British Journal ofHaernatology, 1989, 71, 131-130

In vivo regeneration of red cell 2,3-diphosphoglycerate following transfusion of DPG-depleted AS-1, AS-3 and CPDA-1 red cells

ANDREW HEATON, THAIS KEEGAN A N D STEIN HOLME Tidewater Red Cross Blood Services and Department of Pathology, Eastern Virginia Medical School, Norfolk, Virginia

Received 1 7 March 1988; accepted for publication 25 July 1988

Summary. Regeneration of 2,3-diphosphoglycerate (DPG) was determined following transfusion of DPG-depleted group 0 red cells into group A recipients. Blood from five donors was stored in the adenine-containing solutions CIPDA-1, AS-1 or AS-3 for 35 d at 4°C. Post-transfusion red cell DPG and ATP were measured in separated group 0 red cells over a 7 d period. The studies confirmed rapid in vivo DPG regeneration with 2 50% of the maximum level being achieved within 7 h. An average of 95% of the recipients’ pre-transfusion DPG level was achieved by 72 h and by 7 d mean ( f SEM) DPG levels relative to recipient’s pre-transfusion DPG averaged 8 4 % ( f 1 3 % ) , 9 2 % ( f 1 7 % ) and84%(f21%)forCPDA-l, AS-1 and AS-3 red cells, respectively. Results were compar-

able to those previously reported for blood stored in ACD for 15-20d(Valeri&Hirsch, 1969;Beutler& Wood, 1969). The immediate regeneration rate, V, closely approximated first order regeneration kinetics with AS-3 red cells exhibiting double the rate of CPDA-1 red cells ( P < 0.001). AS-1 red cells exhibited an intermediate rate of regeneration which was not significantly different compared to either CPDA-1 or AS-3 (P>0.05) . V exhibited a significant ( P < 0 . 0 5 ) positive correlation with ATP levels 5-7 h post-infusion. ATP regeneration of the infused cells was rapid with a mean increase of 1.2 pmol/g Hb above post-storage levels being achieved 1 h following transfusion.

Red cell 2.3-diphosphoglycerate (DPG) levels are depleted after the first 14 d of storage in adenine-containing preserva- tives (Chanutin, 1967). The decrease is slower in citrate- phosphate-dextrose (CPD) whole blood than in acid-citrate- dextrose whole blood (Dawson et al, 1970) and addition of adenine, as in CPDA-1, or to an additive solution, such as AS- 1, causes a more rapid decrease (Sugita & Simon, 1965: Heaton et al, 1984). The extent ofDPG loss is related to the fall in pH since the rate of synthesis is dependent on the intracellular pH. DPG levels have been shown to return to normal rapidly following reinfusion of ACD or phosphate buffered red cells stored for 15-20 d (Valeri Kr Hirsch, 1969; Beutler & Wood, 1969). In each of these previous studies, three stable, blood group A, non-surgical anaemic patients, who had a wide range of pre-transfusion DPG values, were transfused with group 0 DPG-depleted red cells and the rate of regeneration was measured in red cells separ,ated by differen- tial agglutination.

Correspondence: Dr Stein Holme. Scientific Director, Research, American Red Cross Blood Services, Tidewater Region, 700 Olney Road, Norfolk, VA 23507. U.S.A.

Preliminary studies (Herve et al, 1980) had suggested that DPG was more rapidly regenerated in stored red cells suspended in an early additive solution, saline-adenine- glucose (SAG), and the current study was undertaken to assess the speed with which DPG was regenerated in vivo following transfusion of red cells stored in one of three other adenine-containing storage solutions. To minimize the effect of biologic variability, a paired study design was utilized in which the same donors would be drawn into each of the study media. Since red cell DPG levels correlated with the degree of anaemia and multi-transfused anaemic patients may exhibit DPG values inappropriate for the level of anaemia (Correra et ul, 1984; Torrance et ul, 1970), the present study was performed using normal recipients to minimize this variability. To evaluate whether prolonged storage or low ATP levels might affect DPG regeneration, a 3 5 d storage period was utilized. The study was conducted in two phases, the first of which involved the selection and verifica- tion of suitability of five group 0 donors: while the second involved five regeneration studies per storage solution using healthy, group A male volunteer recipients. The red cell storage media, CPDA-1, AS-1 and AS-3, were selected to

131

132 A. Heaton, T. Keegan and S. HoZme Table I. Composition of red cell preservative solutions'

CPDA- 1

anticoagulant/ preservative Anticoagulant Preservative Anticoagulant Preservative

combined CPD/AS-1 CP2D/AS-3

Sodium citrate dihydrate 1657 Citric acid, hydrous 206 Phosphate, hydrous 140 Dextrose, hydrous 2000 Adenine 17 Mannitol -

Sodium chloride -

Product volume (mi) 63

1657 - 1657 588 206 - 206 42

140 2 76 140 1607 2200 3220 1100

30

- 900 - 410 63 100 63 100

-

- 27 750

- - - -

Component content as milligrams per product volume.

allow comparison of combined anticoagulant/nutrient stor- age solutions (CPDA-1 packed red cells) with additive- suspended storage (AS-1 or AS-3 red cells), and to compare additive-suspended red cells stored with and without added phosphate (AS-1 versus AS-3).

METHODS

Donor and recipient sezection. Donor selection conformed to conventional American Red Cross health history guidelines. From a group of 11 volunteers, a panel of five group 0 donors was selected based upon the completion of at least two prospective hepatitis follow-up studies in blood recipients who received transfusions of units donated by one or more of the donors. These studies involved testing for hepatitis markers including serum ALT levels and hepatitis B surface antigen, surface antibody, and core antibody at the time of transfusion and approximately 30,60,90 and 180 d later. At least 2 units per donor were followed and confirmed absence of hepatitis transmission. All donors were also screened for HIV antibody.

The regeneration study recipients were adult male volun- teers whose ages ranged from 22 to 3 7, who had no medical disabilities at the time of transfusion, and who had never received any previous blood transfusion. All the recipients were blood group A1, with the exception of one donor who was blood group Az; all gave informed consent. Recipient and donor units were crossmatched after being typed for ABO and Rhesus D antigen. Recipients were rapidly transfused within an average of 15 min, with the exception of CPDA-1 units which required an average of 35 min. Approximately 1 5 ml from the unit was retained for post-storage in vitro testing (3 5 d). Zero time was taken from the point of infusion completion. Mean recipient haemoglobin rose from 1 5 . 3 f l . l g/dl to 15 .6f0 .9 g/dl immediately following infusion and by the end of the study was not significantly different from pre- transfusion levels. No transfusion reactions were observed.

Regeneration units, sampling and testing. For each donor, test units (450f45 ml) of whole blood were drawn in random order at least 56 d apart into CPDA-1, CP2DIAS-3 (Cutter Biologicals, Berkeley, Calif.) or CPD/AS-1 (Fenwal

Laboratories, Deerfield. Ill.) after which they were processed according to conventional American Red Cross procedures into CPDA-1 packed cells, AS-1 red cells and AS-3 red cells, respectively. Table I shows the formulations of the anticoagu- lant preservatives. All units were non-reactive for hepatitis B markers, HIV antibody and syphilis (RPR test) and had alanine aminotransferase (ALT) results< 30 IU/L In vitro analysis was performed on an additional EDTA sample of blood drawn from the donor at the time ofphlebotomy and an aliquot of the unit immediately post-processing (0 d). Hae- moglobin. haematocrit. pH (37"C), red cell ATP and DPG; and supernatant haemoglobin (from which per cent haemo- lysis was calculated), glucose and phosphate were deter- mined by techniques previously described (Heaton et al, 1984). The units were then stored at 4 f 2°C for 3 5 d.

After transfusion, 25 mi EDTA whole blood samples were drawn immediately and at hourly intervals for 7 h. There- after, two samples were collected on day 2, one on day 3, and one on day 7. Samples were kept at 4°C throughout processing to minimize biochemical changes. Whole blood haemoglobin, haematocrit, and red cell ATP and DPG were performed on an unseparated ( 2 ml) aliquot and glucose and phosphate assayed in separated plasma.

CeZZ separation. The majority of the sample (approximately 20 mi) was subjected to differential agglutination at 4OC with anti-A.B blood grouping serum (Immucor Inc., Norcross, Ga.) using a modification of the Ashby technique (Ashby. 1948; Beattie. 1977). The separation process took between 1 and 2 h per sample.

Contamination of separated group 0 cells with residual group A cells was estimated by incubating an aliquot of the separated cell suspension with FITC-conjugated anti-human polyvalent immunoglobulin (Meloy Laboratories, Springfield, Va.) according to the technique of Cohen et al(1960). After washing the cells to remove unbound conjugate, a thin film was examined on a glass slide under fluorescent illumination. Contaminating (non-0) cells were quantified in a similar manner to that of a reticulocyte count with the exception that the stained cells were counted under EPI-fluorescence optics (excitation 485 k 2 0 nm, barrier filter 520-560 nm) and the total cells counted using white light (Brecher & Schneider-

2,3-DPG Regeneration 133 Table 11. In vitro studies of units used for regeneration studies (mean f SD. n = 5 for each)

Phosphate, Glucose, Storage Day of ATP, W.B. 2.3-DPG, W.B. plasma Haemolysis plasma medium storage (pmol/g Hb) ':pmol/g Hb) (mgidl) (%) (g/dl) pH ( 3 7'C)

CPDA-I 0 3.70f0.22 1 3 6 f 1.6 11.1f1.5 0.07f0.05 4 5 3 f 4 4 7,0610.06 35 1.7150.45 0.1 50 .2 27,352.8 0.82f0.55 6 1 f 2 0 6.48f0.06

AS-1 0 3.9450.92 13.052.2 *1.8+0.9 0.0610.03 *11471 147 7.0910.07 3 5 * 2 6 3 1 0 . 3 9 0 .210 .3 *18.012.3 0.37f0.30 *716151 6.4610.07

AS-3 0 3.6950.24 13.1 f l . 1 52.9f7.5 0.0710.04 7993~75 *6.86+0.06 35 3.22f0.58 0 .3 f0 .2 36.0f2.0 0.43f0.17 539A.66 6.46f0.11

* Significantly (P<0.05) different from results for one or both of the other storage media.

Table 111. In vitro studies of recipients pre- and post-transfusion

Phosphate, Glucose, ATP, W.B. 2.3-DPG. W.B. plasma plasma

Storage medium (pmol/g Hb) (pmollg Hb) (mg/dl) Wdl)

CPDA- 1 * 3.8510.35 l l . 8 f 0 . 9 3.250.7 8 6 f 6 :-4.06f0.55 11.4f1.2 3,710.8 9 7 f 1 6

AS- 1 3.8910.35 1 1 . 7 f 1 . 4 4 . 6 f 1 .Z 9 4 f 1 7 t3 .89f0.39 11.651.1 4.0f0.7 96f .17

'3.74f0.60 1 . 4 f 1 . 5 3.5f0.7 9 2 f 2 0 -13.88f0.67 1 2 . 8 f 1.6 3.850.9 102f18

* Mean values f SD prior to infusion, five recipients for each storage medium. t Mean values 5 SD over 7 d post-infusion, five recipients per storage medium,

13 samples per recipient (n=65 each medium).

man, 19 50). The contaminating fluorescent cells were expressed as the percentage of the total cells. Separated group 0 red cells were assayed for haemoglobin, and red cell ATP and DPG which were corrected for contamination by group A cells:

Corrected result (pmol/g Hb) =

[ R e ~ ~ l t l , ~ ~ ~ ~ ~ ~ e d - { [ R e ~ ~ l t l ~ ~ ~ - ~ ~ ~ ~ ~ ~ )(fl where f is the contamination of separated cells: expressed as a fraction.

Preliminary studies which were performed to evaluate the separation protocol by measuring DPG and ATP before mixing and after separation confirmed that neither the DPG nor ATP levels changed significantly during the separation process.

Data analysis. Analysis of the rate of regeneration was performed based on an integrated first-order rate equation (Segal, 1976). Least squares regression plot analysis was conducted on results for the 1-7 h samples (seven data points) according to the formula:

1 [PI v t [Plmax-[PI - k t '-T

where [Plmax is the maximum concentration of DPG found

- 2.3 [Plmax - log

Table IV. Summary of 2.3-DPG regeneration ~

Storage Degree of fit to medium k (pmol/g Hb) V (pmol/g Hb/h) the equation, r

.-. -

CPDA-1 10.3 f 1.1 0.184 f 0.104 0.9935 f 0.0071 AS-1 11.353.1 0.25410.143 0.9761 f0.0454 AS- 3 11.5f2.5 0.381f0.197 0.9966f0.0031

for the separated group 0 cells during the study period; [PI is the DPG concentration at the sample time, t ; V is the maximum rate of regeneration; and k is the constant describing the regeneration. The fit of the data to this equation was determined from the correlation coefficient, r.

RESULTS

Transfusion units: pre- and post-storage data Table I1 shows the pre- and post-storage parameters of the units stored for 3 5 d in each of the three additives. In all units, DPG was effectively depleted after 35 d of storage. The post- storage ATP levels were significantly lower for CPDA-1 red cells than those in either AS-1 or AS-3 ( P < O . O l ) though

134 A. Heaton, T. Keegan and S. Holme

14 r

12

P I" 10 CI)

u)

0

. 32 8 - E a - 6 - g n ' 4 - '? (u

2 -

-

-

I I I I ' f-

0 1 2 3 4 5 6 7 2 4 4 8 7 2 01 '

Time Post-Transfusion (h )

Fig 1. Mean 2,3-DPG in recipient and in transfused red cells (corrected for recipient red cell contamination) during 3 d post- transfusion. Mean (n = 5) red cell DPG separated (-) and unseparated (- - -) values are grouped by anticoagulant CPDA-1 ( O ) , AS-1 (D), AS-3 (A).

Fig 2. Mean ATP in recipient and in

redcell contamination) during the first 24 h 0 4 transfused red cells (corrected for recipient 24

post-transfusion. Same symbols as in Fig 1. Time Post-Transfusion (h)

there was no significant difference in post-storage ATP between AS-1 and AS-3. Haemolysis was slightly, but not significantly, greater in the CPDA-1 units compared to the AS-1 and AS-3 units.

Recipient data throughout study Table I11 shows the results (mean *SD) of in vitro testing on samples drawn from the recipients immediately prior to transfusion and for the 7 d post-transfusion, grouped by the unit storage solution. During the course of the study, an average of 147 f 17 ml (range 11 7-1 98 ml) of red blood cells was transfused to the recipient and approximately 1 50 ml of red cells (1 3 samples of 25 ml whole blood) were withdrawn. All subjects achieved peak regenerated DPG levels by 3 d post- infusion and exhibited a modest decrease by day 7 averaging

0.90 f 1.74 pg DPG/g Hb. The average decrease in DPG from the peak was not significantly different for the three storage media studied. There was no significant difference between recipient in vitro results in the three groups studied nor did regeneration correlate with plasma glucose or phosphate (P>0.05).

Cell separation and regeneration data The average contamination of separated group 0 cells with recipient group Acells was 1.3 f 1.6% (n= 195, meanfSD). Removing results for one individual who was blood group A2, contamination averaged 0.9 *0.5% (n= 182). There was no significant difference (P> 0.05) between the three groups (mean f SD): CPDA-1: 0.8 f 0.5% (n = 5 2 ) , AS-1: 0.9 f 0.5% n=65), and AS-3: 1.0f0.4% (n=65).

2,3-DPG Regenera tion 1 3 5

Table IV summarizes the rate of DPG regeneration during the first 7 h. The regeneration rate (V) was slowest with CPDA-1 and most rapid with AS-3, a difference which was significant (P < 0.05). There was no significant difference between the regeneration constants, k , for the three storage solutions. Fig 1 shows the mean separated and unseparated (recipient) DPG levels, grouped by storage solution, which confirms similar DPG levels within 48 h of transfusion.

The mean red cell ATP in transfused cells increased significantly ( P < 0.05) in all three groups immediately following transfusion to levels that at 1 h averaged 144 f 105% higher than that of the unit value at the time of transfusion. The ATP levels of the separated cells returned to recipient pre-transfusion levels in an average time of 4 h following infusion for CPDA-1 units and after 7 d for both AS- 1 and AS-3 cells. Fig 2 shows the average transfused and recipient red cell ATP grouped by anticoagulant for 24 h following transfusion. During this period, the mean ATP levels of the transfused cells of each additive group were parallel although, after this, the red cell ATF’ of the CPDA-1 group gradually increased and that of the AS-3 group decreased so that, by 7 d, the mean ATP levels were similar.

Linear regression analysis of the relationship between V and the corrected separated ATP concentration after infusion revealed a positive correlation which was significant (P<O.05) at 5, 6 and 7 h post-transfusion. There was no significant correlation between DPG regeneration velocity and any recipient in vitro value; the degree of contamination of the separated cells: post-storage in vitro parameters of the unit: or the recipient’s pre-transfusion DPG ]level.

DISCUSSION

Post-transfusion DPG regeneration is a function of the activity of 2.3-diphosphoglycerate mutase, which could be affected by prolonged periods of 4°C storage as well as changes in the cell environment that occur after transfusion. These studies confirmed excellent post-transfusion regene- ration of DPG in red cell concentrates stored for extended periods of time in solutions containing adenine. DPG regene- ration was rapid, an average of 50% of the peak DPG levels being regenerated in the first 7 h post-transfusion. Although the regeneration rate was slower than that reported by Herve et a1 (1980) for SAG-red cells, results were similar to those reported for ACD whole blood or packed cells stored at 4°C for 15-20 d (Beutler & Wood, 1969; Valeri & Hirsch, 1969).

Peak DPG values which occurred 48-72 h post-trans- fusion averaged 89% of the unit pre-storage levels and 9 5% of the recipient’s pre-transfusion DPG. The final DPG levels in the transfused cells consistently showed a reduction (mean 0.9 pmol/g Hb) from the peak levels. This may be a physiological response to the temporary increase in haemo- globin levels immediately following transfusion or may be explained by changes in the limiting enzymes of the glycolytic pathway which have previously been shown to be altered with cell age (Oski, 1970).

Since the maximal rate of DPG regeneration correlated (P<O.05) only with red cell ATP concentrations 5-7 h post- transfusion, and was greater for the additive with the highest

adenine and phosphate levels, the findings may reflect a more effective glycolytic pathway due to higher intracellular concentrations of these substrates of energy metabolism.

Previously reported studies have shown that viability of greater than 75% survival 24 h post-transfusion is associated with post-storage ATP levels > 1.5 pmol/g Hb (Dern et al, 1967). both for ACD and CPD whole blood. Since all three groups had post-storage mean ATP levels greater than this, and previous studies have shown 35 d post-transfusion recoveries ranging from 71% for CPDA-1 red cells to 86% for AS-1 red cells, the 144% rise of red cell ATP from unit post- storage levels which occurred 1 h following transfusion completion probably represents significant regeneration rather than selective loss of nonviable red cells with low ATP levels. In these studies, the mean increase in red cell ATP was similar for all the storage solutions studied (1.2 pmol/g Hb), though by 7 d this difference had disappeared. Since both DPG and ATP are products of glycolysis, this overshoot may represent a temporary overactivation of this pathway to restore DPG levels; a similar overshoot was observed in one of the three studies reported by Valeri & Hirsch (1 969). Lower intracellular levels of adenine and glucose may explain the slower rate of both DPG and ATP regeneration in CPDA-1 red cells when compared to AS-I or AS-3 suspended red cells.

In summary, the studies reported here confirmed a rate of DPG regeneration comparable to that achieved with the previous generation of red cell anticoagulant-preservatives. The extended storage time available with the adenine-based additive solutions did not appear to compromise the extent of 2.3-diphosphoglycerate regeneration although the rate in the first 7 h was slower with CPDA-1. Further study of regeneration kinetics and metabolic intermediates may be beneficial in developing future storage media and predicting return of red cell oxygen transport function post-transfusion.

ACKN 0 WLEDG MENTS

This research was supported by a matching fund grant from the American Red Cross Blood Services, Washington, D.C., and was approved by the Institutional Review Board of Eastern Virginia Medical School and the Medical Executive Committee of Sentara-Norfolk General Hospital, Norfolk, Virginia.

The authors wish to thank the Blood Bank staff of Sentara- Norfolk General Hospital for their help and support: Mary Tessier, R.N., for nursing assistance: Sharon Dunn, Pamela Whitley, Dale Driver, Susan O’Neil, and Irma Caronen for performing the laboratory procedures: and Diane Garvin for typing the manuscript.

REFERENCES

Ashby. W. (1948) Span of life of red blood cells. Blood, 3, 486a. Beattie. K. (1977) In: The Investigation of Typing and Compatibility

Problems Caused by Red Cells, pp. 35-36. American Association of Blood Banks, Washington, D.C.

Beutler, E.. Meul. A. &Wood, L. (1969) Depletion and regeneration of 2,3-diphosphoglyceric acid in stored red cells. Transfusion, 9.109- 114.

Beutler. E. & Wood, L. (1969) The in vivo regeneration of 2.3-

136 A. Heaton, T. Keegan and S. Holme diphosphoglyceric acid (DPG) after transfusion of stored blood. Journal of Laboratory and Clinical Medicine, 74, 300-304.

Brecher, G. & Schneiderman. M. (1950) A time-saving device for the counting of reticulocytes. American Journal of Clinical Pathology.

Chanutin. A. (1967) The effect of the addition of adenine and nucleosides at the beginning of storage on the concentrations of phosphates of human erythrocytes during storage in acid-citrate- dextrose and citrate-phosphate-dextrose. Transfusion, 7, 120- 132.

Cohen, F., Zuelzer, W.W. & Evans, M.M. (1960) Identification of blood group antigens and minor cell populations by the fluorescent antibody method. Blood, 15, 884-900.

Correra. A., Graziano. J.H., Seaman, C. & Pionnelli, S. (1984) Inappropriately low red cell 2,3-diphosphoglycerate and p50 in transfused B-thalassemia. Blood. 63, 803-806.

Dawson, R.B., Jr. Kocholaty. W.F. & Gray, J.L. (1970) Hemoglobin function and 2.3-DPG levels of blood stored at 4'C in ACD and CPD: pH effect. Transfusion, 10, 299-304.

Dern, R.J.. Brewer, G.J. & Wiorkowski, J.J. (1967) Studies on the preservation of human blood. 11. The relationship of erythrocyte adenosine triphosphate levels and other in vitro measures to red cell storageability. Journal of Laboratory and Clinical Medicine, 69,

20,1079-1083.

968-978.

Heaton. A., Miripol, I., Aster. R., et al (1984) Use of ADSOL@ preservation solution for prolonged storage of low viscosity AS-1 red blood cells. British Journal of Haematology, 57, 467-478.

Herve. P., Lamy. B., Peters, A., et al (1980) Preservation of human erythrocytes in the liquid state: biological results with a new medium. Vox Sanguinis. 39, 195-204.

Oski. F.A. (1970) Normal red cell metabolism and the influences of age and the plasma environment. Blood Cells as a Tissue (ed. by W. L. Holmes). p. 99. Plenum Press, New York.

Segal. LH. (1976) In: Biochemical Calculations, p. 245. John Wiley & Sons, New York.

Sugita, Y. & Simon, E.R. (1 965) The mechanism of action of adenine in red cell preservation. Journal of Clinical Investigation, 44. 629- 642.

Torrance, J., Jacobs, P.. Restrepo. A,. Eschbach. J.. Lenfant, C. & Finch. C.A. (1970) Intraerythrocytic adaptation to anemia. New England Journal of Medicine, 283. 165.

Valeri, C.R. & Hirsch, N.M. (1969) Restoration in vivo of erythrocyte adenosine triphosphate, 2,3-diphosphoglycerate, potassium ion, and sodium ion concentrations following the transfusion of acid- citrate-dextrose-stored human red blood cells. Journal of Labora- tory and Clinical Medicine, 73. 722-733.