Manabu Kinoshita, Department of Immunology and Microbiology, National Defense Medical College, 3-2 Namiki, Tokorozawa, Saitama 359-8513, Japan. Tel.: +81 4 2995 1541; fax: +81 4 2996 5194. E-mail: email@example.com
Summary. Background: We developed a fibrinogen γ-chain (dodecapeptide HHLGGAKQAGDV [H12])-coated, ADP-encapsulated liposome (H12-[ADP]-liposome) that accumulates at bleeding sites via interaction with activated platelets via glycoprotein IIb–IIIa and augments platelet aggregation by releasing ADP.
Objective: To evaluate the efficacy of H12-(ADP)-liposomes for treating liver hemorrhage in rabbits with acute thrombocytopenia.
Methods: Thrombocytopenia (platelets < 50 000 μL−1) was induced in rabbits by repeated blood withdrawal (100 mL kg−1 in total) and isovolemic transfusion of autologous washed red blood cells. H12-(ADP)-liposomes with platelet-poor plasma (PPP), platelet-rich plasma (PRP), PPP, ADP liposomes with PPP or H12-(PBS)-liposomes/PPP, were administered to the thrombocytopenic rabbits, and liver hemorrhage was induced by penetrating liver injury.
Results: Administration of H12-(ADP)-liposomes and of PRP rescued all thrombocytopenic rabbits from liver hemorrhage as a result of potent hemostasis at the liver bleeding site, although rabbits receiving PPP or ADP liposomes showed 20% survival in the first 24 h. Administration of H12-(ADP)-liposomes and of PRP suppressed both bleeding volume and time from the site of liver injury. H12-(phosphate-buffered saline)-liposomes lacking ADP also improved rabbit survival after liver hemorrhage, although their hemostatic effect was weaker. In rabbits with severe thrombocytopenia (25 000 platelets μL−1), the hemostatic effects of H12-(ADP)-liposomes tended to be attenuated as compared with those of PRP treatment. Histologic examination revealed that H12-(ADP)-liposomes accumulated at the bleeding site in the liver. Notably, neither macrothombi nor microthrombi were detected in the lung, kidney or liver in rabbits treated with H12-(ADP)-liposomes.
Conclusions: H12-(ADP)-liposomes appear to be a safe and effective therapeutic tool for acute thrombocytopenic trauma patients with massive bleeding.
Hemorrhage is a major cause of preventable death in trauma victims and combat casualties [1,2]. In particular, non-compressible thoracoabdominal hemorrhage accounts for half of these deaths . Therefore, control of exsanguinating hemorrhage is crucial for reducing mortality in severe trauma victims. Excess blood loss due to uncontrollable hemorrhage often requires massive blood transfusion, which also increases patient mortality . Coagulopathy and thrombocytopenia following massive blood transfusion are frequently observed in such critical patients, resulting from hypothermia, acidosis, and dilution, as well as consumption of hemostatic elements, such as platelets and coagulation factors. Coagulopathy and thrombocytopenia are considered to be the major reasons for the high mortality [4,5].
Transfusion of platelet concentrates and plasma products is effective in attenuating coagulopathy/thrombocytopenia in hemorrhagic patients receiving massive red blood cell (RBC) transfusions, resulting in reduced mortality [6,7]. Replenishment of coagulation factors after massive transfusion can be accomplished by administration of fresh frozen plasma (FFP) containing a normal amount of coagulation factors such as fibrinogen, and this approach is particularly recommended for combat casualties [8,9]. Nevertheless, restoration of coagulation factors by FFP may not terminate the clinical coagulopathy in patients showing severe thrombocytopenia (platelets < 50 000 μL−1) . In these critical cases, platelet therapy is required . However, platelet concentrates are not usually available for emergency use, owing to their short-term viability, which requires stringent storage conditions [12,13]. The utility of platelet concentrates is very limited in difficult-to-reach rural areas, such as remote islands or mountain ranges. Furthermore, the populations of several countries, including Japan, are rapidly aging, and there is concern that a shortage of blood donors will become a serious problem in the near future.
The application of effective blood substitutes, such as oxygen-carrying blood substitutes or platelet substitutes, is therefore expected. It has obvious advantages, particularly in the care of large-scale disaster victims or combat casualties. Although oxygen-carrying blood substitutes appear to be effective in resuscitating patients with hemorrhagic shock , they have no hemostatic effect and are therefore ineffective against thrombocytopenic bleeding. Novel infusible hemostatic agents, namely platelet substitutes, are expected to rescue such critical patients [15,16].
We have developed liposome-based artificial platelet substitutes (mean diameter of 210 nm) bearing synthetic HHLGGAKQAGDV (H12) peptides corresponding to the C-terminus of the fibrinogen γ-chain on the surface, and with the physiologic platelet agonist ADP inside . In primary hemostasis, platelet plug formation is mediated by fibrinogen through the bridging of adjacent platelets via integrin αIIbβ3 (glycoprotein [GP]IIb–IIIa) in an activation-dependent manner [18,19]. With the H12 sequence, a primary recognition site for the GPIIb–IIIa receptor, the synthetic nanocarriers, HHLGGAKQAGDV-coated, ADP-encapsulated liposome (H12-[ADP]-liposomes), specifically target the sites of vascular injury where platelets have been activated [19,20]. H12-(ADP)-liposomes then reinforce platelet aggregation by releasing encapsulated ADP (Fig. 1). H12-(ADP)-liposomes thus work as platelet crosslinkers, and ADP release may be induced by liposomal membrane perturbation and/or destruction elicited by aggregation-driven sheared forces [21,22]. In our previous study, we found that H12-(ADP)-liposomes significantly shortened bleeding time in rats and rabbits with busulfan-induced thrombocytopenia, thus suggesting that they may be able to replace platelet products in prophylactic platelet transfusion . We herein investigated the therapeutic effects of H12-(ADP)-liposomes on non-compressible intra-abdominal hemorrhage resulting from liver injury, using a rabbit model with massive transfusion-induced acute thrombocytopenia.
Materials and methods
This study was conducted according to the guidelines of the Institutional Review Board for the Care of Animal Subjects of the National Defense Medical College.
Rabbits and reagents
New Zealand White rabbits (2 kg, male; Japan SLC, Hamamatsu, Japan) were used. Cholesterol and 1,2-dipalmitoyl-sn-glycero-3-phosphatidylcholine (DPPC) were from Nippon Fine Chemical (Osaka, Japan), 1,2-distearoyl-sn-glycero-3-phosphatidylethanolamine-N-(monomethoxypoly[ethylene glycol]) (PEG-DSPE; 5.1 kDa) was from NOF (Tokyo, Japan), and ADP was from Sigma-Aldrich (St Louis, MO, USA). We synthesized 1,5-dihexadecyl-N-succinyl-l-glutamine (DHSG) and H12-PEG-Glu2C18, where fibrinogen γ-chain dodecapeptide (C-HHLGGAKQAGDV [Cys-H12]) was conjugated to the end of the PEG–lipids, as described elsewhere .
Preparation of H12-(ADP)-liposomes, ADP liposomes, and H12-(phosphate-buffered saline [PBS])-liposomes
H12-(ADP)-liposomes were prepared as described elsewhere . Briefly, DPPC (1 g, 1.36 mmol), cholesterol (527 mg, 1.36 mmol), DHSG (189 mg, 272 μmol), PEG-DSPE (52 mg, 9 μmol) and H12-PEG-Glu2C18 (47 mg, 9 μmol) were dissolved in benzene and freeze-dried. The resulting mixture of lipids was hydrated with PBS containing 1 mmol L−1 ADP or PBS alone, and was filtered twice with Durapore (pore size, 0.45 μm; first and 0.22 μm; second; Millipore, Tokyo, Japan) to prepare H12-(ADP)-liposomes or H12-(PBS)-liposomes, respectively. After washing of liposomes with PBS followed by centrifugation (100 000 × g, 30 min, 4 °C), the remaining ADP was removed with Sephadex G25. ADP liposomes without H12 dodecapeptide were also prepared by skipping the conjugation of fibrinogen γ-chain dodecapeptide to PEG–lipid.
Acute thrombocytopenic rabbit model
Rabbits were anesthetized with intramuscular injections of ketamine (25 mg kg−1) and xylazine (10 mg kg−1), and this was followed by maintenance of anesthesia with intravenous injections of pentobarbital (15 mg kg−1) every 30 min during the experiment. Surgical catheters (polyethylene indwelling needle 22G; Terumo Co., Tokyo, Japan) were inserted into the femoral artery and vein in each rabbit (Fig. S1). Thereafter, 12.5 mL kg−1 blood (sample 1) was drawn from the femoral artery, and the same volume of lactated Ringer’s solution containing 5% human serum albumin (HSA) (Kaketsuken, Kumamoto, Japan) was simultaneously transfused via the femoral vein. Forty minutes later, the next blood sample (12.5 mL kg−1, sample 2) was withdrawn, and the same volume of washed RBCs as prepared with sample 1 was transfused. This isovolemic blood exchange was repeated seven times, and the last transfusion of washed RBCs was performed without simultaneous blood withdrawal (Fig. 2). Thereafter, platelet counts decreased to <5 × 104 μL−1. Furthermore, we repeated this isovolemic blood exchange 14 times in order to ensure a severe thrombocytopenic condition. After blood exchange, platelet counts decreased to (25 ± 1) × 103 μL−1. Body temperature was maintained at 37–38 °C in rabbits, with a heating pad. Arterial blood pH was maintained at 7.35–7.45.
Preparation of washed RBCs, platelet-rich plasma (PRP), and platelet-poor plasma (PPP)
Blood samples drawn with a 10% volume of 3.8% (w/v) sodium citrate were centrifuged at 100 × g for 15 min, and the supernatant was used as PRP (Fig. S1). The remaining sample was further centrifuged at 500 × g for 10 min, and the supernatant was used as PPP. Thereafter, the remaining cells were washed with saline, diluted in 25 mL of lactated Ringer’s solution containing 5% HSA, and transfused into the rabbit as washed RBCs (Fig. S1).
Administration of H12-(ADP)-liposomes, PRP, PPP, ADP liposomes and H12-(PBS)-liposomes
We administered the PRP or PPP that was prepared from the blood taken at the first and second phlebotomies. These PRP and PPP samples showed similar coagulation activity (fibrinogen, ∼ 150 mg dL−1; antithrombin [AT]III activity, 99%; prothrombin time [PT], 12 s; activated partial thromboplastin time [APTT], 32 s). After the last transfusion of washed RBCs, H12-(ADP)-liposomes (20 mg per 4 mL kg−1) were intravenously administered to the rabbits, and this was followed by administration of 11 mL kg−1 PPP (n = 10; Fig. 2). ADP liposomes (without H12) (20 mg per 4 mL kg−1, n = 10) or H12-(PBS)-liposomes (lacking ADP) (20 mg per 4 mL kg−1, n = 5) were also administered to the rabbits, and this was followed by PPP transfusion in the same manner. Similarly, 15 mL kg−1 PRP or PPP was administered to the rabbits (n = 10 in each group, Fig. 1). In addition, another three rabbits were prepared in each group for electron microscopic examination. In the severe thrombocytopenic model (blood exchange repeated 14 times), H12-(ADP)-liposomes/PPP, PRP or PPP were similarly administered to the rabbits.
Liver hemorrhage in thrombocytopenic rabbits
Thirty minutes after administration of H12-(ADP)-liposomes/PPP, PRP, PPP, ADP liposomes/PPP, or H12-(PBS)-liposomes/PPP, rabbits underwent laparotomy, and the liver was penetrated with a DermaPunc (5 mm in diameter; Nipro Medical Industries, Tokyo, Japan) (Fig. S2A,B). To precisely evaluate bleeding volume, we cut a hole in a surgical glove and passed the injured hepatic lobe through the hole to collect exsanguinating blood in the glove (Fig. S2C,D). We measured the bleeding volume from the liver during the initial 5 min and the following 5 min (5–10 min). Bleeding time from the liver-penetrating injury was measured for 20 min. The rabbit abdomen was then closed to monitor survival for 72 h under ad libitum feeding with laboratory diet and water. Postoperative analgesia was performed with two intramuscular injections of buprenorphine (0.02 mg kg−1): immediately after wound closure and after 12 h.
Analyses of whole blood coagulation activity
The coagulation activity of whole blood was examined with a Sonoclot Coagulation & Platelet Function Analyzer (Sienco, Morrison, CO, USA). Briefly, a tubular probe mounted on an ultrasonic transducer and vibrating vertically with a distance of 1 μm and a frequency of 200 Hz is immersed to a fixed depth in a cuvette containing 400 μL of whole blood obtained from the femoral artery without anticoagulant. As the sample clots, the increasing impedance to the vibration of probe is detected by the sensor and converted to an output signal, which reflects the viscoelastic properties of the developing clot. The signal typically describes coagulation parameters including clotting time (CT), which indicates the period up to the beginning of fibrin formation, and clot rate (CR), which indicates the slope of fibrin gel formation, and which is affected by both the rate of the fibrinogen to fibrin conversion and the amount of fibrinogen (Fig. 6A).
Measurements of ear bleeding time, mean arterial pressure, hematologic parameters, and coagulation factors
Ten minutes after administration of H12-(ADP)-liposomes/PPP, PRP, PPP, ADP liposomes/PPP, or H12-(PBS)-liposomes/PPP, the auricle was cut with a 6-mm incision where no vessels were visible, and the ear was immersed in a saline bath. The time required for bleeding to stop was then measured for 20 min. Mean arterial pressure was measured from the cannulated femoral artery with a polygraph recording system (RM-6000; Nihon Kohden, Tokyo, Japan). Blood samples were also collected from the femoral artery. Platelet count, hemoglobin concentration and white blood cell (WBC) count were measured with a hematology analyzer (PEC 170; Erma, Tokyo, Japan). The plasma concentration of fibrinogen, ATIII activity, PT and APTT were measured at the SRL Laboratory (Tokyo, Japan).
Rabbits treated with H12-(ADP)-liposomes, PPP or PRP were killed 24 h after liver hemorrhage (n = 3 in each group). One rabbit treated with PPP that died after ∼ 24 h was also examined. Three rabbits receiving H12-(ADP)-liposomes were also killed at 2 weeks after hemorrhage. Liver (uninjured lobe), spleen, lung and kidney were extracted from all subject rabbits, fixed in 20% formalin for 2 days, and embedded in paraffin. Sections (4 μm) were stained with hematoxylin and eosin.
Electron microscopic examinations
Three rabbits were prepared for each group. Liver specimens were obtained from the injury site at 1 h after injury. These were prefixed with a fixative containing 4% paraformaldehyde and 0.5% glutaraldehyde in 0.1 mol L−1 phosphate buffer (pH 7.4) for 3 h at 4 °C, and this was followed by postfixing in 1% osmium tetroxide in 0.1 mol L−1 phosphate buffer (pH 7.4) for 2 h at 4 °C, dehydration, and embedding in epoxy resin. For selection of the bleeding site lesion, semithin sections were stained with toluidine blue. Ultrathin sections stained with uranyl acetate and lead citrate were then examined under an electron microscope (JEM 1030; JEOL, Tokyo, Japan) at an accelerating voltage of 80 kV.
For immunoelectron microscopy, samples were processed as described elsewhere [23,24]. Briefly, fixed liver specimens were rinsed with 0.1 mol L−1 phosphate buffer and PBS at 4 °C, and then infused with 1 mol L−1 sucrose in PBS for 1 h, 1.84 mol L−1 sucrose in PBS for 2 h, and 1.84 mol L−1 sucrose containing 20% poly(vinylpyrrolidone) in PBS overnight at 4 °C. After being frozen in liquid nitrogen, ultrathin frozen sections were cut and incubated with rabbit anti-H12 (1 : 5000 dilution) in PBS overnight at 4 °C . After being rinsed with PBS five times, sections were incubated with goat anti-rabbit IgG coupled to 15-nm colloidal gold at a dilution of 1 : 100 for 60 min at room temperature. After being rinsed with PBS three times and with distilled water (DW) five times, sections were stained with 1% uranyl acetate and washed with DW, and then adsorption-stained with a mixture of 3% poly(vinyl alcohol) and 0.3% uranyl acetate. Stained sections were examined under a JEM 1230 electron microscope as described above.
Statistical analyses were performed with the Stat View 4.02J software package (Abacus Concepts, Berkeley, CA, USA). Survival rates were compared by use of the Wilcoxon signed rank test. Statistical evaluations were compared by use of one-way analysis of variance, followed by Bonferroni post hoc tests. Data are presented as means ± standard errors, with P <0.05 considered to be statistically significant.
Acute thrombocytopenia in rabbits
Platelet counts decreased gradually in rabbits after repeated blood withdrawal and washed RBC transfusion, reaching 5 × 104 μL−1 (n = 40) at the end of blood exchange (Fig. S3A). Ear bleeding time was prolonged markedly by blood exchange (Fig. 3). Nevertheless, mean arterial blood pressure was maintained as a result of isovolemic exchange (Fig. S3B). The hemoglobin concentration was also maintained at ∼ 8 g dL−1 (Fig. S3C), and the WBC count was minimally altered (Fig. S3D). Coagulation factor levels were, however, markedly decreased (Table S1), probably because of substantial plasma loss.
Ear bleeding time after administration of H12-(ADP)-liposomes in thrombocytopenic rabbits
In our previous studies on rabbits with busulfan-induced thrombocytopenia, H12-(ADP)-liposomes administered at a dose of 20 mg kg−1 gave optimal bleeding time-shortening effects, similar to those of PRP . Therefore, this liposome dose was selected for all of the present experiments. Administration of H12-(ADP)-liposomes/PPP and of PRP efficiently corrected bleeding time in thrombocytopenic rabbits (Fig. 3), although administration of H12-(PBS)-liposomes/PPP, ADP liposomes/PPP or PPP alone did not.
Changes in hematologic parameters after administration of H12-(ADP)-liposomes
Mean arterial blood pressure was not significantly altered after administration of H12-(ADP)-liposomes/PPP, PRP, PPP or ADP liposomes/PPP in all groups (Table 1). Platelet count was markedly increased after PRP administration, but was not affected by other treatments (Table 1). Neither hemoglobin concentration nor WBC count significantly changed in any of the four groups, whereas coagulation factor levels were corrected in all groups (Table 1). Hematologic parameters were also unaffected by administration of H12-(PBS)-liposomes/PPP (data not shown).
Table 1. Changes in hematologic parameters and coagulation factors in rabbits after administration of H12-(ADP)-liposomes/platelet-poor plasma (PPP), platelet-rich plasma (PRP), PPP, or ADP liposomes/PPP
APTT, activated partial thromboplastin time; ATII, antithrombin II; PT, prothrombin time; WBC, white blood cell. Each parameter was measured in rabbits before and after administration of H12-(ADP)-liposomes/PPP, PRP, PPP, or ADP liposomes/PPP. Data are mean ± standard error from 10 rabbits in each group, *P < 0.01 and †P < 0.05 vs. before administration.
Mean blood pressure (mmHg)
90 ± 7
91 ± 7
89 ± 7
88 ± 4
96 ± 10
98 ± 7
87 ± 8
90 ± 7
Hemoglobin concentration (g dL−1)
8.3 ± 0.6
7.9 ± 0.5
8.6 ± 1.3
8.1 ± 0.4
6.4 ± 0.5
6.2 ± 0.6
6.0 ± 0.2
6.1 ± 0.6
Platelet count (×103 μL−1)
50 ± 5
50 ± 6
46 ± 7
50 ± 6
47 ± 5
108 ± 12*
42 ± 4
42 ± 10
WBC count (×103 μL−1)
3.2 ± 0.4
3.5 ± 0.6
3.6 ± 0.3
3.0 ± 0.5
3.3 ± 0.4
4.1 ± 0.8
4.0 ± 0.7
2.8 ± 0.2
Fibrinogen concentration (mg dL−1)
50 ± 1
53 ± 3
54 ± 3
51 ± 1
69 ± 6*
73 ± 7*
73 ± 7*
65 ± 8*
ATIII activity (%)
38 ± 4
40 ± 5
37 ± 4
39 ± 5
52 ± 2†
57 ± 2†
53 ± 4†
53 ± 2†
29 ± 6
30 ± 5
25 ± 3
27 ± 2
18 ± 2†
16 ± 1†
15 ± 1†
19 ± 2†
37 ± 4
39 ± 8
40 ± 4
37 ± 4
34 ± 10
35 ± 1
28 ± 4
29 ± 6
Survival after liver hemorrhage in rabbits receiving H12-(ADP)-liposomes
After 72 h of observation, administration of H12-(ADP)-liposomes/PPP and of PRP rescued all rabbits from lethal liver hemorrhage. H12-(PBS)-liposome/PPP treatment also showed significant therapeutic effects on survival (4/5; 80%). In contrast, rabbits with PPP and ADP liposome treatment showed significantly poorer prognoses, with survival rates of 20% in the first 24 h (Fig. 4). Administration of H12-(PBS)-liposomes/PPP also did not rescue any rabbits (data not shown).
Hemostatic effect of H12-(ADP)-liposomes on liver hemorrhage in rabbits
Administration of H12-(ADP)-liposomes/PPP tended to decrease bleeding volume from the liver injury site in the initial 5-min period as compared with PPP or ADP liposome/PPP treatment (difference not statistically significant, Fig. 5A). Interestingly, H12-(ADP)-liposome/PPP administration significantly reduced bleeding volume in the subsequent 5 min as compared with that in the initial 5 min (Fig. 5A). In contrast, PRP administration significantly reduced liver hemorrhage even in the initial 5 min, but did not further augment hemostatic effects in the subsequent period (Fig. 5A). Administration of H12-(PBS)-liposomes/PPP did not reduce bleeding volume in the initial 5 min, but tended to reduce it in the subsequent 5 min as compared with that in the initial 5 min (difference not significant) (Fig. 5A). Bleeding time from the penetrating liver wound was significantly shortened by administration of H12-(ADP)-liposomes/PPP, as well as of PRP, as compared with PPP, ADP liposome/PPP or H12-(PBS)-liposome/PPP administration (Fig. 5B). Liver hemorrhage was almost fully stopped at 10 min after administration of H12-(ADP)-liposomes/PPP (Fig. S2C), whereas a substantial amount of hemorrhaging was observed at 10 min after PPP administration (Fig. S2D). Administration of ADP liposomes/PPP did not show any obvious hemostatic effects at the liver injury sites (Fig. 5A,B). Administration of H12-(PBS)-liposomes/PPP also showed no significant hemostatic effects (Fig. 5A,B), although it significantly improved recovery from liver hemorrhage in thrombocytopenic rabbits (Fig. 4).
Platelet-based blood coagulation activity in rabbits receiving H12-(ADP)-liposomes
Blood samples obtained after repeated blood withdrawal/RBC transfusion were not coagulated (Fig. 6A). However, supplementation of coagulation factors with PPP administration corrected the clotting parameters CT and CR to some extent, although correction of both parameters was not sufficient to restore normal clotting activity (Fig. 6B). Administration of PRP to supplement platelets, in addition to coagulation factors, further shortened CT and increased CR (Fig. 6C,D), indicating that both parameters measured on the Sonoclot apparatus depended on ex vivo platelet-based coagulation (platelet procoagulant) activity. Notably, H12-(ADP)-liposomes/PPP significantly corrected both parameters to a similar level as PRP (Fig. 6C,D). In contrast, ADP liposomes/PPP showed no significant correcting activity, as compared with that obtained with PRP or H12-(ADP)-liposomes/PPP. H12-(PBS)-liposomes also gave no significant corrections of these parameters. These results demonstrate that H12-(ADP)-liposomes possess the ability to replace platelets in accelerating the coagulation pathway towards fibrin clot formation.
Electron microscopic examination of clot at the site of liver injury
Dense coagulation clots adjacent to the injured site in the liver were observed after PRP administration (Fig. 7A, indicated by circles). These clots had substantial involvement of platelets and fibrin (Fig. 7B). H12-(ADP)-liposome-treated rabbits also showed coagulation clots adjacent to the injured tissue, whereas these appeared to be sparse as compared with those in PRP-treated rabbits (Fig. 7C, indicated by circles). Anhistous particles approximately 0.2–0.4 μm in diameter were observed around the RBC or fibrin deposits of H12-(ADP)-liposome-treated rabbits (Fig. 7D,E, indicated by arrowheads), although such particles were not observed in PRP-treated rabbits (Fig. 7B). On immunoelectron microscopy with anti-H12 polyclonal antibody, gold-labeled liposomes with unit membranes ∼ 0.3 μm in diameter were identified between the coagulation clots in the H12-(ADP)-liposome-treated rabbits (Fig. 7F, indicated by arrows), suggesting that some anhistous particles were H12-(ADP)-liposomes. In fact, anhistous particles were not seen at the liver injury sites of ADP liposome-treated rabbits (Fig. 7G), suggesting that H12-(ADP)-liposomes exert local hemostatic activity at sites of vascular injury through the specific binding of H12 to activated platelets.
Pathologic examination of lung, liver (uninjured lobe), and kidney
Neither macrothombi nor microthrombi were observed in the lung, liver (uninjured lobe) or kidney in the rabbits at 24 h after administration of H12-(ADP)-liposomes, PRP, or PPP (Fig. 8). Although severe pulmonary edema was observed after PPP transfusion, no such lesions were observed after administration of H12-(ADP)-liposomes or PRP (Fig. 8). Severe hepatocyte degeneration was also observed in PPP-treated rabbits, but not in H12-(ADP)-liposome-treated or PRP-treated rabbits (Fig. 8, indicated by arrows). However, some lymphocyte infiltration was observed around the central vein in the liver after administration of H12-(ADP)-liposomes (Fig. 8, indicated by arrowhead), suggesting an immune response to foreign bodies. Remarkable proliferation of neither Kupffer cells (Fig. 8) nor splenic macrophages (not shown) was observed in the rabbits after administration of H12-(ADP)-liposomes, suggesting that no serious damage occurs in the reticuloendothelial system. No significant changes were observed in the kidney in any groups (Fig. 8). We also histologically examined the liver, spleen, lung and kidney at 2 weeks after administration of H12-(ADP)-liposomes/PPP. Neither thrombi nor significant changes indicating immune reactions were observed in any organs. Lymphocyte infiltration around the central vein in the liver was not significant (Fig. S4).
Hemostatic effects of H12-(ADP)-liposomes on liver hemorrhage in the severe thrombocytopenic model
After 14 repeated isovolemic blood exchanges, platelet counts in rabbits decreased to (25 ± 1) × 103 μL−1. Administration of H12-(ADP)-liposomes/PPP rescued three of five rabbits from lethal liver hemorrhage in this severe thrombocytopenic model (60% survival). In contrast, PRP treatment that increased platelet counts to approximately 86 × 103 μL−1 (average) rescued four of five rabbits (80% survival). PPP treatment rescued no rabbits (Table 2). Although the rabbits given H12-(ADP)-liposomes/PPP showed a large amount of liver hemorrhage in the initial 5 min, their bleeding volume was clearly reduced in the subsequent 5 min (9.8 ± 4.4 to 3.1 ± 1.3 mL). In contrast, PRP treatment did not give such an obvious reduction in bleeding volume (average of 6.8–5.7 mL). Bleeding time from the liver injury tended to be prolonged (but not significantly) in rabbits receiving H12-(ADP)-liposomes/PPP as compared with those receiving PRP (1144 ± 50 vs. 875 ± 173 s; PPP treatment, over 1200 s).
Table 2. Survival after non-compressible liver hemorrhage in rabbits with severe thrombocytopenia
PPP, platelet-poor plasma; PRP, platelet-rich plasma. Data are given as no. (%). Severe thrombocytopenia was induced by 14 blood exchanges (12.5 mL kg−1) in rabbits, and their platelet counts decreased to (25 ± 1) × 103 μL−1.
In this study, we demonstrated the feasibility of H12-(ADP)-liposomes as a safe and efficacious synthetic platelet substitute applicable to uncontrollable traumatic bleeding confounded by acute thrombocytopenia after massive RBC transfusion and fluid resuscitation.
The current rabbit model did not exactly reflect a clinical situation with penetrating injuries, as acute thrombocytopenia usually occurs following injury. During the bleeding process, several interventions are essential for life-saving, including volume resuscitation, damage control surgery, and replacement of deficient coagulation factors and platelets. H12-(ADP)-liposomes were able to act as a substitute for platelets to rescue acute thrombocytopenic rabbits from lethal liver hemorrhage by controlling bleeding. When major surgical blood loss is replaced with plasma/platelet-poor packed RBCs, patients show deficiencies in both platelets and coagulation factors [11,25]. Our thrombocytopenic rabbits also showed severe deficiency in coagulation factors. Therefore, on the basis of current hemostatic resuscitation protocols for trauma patients, we administered H12-(ADP)-liposomes together with PPP to treat bleeding [7,26]. In fact, similarly to platelet transfusion with PRP, administration of H12-(ADP)-liposomes with PPP rescued all thrombocytopenic rabbits, although most did not survive when treated with PPP alone (Fig. 4).
Replenishment of coagulation factors by FFP should not be a major therapeutic choice for transfusion-induced coagulopathy [10,11]. Instead, dilutional thrombocytopenia may be a crucial target for coping with coagulopathy, as fibrin clot formation leading to secondary hemostasis through the coagulation pathway is not adequately completed without initial platelet thrombi being generated by platelets [10,11]. This hypothesis was supported by the finding that the levels of Sonoclot clotting activity in whole blood from thrombocytopenic rabbits rescued by treatment were higher with PRP (platelets/PPP) or H12-(ADP)-liposomes/PPP than with PPP alone or ADP liposomes/PPP (Fig. 6).
H12-(PBS)-liposomes potently increased the survival of thrombocytopenic rabbits after liver hemorrhage, whereas ADP liposomes without H12 did not (Fig. 4). Crosslinking platelets by H12 may be crucial for hemostatic resuscitation in this particular lethal liver hemorrhage animal model. However, direct hemostatic effects of H12-(PBS)-liposomes on bleeding time and bleeding volume from the injured liver were not observed (Fig. 5). At present, there is no convincing explanation for the discrepancy between the effects of H12-(PBS)-liposomes on direct hemostatic activity and those on animal survival. However, consistent with our previous observation of the effects of these H12 liposomes with or without ADP on prolonged tail bleeding time in rats with busulfan-induced thrombocytopenia , crosslinking of residual platelets by H12 may be a prerequisite for the hemostatic activity of the liposomes, and local release of ADP appears to have a crucial role in their optimal therapeutic function.
H12-(ADP)-liposomes have several advantages over platelet concentrates in the treatment of acute thrombocytopenia. First, they can be used in emergencies, as they can be stored for long periods of time under simple conditions. In addition, they are entirely synthetic, and as such are free from the risks associated with blood-borne infections and allogeneic host immune reactions. Currently, recombinant factor VIIa (rFVIIa) is widely used as a hemostatic agent to treat massive bleeding in emergency settings, such as the battlefield . However, the effects of rFVIIa on the lifetime prognosis in treated patients remain elusive . Under severe platelet-deficient conditions, rFVIIa may not exert an effective hemostatic function. Combination therapy with rFVIIa and H12-(ADP)-liposomes may be useful for the treatment of thrombocytopenic bleeding in trauma patients.
The most important safety issue for our synthetic platelet substitutes concerns thrombotic complications [15,29]. Our findings to date suggest that this is not a problem with H12-(ADP)-liposomes. In healthy rats and rabbits, there were no changes in coagulation and platelet activation parameters after liposome administration at doses up to 40 mg kg−1 . Pathologic examination revealed that rabbits with acute thrombocytopenia treated with H12-(ADP)-liposomes at a therapeutic dose of 20 mg kg−1 were free of thrombosis in tissues from lung, kidney, and intact liver. This relative safety may stem from the specificity of H12 ligand for GPIIb-IIIa receptor on the activated platelets. Histologic studies have demonstrated that H12-(ADP)-liposomes, but not ADP liposomes, accumulated in clots formed during liver injury. Nevertheless, it is premature to draw conclusions on safety with respect to the induction of intravascular thrombosis, as clinical conditions of patients receiving synthetic platelets may be complex, owing to the hypercoagulable state resulting from activating platelets in the circulation. Therefore, further study is needed to confirm the safety issues related to synthetic platelet substitutes.
As compared with PRP, the hemostatic effects of H12-(ADP)-liposomes on traumatic liver hemorrhage are somewhat slow, but were equally potent after 10 min of observation (Fig. 5). This mode of activity of liposomes may be related to their dependence on residual platelets to exert hemostatic activity in thrombocytopenic animals. Therefore, it is possible that there will be a thrombocytopenic concentration of platelets at which H12-(ADP)-liposomes will be unable to rescue the animals. When tested under severe thrombocytopenic conditions ([25 ± 1] × 103 platelets μL−1), they still showed significant effects on survival, similar to those of PRP, and the mode of their hemostatic effect on liver bleeding was similar to that obtained with (50 ± 1) × 103 platelets μL−1 (Table 2). Thus, H12-(ADP)-liposomes are capable of exerting their hemostatic function through residual platelets ranging from (25 ± 1) × 103 to (50 ± 1) × 103 μL−1 in this particular rabbit model. Another concern is that H12-(ADP)-liposomes will behave like platelet inhibitors in blocking fibrinogen-mediated platelet thrombus formation. In fact, H12-(ADP)-liposomes competed with fibrinogen to bind to activated platelets through GPIIb–IIIa, as determined with in vitro experiments under non-stirring conditions; 50% inhibition of liposome binding was obtained in the presence of a concentration of fibrinogen ∼ 4.3-fold higher than its average blood level (200 mg dL−1) . This dilutional thrombocytopenic animal model was associated with severe hypofibrinogenemia at levels of 50–70 mg dL−1. Nevertheless, H12-(ADP)-liposomes showed potent hemostatic activity, suggesting that crosslinking of the remaining platelets by H12-(ADP)-liposomes is indispensable for platelet thrombus formation. Platelets are first activated at the site of vascular injury, and thereafter, liposomes become vital. Therefore, further careful dose and administration studies are needed to optimize the indications for our platelet substitutes in clinical settings.
K. Nishikawa, K. Hagisawa, M. Kinoshita, D. Saitoh, and M. Handa: conception and design; K. Nishikawa, K. Hagisawa, M. Kinoshita, S. Shono, S. Katsuno, M. Doi, R. Yanagawa, H. Suzuki, and K. Iwaya: acquisition of data; K. Nishikawa, K. Hagisawa, M. Kinoshita, T. Sakamoto, and M. Handa: analysis and interpretation of data; M. Kinoshita and M. Handa: drafting of the manuscript; S. Takeoka and S. Seki: supervision.
We would like to thank M. Arai at Waseda University for the preparation of liposome samples and laboratory assistance. We would also like to thank Y. Ikeda at Waseda University and N. Watanabe at Keio University for critical reading of the manuscript.
Disclosure of Conflict of Interests
This work was supported in part by Health and Labor Sciences Research Grants (Research on Public Essential Drugs and Medical Devices; M. Kinoshita, H. Suzuki, S. Takeoka, and M. Handa) from the Ministry of Health, Labour and Welfare, Japan. The other authors state that they have no conflict of interest.