Microprocessor – controlled vs . “ dump – freezing ” platelet and lymphocyte cryopreservation : a quantitative and qualitative comparative study

Background/Aim. Thermodynamical and cryobiological parameters responsible for cell damages during cryopreservation (cryoinjuries) have not yet been completely explained. Thus, freezing procedures should be revised, exactly optimized to obtain an enhanced structural and functional recovery of frozen–thawed cells. The aim of this study was to compare microprocessor–controlled (controlled–rate) with the compensation of the released fusion heat and “dump–freezing” (uncontrolled–rate) of the platelet and lymphocyte cryopreservation efficacy. Methods. Platelet quantitative recovery (post–thaw vs. unfrozen cell count), viability (using hypotonic shock response – HSR), morphological score (PMS), ultrastructural (electron microscopy) properties and expression of different surface antigens were investigated. In lymphocyte setting, cell recovery and viability (using trypan blue exclusion test) as well as functionality (by plant mitogens) were determined. Controlled–rate freezing and uncontrolled–rate cryopreservation were combined with 6% (platelets) and 10% (lymphocytes) dimethyl sulfoxide (DMSO). Results. Platelet recovery and functionality were superior in the controlled–rate system. The majority of surface antigen expression was reduced in both freezing groups vs. unfrozen cells, but GP140/CD62p was significantly higher in controlled–rate vs. uncontrolled–rate setting. Controlled– rate freezing resulted with better lymphocyte recovery and viability (trypan blue–negative cell percentage). In mitogen–induced lymphocyte proliferative response no significant intergroup difference (controlled–rate vs. uncontrolled–rate) were found. Conclusion. The data obtained in this study showned the dependence of cell response on the cryopreservation type. Controlled–rate freezing provided a superior platelet quantitative and functional recovery. Lymphocyte recovery and viability were better in the controlled–rate group, although only a minor intergroup difference for cell proliferative response was obtained.


Introduction
Current increasing use of different cell-mediated therapeutic approaches has resulted in increasing needs for both blood cells and operating procedures in order to minimize cell damages during their collection, processing and storage in liquid or frozen state.Freezing of isolated cells by various cryopreservation systems has been investigated for many years as a method of long-term storage at cryogenic (subzero) temperatures.Current freezing techniques are developed from the initial report of the cryopreservation of bovine sperm 1 .Cryothermal damages (cryoinjuries) may be the result of extensive cell dehydration ("solution effect") or intracellular ice crystallization ("mechanical damage") and they result in cell destruction [2][3][4][5][6][7][8] .Different freezing procedures (modifications of the Polge-Smith-Parkers method) are in routine use for hematopoietic stem cell and progenitor, as well as mature blood cell cryopreservation.The ability of human stem cells to survive a freezing-thawing procedure results in marrow repopulation and hematopoietic reconstitution after transplantation [9][10][11][12][13][14][15][16][17][18][19][20][21] .Nowadays numerous freezing techniques are used in the practice of platelet [22][23][24][25][26][27][28][29][30][31] and lymphocyte [32][33][34][35][36] cryopreservation.However, the optimal cryopreservation strategy with the greatest quantitative and qualitative cell recovery is still unsolved.
In this study, the influence of the different cryopreservation approaches on the platelet and lymphocyte quantitative and functional recover was compared.It is proposed that the use of controlled-rate vs. uncontrolled-rate ("dump-freezing") procedure will provide a superior cell recovery.Cryopreservation was carried out using our own controlled-rate systems designated for lymphocyte and platelet cryopreservation, as well as by a standard "dump-freezing" technique under optimized dymethyl sulfoxide (DMSO) conditions.

Platelet and lymphocyte collection and testing
The study included 70 whole blood units collected into CPD SAGM plastic bags (Terumo, Japan), designated for platelet (n = 35) or lymphocyte (n = 35) harvest.The primary buffy-coat (BC-I) was obtained by centrifugation (3 890 × g for 10 min) (Hettich, Germany) and processing of whole blood (450±45 ml) collected from donors in steady-state hematopoiesis into a quadruple bag system (Terumo, Japan).Secondary buffy-coat (BC-II) and buffycoat derived platelet concentrates (BC-PC) was prepared by centrifugation (377 × g for 5 min) and processing from BC-I 12,28 .Lymphocytes were isolated from BC-II by centrifugation (320 × g for 35 min) using Ficoll-Paque (Pharmacia Fine Chemicals, Sweden) as a gradient agent.BC-PC samples (2 ml) were dropped by an automatic micropipette into thermo-stable plastic tubes for testing (control group) or cryopreservation.Lymphocytes resuspended in RPMI 1640 culture medium (Serva Feinbiochemica, USA) were taken in the same quantities (2 ml) into plastic tubes for in vitro investigations (control group) or cryopreservation 28,31,34 .
Lymphocyte count in the unfrozen (control) and frozen/thawed samples was quantified manually by a Neubauer chamber.Cell viability was investigated by the trypan blue exclusion test.Based on the quantity and viability of thawed lymphocytes, post-cryopreservation total trypan blue-negative lymphocyte recovery (TBNCR) was determined.To assess functional statement of cryopreserved lymphocytes 0.5 × 10 6 cells were cultivated with different concentration of plant mitogens, phytohemaglutinin (PHA) (Sigma, USA) and concanavalin-A (Con-A) (Sigma, USA), in complete RPMI 1640 culture medium in 96-well microplates (Flow, Irvine, UK).Within the last 18 h cells were pulsed with 3 H-thymidine (1 μCi/well) (Amersham, UK).Radioactivity was determined on a beta scintillation counter (Beckman, Germany).Data are expressed as counts per minute (CPM) ± SE and the stimulation index (SI) [33][34][35][36][37] .

Cell cryopreservation
Cell samples (2 ml) were placed into plastic tubes for cryopreservation.Then, an equal volume (2 ml) of DMSO (Sigma, USA), diluted in autologous plasma (platelet setting) or RPMI 1640 culture medium (lymphocyte setting) was added to cell suspensions.DMSO in 6% (for platelets) and 10% final concentration (for lymphocytes) with both controlled-rate and uncontrolled-rate freezing were combined.Thus, four cryopreservation protocols: platelet and lymphocyte controlled-rate (Con−Plt and Con−Ly) and uncontrolled-rate (Unc−Plt and Unc−Lymphocyte) were investigated.
The uncontrolled-rate freezing without the balanced cooling rate (around 1-2 °C/min) was performed after the equilibration period by simple placing of the cell/cryoprotectant suspension into a mechanical freezer at -90±5 °C [19][20][21] .Our own controlled-rate procedures for platelets 28 and lymphocytes 12  min; IV = -1 °C/min, for 35 min and V = -5 °C, for 5 min.During the phase transition period (liquid to solid state) from -5 °C to -15 °C (platelet setting) and from 0 °C to -10 °C (lymphocyte setting) an intensified (double over) cooling rate (2 °C/min) was used due to the compensation of the released fusion heat 10,28 .
After the completion of controlled-rate freezing procedures, cells were placed into a mechanical freezer (Eurospital, Italy) at the temperature of -90±5 °C till thawing (one to tree months) and the structural and functional investigations that followed.The frozen cell samples were thawed rapidly in a water bath at 37±3 °C.

Statistical analysis
The results were expressed as a mean value ± standard deviation (SD) or mean value ± standard error (SE).Proliferative response of lymphocytes after controlled-rate vs. uncontrolled-rate freezing was compared by the General Linear Model method and Paired Samples t-test.The results obtained before and after freezing of platelets were compared by Student's t-test, using the Original PC Program.The differences were considered as significant at p < 0.05.

Results
The results related to platelet and lymphocyte recovery and viability obtained in this study are presented in Table 1.
The intergroup (controlled-rate vs. uncontrolled-rate) differences showed that there was better platelet and lym-phocyte quantitative recovery (p < 0.05) in the controlledrate groups.Uncontrolled-rate freezing caused higher platelet HSR reduction with the significant intergroup differences (p < 0.01).Regarding cell viability, there were no significant differences neither between the cryopreserved vs. control group, nor between lymphocytes frozen by both freezing methods.On the contrary, the TBNCR values for lymphocytes were significantly (p < 0.05) higher in the controlled-rate setting.
The PMS values were rapidly reduced after cryopreservation, regardless the applied procedure.For controlled-rate cryopreserved platelets, PMS was 81.8±2.8 %, while for those frozen by the uncontrolled-rate technique PMS was 75.7±3.9% (Table 2).
The analysis of PMS recovery confirmed that the intergroup differences were significant (p < 0.05) in favor of the controlled-rate freezing.
The results of platelet surface marker investigations before and after cryopreservation are presented in Table 3.
GPIb/CD42b expression was reduced, in both freezing groups as compared to the control one, with no significant intergroup (controlled-rate vs. uncontrolled-rate) differences.On the contrary, noticable intergroup differences (p < 0.05) for GP140/CD62p expression were observed, suggesting that controlled-rate freezing provided lower platelet activation.
GP53/CD63 showed an increased expression as compared to the control group, but with no significant intergroup   differences.GPIIb-IIIa/CD41 values were lower than in the control, but also without the significant intergroup differences.
Figure 1a shows the normal unfrozen discoid platelet (control) with the typical ultrastructural properties.The electron microscopy appearance of platelets cryopreserved by controlled-rate and uncontrolled-rate freezing are also presented in Figure 1b-e.
Besides discoid and spheric shapes, there were several dendritic platelets, in the controlled-rate freezing group (Figure 1b and 1c).The discoid platelets had maintained intact microtubules and an open canalicular system with a minimal damage of the membrane.Platelets frozen by the uncontrolled-rate procedure (Figures 1d and 1e) showed the membrane damages and altered ultrastructural properties (dendritic or balloon shapes).In addition, these platelets regularly had an unclear edge with numerous large pseudopodes.Unc−Plt = uncontrolled-rate freezing with 6% DMSO; † Cryopreserved vs. unfrozen (control) (p < 0.05); ‡Intergroup (controlled-rate vs. uncontrolled-rate) differences (p < 0.05).

Fig. 1b -The ultrastructural (electron microscopy) properties of cryopreserved platelets
Controlled-rate cryopreserved platelets.Finally, Table 4 shows the mitogen (PHA and Con-A) induced lymphocyte proliferative response in controlled-rate and uncontrolled-rate setting.
The analysis of the obtained results of the lymphocyte proliferative response demonstrated that the cryopreservation resulted in no significant intergroup differences, regardless the used freezing procedure.

Discussion
Our earlier cryoinvestigations 10,12,28 pointed out the importance of the kinetics of the programmed cooling for the total nucleated cell (TNC), lymphocyte and platelet cryopreservation.In this study, we confirmed that the use of the controlled-rate method had significantly influenced the protection of quantity and functionality of frozen/thawed platelets.The platelet postthaw recovery was exactly 91.0±5.5%, which was better than in other studies (62 to 86%) reported 22,23,37,38 .
The results obtained in this research showed that the controlled-rate freezing yielded the improved platelet viabil-ity, as projected across HSR.Although in this study the lower HSR values were observed for both freezing procedures (compared to the control one), the 32% reduction in the controlled-rate setting is in accordance with other reports.However, HSR value depends on some additional parameters, such as cell platelet age, cell membrane integrity, type of platelet shapes, etc 23,31,39 .Finally, we believe that HSR of cryopreserved platelets is not always a result of the unsuccessful response of all cells, but could rather represent an unequal response of different platelet shapes.
The examination of PMS also confirmed clearly that controlled-rate freezing was more efficient than the uncontrolled-rate technique.Better PMS values obtained in the controlled-rate setting were primarily due to a greater proportion of platelets of disc shapes (around 70% in comparison with the control).These data generally agree with those reported in the literature on PMS values after freezing-thawing procedure.It is generally accepted that the high-quality in vivo survival of cells can be expected after transfusion if 50% or more of platelets keep their disc shape after cryopreservation [39][40][41] .
The investigation of ultrastructural properties of platelets confirmed the advantage of the controlled-rate freezing, since it revealed that the discoid platelets cryopreserved by this protocol had kept intact microtubules and an open canalicular system, with the minimal membrane changes.The spheric shaped platelets showed the lower electron density of cytoplasm, gradually changed the interior structure comparing to other shapes, which was manifested by the marginal location of granules and other cytoplasmatic organelles.This finding is in accordance with our previous platelet cryoinvestigation 28 .On the contrary, the samples cryopreserved by the uncontrolled-rate technique showed an abundance of the damaged platelets and cell detritus, along with the lower percentage of platelets with a partially protected membrane integrity.The occurrence of the regular dendritic and balloon shapes, as well as rare spheric platelets indicated reduced morphological properties and the subsequent functionality of thawed cells.In the present study, GPIIb-IIIa/CD41 platelet surface marker was investigated.In addition, platelet activation was assessed by measuring the expression of GPIb/CD42b, GP53/CD63 and GP140/CD62p (P-selectin).The last one is expressed on the cell surface after the release from α-granules.
It is generally accepted that the cold-induced posttransfusion platelet clearance with following compromised clinical effectiveness should be mediated by GPIb/CD42b reorganization into "clusters" [42][43][44][45][46][47] .Our preliminary cryoinvestigation (29) of cold-induced GPIb/CD42b alterations was in agreement with these studies, particularly when uncontrolled-rate freezing was performed.We proposed that the reduction of the GPIb/CD42b expression in the present study (from 11.1 to 14.0%) might be partially caused by the GPIbα/CD42b "relocation".However, these values are markedly better than those reported by Lozano (reduction > 50%) 44 .The mentioned "relocation" could probably make surface antigens less accessible to monoclonal antibodies, resulting in the obtained decrease of GPIb/CD42b expression the of cryopreserved vs. the control platelets.However, Hoffmeister et al. (personal communication) did not find a reduced GPIbα/CD42b expression following cooling using the monoclonal antibody for platelets, so they concluded that clustering did not lead to "pseudo hyporeactivity".In this context, GPIb/CD42b expression decrease should be mediated by other mechanisms such as proteolysis by plasmin, thrombin, neutrophil elastase or the activation of the cell "death machinery", etc 29,46 .The results obtained in our study are comparable with those published by Barnard et al., who also found a reduced GPIb/CD42b expression after cryopreservation 46 .They proposed that decreased antigen expression could be associated with GPIb/CD42b proteolysis, mediated by calcium-activated protease, or calpain.
It is generally accepted that the cold-induced GP140/CD62p expression less than 40% could be an adequate threshold, because higher expression values of GP140/CD62p have been shown to result in platelet in vivo recovery of less than 50% 48 ).As shown in Table 3, freezing promoted a raise in the number of platelets expressing GP140/CD62p antigen.The percentage of platelets expressing GP140/CD62p in the controlled-rate group was significantly lower than in the uncontrolled-rate sample.The elevated GP140/CD62p values for cryopreserved cells pointed to the platelet activation during the freezing/thawing procedure.
The expression of GP140/CD62p was associated with the clearance of platelets by the reticuloendothelial system 48 .More than 50% of platelets stored at 22 °C expressed GP140/CD62p on their surface after 5 days, which indicated that the cryopreserved cells had the superior retention of intact granules, which correlated with their in vivo functionality 49 .With respect to GP53/CD63 antigen (activation marker of the externalization of the contents from lysosomes), a considerable increase in the expression of this antigen after freezing-thawing process was observed.However, this increase in the GP53/CD63 expression was not as high as for GP140/CD62p antigen.This finding might reflect the more intensive activator stimulus necessary for the release of lysosomes compared to that of α-granules 50 .
In this study, the GPIIb-IIIa/CD41 (the main platelet aggregation protein) examination showed the decrease in cryopreserved platelets in both groups.However, the reduction of GPIIb-IIIa/CD41 was not significant after controlled-rate freezing vs. the control group (opposite to uncontrolled-rate freezing), which is in agreement with data from the literature 51,52 .GPIV/CD36 is an accelerator of the interaction between platelets and collagen and it is a mediator of the platelet adhesion to subendothelial surfaces 53 .Cryopreserved platelets in the present investigation obtained a significant reduction of the expression of GPIV/CD36 as compared to unfrozen platelets, but without intergroup differences.These data are in accordance with data from the literature 44 .
For a successful cryopreservation not only the use of appropriate freezing strategy, but also the selection of a cryoprotectant in the optimized concentration is required.For platelet and lymphocyte freezing, DMSO and/or hydroxyethyl starch (HES) dissolved in a cell-specific "freezing medium" is commonly used, but with different concentrations of cryoprotectant 54,55 .In the present study, 6% DMSO in autologous plasma (platelet setting), and 10% DMSO in RPMI 1640 (lymphocyte setting) were used.Our earlier cryoinvestigation showed that the TNC and the committed progenitor recovery was better when the lower (5%) DMSO and controlled-rate freezing were used.In contrast, pluripotent Makrow Repopulating Ability (MRA) cells could only be well cryopreserved in the presence of 10% DMSO 10 .These data implied a different "cryobiological request" of MRA cells in comparison to the committed progenitors or TNCs.However, as recently presented, the postthaw cell quality is better when rHu-DNase, membrane stabilizers or bioantioxidants are added to the Conventional Freezing Medium (CFM) [56][57][58] .Accordingly, it is possible that the use of these additives is more successful than the reduction of DMSO concentration.We also showed that the differences in the cell recovery and functionality rates were not related to the changes in the total number of frozen/thawed cells 10,11 .
In this research, the obtained lymphocyte recovery was significantly better in the controlled-rate vs. uncontrolledrate settings.These results are in agreement with our earlier findings for TNCs and lymphocytes 11,35 , and they are also in accordance with the data from other authors 36,37,59 .On the other hand, we obtained no significant differences in mitogen induced proliferation of lymphocytes after the controlled-rate vs. uncontrolled-rate procedure.Proliferative response to mitogens or microbial antigens is often used as a measure of lymphocyte functionality.Weinberg et al. 60 showed that the results of functional assays on cryopreserved peripheral blood MNCs are associated with cell viability.Although the obtained TBNCR values were markedly higher in the controlled-rate group, the proportion of viable cells was high (≈ 90%) for both procedures (Table 1) which can explain a similar proliferative response of lymphocytes after different freezing methods.

Conclusion
In conclusion, the results presented here point undoubtedly out the controlled-rate system (with the compensated fusion heat) which resulted in a superior platelet recovery of the tested parameters.A simultaneous analysis of platelet surface antigens and cell micro-integrity markers could predict certain in vivo properties of the cooled, especially cryo-preserved platelets, however these postulates should be confirmed.The lymphocyte recovery and TBNCR values were obviously lower in the uncontrolled-rate setting.The obtained high-level postthaw cell viability can explain their similar proliferative response regardless the used freezing.Further researches in various blood-derived cell settings are recommended to define the optimal cell-specific cryopreservation strategy with the best cell recovery.