Cumulative erythrocyte damage in blood storage and relevance to massive transfusions: selective insights into serial morphological and biochemical findings

Abstract

Elucidating the precise mechanisms of cumulative red cell damages during storage and the potential harmful consequences after transfusion are achievable by exacting laboratory science and well-defined clinical studies in progress. Accordingly, for larger magnitude blood transfusions (i.e. 8–12 U in 24 hours), the quality of the stored blood and its characterisation are of special academic and clinical importance. Our main objectives in this review are to illuminate facets of the red cell storage lesion for prolonged storage (0–42 days) by concentrating on various hallmarks of the disorder: 1) identifying and characterising serial markers of the progressive lesion with respect to red cell dysmorphology, deformability, haemolytic fragility and dysfunction both in storage and the microcirculation; and 2) relevant biochemical findings of redox status correlated to oxidative stress of erythrocyte proteins. This is accomplished in part by reliance on advanced metabolomic and proteomic technologies using various sophisticated tools such as high pressure liquid chromatography in combination with mass spectrometry of proteins and small molecule metabolites. It is anticipated that these sophisticated methodologies and the experimental results therein shall lead to further advances in the quality improvement of red cell storage.

Morphological features of red cells, with reference to their deformability, haemolytic fragility and dysfunction in storage and the microcirculation: a historical perspective and modern studies

Introduction to erythrocyte shapes, deformability and ektacytometry

The medical implications of the red blood cell (RBC) storage lesion (RSL) and its potential adverse consequences to cell function warrant further study, particularly in massive transfusion scenarios. The perturbations in morphology of erythrocytes are closely and inextricably linked to their functions in the circulation. Historically, it is relevant to refer to a widely accepted morphological classification system of red cell shapes in evolution, the “Bessis classification” of nomenclature, formulated in the 1970s^1,2.

Consistent with the Bessis classification, erythrocytes contained in optimised storage media are well-documented to evolve and degrade progressively, from day 0 to day 42 of routine storage. Transformations that are reliably observed by the microscopist are discocytes to disco-echinocytes I, disco-echinocytes II, echinocytes, sphero-echinocytes I and sphero-echinocytes II (Figure 1)³. There are other recognised cell shapes, but these are the predominant types.

Figure 1.

The general Bessis morphology classification of red cell shapes.

Reprinted with permission from Springer International^©3.

The Bessis nomenclature is relatively intuitive (Figures 1 and 2)³. The dumbbell-shaped discocyte develops progressive spiculation (disco-echinocyte), completely loses the discoid shape (echinocyte), and subsequently accrues partial spherical aberrations (sphero-echinocyte). In the illustration of red cell shapes, there is a progressive decrease in the diameter of the red cell, which correlates to processes that degrade and lose portions of the cell membrane. These processes are described as extracellular microvesiculation (MV) in the sub-micron range, as low as 200 nm or 0.2 μm⁴. To appreciate the minute nature of these particles by comparison, the diameter of an RBC discocyte is in the range of 7.82 μm ±0.62. In addition to red cells, white cells and platelets give rise to exosomes and microvesicles in storage media. Exosomes are even smaller submicron particles that bud off from residual non-red blood cells in the range of 40–150 nm (0.04–0.15 μm)⁴.

Figure 2.

Scanning electron micrographs of discocyte transformation into echinocyte.

(A) Discocyte. (B) Echinocyte II. (C) Echinocyte III³.

Reprinted with permission from Springer International^©.

Nearly 50 years ago, Bessis and others demonstrated the utility of optical methods to measure RBC deformability, an ektacytometer, the methodology being discodiffractometry and still quite relevant today^5,6. The loss of RBC deformability is one signature feature to define the storage lesion and is germane to the first study of this factor in a model of haemorrhagic shock in 2003⁷. The clinical significance of the ektacytometer is that RBC deformability translates to an erythrocyte’s capability to transit the microcirculation and effectively deliver its payload of oxygen⁷.

The morphology index and an assessment of its utility in bioassays of the storage lesion

We engage phase contrast microscopy with facility for erythrocyte morphological observations while others utilise scanning electron microscopy. We developed a simple numerical scoring system or morphology index (MI) which grades the gradual increasing severity of the red cell degradation from 1 to 8, from discocytes to echinocytes/stomatocytes, spherocyte degraded forms and ultimately cells in the process of haemolysis. A smaller MI range (1.0–3.0) is indicative of cells classified in morphology similar to discocytes or disco-echinocytes and a larger MI range (5.0–7.0) reveals degraded cells of the sphero-echinocyte/stomatocyte classification.

The MI should have practicable utility when, for example, a drug, chemical or biological influence being measured in the storage bioassay has a known predilection for sphero-echinocytes to form. In our unpublished research regarding the effects being tested in particular bioassays of the storage lesion, it is determined that the percentage of red cell survival, and potentially the extent of haemolysis, to be a more sensitive indicator of the storage lesion. When the percentage mix of cell shapes has a tendency to be equivalent between the untreated control and the treated sample, there is less difference between the groups. The utility of the MI is for tracking significant morphological changes in storage media when correlated to other biomarkers of the storage lesion.

Spherocyte morphology in storage and the process of cell swelling leading to end-stage haemolysis

Not depicted in the Bessis nomenclature are the shape changes or relative size of the evolving spherocyte, a non-spiculated end-storage red cell. Multiple phase contrast observations are made by the Authors of this paper to identify morphology at magnifications of 100x, 400x, and 1,000x oil immersion. We use 20 μL blood samples, a 1/200 haemodiluted mix of non-leucoreduced whole blood, 0.9 N normal saline and SAG-M storage media, and refrigerated at 4–6 °C (Putter and Seghatchian, unpublished research; 2017). The studies can be repeated with leucodepletion and pathogen reduction. In a bioassay, we observe a few discernable end-stage spherocytes, all larger in diameter than the numerous recognisable sphero-echinocytes (SEs) but none smaller. The SE cells are trending towards end-storage life. Some end-of-life red cells develop crenation, a shrunken appearance to the cells with notched or scalloped edges. This precedes transition to a precipitous expansion of the cell diameter, and spherocytosis and haemolysis.

In bioassays of RBC storage ex vivo, we are distinguishing this observation of expanding/swelling of spherocytes, which are larger than the small spiculated sphero-echinocytes. Morphologically, the microscopic evidence gives credence to the theory that sphero-echinocytes of prolonged storage are physically degraded, taking on water passively to the inner cell through a leaky or porous cell membrane with observable punctate holes⁸. Concomitantly with vesiculation and swelling of the erythrocyte, spiculation disappears and the red cell expands prolifically to the point of rupture or haemolysis. This yields a haemoglobin-deficient remnant: the red cell ghost. Under phase contrast microscopy, the expanding spherocytes, and less frequently elliptocytes, present as non-spiculated cells with a whitish hue to their cytoplasm. The coloration is specific to the set up of the microscope, for example, a 45 mm blue filter LBD-IF positioned topside of the light exit glass of the CX41 scope (Olympus Corporation of the Americas, Scientific Solutions Group, Waltham, MA, USA). A blue filter reduces the yellowing effect of the lamp light source at the slide.

A novel transfusion stress model: ex vivo assays of red cell concentrates and the degradation of RBC shapes

Glucose-6 phosphate dehydrogenase deficiency (G6PD⁻) is concentrated geographically in the Mediterranean, sub-Sahara, Middle-East and Asia regions. It is the most ubiquitous of human enzyme defects. It effects approximately 3.5–7.5% of the population, or 400 million persons. Class II effected individuals have less than 10% G6PD activity marked by non-spherocytic haemolytic anaemias compared to class III (10% to <60% of normal activity), and class IV (>60% of normal enzyme activity while asymptomatic clinically)⁹. As blood center testing for G6PD deficiency is not routine, this is potentially clinically relevant causing suboptimal donor resuscitation of massive transfusions, i.e. haemorrhagic shock states and the critically ill. As a result, the suitability of class II donors for large transfusions comes into question and warrants further enquiry in the testing laboratory¹⁰.

A principal reason for concern is that the G6PD enzymatic deficiency is the first metabolic rate-limiting step of the pentose phosphate pathway (PPP). In red cells, the PPP is a generator and source of the protective reducing agent nicotinamide adenine dinucleotide phosphate (NADPH). NADPH is thought to limit the negative consequences of in vitro-generated hydrogen peroxide, H₂O₂ and organic peroxides¹⁰. The peroxides are capable of initiating damage to red cell membranes and haemoglobin by the formation of harmful reactive oxygen species (ROS), causing protein fragmentation and oxidative carbonylation reactions. Primary protein carbonylation reactions involve oxidation of side chains of amino acids, such as arginine, lysine, proline and threonine, to form ketones. Secondary carbonylation reactions involve lipid peroxidation of polyunsaturated fatty acids of the cell membrane. This leads to the formation of cytotoxic three to nine carbon chain reactive aldehydes such as the stress marker malondialdehyde¹¹.

Innovative research involves the development of a stress model of transfusion ex vivo whereby G6PD⁻deficient G6P⁻ RBCs are compared to sufficient G6PD⁺ RBCs while simulating “stored donor” to “non-stored recipient” transfusions. The stored donor blood is artificially stimulated with chemical agents tert-butyl hydroperoxide, tBHP or diamide to simulate oxidative stress conditions. The G6PD⁺ or G6PD⁻ RBCs are studied serving as either donors or recipients¹⁰. As the intriguing methodology is subtlety complex, it is better comprehended as formulated in Table I.

Table I.

Glucose-6 phosphate dehydrogenase (G6PD) deficient red cells exposed to oxidative stress serving as potentially suboptimal donors or serving as better recipients to healthy red cells.

Severely deficient G6PD blood may be sub-optimal as the donor
G6PD−	G6PD+	Supernatant G6P− or G6PD+	Stored 21, 30, 42 d	Donor	tBHP or diamide induced oxidative stress
FFP G6PD+ or −	FFP G6P+ or −	Healthy RBCs	Non-stored	Recipient
Severely deficient G6PD blood serving as better recipients of healthy RBCs
Supernatant	Supernatant	Healthy RBCs	Stored 21, 30, 42 d	Donor	tBHP or diamide induced oxidative stress
G6PD−	G6PD+	FFP G6PD+ or −	Non-stored	Recipient

Regarding the variants, G6PD⁻ are deficient and G6PD⁺ are sufficient in enzyme activity¹⁰. Table columns define the mixture of components tested. d: day; tBHP: tert-butyl hydroperoxide; FFP: fresh-frozen plasma; RBC: red blood cell. Assays include measurements of cell fragility, haemolysis, anti-oxidant capacity, reactive oxygen species, supernatant potassium, microparticles, pro-coagulant activity, protein carbonylation and metabolomic analysis.

Beginning in the middle stage of storage, day 21, and extending through the later course, day 42, irreversible sphero-echinocyte species propagate in G6PD⁻ stored red cells but counterintuitively at lower levels than G6PD⁺ cells. G6PD⁻ red cells actually tolerate end storage well, at least based upon preservation of a more normal morphology. Towards end-stage storage, the rate of progressive fragility in osmotic and mechanical stress conditions is found to be comparable between G6PD⁺ and G6PD⁻ RBCs. Yet the intrinsic stability of G6PD⁻ cells becomes significantly compromised by oxidative stress simulated in a stress transfusion model. Individuals severely deficient in G6PD activity but adapted to oxidative stress chronically may, in fact, serve as better recipients of healthy blood rather than serving as donors¹⁰.

Dysmorphology of red cells correlated to diverse markers of the storage lesion in SAG-M

The serial degradation of red cell shapes is particularly accelerated at three weeks or day 21 of storage onward. In whole blood of healthy donors mixed with CPD-A anticoagulant, processed and suspended in SAG-M, liquid storage of cells is followed by scanning electron images at 2,000x. Random counts of 600 cell shapes and their percentage in the Bessis classification are measured weekly for 0–42 days. The data collection adheres to a modified classification system of reversible and irreversible forms by Berezina et al.¹² After several weeks, 33% are considered partly reversible forms of echinocytes and stomatoctyes while probably carrying forward cumulative damage of cytosolic and membrane protein/lipid structures. The residual 16% are irreversible forms of spherocytes, ovalocytes, sphero-stomatocytes, sphero-echinocytes and crenated cells. By day 21, a phenomenal 50% of erythrocytes shapes reveal non-discocyte phenotypes as an indicator of the imperfect storage conditions¹³.

In the testing laboratory, these degraded cell shapes correlate with other demonstrable markers of a progressive storage lesion. Representative examples are: increased osmotic fragility tested in diluted saline solutions, a rising supernatant K⁺ electrolyte, elevated haemolysis, declining intracellular and supernatant pH with attendant lactic acid production, a rising erythrocyte sedimentation rate, a decline of reduced glutathione (GSH), an elevation of oxidised glutathione (GSSG), and increased carbonyl biomarkers of lipid peroxidation, i.e. malondialdehyde¹³.

Dysmorphology and haemorheological findings of pRBCs stored in adenine-saline

In a morphology analysis of red cell concentrates by other investigators in AS storage media, congruous results of shape degradation are observed at day 21. These are counts on 500–1,500 RBCs in randomly chosen fields, finding 46% non-discocytes, 31% reversible RBCs, and 16% irreversible forms¹⁴.

Haemorheological changes over the time course of refrigerated adenine saline stored red cells are reported by researchers to include: abnormal shapes according to the Bessis classification, a concomitant deformability index determined using micropore filtration, extent of haemolysis and acidosis, assayed by pH and base deficit BE of packed red blood cells (pRBCs). By day 21, in addition to the progression of abnormal RBC shapes, there is a declining deformability, and increased haemolysis and acidosis¹⁴.

Red cell survival and haemolysis: the standard quality indicators of stored red cell product

In hypothermic blood storage, there is recognition that the RBC storage lesion involves a variety of factors. The increasing formation of extracellular microvesicles (MVs) released into the product’s supernatant is one such manifestation of the storage lesion. Consequential transfusion-related immunomodulatory effects (TRIM) are reported in association with microvesiculation that occurs during component manufacturing and in the circulation. TRIM is considered to be one mechanism of the possible adverse consequences of transfusion, such as infections, multi-organ dysfunction and mortality. Transfusion-associated lung injury (TRALI) is one example, and is part of multi-organ dysfunction. It is believed that the accumulation of MVs in the storage media during component processing and in prolonged storage is a significant contributing factor to post-transfusion TRALI¹⁵. This is related to neutrophil activation and proinflammatory cytokine and chemokine mediators of peripheral blood mononuclear cells (PBMCs)^16–18. Examples of PBMCs are T-cell, B-cell, and natural killer (NK) lymphocytes, and monocytes with rounded nuclei.

As red cells age under various storage conditions, there is ongoing haemolysis and the presence of either free haemoglobin or bound haemoglobin to these extracellular MVs. In transfusions, vasoconstrictive effects may predominate at the capillary bed level as a result of a depletion in nitric oxide (NO), as this is scavenged by free haemoglobin^4,19.

Application of the dilution assay: a unique stress test of haemolysis

As an indicator of incipient red cell failure, the extent of haemolysis is an old but well established quality standard of the processed red cell product and the storage lesion. The processing of red cells in the blood bank has to be optimised in order to consistently limit end-of storage haemolysis, which in Europe must not exceed 0.8%²⁰. The extent of haemolysis can also be adapted to bioassays that measure the red cell survival, the percentage of non-haemolysed red cells of the initial total RBC count/μL. This can be easily determined with acceptable accuracy using a haemocytometer and phase contrast microscopy.

In a storage SAG-M/0.9 N NaCl mix optimised in composition microscopically to favour discocyte vs abnormal shapes, a non-anticoagulated 1/200 diluted sample of whole blood has a tendency to unpredictable clot and rouleaux formation. Visually, there is increased “stickiness” of RBCs aggregated to the reservoir wall and accelerated haemolysis at week 5 of end storage. As a result, spurious total RBC counts are apt to occur using a haemocytometer (Putter and Seghatchian, unpublished observations; 2017).

An alternative methodology is practicable to monitor low levels of haemoglobin accumulating in the supernatant instead of the percentage red cell survival. The use of a compact portable spectrophotometer, the HemoCue^® Plasma/Low Hb System (Brea, CA, USA) is relevant in the bench laboratory.

In this stress dilution assay, a 1/200 dilution of whole blood when compared to normal storage media is a highly stringent environment for red cell survival. Nearly 100% haemolysis occurs at week 5 of storage compared to levels not exceeding 0.8% in typical storage. As a vigorous stressor, this assay is adaptable to measure whether novel treatment effects to augment the quality of storage can slow the significant rate of haemolysis. The 1/200 stress dilution assay complements other tests such as osmotic fragility (using increasingly diluted saline constituted blood samples) or mechanical fragility (employing rocker balls).

Red cell damage in storage media and the microcirculation: the double hit hypothesis

Morphologically, damaged red cells have been visualised in the systemic circulation by intravital video microscopy, one example being intra-abdominal visualisation of the greater omentum. These echinocytes, sphero-echinocytes and crenated RBCs are observable within the arterioles, capillaries, and venules of the microcirculatory bed during low flow circulatory states²¹. Although the causes are different, the proliferation of crenated spheres in shock states mirrors the abnormal cell shapes of the evolving storage lesion. The joint presence of damaged red cells in storage and the microcirculation raises concern about their synergistic effects, potentially causing a double hit of injury.

Nature of red cell dysfunction in the microcirculation

Discocytes have an optimal surface to volume ratio for oxygen exchange, which declines as degraded cells evolve into spherical shapes. The proposed etiologies of red cell damage and microcirculatory failure in critically ill patients are multifactorial: depletion of intracellular ATP reserves, tissue hypoxia, oxygen free radicals, complement activation, cytokine immunomodulatory mediators of inflammation, adverse shifts in intracellular ionic composition and extracellular hypertonicity causing shrinkage or hypotonicity causing swelling of the red cells^22,23.

Imbalances in the intracellular ionic milieu of cytosolic Na⁺, K⁺, CA⁺⁺ and Mg⁺⁺ have the potential to change the normal conformation of the haemoglobin protein tetramer, and disrupt the internal distribution of haemoglobin, including its attachment to the inner leaflet of the plasma membrane. The inner plasma membrane of the red cell is thought to be integral to maintaining the discoid shape. Ionic shifts have the capacity to impair the function of the cytoskeleton proteins such as spectrin in the inner membrane. Inducible distortions of red cell shape during shock are postulated to reduce cell deformability, capability to successfully pass through the microcirculation, and oxygen delivery capacity.

First serial measurements of red cell deformability in haemorrhagic shock

Since the inception of ektacytometry, researched by Bessis and others in the 1970s and 1980s, this haemorheology methodology is currently recognised as a valuable laboratory asset. Its utility to measure RBC deformability is important to study the pathophysiology of damaged red cells both in prolonged storage and in the microcirculation. Of special interest are the medical implications of changes in red blood cell deformability and their effect on tissue oxygen delivery in haemorrhagic shock states. In transfusion of these cases, the patient is exposed to damaged RBCs from two sources: 1) storage, and 2) those originating de novo in the microcirculation. The end result of refractory haemorrhagic shock is microcirculatory failure leading to multi-organ dysfunction.

In 2003, university researchers published the first study of red cell deformability measured serially in haemorrhagic shock states. Pathophysiologically, shock manifests as a low flow state and tissue hypoxia and has been demonstrated to reliably cause red cell damage in the post-shock period. A standardised haemorrhagic shock model uses Sprague-Dawley rats exsanguinated to a mean arterial pressure, MAP of 30 mm Hg for 90 minutes. Shock is induced by controlled bloodshed through cannulation of the external jugular vein and periodic resuscitation to maintain the target arterial pressure at 30 mm Hg. This is achieved by reinfusion of shed blood that has been heparinised in a syringe, and after 90 minutes, there is reinfusion of the residual shed blood, i.e. the final resuscitation²⁴.

The RBC mean elongation index (EI) is computer calculated at various shear stresses in Pascal (Pa) using ektacytometry. Assays of erythrocyte samples are collected serially before and after the induction of shock. Scanning electron microscopy documents that are predominantly normal show biconcave discocytes in the pre-shock period. In the post-shock period, measured up to 6.0 hours, there is a persistent decrease in the RBC EI in the lower range of shear stress (0.5–1.58 Pa); this is an indicator of impaired erythrocyte deformability. Impaired deformability is worse in the decompensated rats, 50% of which died in refractory shock compared to the compensated rats that survived the 6.0-hour post-shock period²⁴.

Regression analyses reveals a tight relationship between an increasing percentage of abnormally-shaped cells in the circulation and decreasing red cell deformability. Impaired deformability correlates with two abnormal populations of cells, primarily echinocytes, thought to be potentially reversible in form to discocytes and a smaller subgroup of sphero-echinocytes, by 3.0%, thought to have irreversible damage²⁴. Supported by intravital video microscopic and red cell radiolabelling studies, the advanced degraded spherical forms of rigid red cells are thought to become entrapped in the capillary circulatory bed of the lung, bone, liver and spleen^25,26.

Biochemical findings of the red cell storage lesion. Advanced metabolomic and proteomic studies

Dysregulation of the red cell inner leaflet AE1-Band 3 respiratory metabolism complex and potential cytotoxic effects ex vivo

One of the most intricate of red cell storage lesions is protein fragmentation and subsequent dysregulation of the anion exchanger AE1-Band 3 respiratory metabolism complex. The putative damage is by the proteases calpain, caspase and reactive oxygen species (ROS), such as hydroxyl radicals and superoxide. The latter ROS are known products of Haber Weiss and Fenton biochemical reactions, which are dependent upon haeme iron²⁷. This respiratory complex is stationed on the inner leaflet of the cell membrane, an N-terminal cytosolic domain identified as CDB3. CDB3 couples key enzymes of the glycolytic and PPP pathways. The chief function of AE1 is the chloride shift reaction, exchanging the influx of one Cl⁻ for the efflux of one HCO₃⁻ anion and strong acidification of the cell (HCl). By the Bohr effect, this unloads critical oxygen from haemoglobin, a reduced affinity for that molecule and its preferential delivery to oxygen-deficient tissues²⁸.

The CDB3 respiratory complex is thought to modulate which metabolic pathway prevails, either glycolytic or PPP, dependent on the competitive binding of either deoxygenated or oxygen rich haemoglobin. Deoxygenation favors anaerobic glycolysis and ATP energy production while O₂ saturation and oxidative stress is inclined to PPP and the production of antioxidant reducing units (NADPH)²⁸. It is predictable that storage lesions targeting Band 3 would induce cytotoxic dysregulation of respiratory metabolism along with reported abnormal clustering of Band 3 in the cell membrane. Conceivably, the dysregulation could be potentiated and have harmful consequences under significant oxidative stress conditions, massive transfusions for haemorrhagic shock being a prime example.

Dysfunction of red cell glyceraldehyde 3-phosphate dehydrogenase activity under hyperoxic storage conditions

Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) is a key regulatory enzyme of red cell glycolysis in storage. It is naturally regulated by a complex interaction with the CDB3 respiratory complex and the degree of oxygenation. This effects the downstream generation of the substrate diphosphoglycerate 1,3 DPG and sequentially 2,3 DPG via a translocating phosphate mutase, the Rapoport-Luebering shunt reaction. Maintenance of sufficient 2,3 DPG levels equates to preserving oxygen unloading capacity of haemoglobin to the tissues, the microcirculatory bed. In hyperoxic conditions of storage, the coupling of GAPDH to CDB3 compromises GAPDH activity, favouring a metabolic shift to the PPP pathway and synthesis of antioxidant NADPH-reducing units. Hypoxia has the reverse effect to favour glycolysis and generation of 2,3 DPG²⁹.

An elegant array of gel electrophoresis, high pressure liquid chromatography and mass spectrometry are utilised for metabolomic/proteomic experiments of erythrocyte concentrates stored in additive solution-3. The results support the theory that irreversible oxidative damage occurs to three disparate functional cysteine thiol residues, 152, 156 and 247, of the GAPDH in its active protein pocket. Not only does oxidation reduce GAPDH functional activity, it promotes protective extrusion of the irreversibly damaged enzyme by a process of extracellular membrane vesiculation to the supernatant²⁹.

Potential negative effects of multiple oxidative modifications to haemoglobin in storage

In addition to oxidative damage to GADPH, the results of comparable metabolomic/proteomic experiments indicate more than 20 oxidative modifications to haemoglobin, including histidine/cysteine residues H93, C94 and H144 of the Hb β chain. The net effects of these oxidative modifications to haemoglobin, which require further elucidation, are potential negative impacts concerning redox homeostasis, modulation of oxygen affinity by 2,3 DPG, and vasodilation mediated by NO³⁰.

Impairment of oxidative/reduction recycling of peroxiredoxin-2 protein in aged blood compromising antioxidant capacity

The third most abundant protein in red blood cells is peroxiredoxin-2, a key antioxidant homodimer (2-Cys Prx, shortened to Prx-2). Homodimers have a duplicate amino acid chain sequence but are not usually covalently bound to each other. Prx-2 plays a primary role in an endogenous high capacity non-catalytic recycling pathway to detoxify harmful generated intracellular oxidants such as the cell’s intrinsic peroxide H₂O₂. Essentially a cysteine sulfhydryl group becomes oxidised to form a disulfide bridge, identified as the inactive Prx-2 dimer, with no antioxidant capacity. The oxidised Prx-2 dimer must subsequently be converted or reduced back to the Prx-2 monomeric form in order to restore antioxidant capacity. We have a reversible cleaving of the disulfide bridge, now a reduced Prx-2. Recycling to the Prx-2 reduced state occurs by the flavoenzyme thioredoxin reductase in concert with another ubiquitous antioxidant protein, thioredoxin³¹.

Progressive storage of leucoreduced human pRBCs in Adsol leads to quantitating the recovery fraction of the residual oxidised Prx-2 day 7 compared to day 35 of storage when exposed to H₂O₂ [10 μM]/glucose. The results conform with the hypothesis that the recycling reduction reaction of oxidised Prx-2 is appreciably slowed in aged blood. Coincident with this concept of decreased recycling capacity of oxidised Prx-2, being concomitant to prolonged storage, is a corresponding increase in the susceptibility to haemolysis. This is artificially induced by exposure of red cell concentrates to H₂O₂ [10 μM], and the susceptibility to haemolysis noted to be greater for day 35 compared to day 7 concentrates³¹.

It has been proposed that oxidised Prx-2 is coupled to Hb and the cell membrane as oligomeric complexes. The oxidised Prx-2 conceivably has a red cell membrane protective effect regarding its preferential oxygenation reaction with H₂O₂. Innovative experiments of leucoreduced RBC storage in Adsol days 1 and 10 use a transgenic mouse model expressing wild-type human Hb, either Cys 93 or Ala 93 substituted in the β chain. This 10-day storage murine model is thought to mimic a 42-day storage period of human blood. Susceptibility to haemolysis is noted in the murine model with older blood and the substitution of a β 93-Ala residue for a β 93-Cys residue. While the precise mechanisms remain to be elucidated, it is believed that β 93-Cys of Hb is integral to modulating oxidative stress, possibly in part by being capable to up-regulate reductive recycling of Prx-2 in early storage. This recycling capacity is compromised in older cells³¹.

Anaerobic red cell storage to abrogate oxidative damage to intracellular and membrane proteins

In hypothermic storage of human red cells, their progressive deterioration can be demonstrably slowed such that storage life is reportedly increased by 50% up to 63 days. Artificial microvascular networks, AMVN have been developed to mimic the rat mesentery microcirculation and study the application of anaerobic storage conditions to abrogate oxidative damage^32,33. Microchannel networks are constructed to specifications 5 μm depth and a 5–70 μm range in width to simulate arterioles, capillaries and venules of the microcirculatory beds. Whole blood is anticoagulated in citrate phosphate double dextrose CP2D and leucoreduced. RBCs stored aerobically in AS-3 are the control arm representing conventional storage and compared to RBCs stored anaerobically in an experimental additive OFAS3 (New Health Sciences, Bethesda, MD, USA)³³.

Sample depletion of O₂ is achieved with a prototype custom device: 150 μm inner diameter microporous fibers (Celguard; 3M, Charlotte, NC, USA) and contiguous oxygen sorbent packets (ZB; Mitsubishi Gas Chemical Co., New York, NY, USA) Under hydrostatic pressure, bulk perfusion capillary rates are demonstrably higher for anaerobic vs aerobically stored blood³³. In the AMVN model, the positive effects of anaerobic storage are linked to improved mechanical rheological properties of the blood, specifically conserving the discocyte shape, decreasing the population of least deformable cells, and favouring unimpeded transit of normal shaped cells via the artificial network.

Identification of candidate protein biomarkers of the red cell storage lesion located in the supernatant

In a demanding initial experimental set up, the synthesis of a chimeric gene based upon a protein template is demonstrated and inserted into Escherichia Coli to trigger protein expression. The proteins are then purified as internal standards and tandem HPLC/mass spectrometry enables absolute quantification of Hb and other byproduct proteins in the supernatant. By this assay, candidate protein biomarkers of the storage lesion are identifiable, such as the haemoglobin chains, peroxiredoxins, catalase, carbonic anhydrases, GADPH, selenium-binding protein, peptidyl-prolyl isomerase A and aminolevolinate dehydratase (ALAD)³⁴.

Energy depletion and microvesiculation as harbingers of the storage lesion and the role of antioxidant blood additives as protective agents

High energy ATP depletion, rising intracellular acidosis, and a decline in 2,3 DPG are representative markers of the evolving red cell storage lesion. Attendant to these biochemical changes are disturbances in the ATP dependent NA⁺/K⁺ cationic pump of the cell, elevations in supernatant K⁺, decreased RBC deformability and increased osmotic fragility^12,13. In an effort to ablate the propagation of the storage lesion experimentally, investigators have incorporated anti-oxidant additives to blood storage media such as vitamin C (ascorbic acid) and N-acetyl cysteine (NAC)³⁵. These additives are associated with a reported reduction in oxidative stress to the red cell. This is manifested as decreased haemolysis, especially up to storage day 28, and less accumulation of oxidised byproducts such as malondialdhyde and oxidised glutathione GSSG and prostaglandins. The tripeptide glutathione GSH is maintained in a reduced state as a key antioxidant. It is integral to the protection of RBC cytosolic and membrane proteins by negating the harmful effects of reactive oxygen species. Moreover, antioxidant additives seem to positively maintain cellular ATP levels, possibly mediated by cyclic AMP (cAMP) linked to a decrease in the release of erythrocyte ATP³⁵.

As a result of uncompensated oxidative stresses under storage conditions, this promotes microvesiculation and the incorporation of bioactive proteins, lipids and microRNA into microparticles¹⁹. These donor-based bioactive particles have intercellular signalling capacity and the propensity of recruiting adverse pro-inflammatory and pro-thrombotic pathways in the transfused recipient³⁶.

Conclusions

Because red cell storage conditions cannot yet replicate the human microcirculation, the storage lesion progresses inexorably from days 0–42. After two weeks in early storage, morphological degradation of erythrocyte shapes begins to accelerate along with discernable changes in various metabolic biomarkers of the storage lesion, i.e. elevations in supernatant K⁺, lactic acid, and Hgb to name just a few. The acceleration of the lesion becomes pronounced at 21 days while irreversible RBC spherocytic and crenated cells ensue preceding the haemolysis event. Cell membrane damage is evident and can be observed as punctate holes at high magnification SEM. This serves as an indicator of the increasingly porous nature of the erythrocyte, ongoing vesiculation, and presages end-stage swelling of the cell. Haemolytic fragility and decreased deformability of the red cell contribute to decreased RBC survival in the circulation, increased erythrophagocytosis after transfusion and decreased oxygen unloading capacity to the microcirculation.

Coincident with the evolving storage lesion are the deleterious effects of reactive oxygen species that likely oxidise key regulatory proteins of red cell respiratory metabolism, such as cytosol domain Band 3, glyceraldehyde 3-phosphate dehydrogenase, peroxiredoxin-2 and haemoglobin. An array of metabolomic and proteomic studies and tools have enabled experimentalists to unlock the complex biochemistry and identify key protein biomarkers of the storage lesion. An increased understanding of this biochemistry heralds the development of new interventions such as anaerobic storage and antioxidant blood additives to storage media that can be applied practicably to improve the quality of donor red cells. Such research is expected to benefit various patient populations in the future, including the critically ill, those patients exposed to the adverse consequences of larger transfusions.