QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.104.073049v1 311/3/874    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sakai, H. Articles by Tsuchida, E. Search for Related Content PubMed PubMed Citation Articles by Sakai, H. Articles by Tsuchida, E. Pubmed/NCBI databases Compound via MeSH Substance via MeSH

ABSORPTION, DISTRIBUTION, METABOLISM, AND EXCRETION

Physiological Capacity of the Reticuloendothelial System for the Degradation of Hemoglobin Vesicles (Artificial Oxygen Carriers) after Massive Intravenous Doses by Daily Repeated Infusions for 14 Days

Advanced Research Institute for Science and Engineering (H.S., Y.M., K.S., S.T., E.T.), Waseda University, Tokyo, Japan; Departments of Surgery (H.H., K.K.), Pathology (E.I.), and Biochemistry (M.S.), School of Medicine, Keio University, Tokyo, Japan; and East Takarazuka Satoh Hospital (M.T.), Takarazuka, Japan

Received June 21, 2004; accepted August 4, 2004.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
A hemoglobin vesicle (HbV; diameter 252 ± 53 nm) or liposome-encapsulated Hb is an artificial oxygen carrier developed for use as a transfusion alternative, and its oxygen-transporting capacity has been well characterized, although critical physiological compartments for the Hb degradation after a massive infusion of HbV and the safety outcome remain unknown. In this study, we aimed to examine the compartments for its degradation by daily repeated infusions (DRI) of HbV, focusing on its influence on the reticuloendothelial system (RES). Male Wistar rats intravenously received the HbV suspension at 10 ml/kg/day for 14 consecutive days. The cumulative infusion volume (140 ml/kg) was equal to 2.5 times the whole blood volume (56 ml/kg). The animals tolerated the DRI well and survived, and body weights continuously increased. One day after DRI, hepatosplenomegaly occurred significantly through the accumulation of large amounts of HbV. Plasma clinical chemistry was overall normal, except for a transient elevation of lipid components derived from HbV. These symptoms subsided 14 days after DRI. Hemosiderin deposition and up-regulation of heme oxygenase-1 coincided in the liver and spleen but were not evident in the parenchyma of these organs. Furthermore, the plasma iron and bilirubin levels remained unchanged, suggesting that the heme-degrading capacity of the RES did not surpass the ability to eliminate bilirubin. In conclusion, phospholipid vesicles for the encapsulation of Hb would be beneficial for heme detoxification through their preferential delivery to the RES, a physiological compartment for degradation of senescent RBCs, even at doses greater than putative clinical doses.

In a series of safety studies of HbV, it has been clarified that the cellular structure and the size of the HbV are advantageous for maintaining a steady blood circulation without vasoconstriction and hypertension (Sakai et al., 2000a) and that the surface modification of HbV with polyethylene glycol (PEG) is beneficial not only for a longer circulation time (Phillips et al., 1999) but also for suppression of intervesicular aggregation of HbV during preservation for years and in the plasma phase in the peripheral tissues after intravenous infusion (Sakai et al., 1998, 2000b). In our previous report on the histopathological analysis of rats receiving a bolus HbV infusion (20 ml/kg), the HbV particles are recognized as foreign materials and finally captured mainly by the reticuloendothelial system (RES, or mononuclear phagocytic system) in the spleen and liver, and they are promptly degraded (Sakai et al., 2001). These are outstanding characteristics in comparison with molecular Hb that shows a shorter circulation time because it is filtered through the kidneys when the Hb concentration exceeds the haptoglobin concentration and induces hemoglobinuria and eventually renal failure, and it extravasates across the fenestrated endothelium in the liver and induces excess heme catabolism in the hepatocytes and marked sinusoidal constriction (Goda et al., 1998, Kyokane et al., 2001). However, it is not clear whether the physiological capacity of the RES for the degradation and excretion of the components of HbV would be sufficient even after a massive infusion of HbV. The circulation half-life of HbV is within a few days, which is significantly shorter than that of red blood cells (RBCs), and it is anticipated that a massive infusion of HbV would burden the RES and result in abrupt heme degradation and iron overload having the potential to cause deleterious effects.

One of the safety studies of a new drug in the preclinical stage should be a massive dose by daily repeated infusions (DRI) in rodents and nonrodents for at least 14 days at three different dosages; a guideline decided by the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use. However, the documentation of the DRI studies to the public is scarce, especially in the research field of artificial oxygen carriers (Biro and Greenburg, 1999). In a clinical setting, the amount of an artificial oxygen carrier to be infused should be at least several hundred milliliters, which is significantly greater than the dose of conventional drugs; therefore, it is not clear whether a preclinical protocol for a conventional drug is appropriate for the safety evaluation of artificial oxygen carriers. On the other hand, there may be a need for a repetitive infusion of an oxygen carrier in a clinical situation, such as chronic anemia (Hamilton et al., 2001) or cancer therapy (Teicher et al., 1997). Based on these backgrounds, we tested the DRI of HbV into Wistar rats at one dose rate as a preliminary study to confirm the safety of HbV. Because the dose amount of phospholipid vesicles for use as an oxygen carrier is significantly greater than that used for conventional drugs, the influence of a massive infusion of HbV on the RES and the excretion of the components, especially after heme degradation, are of great concern.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Preparation of HbV Suspension. The test fluid, the HbV suspension, was prepared under sterile conditions as reported previously (Sakai et al., 2000b; Sou et al., 2000, 2003). Human Hb was purified from outdated, donated blood provided by the Hokkaido Red Cross Blood Center (Sapporo, Japan) and the Japanese Red Cross Society (Tokyo, Japan). The encapsulated Hb (38 g/dl) contained 14.7 mM pyridoxal 5'-phosphate (Aldrich Chemical Co., Milwaukee, WI) as an allosteric effector at a molar ratio of pyridoxal 5'-phosphate/Hb = 2.5. The lipid bilayer was composed of a mixture of 1,2-dipalmitoyl-sn-glycero-3-phosphatidylcholine (DPPC), cholesterol, and 1,5-O-dihexadecyl-N-succynyl-L-glutamate (DHSG) at a molar ratio of 5:5:1 (Nippon Fine Chemicals Co., Osaka, Japan), and 1,2-distearoyl-sn-glycero-3-phosphatidylethanolamine-N-PEG₅₀₀₀ (NOF Co., Tokyo, Japan). Thus, the vesicular surface is covered with PEG chains. The molar composition of DPPC/cholesterol/DHSG/1,2-distearoyl-sn-glycero-3-phosphatidylethanolamine-N-PEG₅₀₀₀ was 5:5: 1:0.033. HbVs were suspended in a physiological salt solution, sterilized using filters (pore size, 0.45 µm, Dismic; Toyo Roshi Co., Tokyo, Japan), and deoxygenated with bubbling N₂ for storage (Sakai et al., 2000b). The physicochemical parameters of the HbV are as follows: particle diameter, 252 ± 53 nm; [Hb], 9.5 g/dl; [metHb], 2.3%; [HbCO], <2%; [lipids], 5.3 g/dl; and oxygen affinity (P₅₀), 30 Torr. The endotoxin content was measured by a modified Limulus amebocyte lysate gel-clotting analysis (Wako Pure Chemicals, Tokyo, Japan) and was less than 0.2 endotoxin unit/ml (Sakai et al., 2004a).

In our previous reports on resuscitation from hemorrhagic shock or extreme hemodilution, the HbV was suspended in a 5-g/dl albumin solution as a plasma expander to regulate the colloid osmotic pressure to 20 mm Hg (Sakai et al., 2004c). However, it is anticipated that the DRI of HbV suspended in albumin would result in enhanced hypervolemia. Because the main purpose of this DRI study was to clarify the safety of HbV and not albumin, HbV was simply suspended in a physiological saline solution.

Daily Repeated Infusion of HbV. All animal studies were approved by the Animal Subject Committee of the Keio University School of Medicine and performed according to National Institutes of Health Guidelines for the Care and Use of Laboratory Animals (NIH Publication #85-23 rev. 1985).

The experiments were carried out using 34 male Wistar rats (145 ± 4 g; Saitama Experimental Animals, Kawagoe, Japan). All the rats were housed in cages and provided with food and water ad libitum in a temperature-controlled room on a 12-h dark/light cycle. At first, the rats were anesthetized lightly with diethyl ether inhalation and then 1.5% sevoflurene (Maruishi Pharmaceutical Co., Osaka, Japan) using a vaporizer (model TK-4 Biomachinery; Kimura Medical, Tokyo, Japan) to immobilize them for every infusion. Every day for 14 days, the rats received HbV (n = 12) or saline (n = 12) via the tail vein using an indwelling needle (24-gauge; Nipro Co., Osaka, Japan) at a dose rate of 10 ml/kg with an injection rate of 1 ml/min. The total volume of the infused HbV into a rat for 14 days reached 140 ml/kg, which was equal to 2.5 times the actual blood volume of the rat (56 ml/kg). The infused total solid material (Hb and lipids) is calculated to be 20,689 mg/kg (1478 mg/kg/day x 14 days). The rats were weighed every day just before every infusion to calculate the amount of the infusion. After every infusion, the needle was immediately removed and the bleeding was stopped by applying pressure for a short time. The two groups (n = 12) were divided in half (n = 6 x 2), and six rats were sacrificed 1 day after the final 14th infusion. The remaining six rats were sacrificed at 14 days after the final infusion. Ten animals without the infusions were used to obtain control values.

Hematological Test. A hematological examination was performed at 1, 3, 7, 9, and 12 days during the DRI and at 1, 7, and 14 days after the final infusion. About 200 µl of blood was collected from a tail vein when an indwelling needle was inserted for HbV infusion. Seventy microliters was immediately diluted with 200 µl of citrate solution for a blood cell counter (Sysmex KX-21, Kobe, Japan), and the rest of the blood was inserted into a glass capillary (Terumo Co., Tokyo, Japan) for hematocrit (Hct) measurements. In this study, Hct indicates the volume of RBC and does not include the volume of HbV. The concentration of HbV in the plasma was measured by a cyanomethemoglobin method.

Hemodynamic and Blood Gas Parameters, Blood Glucose Level, and Urinalysis. One day or 2 weeks after the final infusion, the rats were anesthetized with 1.5% sevoflurene inhalation. A polyethylene tube (PE-50; Natsume Co., Tokyo, Japan) was inserted into the carotid artery for measurement of the mean arterial pressure (MAP), the heart rate (HR) by a recording system (Polygraph system 1000; Nippon Koden, Tokyo, Japan) and for withdrawing blood for various measurements. For the blood gas analysis, blood samples were collected in 70 IU/ml heparinized microtubes (125 µl, Clinitubes; Radiometer Nederland, Copenhagen, Denmark) and injected into a pH/blood gas analyzer (model ABL 555; Radiometer Nederland) for analyses of the arterial blood O₂ tension, arterial blood carbon dioxide tension, pH, base excess, and lactate. The blood glucose level was measured with a Medisafe Reader (GR-101; Terumo Co., Tokyo, Japan). Urinalysis was performed by dip-sticktesting (UA-L08M; Terumo Co.) as a qualitative measurement. A urine specimen of a rat was collected in a transparent plastic bag when the rat was lightly anesthetized with diethyl ether, and a test stick was dipped in the collected urine. In each item, the levels were judged by visual examination of the color identification after a specific time of exposure according to the instructions, in the order of protein (10 s), pH (10 s), occult blood (20 s), ketone body (20 s), urobilinogen (20 s), glucose (30 s), nitrite (30 s), and bilirubin (40 s).

Plasma Clinical Chemistry. A part of the withdrawn blood (6 ml) was centrifuged to obtain plasma that was turbid and red/brown colored due to the presence of PEG-modified HbV particles, especially in the samples taken one day after DRI. The plasma was ultracentrifuged (50,000g; 20 min) to remove the HbV particles (Sakai et al., 2003). The obtained transparent plasma specimens were stored in a freezer at –80°C until the clinical chemistry tests (BML, Kawagoe, Japan). The selected analytes were total protein, albumin, total bilirubin, aspartate aminotransferase, alanine aminotransferase, -glutamyltransferase, alkaline phosphatase, cholinesterase, leucine amino peptidase, creatine phosphokinase, amylase, lipase, aldosterone, total cholesterol, cholesterol ester, free cholesterol, HDL-cholesterol, -lipoprotein, triglyceride, free fatty acid, phospholipids, total lipids, uric acid (UA), urea nitrogen (BUN), creatinine (CRE), K⁺, Ca²⁺, inorganic phosphate, unsaturated iron binding capacity, and Fe³⁺. All the analytical methods were described in our previous articles (Sakai et al., 2003, 2004b).

Blood Coagulation Test and Fibrinogen Concentration. For the blood coagulation test at 1 and 14 days after the final infusion of HbV or saline, 1.8 ml of the withdrawn blood was immediately mixed with 0.2 ml of 3.8% sodium citrate solution. The plasma fraction, obtained by gentle centrifugation, was analyzed for prothrombin time (PT), activated partial thromboplastin time (APTT), and fibrinogen (BML).

Histopathological Examination. The animals were finally laparotomized and sacrificed by acute bleeding from the abdominal aorta, and the liver, spleen, and kidney were resected for weight measurements and also all the other organs were obtained for a histopathological study. They were fixed in 10% buffered formalin (Wako Pure Chemicals) immediately after removal, and the paraffin sections were stained with hematoxylin & eosin, and Berlin blue.

Immunohistochemistry was performed to detect rat heme oxygenase-1 (HO-1) and human Hb from the injected HbV in the rat spleen and liver. Four-micrometer-thick paraffin sections were mounted on 3-aminopropyl triethoxysilane-coated glasses. The sections were treated with 0.03% H₂O₂ in methanol for 10 min at room temperature to block the endogenous peroxidase activity. For antigen retrieval, the sections were also treated with proteinase K (0.4 mg/ml; DakoCytomation California Inc., Carpinteria, CA) for 10 min at room temperature. After blocking the nonspecific binding with 5% normal goat serum, they were incubated with mouse monoclonal antibody against rat HO-1 (20 µg/ml; GTS-3, TaKaRa, Tokyo, Japan) at 4°C overnight. They were then incubated for 30 min at room temperature with goat antibodies against mouse immunoglobulins conjugated to the amino acid polymer [no dilution; Histofine Simple Stain MAX-PO(M), Nichirei Co., Tokyo, Japan]. Color was developed with 3,3'-diaminobenzidine tetrahydrochloride (0.2 mg/ml; Dojindo Laboratories, Kumamoto, Japan) in 0.05 M Tris-HCl, pH 7.4, containing 0.003% hydrogen peroxide. Subsequently, the sections were treated with 5% normal swine serum for 30 min at room temperature and reacted with rabbit polyclonal antibodies against human Hb (1:500 dilution; DakoCytomation A/S, Glostrup, Denmark) for 60 min at room temperature. They were further incubated with alkaline phosphatase-conjugated swine antibodies against rabbit immunoglobulins (1:100 dilution; DakoCytomation A/S). Color development was performed using a New Fuchsin Substrate kit (Nichirei Co.), and the sections were counterstained with hematoxylin.

Data Analysis. Differences between the control and the treatment group were analyzed using a one-way analysis of variance followed by Fisher's protected least significant difference test. The changes were considered statistically significant if p < 0.05. All the data are shown as mean ± S.D. For the results of the plasma clinical chemistry, the allowance of twice the standard deviation (2x S.D.) of the baseline values is indicated in the figures in considering the variable nature of these parameters.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Body Weight. The body weight of rats in the HbV group (baseline, 144 ± 3 g) showed a monotonous increase during the 14 days of the DRI period and reached 195 ± 12 g (Fig. 1); however, this was slightly but significantly suppressed (p < 0.05) in comparison with the control saline group (220 ± 13 g). The body weight in the HbV group increased to 265 ± 14 g at 14 days after DRI. No significant difference was noted in the body weight compared with that of the saline control group (280 ± 22 g).

View larger version (24K): [in this window] [in a new window]   Fig. 1. Time course of the gain in body weight during and after DRI of HbV and saline for 14 days at a dose rate of 10 ml/kg/day. Both groups showed monotonous increases; however, after the 5th day, a significant difference was observed. Seven days after the final infusion, there were no significant differences between the two groups. The values are mean ± S.D. *, significantly different between the groups (p < 0.05).

Hematological Changes and Concentration of HbV in Blood. The Hct of the HbV group (baseline, 41.7 ± 2.1%) tended to decrease to 37.5 ± 0.9% 1 day after DRI, which was lower than that of the saline group (44.7 ± 2.0%) (Fig. 2). However, after 14 days, the Hct of the HbV group increased to 45.1 ± 1.9%, which was comparable with that of the saline group (47.8 ± 2.7%). The numbers of white blood cells and platelets were comparable with those of the saline control group throughout the observation period.

View larger version (25K): [in this window] [in a new window]   Fig. 2. Time course of Hct and blood cell counts during and after DRI of HbV and saline for 14 days at a dose rate of 10 ml/kg/day. The values are mean ± S.D. #, significantly different between the groups (p < 0.05). *, significantly different versus the baseline values (p < 0.05).

The concentration of HbV immediately after every infusion was estimated from the volumes of the whole blood (56 ml/kg) and the infused volume of HbV (10 ml/kg) and was plotted with open circles, and the measured concentration of HbV just before every infusion was plotted with solid circles (Fig. 3). The concentration of HbV just after the first infusion was estimated to be about 3 g/dl, and 1 day later, it decreased to 1.3 g/dl, with a half-life of about 22 h. The half-life of the second infusion seemed to become shorter. The DRI resulted in the accumulation of HbV in the plasma and it increased to 5.2 g/dl 1 day after the 14th infusion. However, 1 week after the final infusion, no HbV was confirmed in the plasma, indicating that all of the HbV was captured by the RES.

View larger version (14K): [in this window] [in a new window]   Fig. 3. Time course of the HbV concentration in the plasma phase during and after DRI of HbV for 14 days at a dose rate of 10 ml/kg/day. It is assumed that immediately after the first infusion, the concentration should be around 3 g/dl, and 1 day later, it decreased to about 1.2 g/dl. Immediately after the second infusion, the concentration should be about 4.3 g/dl and decreased to 2 g/dl 1 day later. The half-life of the HbV apparently decreases with multiple infusions. The concentration tended to reach a plateau. The final HbV concentration reached 5.3 g/dl, and this completely disappeared 7 days after DRI.

Coagulation Test. A coagulation test indicated that the HbV group showed a slight prolongation in APTT from 24.3 ± 1.3 s at the baseline to 36.0 ± 11.8 s 1 day after the DRI but changed to 40.3 ± 20.3 s after 14 days with marked individual variations. On the other hand, there was no noticeable change in the PT for the HbV group (Fig. 4). The fibrinogen concentration (baseline, 223 ± 12 mg/dl) significantly increased for all groups (HbV group, 1 day after DRI, 289 ± 22 mg/dl; 14 days after DRI, 255 ± 11 mg/dl), probably due to the stress of infusion and influence on the liver function.

View larger version (18K): [in this window] [in a new window]   Fig. 4. Parameters for blood coagulation, APTT and PT, and fibrinogen concentration after DRI of HbV and saline for 14 days at a dose rate of 10 ml/kg/day. The values are mean ± S.D. *, significantly different versus the baseline group (p < 0.05).

Blood Pressure, Heart Rate, and Blood Gas Parameters. Table 1 summarizes the blood gas parameters, blood glucose level, MAP, and HR 1 day after the DRI of HbV. There is no abnormal value except for an increase in MAP of the HbV group 1 day after DRI (125 ± 4 mm Hg) in comparison with the baseline value (101 ± 8 mm Hg).

Urinalysis. The color of the urine was normal for all groups, and there was no sign of hemoglobinurea. Table 2 summarizes the results of the urinalysis. The HbV group showed a slight increase in the protein concentration. There were no significant signs of any organ damage. Urobilinogen and bilirubin were within the normal range in spite of the large amount of HbV infusion.

Organ Weights. The liver and spleen are thought to be the main organs that trap and degrade HbV. As shown in Fig. 5, significant splenomegaly and hepatomegaly were confirmed 1 day after DRI. The percentage of spleen weight relative to the body weight increased from 0.33 ± 0.04% at the baseline to 0.94 ± 0.07 1 day after DRI, about 2.9 times the baseline value. This returned to 0.41 ± 0.03% after 14 days. The percentage of liver weight relative to the body weight increased from 4.81 ± 0.15% at the baseline to 5.83 ± 0.37% 1 day after DRI; and it returned to 4.33 ± 0.20%, comparable with the baseline after 14 days. The color of the liver was darkened just after DRI, however, it returned to its normal color 14 days after DRI. The kidney weight did not show any significant increase but tended to show a slight decrease for all groups.

View larger version (19K): [in this window] [in a new window]   Fig. 5. Changes in organ weights (spleen, liver, and kidney) after DRI of HbV and saline for 14 days at a dose rate of 10 ml/kg/day. The values are mean ± S.D. *, significantly different versus the baseline group (p < 0.05).

Plasma Clinical Chemistry. A significant amount of the HbV particles was present in the plasma one day after DRI. However, they could be easily removed from the plasma by ultracentrifugation (50,000g; 20 min) (Sakai et al., 2003), and we could avoid any interference effect of HbV in the colorimetric and turbidimetric analyses in the plasma clinical chemistry. The parameters affecting the liver function (total protein, albumin, aspartate aminotransferase, alanine aminotransferase, leucine amino peptidase, alkaline phosphatase, -glutamyltransferase, and cholinesterase) did not show any noteworthy changes in the HbV group (Fig. 6).

View larger version (55K): [in this window] [in a new window]   Fig. 6. Plasma clinical chemistry tests reflecting the organ functions such as liver, pancreas, and kidneys and the metabolism of Hb after DRI of HbV or saline. The values are mean ± S.D. *, significantly different versus the baseline group (p < 0.05). The dotted lines indicate the levels of 2x S.D. LAP, leucin amino peptidase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; ChE, cholinesterase; ALP, alkaline phosphatase; -GTP, -glutamyltransferase; CPK, creatine phosphokinase; CRE, creatinine; UA, uric acid; BUN, urea nitrogen.

The parameters affecting the function of the kidneys (CRE, uric acid, and BUN) varied within the normal ranges. The parameter reflecting the heart and skeletal muscle, creatine phosphokinase, did not show any noticeable change. Aldosterone, released from adrenal gland to regulate ionic balance, did not show noticeable change. A significant change was observed for the parameters of pancreatic function. In the HbV group, the lipase activity (baseline, 8.5 ± 1.4 U/l) increased to 48.5 ± 16.8 U/l 1 day after DRI and tended to decrease to 33.2 ± 29.4 U/l after 14 days. On the contrary, amylase activity (baseline, 1613 ± 74 U/l) did not show an increase but a slight decrease to 1455 ± 28 U/l 1 day after DRI and returned to 1546 ± 77 U/l after 14 days.

The concentrations of the cholesterol components (total and free cholesterols, and cholesterol ester) and lipids (-lipoprotein, total lipids, and phospholipids) significantly increased 1 day after the final infusion (Fig. 7). For example, total cholesterol (baseline, 72.6 ± 7.5 mg/dl) increased to 182.2 ± 22.6 mg/dl after DRI. However, they returned to the original values 14 days after DRI. These increases should indicate that cholesterol and phospholipid (probably DPPC) are released from the RES after entrapping the HbV particles. Bilirubin and ferric iron, which should be released from the Hb decomposition, were minimal. Unsaturated iron binding capacity did not show noticeable changes. The electrolyte concentrations varied within the normal range.

View larger version (59K): [in this window] [in a new window]   Fig. 7. Plasma clinical chemistry tests reflecting the metabolism of lipids and Hb and electrolytes 1 or 14 days after DRI of HbV or saline. The values are mean ± S.D. *, significantly different versus the baseline group. The dotted lines indicate the levels of 2x S.D. TG, triglyceride; FFA, free fatty acid; UIBC, unsaturated iron-binding capacity; IP, inorganic phosphate.

Histopathological Examination. Histopathological examination 1 day after DRI showed a significant amount of HbV accumulated in the red pulp zone in the spleen, and very few RBCs were seen (Fig. 8a). In the liver, the presence of Kupffer cells that captured a large amount of HbV was seen (Fig. 8b). In the kidneys, the mesangial cells in the renal glomerulus seemed to entrap HbV (Fig. 8c). These organs and the adrenal gland were slightly stained with Berlin blue (data not shown), indicating that the decomposition of heme should have already started. No morphological change was noted in the myocardium; however, some slightly stained particles were observed (Fig. 8d). The pancreas (Fig. 8e), lungs (Fig. 8f), intestine, stomach, brain, thymus, testis, and skin did not show significant abnormalities.

View larger version (92K): [in this window] [in a new window]   Fig. 8. Histology of spleen (a), liver (b), kidneys (c), heart (d), pancreas (e), and lungs (f) 1 day after DRI of HbV. A significant amount of HbV was accumulated in the red pulp zone of the spleen. The invasion of a significant number of Kupffer cells with HbV was seen in the liver. In the kidneys, the mesangial cells in the renal glomeruli seemed to entrap HbV. The myocardium showed slight staining with Berlin blue. No significant pathological changes are noted in the pancreas and lungs. Scale bar, 100 µm. Hematoxylin and eosin stains (a, b, c, e, and f) and Berlin blue stain (d).

Fourteen days after DRI, the images of the accumulated HbV almost disappeared in all organs. However, there were materials that were moderately stained with Berlin blue in the red pulp zone of the spleen (Fig. 9a), liver (Fig. 9b), bone marrow (Fig. 9c), and slightly in the kidney (Fig. 9d) and adrenal gland (Fig. 9e)

View larger version (98K): [in this window] [in a new window]   Fig. 9. Histology of spleen (a), liver (b), bone marrow (c), kidneys (d), and adrenal gland (e) 14 days after DRI. Berlin blue staining was performed to examine the presence of hemosiderin. Scale bar, 100 µm.

Immunohistochemical analysis of the liver and the spleen clarified the presence of human Hb in HbV as pink-colored areas that were stained with anti-human Hb-antibody 1 day after DRI (Fig. 10, a and b). In the spleen, the presence of HO-1 was confirmed as brown-colored stains in the cytoplasm of the macrophages in the red pulp zone both at 1 and 14 days after DRI (Fig. 10, a and c). In the liver, the presence of HO-1 was confirmed in the Kupffer cells only at 14 days after DRI (Fig. 10d). No HO-1 was confirmed in the parenchyma of these organs.

View larger version (171K): [in this window] [in a new window]   Fig. 10. Double immunohistochemical staining for HO-1 and human Hb in HbV in the rat spleen (a and c) and liver (b and d), 1 (a and b) and 14 days (c and d) after DRI of HbV. The tissues were stained with anti-rat HO-1 monoclonal antibody (GTS-3). The brown-colored portions (a, c, and d) indicate the presence of HO-1, and the pink or gray-beige areas (a and b) indicate the presence of a large amount of HbV. Scale bar, 100 µm.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
Our primary finding is that all the rats tolerated the DRI of HbV well for 14 days with no deteriorative signs in organ functions, due to the preferable effect of Hb encapsulation in phospholipid vesicles that minimizes the toxicity of molecular Hbs and delivers them to the RES as a physiological compartment for degradation and detoxification of foreign materials. The RES had sufficient capacity for the degradation of HbV, even though the total infused volume reached 140 ml/kg, which was equal to 2.5 times the actual blood volume of the rat (56 ml/kg) and was significantly larger than the dose of multiple infusions of liposomes for antifungal and antitumor targeting (Fielding et al., 1999; Charrois and Allen, 2003).

The body weight of the HbV group monotonously increased, whereas the rate was slightly slower than that of the saline control group. It is speculated that the infusion of HbV, which could not be excreted easily in the urine and remain in circulation, could 1) disturb physiological functions and suppress the growth of the animals, 2) put the animals under stress and reduce their appetite, or 3) tend to accelerate the catabolism. In spite of such a condition, the components of HbV could be used as a part of the cellular components for the growth of rats. One and 2 weeks after DRI, there was no significant difference in the body weight between the groups.

The numbers of RBCs, whole blood cells, and platelets showed moderate changes, even though there were some significant differences between the HbV and the saline groups. Hct and RBC counts decreased significantly for the HbV group, probably due to the dilution of blood by hypervolemia, or suppression of erythropoiesis (release of erythropoietin) because the renal cortex would be exposed to the increased oxygen content in the blood during DRI of HbV as oxygen carriers. The slight hypertension 1 day after DRI would be related to the blood hyperviscosity or hypervolemia due to the presence of HbV. However, the Hct and RBC counts returned to levels similar to those of the saline group 14 days after DRI. The time course of the HbV concentration in plasma indicates that the rate of HbV clearance gradually increased and the concentration reached a plateau, probably due to the nonspecific phagocytic activation of the RES that was clarified previously by a carbon clearance measurement (Sakai et al., 2001). The accelerated liposome clearance of the second infusion was well characterized (Claassen et al., 1988; Laverman et al., 2001); however, its mechanism, antibody formation or complement activation is controversial (Dams et al., 2000; Ishida et al., 2003).

In our previous report, the bolus HbV infusion (20 ml/kg) resulted in significant splenomegaly (about 100% increase) and hepatomegaly (13%) (Sakai et al., 2004b). In the present DRI study, splenomegaly was enhanced (190%), whereas hepatomegaly was similar (14%), indicating that the spleen had a larger capacity for HbV clearance. A large amount of HbV accumulated in the red pulp zone of the spleen and in Kupffer cells of the liver; however, 14 days later it disappeared and the splenohepatomegaly completely subsided. The spleen and the liver showed significant hemosiderin deposition; however, the enzyme concentrations that reflect the liver function did not show any abnormal values.

One day after DRI, the mesangial cells in the renal glomerulus seemed to entrap HbV in their intracellular spaces, and the same portion was stained with Berlin blue 1 and 14 days after DRI. In our previous report on the bolus HbV infusion, there was no abnormality in the kidneys (Sakai et al., 2004b). According to Rudolph et al. (1995), liposomeencapsulated Hb without PEG-modification aggregated in the plasma and showed a slight accumulation in the kidneys. Even though our PEG-modified HbV does not induce intervesicular aggregation, HbV would tend to be aggregated during the longer circulation time due to the DRI. No abnormal value was noted for UA, BUN, and CRE, although urinalysis showed a slight increase in protein levels.

Lipase activity, but not that of amylase, significantly increased in the HbV group, whereas there was no histopathological abnormality in the pancreas. A similar tendency was observed after the bolus HbV infusion (20 ml/kg) (Sakai et al., 2004b). This level of increment was significantly smaller than the value for the Wistar rats with acute necrotizing pancreatitis that increased the lipase activity from 10 to 475 to 5430 IU/l (Hofbauer et al., 1996). One possible reason for the moderate and specific increase in lipase activity would be related to the enzyme induction in the pancreas by the presence of a large amount of lipids from the liposomes (Stuecklin-Utsch et al., 2002), because pancreatic lipase hydrolyzes not only triglyceride but also phosphatidylcholine (Rowland and Woodley, 1980). However, the mechanism is not clear, and the pancreatic function should be carefully monitored in the ongoing safety studies.

The plasma lipid components significantly increased after the DRI of HbV. They should be derived from HbV because it contains a large amount of cholesterol and DPPC, and they would be liberated after the HbV particles are captured and degraded in the RES. It is reported that once liposome is captured in the Kupffer cells, the diacylphosphatidylcholine is metabolized and is reused as a cell membrane component or excreted in the bile (Dijkstra et al., 1985; Verkade et al., 1991). Cholesterol is finally catabolized as bile acids in the parenchymal hepatocytes. There should be no direct contact of HbV and the hepatocytes because HbV (diameter, 250 nm) cannot diffuse across the fenestrated endothelium into the space of Disse (Goda et al., 1998). Cholesterol of the vesicles should reappear in the blood mainly as lipoprotein cholesterol after entrapment in the Kupffer cells and should then be excreted in the bile after entrapment of the lipoprotein cholesterol by the hepatocytes (Kuipers et al., 1986). Judging from the results showing that the increases in the plasma lipid components were transient, the lipid components of HbV would gradually be redistributed, metabolized, and excreted in the same manner within 14 days after DRI. However, the details have to be confirmed by the biodistribution of the radiolabeled components.

In spite of the massive HbV infusions, the plasma bilirubin and iron levels did not increase. Urinalysis also showed no increase in the urobilinogen and bilirubin. The anti-human Hb antibody staining detected temporal distributions of HbV in the spleen and liver. The excess amount of heme from Hb in HbV should be metabolized by the inducible form of HO-1 in the spleen macrophages and the liver Kupffer cells, as shown in Fig. 10 (Braggins et al., 1986; Goda et al., 1998). Bilirubin should be excreted in the bile as a normal physiological pathway even during the massive doses of HbV. No increase in the plasma bilirubin level indicated that there was no obstruction or stasis of bile in the biliary tree and that the heme-degrading capacity of the RES did not surpass the ability to eliminate bilirubin. Berlin blue staining revealed the presence of hemosiderin in the liver, spleen, kidneys, adrenal gland, and bone marrow 14 days after DRI and also in the myocardium 1 day after DRI. Both ferritin and hemosiderein store and release iron molecules, and they are anticipated to induce hydroxyl radical production and succeeding lipid peroxidation. However, iron release from hemosiderin is substantially less than that from ferritin, thus iron molecules in hemosiderin are relatively inert (O'Connell et al., 1989). Multiple blood transfusions often induce hemosiderosis in many organs. Accordingly, Hb encapsulation in the phospholipid vesicles would guarantee the smooth metabolic route of HbV that is similar to the well characterized metabolic route of senescent RBCs in the liver Kupffer cells and spleen macrophages (Bennett and Kay, 1981; Hirano et al., 2001). This would be a great advantage over molecular Hb that incurs not only filtration across the fenestrated endothelium of the glomerular capillary in the kidneys resulting in shorter circulation time and renal failure but also extravasation from the sinusoidal caliber in the liver, causing cancellation of the CO-mediated fail-safe mechanism for conserving sinusoidal patency and bile formation (Kyokane et al., 2001).

In conclusion, all the rats tolerated the DRI of HbV with no deteriorative signs of the organ functions. The phospholipid vesicles for Hb encapsulation would be beneficial for heme detoxification through their preferential delivery to the RES, a physiological compartment for degradation of not only foreign materials but also the senescent RBCs. However, it has to be considered that in humans the circulation time of HbV and its degradation rate in the RES would be different compared with those in rats, because the circulation time of stealth liposomes and the life span of RBCs are different between rodents and humans (Landaw, 1988; Gabizon et al., 2003). A shock condition may also influence on the RES function.

Our results would provide important information not only for the ongoing safety studies of HbV but also for the overall research on liposomal drugs, because this study is the first attempt to infuse repetitively such a large amount of phospholipid vesicles.

	Acknowledgements

 
We acknowledge researchers in the School of Medicine, Keio University; H. Abe, T. Yamaguchi, and S. Kurasaki (Department of Pathology) for excellent histopathological techniques; and Dr. Y. Izumi, Dr. M. Watanabe, and T. Ohba (Department of Surgery) and Dr. M. Kajimura (Department of Biochemistry) for discussions on the experimental procedures.

	Footnotes

 
This work was supported by Health Sciences Research grants (Research on Pharmaceutical and Medical Safety, Artificial Blood Project); the Ministry of Health, Labor and Welfare, Japan; grants-in-aid for Scientific Research from the Japan Society for the Promotion of Science (B16300162); 21 COE "Practical Nano-Chemistry" from the Ministry of Education, Culture, Sports, Science and Technology of Japan; and Oxygenix Inc. (Tokyo, Japan). S.T., M.S., K.K., and E.T. are consultants of Oxygenix Inc.

doi:10.1124/jpet.104.073049.

ABBREVIATIONS: HbV, Hb vesicle(s); Hct, hematocrit; RES, reticuloendothelial system; PEG, poly(ethylene glycol); RBC, red blood cell; DRI, daily repeated infusion(s); DPPC, 1,2-dipalmitoyl-sn-glycero-3-phosphatidylcholine; DHSG, 1,5-O-dihexadexyl-N-syccynyl-L-glutamate; MAP, mean arterial pressure; HR, heart rate; UA, uric acid; BUN, urea nitrogen; CRE, creatinine; PT, prothrombin time; APTT, activated partial thromboplastin time; HO-1, heme oxygenase-1.

Address correspondence to: Prof. Eishun Tsuchida, Advanced Research Institute for Science and Engineering, Waseda University, Tokyo 169-8555, Japan. E-mail: eishun{at}waseda.jp

	References

Top Abstract Materials and Methods Results Discussion References

 

Bennett GD and Kay MM (1981) Homeostatic removal of senescent murine erythrocytes by splenic macrophages. Exp Hematol 9: 297–307.[Medline]

Biro GP and Greenburg AG (1999) Safety toxicology evaluation of o-raffinose cross linked hemoglobin solution by daily repeated infusions in rats and dogs (Abstract). Crit Care Med 27 (Suppl): 479.

Braggins PE, Trakshel GM, Kutty RK, and Maines MD (1986) Characterization of two heme oxygenase isoforms in rat spleen: comparison with the hematin-induced and constitutive isoforms of the liver. Biochem Biophys Res Commun 141: 528–533.[CrossRef][Medline]

Chang TM, Lister C, Nishiya, and Varma R (1992) Immunological effects of hemoglobin, encapsulated hemoglobin, polyhemoglobin and conjugated hemoglobin using different immunization schedules. Biomater Artif Cells Immobil Biotechnol 20: 611–618.[Medline]

Charrois GJR and Allen TM (2003) Multiple injection of pegylated liposomal doxorubicin: pharmacokinetics and therapeutic activity. J Pharmacol Exp Ther 306: 1058–1067.[Abstract/Free Full Text]

Claassen E, Westerhof Y, Versluis B, Kors N, Schellekens M, and van Rooijen N (1988) Effect of chronic injection of sphingomyelin-containing liposomes on lymphoid and non-lymphoid cells in the spleen. Transient suppression of marginal zone macrophages. Br J Exp Pathol 69: 865–875.[Medline]

Dams ETM, Laverman P, Oyen WJG, Storm G, Scherphof GL, van der Meer JWM, Corsten FHM, and Boerman OC (2000) Accelerated blood clearance and altered biodistribution of repeated injections of sterically stabilized liposomes. J Pharmacol Exp Ther 292: 1071–1079.[Abstract/Free Full Text]

Dijkstra J, van Galen M, Regts D, and Scherphof G (1985) Uptake and processing of liposomal phospholipids by Kuppfer cells in vitro. Eur J Biochem 148: 391–397.[Medline]

Djordjevich L, Mayoral J, Miller IF, and Ivankovich AD (1987) Cardiorespiratory effects of exchanging transfusions with synthetic erythrocytes in rats. Crit Care Med 15: 318–323.[Medline]

Fielding RM, Moon-Mcdermott L, Lewis RO, and Horner MJ (1999) Pharmacokinetics and urinary excretion of amikacin in low-clearance unilamellar liposomes after a single or repeated intravenous administration in the rhesus monkey. Antimicrob Agents Chemother 43: 503–509.[Abstract/Free Full Text]

Gabizon A, Shmeeda H, and Barenholz Y (2003) Pharmacokinetics of pegylated liposomal Doxorubicin: review of animal and human studies. Clin Pharmacokinet 42: 419–436.[CrossRef][Medline]

Goda N, Suzuki K, Naito S, Takeoka S, Tsuchida E, Ishimura Y, Tamatani T, and Suematsu M (1998) Distribution of heme oxygenase isoform in rat liver: topographic basis for carbon monoxide-mediated microvascular relaxation. J Clin Investig 101: 604–612.[Medline]

Hamilton RG, Kelly N, Gawryl MS, and Rentko VT (2001) Absence of immunopathology associated with repeated IV administration of bovine Hb-based oxygen carrier in dogs. Transfusion 41: 219–225.[CrossRef][Medline]

Hirano K, Kobayashi T, Watanabe T, Yamamoto T, Hasegawa G, Hatakeyama K, Suematsu M, and Naito M (2001) Role of heme oxygenase-1 and Kupffer cells in the production of bilirubin in the rat liver. Arch Histol Cytol 64: 169–178.[CrossRef][Medline]

Hofbauer B, Friess H, Weber A, Baczako, Kisling P, Schilling M, Uhl W, Dervenis C, and Buchler MW (1996) Hyperlipaemia intensifies the course of acute oedematous and acute necrotising pancreatitis in the rat. Gut 38: 753–758.[Abstract/Free Full Text]

Ishida T, Maeda R, Ichihara M, Irimura K, and Kiwada H (2003) Accelerated clearance of PEGylated liposomes in rats after repeated infusion. J Controlled Release 88: 35–42.[CrossRef][Medline]

Izumi Y, Sakai H, Hamada K, Takeoka S, Yamahata T, Kato R, Nishide H, Tsuchida E, and Kobayashi K (1997) Physiologic responses to exchange transfusion with hemoglobin vesicles as an artificial oxygen carrier in anesthetized rats: changes in mean arterial pressure and renal cortical tissue oxygen tension. Crit Care Med 24: 1869–1873.[CrossRef]

Kuipers F, Spanjer HH, Havinga R, Scherphof GL, and Vonk RJ (1986) Lipoproteins and liposomes as in vivo cholesterol vehicles in the rat: preferential use of cholesterol carried by small unilamellar liposomes for the formation of muricholic acids. Biochim Biophys Acta 876: 559–566.[Medline]

Kyokane T, Norimuzu S, Taniai H, Yamaguchi T, Takeoka S, Tsuchida E, Naito M, Nimura Y, Ishimura Y, and Suematsu M (2001) Carbon monoxide from heme catabolism protects against hepatobiliary dysfunction in endotoxin-treated rat liver. Gastroenterology 120: 1227–1240.[CrossRef][Medline]

Landaw SA (1988) Factors that accelerate or retard red blood cell senescence. Blood Cells 14: 47–59.[Medline]

Laverman P, Carstens MG, Boerman OC, Dams ETM, Oyen WJG, Rooijen NV, Corstens FHM, and Storm G (2001) Factors affecting the accelerated blood clearance of polyethylene glycol-liposomes on repeated injection. J Pharmacol Exp Ther 298: 607–612.[Abstract/Free Full Text]

Lian T and Ho RJY (2001) Trends and developments in liposome drug delivery systems. J Pharm Sci 90: 667–680.[CrossRef][Medline]

O'Connell MJ, Ward RJ, Baum H, and Peters TJ (1989) Iron release from haemosiderin and ferritin by therapeutic and physiological chelators. Biochem J 260: 903–907.[Medline]

Phillips WT, Klipper RW, Awasthi VD, Rudolph AS, Cliff R, Kwasiborski V, and Goins BA (1999) Polyethylene glycol-modified liposome-encapsulated hemoglobin: a long circulating red cell substitute. J Pharmacol Exp Ther 288: 665–670.[Abstract/Free Full Text]

Rowland RN and Woodley JF (1980) The stability of liposomes in vivo to pH, bile salts and pancreatic lipase. Biochim Biophys Acta 620: 400–409.[Medline]

Rudolph AS, Spielberg H, Spargo BJ, and Kossovsky N (1995) Histopathologic study following administration of liposome-encapsulated hemoglobin in the normovolemic rat. J Biomed Mater Res 29: 189–196.[CrossRef][Medline]

Sakai H, Hara H, Yuasa M, Tsai AG, Takeoka S, Tsuchida E, and Intaglietta M (2000a) Molecular dimensions of Hb-based O₂ carriers determine constriction of resistance arteries and hypertension in conscious hamster model. Am J Physiol 279: H908–H915.

Sakai H, Hisamoto S, Fukutomi I, Sou K, Takeoka S, and Tsuchida E (2004a) Detection of lipopolysaccharide in hemoglobin-vesicles by Limulus amebocyte lysate test with kinetic-turbidimetric gel clotting analysis and pretreatment with a surfactant. J Pharm Sci 93: 310–321.[CrossRef][Medline]

Sakai H, Horinouchi H, Masada Y, Takeoka S, Kobayashi K, and Tsuchida E (2004b) Metabolism of hemoglobin-vesicles (artificial oxygen carriers) and their influence on organ functions in a rat model. Biomaterials 25: 4317–4325.[CrossRef][Medline]

Sakai H, Horinouchi H, Tomiyama K, Iikeda E, Takeoka S, Kobayashi K, and Tsuchida E (2001) Hemoglobin-vesicles as oxygen carriers: influence on phagocytic activity and histopathological changes in metabolism. Am J Pathol 159: 1079–1088.[Abstract/Free Full Text]

Sakai H, Masada Y, Horinouchi H, Yamamoto M, Ikeda E, Takeoka S, Kobayashi K, and Tsuchida E (2004c) Hemoglobin-vesicles suspended in recombinant human serum albumin for resuscitation from hemorrhagic shock in anesthetized rats. Crit Care Med 32: 539–545.[CrossRef][Medline]

Sakai H, Tomiyama K, Masada Y, Takeoka S, Horinouchi H, Kobayashi K, and Tsuchida E (2003) Pretreatment of serum containing Hb-vesicles (oxygen carriers) to avoid their interference in laboratory tests. Clin Chem Lab Med 41: 222–231.[CrossRef][Medline]

Sakai H, Tomiyama K, Sou K, Takeoka S, and Tsuchida E (2000b) Polyethyleneglycol-conjugation and deoxygenation enable long-term preservation of hemoglobinvesicles as O₂ carriers in a liquid state. Bioconjug Chem 11: 425–432.[CrossRef][Medline]

Sakai H, Tsai AG, Kerger H, Takeoka S, Tsuchida E, and Intaglietta M (1998) Subcutaneous microvascular responses to hemodilution with a red cell substitute consisting of polyethyleneglycol-modified vesicles encapsulating hemoglobin. J Biomed Mater Res 40: 66–78.[CrossRef][Medline]

Sou K, Endo T, Takeoka S, and Tsuchida E (2000) Poly(ethylene glycol)-modification of the phospholipid vesicles by using the spontaneous incorporation of poly(ethylene glycol)-lipid into the vesicles. Bioconjug Chem 11: 372–379.[CrossRef][Medline]

Sou K, Naito Y, Endo T, Takeoka S, and Tsuchida E (2003) Effective encapsulation of proteins into size-controlled phospholipid vesicles using freeze-thawing and extrusion. Biotechnol Prog 19: 1547–1552.[CrossRef][Medline]

Stuecklin-Utsch A, Hasan C, Bode U, and Fleischhack G (2002) Pancreatic toxicity after liposomal amphotericin B. Mycoses 45: 170–173.[CrossRef][Medline]

Teicher BA, Ara G, Herbst R, Takeuchi H, Keyes S, and Northey D (1997) PEG-hemoglobin: effects on tumor oxygenation and response to chemotherapy. In Vivo 11: 301–311.[Medline]

Verkade HJ, Derksen JT, Gerding A, Scherphof GL, Vonk RJ, and Kuipers F (1991) Differential hepatic processing and biliary secretion of head-group and acyl chains of liposomal phosphatidylcholines. Biochem J 275: 139–144.

This article has been cited by other articles:

				K. Taguchi, Y. Urata, M. Anraku, T. Maruyama, H. Watanabe, H. Sakai, H. Horinouchi, K. Kobayashi, E. Tsuchida, T. Kai, et al. Pharmacokinetic Study of Enclosed Hemoglobin and Outer Lipid Component after the Administration of Hemoglobin Vesicles as an Artificial Oxygen Carrier Drug Metab. Dispos., July 1, 2009; 37(7): 1456 - 1463. [Abstract] [Full Text] [PDF]

				J. A. Plock, C. Contaldo, H. Sakai, E. Tsuchida, M. Leunig, A. Banic, M. D. Menger, and D. Erni Is hemoglobin in hemoglobin vesicles infused for isovolemic hemodilution necessary to improve oxygenation in critically ischemic hamster skin? Am J Physiol Heart Circ Physiol, December 1, 2005; 289(6): H2624 - H2631. [Abstract] [Full Text] [PDF]

				H. Sakai, P. Cabrales, A. G. Tsai, E. Tsuchida, and M. Intaglietta Oxygen release from low and normal P50 Hb vesicles in transiently occluded arterioles of the hamster window model Am J Physiol Heart Circ Physiol, June 1, 2005; 288(6): H2897 - H2903. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.104.073049v1 311/3/874    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sakai, H. Articles by Tsuchida, E. Search for Related Content PubMed PubMed Citation Articles by Sakai, H. Articles by Tsuchida, E. Pubmed/NCBI databases Compound via MeSH Substance via MeSH