibrinogen binding in PLTs from WB and APCs on Day 0 or Day 2 (data not shown), whereas in APCs stored for 5 days, it caused a weak, but significant, inhibition. DEA/NO alone did not induce unspecific antibody binding (Table 1). 1786

TRANSFUSION Volume 54, July 2014

DEA/NO inhibits collagen-induced aggregation in stored PLTs An additional PLT activator, collagen, was used to confirm the effects of DEA/NO on TRAP-6–induced aggregation. As expected, 5 μg/mL collagen induced a strong PLT aggregation (87.5 ± 2.8%) in fresh PRP from WB and in PLTs from APCs on Day 0 (88.5 ± 2.5%; Fig. 4). This aggregation was unaffected by 5 nmol/L DEA/NO (85.5 ± 1.6 and 88.0 ± 2.7%). However, 5 μg/mL collagen-induced aggregation significantly decreased (74.8 ± 6.4%) in APCs stored for 2 days. A quantity of 5 nmol/L DEA/NO further inhibited the aggregation response (68.0 ± 6.9%). Storage of APCs for 5 days resulted in an even more pronounced decrease of PLT aggregation induced by 5 μg/mL collagen (61.3 ± 4.9%) and 5 nmol/L DEA/NO reduced aggregation to a greater extent (41.8 ± 14.9%; Fig. 4). The higher collagen concentration of 10 mg/mL leads to similar results. However, 5 nmol/L DEA/NO reduced the aggregation values only in APCs on Day 5 (Fig. 5). A quantity of 10 μg/mL collagen induced comparable PLT aggregation in both PRP from WB and APCs on Day 0 (88.4 ± 2.7 and 89.0 ± 2.8%), which was not inhibited by 5 nmol/L DEA/NO (88.3 ± 2.7 and 91.5 ± 23.2%). In APCs on Day 2, mean aggregation significantly decreased to 77.5 ± 4.3%, but was not influenced by 5 nmol/L DEA/NO (78.0 ± 2.9%). In APCs on Day 5, PLT aggregation after stimulation with 10 μg/mL collagen was dramatically decreased (57.1 ± 13.9%) compared to APCs on Day 2 and, in addition, significantly inhibited by 5 nmol/L DEA/NO (50.6 ± 17.3%; Fig. 5).

NO-MEDIATED SIGNALING IN STORED PLTs

120

WB/PRP, Day 0

APCs, Day 0

APCs, Day 2

APCs, Day 5

-

-

-

10 mg/mL collagen-induced PLT aggregation, %

100 80 60 40 20

0 DEA/NO -

+

+

+

+

Fig. 5. A 10 μg/mL collagen-induced aggregation is reduced by DEA/NO stimulation in PLTs from APCs on Day 2 and Day 5. The box-and-whisker plots show the distribution of the 10 μg/mL collagen-induced aggregation values of WB/PRP and of APCs on Days 0, 2, or 5 without (●) and after 2 minutes of incubation with 5 nmol/L DEA/NO (○). Results are presented in percent aggregation.

DISCUSSION To date, there are limited data on changes of inhibitory signaling pathways occurring under storage of PLT concentrates. Recently, proteomic analysis of stored human PLTs demonstrated an increased level of PLT VASP phosphorylation in APCs.20 In our previous study, this observation was confirmed and demonstrated that the decrease of PDE5 activity during storage of PLTs may contribute to elevated basal cGMP concentrations and the associated increase of VASP phosphorylation at Ser239 and Ser157.15 This study investigated the modulation capacity of the NO-mediated inhibitory signaling pathway in PLTs by use of DEA/NO. DEA/NO induced increasing shifts in the levels of VASP phosphorylation at Ser239 (up to fivefold, Fig. 1) and at Ser157 (up to threefold, data not shown) over 5 days of APC storage with both DEA/NO concentrations. The basal levels of VASP phosphorylation also increased, confirming results of previously published studies.15,20 The excess of induced VASP phosphorylation in stored PLTs may be caused by enlarged cGMP or cAMP generation. The direct measurement of cyclic nucleotides revealed that DEA/NO-stimulated cGMP (Fig. 2) and cAMP levels (data not shown) increased in a concentration-dependent manner in PLTs from freshly collected blood and in PLTs from stored APCs. However, DEA/NO treatment resulted in (approx. twofold) higher cGMP levels in stored PLTs than in fresh PLTs. In contrast to cGMP, the increase of cAMP was similar in freshly collected PLTs and in stored APCs (approx. 1.5-fold with

2 nmol/L DEA/NO and approx. twofold with 5 nmol/L in all investigated samples). In conclusion, the excessive increase of DEA/NOinduced VASP phosphorylation at Ser239 and Ser157 was most likely caused by greater cGMP accumulation in stored PLTs. DEA/NO-induced VASP phosphorylation and accumulation of cyclic nucleotides were completely blocked with ODQ, a sGC inhibitor, indicating that cGMP was likely the prime signaling agent in this process. VASP phosphorylation and cGMP levels did not differ between PLTs from WB and PLTs from APCs on Day 0 (directly after apheresis), suggesting that the manufacturing process itself has no relevant influence on the NO-mediated inhibitory signaling system. Interestingly, the stimulated levels of VASP phosphorylation approached in the course of storage and reached similar values on Day 5 despite two different concentrations of the NO donor. Besides, cGMP levels on Day 2 and on Day 5 were comparable. These observations indicate that the enhanced sensitivity to NO is a restricted alteration of the inhibitory signaling system and does not appear as an absolute loss of function, which is supported by the results of the recent study showing that the cGMP degrading enzyme PDE5A was reduced only to a certain degree (by 23%) on Day 5 of storage.15 The functional implications of increased susceptibility to NO release in stored PLTs were investigated by aggregation and fibrinogen binding studies with and without 5 nmol/L DEA/NO stimulation of PLTs. There was no significant difference in aggregation between DEA/NOtreated and untreated PLTs after activation with TRAP-6 and collagen (Figs. 3-5), despite strong VASP phosphorylation induced by 5 nmol/L DEA/NO in fresh PLTs from WB and APCs on Day 0 (Fig. 1). These results are in accordance with recently published data demonstrating that inhibitory effects of DEA/NO on PLT aggregation in PRP are expected only with higher concentrations of 0.1 to 1 μmol/L.21-23 In general, most of the known NO donors used for scientific research or for therapy have only limited inhibitory potential on PLT aggregation in vitro. For glyceryl trinitrate it was demonstrated that it has no effect on collagen-induced aggregation and weak effects on ADP induced aggregation.16 High concentrations of sodium nitroprusside (SNP), S-nitrosothiol (RIG200), S-nitrosoglutathione (GSNO), or DEA/NO were able to decrease aggregation by 40% to 60%. The inhibitory effects of such high NO donor concentrations were only partially inhibited with ODQ, indicating that additional cGMPindependent mechanisms were activated.16,17 In our study, DEA/NO was chosen in a low concentration of 2 or 5 nmol/L to specifically stimulate the inhibitory signaling pathways in a cGMP-dependent manner without affecting aggregation in freshly prepared PLTs (Figs. 1-5). TRAP-6 and collagen-induced aggregation decreased during APC storage (Figs. 3-5) reflecting the proceeding Volume 54, July 2014

TRANSFUSION

1787

KOBSAR ET AL.

functional impairment as reported in earlier studies.5-7 On Day 2 of APC storage, mean aggregation induced with 5 μmol/L TRAP-6 and low-dose collagen was significantly reduced (by 9%) in the presence of 5 nmol/L DEA/NO, again indicating an enhanced sensitivity of stored PLTs to the NO donor. On Day 5 of APC storage, treatment of PLTs with DEA/NO resulted in even stronger inhibition of aggregation, with 5 μmol/L TRAP-6 by 19% and with 5 μg/mL collagen by 32%. Flow cytometric analysis of TRAP-6–stimulated fibrinogen binding demonstrated a weak inhibition (of approx. 7%) only in DEA/NO-treated PLTs from APCs on Day 5 (Table 1), supporting the results received for TRAP6–induced PLT aggregation. The apparent discrepancy between aggregometry and flow cytometric analysis on Day 2 of APC storage may be explained by differences in the sensitivity of both methods, since the effect on aggregation was only moderate at this time point. The modulation of PLT aggregation by DEA/NO was dependent on the concentration of collagen. DEA/NO significantly decreased aggregation induced by 10 μg/mL collagen in PLTs from 5-day-stored APCs (by 11%), whereas the same DEA/NO concentration did not inhibit 10 μg/mL collagen-induced aggregation in freshly collected blood and in APCs on Days 0 and 2. Obviously, cGMP accumulation after NO donor stimulation particularly influences the reactivity of stored PLTs to weak activating signals, for example, TRAP-6 and collagen in low concentrations. Higher collagen concentrations caused more stable PLT aggregation overacting DEA/NO induced inhibition. It must be mentioned that the study also has some limitations. The study was performed with apheresisderived PLT concentrates. The confirmation of the results in differently manufactured PLT concentrates (e.g., buffy coat–derived PLT concentrates) could emphasize their significance for the contribution to storage lesion. Although DEA/NO was used because of its characteristics of a short half-life and in low concentrations to simulate physiologic conditions, it remains an artificial NO releasing agent. After retransfusion of stored PLTs, the modulation of inhibitory PLT signaling by endothelial NO release in vivo may be different. In addition, it cannot be ruled out that the observed functional effects of DEA/NO on aggregation and fibrinogen binding in stored PLTs have been influenced by other unspecific factors of storage lesion. In summary, the study demonstrated that the inhibitory cGMP-dependent signaling pathway becomes more susceptible for NO release during storage of APC-derived PLTs. A possible explanation may be the degradation of PDE5A resulting in higher cGMP levels, as previously described.15 The alterations are associated with functional impairment as shown in exceedingly reduced agonist responses in the presence of NO release by DEA/NO. The observed effects may be an important mechanism con1788

TRANSFUSION Volume 54, July 2014

tributing to the storage lesion with reduced activating potential of PLTs. Further studies are required to determine whether these phenomena also play a role in vivo after retransfusion of stored PLTs. ACKNOWLEDGMENTS The authors thank their colleagues of the Institute of Transfusion Medicine and Haemotherapy for donation management and preparation of PLT concentrates. They also thank their colleagues of the Department of Nephrology, Internal Medicine 1, University Clinic of Wuerzburg, for technical support.

CONFLICT OF INTEREST The authors report no conflicts of interest or funding sources.

REFERENCES 1. Cauwenberghs S, van Pampus E, Curvers J, et al. Hemostatic and signaling functions of transfused platelets. Transfus Med Rev 2007;21:287-94. 2. Lozano ML, Rivera J, Gonzalez-Conejero R, et al. Loss of high-affinity thrombin receptors during platelet concentrate storage impairs the reactivity of platelets to thrombin. Transfusion 1997;37:368-75. 3. Schlagenhauf A, Kozma N, Leschnik B, et al. Thrombin receptor levels in platelet concentrates during storage and their impact on platelet functionality. Transfusion 2012;52: 1253-9. 4. Perez-Pujol S, Aras O, Lozano M, et al. Stored platelets contain residual amounts of tissue factor: evidence from studies on platelet concentrates stored for prolonged periods. Transfusion 2005;45:572-9. 5. Rudderow D, Soslau G. Permanent lesions of stored platelets correlate to pH and cell count while reversible lesions do not. Proc Soc Exp Biol Med 1998;217:219-27. 6. Shapira S, Friedman Z, Shapiro H, et al. The effect of storage on the expression of platelet membrane phosphatidylserine and the subsequent impacton the coagulant function of stored platelets. Transfusion 2000;40: 1257-63. 7. Albanyan AM, Murphy MF, Harrison P. Evaluation of the Impact-R for monitoring the platelet storage lesion. Platelets 2009;20:1-6. 8. Schwarz UR, Walter U, Eigenthaler M. Taming platelets with cyclic nucleotides. Biochem Pharmacol 2001;62:115361. 9. Butt E, Abel K, Krieger M, et al. cAMP- and cGMPdependent protein kinase phosphorylation sites of the focal adhesion vasodilator-stimulated phosphoprotein (VASP) in vitro and in intact human platelets. J Biol Chem 1994;269:14509-17. 10. Schwarz UR, Geiger J, Walter U, et al. Flow cytometry analysis of intracellular VASP phosphorylation for the

NO-MEDIATED SIGNALING IN STORED PLTs

assessment of activating and inhibitory signal transduction

17. Crane MS, Rossi AG, Megson IL. A potential role for extra-

pathways in human platelets—definition and detection of ticlopidine/clopidogrel effects. Thromb Haemost 1999;82:

cellular nitric oxide generation in cGMP-independent inhibition of human platelet aggregation: biochemical and

1145-52.

pharmacological considerations. Br J Pharmacol 2005;144:

11. Smolenski A, Bachmann C, Reinhard K, et al. Analysis and regulation of vasodilator-stimulated phosphoprotein serine 239 phosphorylation in vitro and in intact cells using a phosphospecific monoclonal antibody. J Biol Chem 1998;273:20029-35. 12. Aszodi A, Pfeifer A, Ahmad M, et al. The vasodilator-

849-59. 18. Friebe A, Koesling D. The function of NO-sensitive guanylyl cyclase: what we can learn from genetic mouse models. Nitric Oxide 2009;21:149-56. 19. Ferrero R, Rodriguez-Pascual F, Miras-Portugal MT, et al. Comparative effects of several nitric oxide donors on intra-

stimulated phosphoprotein (VASP) is involved in cGMP-

cellular cyclic GMP levels in bovine chromaffin cells: corre-

and cAMP-mediated inhibition of agonist-induced platelet aggregation, but is dispensable for smooth muscle func-

lation with nitric oxide production. Br J Pharmacol 1999; 127:779-87.

tion. EMBO J 1999;18:37-48. 13. Massberg S, Gruner S, Konrad I, et al. Enhanced in vivo platelet adhesion in vasodilator-stimulated phosphoprotein (VASP)-deficient mice. Blood 2004;103:136-42. 14. Eigenthaler M, Nolte C, Halbrugge M, et al. Concentration

20. Schubert P, Culibrk B, Coupland D, et al. Riboflavin and ultraviolet light treatment potentiates vasodilatorstimulated phosphoprotein Ser-239 phosphorylation in platelet concentrates during storage. Transfusion 2012;52: 397-408.

and regulation of cyclic nucleotides, cyclic-nucleotidedependent protein kinases and one of their major substrates in human platelets. Estimating the rate of cAMP-

21. Gudmundsdottir IJ, McRobbie SJ, Robinson SD, et al. Sildenafil potentiates nitric oxide mediated inhibition of human platelet aggregation. Biochem Biophys Res

regulated and cGMP-regulated protein phosphorylation in intact cells. Eur J Biochem 1992;205:471-81.

Commun 2005;337:382-5. 22. Roger S, Badier-Commander C, Paysant J, et al. The anti-

15. Kobsar A, Putz E, Yilmaz P, et al. Decreasing phosphodiesterase 5A activity contributes to platelet cGMP accumulation during storage of apheresis-derived platelet concentrates. Transfusion 2014;54:1008-14. 16. Sogo N, Magid KS, Shaw CA, et al. Inhibition of human

aggregating effect of BAY 41-2272, a stimulator of soluble guanylyl cyclase, requires the presence of nitric oxide. Br J Pharmacol 2010;161:1044-58. 23. Chlopicki S, Lomnicka M, Fedorowicz A, et al. Inhibition of platelet aggregation by carbon monoxide-releasing mol-

platelet aggregation by nitric oxide donor drugs: relative contribution of cGMP-independent mechanisms. Biochem

ecules (CO-RMs): comparison with NO donors. Naunyn Schmiedebergs Arch Pharmacol 2012;385:

Biophys Res Commun 2000;279:412-9.

641-50.

Volume 54, July 2014

TRANSFUSION

1789

Increasing susceptibility of nitric oxide-mediated inhibitory platelet signaling during storage of apheresis-derived platelet concentrates.

Storage of platelets (PLTs) affects PLT integrity and functionality, a process named the PLT storage lesion. Normal PLT function essentially depends o...
271KB Sizes 0 Downloads 3 Views