Hypoxanthine: a new paradigm to interpret the origin of transfusion toxicity

Dear Sirs,

For blood banks, the possibility to store blood components for prolonged periods, preserving their clinical effectiveness, is an issue of paramount relevance for which a reliable solution has not been found, yet. In 1978, the standard conservation medium was enriched with adenine to sustain an adequate level of adenosine 5′-triphosphate (ATP); as a result, the potential duration of red blood cell (RBC) storage was extended up to 6 weeks over the years. The amount of adenine used in different countries currently ranges from 1.25 to 2.2 mM (Table I).

Table I.

Adenine concentration (mM) in additive solutions^*.

Additive solution	Adenine (mM)	Country
SAGM(1)	1.25	Europe, Australia, Canada
PAGGSM (MacoPharma)	1.4	Germany
MAP(2)	1.5	Japan
AS-1 (Adsol Baxter)	2.0	USA
AS-3 (Nutricel Pall Medical)	2.0	USA, Canada
CPDA-1(3)	2.0	Brazil
AS-5 (Optisol Terumo)	2.2	USA

^*Adapted from Sparrow RL⁶.

⁽¹⁾Saline-adenine-glucose-mannitol;

⁽²⁾mannitol-adenine-phosphate;

⁽³⁾citrate-phosphate-dextrose-adenine.

Parallel with the prolongation of RBC shelf-life, however, a number of studies highlighted an increased risk of adverse reactions. Those events lead to an increase in morbidity and mortality in critically ill patients who receive RBC that have been stored for a long time¹. These worrying reports stimulated a number of researchers to seek a connection between transfusion toxicity and RBC modifications, commonly indicated as “storage lesions”. Documented effects of RBC storage lesions¹ include increases in free iron and free haemoglobin, altered nitric oxide metabolism, neutrophil priming compounds (from stored cells) and generation of reactive oxygen species.

Attention has also been focused on the potential toxic effects of adenine, a product of ATP and nicotinamide adenine dinucleotide (NAD) breakdown, and its final metabolic product, uric acid. Those studies discarded the possibility that adenine could be harmful. However, the degradation pathway of adenine produces an intermediate, hypoxanthine whose presence in conservation medium together with its potential toxic effects have been disregarded, so far.

Under physiological conditions the concentration of hypoxanthine is very low, both inside erythrocytes (9.3 nM) and in human plasma (1–8 μM) (www.hmdb.ca). Our metabolomic profiling of leucodepleted RBC units revealed a significant increase of hypoxanthine concentration during storage², with the concentrations reaching very high levels compared to those in normal human serum. As shown in Figure 1, in both leucodepleted and non-leucodepleted RBC units, hypoxanthine progressively accumulates inside RBC (reaching concentrations up to 300 μM) and in the suspension medium (up to 1 mM). The build-up of hypoxanthine in RBC units is due to the absence of xanthine oxidase (XO), the enzyme that catalyses hypoxanthine metabolism to xanthine and then to uric acid with the concomitant production of reactive oxygen species (H₂O₂, O₂^•−)³.

Figure 1.

Hypoxanthine (HX) quantification by ¹H-NMR in RBC units during storage.

Leucodepleted RBC (n=8): (□) HX inside RBC and (■) HX in the suspension medium. Non-leucodepleted RBC (n=8): (○) HX inside RBC and (●) HX in the suspension medium. Data are expressed as the mean ± standard deviation.

RBC: red blood cells; ¹H-NMR: proton nuclear magnetic resonance.

In humans, XO is abundant in the intestine, liver and also in endothelial cells of the microvasculature. It is worth noting that the concentration of circulating XO is very low in physiological conditions, but its concentration increases in several pathologies in which liver or intestine are damaged. This effect has been indicated as responsible for the tissue damage associated with a number of pathologies^3,4. In fact, it is known that circulating XO may reach distant organs by binding to the proteoglycans on the outer surface of endothelial cells; thus, many tissue injuries observed after the increase of XO in the plasma have been related to XO-released reactive oxygen species^3,4. In turn, this vascular damage stimulates the production of cytokines that play a critical role in systemic inflammatory responses and multiple organ dysfunction syndromes.

On the basis of the biochemical and clinical observations reported, we believe that multiple transfusions performed using “long-stored” RBC units produce a significant, sudden increase of hypoxanthine in the blood that, even in the presence of medium-low concentrations of circulating XO, will lead to increased generation of reactive oxygen species, a fact that may contribute to transfusion-related tissue damage. When using RBC units at the end of their storage period, the concentration of hypoxanthine in the plasma after the transfusion of three units in a normovolaemic patient is in the range of 100 μM. It is worth noting that an increase of hypoxanthine in the plasma, reaching levels up to 50–100 μM, has been reported under hypoxic/inflammatory conditions³; moreover, that concentration of hypoxanthine is used as a free radical-generating system in cytotoxic assays on human cells in vitro⁵.

For decades, it has been hypothesised that the higher incidence of morbidity and/or mortality following transfusion therapy might be associated with the use of “long-stored” RBC units. We suggest that transfusion of “old” RBCs units, containing a high concentration of hypoxanthine, promotes bursts of reactive oxygen species and consequent neutrophil activation. These events may initiate a cascade of inflammatory reactions concurring with the observed higher incidence of negative outcomes in transfused patients, such as acute lung injury and multiple organ failure. In conclusion, our metabolomic study draws attention to the age of RBC units, particularly when performing multiple transfusions, since hypoxanthine can reach hazardous concentrations.