ORIGINAL ARTICLE Metabolomic analysis of platelets during storage: a comparison between apheresis- and buffy coat–derived platelet concentrates Giuseppe Paglia,1 Ólafur E. Sigurjónsson,2,3 Óttar Rolfsson,1 Morten Bagge Hansen,4 Sigurður Brynjólfsson,1 Sveinn Gudmundsson,2 and Bernhard O. Palsson1

BACKGROUND: Platelet concentrates (PCs) can be prepared using three methods: platelet (PLT)-rich plasma, apheresis, and buffy coat. The aim of this study was to obtain a comprehensive data set that describes metabolism of buffy coat–derived PLTs during storage and to compare it with a previously published parallel data set obtained for apheresis-derived PLTs. STUDY DESIGN AND METHODS: During storage we measured more than 150 variables in 8 PLT units, prepared by the buffy coat method. Samples were collected at seven different time points resulting in a data set containing more than 8000 measurements. This data set was obtained by combining a series of standard quality control assays to monitor the quality of stored PLTs and a deep coverage metabolomics study using liquid chromatography coupled with mass spectrometry. RESULTS: Stored PLTs showed a distinct metabolic transition occurring 4 days after their collection. The transition was evident in PLT produced by both production methods. Apheresis-derived PLTs showed a clearer phenotype of PLT activation during early days of storage. The activated phenotype of apheresis PLTs was accompanied by a higher metabolic activity, especially related to glycolysis and the tricarboxylic acid cycle. Moreover, the extent of the activation differed between bags resulting in interbag variability in the storage lesion of apheresis-prepared PLTs. This may be related to donor-related polymorphism. CONCLUSION: This study demonstrated two discrete metabolic phenotypes in stored PLTs prepared with both apheresis and buffy coat methods. PLT activation occurs during the first metabolic phenotype and might lead to a low reproducibility of the apheresis PCs.

P

latelet concentrates (PCs) are transfused to treat several clinical situations.1 Defining and improving the quality of PCs during production and storage are important aspects of modern transfusion medicine.2 PCs are obtained by harvesting platelets (PLTs) from donors by apheresis or by whole blood using the buffy coat or the PLT-rich plasma method. PLTs derived from the PLT-rich plasma method are mainly used in the United States, while in Europe the buffy coat method is preferred.3 In apheresis procedures, PLTs are obtained from a single donor, while in both PLT-rich plasma and buffy coat PLTs derived from four to six donors are pooled to obtain a single unit equivalent to an apheresis product.3-5 PCs are routinely stored for 5 days.6-8 In some countries, the introduction of new technologies for detecting, limiting, or inactivating bacterial contamination has made a further extension of the PC expiration date possible, until 7 days of storage.9,10 Besides the risk of bacterial contamination, the limiting factor affecting quality and

ABBREVIATIONS: PC(s) = platelet concentrate(s); PCA = principal component analysis; PSL = platelet storage lesion; sP-selectin = soluble P-selectin; TCA = tricarboxylic acid; UPLC = ultra performance liquid chromatography. From the 1Center for Systems Biology, University of Iceland; 2 The Blood Bank, Landspitali-University Hospital; and the 3 School of Science and Engineering, Reykjavik University, Reykjavik, Iceland; and the 4Department of Clinical Immunology, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark. Address reprint requests to: Ólafur E. Sigurjónsson, The Blood Bank, Landspitali University Hospital, Snorrabraut 60, 105 Reykjavik, Iceland; e-mail: [email protected] This work was supported by the European Research Council Grant Proposal 232816. Received for publication May 10, 2014; revision received July 15, 2014, and accepted July 16, 2014. doi: 10.1111/trf.12834 © 2014 AABB TRANSFUSION **;**:**-**. Volume **, ** **

TRANSFUSION

1

PAGLIA ET AL.

time of storage of PCs is the development of a series of morphologic and metabolic changes, known as the PLT storage lesion (PSL),11 that are associated with loss of the PLT functions.12,13 PSL has been associated with storage conditions as well as production methods. In fact, several in vitro studies reported that PLTs derived from apheresis or whole blood present different degrees of PSL, with specific differences in the profiles of activation markers and mitochondrial activity.14-17 Despite these results, there is an ongoing discussion as to the preferred method for obtaining PCs, and there is no agreement as to whether apheresis- and whole blood–derived PLTs can be considered comparable in quality.3,18,19 A combination of several factors, including production methods and composition of the PLT storage solution needs to be considered for defining and improving the quality of PCs.4,7,19-26 To further uncover the complex mechanisms involved in the development of PSL, and to obtain comprehensive data sets, new integrated omics and systems biology approaches may be useful.27-35 In a previous study, we reported the first metabolomics study on apheresis-derived PLTs, resulting in the most comprehensive data set reported for PLT metabolism to date. The results highlighted that PLTs undergo complex metabolic changes during storage that lead to the expression of two different metabolic phenotypes associated with the storage time.36 A recent retrospective study suggested that transfusion of PCs stored for 4 or 5 days resulted in poorer clinical outcomes, supporting the idea that the quality of PCs is associated with the time of storage.37 In this study we have used experimental approaches as previously described36 to investigate the metabolism of stored buffy coat–derived PLTs. We systematically compared the data obtained in this study with the data obtained from apheresis derived PLTs.36

MATERIALS AND METHODS Sample collection PCs were obtained from a pool of eight healthy blood donors by the buffy coat method at the Blood Bank, Landspitali-University Hospital, Iceland. In total, 8 units of buffy coat PLTs were obtained from donors of mixed age, sex, and blood types. PLTs were stored in additive solution (AS; T-SOL, Fenwal, Lake Zurich, IL) in a plastic container (PL 2410, Fenwal). After 1 day of resting, sampling was performed on Days 1, 3, 4, 5, 6, 7, and 10. Units were stored in a shaking incubator set at 22°C and were discarded after completion of sampling on Day 10. No more than 4.2 mL was removed on each day of sampling using a sterile connected clave valve; the weight of the bag went from approximately 290 g in the 2

TRANSFUSION Volume **, ** **

beginning down to approximately 258 g on Day 10 (see Table S1, available as supporting information in the online version of this paper). PLTs were stored in a PLT incubator (PC900h Horizon, Helmer, Noblesville, IN) at 22°C. To reduce the possibility of bacterial contamination during sampling, the clave valve was wiped with an 85% alcohol solution and approximately 1 mL of PC was collected into a syringe with a Luer-Lock connector and discarded. After the completion of sampling on Day 10, a 10-cm-long section of the sampling line was removed with a sterile tubing sealer and sent for pathogen analysis at Landspitali-University Hospital, where it was confirmed that no bacteria were present in the PCs. In the apheresis PLT units, samples were taken on Days 0, 1, 3, 4, 5, 6, 7, and 10 where Day 0 is the day the apheresis was performed (for more detailed methods see Paglia et al.36) All units were checked for bacterial contamination on Day 10. The National Bioethics Committee of Iceland and the Icelandic Data Protection Authority approved the study.

Blood bank quality controls A blood gas analyzer (ABL90 FLEX, Radiometer Medical ApS, Brønshøj, Denmark) was used for determination of pH, pO2, and pCO2; concentration of total hemoglobin; and concentrations of K+, Na+, Cl–, glucose, and lactate in the PLT medium, immediately after sample collection. PLT concentration, mean PLT volume, plateletcrit, PLT distribution width, and white blood cell count were assayed using an automated hematology analyzer (CELL-DYN Ruby, Abbott Diagnostics, Abbott Park, IL). Flow cytometry (FACSCalibur, Becton Dickinson, La Jolla, CA) was used to determine expression of CD41, CD42b, CD62P, CD63, annexin V (all from Becton Dickinson), and JC1 to indicate the level of mitochondrial polarization (Life Technologies, Boston, MA). Lactate dehydrogenase (LDH) activity in the PC medium was assessed by an LDH assay kit (ab102526, Abcam, Cambridge, UK). The soluble CD40 ligand concentration was determined using a quantikine enzyme-linked immunosorbent assay (ELISA) kit (DCDL40, R&D Systems, Minneapolis, MN). Soluble P-selectin (sPselectin) concentration was determined via ELISA (BBE6, R&D Systems). Aliquots of 0.25 mL were used to determine the adenosine triphosphate (ATP) concentration employing a luminescent cell viability assay (CellTiterGlo, Promega, Madison, WI) using a microplate reader (Spectramax M3, Molecular Devices, Sunnyvale, CA). For more detailed methods refer to the Supplemental Data (available as supporting information in the online version of this paper).

METABOLIC PROPERTIES AND PROFILES OF STORED PLTs

Metabolomics Samples for metabolomic analysis were processed starting from 0.5 mL of buffy coat PC using the procedures previously described for the analysis of apheresis PLTs.36 Metabolomic analysis was performed using an ultra performance liquid chromatography (UPLC; Acquity, Manchester, UK) system, coupled with a quadrupole– time-of-flight mass spectrometer (Synapt G2, Waters). UPLC separation was achieved by hydrophilic interaction liquid chromatography using an amide column, 1.7 μm (2.1 × 150 mm) (Acquity, Waters).38,39 The UPLC-MS method previously described for the analysis of apheresis PLTs was employed in this work for the analysis of buffy coat PLTs.36

Data processing Unexpected metabolites were identified by integration, alignment, and conversion of MS data points into exact mass retention time pairs (MarkerLynx, v4.1, Waters). The identity of the unexpected metabolites was established by verifying peak retention time, accurate mass measurements, and tandem MS against our in-house database and/or online databases, including HMDB40 and METLIN.41 TargetLynx (v4.1, Waters) was used to integrate chromatograms of targeted metabolites. Ion chromatograms were extracted using a 0.02-mDa window centered on the expected m/z for each targeted compound. Each metabolite was normalized using a proper isotopically labeled internal standard and the ratio metabolites : internal standard was then used for deriving concentration using external calibration with reference standards, as described in the previous paper.36

Statistical analysis The data set employed consisted of the apheresis data published recently by our group36 and the new buffy coat data obtained in this study (Appendix S1, available as supporting information in the online version of this paper). Intracellular metabolomics data were normalized by cell number. Since apheresis PLT count increased during the first 4 days, probably due to a previously described artifact of PLT aggregation,42,43 an average of the cell count in the first 5 days was used to normalize intracellular apheresis metabolomics data. A further normalization was performed to minimize differences between two different metabolomics experiments (apheresis data set—Experiment 1, Units 1, 2, and 3; Experiment 2, Units 4, 5, and 6; buffy coat data set— Experiment 1, Units 1, 2, 3, and 4; Experiment 2, Units 5, 6, 7, and 8). Normalization was performed by dividing each variable of each experiment by the square root of the sum of the squares of all original values of that experiment. Concentrations of extracellular measurements were computed to obtain consumption and secretion rates

(ΔC/Δt) during the storage. External lactate theoretical secretion was derived by estimating stoichiometric conversion of the experimental external glucose consumed. Pathway analysis was performed using metaboanalyst,44 and it combines pathway enrichment and pathway topology analysis. Every variable having an HMDB ID was included in the analysis using the Homo sapiens KEGG library. The pathway enrichment analysis method was global test and the pathway topology analysis used relative betweenness centrality. The fold change (apheresis, normalized to Day 0; buffy coat, normalized to Day 1) was calculated for each variable monitored and one-way analysis of variance (ANOVA) test was used to find measurements presenting a significant change with the time of storage (p ≤ 0.05). Principal component analysis (PCA) (EZinfoUmetrics, http://www.umetrics.com/kb/about-ezinfo20) was performed on all significant metabolomics measurements (p < 0.05 one-way ANOVA test) and all blood bank quality controls. Before PCA, data were scaled (unit variance scaling).

RESULTS This study compared PLTs from two production methods routinely used in transfusion medicine with regard to changes in metabolic states and PSL during storage for 10 days. To achieve this goal we generated the most comprehensive data set available to describe metabolic properties of buffy coat PLTs and we compared it with the data set that we have recently published on apheresis PLTs (Appendix S1).36

Apheresis and buffy coat PLTs exhibit discrete metabolic phenotypes during storage We previously reported that apheresis PLTs experience, during storage, a metabolic decay process composed of discrete metabolic phenotypes.36 In this work, using PCA, we demonstrated that a similar metabolic shift could be identified in buffy coat–prepared PLTs (Fig. 1). We observed a metabolic transition occurring 4 days after the PLT production that pointed out two potential clinically relevant, metabolic phenotypes (short-term-stored PLTs and medium-term-stored PLTs; Fig. 1). A similar transition was also observed in apheresis-produced PLTs. The first principal component accounted for 41% of the total variance in the buffy coat data set and for 35% in the apheresis data set. In both cases, first principal component mainly describes how extracellular metabolic concentrations changed over time (Tables S3 and S4, available as supporting information in the online version of this paper), such as external lactate, hypoxanthine, arginine, and nicotinamide (Tables S3 and S4). The second principal component (15% of total variance in apheresis Volume **, ** **

TRANSFUSION

3

PAGLIA ET AL.

Fig. 1. PCA of data obtained from deep metabolomics study. (A) Apheresis PLTs; PCA was performed considering each single sample, the trajectories of each PLTs unit, and the average of 6 units for each day. (B) Buffy coat PLTs; PCA was performed considering each single sample, the trajectories of each PLTs unit, and the average of 8 units for each day. The metabolomics data (Appendix S1) was screened by using ANOVA test and normalized to minimize analytical variability coming from different experiments as described under Material and Methods. PC1 = first principal component; PC2 = second principal component.

PLTs and 10% in buffy coat PLTs) showed a certain variability, as detailed below, within the first metabolic phenotype (Tables S3 and S4). The individual trajectories of each single unit clearly showed that apheresis PLTs have a higher interbag variability compared to buffy coat PLTs. Moreover, in apheresis PLTs, PCA indicated a third metabolic phenotype (from Day 7 to Day 10) that was absent in buffy coat PLTs. Pathway analysis was then performed on the endoand exometabolome to isolate pathways involved in the transition phase occurring on Day 3 from short-termstored PLTs to medium-term-stored PLTs (Fig. S1, available as supporting information in the online version of 4

TRANSFUSION Volume **, ** **

this paper). This analysis identified glutathione metabolism, purine metabolism, glycolysis, tricarboxylic acid (TCA) cycle, pentose phosphate pathway, lipid metabolism, and nicotinamide metabolism as the main representative pathways of the metabolic transition from short-term-stored PLTs and medium-term-stored PLTs in both apheresis and buffy coat PLTs (Fig. S1). Based on these results, we selected significant metabolites (ANOVA test p ≤ 0.05) representative of the transition phase and we evaluated their time profiles (Fig. 2). The transition phase at Day 3 was evident in both the exo- and endometabolomic profiles as shown in the heat map (Fig. 2). In the exometabolome, the transition phase showed a significant compositional change of the storage solution

METABOLIC PROPERTIES AND PROFILES OF STORED PLTs

Fig. 2. Heat map of selected metabolites and quality control variables. The time profiles are presented as average of 6 units for apheresis and 8 units for buffy coat. The metabolites were selected based on pathway analysis and ANOVA test (p ≤ 0.05). Normalized signals as described under Material and Methods were used in these time profiles.

after 4 days from PLTs production. In fact, several metabolites such as lactate, hypoxanthine, malate, xanthine, and nicotinamide accumulate in the extracellular environment, and these extracellular changes coincided with the expression of surface activation markers (Fig. 2). In the endometabolome the transition phase showed how the intracellular pool of many metabolites was almost completely depleted in the second phenotype, suggesting that the metabolic decay initiated 4 days after the production (Fig. 2). Even though apheresis and buffy coat PLTs showed similar exo- and endometabolic profiles, some differences were observed according to the production methods. For instance, in the exometabolome the secretion and consumption rates were faster in apheresis PLTs, suggesting a higher metabolic activity in the first phase during apheresis, which was consistent with activation

marker profiles (Fig. 2). Also the endometabolic profiles showed differences in the first phase, especially related to metabolites involved in glycolysis and the TCA cycle (Fig. 2). These findings suggest higher metabolic activity in the first phase in apheresis PLTs, while in the second phase both endo- and exometabolic profiles appear to be very similar between the two production methods.

Apheresis triggers activation of PLTs Flow cytometry of selected activation markers revealed that apheresis-prepared PLTs were more activated at the beginning of storage compared to buffy coat–prepared PLTs, whose activation occurred more gradually over time (Fig. 3). We monitored six different surface activation markers and found, in all cases, higher expression in apheresis PLTs. Moreover, enhanced activation at the Volume **, ** **

TRANSFUSION

5

PAGLIA ET AL.

Fig. 3. Profiles of activation markers in apheresis and buffy coat PLTs. Apheresis, n = 6 units for apheresis and 8 units for buffy coat. sCD40L = soluble CD40 ligand.

beginning of storage resulted in a higher metabolic activity especially during the first phase (short-term-stored PLTs). This enhanced metabolic activity, which coincided with the activation in apheresis PLTs, affected both glycolysis and the TCA cycle (Fig. 2). Besides the metabolic consequences, increased activation during the first phase of apheresis PC storage was a potential source of a higher interbag variability, as is shown in PCA of all variables (Fig. 1) and in the PCA performed on the activation markers alone (Fig. 4). PCA of activation markers showed two main results: 1) apheresis and buffy coat activation markers both demonstrated the metabolic transition 4 days from production (Fig. 4) consistently with PCA of the whole metabolome (Fig. 1) and 2) PCA of buffy coat PLT activation markers has a similar characteristic profile for all units investigated (Fig. 4B). In contrast, in apheresis-produced PLTs, there was no overlap in the activation marker profiles of the single units, demonstrating again the higher variability of individual PLT units collected by apheresis (Fig. 4C). 6

TRANSFUSION Volume **, ** **

Differences in consumption and secretion rates Differences in glucose consumption and lactate secretion were detected especially at the beginning of the storage, consistent with the differences in the activation marker profiles (Figs. 3 and 5). In fact, in the first phase, apheresisactivated PLTs consumed more glucose (approx. 0.8 mmol/L/day) and secreted more lactate. Consequently, in the first 4 days after the production, apheresis PLTs consumed approximately 3 mmol/L glucose, while buffy coat PLTs needed almost twice the amount of time to consume the same amount of glucose (Fig. 5). Differences in glucose consumption rate were also reflected in the secretion rate of lactate, showing how the higher apheresis PLTs lactate secretion rate resulted in a different pH profile (Fig. 5). Data showed that buffy coat PLTs had a higher secretion rate of lactate from Day 7 to Day 10. This difference was due to the fact that apheresis PLTs consumed almost all glucose in the first 7 days of storage and its conversion in lactate dropped at the end of storage (Fig. 5). Finally, it was interesting to note that in

METABOLIC PROPERTIES AND PROFILES OF STORED PLTs

Fig. 4. PCA performed on the measured levels of the activation markers. (A) PCA performed on both apheresis and buffy coat PLTs. (B) PCA performed on apheresis PLTs. (C) PCA performed on buffy coat PLTs. In this analysis data reported in Fig. 3 and in Table S2 were used. PC1 = first principal component; PC2 = second principal component.

apheresis PLTs there is a stoichiometric conversion of glucose into lactate in the first phase, while during the rest of the storage, the glucose conversion into lactate does not seem to be complete, suggesting more activity of the TCA cycle as previously reported.36 This transition on Day 3 in the glucose–lactate conversion was not evident in buffy coat PLTs (Fig. 5).

Correlation between activation markers and external metabolites In the previous article we found a strong correlation between sP-selectin and the concentration of certain

external metabolites.36 In this study we expanded this finding and systematically correlated the expression of surface activation markers with external metabolites (Fig. S2, available as supporting information in the online version of this paper). Buffy coat PLT activation markers in general had a stronger correlation with external metabolites (Fig. S2), which again correlated stronger with each other (Supporting Information Fig. S3). This result was consistent with the difference in the interbag variability described in the PCA profiles of Fig. 4. In both apheresis and buffy coat PLTs, sP-selectin was the activation marker that had the highest correlation with external metabolites. Volume **, ** **

TRANSFUSION

7

PAGLIA ET AL.

Fig. 5. Profiles of external glucose, external lactate, and external pH during PLT storage. Lactate and glucose consumption and secretion rates were computed from the concentration profiles as ΔC/Δt (see Materials and Methods). Theoretical consumption and secretion refer to complete conversion of glucose to lactate. Apheresis n = 6; buffy coat n = 8. Bars represent standard deviations (SDs).

In particular, we found that that sP-selectin strongly correlated (r > 0.8) with four metabolites (hypoxanthine, malate, lactate, and glucose) in both production methods (Fig. S2). Their correlations are shown in Fig. 6.

Buffy coat PLTs exhibit a linear decay of mitochondrial functionality Buffy coat and apheresis PLTs performed nonoxidative glycolysis with conversion of glucose to lactate. The metabolic transition occurring 4 days after PLT production affected glycolysis and mitochondrial activity in apheresis PLTs but not in buffy coat PLTs (Figs. 5 and 7). Different TCA activity between buffy coat and apheresis PLTs was particularly evident by examining the profile of the mitochondrial activity (JC1) and of the expression of CD41+ and annexin V+ (marker of apoptosis). More8

TRANSFUSION Volume **, ** **

over, it was consistent with several profiles of extracellular and intracellular intermediates of TCA cycle (Fig. 7). With the exception of the different TCA cycle and glycolysis rates at the beginning of the storage, apheresis and buffy coat PLTs had a similar profile for other metabolites and pathways. Both extracellular and intracellular metabolites involved in the pentose phosphate pathway, purine metabolism, glutathione metabolism, and lipid metabolism had very similar profiles (Fig. 2, Figs. S4 and S5, available as supporting information in the online version of this paper).

DISCUSSION In Europe, apheresis PLTs and buffy coat PLTs are considered equivalent blood components for patient therapy.3

METABOLIC PROPERTIES AND PROFILES OF STORED PLTs

Fig. 6. sP-selectin correlates with external metabolites; r is the Pearson correlation coefficient (apheresis, blue; buffy coat, red).

Apheresis PLTs are obtained from a single donor and thus may provide the advantage of reducing the risk of HLA immunization once transfused, especially in multiply transfused patients. Different production methods may affect PLT metabolism during the storage resulting in different degrees of PSL and consequently to a noncomparable quality of the product.14-17 Considering the open discussion about potential differences in quality of PLTs produced with different methods,3,18,19 we performed a deep metabolomic study on buffy coat PLTs and then investigated the influence of production methods on

PLT metabolism during storage by comparing the two most comprehensive data sets available for stored apheresis and buffy coat PLTs.36 Indeed, in the preceding article we investigated the metabolism of apheresis PLTs during storage demonstrating that the metabolic decay occurring during storage was not a linear process, but instead we described discrete metabolic phenotypes during different stages of storage.36 Not surprisingly, we observed a similar metabolic transition after 4 days in the buffy coat–derived PLTs production as observed previously for apheresis-derived PLTs Volume **, ** **

TRANSFUSION

9

PAGLIA ET AL.

Fig. 7. Intracellular and extracellular profiles of intermediates of TCA cycle variables. Apheresis n = 6; buffy coat n = 8. Bars represent SDs.

(Fig. 1).36 Thus both production methods exhibited two distinct metabolic phenotypes during storage. During the first phase, the initial 4 days of storage, the PLTs displayed signs of active metabolism. During the second phase, however, the metabolic decay of PLT set in, with a metabolic signature characterized by accumulation of several compounds in the medium and the depletion of many intracellular intermediates (Figs. 1 and 2). These data may represent a metabolic footprint of the clinical 10

TRANSFUSION Volume **, ** **

findings that PLTs transfused before this metabolic transition may be of preferable quality compared to PLTs transfused later during storage.37 PCA of apheresis PLTs also described a second metabolic transition on Day 6, leading to a third metabolic phenotype (Fig. 1A). This third phenotype in apheresis PLTs was associated with the complete depletion of glucose and it may represent nonreversible metabolic changes leading to a faster metabolic decay (Figs. 1 and 5). Even if the third

METABOLIC PROPERTIES AND PROFILES OF STORED PLTs

metabolic phenotype has less, if any, clinical relevance, it highlights an important difference in the consumption and secretion rates. In fact, apheresis PLTs consumed glucose twice as fast as buffy coat PLTs (Fig. 5) in the first phase. The higher metabolic activity of apheresis PLTs at the beginning of storage is consistent with the different expression of activation markers (Fig. 3), confirming that the apheresis method triggered PLTs activation.14-17 The method and type of apheresis machine may play a pivotal role in this respect and should be addressed further. Activation of PLTs during storage plays an important role in the expression of the two metabolic phenotypes described. In fact, PCA of activation markers defined in a consistent way a transition point 4 days after the production with both production methods (Fig. 4). Since the development of two discrete metabolic phenotypes was strongly associated with the activation in PLTs during storage (Figs. 1 and 4), it was not surprising to observe a strong correlation between sP-selectin and some external metabolites (Fig. 6). In particular, we believe that the secreted malate and hypoxanthine found in the storage solution should be further investigated as potential biomarkers of PSL. The results presented here are representative of the level of activation of PLTs prepared by our specific method of apheresis and may not be representative of other apheresis devices. It has previously been reported that apheresis cell separators induce various degrees of PLT activation.15,45 Since the activation was intimately linked to the development of the PSL, and since it influenced the PLT metabolism in the first 4 days of storage, different degrees of activation between different apheresis PLTs units might lead to a nonhomogenous population of PCs that differ in quality. In fact, we demonstrated here that apheresis causes higher interbag variability (Figs. 1 and 4) and potentially clinical relevant differences in PLT quality between bags. These findings suggest that a careful standardization of the apheresis process is necessary to achieve PLT units of comparable quality, since different degrees of activation lead to different metabolism in the first phase and might accelerate the decay process, resulting in low quality of the transfusion product. Conversely, this may also represent donor dependent polymorphism as previously suggested.46 Different degrees of activation were associated with specific pathway differences in the first phase. In fact, apheresis-activated PLTs showed differences in glycolysis and the TCA cycle (Figs. 2, 5, and 7). We reported in the previous study that short-term-stored apheresis PLTs exhibited a down regulation of the TCA cycle, while medium-term-stored PLTs showed a moderate increase of the TCA cycle.36 In buffy coat PLTs the mitochondrial activity had a linear decay with time (Fig. 7). These new data suggest that the overall effect of TCA in defining this metabolic transition is not as strong as previously proposed36

and that a different degree of activation resulted in different mitochondrial activity. These new findings move some attention to other metabolic pathways, in particular, intermediates of the pentose phosphate pathway, glutathione metabolism, and some other metabolites involved in lipid metabolism that showed a strong transition after Day 3. In all cases, the intracellular pool of these intracellular metabolites was almost depleted (Figs. S4 and S5). The metabolic shift after 4 days of storage seems to be mainly related with changes in the energy metabolism. Changes in the energy metabolism are likely affecting PLT function, as demonstrated by the similar shift after 4 days observed for the activation markers. Collectively our data support a two-stage aggregation process47 whereby during the first 3 days of storage, PLTs undergo reversible activation that on Day 4, due to the augmented nature in which PLT activation occurs and increased receptor occupancy, the PLTs arrive at a point of no return and irreversible activation takes place. Here, we have specifically and quantitatively defined metabolic alterations associated with PSL; nevertheless, further studies are necessary to fully understand the implication of the metabolic shift on PLT function. Systems biology approaches, where metabolomics data are integrated in metabolic network reconstructions, might help to elucidate the metabolic properties of stored PLTs and characterize the specific metabolic phenotypes of short- and medium-term-stored PLTs produced with both apheresis and buffy coat methods.28,48 Indeed, a detailed knowledge of PLTs’ metabolism during storage is likely to advance our understanding of important aspects associated with storage lesions in PLTs and give rise to strategies and methods to improve patient care. Consideration of these may open ways to improve storage of PLTs with addition of new substances to PLT ASs and/or depletion of some critical metabolites appearing during storage. ACKNOWLEDGMENTS The authors thank Soley Valgeirsdottir, Kristine Wichuk, Manuela Magnusdottir, and staff at the blood bank for technical support. The authors would especially thank Sirus Palsson, Sisse Ostrowski, Marc Abrams, and Hans Gulliksson for their input. GP designed and performed the experiments, analyzed the data, and wrote the paper; OES designed and performed the experiments and wrote the paper; OR analyzed the data and wrote the paper; MBH designed experiments and wrote the paper; SB designed the experiments; SG designed the experiments and wrote the paper; and BOP designed the experiments and wrote the paper.

CONFLICT OF INTEREST The authors have disclosed no conflicts of interest. Volume **, ** **

TRANSFUSION

11

PAGLIA ET AL.

REFERENCES 1. Michelson AD. Platelets. San Diego (CA): Academic Press; 2002. 2. Goodnough LT, Shander A, Brecher ME. Transfusion medicine: looking to the future. Lancet 2003;361:161-9. 3. Vassallo RR, Murphy S. A critical comparison of platelet preparation methods. Curr Opin Hematol 2006;13:323-30. 4. Eriksson L, Hogman CF. Platelet concentrates in an additive solution prepared from pooled buffy coats. 1. In vitro

19. Schrezenmeier H, Seifried E. Buffy-coat-derived pooled platelet concentrates and apheresis platelet concentrates: which product type should be preferred? Vox Sang 2010;99: 1-15. 20. Gulliksson H. Defining the optimal storage conditions for the long-term storage of platelets. Transfus Med Rev 2003; 17:209-15. 21. de Wildt-Eggen J, Nauta S, Schrijver JG, et al. Reactions and platelet increments after transfusion of platelet concentrates in plasma or an additive solution: a prospective,

studies. Vox Sang 1990;59:140-5. 5. Murphy S, Heaton WA, Rebulla P. Platelet production in

randomized study. Transfusion 2000;40:398-403. 22. Gulliksson H. Storage of platelets in additive solutions: the

the Old World—and the New. Transfusion 1996;36:751-4.

effect of citrate and acetate in in vitro studies. Transfusion

6. Hogman CF. Storage of blood components. Curr Opin Hematol 1999;6:427-31. 7. Vasconcelos E, Figueiredo AC, Seghatchian J. Quality of platelet concentrates derived by platelet rich plasma, buffy

1993;33:301-3. 23. Gulliksson H, Eriksson L, Hogman CF, et al. Buffy-coatderived platelet concentrates prepared from half-strength citrate CPD and CPD whole-blood units. Comparison

coat and apheresis. Transfus Apher Sci 2003;29:13-6. 8. Dijkstra-Tiekstra MJ, Pietersz RN, Hendriks EC, et al. In vivo PLT increments after transfusions of WBC-reduced

between three additive solutions: in vitro studies. Vox Sang 1995;68:152-9. 24. Gulliksson H, Larsson S, Kumlien G, et al. Storage of plate-

PLT concentrates stored for up to 7 days. Transfusion 2004; 44:330-6. 9. Hess JR. Red cell changes during storage. Transfus Apher

lets in additive solutions: effects of phosphate. Vox Sang 2000;78:176-84. 25. Gulliksson H, AuBuchon JP, Vesterinen M, et al. Storage of

Sci 2010;43:51-9. 10. Bashir S, Cookson P, Wiltshire M, et al. Pathogen inactivation of platelets using ultraviolet C light: effect on in vitro function and recovery and survival of platelets. Transfusion 2013;53:990-1000. 11. Shrivastava M. The platelet storage lesion. Transfus Apher Sci 2009;41:105-13. 12. Kocazeybek B, Arabaci U, Akdur H, et al. Prospective evaluation of platelets prepared by single and random methods during 5 days of storage: aspects related to quality and quantity. Transfus Apher Sci 2002;26:29-34. 13. Akay OM, Gunduz E, Basyigit H, et al. Platelet function testing during 5-day storage of single and random donor plateletpheresis. Transfus Apher Sci 2007;36:285-9. 14. Sloand EM, Yu M, Klein HG. Comparison of random-donor platelet concentrates prepared from whole blood units and platelets prepared from single-donor apheresis collections. Transfusion 1996;36:955-9. 15. Krailadsiri P, Seghatchian J. Are all leucodepleted platelet concentrates equivalent? Comparison of Cobe LRS Turbo, Haemonetics MCS+ LD, and filtered pooled buffy-coatderived platelets. Vox Sang 2000;78:171-5. 16. Bock M, Rahrig S, Kunz D, et al. Platelet concentrates derived from buffy coat and apheresis: biochemical and functional differences. Transfus Med 2002;12:317-24. 17. Albanyan AM, Harrison P, Murphy MF. Markers of platelet activation and apoptosis during storage of apheresis- and buffy coat-derived platelet concentrates for 7 days. Transfusion 2009;49:108-17. 18. van der Meer PF. Platelet concentrates, from whole blood or collected by apheresis? Transfus Apher Sci 2013;48:12931.

12

TRANSFUSION Volume **, ** **

platelets in additive solutions: a pilot in vitro study of the effects of potassium and magnesium. Vox Sang 2002;82: 131-6. 26. Saunders C, Rowe G, Wilkins K, et al. Impact of glucose and acetate on the characteristics of the platelet storage lesion in platelets suspended in additive solutions with minimal plasma. Vox Sang 2013;105:1-10. 27. Thiele T, Iuga C, Janetzky S, et al. Early storage lesions in apheresis platelets are induced by the activation of the integrin alphaIIbbeta(3) and focal adhesion signaling pathways. J Proteomics 2012;76 Spec No.:297-315. 28. Paglia G, Palsson BO, Sigurjonsson OE. Systems biology of stored blood cells: can it help to extend the expiration date? J Proteomics 2012;76 Spec No.:163-7. 29. Gevi F, D’Alessandro A, Rinalducci S, et al. Alterations of red blood cell metabolome during cold liquid storage of erythrocyte concentrates in CPD-SAGM. J Proteomics 2012;76 Spec No.:168-80. 30. Pienimaeki-Roemer A, Ruebsaamen K, Boettcher A, et al. Stored platelets alter glycerophospholipid and sphingolipid species, which are differentially transferred to newly released extracellular vesicles. Transfusion 2013;53: 612-26. 31. Antonelou MH, Tzounakas VL, Velentzas AD, et al. Effects of pre-storage leukoreduction on stored red blood cells signaling: a time-course evaluation from shape to proteome. J Proteomics 2012;76 Spec No.:220-38. 32. Parguina AF, Garcia A. Platelet proteomics in transfusion medicine: a reality with a challenging but promising future. Blood Transfus 2012;10(Suppl 2):s113-4. 33. Zolla L, D’Alessandro A. Shaking hands with the future through omics application in transfusion medicine and

METABOLIC PROPERTIES AND PROFILES OF STORED PLTs

clinical biochemistry. Blood Transfus 2012;10(Suppl 2): s1-3. 34. Schubert P, Devine DV. De novo protein synthesis in mature platelets: a consideration for transfusion medicine. Vox Sang 2010;99:112-22. 35. D’Alessandro A, Gevi F, Zolla L. Red blood cell metabolism under prolonged anaerobic storage. Mol Biosyst 2013;9: 1196-209. 36. Paglia G, Sigurjónsson OE, Rolfsson O, et al. Comprehensive metabolomic study of platelets reveals the expression of discrete metabolic phenotypes during storage. Transfusion 2014. doi: 10.1111/trf.12710. 37. Heuft HG, Goudeva L, Krauter J, et al. Effects of platelet concentrate storage time reduction in patients after blood stem cell transplantation. Vox Sang 2013;105:18-27. 38. Paglia G, Hrafnsdottir S, Magnusdottir M, et al. Monitoring metabolites consumption and secretion in cultured cells using ultra-performance liquid chromatography quadrupole-time of flight mass spectrometry (UPLC-QToF-MS). Anal Bioanal Chem 2012;402:1183-98. 39. Paglia G, Magnusdottir M, Thorlacius S, et al. Intracellular metabolite profiling of platelets: evaluation of extraction processes and chromatographic strategies. J Chromatogr B Analyt Technol Biomed Life Sci 2012;898:111-20. 40. Wishart DS, Jewison T, Guo AC, et al. HMDB 3.0—the human metabolome database in 2013. Nucleic Acids Res 2013;41(Database issue):D801-7. 41. Smith CA, O’Maille G, Want EJ, et al. METLIN: a metabolite mass spectral database. Ther Drug Monit 2005;27: 747-51. 42. Macher S, Sipurzynski-Budrass S, Rosskopf K, et al. Function and activation state of platelets in vitro depend on apheresis modality. Vox Sang 2010;99:332-40. 43. Hagberg IA, Akkok CA, Lyberg T, et al. Apheresis-induced platelet activation: comparison of three types of cell separators. Transfusion 2000;40:182-92. 44. Xia J, Mandal R, Sinelnikov IV, et al. MetaboAnalyst 2.0—a

45.

46.

47.

48.

comprehensive server for metabolomic data analysis. Nucleic Acids Res 2012;40(Web Server issue):W127-33. Holme S, Sweeney JD, Sawyer S, et al. The expression of p-selectin during collection, processing, and storage of platelet concentrates: relationship to loss of in vivo viability. Transfusion 1997;37:12-7. Schubert P, Devine DV. Proteomics meets blood banking: identification of protein targets for the improvement of platelet quality. J Proteomics 2010;73:436-44. Maxwell MJ, Westein E, Nesbitt WS, et al. Identification of a 2-stage platelet aggregation process mediating sheardependent thrombus formation. Blood 2007;109:566-76. Thomas A, Rahmanian S, Bordbar A, et al. Network reconstruction of platelet metabolism identifies metabolic signature for aspirin resistance. Sci Rep 2014;4:3925.

SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s Web site: Fig. S1. Pathway analysis was performed to highlight pathways involved in the transition phase. a) Pathway analysis of the endo-metabolome. b) Pathway analysis of the exo-metabolome. Apheresis Pathway Analysis: samples from Day 0, 1 and 3 were considered as Phase 1 (short-term stored PLTs) and samples from Day 4, 5, and 6 were considered as Phase 2 (medium-term stored PLTs). Buffy Coat Pathway Analysis: samples from Day 1 and 3 were considered as Phase 1 (short-term stored PLTs) and samples from Day 4 and 5 were considered as Phase 2 (medium-term stored PLTs). Fig. S2. Systematic correlation of activation markers versus external metabolites. Pearson correlation coefficients (r) obtained from apheresis and buffy were plotted against each other. Fig. S3. Correlation of activation markers in apheresis and buffy coat PLTs. r = Pearson correlation coefficient. (Apheresis: Blue; Buffy Coat: Red). Fig. S4. Intracellular profiles of selected metabolites associated with the metabolic transition after 4 days from the PLT production. Signals were normalized as described in material and methods. Fig. S5. Extracellular profiles of selected metabolites associated with the metabolic transition after 4 days from the PLT production. Signals were normalized as described in Material and Methods. Table S1. Additional in vitro blood bank quality control parameters in PC samples stored for a period of 10 days. Results are presented as mean ± standard deviation. Unless otherwise noted, n = 6 for the apheresis experiment and n = 8 for the buffy coat experiment. Table S2. Additional markers of PLT activation and metabolism in PCs stored for a period of 10 days. Results are presented as mean ± standard deviation. n = 6 for the apheresis experiment and n = 8 for the buffy coat experiment. Table S3. Parameter contribution in principal component analysis of AP PLTs Table S4. Parameter contribution in principal component analysis of BC PLTs Table S5. Activation markers’ contribution in principal component analysis Appendix S1. Data set used in this work.

Volume **, ** ** TRANSFUSION

13

Metabolomic analysis of platelets during storage: a comparison between apheresis- and buffy coat-derived platelet concentrates.

Platelet concentrates (PCs) can be prepared using three methods: platelet (PLT)-rich plasma, apheresis, and buffy coat. The aim of this study was to o...
2MB Sizes 0 Downloads 7 Views