Jacobs Journal of Hematology

The Effect of Storage on Various Haematological and Biochemical Parameters of NHSBT Donor Blood

*Peter Ella-Tongwiis
Department Of Biological Sciences, University Of Chester, United Kingdom

*Corresponding Author:
Peter Ella-Tongwiis
Department Of Biological Sciences, University Of Chester, United Kingdom
Email:p.ellatongwiis@chester.ac.uk

Published on: 2018-09-10

Abstract

Background: Recent studies have reported that blood stored for longer periods (>14days) are associated with the development of post-transfusion complications. Most changes taking place in stored blood are biochemical, biomechanical and haematological and are known as storage lesion. The aim of this study was to investigate the effects of storage on the various haematological and biochemical parameters of National Health Service Blood and Transplant (NHSBT), whole donor blood.
Methods: 5 units of whole blood suspended in SAGM additive solution was acquired from the NHSBT, United Kingdom (UK). Units were stored at standard blood bank conditions (2-6oC) and analysed on days 1, 7, 14, 21 and 28. Full blood count was performed using Coulter® MicoDiff18 (Beckman Coulter, UK) analyser. Other biological parameters including Prothrombin time (PT), von willibrand factor (vWF), Potassium (K+), Sodium (Na), glucose, ferritin, pH and lactate were measured via various methods, such as the Randox RX Monza biochemistry analyser and ELISA. Results: Our study demonstrates that noticeable changes occur during storage of donor blood. Specifically, red cell membrane damage was observed by increased accumulation of plasma haemoglobin (Hb) (p=0.014) and increases in K+ (p=0.001) and Na+ (p-0.070), whilst glucose (p=0.001) levels decreased at day 28 storage. Significantly higher lactate levels (p=0.002) resulted in a fall in pH especially after day 21. With regards to blood coagulation, PT levels significantly increased during storage, indicating a reduced clotting ability (p=0.001).
Conclusion: This study demonstrates that changes to several haematological and biochemical parameters occur during the storage of blood, and subsequently may cause untoward risks to patients. These parameters, however, need further work employing a larger study to validate reliability in a clinical setting.

Keywords

Biochemical Parameters; NHSBT; Hematology; Haemolysis; Blood transfusion

Introduction

Globally, almost 105 million blood donations are made annually , while according to the National Health Service Blood and Transplant (NHSBT), about 2.1 million blood donations are taken every year in the United Kingdom. In the United States, somebody requires blood transfusion every two seconds. Recent findings from both in vitro and in vivo studies, have raised some concerns about the safety of the transfusion process. Blood stored for more than 30 days was associated with increased death rate in the elderly, severely sick and cardiac surgery patients, have demonstrated an association between duration of storage and high risk of pneumonia in patients undergoing cardiac surgery.

Recently, Mukherjee et al. [5] proved that stored whole blood less than seven days can be considered as fresh. The observed in vivo negative impacts of “old blood” transfusion have been linked with various biochemical, biomechanical and haematological changes occurring during storage, collectively known as “storage lesion” [6,7]. Most red cell indices are altered during storage mostly due to mechanical changes to the cell membrane. Chaudhary [8], have demonstrated that an increased plasma haemoglobin (Hb) was caused by a higher rate of haemolysis during storage (p<0.001). This increased rate of haemolysis however did not decrease the total RBC count [9]. Jobes et al. [10] reported significant reductions in glucose and ATP levels in blood stored in CPDA-1 over 35 days, with lowest levels being recorded in the last week of storage (28-35 days). Recently, Saunders, Rowe, Wilkins, & Collins [11], have also reported a significant association between glucose/ATP depletion and platelet dysfunction. Prothrombin time (PT) is a clotting profile used in conjunction with other factors to estimate extrinsic pathway of coagulation. During storage, significant changes in various clotting factors including prothrombin time have been reported [12]. von Willibrand Factor (vWF) is a large protein molecule about 20,000kD in size, typically made up of around 200kD monomers and is produced by platelets, endothelial cells and megakaryocytes. Weiss et al. [13] observed an increase in vWF levels (2% per day) during storage of blood. Changes in vWF levels have also been associated with risks of bleeding and thrombotic complications [14]. The effect of blood storage on coagulation has primarily been studied in patients. Patients, especially children, who receive frequent transfusions, have been associated with increased bleeding tendencies and complications during surgery [15]. In a recent study, no significant difference was observed in the pulmonary function or coagulation status of patients who received fresh or old blood [16].

The sodium/potassium pump (Na-K-ATPase) maintains the electrolyte balance between the red cell membrane and the surrounding plasma. It achieves this by pumping out sodium (Na) generated within the cell while keeping potassium (K) in the cell [17]. A progressive decline in sodium and a rise in potassium levels have been reported in donor blood throughout storage duration in a study by Jobes et al [10]. The association between pH and lactate levels have previously been reported by Uvizl, Klementa, Adamus & Neiser [18]. They observed high lactate concentrations and low pH levels in patients who were transfused old blood. This is in agreement with findings by Jobes et al. [10] who reported similar trends. With millions of stored donor blood being used across the globe on a daily basis, the overall safety of stored blood and blood transfusion should be vigorously examined. In that light, storage lesion and the complications associated with old blood should be properly understood. The main aim of this study was to investigate the effects of storage on various biochemical and haematological parameters of NHSBT donor blood. It is anticipated that any significant changes to the parameters investigated, may provide valuable information, where current NHSBT guidelines for the storage and clinical uses of blood components may need revising and updating.

Methods

Approval for this study was granted by the research ethics committee of the Faculty of Applied Sciences of the University of Chester (ref. no. 765/13/PE/BS). All laboratory estimations and analyses were undertaken at the Thomas building within the University of Chester and conformed to the rules and conditions set out by the research ethics committee.

Nhsbt Donor Blood

Whole Donor blood (n=5) was acquired from the National Health Service Blood and transplant (NHSBT) at Speke, Liverpool UK. All units were whole blood collected in SAGM additive solution and were “fresh” since they were less than 7 days old [5]. Units were stored at 2-60C. Day of arrival was recorded as baseline (day 1 of storage).

Blood Sampling

On days 1, 7, 14, 21 and 28 all five units were individually mixed by gently rotating the donor blood bag. Haemolysis was prevented by avoiding vigorous shaking. Samples were then drawn from the units into 7ml tubes prior to further laboratory testing.

Full Blood Count

Full Blood Counts (FBC) were performed at days 1, 7, 14, 21 and 28 using a Coulter® MicoDiff18 (Beckman Coulter, U.K.) analyser. RBC, WBC and platelet parameters were assessed using the fully automated system. 

Prothrombin Time, Sodium, Potassium And Lactate Estimation

Prothrombin time, sodium, potassium and lactate were measured at days 1, 7, 14, 21 and 28 using the Randox RX Monza semi-automated chemistry analyser (Randox laboratories), which employed an enzymatic colorimetric test. Following manufacturers recommendations, potassium was always assayed before sodium to avoid interference.

Measurement of von Willebrand Factor (Vwf)

Plasma vWF concentration was measured as described previously by a sandwich ELISA technique using rabbit anti-human vWF and rabbit anti-human vWF peroxidase conjugate (Dako, UK), [19,20].

pH Estimation And Glucose Estimation

pH measurements were taken using a pocket-sized pH meter (pHep®) on each sampling day. Glucose was also measured using a blood glucose meter, aviva nano (ACCU-CHEK®), with a system measurement requirement range of 0.6-33.3 mmol/L.

Ferritin

The miniVIDAS automated analyser (BioMerieux), was used to estimate ferritin throughout the study. The analyser employes a one-step enzyme immunoassay sandwich ELISA with a final fluorescent detection.

Figure1(A-D). The Effect of Storage on Various Red Cell Indices of NHSBT Whole Blood Donor Samples. (A) Hb, p= 0.014, Friedman; (B) MCV, p< 0.001, ANOVA; (C) RBC, p=0.526, Friedman; (D) HCT, p< 0.001, ANOVA ( = P< 0.05 compared to baseline), (n=5).

Statistics Analysis

Data analysis was performed using SPSS software (version 21). All results were presented as mean (± standard error (SEM)) or median (± interquartile (Iqr). Where data were normally distributed, repeated measures one way-anova analysis of variance (ANOVA) between samples test was employed. Post hoc testing was undertaken employing the tukey test for pairwise comparison of the means. Friedman test was used to analyse data that were not normally distributed. Significant statistical differences from the Friedman test were further analysed using the Wilcoxon test. p≤0.05 was accepted as statistical significance.

Results

Effect of Storage on Haematological Parameters

Full Blood Count

During the 28-day storage, significant changes were observed in red blood cell indices (Figures 1A-D). There was a significant change in Hb following 28 days of storage (p = 0.014, as determined by Friedman test), (Figure 1A). There were significant changes in mean cell volume (MCV) during the storage period (p < 0.001, ANOVA) (figure 1B). Specifically, significant increases between day 1 storage versus days 7, 14, 21 and 28 (p< 0.05) were reported (Figure 1B). Similarly, the haematocrit (Hct), an indicator of plasma cell volume, demonstrated significant increases during storage (p < 0.001, Friedman) (Figure 1D).

Compared with day 1 storage, significant increases were observed versus days 7 (p= 0.028), 14 (p = 0.04), 21 (p = 0.008) and 28 (p= 0.030) (figure 1D). However, no significant changes were reported for red blood cell (RBC) levels during storage (p = 0.526, Friedman) (figure 1C). Total white cell count (WBC) count significantly decreased from day 1 to day 28 (p = 0.026). Specifically, significant variations between day 1 and day 21 (p = 0.042) and day 28 (p = 0.041) were observed (figure 2A).

Figure 2. The Effect Of Storage On Total WBC And Platelet Counts Of NHSBT Whole Blood Donor Samples. (A) WBC count (p= 0.026 as determined by Friedman test). (B) Platelet count (p= 0.040 as determined by ANOVA) ( , p<0.05 compared to baseline), (n=5).

With respect to platelet concentration (Figure 2b), a significant decrease was observed during storage (p= 0.040 as determined by ANOVA). Specifically, platelet concentration decreased at day 7 (146.8 ± 28.10) and day 21 (157.8 ± 40.68). Upon further statistical testing using pairwise comparisons, a significant decrease was observed at day 7 storage versus baseline values (p= 0.048).

Prothrombin time

There was a significant increase in prothrombin time (p = 0.001, Friedman test) during storage.Further analysis employing the Wilcoxon test demonstrated significant differences between day 1 storage versus days 7, 14, 21 and day 28 (p<0.005 as indicated in Figure 3). 

Figure 3. The Effect of Storage of NHSBT Whole Donor Blood on Prothrombin Time (PT). p = 0.001 as determined by Friedman test, ( * p< 0.05 compared to baseline), (n=5).

von Willebrand Factor (vWF)

vWF is a large protein, acting as a carrier for factor VIII during haemostasis. No significant changes were observed in vWF levels following 28 days storage (P= 0.062, as determined by ANOVA) (Figure 4).

Figure 4. The Effect of Storage of NHSBT Whole Donor Blood on Vwf. p= 0.062 as determined by ANOVA test, n=5.

Figure 5A-B. The Effect of Storage of NHSBT Whole Donor Blood on Potassium and Sodium Concentration. p = 0.001 for potassium, using the Friedman test and p = 0.070 for sodium using ANOVA, (* p< 0.05 compared to baseline), (n=5).

Effect of Storage on Biochemical Parameters

Potassium and sodium

Results from this study showed an increase in K+ during storage (p = 0.001, determined by Friedman test) (Figure 5A). Upon further analysis, the Wilcoxon test demonstrated significant increases between day 1 and day 7 (p = 0.043), day 14 (p = 0.043), day 21 (p = 0.043) and day 28 (p = 0.043) (figure 5A). Although increasing sodium concentrations were observed up to day 21, these changes were statistically not significant (p = 0.070 as determined by ANOVA) (Figure 5B).

Glucose

Glucose is the main energy source of stored RBCs, the depletion of which increases the rate of haemolysis and lowers cell survivability. Evaluation of plasma glucose levels showed significant depletions starting from day 1 (p = 0.001, ANOVA) (figure 6A). Upon further analysis significant decreases were observed between storage day 1 (baseline) versus day 7 (p = 0.001), day 14 p = 0.006), day 21 (p = 0.001) and day 28 (p = 0.005), (figure 6A). 

pH, lactate and ferritin

pH levels in blood showed significant reductions during storage (p = 0.001, ANOVA) (Figure 6B). Upon further testing, significant changes were observed between day 1 storage versus day 7 (p = 0.004), day 14 (p = 0.004), day 21 (p = 0.015) and day 28 (p = 0.002) (figure 6B). Evidently, there was a shift in pH from physiologically neutral to acidic. Analysis of lactate results demonstrated a significant accumulation of lactate in the blood bag (p = 0.002, determined by Friedman test) (figure 6C). Further pairwise analyses using the Wilcoxon test showed significant increases from on day 7 (p = 0.043), 14 (p = 0.043), 21 (p = 0.043) and 28 (p = 0.043) (figure 6C), suggesting a rise in H- ions with the potential of reducing pH. Ferritin, a primary form of iron storage, is essential in regulating the amount of iron in circulation. With regards to ferritin levels in stored red cells, 

Figure 6A-B. The Effect of Storage on the Concentrations of Various Biochemical Parameters of NHSBT Whole Blood Donor Samples. (A) Glucose levels (p= 0.001 as determined by ANOVA). (B) pH (p= 0.001 as determined by ANOVA) (C) Lactate concentrations (p= 0.002, determined by Friedman) (D) ferritin concentrations (p= 0.009, determined by Friedman), (* p< 0.05 compared to baseline), (n=5).

Discussion

The main aim of this study was to investigate the effects of storage on various haematological and biochemical parameters of NHSBT whole donor blood. Recent studies have reported that blood stored for longer periods (>14days) are associated with the development of post-transfusion complications. This has re-ignited the debate about the potentially harmful changes occurring in stored blood. It has also highlighted the need to further study storage and processing guidelines, with the hope of making blood transfusions safer and reliable. Results from this study show evidence of many potentially harmful changes that occur in blood during storage. There was a significant decrease in glucose and pH while K, lactate, PT, ferritin and some red cell indices show a significant trend of increasing concentrations. Most of these changes are among a group of biochemical, biomechanical and haematological changes taking place in stored blood known as storage lesion.

Changes to Hb concentrations during the present study agree with a study by Chaudhary & Katharia [8], who reported significantly increased plasma Hb caused by a higher haemolysis rate. The increased Hb levels observed in this study correspond with falls in glucose and pH, which may be the reason for the high haemolysis rate. High levels of free heme have been associated with inducing post-transfusion complications, especially with high dependency and immunocompromised there was a significant change in concentration during storage (p= 0.009, determined by Friedman). Compared to basal values, subsequent pairwise examination with the Wilcoxon test demonstrated significant increases on days 7 (p= 0.043), 14 (p= 0.043) and a slight fall on day 21 (20.26 ± 13.24) albeit not significantly. Values then significantly increased on day 28 (p= 0.043). 

patients [21]. Transfusion of blood with high levels of Hb especially those stored between 7 and 21 days could increase the patient’s risk of developing hyperhaemoglobinaemia [21]. Insignificant changes were observed following 28 days storage in RBC counts during this study. However, the patterns reported in this study were similar to observations by Sawant et al. [22] and Urbina & Palomino, [9] who reported slight increases in RBC count in various storage conditions. The effect of storage on platelet structure and function has previously been reported by Saunders et al [11], where they demonstrated a relationship between glucose/ATP depletion and decreased platelet number and dysfunction. Our study supports findings from Saunders et al [11], although we did not specifically investigate platelet function during the storage period.

Decreasing WBC counts observed during the present study, are in agreement with findings by Jobes et al. [10], who investigated the characterization of coagulation properties in refrigerated whole blood for transfusion in an in-vitro study. Glucose is the main energy source of stored blood, and is responsible for maintaining cell viability. Glucose depletion in stored blood, arises mostly due to a shutdown of the glycolytic pathway [23]. A significant decrease in glucose levels were noted during the study and are in agreement with similar findings reported by Jobes et al. [10]. Different additive solutions and anticoagulants possess different pH ranges and may eventually affect the pH of the blood being transfused [24]. The influence of lactate in reducing pH has previously been described [25]. The pH levels in the present study increased significantly with storage, and are in agreement with others [24,25], who have investigated stored donor blood and various additive solutions. The observed decline in pH during the present study compliments findings of increasing lactate levels with storage. Increased potassium concentrations in stored whole blood have been associated with the breakdown of the Na-K-ATPase pathway [26]. During our study, significant elevations in Potassium levels were observed and agree with findings from previous studies. 

Although no significant changes were seen in Sodium during the present study, trends of increasing Sodium levels were recorded. Findings here differ from a similar study undertaken by Jobes et al. [10], where they reported on decreasing levels of sodium during the storage of blood. A possible explanation for the differences between the studies is that Jobes et al. [10] measured intracellular sodium rather than plasma levels. It can therefore be appreciated that biochemically; Sodium can diffuse across the red cell membrane into plasma during storage, accounting for the decline in intracellular sodium levels whilst increasing plasma (extracellular) Sodium levels, as noted in the present study. Another likely cause for the increased Sodium levels noted during the present study could be due to the effect of the additive solution, SAGM. Transferrin and ferritin, play an important role in maintaining iron homeostasis. As the main storage form of iron, ferritin can be used to estimate donor eligibility and also enhance erythropoiesis for the patient requiring a blood transfusion [27]. In the present study, a significant increase in ferritin levels was observed following 28 days storage of whole blood. It has previously been reported that high iron and transferrin levels have been observed in patients transfused with old stored blood [27].

Our findings, therefore compliment this observation. The effect of storage on coagulation (prothrombin time) has been the focus of much research attention. Prothrombin times in the present study significantly increased with storage, indicating a compromise in the clotting ability of the stored blood. vWF, which is a carrier for factor VIII during the coagulation cascade, plays important roles in coagulation, and levels reported in the present study increased over the storage time. Both the PT and vWF results from our study highlight the potential reduced haemostatic state and possible increased bleeding tendency in patients transfused with old stored blood units. These findings are in agreement with previous studies undertaken by Lazzari et al. [14] and Weiss et al. [13]. This study, however, is not without its limitations. It is noted that only a small number of whole blood units (n= 5) were recruited for the purpose of the study. Additionally, studies focussing on specific aspects of the blood (e.g. biochemical markers) could provide a more focused approach.

Conclusion

The aim of this study was to investigate the effects of storage on the various haematological and biochemical parameters of NHSBT donor blood. Significant changes in several haematological and biochemical parameters were observed, highlighting the effect that storage does have on blood and blood components scheduled for transfusion. Clearly, further work is needed to fully understand an area of clinical significance. This study, however, needs to be expanded employing a larger sample size to fully validate reliability in a clinical setting.

Acknowledgments

The authors would like to acknowledge the Department of Biological Sciences, University of Chester for their financial support.

Competing Interests

The authors have declared that no competing interests exist.

References

1. World Health Organisation, 2013. Global Database on Blood Safety. Summary Report, 2011.

2. American Red Cross. (2013). Blood Facts and Statistics. Assessed on 15th September, 2013

3. Edgren Gustaf, Kamper-Jørgensen Mads, Eloranta Sandra, Rostgaard Klaus, Custer Brian et al. Duration of red blood cell storage and survival of transfused patients (CME). Transfusion (2010), 50(6): 1185-1195.

4. Koch CG, Li L, Sessler DI, Figueroa P, Hoeltge GA et al. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med. (2008), 358(12): 1229-1239.

5. Mukherjee Somnath, Marwaha Neelam, Prasad Rajendra, Sharma Ratti Ram, Thakral Beenu. Serial assessment of biochemical parameters of red cell preparations to evaluate safety for neonatal transfusions. The Indian journal of medical research. (2010), 132(6): 715-720.

6. Donadee Chenell, Raat Nicolaas JH, Kanias Tamir, Tejero Jesús, Lee Janet S et al. Nitric oxide scavenging by red blood cell microparticles and cell-free hemoglobin as a mechanism for the red cell storage lesion. Circulation, (2011), 124(4): 465- 476.

7. Kim-Shapiro DB, Lee J, Gladwin MT. Storage lesion: role of red blood cell breakdown. Transfusion. (2011), 51(4): 844- 851.

8. Chaudhary Rajendra, Katharia Rahul. Oxidative injury as contributory factor for red cells storage lesion during twenty eight days of storage. Blood Transfusion. (2012), 10(1): 59-62.

9. Urbina A, Palomino F. Reticulocyte count in red-blood-cell units stored in AS-1. Vox sanguinis, (2013), 104(4): 331-336.

10. Jobes David, Wolfe Yanika, O’Neill Daniel, Calder Jennifer, Jones Lisa et al. Toward a definition of “fresh” whole blood: an in vitro characterization of coagulation properties in refrigerated whole blood for transfusion. Transfusion. (2011), 51(1): 43-51.

11. Saunders C, Rowe G, Wilkins K, Collins P. 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): 1-10.

12. Zhao Y, Lv G. Influence of temperature and storage duration on measurement of activated partial thromboplastin time, D-dimers, fibrinogen, prothrombin time and thrombin time, in citrate-anticoagulated whole blood specimens. International Journal of Laboratory Hematology, (2013), 35(5): 566-570.

13. Weiss DominikR., Franke D, Strasser ErwinF, Ringwald Juergen, Zimmermann Robert et al. von Willebrand factor, clotting factors, and clotting inhibitors in apheresis platelet concentrates. Transfusion, 2014, 54(3): 633-639.

14. Lazzari MA, Sanchez-Luceros A, Woods AI, Alberto MF, Meschengieser SS. Von Willebrand factor (VWF) as a risk factor for bleeding and thrombosis. Hematology. 2012, 17(1): 150-152.

15. Manno CatherineS, Hedberg KathleenW, Kim HaewonC, Bunin GretaR, Nicolson Susan et al. Comparison of the hemostatic effects of fresh whole blood, stored whole blood, and components after open heart surgery in children. Blood. 1991, 77(5): 930-936.

16. Kor DarylJ, Kashyap Rahul, Weiskopf RichardB, Wilson GregoryA, van Buskirk Camille M et al. Fresh red blood cell transfusion and short-term pulmonary, immunologic, and coagulation status: a randomized clinical trial. American journal of respiratory and critical care medicine. 2012, 185(8): 842- 850.

17. Kaplan, JH. Biochemistry of na, K-ATPase. Annual review of biochemistry. 2002, 71(1): 511-535.

18. Uvizl R, Klementa B, Adamus M, Neiser J. Biochemical changes in the patient’s plasma after red blood cell transfusion. Signa Vitae. 2011, 6(2): 64-71.

19. Blann Andrew D, McCollum, Charles N. von Willebrand factor, endothelial cell damage and atherosclerosis. European journal of vascular surgery. 1994, 8(1): 10-15.

20. Blann, Andrew D, WaiteMalcolm A. von Willebrand factor and soluble E-selectin in hypertension: influence of treatment and value in predicting the progression of atherosclerosis. Coronary artery disease. 1996, 7(2): 143-147.

21. Weinberg JordanA, McGwin Jr Gerald, Vandromme MarianneJ, Marques MarisaB, Melton SherryM. Duration of red cell storage influences mortality after trauma. The Journal of trauma, (2010), 69(6): 1427-1431.

22. Sawant RB, Jathar SK, Rajadhyaksha SB, Kadam PT. Red cell hemolysis during processing and storage. Asian Journal of Transfusion Science. (2007), 1(2): 47-51.

23. O’Neill ohnS, Reddy AkhileshB. Circadian clocks in human red blood cells. Nature. (2011), 469(7331): 498-503.

24. Sparrow RosemaryL, Sran Amrita, Healey, Geraldine, Veale, Margaret F., & Norris, Philip J. in vitro measures of membrane changes reveal differences between red blood cells stored in saline-adenine-glucose-mannitol and AS-1 additive solutions: a paired study. Transfusion (2013), 54(3): 560-568.

25. Högman CF, Knutson F, Lööf H, Payrat JM. Improved maintenance of 2,3 DPG and ATP in RBCs stored in a modified additive solution. Transfusion. (2002), 42(7): 824-829.

26. Burger Patrick, Kostova Elena, Bloem Esther, Hilarius-Stokman Petra, Meijer Alexander B. Potassium leakage primes stored erythrocytes for phosphatidylserine exposure and shedding of pro-coagulant vesicles. British Journal of Haematology. (2013), 160(3): 377-386.

27. Hod Eldad A, Brittenham Gary M, Billote Genia B, Francis Richard O, Ginzburg Yelena Z et al. Transfusion of human volunteers with older, stored red blood cells produces extravascular hemolysis and circulating non-transferrin-bound iron. Blood. (2011), 118(25): 6675-6682.