Vox Sanguinis (2014) 107, 360–367 © 2014 International Society of Blood Transfusion DOI: 10.1111/vox.12174
ORIGINAL PAPER
Addition of sialidase or p38 MAPK inhibitors does not ameliorate decrements in platelet in vitro storage properties caused by 4 °C storage A. Skripchenko, D. Thompson-Montgomery, H. Awatefe, A. Turgeon & S. J. Wagner American Red Cross Biomedical Services, Holland Laboratory, Rockville, MD, USA
Background and Objectives Bacterial proliferation is inhibited in platelets (PLTs) stored at refrigerated temperatures, but also dramatically decreases PLT in vivo survival. Recent studies have demonstrated that cold temperature (CT) stored PLTs secrete sialidases upon re-warming, removing sialic acid from the PLT surface, which may be responsible for clustering of GPIba and PLT clearance from circulation. In this study, the influence of a sialidase inhibitor or a p38 MAP kinase inhibitor was evaluated in units stored at 4 °C. Materials and Methods After collection of a single Trima apheresis unit (n = 12), PLTs were aliquoted into four 60-ml CLX storage bags. One bag was stored at 20–24 °C (RT) with continuous agitation; a second bag was stored at 4 °C without agitation; a third bag was held at 4 °C without agitation with sialidase inhibitor, a fourth bag was incubated at 4 °C with a p38 MAPK inhibitor without agitation. Results Beginning from Day 1, all in vitro PLT parameters were adversely affected by CT compared to those of RT. Similar in vitro storage properties were observed in CT PLT in the presence or absence of sialidase or p38 MAPK inhibitors. P38 MAPK phosphorylation inhibition was not observed at CT. Decrease of sialidase activity was observed for 2 days in PLTs stored in additive solution but not in plasma. Received: 20 December 2013, revised 9 May 2014, accepted 2 June 2014, published online 27 June 2014
Conclusion Addition of either sialidase or p38 MAPK inhibitors do not improve any in vitro parameters of PLTs stored at 4 °C in 100% plasma. Key words: p38 MAPK, platelets properties, platelets storage lesion, sialidase inhibitor.
Introduction Platelets (PLTs) initially were stored at cold refrigerated temperature (4 °C) before Murphy and Gardner demonstrated that PLT storage at room temperature (22 °C) dramatically improved PLT recovery and survival after transfusion [1]. With introduction of highly gas permeable storage containers, PLT storage could be potentially extended to 7 days [2]. Unfortunately, 7 day storage was later found to be associated with an increased Correspondence: Andrey Skripchenko, American Red Cross, 15601 Crabbs Branch Way, Rockville, MD 20855, USA E-mail: [email protected]
360
rate of bacterial contamination [3], resulting in the reduction of the storage time back to 5 days [4]. Despite the introduction of automated bacterial culture, a substantial risk for transfusion-transmitted bacterial infection remains. Based on the recent studies, the prevalence of bacterial contamination is estimated as one in 750 to one in 1000 PLT concentrates [5]. Cold temperature (CT) storage would be expected to slow bacterial growth, but PLTs do not tolerate CT and are rapidly cleared in the liver following transfusion [6, 7]. One of the pathways of PLT clearance is the removal of sialic acid from GPIba, a subunit of PLT von Willebrand factor receptor, by active sialidases secreted from the PLTs upon re-warming after CT storage [8]. Sialidase
Inhibitors with 4 C PLT storage 361
activation also stimulates metalloproteinase ADAM17, also known as TACE (TNF-a-converting enzyme), which cleaves GPIba and GPV from the PLT surface [9]. Shedding desialylated GPIba and GPV have been correlated with poor recovery and survival in a mouse model [9]. Inhibition of sialidase activity by N-acetylneuraminic acid, 2,3-dehydro-2-deoxy-sodium salt (DANA) diminished shedding of GPIba and GPV and greatly improved PLT recovery and survival in mice after refrigerated storage for 48 h [9]. On other hand, activation of ADAM17 is regulated by p38 mitogen-activated protein kinase (MAPK) activation [10]. It has been shown in animal models that inhibition of ADAM17 and p38 MAPK activity enhances the recovery of PLT stored at room temperature [10]. The goal of this study was to compare PLTs stored at CT in 100% plasma for 7 days with or without a competitive sialidase inhibitor, DANA, or with or without a second generation of p38 MAPK inhibitor, VX-702.
Materials and methods Materials The sialidase inhibitor, DANA, was purchased from EMD Chemicals, subsidiary of EMD Millipore (Billerica, MA, USA). A second generation of p38 MAPK inhibitor, VX-702, was obtained from Selleck Chemicals LLC (Houston, TX, USA). Dimethylsulphoxide (DMSO) was purchased from Sigma-Aldrich (Sigma-Aldrich Corporation, St. Louis, MO, USA). Phosphate-buffered solution (PBS) pH 74 was obtained from Life Technologies (Life Technologies Corporation, NY, USA).
PLT collection and study design Twelve independent experiments were carried out. In each, a single apheresis unit was collected under informed consent using the Trima cell separator from healthy volunteers with a targeted yield of 40 9 1011 PLTs and targeted PLT concentration 14 9 106/ll. The unit was divided into four identical 60-ml aliquots in CLX storage bags (Medsep Corporation, a Subsidiary of Haemonetics Corporation, Covina, CA, USA) immediately after collection. Aliquot A) was stored at 20–24 °C with continuous agitation (RT); aliquot B) was designated as a cold temperature control and incubated at 4 °C without agitation (CT); aliquot C) PLTs were stored with the addition of a sialidase inhibitor without agitation at 4 °C (DANA); aliquot D) was stored at 4 °C with the addition of a p38 MAPK inhibitor without agitation (VX-702). To assess effectiveness of added inhibitors, a satellite study with six experiments was carried out. In each, © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
a single apheresis unit was collected using the same separator and the same parameters. The unit was divided into six identical 60-ml aliquots in CLX storage bags. All aliquots were sedimented with 4200 g and ACE 550 9 107 (Sorvall RC3BP; Thermo Fisher Scientific Inc., Waltham, MA, USA). After 1 h rest, plasma was replaced with PLT additive solution (PAS) M-Sol [11] in one bag resulting in 5% plasma. All other aliquots were resuspended in plasma. Aliquot 1) PLTs were stored with addition of DANA at 4 °C without agitation in 100% plasma (CT-DANA-PL); aliquot 2) PLTs were stored with addition of DANA in 95% PAS/5% plasma without agitation (CT-DANA-PAS); aliquot 3) PLTs were stored with addition of VX-702 at 4 °C without agitation (CT-VX); aliquot 4) PLTs were stored with addition of VX-702 at 20–24 °C with continuous agitation (RT-VX); aliquot 5) PLTs were stored at 4 °C without agitation (CT-CNTR); aliquot 6) PLTs were stored at 20–24 °C with continuous agitation (RT-CNTR). In these experiments, the following parameters were measured: phosphorylated (Thr180/Tyr182) and total p38 MAPK, b-N-acetylglucosamine (b-GlcNAc)terminated glycans exposure on GPIba, p-selectin exposure on the PLT membrane and Annexin V binding.
Drug delivery A stock solution of DANA was prepared in PBS at 319 mM (10 mg/ml). A stock solution of VX-702 was prepared in DMSO at 618 mM (25 mg/ml). Final drug concentrations of DANA and VX-702 in the PLT aliquots were 1 mM and 1 lM, respectively, and were identical to those previously reported [9, 12]. The final DMSO concentration in PLT aliquots D, 3 and 4 (satellite study) was 0016%; this concentration has been previously shown not to affect PLT storage properties [13].
PLT assays All PLT units were sampled by syringe (approximate volume, 38 ml) on Days 1, 2, 4 and 7 using a CBS needlefree spike (OriGen Biomedical, Austin, TX, USA). PLT concentrations were measured using a Sysmex XE 2100D haematology analyser (Sysmex Corporation, Lincolnshire, IL, USA). Metabolic parameters were assessed measuring pH, glucose, lactate, bicarbonate, oxygen and carbon dioxide levels. PLT extracellular pH was measured using a bench top meter (Orion, Thermo Scientific, Beverly, MA) and pH electrode (Accu-pHast; Fisher Scientific, Pittsburg, PA, USA). Other parameters were determined using a blood gas analyzer (Cobas b221; Roche Diagnostics, Indianapolis, IN, USA). The bicarbonate concentration was calculated automatically from pH and CO2 (37 °C) levels. Rates
362 A. Skripchenko et al.
for bicarbonate neutralization, glucose consumption and lactate generation were calculated from measurements on Day 1 and Day 7 and normalized by day and 1010 PLTs. Structural parameters were evaluated by measuring morphology, Annexin V binding, and expression of CD62P and CD42b. PLT morphology was expressed as the percentage of PLTs with discoid morphology using a phase microscopy as previously described [1]. Annexin V binding and expression of CD62P (p-selectin) and CD42b (GPIba) were measured by flow cytometry (FACSCalibur; BD Biosciences, San Jose, CA, USA). To measure the percentage of Annexin V positive PLTs, samples were washed and resuspended to 1 9 106 PLTs/ml with manufacturer supplied binding buffer. PLTs were incubated with fluorescein isothiocyanate (FITC)-conjugated CD61 (Biolegend, San Diego, CA, USA) monoclonal antibodies and with phycoerythrin (PE)-conjugated Annexin V (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA) at saturating concentrations for 15 min at 22 °C. PLTs were gated by forward and side scattering and binding of CD61. To measure the percent of PLTs expressing membrane-bound CD62P and CD42b, PLT samples were diluted to 1 9 106 PLTs/ml with PBS supplemented with 01% human albumin (HA) (American Red Cross, Blood Services; Washington DC, USA). Sample aliquots were incubated with FITC-conjugated CD61 (Biolegend) monoclonal antibodies and with PE-conjugated CD62P (Biolegend) or PE-conjugated CD42b monoclonal antibodies (Biolegend) at saturating concentrations for 15 and 120 min, respectively, at 22 °C. Mouse IgG1 PE and FITC PLT isotype control (Biolegend) were used at saturating concentrations as negative controls. PLTs were gated by forward and side scattering and binding of CD61. Appropriate color compensation was set for FITC fluorescence and PE fluorescence by using 530 and 585 nm bandpass filters respectively. Standard three-colour beads (Calibrite 3; BD Biosciences) were used for daily instrument calibration and FITC and PE colour compensation. Functional parameters were assessed measuring hypotonic shock response (HSR), extent of shape change (ESC) and PLT aggregation as previously described [13, 14]. Briefly, the ESC and HSR levels were measured turbimetrically using a Chrono-Log SPA 2000 (Chrono-Log Corporation, Havertown, PA, USA). PLT aggregation was measured using a lumiaggregometer (Chrono-Log) and 10 lM of ADP (Chrono-Log) with 10 lg/ml of collagen (Chrono-Log), as final concentrations. Agonists were added simultaneously to PLT suspensions. Aggregation was measured according to manufacturer’s protocol and expressed relative to a maximum slope and amplitude of the instrument’s internal electronic standard. Platelets mitochondrial parameters were evaluated by measuring changes in mitochondrial membrane potential
(MMP) and intracellular reactive oxygen species (ROS) generation. The MMP was determined using a MitoProbe JC-1 Assay Kit (Molecular Probes; Invitrogen, Eugene, OR, USA). Samples were diluted to 1 9 106 PLTs/ml with PBS supplemented with 01% HA and stained with 1 lM of JC-1 for 15 min at 37 °C. Samples were analysed using a flow cytometer (FACSCalibur). ROS generation was monitored by measuring changes in fluorescence resulting from the oxidation of intracellular probes, dihydroethidium (DHE) (Molecular Probes; Invitrogen) and 5(and-6)-chloromethyl-20 ,70 -dichlorodihydrofluorescein diacetate, acetyl ester (CM-H2DCFDA, Molecular Probes; Invitrogen). Samples were diluted to 1 9 106 PLTs/ml with PBS supplemented with 01% HA and stained with 5 lM DHE and 5 lM CM-H2DCFDA for 15 min at 37 °C. Fluorescence of DHE and CM-H2DCFDA was detected in the FL2 and FL1 region of a FACSCalibur flow cytometer, respectively. The amount of phosphorylated (Thr180/Tyr182) p38 MAPK was measured using a STAR (Signal Transduction Assay Reaction) ELISA kit (EMD Millipore). Total p38 MAPK content was measured (InstantOne ELISA kit; eBioscience, San Diego, CA, USA). To measure total and phospho-p38 MAPK, PLTs were centrifuged at 850 g in the presence of 1 lg/ml prostaglandin E1, the pellet was resuspended in ice cold PBS supplemented with 01% HA and lysed at 0 °C using 2X platelet lysis buffer (2% NP40, 30 mM Hepes, 150 mM NaCl, 2 mM EDTA, pH 74). Cell debris were sedimented at 18 000 g for 5 min, and supernatant was aliquoted and stored at -80 °C before immunoassays. All assays were performed in duplicate according to manufacturer’s protocols. Concentrations of phospho-p38 MAPK were expressed as Units/ml per 1010 PLTs to normalize measurements to PLT count. To measure the effectiveness of DANA, a flow cytometric analysis of b-GlcNAc-terminated glycans exposure on GPIba was performed using a FACSCalibur flow cytometer and 01 lg/ml succinyl Triticum vulgare lectin from wheat germ (s-WGA; EY Laboratories, Inc., San Mateo, CA, USA) conjugated with FITC as previously described [9].
Statistics Determination of means and standard deviations of experimental values and Analysis of Variance with repeated measures were carried out using standard software (Instat; GraphPad Software, San Diego, CA, USA). A value of P ≤ 0001 was considered significant taking into account repeated measures for the 18 PLT assays utilized in this study and three testing days [15]. For a satellite study, a value of P ≤ 00125 was considered significant taking into account repeated measures of 4 days of testing [15]. Statistical differences between all paired values © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Inhibitors with 4 C PLT storage 363
and VX-702 aliquots on any given day of storage. GPIba expression was similar in all PLT aliquots on Days 1 and 2 and the GPIba expression in the RT aliquots was greater than those in CT, DANA and VX-702 aliquots on Days 4 and 7 (P < 001). Results of functional PLT parameters measured during storage are given in Fig. 2. HSR, ESC and both aggregation measurements were less in CT, DANA and VX-702 aliquots throughout storage compared to those of RT (P < 0001). Values of all functional parameters were similar between CT, DANA and VX-702 aliquots on any given day during storage except for the values of aggregation slope for DANA and VX-702 on Day 1, which were greater than those of CT (P < 0001). Results of the measurements of mitochondrial function during storage are shown in Fig. 3. No differences were observed in PLT MMP levels during storage between CT, DANA and VX-702 aliquots. All aliquots stored at cold temperature had MMP levels greater than those of RT on Day 1 (P < 0001). On Day 2 only, the PLT MMP levels were greater in DANA and VX-702 compared to those of RT (P < 001). On Day 7, the MMP levels of CT PLTs were less than those of RT PLTs (P < 005). ROS accumulations were similar between all aliquots during storage with the exceptions that the level of DHE which was greater in DANA and VX-702 aliquots on Day 7 (P < 001) and the CM-H2DCFDA level was greater in CT and DANA aliquots on Day 4 compared to those of RT (P < 001).
of RT, CT, DANA and VX-702 aliquots were determined by post Hoc tests with Bonferroni corrections.
Results Results of this study which provide data for PLT storage parameters measured on Days 1–7 are given in Table 1 and Figs 1–3. Assessments of metabolic parameters of PLTs stored at RT, CT, CT with DANA or CT with VX-702 are shown in Table 1. All parameters of CT, DANA and VX-702 were significantly different compared to those of RT. The pH levels of DANA PLT aliquots were less than those of VX-702 on Days 1 and 2. No differences were observed in glucose, lactate and bicarbonate levels between CT, DANA and VX-702 during storage except for bicarbonate levels on Day 1 between RT and all CT treatments (Table 1). Results of structural PLT parameter measurements are presented in Fig. 1. Morphology score was less in all aliquots stored at cold temperature compared to those of RT (P < 0001), but those stored at CT in the presence or absence of DANA or VX-702 did not differ from each other. The percent Annexin V positive PLT stored at CT were greater (P < 005), with the exception of VX-702 aliquots on Days 2 and 7, than those stored at RT. PLT p-selectin expression was less on Days 1 and 2 (P < 005) and greater on Day 7 in RT aliquots compared to those stored at cold temperature (P < 0001). No differences in p-selectin expression were observed between CT, DANA
Table 1 Assessments of metabolical parameters of PLTs stored for 7 days at RT, CT, CT with DANA or CT with VX-702 pH Day 1
Day 2
Day 4
Day 7
RT CT DANA VX-702 RT CT DANA VX-702 RT CT DANA VX-702 RT CT DANA VX-702
747 730 727 732 752 736 734 737 734 745 742 746 714 747 745 750
Glucose (mM) – – – – – – – – – – – – – – – –
004 005a 005ab 006a 003 005a 004ab 005a 005 006a 006a 006a 010 008a 008a 009a
193 198 195 197 183 193 193 195 147 184 181 183 118 177 174 176
– – – – – – – – – – – – – – – –
31 30a 30a 31a 33 32a 30a 30a 30 31a 31a 29a 30 32a 31a 30a
Data are presented as mean – one standard deviation. a P < 0001, compared to that of RT values. b P < 001, compared between DANA and VX-702 values.
© 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Lactate (mM) 35 29 29 29 55 39 38 38 128 64 63 62 178 81 78 77
– – – – – – – – – – – – – – – –
06 06a 06a 06a 09 07a 07a 07a 22 11a 12a 12a 34 15a 14a 15a
Bicarbonate (mM) 177 195 193 198 156 181 177 183 103 153 152 154 67 136 134 139
– – – – – – – – – – – – – – – –
18 21a 22a 20a 13 20a 20a 19a 11 14a 15a 14a 13 11a 15a 12a
pO2 (mm Hg) 124 203 202 205 126 205 204 209 138 212 211 211 146 218 217 220
– – – – – – – – – – – – – – – –
10 6a 7a 7a 8 10a 9a 10a 12 9a 9a 9a 8 10a 10a 8a
pCO2 (mm Hg) 402 655 666 642 336 541 556 537 287 392 405 382 256 346 355 340
– – – – – – – – – – – – – – – –
39 96a 67a 66a 34 63a 59a 59a 36 54a 52a 50a 25 35a 41a 35a
364 A. Skripchenko et al.
Fig. 1 Assessments of structural parameters of PLTs stored for 7 days at RT, CT, CT with DANA or CT with VX-702. Data are presented as mean – one standard deviation. Opened circle – RT-CNTR, closed circle – CT-CNTR, opened square – CT-VX, opened triangular – CT-DANA. [Correction added on 14 July 2014, after first online publication: the title of the fourth panel has been corrected].
Results of the satellite study are presented in Fig. 4 and Table 2. Measurements of phosphorylated p38 MAPK are shown in Fig. 4. Levels of phosphorylated p38 MAPK were greater in CT-VX PLTs on Day 4 and Day 7 (P ≤ 0034) compared to those of RT-VX but was comparable on Day 1. Levels of phosphorylated p38 MAPK in CT-VX were less than that of CT-CNTR on Day 1 (P < 001) but were not different on Days 4 and 7. RT-VX samples had less phosphorylation on Days 4 and 7 compared to those of RT-CNTR (P < 005 and P < 0001, respectively). Phos-
Fig. 2 Assessment of functional parameters of PLTs stored for 7 days at RT, CT, CT with DANA or CT with VX-702. Data are presented as mean – one standard deviation. Opened circle – RT-CNTR, closed circle – CT-CNTR, opened square – CT-VX, opened triangular – CT-DANA.
phorylation was greater in CT-CNTR PLTs compared to that of RT-CNTR during storage (P < 0001). Total p38 MAPK content was constant during storage and similar between all PLT aliquots (data not shown). S-WGA binding to PLTs is presented in Table 2. S-WGA binding was greater in CT-CNTR PLTs compared to that of RT-CNTR on Day 1 and was comparable to that of CT-DANA-PL. S-WGA binding in CT-DANA-PAS PLTs was less than that of CT-DANA-PL on Days 1 and 2 but similar on Day 4 and 7. In addition, S-WGA binding in CT-DANA-PAS was less than that of CT-CNTR on Days 1–2 but was comparable on Days 4 and 7. Results of © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Inhibitors with 4 C PLT storage 365
Fig. 4 Levels of phosphorylated p38 MAPK of PLTs stored for 7 days at RT and CT with or without VX-702. Data are presented as mean – one standard deviation. Opened circle – RT, closed circle – RT-VX, opened square – CT, closed square – CT-VX.
Fig. 3 Assessment of mitochondrial parameters of PLTs stored for 7 days at RT, CT, CT with DANA or CT with VX-702. Data are presented as mean – one standard deviation. Opened circle – RT-CNTR, closed circle – CT-CNTR, opened square – CT-VX, opened triangular – CT-DANA.
Annexin V binding and p-selectin surface exposure on PLT were comparable to those obtained in the main study (data not shown).
Discussion Refrigerated storage results in PLT decrements known as the ‘PLT cold storage lesion’ [16]. P38 MAPK is implicated in the PLT storage lesion under normal conditions, and appears to be associated with recovery and survival of mouse PLTs [10]. When PLTs are stored at CT, their recovery and survival are very low [6]. In a mouse model, desialylation of GPIba accelerates PLT removal after CT storage [9]. Addition of the competitive sialidase inhibitor, DANA, to the refrigerated mouse PLTs enhanced recoveries and circulation time after transfusion [9, 17]. In this study, the in vitro parameters of PLTs stored at 4 °C with addition of DANA or VX-702 were evaluated. © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Addition of either DANA or VX-702 to PLTs stored at 4 °C did not affect any PLT metabolic parameters, which were all comparable to those of CT. In contrast, addition of VX-702 to the PLTs stored at 20–24 °C decreased the glucose consumption and lactate accumulation rates [13]. In a different study, investigators evaluated the possibility of ThromboSol, a mixture of second messengers, effectors, or PLT inhibitors, to ameliorate changes caused by cold storage. Treating PLTs with ThromboSol during cold storage also did not also affect metabolic parameters [18]. Assessment of changes in structural PLT parameters (morphology and CD62P) with the addition of either DANA or VX-702 inhibitors did not reveal differences between any of the Day 7 aliquots stored at cold temperature. In contrast, addition of p38 MAPK inhibitor to
Table 2 S-WGA binding to PLTs stored for 7 days at RT and CT in 100% plasma or PAS (5% plasma level)
CT-DANAPL CT-DANAPAS CT-CNTR RT-CNTR
Day 1
Day 2
Day 4
Day 7
414 – 125ab
500 – 154ab
435 – 155
434 – 207
292 – 103bc
355 – 214bd
278 – 102
326 – 122
472 – 97a 370 – 96
469 – 77 344 – 88
430 – 135 317 – 116
418 – 152 267 – 78
Data are presented as mean – one standard deviation. P < 005, compared to that of RT-CNTR values. b P < 001, compared between CT-DANA-PL and CT-DANA-PAS. c P < 0001, compared to that of CT-CNTR. d P < 005, compared to that of CT-CNTR. a
366 A. Skripchenko et al.
PLTs stored at room temperature significantly reduced CD62P expression and greatly improved morphology score [13]. A future study would benefit from assessing p-selectin concentrations in the supernatant of PLT suspensions [19]. No differences were observed in Annexin V binding between any cold stored aliquots. It has been previously shown that the addition of a p38 MAPK inhibitor reduced the percentage of Annexin V positive PLTs during 7 daystorage at 20–24 °C [13]. Our findings of elevated Annexin V binding at refrigerated temperature compared to room temperature are in accordance with previously published data [20, 21]. Our results also demonstrated a loss of GPIba expression during 7 days at cold temperature with or without inhibitors compared to RT storage. Rivera and coauthors reported similar results in GPIba expression loss during cold storage with the ThromboSol treatment [18]. As expected, all PLT functional parameters were decreased during refrigerated storage starting on Day 1 as previously reported [22]. Addition of either p38 MAPK or sialidase inhibitors did not improve ESC, HSR and aggregation. These results contrasted to previously reported data when addition of VX-702 to PLTs during 7-day RT storage markedly improved ESC and HSR [13]. Inhibition of p38MAPK does not affect human PLT aggregation during storage in 100% plasma or freshly prepared mouse or human platelets suspended in either 100% plasma or buffer in response to one or more synergistic agonists [10, 13]. In our experiments, cold storage of PLTs triggered more rapid loss of MMP levels and greater ROS accumulation in PLTs. Addition of DANA or VX-702 did not prevent MMP loss or ROS accumulation. In contrast to these results, addition of VX-702 to PLTs dramatically improved mitochondrial function when PLTs suspended in 100% plasma were stored at RT [13]. Addition VX-702 to PLTs stored at RT diminishes phosphorylation of p38 MAPK during storage, which is associated with improved in vitro storage parameters. As expected, CT storage of platelets caused a significant increase in phosphorylation of p38 MAPK [23]. Addition of VX-702 decreased p38 MAPK phosphorylation only on Day 1 of CT storage, but failed to overcome the level of p38 MAPK phosphorylation caused by CT storage. GPIba is glycosylated with N-linked carbohydratechains containing sialic acids [24]. PLTs stored at refrigerated temperature release sialidase and subsequently remove sialic acid residues upon re-warming [9]. The addition of DANA prevented sialidase activity and GPIba shedding with enhanced PLT recovery and survival in an animal model [9]. In our experiments, DANA failed to improve any in vitro properties of PLTs stored at refrigerated temperature in 100% plasma, including levels of GPIba. At the same time, sialidase activity was not
inhibited by DANA for CT PLTs suspended in 100% plasma. Furthermore, DANA inhibited the sialidase activity for CT PLTs suspended in PAS (5% residual plasma) during the first 2 days of storage, which corroborates previously published data of sialidase inhibition reported by Jansen et al. [9]. In their study, experiments were performed in a glucose containing buffer rather than the 100% plasma utilized in this study. More work needs to be done to understand how plasma proteins and plasma levels affect DANA activity on PLT in vitro storage properties. The breakdown of DANA during storage also cannot be ruled out. Some sialidase inhibitors bind to plasma proteins depending on their lipophilicity and charge [25]. Moreover, the potency of DANA to inhibit platelet sialidase may not be optimal to affect PLT in vitro properties; more potent analogues of DANA have been developed [26]. Finally, a possible interaction between DANA and other PLT receptors cannot be ruled out [25]. The addition of p38 MAPK inhibitor prevents shedding of GPIba and improved PLT recovery and survival during RT storage [10]. It has been previously reported that cooling PLTs to < 10 °C activated both p38 MAPK and phospholipase A2, releasing arachidonic acid (AA) [23, 27, 28]. In room temperature stored PLTs, AA is metabolized by cyclooxygenase-1 (COX-1) to endoperoxides and further converted to thromboxane A2 by thromboxane synthase. Refrigerated temperature decreases COX-1 activity which causes AA accumulation [23]. Free AA has apoptotic properties manifested in caspase-9 induction followed by phosphatidylserine surface exposure [23]. It is possible that blockage of p38 MAPK phosphorylation by VX-702 does not involve the AA pathway because it has been previously reported that selective inhibition of p38 MAPK by VX-702 at physiological temperature did not impact thromboxane A2 production in human PLTs and had no effect on collagen-mediated PLT aggregation [12]. Although inhibition of sialidase activity by DANA improves recovery and survival without loss of hemostatic function of mouse PLTs suspended in buffer, [17, 23], it does not diminish the cold storage in vitro decrements on human PLT stored in 100% plasma. Alternative storage conditions, including the use of additive solutions and the use of sialidase inhibitors with improved stability must be identified to abrogate some of the decrements from prolonged cold storage. Clinical trials would then be appropriate to determine if platelet recovery and survival in humans are similarly improved.
Acknowledgements AS did study design, performed assays, contributed to the analysis and interpretation of data and wrote the manuscript. AT, DT-M, HA performed assays, acquired © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Inhibitors with 4 C PLT storage 367
and critically reviewed the data and manuscript, SJW oversaw the project, reviewed the data and revised manuscript. All authors approved the submitted version.
Conflict of interest None of the authors have any conflict of interest.
References 1 Holme S, Vaidja K, Murphy S: Platelet storage at 22 C: effect of type of agitation on morphology, viability, and function in vitro. Blood 1978; 52: 425–435 2 Simon TL, Nelson EJ, Carmen R, et al.: Extension of platelet concentrate storage. Transfusion 1983; 23:207–212 3 Heal JM, Singal S, Sardisco E, et al.: Bacterial proliferation in platelet concentrates. Transfusion 1986; 26: 388–390 4 Esber ES. Reduction of the Maximum Platelet Storage Period to 5 days in an Approved Container. FDA Memorandum. Rockville, MD: U.S. Food and Drug Administration, 1986. 5 Corash L: Bacterial contamination of platelet components: potential solutions to prevent transfusion-related sepsis. Expert Rev Hematol 2011; 4:509–525 6 Murphy S, Gardner FH: Effect of storage temperature on maintenance of platelet viability: deleterious effect of refrigerated storage. N Engl J Med 1969; 280:1094–1098 7 Grewal PK, Ashwell-Morell T, receptor. : Methods Enzymol 2010; 479: 223–241 8 Sørensen AL, Rumjantseva V, NayebHashemi S, et al.: Role of sialic acid for platelet life span: exposure of b-galactose results in the rapid clearance of platelets from the circulation by asialoglycoprotein receptorexpressing liver macrophages and hepatocytes. Blood 2009; 114:1645–1654 9 Jansen AJG, Josefsson EC, Rumjantseva V, et al.: Desialylation accelerates platelet clearance after refrigeration and initiates GPIba metalloproteinasemediated cleavage in mice. Blood 2012; 119:1263–1273 10 Canault M, Duerschmied D, Brill A, et al.: p38 mitogen-activated protein kinase activation during platelet storage: consequences for platelet
11
12
13
14
15
16
17
18
19
recovery and hemostatic function in vivo. Blood 2010; 115:1835–1842 Wagner SJ, Skripchenko A, Myrup A, et al.: Calcium is a key constituent for maintaining the in vitro properties of platelets suspended in the bicarbonatecontaining additive solution M-sol with low plasma levels. Transfusion 2010; 50:1028–1035 Kuliopulos A, Mohanial R, Covic L: Effect of selective inhibition of the p38 MAP kinase pathway on platelet aggregation. Thromb Haemost 2004; 92:1387–1393 Skripchenko A, Awatefe H, ThompsonMontgomery D, et al.: An inhibition of p38 mitogen activated protein kinase delays the platelet storage lesion. PLoS ONE 2013; 8:e70732 Holme S, Moroff G, Murphy S: A multi-laboratory evaluation of in vitro platelet assays: the tests for extent of shape change and response to hypotonic shock. Biomedical Excellence for Safer Transfusion Working Party of the International Society of Blood Transfusion. Transfusion 1998; 38: 31–40 Gordi T, Khamis H: Simple solution to a common statistical problem: interpreting multiple tests. Clin Ther 2004; 26:780–786 Kaufman RM: Uncommon cold: could 4 °C storage improve platelet function? Transfusion 2005; 45:1407–1414 Gitz E, Koekman CA, van den Heuvel DJ, et al.: Improved platelet survival after cold storage by prevention of glycoprotein Iba clustering in lipid rafts. Haemotologica 2012; 97:1873–1881 Rivera J, Lozano ML, Corral J, et al.: Quality assessment of platelet concentrates supplemented with second-messenger effectors. Transfusion 1999; 39:135–143 Kostelijk EH, Fijnheer R, Nieuwenhuis HK, et al.: Soluble p-selectin as parameter for platelet activation
© 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
20
21
22
23
24
25
26
27
28
during storage. Thromb Haemost 1996; 76:1086–1089 Babic A, Josefsson EC, Bergmeier W, et al.: In vitro function and phagocytosis of galactosylated platelet concentrates after long-term refrigeration. Transfusion 2007; 47:442–451 van der Wal DE, Du VX, Lo KS, et al.: Platelet apoptosis by cold-induced glycoprotein Ibalpha clustering. J Thromb Haemost 2010; 8:2554–2562 Hornsey VS, Drummond O, McMillan L, et al.: Cold storage of pooled, buffycoat-derived, leukoreduced platelets in plasma. Vox Sang 2008; 95:26–32 van der Wal DE, Gitz E, Du VX, et al.: Arachidonic acid depletion extends survival of cold stored platelets by interfering with [Glycoprotein Iba-143-3f] association. Haematologica 2012; 97:1514–1522 Soslau G, Giles J: The loss of sialic acid and its prevention in stored human platelets. Thromb Res 1982; 26:443–455 Davis AM, Teague SJ, Kleywegt GJ: Structure-guided ligand design and drug design; in Stroud RM, FinerMoore J (eds): Computational and Structural Approaches to Drug Discovery. Ligand Protein Interactions. Cambridge, UK, Royal Society of Chemistry, 2008:74–79 Hinou H, Miyoshi R, Takasu Y, et al.: A strategy for neuraminidase inhibitors using mechanism-based labeling information. Chem Asian J 2011; 6:1048–1056 Gousset K, Tsvetkova NM, Crowe JH, et al.: Important role of raft aggregation in the signaling events of coldinduced platelet activation. Biochim Biophys Acta 2004; 1660:7–15 Gousset K, Wolkers WF, Tsvetkova NM, et al.: Evidence for a physiological role for membrane rafts in human platelets. J Cell Physiol 2002; 190:117–128
Copyright of Vox Sanguinis is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL PAPER
Addition of sialidase or p38 MAPK inhibitors does not ameliorate decrements in platelet in vitro storage properties caused by 4 °C storage A. Skripchenko, D. Thompson-Montgomery, H. Awatefe, A. Turgeon & S. J. Wagner American Red Cross Biomedical Services, Holland Laboratory, Rockville, MD, USA
Background and Objectives Bacterial proliferation is inhibited in platelets (PLTs) stored at refrigerated temperatures, but also dramatically decreases PLT in vivo survival. Recent studies have demonstrated that cold temperature (CT) stored PLTs secrete sialidases upon re-warming, removing sialic acid from the PLT surface, which may be responsible for clustering of GPIba and PLT clearance from circulation. In this study, the influence of a sialidase inhibitor or a p38 MAP kinase inhibitor was evaluated in units stored at 4 °C. Materials and Methods After collection of a single Trima apheresis unit (n = 12), PLTs were aliquoted into four 60-ml CLX storage bags. One bag was stored at 20–24 °C (RT) with continuous agitation; a second bag was stored at 4 °C without agitation; a third bag was held at 4 °C without agitation with sialidase inhibitor, a fourth bag was incubated at 4 °C with a p38 MAPK inhibitor without agitation. Results Beginning from Day 1, all in vitro PLT parameters were adversely affected by CT compared to those of RT. Similar in vitro storage properties were observed in CT PLT in the presence or absence of sialidase or p38 MAPK inhibitors. P38 MAPK phosphorylation inhibition was not observed at CT. Decrease of sialidase activity was observed for 2 days in PLTs stored in additive solution but not in plasma. Received: 20 December 2013, revised 9 May 2014, accepted 2 June 2014, published online 27 June 2014
Conclusion Addition of either sialidase or p38 MAPK inhibitors do not improve any in vitro parameters of PLTs stored at 4 °C in 100% plasma. Key words: p38 MAPK, platelets properties, platelets storage lesion, sialidase inhibitor.
Introduction Platelets (PLTs) initially were stored at cold refrigerated temperature (4 °C) before Murphy and Gardner demonstrated that PLT storage at room temperature (22 °C) dramatically improved PLT recovery and survival after transfusion [1]. With introduction of highly gas permeable storage containers, PLT storage could be potentially extended to 7 days [2]. Unfortunately, 7 day storage was later found to be associated with an increased Correspondence: Andrey Skripchenko, American Red Cross, 15601 Crabbs Branch Way, Rockville, MD 20855, USA E-mail: [email protected]
360
rate of bacterial contamination [3], resulting in the reduction of the storage time back to 5 days [4]. Despite the introduction of automated bacterial culture, a substantial risk for transfusion-transmitted bacterial infection remains. Based on the recent studies, the prevalence of bacterial contamination is estimated as one in 750 to one in 1000 PLT concentrates [5]. Cold temperature (CT) storage would be expected to slow bacterial growth, but PLTs do not tolerate CT and are rapidly cleared in the liver following transfusion [6, 7]. One of the pathways of PLT clearance is the removal of sialic acid from GPIba, a subunit of PLT von Willebrand factor receptor, by active sialidases secreted from the PLTs upon re-warming after CT storage [8]. Sialidase
Inhibitors with 4 C PLT storage 361
activation also stimulates metalloproteinase ADAM17, also known as TACE (TNF-a-converting enzyme), which cleaves GPIba and GPV from the PLT surface [9]. Shedding desialylated GPIba and GPV have been correlated with poor recovery and survival in a mouse model [9]. Inhibition of sialidase activity by N-acetylneuraminic acid, 2,3-dehydro-2-deoxy-sodium salt (DANA) diminished shedding of GPIba and GPV and greatly improved PLT recovery and survival in mice after refrigerated storage for 48 h [9]. On other hand, activation of ADAM17 is regulated by p38 mitogen-activated protein kinase (MAPK) activation [10]. It has been shown in animal models that inhibition of ADAM17 and p38 MAPK activity enhances the recovery of PLT stored at room temperature [10]. The goal of this study was to compare PLTs stored at CT in 100% plasma for 7 days with or without a competitive sialidase inhibitor, DANA, or with or without a second generation of p38 MAPK inhibitor, VX-702.
Materials and methods Materials The sialidase inhibitor, DANA, was purchased from EMD Chemicals, subsidiary of EMD Millipore (Billerica, MA, USA). A second generation of p38 MAPK inhibitor, VX-702, was obtained from Selleck Chemicals LLC (Houston, TX, USA). Dimethylsulphoxide (DMSO) was purchased from Sigma-Aldrich (Sigma-Aldrich Corporation, St. Louis, MO, USA). Phosphate-buffered solution (PBS) pH 74 was obtained from Life Technologies (Life Technologies Corporation, NY, USA).
PLT collection and study design Twelve independent experiments were carried out. In each, a single apheresis unit was collected under informed consent using the Trima cell separator from healthy volunteers with a targeted yield of 40 9 1011 PLTs and targeted PLT concentration 14 9 106/ll. The unit was divided into four identical 60-ml aliquots in CLX storage bags (Medsep Corporation, a Subsidiary of Haemonetics Corporation, Covina, CA, USA) immediately after collection. Aliquot A) was stored at 20–24 °C with continuous agitation (RT); aliquot B) was designated as a cold temperature control and incubated at 4 °C without agitation (CT); aliquot C) PLTs were stored with the addition of a sialidase inhibitor without agitation at 4 °C (DANA); aliquot D) was stored at 4 °C with the addition of a p38 MAPK inhibitor without agitation (VX-702). To assess effectiveness of added inhibitors, a satellite study with six experiments was carried out. In each, © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
a single apheresis unit was collected using the same separator and the same parameters. The unit was divided into six identical 60-ml aliquots in CLX storage bags. All aliquots were sedimented with 4200 g and ACE 550 9 107 (Sorvall RC3BP; Thermo Fisher Scientific Inc., Waltham, MA, USA). After 1 h rest, plasma was replaced with PLT additive solution (PAS) M-Sol [11] in one bag resulting in 5% plasma. All other aliquots were resuspended in plasma. Aliquot 1) PLTs were stored with addition of DANA at 4 °C without agitation in 100% plasma (CT-DANA-PL); aliquot 2) PLTs were stored with addition of DANA in 95% PAS/5% plasma without agitation (CT-DANA-PAS); aliquot 3) PLTs were stored with addition of VX-702 at 4 °C without agitation (CT-VX); aliquot 4) PLTs were stored with addition of VX-702 at 20–24 °C with continuous agitation (RT-VX); aliquot 5) PLTs were stored at 4 °C without agitation (CT-CNTR); aliquot 6) PLTs were stored at 20–24 °C with continuous agitation (RT-CNTR). In these experiments, the following parameters were measured: phosphorylated (Thr180/Tyr182) and total p38 MAPK, b-N-acetylglucosamine (b-GlcNAc)terminated glycans exposure on GPIba, p-selectin exposure on the PLT membrane and Annexin V binding.
Drug delivery A stock solution of DANA was prepared in PBS at 319 mM (10 mg/ml). A stock solution of VX-702 was prepared in DMSO at 618 mM (25 mg/ml). Final drug concentrations of DANA and VX-702 in the PLT aliquots were 1 mM and 1 lM, respectively, and were identical to those previously reported [9, 12]. The final DMSO concentration in PLT aliquots D, 3 and 4 (satellite study) was 0016%; this concentration has been previously shown not to affect PLT storage properties [13].
PLT assays All PLT units were sampled by syringe (approximate volume, 38 ml) on Days 1, 2, 4 and 7 using a CBS needlefree spike (OriGen Biomedical, Austin, TX, USA). PLT concentrations were measured using a Sysmex XE 2100D haematology analyser (Sysmex Corporation, Lincolnshire, IL, USA). Metabolic parameters were assessed measuring pH, glucose, lactate, bicarbonate, oxygen and carbon dioxide levels. PLT extracellular pH was measured using a bench top meter (Orion, Thermo Scientific, Beverly, MA) and pH electrode (Accu-pHast; Fisher Scientific, Pittsburg, PA, USA). Other parameters were determined using a blood gas analyzer (Cobas b221; Roche Diagnostics, Indianapolis, IN, USA). The bicarbonate concentration was calculated automatically from pH and CO2 (37 °C) levels. Rates
362 A. Skripchenko et al.
for bicarbonate neutralization, glucose consumption and lactate generation were calculated from measurements on Day 1 and Day 7 and normalized by day and 1010 PLTs. Structural parameters were evaluated by measuring morphology, Annexin V binding, and expression of CD62P and CD42b. PLT morphology was expressed as the percentage of PLTs with discoid morphology using a phase microscopy as previously described [1]. Annexin V binding and expression of CD62P (p-selectin) and CD42b (GPIba) were measured by flow cytometry (FACSCalibur; BD Biosciences, San Jose, CA, USA). To measure the percentage of Annexin V positive PLTs, samples were washed and resuspended to 1 9 106 PLTs/ml with manufacturer supplied binding buffer. PLTs were incubated with fluorescein isothiocyanate (FITC)-conjugated CD61 (Biolegend, San Diego, CA, USA) monoclonal antibodies and with phycoerythrin (PE)-conjugated Annexin V (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA) at saturating concentrations for 15 min at 22 °C. PLTs were gated by forward and side scattering and binding of CD61. To measure the percent of PLTs expressing membrane-bound CD62P and CD42b, PLT samples were diluted to 1 9 106 PLTs/ml with PBS supplemented with 01% human albumin (HA) (American Red Cross, Blood Services; Washington DC, USA). Sample aliquots were incubated with FITC-conjugated CD61 (Biolegend) monoclonal antibodies and with PE-conjugated CD62P (Biolegend) or PE-conjugated CD42b monoclonal antibodies (Biolegend) at saturating concentrations for 15 and 120 min, respectively, at 22 °C. Mouse IgG1 PE and FITC PLT isotype control (Biolegend) were used at saturating concentrations as negative controls. PLTs were gated by forward and side scattering and binding of CD61. Appropriate color compensation was set for FITC fluorescence and PE fluorescence by using 530 and 585 nm bandpass filters respectively. Standard three-colour beads (Calibrite 3; BD Biosciences) were used for daily instrument calibration and FITC and PE colour compensation. Functional parameters were assessed measuring hypotonic shock response (HSR), extent of shape change (ESC) and PLT aggregation as previously described [13, 14]. Briefly, the ESC and HSR levels were measured turbimetrically using a Chrono-Log SPA 2000 (Chrono-Log Corporation, Havertown, PA, USA). PLT aggregation was measured using a lumiaggregometer (Chrono-Log) and 10 lM of ADP (Chrono-Log) with 10 lg/ml of collagen (Chrono-Log), as final concentrations. Agonists were added simultaneously to PLT suspensions. Aggregation was measured according to manufacturer’s protocol and expressed relative to a maximum slope and amplitude of the instrument’s internal electronic standard. Platelets mitochondrial parameters were evaluated by measuring changes in mitochondrial membrane potential
(MMP) and intracellular reactive oxygen species (ROS) generation. The MMP was determined using a MitoProbe JC-1 Assay Kit (Molecular Probes; Invitrogen, Eugene, OR, USA). Samples were diluted to 1 9 106 PLTs/ml with PBS supplemented with 01% HA and stained with 1 lM of JC-1 for 15 min at 37 °C. Samples were analysed using a flow cytometer (FACSCalibur). ROS generation was monitored by measuring changes in fluorescence resulting from the oxidation of intracellular probes, dihydroethidium (DHE) (Molecular Probes; Invitrogen) and 5(and-6)-chloromethyl-20 ,70 -dichlorodihydrofluorescein diacetate, acetyl ester (CM-H2DCFDA, Molecular Probes; Invitrogen). Samples were diluted to 1 9 106 PLTs/ml with PBS supplemented with 01% HA and stained with 5 lM DHE and 5 lM CM-H2DCFDA for 15 min at 37 °C. Fluorescence of DHE and CM-H2DCFDA was detected in the FL2 and FL1 region of a FACSCalibur flow cytometer, respectively. The amount of phosphorylated (Thr180/Tyr182) p38 MAPK was measured using a STAR (Signal Transduction Assay Reaction) ELISA kit (EMD Millipore). Total p38 MAPK content was measured (InstantOne ELISA kit; eBioscience, San Diego, CA, USA). To measure total and phospho-p38 MAPK, PLTs were centrifuged at 850 g in the presence of 1 lg/ml prostaglandin E1, the pellet was resuspended in ice cold PBS supplemented with 01% HA and lysed at 0 °C using 2X platelet lysis buffer (2% NP40, 30 mM Hepes, 150 mM NaCl, 2 mM EDTA, pH 74). Cell debris were sedimented at 18 000 g for 5 min, and supernatant was aliquoted and stored at -80 °C before immunoassays. All assays were performed in duplicate according to manufacturer’s protocols. Concentrations of phospho-p38 MAPK were expressed as Units/ml per 1010 PLTs to normalize measurements to PLT count. To measure the effectiveness of DANA, a flow cytometric analysis of b-GlcNAc-terminated glycans exposure on GPIba was performed using a FACSCalibur flow cytometer and 01 lg/ml succinyl Triticum vulgare lectin from wheat germ (s-WGA; EY Laboratories, Inc., San Mateo, CA, USA) conjugated with FITC as previously described [9].
Statistics Determination of means and standard deviations of experimental values and Analysis of Variance with repeated measures were carried out using standard software (Instat; GraphPad Software, San Diego, CA, USA). A value of P ≤ 0001 was considered significant taking into account repeated measures for the 18 PLT assays utilized in this study and three testing days [15]. For a satellite study, a value of P ≤ 00125 was considered significant taking into account repeated measures of 4 days of testing [15]. Statistical differences between all paired values © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Inhibitors with 4 C PLT storage 363
and VX-702 aliquots on any given day of storage. GPIba expression was similar in all PLT aliquots on Days 1 and 2 and the GPIba expression in the RT aliquots was greater than those in CT, DANA and VX-702 aliquots on Days 4 and 7 (P < 001). Results of functional PLT parameters measured during storage are given in Fig. 2. HSR, ESC and both aggregation measurements were less in CT, DANA and VX-702 aliquots throughout storage compared to those of RT (P < 0001). Values of all functional parameters were similar between CT, DANA and VX-702 aliquots on any given day during storage except for the values of aggregation slope for DANA and VX-702 on Day 1, which were greater than those of CT (P < 0001). Results of the measurements of mitochondrial function during storage are shown in Fig. 3. No differences were observed in PLT MMP levels during storage between CT, DANA and VX-702 aliquots. All aliquots stored at cold temperature had MMP levels greater than those of RT on Day 1 (P < 0001). On Day 2 only, the PLT MMP levels were greater in DANA and VX-702 compared to those of RT (P < 001). On Day 7, the MMP levels of CT PLTs were less than those of RT PLTs (P < 005). ROS accumulations were similar between all aliquots during storage with the exceptions that the level of DHE which was greater in DANA and VX-702 aliquots on Day 7 (P < 001) and the CM-H2DCFDA level was greater in CT and DANA aliquots on Day 4 compared to those of RT (P < 001).
of RT, CT, DANA and VX-702 aliquots were determined by post Hoc tests with Bonferroni corrections.
Results Results of this study which provide data for PLT storage parameters measured on Days 1–7 are given in Table 1 and Figs 1–3. Assessments of metabolic parameters of PLTs stored at RT, CT, CT with DANA or CT with VX-702 are shown in Table 1. All parameters of CT, DANA and VX-702 were significantly different compared to those of RT. The pH levels of DANA PLT aliquots were less than those of VX-702 on Days 1 and 2. No differences were observed in glucose, lactate and bicarbonate levels between CT, DANA and VX-702 during storage except for bicarbonate levels on Day 1 between RT and all CT treatments (Table 1). Results of structural PLT parameter measurements are presented in Fig. 1. Morphology score was less in all aliquots stored at cold temperature compared to those of RT (P < 0001), but those stored at CT in the presence or absence of DANA or VX-702 did not differ from each other. The percent Annexin V positive PLT stored at CT were greater (P < 005), with the exception of VX-702 aliquots on Days 2 and 7, than those stored at RT. PLT p-selectin expression was less on Days 1 and 2 (P < 005) and greater on Day 7 in RT aliquots compared to those stored at cold temperature (P < 0001). No differences in p-selectin expression were observed between CT, DANA
Table 1 Assessments of metabolical parameters of PLTs stored for 7 days at RT, CT, CT with DANA or CT with VX-702 pH Day 1
Day 2
Day 4
Day 7
RT CT DANA VX-702 RT CT DANA VX-702 RT CT DANA VX-702 RT CT DANA VX-702
747 730 727 732 752 736 734 737 734 745 742 746 714 747 745 750
Glucose (mM) – – – – – – – – – – – – – – – –
004 005a 005ab 006a 003 005a 004ab 005a 005 006a 006a 006a 010 008a 008a 009a
193 198 195 197 183 193 193 195 147 184 181 183 118 177 174 176
– – – – – – – – – – – – – – – –
31 30a 30a 31a 33 32a 30a 30a 30 31a 31a 29a 30 32a 31a 30a
Data are presented as mean – one standard deviation. a P < 0001, compared to that of RT values. b P < 001, compared between DANA and VX-702 values.
© 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Lactate (mM) 35 29 29 29 55 39 38 38 128 64 63 62 178 81 78 77
– – – – – – – – – – – – – – – –
06 06a 06a 06a 09 07a 07a 07a 22 11a 12a 12a 34 15a 14a 15a
Bicarbonate (mM) 177 195 193 198 156 181 177 183 103 153 152 154 67 136 134 139
– – – – – – – – – – – – – – – –
18 21a 22a 20a 13 20a 20a 19a 11 14a 15a 14a 13 11a 15a 12a
pO2 (mm Hg) 124 203 202 205 126 205 204 209 138 212 211 211 146 218 217 220
– – – – – – – – – – – – – – – –
10 6a 7a 7a 8 10a 9a 10a 12 9a 9a 9a 8 10a 10a 8a
pCO2 (mm Hg) 402 655 666 642 336 541 556 537 287 392 405 382 256 346 355 340
– – – – – – – – – – – – – – – –
39 96a 67a 66a 34 63a 59a 59a 36 54a 52a 50a 25 35a 41a 35a
364 A. Skripchenko et al.
Fig. 1 Assessments of structural parameters of PLTs stored for 7 days at RT, CT, CT with DANA or CT with VX-702. Data are presented as mean – one standard deviation. Opened circle – RT-CNTR, closed circle – CT-CNTR, opened square – CT-VX, opened triangular – CT-DANA. [Correction added on 14 July 2014, after first online publication: the title of the fourth panel has been corrected].
Results of the satellite study are presented in Fig. 4 and Table 2. Measurements of phosphorylated p38 MAPK are shown in Fig. 4. Levels of phosphorylated p38 MAPK were greater in CT-VX PLTs on Day 4 and Day 7 (P ≤ 0034) compared to those of RT-VX but was comparable on Day 1. Levels of phosphorylated p38 MAPK in CT-VX were less than that of CT-CNTR on Day 1 (P < 001) but were not different on Days 4 and 7. RT-VX samples had less phosphorylation on Days 4 and 7 compared to those of RT-CNTR (P < 005 and P < 0001, respectively). Phos-
Fig. 2 Assessment of functional parameters of PLTs stored for 7 days at RT, CT, CT with DANA or CT with VX-702. Data are presented as mean – one standard deviation. Opened circle – RT-CNTR, closed circle – CT-CNTR, opened square – CT-VX, opened triangular – CT-DANA.
phorylation was greater in CT-CNTR PLTs compared to that of RT-CNTR during storage (P < 0001). Total p38 MAPK content was constant during storage and similar between all PLT aliquots (data not shown). S-WGA binding to PLTs is presented in Table 2. S-WGA binding was greater in CT-CNTR PLTs compared to that of RT-CNTR on Day 1 and was comparable to that of CT-DANA-PL. S-WGA binding in CT-DANA-PAS PLTs was less than that of CT-DANA-PL on Days 1 and 2 but similar on Day 4 and 7. In addition, S-WGA binding in CT-DANA-PAS was less than that of CT-CNTR on Days 1–2 but was comparable on Days 4 and 7. Results of © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Inhibitors with 4 C PLT storage 365
Fig. 4 Levels of phosphorylated p38 MAPK of PLTs stored for 7 days at RT and CT with or without VX-702. Data are presented as mean – one standard deviation. Opened circle – RT, closed circle – RT-VX, opened square – CT, closed square – CT-VX.
Fig. 3 Assessment of mitochondrial parameters of PLTs stored for 7 days at RT, CT, CT with DANA or CT with VX-702. Data are presented as mean – one standard deviation. Opened circle – RT-CNTR, closed circle – CT-CNTR, opened square – CT-VX, opened triangular – CT-DANA.
Annexin V binding and p-selectin surface exposure on PLT were comparable to those obtained in the main study (data not shown).
Discussion Refrigerated storage results in PLT decrements known as the ‘PLT cold storage lesion’ [16]. P38 MAPK is implicated in the PLT storage lesion under normal conditions, and appears to be associated with recovery and survival of mouse PLTs [10]. When PLTs are stored at CT, their recovery and survival are very low [6]. In a mouse model, desialylation of GPIba accelerates PLT removal after CT storage [9]. Addition of the competitive sialidase inhibitor, DANA, to the refrigerated mouse PLTs enhanced recoveries and circulation time after transfusion [9, 17]. In this study, the in vitro parameters of PLTs stored at 4 °C with addition of DANA or VX-702 were evaluated. © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Addition of either DANA or VX-702 to PLTs stored at 4 °C did not affect any PLT metabolic parameters, which were all comparable to those of CT. In contrast, addition of VX-702 to the PLTs stored at 20–24 °C decreased the glucose consumption and lactate accumulation rates [13]. In a different study, investigators evaluated the possibility of ThromboSol, a mixture of second messengers, effectors, or PLT inhibitors, to ameliorate changes caused by cold storage. Treating PLTs with ThromboSol during cold storage also did not also affect metabolic parameters [18]. Assessment of changes in structural PLT parameters (morphology and CD62P) with the addition of either DANA or VX-702 inhibitors did not reveal differences between any of the Day 7 aliquots stored at cold temperature. In contrast, addition of p38 MAPK inhibitor to
Table 2 S-WGA binding to PLTs stored for 7 days at RT and CT in 100% plasma or PAS (5% plasma level)
CT-DANAPL CT-DANAPAS CT-CNTR RT-CNTR
Day 1
Day 2
Day 4
Day 7
414 – 125ab
500 – 154ab
435 – 155
434 – 207
292 – 103bc
355 – 214bd
278 – 102
326 – 122
472 – 97a 370 – 96
469 – 77 344 – 88
430 – 135 317 – 116
418 – 152 267 – 78
Data are presented as mean – one standard deviation. P < 005, compared to that of RT-CNTR values. b P < 001, compared between CT-DANA-PL and CT-DANA-PAS. c P < 0001, compared to that of CT-CNTR. d P < 005, compared to that of CT-CNTR. a
366 A. Skripchenko et al.
PLTs stored at room temperature significantly reduced CD62P expression and greatly improved morphology score [13]. A future study would benefit from assessing p-selectin concentrations in the supernatant of PLT suspensions [19]. No differences were observed in Annexin V binding between any cold stored aliquots. It has been previously shown that the addition of a p38 MAPK inhibitor reduced the percentage of Annexin V positive PLTs during 7 daystorage at 20–24 °C [13]. Our findings of elevated Annexin V binding at refrigerated temperature compared to room temperature are in accordance with previously published data [20, 21]. Our results also demonstrated a loss of GPIba expression during 7 days at cold temperature with or without inhibitors compared to RT storage. Rivera and coauthors reported similar results in GPIba expression loss during cold storage with the ThromboSol treatment [18]. As expected, all PLT functional parameters were decreased during refrigerated storage starting on Day 1 as previously reported [22]. Addition of either p38 MAPK or sialidase inhibitors did not improve ESC, HSR and aggregation. These results contrasted to previously reported data when addition of VX-702 to PLTs during 7-day RT storage markedly improved ESC and HSR [13]. Inhibition of p38MAPK does not affect human PLT aggregation during storage in 100% plasma or freshly prepared mouse or human platelets suspended in either 100% plasma or buffer in response to one or more synergistic agonists [10, 13]. In our experiments, cold storage of PLTs triggered more rapid loss of MMP levels and greater ROS accumulation in PLTs. Addition of DANA or VX-702 did not prevent MMP loss or ROS accumulation. In contrast to these results, addition of VX-702 to PLTs dramatically improved mitochondrial function when PLTs suspended in 100% plasma were stored at RT [13]. Addition VX-702 to PLTs stored at RT diminishes phosphorylation of p38 MAPK during storage, which is associated with improved in vitro storage parameters. As expected, CT storage of platelets caused a significant increase in phosphorylation of p38 MAPK [23]. Addition of VX-702 decreased p38 MAPK phosphorylation only on Day 1 of CT storage, but failed to overcome the level of p38 MAPK phosphorylation caused by CT storage. GPIba is glycosylated with N-linked carbohydratechains containing sialic acids [24]. PLTs stored at refrigerated temperature release sialidase and subsequently remove sialic acid residues upon re-warming [9]. The addition of DANA prevented sialidase activity and GPIba shedding with enhanced PLT recovery and survival in an animal model [9]. In our experiments, DANA failed to improve any in vitro properties of PLTs stored at refrigerated temperature in 100% plasma, including levels of GPIba. At the same time, sialidase activity was not
inhibited by DANA for CT PLTs suspended in 100% plasma. Furthermore, DANA inhibited the sialidase activity for CT PLTs suspended in PAS (5% residual plasma) during the first 2 days of storage, which corroborates previously published data of sialidase inhibition reported by Jansen et al. [9]. In their study, experiments were performed in a glucose containing buffer rather than the 100% plasma utilized in this study. More work needs to be done to understand how plasma proteins and plasma levels affect DANA activity on PLT in vitro storage properties. The breakdown of DANA during storage also cannot be ruled out. Some sialidase inhibitors bind to plasma proteins depending on their lipophilicity and charge [25]. Moreover, the potency of DANA to inhibit platelet sialidase may not be optimal to affect PLT in vitro properties; more potent analogues of DANA have been developed [26]. Finally, a possible interaction between DANA and other PLT receptors cannot be ruled out [25]. The addition of p38 MAPK inhibitor prevents shedding of GPIba and improved PLT recovery and survival during RT storage [10]. It has been previously reported that cooling PLTs to < 10 °C activated both p38 MAPK and phospholipase A2, releasing arachidonic acid (AA) [23, 27, 28]. In room temperature stored PLTs, AA is metabolized by cyclooxygenase-1 (COX-1) to endoperoxides and further converted to thromboxane A2 by thromboxane synthase. Refrigerated temperature decreases COX-1 activity which causes AA accumulation [23]. Free AA has apoptotic properties manifested in caspase-9 induction followed by phosphatidylserine surface exposure [23]. It is possible that blockage of p38 MAPK phosphorylation by VX-702 does not involve the AA pathway because it has been previously reported that selective inhibition of p38 MAPK by VX-702 at physiological temperature did not impact thromboxane A2 production in human PLTs and had no effect on collagen-mediated PLT aggregation [12]. Although inhibition of sialidase activity by DANA improves recovery and survival without loss of hemostatic function of mouse PLTs suspended in buffer, [17, 23], it does not diminish the cold storage in vitro decrements on human PLT stored in 100% plasma. Alternative storage conditions, including the use of additive solutions and the use of sialidase inhibitors with improved stability must be identified to abrogate some of the decrements from prolonged cold storage. Clinical trials would then be appropriate to determine if platelet recovery and survival in humans are similarly improved.
Acknowledgements AS did study design, performed assays, contributed to the analysis and interpretation of data and wrote the manuscript. AT, DT-M, HA performed assays, acquired © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
Inhibitors with 4 C PLT storage 367
and critically reviewed the data and manuscript, SJW oversaw the project, reviewed the data and revised manuscript. All authors approved the submitted version.
Conflict of interest None of the authors have any conflict of interest.
References 1 Holme S, Vaidja K, Murphy S: Platelet storage at 22 C: effect of type of agitation on morphology, viability, and function in vitro. Blood 1978; 52: 425–435 2 Simon TL, Nelson EJ, Carmen R, et al.: Extension of platelet concentrate storage. Transfusion 1983; 23:207–212 3 Heal JM, Singal S, Sardisco E, et al.: Bacterial proliferation in platelet concentrates. Transfusion 1986; 26: 388–390 4 Esber ES. Reduction of the Maximum Platelet Storage Period to 5 days in an Approved Container. FDA Memorandum. Rockville, MD: U.S. Food and Drug Administration, 1986. 5 Corash L: Bacterial contamination of platelet components: potential solutions to prevent transfusion-related sepsis. Expert Rev Hematol 2011; 4:509–525 6 Murphy S, Gardner FH: Effect of storage temperature on maintenance of platelet viability: deleterious effect of refrigerated storage. N Engl J Med 1969; 280:1094–1098 7 Grewal PK, Ashwell-Morell T, receptor. : Methods Enzymol 2010; 479: 223–241 8 Sørensen AL, Rumjantseva V, NayebHashemi S, et al.: Role of sialic acid for platelet life span: exposure of b-galactose results in the rapid clearance of platelets from the circulation by asialoglycoprotein receptorexpressing liver macrophages and hepatocytes. Blood 2009; 114:1645–1654 9 Jansen AJG, Josefsson EC, Rumjantseva V, et al.: Desialylation accelerates platelet clearance after refrigeration and initiates GPIba metalloproteinasemediated cleavage in mice. Blood 2012; 119:1263–1273 10 Canault M, Duerschmied D, Brill A, et al.: p38 mitogen-activated protein kinase activation during platelet storage: consequences for platelet
11
12
13
14
15
16
17
18
19
recovery and hemostatic function in vivo. Blood 2010; 115:1835–1842 Wagner SJ, Skripchenko A, Myrup A, et al.: Calcium is a key constituent for maintaining the in vitro properties of platelets suspended in the bicarbonatecontaining additive solution M-sol with low plasma levels. Transfusion 2010; 50:1028–1035 Kuliopulos A, Mohanial R, Covic L: Effect of selective inhibition of the p38 MAP kinase pathway on platelet aggregation. Thromb Haemost 2004; 92:1387–1393 Skripchenko A, Awatefe H, ThompsonMontgomery D, et al.: An inhibition of p38 mitogen activated protein kinase delays the platelet storage lesion. PLoS ONE 2013; 8:e70732 Holme S, Moroff G, Murphy S: A multi-laboratory evaluation of in vitro platelet assays: the tests for extent of shape change and response to hypotonic shock. Biomedical Excellence for Safer Transfusion Working Party of the International Society of Blood Transfusion. Transfusion 1998; 38: 31–40 Gordi T, Khamis H: Simple solution to a common statistical problem: interpreting multiple tests. Clin Ther 2004; 26:780–786 Kaufman RM: Uncommon cold: could 4 °C storage improve platelet function? Transfusion 2005; 45:1407–1414 Gitz E, Koekman CA, van den Heuvel DJ, et al.: Improved platelet survival after cold storage by prevention of glycoprotein Iba clustering in lipid rafts. Haemotologica 2012; 97:1873–1881 Rivera J, Lozano ML, Corral J, et al.: Quality assessment of platelet concentrates supplemented with second-messenger effectors. Transfusion 1999; 39:135–143 Kostelijk EH, Fijnheer R, Nieuwenhuis HK, et al.: Soluble p-selectin as parameter for platelet activation
© 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 360–367
20
21
22
23
24
25
26
27
28
during storage. Thromb Haemost 1996; 76:1086–1089 Babic A, Josefsson EC, Bergmeier W, et al.: In vitro function and phagocytosis of galactosylated platelet concentrates after long-term refrigeration. Transfusion 2007; 47:442–451 van der Wal DE, Du VX, Lo KS, et al.: Platelet apoptosis by cold-induced glycoprotein Ibalpha clustering. J Thromb Haemost 2010; 8:2554–2562 Hornsey VS, Drummond O, McMillan L, et al.: Cold storage of pooled, buffycoat-derived, leukoreduced platelets in plasma. Vox Sang 2008; 95:26–32 van der Wal DE, Gitz E, Du VX, et al.: Arachidonic acid depletion extends survival of cold stored platelets by interfering with [Glycoprotein Iba-143-3f] association. Haematologica 2012; 97:1514–1522 Soslau G, Giles J: The loss of sialic acid and its prevention in stored human platelets. Thromb Res 1982; 26:443–455 Davis AM, Teague SJ, Kleywegt GJ: Structure-guided ligand design and drug design; in Stroud RM, FinerMoore J (eds): Computational and Structural Approaches to Drug Discovery. Ligand Protein Interactions. Cambridge, UK, Royal Society of Chemistry, 2008:74–79 Hinou H, Miyoshi R, Takasu Y, et al.: A strategy for neuraminidase inhibitors using mechanism-based labeling information. Chem Asian J 2011; 6:1048–1056 Gousset K, Tsvetkova NM, Crowe JH, et al.: Important role of raft aggregation in the signaling events of coldinduced platelet activation. Biochim Biophys Acta 2004; 1660:7–15 Gousset K, Wolkers WF, Tsvetkova NM, et al.: Evidence for a physiological role for membrane rafts in human platelets. J Cell Physiol 2002; 190:117–128
Copyright of Vox Sanguinis is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
