http://informahealthcare.com/plt ISSN: 0953-7104 (print), 1369-1635 (electronic) Platelets, Early Online: 1–8 ! 2014 Informa UK Ltd. DOI: 10.3109/09537104.2013.872772

ORIGINAL ARTICLE

Quality assessment of platelets stored in a modified platelet additive solution with trehalose at low temperature (10  C) and in vivo effects on rabbit model of thrombocytopenia Xin Wang, Yahan Fan, Ronghua Shi, Jing Li, & Shuming Zhao

Platelets Downloaded from informahealthcare.com by University of Auckland on 10/16/14 For personal use only.

Department of Blood Transfusion, Southwest Hospital, the Third Military Medical University, Chongqing, P.R. China

Abstract

Keywords

Trehalose is widely used as a cryoprotective reagent to preserve various cells. Platelet additive solution-III (PAS) has been used to maintain platelet function, benefit the virus inactivation, and extend the storage period. PAS with trehalose (PAS-III M þ T) may effectively protect platelets (PLTs) at a relatively low temperature (10  C). The apheresis PLTs from six donors were divided into two groups. Group A was stored in PAS-III M þ T at 10  C as experimental group and group B in plasma at 22  C as control group. The samples were collected on different storage dates, and multiple parameters were determined or investigated for in vitro studies. The in vivo recovery and survival of rabbit PLTs stored in the same conditions, and then labeled with 51Cr were measured and evaluated using a rabbit model of thrombocytopenia. Over 9 days, P-selectin expression increased significantly in a time-dependent manner in both groups (n ¼ 6). The levels of the hypotonic shock reaction and PLT aggregation rate decreased in both groups and were significantly higher in group A than B after 1 day of storage. The lactate dehydrogenase (LDH) release and glucose (GLU) consumption increased similarly, but the levels were significantly lower in group A than B. The pH decreased significantly after 5 days of storage in group B but did not change in group A. After 5 days, the morphology of the PLTs in group B maintained a more normal shape than that of group A. The recovery and survival of PLTs stored in both groups were not significantly different (p40.05). The bacteria growth was not examined out in both groups for up to 5 (group A) and 9 (group B) days. Storage of PLTs in the modified PAS at low temperature was more effective in protecting PLT functions than that of standard storage method and may have the potential to decrease the risk of PLT activation and bacterial contamination.

Cold storage, PAS, platelet, trehalose

Introduction Platelet (PLT) transfusion plays an important role in clinical treatment of the patient and is widely used to treat thrombocytopenia, which is generally caused by chemo/radiotherapy, and occasionally due to massive trauma. Currently, there are two conventional methods used to prepare PLTs for clinical transfusion. One method is to manually or automatically prepare platelet concentrates (PCs) from whole blood. Normally, five to six units of whole blood (400 ml/unit) are needed to prepare one therapeutic dose (2.5  1011) of PCs. The other method is to collect PLTs from a single donor (single-donor PLTs or SDPs) by apheresis. One or two therapeutic doses of PLTs can be collected from a single donor using a blood cell separating apparatus. PLTs prepared by apheresis are highly concentrated and have the advantage of decreasing the chance of exposure to donors’ infectious diseases and the risk of blood-borne diseases by reducing routine leukocytes. SDPs stored in platelet additive

Correspondence: S. Zhao, Department of Blood Transfusion, Southwest Hospital, the Third Military Medical University, Chongqing 400038, P.R. China. Tel: 86 23 68754270. Fax: 86 23 68754270. E-mail: [email protected]

History Received 14 October 2013 Revised 19 November 2013 Accepted 2 December 2013 Published online 5 February 2014

solutions (PASs) with about 30% plasma remained for resuspension can be used for ABO-incompatible PLT transfusions [1]. In addition, the use of PASs can conserve plasma resources and thereby decrease plasma-related transfusion reactions and disease transmission [2, 3]. On the contrary, PCs face relatively high risk of blood-borne viruses due to the blood process of more than one donor. Moreover, there are at least three times the numbers of chance for bacterial contamination of the pools during manufacturing PCs as opposed to SDPs and the number of PCs with positive bacterial cultures is at least two times that of SDPs [4]. In recent years, with the wide use of automatic blood cell separators, the use of SDPs has been increasing in clinical practice in many countries. Meanwhile, the quality of manually whole blood-derived PCs has also improved with the emergence of storage bags, additive solutions and automatic in vitro whole blood cell separators, and reached to clinically equivalent to a SDPs [3, 4]. Currently, PLTs are conventionally stored at 22  C on a shaker with continuous gentle agitation for the optimal maintenance of PLT viability and function. When PLTs are stored at a low temperature, their function can be affected, so they may not circulate for an acceptable period of time in the recipient. Cold storage does cause PLTs to lose their discoid shape, but this does not appear to adversely affect either PLT function or survival.

Platelets Downloaded from informahealthcare.com by University of Auckland on 10/16/14 For personal use only.

2

X. Wang et al.

Platelets, Early Online: 1–8

Past in vivo studies provide evidence that at PLTs cryopreserved can, in fact, function in human recipients [5], especially for treating active bleeding [6, 7], but there was a decrease of at least 50% in recovery and survival when the PLTs were stored in the cold [8]. Storing PLTs at cold temperatures is now known to reduce metabolic activity [9, 10], alter PLT morphology from discoid to spherical [11, 12] and decrease responses in various in vitro tests, for example, the hypotonic shock response (HSR) and the extent of shape change [8, 13]. Based on these data, it was suggested that PLTs specifically intended to treat active bleeding should be stored at cold temperatures, whereas PLTs used for bleeding prophylaxis should be stored at room temperature [9]. Trehalose is widely used as a biomacromolecular protective agent, which decomposes into two GLU molecules in vivo and has no side or toxic effects [14]. At low temperatures, trehalose can be used to replace water, prevent the solid–liquid transition phase of the PLT membrane, and halt GPIba polymerization [15]. Hence, trehalose is considered a satisfactory, potential, protective agent for PLTs stored in PAS at low temperatures [10, 16]. In the present study, we modified the PAS by adding trehalose and replaced 70% of the plasma with this modified PAS (PAS-III M þ T) to preserve SDPs at 10  C. PLT membrane glycoprotein P-selectin (CD62p) expression, HSR, PLT aggregation rate (PAgT), pH, LDH release, GLU consumption, PLT count, mean platelet volume (MPV), and platelet distribution width (PDW) were measured during 9 days of cold (10  C) and routine (22  C) storage conditions.

Materials and methods In vitro function of PLT stored on PAS with trehalose Preparation of the modified PAS with trehalose Trehalose was added in the PAS to prepare the modified PAS [14, 16]. Table I lists the constituents of the modified PAS-III M þ T. All chemical components were purchased from Shanghai Chemical Reagent Co. Ltd (Shanghai, China) and ultra pure trehalose was from Sinozyme Biotechnology Co. Ltd (Nanning, China). Sample collection and experimental groups Due to the difficulty in the collection of multiple-doses SDPs, we designed a paired, case–control experiment to reach the practical PLT storage. SDPs were collected from six healthy volunteers using a blood cell separating apparatus (XCF 3000, Blood component separator, Sichuan Nigale Biomedical Co. Ltd. Chengdu, China) according to the manufacturer’s operating procedure and stored in the PLT preserve bag (B-2000, transfer bag, Sichuan Nigale Biomedical Co. Ltd. Chengdu, China). The SDPs for this study were donated by the blood volunteers and approved by the ethic committee of the Southwest Hospital of China (ID 201104). All the participants assigned the blood donor questionnaire to participate this study, without additional written informed consent. All the blood units donated to the blood center Table I. Constituents of the PAS-III M þ T. Constituent Na3-citrate, 2H2O Na-citrate, 3H2O NaH2PO4, 2H2O Na2HPO4 KCl MgCl2, 6H2O NaCl Trehalose

Content (g/l) 3.18 4.42 1.05 3.05 0.37 0.30 4.05 75.00

or bank can be used for patients in hospital or for blood relative medical researches in blood center or bank according to the blood donation law of China. The ethic committee of the Southwest Hospital approved blood donor questionnaire to replace the written consent for this research project. Each double-doses SDP unit was equally split into two equal parts (about one adult therapeutical dose) and preserved separately. In group A, about 2.5  1011 PLTs in 250 ml of plasma for each SDPs was centrifuged at 4000g for 8 minutes (22  C; J-6M centrifuge, Beckman, Palo Alto, CA) and about 180 ml plasma (70%) was removed replaced with an equal volume of PAS-III M þ T using a sterile connector device (TSCD-II, Terumo, China), while in group B there were no changes. At the beginning, group A was incubated at 37  C incubator for 2 hours to internalize trehalose into PLTs. After that, group A was preserved at 10  C in agitation. Group B was preserved at an ordinary temperature (22  C) in agitation. Samples from groups A and B were collected and analyzed after 1, 3, 5, 7, and 9 days of storage. PLT parameters PLT count, MPV, and PDW were determined using a KX-21N blood cell counter (Sysmex Company, Kobe, Japan).

Assays Activation marker Determination of membrane glycoprotein CD62P on apheresis PLTs [17] For each sample, three tubes (A–C) were used. In each tube, 7 ml of the sample and 993 ml PAS were added, and the PLT concentration was adjusted to 106 PLTs/ml. In tube A, mouse IgG1/FITC and mouse IgG1/PE were added (Beckman Coulter, Inc., Mississauga, ON, Canada); in tube B, CD41a-FITC and Mouse IgG1/PE were added (Beckman Coulter, Inc.); and in tube C, CD41a-FITC and CD62p-PE were added (Beckman Coulter, Inc.). The tubes were incubated in the dark at 20  C for 20 minutes; then 2 ml phosphate-buffered saline (PBS) was added into each tube to wash the PLTs, followed by centrifugation at 1000 rpm for 5 minutes and removal of supernatants. PLTs were resuspended in 0.3 ml 1% cold paraformaldehyde (4  C), kept for 30 minutes at 2–8  C in the dark, and then analyzed on a flow cytometer (FACS Calibur flow cytometer, BD Biosciences, San Jose, CA). PLT metabolism pH The pH of the sample was measured at room temperature using a blood gas analyzer (Medical EasyBloodGas Blood Gas Analyzer, GMI, Inc., MN). LDH and GLU determination LDH and GLU of the sample were determined using Cobas C 501 automatic biochemistry analyzer (Roche Diagnosis Limited, Indianapolis, IN, USA). LDH release and GLU consumption were calculated by subtracting the baseline values at day 0 from the values at day 1, 5, 7, and 9, respectively. PLT hypotonic shock response HSR was determined described by Holme et al. with minor modifications [18]. Each sample was divided into two test tubes. One was centrifuged at 3000 rpm for 10 minutes, and the supernatant was platelet-poor plasma (PPP). The PLT concentration in the other tube was adjusted to 3.0  1011 l1 to obtain

DOI: 10.3109/09537104.2013.872772

PLT-rich plasma, and the pH was adjusted to 7.0 with HEPES buffer. Isotonic PBS and hypotonic deionized water were warmed to 37  C. The blood agglutometer (TYXN-91 Blood Agglutometer, General Machinery & Electronics Company, Shanghai, China) was also warmed. In one test tube, 200 ml PPP was added, and the testing well was read. In another test tube, 200 ml PRP was added, and a stirring rod was placed before the test tube was inserted into the testing well. Two-hundred microliters of isotonic PBS was injected quickly into the test tube, immediately followed by a reading and recording of the transmittance change in the agglutometer after 1 minute. Then, with isotonic PBS replaced by hypotonic deionized water, the reading and recording of the transmittance change after 5 minutes were repeated. HSR can be calculated as follows:

Platelets Downloaded from informahealthcare.com by University of Auckland on 10/16/14 For personal use only.

HSR ¼

Tmax  T5 min  100%, Tmax  TPBS

where TPBS is the transmittance value after the first minute with isotonic PBS; Tmax is the maximum transmittance value with hypotonic deionized water; T5 min is the transmittance value after 5 minutes with hypotonic deionized water. PLT aggregation rate The response of the PLTs to the agonists ADP and adrenaline was measured by the method similar to that of the PLT hypotonic shock assay. In one test tube, 200 ml PPP was added, and the test tube was inserted into the testing well for zeroing. In another test tube, 200 ml PRP was added, and a stirring rod was placed before the test tube was inserted into the testing well. In the test tube, 10 ml ADP (40 mmol/l) and 10 ml adrenaline (1 mg/ml) were added, and subsequently, the maximum PAgT within 5 minutes was immediately determined.

Morphology The morphology of the apheresis PLTs was evaluated under the scanning electron microscope (S-3400N, HITACHI Corp., Tokyo, Japan). The 200 ml samples of PLTs were mixed with 2% glutaraldehyde (0.053 M cacodylate buffer), allowed to fix at room temperature for 2 hours, and then centrifuged at 2500g for 5 minutes. The supernatant was removed, and the pellet was slightly washed in 1 ml PBS buffer. The samples were then dehydrated in an ethanol and anhydrous tert-butanol series, dried by the critical point method, mounted on aluminum or copper stubs, and coated with 20 nm gold in a sputter coater. Observations were made and evaluated under the scanning electron microscope. Bacteria assays To test the bacteria growing, a 1.0-ml aliquot of the PLTs was inoculated into 9 ml of Luria Bertani broth and incubated overnight. Viable bacteria were detected by spreading the overnight culture on a Luria Bertani agar plate with incubation at 37  C overnight. In vivo PLT recovery and survival in rabbit model Preparation of guinea pig anti-rabbit PLT antisera Adult white rabbits (2.5–3.0 kg New Zealand white rabbit, Chongqing Animal Corporation) were given free access to food and water. Ethical permission for the studies was granted by the Animal Research Welfare Committee of the Third Military Medical University of China (ID 2005087) [17]. Eighteen milliliters of blood was drawn from each rabbit by cardiac puncture into plastic syringes containing ethylenediaminetetraacetic acid (EDTA)-K2. The anticoagulated blood was centrifuged at 1000g

Cold-stored platelets in PAS with trehalose

3

for 10 minutes at 4  C to prepare PLT-rich plasma, which was then heavily centrifuged at 3750g for 20 minutes. The PLT pellet was washed three times in PBS and then resuspended in saline to reach a PLT level of 2  1012 l1. The 2 ml suspension of PLTs and the same volume of incomplete Freund’s adjuvant (Sigma-Aldrich Shanghai Trading Co Ltd Shanghai, China) were homogenized completely with a homogenizer (FSH-2, Jiangshu Huanyu Instrument Factory, Nanjing, China). Each guinea pig (from the animal center of the Third Military Medical University) was subcutaneously immunised 500 ml of the PLT mixture at multiple points on the skin of the back once a week. After the first injection, the antibody titer was measured every week. Blood was drawn from the carotid arteries about 45 days after the first injections when the titer reached 1:51 200. The blood was placed in a 37  C water bath for 2 hours and centrifuged at 3500g for 10 minutes to collect serum, which was then aliquoted for storage in a 80  C freezer. The antiserum was then thawed at 56  C for 45 minutes to inactivate complements before use. Preparation of the thrombocytopenic rabbit model White rabbits were restrained, and their lateral ear veins were swabbed with 95% ethanol [17, 19]. A 25-gauge scalp vein needle was inserted into the vein, and adequate blood return was demonstrated. Doxorubicin (Haizheng pharmaceutical limited corporation, Zhejiang Province, China) was injected at 3 mg/kg of body weight on the first day and then at 2 mg/kg on the third day. The guinea pig anti-rabbit PLT antibody was filtered through a 0.22-pm filter (Millex-HA; Millipore Products Division, Bedford, MA), and about 1.0 ml was injected into the scalp vein over 3–5 minutes. The residual antibody in the infusion set was delivered by flushing the set with 3 ml sterile 0.15-mol/l NaCl before removing the needle. Animals were observed to reach PLT counts less than 20  109/l1 1 hour after the antibody injection. The antisera dramatically reduced PLT counts to less than 5% of normal levels but had no significant effect on either red blood cell or white blood cell counts. This thrombocytopenic rabbit model was successfully established for determination of PLT recovery and survival. Preparation of rabbit PCs stored in modified PAS-III M þ T at 10  C or in plasma at 22  C About 16–18 ml of white rabbit blood was drawn under anesthesia by an aseptic cardiac puncture into plastic syringes and placed into 4 ml of ACD in an anticoagulant tube. The blood was lightly centrifuged at 1100g for 5 minutes at 22  C to prepare PLT-rich plasma, which was then heavily centrifuged at 3750g for 6 minutes. One unit (5  1011 PLTs) of the rabbit PLT mixture (60 ml) used to study in vivo PLT recovery and survival was produced with six rabbit PCs preparations. The PLT pellet was resuspended in plasma (10 ml) or modified PAS-III M þ T containing trehalose (10 ml) to reach PLT counts of 1.5  1012 /L or L1. In vivo PLT recovery and survival in thrombocytopenic rabbit model The mixed rabbit PCs suspension was infused into thrombocytopenic rabbits through the lateral ear vein at a dose of (20–30)  109 PLTs per rabbit in a volume of 20–25 ml at a rate of 1 ml/minute [19, 20]. After 24 hours of infusion, the approximately 0.8 ml blood samples were collected in 1.5-ml polypropylene Eppendorf tubes containing 0.1 ml of 5% EDTA and counted PLTs. The percentage of platelet recovery (PPR) was measured according to the following formula: PPR ¼

PI  W  0:07  100%, NF

4

X. Wang et al.

Platelets, Early Online: 1–8

Table II. In vitro assay results of PLTs stored in modified PAS-III M þ T containing trehalose at 10  C (Group A) and in plasma at 22  C (Group B) (x  SD, n ¼ 6). Storage time day Group A 0

Platelets Downloaded from informahealthcare.com by University of Auckland on 10/16/14 For personal use only.

LDH increase (U/1011PLT) GLU consumption (mmol/1011 PLT) pH PLT (109 l–1) PDW (fl) MPV (fl) CD62p (%) HSR (%) PAgT (%)

– – 7.02  0.15 1435  129 10.1  1.4 8.5  0.9 6.2  1.5 73.0  2.2 91.3  1.5a

1

0

1

a

–

10.0  1.9

0.09  0.03a 0.23  0.06a 0.30  0.03a 0.41  0.07a

–

0.24  0.10 0.94  0.15

2.9  1.6

5

Group B

a

6.96  0.21 1401  179 11.2  1.0 9.3  0.7 8.8  2.2ab 69.4  1.8b 65.8  4.4ab

9.0  2.9

7 a

6.96  0.06 1371  212 10.5  1.22 8.7  0.9 33.6  4.0ab 61.7  3.2ab 9.2  1.3ab

13.3  2.6

9 a

7.03  0.13 1370  161 11.6  1.0 9.9  0.6 42.4  2.3ab 45.4  4.5ab 4.8  2.2ab

18.6  3.4

7.05  0.08 1387  127 10.4  1.2 8.7  0.9 53.8  2.8ab 37.9  3.4ab 2.6  1.3ab

7.42  0.08 1400  238 12.6  1.9 9.7  1.4 7.3  1.8 75.5  4.2 97.9  0.7

7.53  0.12 1446  120 10.3  0.8 8.2  0.6 15.2  2.0b 70.5  2.7b 86.5  1.5b

5 20.3  2.4

7.11  0.31b 1239  182 14.6  1.6 10.3  0.7 42.2  2.1b 58.6  2.8b 23.4  2.6b

7

9

28.0  7.2

30.4  3.2

1.12  0.19

1.25  0.15

7.02  0.12b 1297  280 11.1  2.0 9.8  1.0 46.4  2.6b 34.5  3.7b 12.8  1.8b

7.06  0.10b 1377  111 11.4  1.1 10.5  1.0 59.2  2.4b 32.2  4.2b 5.7  1.3b

Note: GLU consumption and LDH release are expressed as mmol (U)/1011 PLT/day. a p50.05 vs. group B at the same time point. b p50.05 vs. the same group on day 0.

where PPR is the percentage of PLT recovery, PI is the PLT count before transfusion minus after transfusion, W is the body weight in kg, N is the number of PLTs transfused, and F is the coefficient factor 0.62. Ten milliliters of rabbit PCs was centrifuged at 3000g for 10 minutes at 20–24  C. The supernatant of the PPP was removed and placed in another tube for later use. The pellet of the PLTs was resuspended in 1 ml of saline, labeled with 50 mCi Na251CrO4 (PerkinElmer Life and Analytical Sciences, Inc., Wellesley, MA, USA), and incubated at 37  C for 45 minutes with gentle shaking several times during the incubation. The mixture was centrifuged at 3000g for 10 minutes at 20–24  C again, and the supernatant was removed. The radiolabeled pellet was resuspended with PPP (10 ml) and infused into a rabbit through the lateral ear vein at a rate of 1 ml/minute. Serial blood samples of approximately 0.5 ml were collected from a marginal ear vein at 1, 24, 48, and 72 hours post-transfusion, respectively, and mixed with 50 ml of 5% EDTA. The radioactivity of the labeled, circulating rabbit PLTs was counted in a gamma counter (GC-911g Scintillation Counting System, USTC Chuangxin Co., Ltd., ZONKIA Branch, Anhui, China). PLT survival was calculated with radioactivity 1 hour post-transfusion being 100%. Statistical analysis Data are expressed as mean  standard deviation (x  SD), and paired t test (1:1) was performed using SPSS 13.0 (Chicago, IL, USA). The p value of less than 0.05 was regarded as significant.

Results In vitro results of apheresis PLTs stored in modified PAS-III M þ T containing trehalose at 10  C (group A) and in plasma at 22  C (group B) A minority of the apheresis PLTs stored in PAS-III M þ T containing trehalose at 10  C lost their ability to swirl, whereas the PLTs stored at 22  C maintained their swirl ability after 5 days of storage. Table II shows results for the in vitro assays. For PLTs stored at 10  C, there was a mean loss of PLT counts of 4.5% and 4.5% by days 5 and 7 compared to 11.5% and 7.5% by days 5 and 7 at 22  C. During storage, PLT count, MPV, and PDW changed insignificantly in both groups A and B.

At each time point over 9 days, GLU consumption and LDH release of the apheresis PLTs at 10  C (group A) were lower than that at 22  C (group B). The pH of apheresis PLTs at 22  C (group B) was significantly reduced by day 5 (7.11 vs. 7.42 by day 0, p50.01). However, the pH of apheresis PLTs at 10  C (group A) was not significantly different between days 0 and 5. During days 7–9 of storage, the pH was similar between groups A and B. During storage, the expression of PLT membrane glycoprotein CD62p in both groups tended to increase over time. The percentage of CD62p expression of the apheresis PLTs at 10  C (group A) was slightly lower than that of apheresis PLTs at 22  C at corresponding time points (group B). The PLT HSR tended to reduce over time during storage in apheresis PLTs at both 10 and 22  C. HSR was similar between groups A and B at days 0 and 1 but was higher in group A than group B at corresponding time points after storage day 5 (p50.05). The maximum PLT aggregation rate tended to decrease over time in apheresis PLTs in both groups at 10 and 22  C. During storage, the maximum PLT aggregation rate in the apheresis PLTs at 10  C (group A) was significantly lower than that in group B at each corresponding time point (p50.01). The morphology of human apheresis PLTs was observed with the scanning electron microscope. The shape of the fresh apheresis PLTs was discrete, discoid, plastic, and normally sized with regular and smooth margins. Some had aggregated due to physical accumulation. A few PLTs were seen in pseudopodium formation (Figure 1A). At days 3 and 5, the morphology of the apheresis PLTs stored according to the routine method in plasma at 22  C was similar to that of fresh samples. The aggregation of apheresis PLTs was more likely due to physical accumulation. The dendritic, spread, and aggregated PLT forms were seldom seen (Figure 1B). On days 3 and 5, the morphology of the apheresis PLTs stored in modified PAS-III M þ T containing trehalose at 10  C had a more irregular shape with long and slender pseudopodia. A minority of the PLTs transformed from the discoid to spherical form with long and slender pseudopodia and irregular margins although most of the PLTs still had a normal size and shape (Figure 1C). More than half of the PLTs stored in the plasma for 9 days at 22  C were aggregated and had irregular margins. A minority of the PLTs exhibited a normal shape (Figure 1D). With the prolongation of

Cold-stored platelets in PAS with trehalose

Platelets Downloaded from informahealthcare.com by University of Auckland on 10/16/14 For personal use only.

DOI: 10.3109/09537104.2013.872772

5

Figure 1. The morphologies of human PLTs stored at different conditions were seen under the scanning electron microscope (5000). (A) Fresh apheresis PLTs. The shape of most of the fresh human apheresis PLTs is independent, discoid, regular smooth margin, plastic effect,and normal size. Some are aggregated or fused together due to physical accumulation. (B) Apheresis PLTs stored in plasma on day 3 at 22  C. The shape of the PLTs is similar to that of fresh PLTs. A few PLTs are seen with slender pseudopodia. (C) Apheresis PLTs stored in PAS-III M þ T containing trehalose on day 3 at 10  C. The shape of some PLTs has changed from discoid to spherical with long and slender pseudopodia and irregular margins. Most of the PLTs still presented with normal sizes and shapes. (D) Apheresis PLTs stored in plasma on day 9 at 22  C. The shape of more than half of the PLTs is aggregated and formed together with irregular margins. A minority of PLTs is normally shaped. (E) Apheresis PLTs stored in PAS-III M þ T containing trehalose on day 9 at 10  C. Most PLTs were aggregated and fused together, forming a viscous, irregular shape.

storage time, more and more PLTs changed from the normal discoid shape to irregular spherical form. On day 9, most of the apheresis PLTs stored in PAS-III M þ T containing trehalose at 10  C were aggregated and fused together, forming viscous irregular shapes (Figure 1E). After 5 days, the morphology of the PLTs in group B maintained a more normal shape than that of group A. In vivo PLT recovery and survival in the rabbit model The results for the in vivo rabbit PLT recovery are shown in Table III. For six rabbits infused with fresh PLTs, the mean peak recovery of rabbit PLTs was (50.8  3.9)%. The recovery of fresh PLTs was not significantly different than the mean peak recovery of rabbit PLTs stored in plasma on day 3 at 22  C or in PAS-III

Table III. Mean peak recovery (%) of rabbit PLTs stored at different conditions in the rabbit model (x  SD). Groups Group Group Group Group Group

B on day B on day A on day A on day A on day

Mean peak recovery (%) 1 3 3 7 9

50.8  3.9 45.7  4.4 43.4  5.1 42.8  0.9 42.7  3.3

Note: Results are given as mean x  SD, n ¼ 6. There is no significantly difference among the groups.

6

X. Wang et al.

Platelets, Early Online: 1–8

Table IV. Survival (%) of rabbit PLTs thrombocytopenic rabbit model (x  SD).

after

transfusion

in

Survival (%) Groups Group Group Group Group Group

B on day B on day A on day A on day A on day

1 3 3 7 9

24 hours

48 hours

72 hours

80.3  2.0 78.2  2.9 79.9  2.6 74.4  3.2a 72.3  1.9a

70.8  4.9 66.3  6.6 63.3  2.3 59.1  3.3a 56.9  4.3a

54.4  1.1 51.4  2.0 50.9  2.9 48.8  1.8 44.8  4.2a

Platelets Downloaded from informahealthcare.com by University of Auckland on 10/16/14 For personal use only.

Note: Results are given as mean x  SD, n ¼ 6. a The survival percentages of the PLTs 24, 48, and 72 hours posttransfusion for group A on days 7 and 9 of storage were significantly lower than group B on day 1, respectively (p50.05).

M þ T containing trehalose on day 3 at 10  C. Over 9 days, for PLTs stored in PAS-III M þ T containing trehalose at 10  C, there was no significant difference in percentages of PLT recovery on days 3, 7, and 9, which were (43.4  5.1)%, (42.8  0.9)%, and (42.7  3.3)%, respectively (Table III). The percentages of in vivo PLT survival for fresh rabbit PLT suspensions and those stored in plasma at 22  C on day 3 and in PAS-III M þ T containing trehalose at 10  C on day 3 were (80.3  2.0)%, (78.2  2.9)%, and (79.9  2.6)%, respectively, 24 hours post-transfusion. These were compared to PLTs infused into rabbits as 1 hour post-transfusion with a survival of 100%. Seventy-two hours post-transfusion, the survival of PLTs stored up to 3 days tended to decrease but with no significant difference among the different storage conditions at the same time points (Table IV). On day 9, the survival of PLTs stored in PAS-III M þ T containing trehalose at 10 C decreased significantly than that within 24 hours (Table IV).

Discussion PLT transfusion is very important in transfusion medicine for the treatment of thrombocytopenia and hemorrhage due to various causes. Conventionally, apheresis PLTs can be stored at (22  2)  C in shaking for 5 days. During storage, PLTs undergo structural and biochemical changes that are collectively referred to as ‘‘PLT storage lesions.’’ Extending the storage period of PLTs will probably have some effects on PLT quality, and bacterial contamination also occurs easily at (22  2)  C in plasma media [8]. In addition, plasma used to store SDPs greatly increases the risk of transfusion reactions, such as allergy and fever. Partly replacing plasma with PAS to store SDPs can extend PLT storage for 7 days and reduce the risk of transfusion reactions and the incidence of blood-borne pathogens, particularly in Northern European countries [8, 20, 21]. Therefore, it would be valuable to utilize PAS to maintain PLT function and extend the storage period [22, 23]. We adapted PAS-III M þ T to increase the content of potassium chloride and magnesium chloride because potassium and magnesium can stabilize the cell membrane and inhibit PLT aggregation [20]. We added trehalose to decrease lesions due to cold storage and preserve PLT function in additive solution at a low temperature [10, 21, 24, 25]. Therefore, this study was designed a paired, case control experiment to investigate the in vitro and in vivo effects of PLTs stored in PAS-III M þ T with trehalose at 10  C and in plasma at 22  C. Due to the difficult of collection of multiple-doses SDPs, we cannot design multiple groups including PAS-III M þ T at 22  C and plasma preservation at 10  C for control groups. It is the shortage of this study. It is not easy to find a balance among extending the storage period of PLTs, protecting PLT quality, and avoiding bacterial

contamination during storage [4, 8, 9]. In this study, we stored PLTs in modified PAS with trehalose at 10  C due to damage theoretically to PLTs is milder when they are stored at 10  C rather than at 4  C, and bacteria grow and proliferate slowly at 10  C, thus reducing the risk of bacterial contamination of the PLTs. Meanwhile, we added trehalose to the PAS to protect PLTs against the adverse effects of low-temperature storage. Trehalose is accumulated at high concentrations, as much as 20% of the dry weight, by many organisms and is widely used to protect cells or organs [14, 25]. Trehalose is a non-reducing sugar formed from two GLU units in plants and animals, and is broken down into GLU by the catabolic enzyme trehalase for use. Trehalose is a non-permeable cell preservative agent, and is used to up to 50 mg/ml level in PCs as a best condition for PLTs preservation at cold temperature [10]. The function of rabbit PLTs stored in plasma with trehalose at cold temperature showed no difference, compared to the control group (PLTs stored with trehalose at 20–24  C for 24 hours) [10]. The two prevalent theories as to how trehalose works within the organism in the state of cryptobiosis are the vitrification theory, a state that prevents ice formation, or the water displacement theory, whereby water is replaced by trehalose, although it is possible that a combination of the two mechanisms is at work. In our study, the final concentration of trehalose is about 50 mg/ml level (30% plasma and 70% PAS-III M þ T). Cryophysiology demonstrated that at low temperatures, PLTs changed from a normal, discoid shape to spherical form [21], and after 24 hours of storage at 4  C with shaking, the release of a2 granules and activation of PLTs possibly occurred [21]. In addition, a drastic increase in intracellular calcium ions and actin assembly not only changed PLT shape but also caused lipid phase transitions of the cell membrane and lateral separation as well as an increase in cell membrane permeability. PLTs released both a granules and dense body contents during storage. Then, the ADP released from the dense bodies caused the PLTs to undergo shape change and aggregation. PLTs stored at low temperatures may decrease their release of dense bodies, but cold storage of PLTs easily causes PLTs to activate and lose their normal shape [26]. Membrane lipid phase transitions that occur when PLTs are cooled to 4  C also negatively affect PLT membrane integrity. PLTs stored at cold will be removed by mononuclear macrophages (Kupffer cells) for short-term-cooled PLTs and by hepatocytes for the prolonged storage PLTs [22, 26]. Mondoro and Vostal [23] also demonstrated that cold-stored PLTs have a higher sensitivity to agonist-induced aggregation when compared to room temperature-stored PLTs. The results indicate that PLT GLU consumption was significantly lower at 10  C than at 22  C. LDH resides mainly in the cell and is released from injured cells. LDH release was lower in the PAS-III M þ T group than the plasma group, indicating that PLT integrity was better maintained in PAS-III M þ T at 10  C than in plasma at 22  C. HSR is a putative in vitro parameter that ideally reflects the functional status of PLTs in vivo [17, 23]. In this study, PLT HSR was significantly higher in the PAS-III M þ T group than the plasma group, demonstrating that trehalose protects PLTs stored in a liquid at a low (10  C) temperature. PLT CD62p expression is a parameter measuring PLT activation. The results indicate that the expression of PLT membrane glycoprotein CD62p in both temperature conditions tended to increase over time during storage. PLT CD62p expression was lower in the PAS-III M þ T group at 10  C than the plasma group at 22  C, demonstrating that the number of activated PLTs was slightly lower at 10  C than at 22  C. These results show that PLTs stored in the cold were less activated than those at 22  C. This was also observed in a study by Hornsey et al., which found that P-selectin expression (binding sites/PLT), soluble P-selectin and RANTES concentrations

Cold-stored platelets in PAS with trehalose

Platelets Downloaded from informahealthcare.com by University of Auckland on 10/16/14 For personal use only.

DOI: 10.3109/09537104.2013.872772

increased less during cold storage than at 22  C [8]. The maximum PLT aggregation rate was lower in the PAS-III M þ T group than the plasma group, suggesting that Kþ, Mg2þ, and trehalose in the PAS may inhibit PLT aggregation and activation. The GLU content was high, and the pH tended to decrease in the plasma group during storage at 22  C, demonstrating that the pH of the media usually reduces in the presence of high glycolytic activity and low buffer capacity, which is supported by other reports [22]. After 3 days, the normal morphology of the apheresis PLTs stored in modified PAS-III M þ T containing trehalose at 10  C and their ability to function were maintained compared to the fresh PLTs at 22  C. The shape of the fresh apheresis PLTs as visualized under the scanning electron microscope was discrete, discoid, plastic and normally sized with regular, smooth margins. As storage time increased, more apheresis PLTs were activated, and the shape of PLTs changed from a normal discoid to irregular spherical form. Some apheresis PLTs aggregated and fused together to form viscous, irregular shapes. The degree and number of lesioned PLTs in PAS-III M þ T containing trehalose on day 9 at 10  C were greater than that of the apheresis PLTs stored in plasma on day 9 at 22  C. There were no bacteria growing in both groups of PLTs. The in vivo rabbit model shows very clearly that both survival and recovery of PLTs stored in PAS-III M þ T with trehalose at 10  C and in plasma at 22  C changed similarly although there was a greater reduction of survival and recovery of PLTs stored at 10  C. This study used rabbit PLTs to investigate survival and recovery, avoiding the shortage of human PLT samples in other rabbit models. The rabbit model used in human PLT survival and recovery research blocks the reticuloendothelial system and may lead to variability in clearance. Moroff et al. [12] reported a 10% decrease in recovery and a decrease in survival from 6.5  1.4 to 2.0  0.3 days when PLTs stored at 12  C were transfused into humans. Becker et al. [6] saw a 20% decrease in recovery when PLTs stored at 4  C were transfused, and survival was shortened by 2–3 days. Recently, in an in vivo model, Hornsey et al. [8] demonstrated that there was a decrease of at least 50% in recovery and survival when the PLTs were stored in the cold. Huang et al. [10] showed the rabbit PLTs stored with trehalose at cold temperature caused significantly decrease of the survival and recovery of PLTs, compared with the control group stored at 22  C. Although the recoveries and survivals of PLTs stored at cold temperature in animals and humans are very different, and the rabbit PLTs might behave differently to cold storage as human PLTs, the change tendency is very similar. In our study, the in vivo recovery of rabbit PLTs when they were fresh and at 22 and 10  C for 3 days tended to decrease. PLTs stored until day 9 at 10  C showed a decrease of about 8% compared to fresh PLTs. The in vivo survival of PLTs when they were fresh and stored at 22 and 10  C showed similar changes 24, 48 and 72 hours posttransfusion into rabbits. It was demonstrated that with respect to in vitro human PLT HSR, CD62p expression, PLT count, PDW, MPV, pH, LDH release, GLU consumption, morphology, and in vivo rabbit PLT survival and recovery, PAS-III M þ T exhibited a preservative performance better than that of plasma. This study suggests that PAS-III M þ T can satisfactorily maintain in vitro and in vivo PLT quality. Hoffmeister et al. [22] showed that the chilled PLTs transfused in humans were cleaned quickly and associated with GpIb clustering. In order for the PAS with trehalose to be considered to be used in humans, more studies, such as PLT recovery and survival, the corrected count increments after transfusion, bleeding time, in human volunteers, are obviously needed in the future research. In summary, compared to plasma, the modified PAS containing trehalose can protect PLT function satisfactorily at 10  C and

7

potentially reduce the risk of bacterial proliferation. Furthermore, the modified PAS reduces the incidence of transfusion-related disease transmission by reducing plasma volume and prolongs the shelf life of PLTs. Despite the insufficient information about the storage of apheresis PLTs with PAS and the standard in vitro quality of apheresis PLTs stored in PAS, this study indicates that the storage of apheresis PLTs using the modified PAS at a low temperature is superior to the traditional storage of PLTs using plasma at an ordinary temperature. This study also provides the experimental basis for preserving manually prepared PCs using the modified PAS.

Acknowledgements The findings and conclusions in this paper are those of the authors and do not necessarily represent the views of the Southwest Hospital.

Declaration of interest The authors declare no conflicts of interest relevant to this manuscript. Funding This work was in part supported by the National Natural Science Foundation of China (NSFC 81270649) and the ‘‘12.5’’ Military Medical Important Project of China (ASW11J007-03; ASW11J007-05). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author contributions Conceived and designed the experiments: Z.S.M. Performed the experiments: W.X., L.J., S.R.H., F.Y.H. Analyzed the data: W.X., F.Y.H. Contributed reagents/materials/analysis tools: L.R.Q. Wrote the paper: W.X., Z.S.M.

References 1. Shehata N, Tinmouth A, Naglie G, Freedman J, Wilson K. ABOidentical versus nonidentical platelet transfusion: A systematic review. Transfusion 2009;49:2442–2453. 2. Gulliksson H. Platelet preservation media. TransfusApher Sci 2001; 24:241–244. 3. Rivera J, Lozano ML, de la Calle JC, Connor J, Go´nza´lez-Conejero R, Ferrer F, Currie L, Garcia VV. Quality assessment of platelet concentrates supplemented with second-messenger effectors. Transfusion 1999;39:135–143. 4. Rouault CL. Platelet therapy is always changing. Transfusion 2007; 47:192–193. 5. Murphy S, Gardner FH. Platelet preservation: Effect of storage temperature on maintenance of platelet viability—Deleterious effect of refrigerated storage. New Engl J Med 1969;280:1094–1098. 6. Becker GA, Tuccelli M, Kunicki T, Chalos MK, Aster RH. Studies of platelet concentrates stored at 22 C and 4  C. Transfusion 1973; 13:61–68. 7. Valeri CR. Circulation and hemostatic effectiveness of platelets stored at 4 C or 22 C: Studies in aspirin-treated normal volunteers. Transfusion 1976;16:20–23. 8. Hornsey VS, Drummond O, McMillan L, Morrison A, Morrison L, MacGregor IR, Prowse CV. Cold storage of pooled, buffy-coatderived, leucoreduced platelets in plasma. Vox Sang 2008;95:26–32. 9. Kaufman RM. Uncommon cold: could 4 C storage improve platelet function? Transfusion 2005;45:1407–1412. 10. Huang CY, Tang RC, Cai L, Cui Y, Liu L. Cold storage of platelet suspension by adding trehalose. Zhongguo Shi Yan Xue Ye Xue Za Zhi. 2005;13:904–906 [article in Chinese]. 11. Holme S, Sawyer S, Heaton A, Sweeney JD. Studies on platelets exposed to or stored at temperatures below 20 C or above 24 C. Transfusion 1997;37:5–11. 12. Moroff G, Holme S, George VM, Heaton WA. Effect on platelet properties of exposure to temperatures below 20 C for short periods during storage at 20–24 C. Transfusion 1994;34:317–321. 13. Tablin F, Wolkers WF, Walker NJ, Oliver AE, Tsvetkova NM, Gousset K, Crowe LM, Crowe JH. Membrane reorganization during chilling: Implications for long-term stabilization of platelet. Cyobiology 2001;43:114–123.

8

X. Wang et al.

Platelets Downloaded from informahealthcare.com by University of Auckland on 10/16/14 For personal use only.

14. Crowe JH, Crowe LM, Oliver AE, Tsvetkova N, Wolkers W, Tablin F. The trehalose myth revisited: Introduction to a symposium on stabilization of cells in the dry state. Cryobiology 2001;43:89–105. 15. Italiano JE, Bergmeier W, Tiwari S, Falet H, Hartwig JH, Hoffmeister KM, Andre´ P, Wagner DD, Shivdasani RA. Mechanisms and implications of platelet discoid shape. Blood 2003;101:4789–4796. 16. Dumont LJ, Hay SN, Herschel L, Brantigan B, Houghton J, Elfath MD, Brecher ME. Validation of a microbial detection system for use with ACD-A platelets with PAS III platelet additive solution. Transfusion 2011;51:2219–2227. 17. Kostelijk EH, Fijnheer R, Nieuwenhuis HK, Gouwerok CW, de Korte D. Soluble P-selectin as a parameter for platelet activation during storge. Throm Haemost 1996;76:1086–1089. 18. Holme S, Moroff G. Murphy S. A multi-laboratory evaluation of in vitro platelet assays: The tests for extent of shape change and response to hypotonic shock. Transfusion 1998;38:31–40. 19. Sandgren P, Shanwell A, Gulliksson H. Storage of buffy coat derived platelets in additive solutions: In vitro effects of storage at 4 C. Transfusion 2006;46:828–834. 20. Diedrich B, Sandgren P, Jansson B, Gulliksson H, Svensson L, Shanwell A. In vitro and in vivo effects of potassium and magnesium

Platelets, Early Online: 1–8

21.

22.

23. 24. 25. 26.

on storage up to 7 days of apheresis platelet concentrates in platelet additive solution. Vox Sang 2008;94:96–102. Ringwald J, Walz S, Zimmermann R, Zingsem J, Strasser E, Weisbach V, Eckstein R. Hyperconcentrated platelet stored in additive solution: Aspects on productivity and in vitro quality. Vox Sang 2005;89:8–11. Hoffmeister KM, Felbinger TW, Falet H, Denis CV, Bergmeier W, Mayadas TN, von Andrian UH, Wagner DD, Stossel TP, Hartwig JH. The clearance mechanism of chilled blood platelet. Cell 2003; 112:87–97. Mondoro TH, Vostal JG. Cold temperatures reduce the sensitivity of stored platelets to disaggregating agents. Platelets 2002;13:11–20. Wolkers WF, Walker NJ, Tablin F, Crowe JH. Human platelets loaded with trehalose survive freeze-drying. Cryobiology 2001;42: 79–87. Crowe JH, Tablin F, Wolkers WF, Gousset K, Tsvetkova NM, Ricker J. Stabilization of membranes in human platelets freeze-dried with trehalose. Chem Phys Lipids 2003;122:41–52. Rumjantseva V, Grewal PK, Wandall HH, Josefsson EC, Sørensen AL, Larson G, Marth JD, Hartwig JH, Hoffmeister KM. Dual roles for hepatic lectin receptors in the clearance of chilled platelets. Nat Med 2009;15:1273–1280.