J Clin Pathol 1981 ;34:71-75

Cryopreservation of : an in-vitro comparison of four methods

M ANN TAYLOR From the South Western Regional Transfusion Centre, Bristol BSIO SND, UK

SUMMARY Platelets were cryopreserved in four different cryoprotectives-two intracellular (dimethylsulphoxide, glycerol) and two extracellular (hydroxyethyl starch, ). The platelets were evaluated according to their yields, hypotonic response, and serotonin uptake, which are useful parameters for establishing optimum storage conditions. Hydroxyethyl starch and dextran were found to be poor cryoprotectives while 50% dimethylsulphoxide was the most suitable.

Blood transfusion centres are under considerable preserved platelets as these tests are excellent for pressure to produce an ever-increasing number of evaluating preserved platelets.9-13 concentrates for treating a wide variety of clinical conditions. The maximum shelf-life of Methods platelet concentrates stored at 40C or 220C is 72 hours, which makes inventory control very difficult. PLATELET CONCENTRATE PREPARATIONS The successful long-term cryopreservation of plate- Platelet-rich plasma (PRP) was prepared from fresh lets, prepared on a large scale either from single or whole blood, collected in ACD anticoagulant, within pooled donations or from donations collected from six hours of collection by centrituging at 2900 rpm cell separators, would be a considerable advance. (2260 g) for 90 s at 220C in a Damon PR6000 An HLA-typed platelet bank would also be invalu- centrifugewith interiorwind-shielded, swing-outhead. able for providing platelets for patients requiring The PRP was acidified by adding 35 ml ACD HLA compatible platelets. anticoagulant and then agitated to ensure mixing. Much work has already been carried out towards The platelet concentrate (PC) was prepared by establishing the optimum method for cryopreserva- centrifuging the acidified PRP at 3500 rpm (3000 g) tion of platelets. Dimethylsulphoxide (DMSO) has for 20 min at 22°C. The platelet button was resus- been shown by many workers'-4 to be a suitable pended immediately in 20 ml autologous plasma and cryoprotective agent, while others5 6 have shown the remaining acidified platelet-poor plasma (PPP) glycerol- to be successful. With these methods was retained for use in further processing of the the thawed platelets need to be washed to remove the platelets. cryoprotective before transfusion. Choudhury and Gunstone,7 however, described a method for freezing CRYOPRESERVATION platelets using hydroxyethyl starch (HES) as a The platelets were frozen in polyolefin platelet cryoprotective agent. The platelets could be trans- freezing bags (Ucar blood freezing bags, Union fused after thawing without prior washing. Elimina- Carbide). The bags were placed in an aluminium ting the need to wash platelets before transfusion has container for both freezing and storage. A Union many advantages, but Raymond et C£4.8 found that Carbide controlled rate freezer was used to obtain recovery of platelets preserved in 5Y% HES and the required freezing rates and the units were cooled 5 % dextran was poor. to - 80°C. Then they were transferred, in the In this present study platelet cryopreservation was aluminium containers, to the vapour phase of liquid carried out using two intracellular (DMSO, glycerol) nitrogen, where they were stored until required for and two extracellular (HES, dextran) cryoprotective reconstitution. agents. Serotonin (5HT) uptake and hypotonic shock The platelet freezing methods were as follows. response (HSR) have been carried out on the cryo- (1) Cryopreservation in 4% (w/v) HES was carried out by the method of Choudhury and Gunstone.7 Accepted for publication 9 June 1980 Two volumes of 6 % HES in normal saline (Plasma- 71 72 Taylor steril MW 450 000, Fresenins, Bad Homburg, W. IN-VITRO TESTING Germany) were added to one volume of platelet Platelet counts were determined on the original concentrate and the platelets were frozen at 1 C/min. freshly prepared platelet concentrates and on the The platelets were ready for testing immediately reconstituted cryopreserved platelets using a Techni- after thawing in a 370C water bath. con Autocounter (dark ground illumination). From (2) Dextran (MW 10 000, Sigma Co Ltd) was these counts and the PC volumes the percentage mixed with phosphate buffered saline to make a final yield of platelets could be determined. concentration of 15% (w/v). After preparing the The method used for serotonin uptake was based platelet concentrate all the acidified plasma was on the method of Weiss et al.'7 as described by removed and the platelets were suspended in 20 ml Tandy and Taylor.13 Tandy and Taylor'3 showed 15 % dextran and then frozen at 50C/min to - 80'C. that for fresh platelets the concentration of 14C- After cryopreservation the platelets were thawed at serotonin was in excess when 20 min incubation at 370C and were ready for testing. 37°C and a platelet count of /- 200 x 109/1 were (3) The method used for cryopreserving platelets used. in glycerol-sugar was that of Herve et al.6 14 Three The method for HSR was based on the method of different glycerol-sugar concentrations in isotonic Valeri et al.10 as described by Tandy and Taylor.13 saline were used: 3 % glycerol (v/v) 3 % glucose The results were calculated relative to fresh platelets. (w/v), 5% glycerol (v/v) 4% (w/v), and 6% glycerol (v/v) 3% glucose (w/v). The platelets were Results frozen at 10C/min to -3C and then at 350C/min to - 80'C. After thawing autologous PPP was added The yields of platelets obtained with the four cryo- to the platelets. They were spun to pellet the plate- protective reagents are shown in Table 1. Platelets lets, all the plasma was removed, and the platelets cryopreserved in DMSO or glycerol had higher were resuspended in autologous PPP. yields than those in HES. For the platelets cryo- (4) Platelets were cryopreserved in 5 % DMSO by preserved in dextran the count recoveries on thawing a method similar to those published.415 DMSO 10% were quite high. However, when the platelets were in autologous PPP was added slowly and with washed to remove the dextran and resuspended in agitation to the PC. The platelets in 5% DMSO autologous PPP the platelet count fell considerably. were frozen at 1°C/min. They were thawed in a This large reduction in numbers during washing was 37°C water bath and about 100 ml autologous, not observed with the other cryoprotectives (unpub- acidified PPP was added slowly and with gentle lished results). agitation. This PRP was spun to pellet the platelets, Serotonin uptake was measured over one hour all the plasma was removed and the platelets were (Table 2). For the platelets frozen in DMSO, resuspended in 20 ml autologous acidified plasma. glycerol, or dextran a platelet count of about Platelets cryopreserved in 5 % DMSO but frozen by 200 x 109/1 was used in the assay. Since the count placing them immediately in the vapour phase of recovery was low for platelets preserved in 4 % HES liquid nitrogen were prepared and resuspended as the average count used in this assay was 100 x 109/1. above. Residual DMSO levels were measured by None of the platelet samples recovered after cryo- gas liquid chromatography using a Pye GCD preservation took up so much serotonin as fresh machine. The stationary phase was 10% polypropy- platelets. The highest uptake obtained was for plate- lene glycol adipate with a column temperature of lets cryopreserved in DMSO. 170°C. DMSO in plasma was extracted with acetone Relative HSR determinations are shown in the to precipitate the protein before being added to the Figure. Also shown are the relative HSRs for fresh column.16 platelets incubated at room temperature for one hour

Table 1 Mean (± SD) platelet yields with four different cryoprotective agents 4% HES 15y% Dextran Glycerol-sugar 5% DMSO 3% Glycerol 5% Glycerol 6% Glycerol PC/min Vapour phase 3% glucose 4% sorbitol 3 % glucose Yield(%) 11 57 ± 11 30 ± 13 32 ± 10 51 ± 15 54 ± 20 42 ± 9 21 ± 9t Number of units tested- for all assays 20 21 20 14 7 18 18

*Yield immediately on thawing. tYMeld after washing and resuspending the platelets. Cryopreservation ofplatelets: an in-vitro comparison offour methods 73 Table 2 Mean (± SD) serotonin uptake over one hour by platelets cryopreserved infour different cryoprotective agents 4% HES* 15% Dextraig Glycerol-sugar 5% DMSO Fresht platelets 3% Glycerol 5% Glycerol 6% Glycerol I'Clmin Vapour phase (n -1)= *lc-E 3% glucose 4%sorbitol 3% glucose 0* O-0-- 0 100 100 100 100 100 100 100 100 o~c X 10 100 94 4 95 5 84 7 87 8 62 26 72 29 34-5 + 6-4 &. E~ 20 100 92 4 44 23 61 15 16 ± 3-2 iE 0 30 85 6 65 11 73 14 ea.9 60 100 87 7 69 19 52 15 62 15 28 17 28 16

*Platelet counts for 5HT assay 200 x 109/1 except 4% HES where platelet count 100 x 109/1. tFrom Tandy and Taylor.13 in the presence of cryoprotective, but without Of the four cryoprotective agents tested DMSO freezing, and then washed and resuspended. The appears the most suitable. There is little difference HSR was not carried out for platelets cryopreserved between the percentage yield (Table 1) and the sero- in 6 % glycerol and 3 % glucose and dextran as these tonin uptake (Table 2) for the platelets cryopreserved compounds substantially reduced the HSR even in 5% DMSO with freezing either at a controlled before cryopreservation (Figure). The frozen plate- rate of r°C/min or by placing directly into the lets cryopreserved in 5 % DMSO with a freezing rate vapour phase of liquid nitrogen. However, the of 1 C/min gave the highest response, whereas relative HSR is noticeably higher for platelets frozen freezing with 4% HES gave no response at all. at 1VC/min (Figure). Gas liquid chromatography (GLC) was carried out to determine the residual DMSO concentration in cryopreserved platelets, Control | jpxtr~ r Glycerol sugar .5%DMS 100 recovered and prepared for transfusion. Twelve units were tested and a mean value of 113 mg (± 29 mg, 1 SD) DMSO per unit of platelets was found, which is about 5% of the original DMSO. 50 Discussion The results of this present study have indicated that

0 tL11h3%<3%9Y+4%6Yt+% hydroxyethylr3m \4_ starch and dextran are poor cryo- plaltets Glicose Stitbc Gkcose phase protective agents for the long-term storage of plate- lets. Choudhury and Gunstone7 reported satisfactory NT not tested Platelets not frozen results using HES but their results are also at variance Platelets frozen with Odink and Sprokholt.19 Different molecular Relative hypotonic shock responses. weight HESs (150 000-450 000) and faster freezing rates (from 40C/min to direct immersion in liquid Altering the glycerol-sugar concentrations had nitrogen) were tried in this present study but they little effect on the serotonin uptake of the cryo- also gave poor results (unpublished results). In the preserved platelets. The effect of glycerol on HSR present work the methods of testing the platelets in increases with increase in glycerol concentration vitro differed from those used by Choudhury and (Figure) such that there is little HSR for platelets Gunstone.7 Platelet factor 3 availability was not incubated with 6% glycerol and 3% glucose. The determined because it was considered unsuitable for relative HSR for platelets cryopreserved in 5% monitoring platelet production and storage.13 glycerol and 4% glucose was marginally better than Choudhury and Gunstone counted platelets using for platelets in 3Y% glycerol and 3% glucose. While the Thrombocounter (Coulter counter), which there was little difference in the percentage yield of counts all particles within a chosen size range platelets preserved at these concentrations, the best whether they are intact platelets or not. In this work yeilds were obtained when 6% glycerol and 3% and previously'3 a Technicon system was used which glucose was used (Table 1). Other glycerol-sugar is based on dark ground illumination of platelets concentrations (in the range 5%-10% glycerol- and not size. Ammonia, a dispersant, is used in the glucose and 3 %-6 % glycerol-sorbitol) were tried system to clear cell debris, thus making the counts but the results were not improved (unpublished more accurate. Vecchione and Valeri20 reported a results). difference in counts obtained when the Technicon

3* 74 Taylor and Coulter counters were used to count cryo- relative HSR of 44% reported by Kim and Baldini'5 preserved platelets. The platelet counts using the and 36% reported by Odink and Brand.1" The sero- Technicon system were about 65 % of those obtained tonin uptake compared with fresh, control platelets with the Coulter when resuspended platelets which was 38 % after two hours15 and 58 % after 10 minutes had been cryopreserved in 5 % DMSO were counted. in this report. Odink and Brand" reported a relative A more sensitive method for HSR was also used in serotonin uptake velocity of 23%. Valeri et al.10 this present study and previously.13 reported a mean total in-vitro recovery of 43 ± 100 Platelets cryopreserved in glycerol gave lower for platelet cryopreservation in 5% DMSO. yields and poor in-vitro viability compared with those Many workers have transfused platelets which cryopreserved in DMSO. Attempts were made to have been cryopreserved in 5% DMSO using a freeze platelets in glycerol at 30°C/min5 but a controlled rate of freezing. Satisfactory in-vivo consistent freezing curve could not be obtained. results have been obtained. Handin and Valeril Again, the method of counting platelets in this showed that in normal volunteers the recovery of present study was different from those of other frozen platelets was 47% ± 3% (70% of the value workers using glycerol.5 6 14 21 Herve et al.'4 used the for fresh platelets) with a lifespan of 8-8 ± 0 3 days Coulter while the others counted the platelets in a (8-9 ± 0 3 days for fresh platelets). Kim and haemocytometer chamber by phase contrast micro- Baldini2 showed that platelets had a mean viability scopy. The platelet recoveries were high, four of the 17 index of 69%, while Odink and Brand" showed units reported by Dayian and Pert2' having recoveries cryopreserved platelets had a transfusion efficiency over 100%. They reported that most ofthe platelet loss of 36%. They both showed that on infusion some occurred during processing and not freezing. Manual cryopreserved platelets have a short lifespan while counting was not carried out in this present study the rest survived almost normally. Results from because of difficulties in identifying the intact transfusion of cryopreserved platelets to patients platelets by phase microscopy in some samples with leukaemia25 and thrombocytopenia34 showed which had been cryopreserved. It has been reported5 21 that the platelets can circulate and function haemo- that platelets cryopreserved in glycerol show a high statically. Schiffer et al.25 transfused with satisfactory uptake of serotonin after 30 min incubation. How- results platelets cryopreserved in 5% DMSO and ever, these workers used a much lower concentration frozen by placing directly into the vapour phase of of serotonin in their test than is used in the present liquid nitrogen, but they did not provide sufficient study and the platelet count of the test samples was data to indicate which freezing rate was preferable. either not specified2' or covered a wide range.5 Platelets frozen by placing directly into a -80'C This raises the possibility that in their experiments freezer in 4% DMS026 had an in-vivo recovery of the serotonin concentration may not have been in 35% and a lifespan of 6-4 days (65% and 8 days for excess for the total time of incubation with fresh fresh platelets), while with 6Y% DMS18 a 450% platelets. The serotonin concentration used in this recovery and a lifespan of 8-5 days was shown. work and previously'3 was chosen to ensure serotonin In the first experiments with transfusion of was in excess. Neither Herve et al.614 nor Dayian DMSO-preserved platelets the DMSO was not and coworkers5 21 used the HSR test. removed from the thawed platelets before trans- Kim and Baldini9 have shown that glycerol has a fusion and there were side effects from the infusion.27 marked effect on HSR and the effect increases with Handin and Valeri' showed that platelet recovery increasing glycerol concentration. Bakry and McGill22 in vivo was adversely affected by the presence of reported that the ultrastructural integrity appeared DMSO at the time of transfusion. In addition, since to be maintained in platelets incubated in 3% glycerol. the long-term toxicity of large volumes of intra- They were discoid shaped, contained normal num- venous DMSO is uncertain, the platelets are now bers of granules, and their surface membranes washed to remove the DMSO before infusion. In appeared intact. However, at 6 % glycerol and the present work only 5% of the DMSO remained. higher the number and electron density of the Kim and Baldini'6 have also reported low concentra- granules were significantly reduced. Meryman and tions of residual DMSO-a 20-ml PC contained Burton23 and Kahn24 state that other workers have 45 mg (21-78 mg) of residual DMSO after a single found cryopreservation of platelets using glycerol to washing with 120-150 ml PPP. be unsuccessful. Kim and Baldini9 showed that 5% DMSO had a In-vitro testing of platelets cryopreserved in 5% definite though not large influence on the reversal DMSO and frozen at 1°C/min has been carried out reaction, while 10% and 15% had a greater effect. by some laboratories, although the exact methods Odink and Sprokholt28 showed that incubation with used for each test has varied. The relative HSR of 5% DMSO had more effect on HSR than on the 38 % reported here compares favourably with the serotonin uptake velocity. Murphy et al.3 reported Cryopreservation ofplatelets: an in-vitro comparison offour methods 75 that as the DMSO concentration increased the blood platelets. The serotonin uptake and hypotonic was shock response as in-vitro viability test. Thromb Haemo- theoretical increase in cryoprotective effect stas 1976;32 :405-16. apparently offset by the agents inherent toxicity 13 Tandy NP, Taylor MA. Platelet concentrates for trans- and they concluded that 5% DMSO was optimal. fusion: control of production and storage. Med Lab Sci From the results presented here DMSO is clearly 1980;37:127-36. 14 Herve P, Masse P, Porton G, et al. Methode de cryo- the most suitable cryoprotective even though many conservation des plaquettes a 196°C. Rev Fr Transfus platelets are lost during the process. Cryopreserva- Immunohematol 1977;20:603-1 5. tion of platelets using 5% DMSO may be of great 15 Kim BK, Baldini MG. Biochemistry, function and haemo- use when fresh platelets are not available and when static effectiveness of frozen human platelets. Proc Soc cryo- Exp Biol Med 1974;145:830-5. compatible platelets are required. Platelets 16 Kim BK, Baldini MG. Preservation of viable platelets by preserved in 5 % DMSO in this laboratory are being freezing. Effect of plastic containers. Proc Soc Exp Biol transfused and a further report will follow. Med 1973;142:345-50. 17 Weiss HJ, Tschopp TB, Rogers J, Brand H. Studies of platelet 5-hydroxytryptamine (serotonin) in storage pool I thank Mr J Allen for the gas liquid chromatography disease and albinism. J Clin Invest 1974;54:421-32. work and Dr ID Fraser for his aid in preparing this 18 Valeri CR, Feingold H, Marchionni LD. A simple method paper. for freezing human platelets using 6 % dimethyl- sulphoxide and storage at -80°C. Blood 1974;43:131-6. 19 Odink J, Sprokholt R. Platelet preservation. I. The use of decrease in light absorbance as a screening method in References cryopreservation studies on human platelets. Cryo- biology 1977;14:519-28. Handin RI, Valeri CR. Improved viability of previously 20 Vecchione JJ, Valeri CR. Status report on cryopreserva- frozen platelets. Blood 1972;40:509-1 3. tion of human platelets. Proceedings of the Haemonetics 2 Kim BK, Tanoue K, Baldini MG. Storage of human Research Institute advanced component seminar. Boston, platelets by freezing. Vox Sang 1976;30:401 - 11. Massachusetts, 1979. 3 Murphy S, Sayer SN, Abdou NL, Gardner FH. Platelet 21 Dayian G, Pert JH. A simplified method for freezing preservation by freezing. Use of dimethylsulphoxide as human blood platelets in glycerol-glucose using a cryoprotective agent. Transfusion 1974 ;14:139-44. statically controlled cooling rate device. Transfusion 4 Schiffer CA, Aisner J, Wiernik PH. Clinical experience 1979 ;19 :255-60. with transfusion of cryopreserved platelets. BrJHaematol 22 Bakry M, McGill M. Glycerol effects on human platelet 1976;34:377-85. volume shape and ultrastructure. Cryobiology 1977; Dayian G, Rowe AW. Cryopreservation of human plate- 14:699. lets for transfusion; a glycerol-glucose moderate rate 23 Meryman HT, Burton JL. Cryopreservation of platelets. cooling procedure. Cryobiology 1976;13:1-8. In: The BloodPlatelet in Transfusion Therapy. New York: 6 Herve P, Masse M, Kieffer Y, et al. Cryoconservation des Alan R Liss, 1978:153-65. plaquettes dans l'azote liquide en presence de glycerol- 24 Kahn RA. Discussion. In: The Blood Platelet in Trans- glucose. Colloque de Cryoimmunologie, Dijon, 1976. fusion Therapy. New York: Alan R Liss, 1978:192. INSERM, 62, 163-70. 21 Schiffer CA, Aisner J, Wiernik PH. Frozen autologous 7Choudhury C, Gunstone MJ. Freeze preservation of platelet transfusion for patients with leukaemia. N Engl platelets using hydroxyethyl starch (HES); a prelimin- J Med 1978;299:7-12. ary report. Cryobiology 1978;15:493-501. 26 Kahn RA, Dionne J, Becker G. A procedure for the freeze 8 Raymond SL, Pert JM, Dodds WJ. Evaluation of platelet preservation of platelets. Cryobiology 1975 ;12 :565. cryopreservation technique by isolated per- 27 Djerassi I, Farber G, Roy A, Cavins J. Preparation and fusion. Transfusion 1975;15:219-25. in vivo circulation of human platelets preserved with Kim BK, Baldini MG. The platelet response to hypotonic combined dimethylsulphoxide and dextrose. Transfusion shock. Its value as an indicator of platelet viability after 1966;6:572-6. storage. Transfusion 1974;14:130-8. 28 Odink J, Sprokholt R. Platelet preservation. III. The 10 Valeri CR, Feingold H, Marchionni LD. The relation influence of dimethylsulphoxide and cooling conditions between response to hypotonic stress and the 61Cr on serotonin uptake velocity and response to hypotonic recovery in vivo of preserved platelets. Transfusion stress of human platelets. Thromb Haemostas 1976; 1974;14:331-7. 36:192-9. 11 Odink J, Brand A. Platelet preservation V survival, sero- tonin uptake velocity and response to hypotonic stress Blood of fresh and cryopreserved human platelets. Transfusion Requests for reprints to: MA Taylor, National 1977;17:203-9. Transfusion Service, South Western Regional Trans- 12 Hardeman MR, Heynens Carina JL. Storage of human fusion Centre, Southmead Road, Bristol BS1O 5ND.