Polyethylene Glycol Conjugated Bovine Hemoglobin Containing 15% MetHb Plays Approving Effect in Exchange Transfusion Rabbit Model

Abstract

PEG-bHb was developed by Kaizheng Biotech (Beijing, China), and pre-clinical research was completed. The objective of this study was to investigate the safe concentration of MetHb in PEG-bHb. The study was accomplished by examining the effects of PEG-bHb containing 5%, 8%, 15%, and 25% methemoglobin (MetHb), respectively, on cardiovascular system, blood chemistry, pathology of liver and kidney in rabbits following a 50% exchange transfusion. The results showed that PEG-bHb containing 5%, 8%, 15%, and 25% MetHb could keep four groups of experimental rabbits (5/5) alive until the 8th day after 50% exchange infusion as autologous whole blood did, and were superior to dextran 40 (2/5). MetHb concentration in PEG-bHb, no more than 25%, did not affect the PEG-bHb function on resuscitation of hemorrhaged rabbits by physiological measurements and blood chemistry assays. Histology study using optic and electron microscopy showed that there were slight pathological changes in hepatocytes and renal tubule epithelia in rabbits, which were infused by PEG-bHb containing 5%, 8%, and 15% MetHb. Partial organelles collapse was observed in rabbits resuscitated by PEG-bHb containing 25% MetHb. In conclusion, PEG-bHb is safe and effective when the MetHb concentration is at or below 15%.

Keywords:

• Exchange transfusion
• Methemoglobin
• PEG-hemoglobin
• Polyethylene glycol
• Resuscitation

View correction statement:

Erratum

INTRODUCTION

The hemoglobin based oxygen carriers (HBOCs) are an interesting class among the large family of oxygen therapeutics that includes perfluorocarbon emulsions and liposome-encapsulated Hb. The primary function of all these agents is to provide oxygen-carrying capacity. PEG-bHb, produced by Kaizheng Biotech Ltd. (Beijing, China), is a bovine hemoglobin conjugated by polyethylene glycol. The previous study testified that PEG-bHb is an effective blood substitute with powerful tissue oxygenation and blood volume expansion using exchange transfusion and hemorrhaged shock models in rats [[1]]; the oxygen-carrying capability of PEG-bHb has also been checked by quite a few other experiments on dogs and rabbits (data not published).

Heme iron of HBOCs in the reduced state is necessary for the reversible binding/release of molecular oxygen. Minimization of methemoglobin (Fe⁺³) formation is the major requirment for HBOCs function. MetHb is unable to bind oxygen and thus lowers the carrying capability of HBOC. In addition, MetHb has a potential to cause endothelial and surrounding tissue damage owing to rising free radicals [[2]]. If compounds such as sodium nitrite and anyl nitrite react with heme iron in vivo, blood MetHb will rise. Blood MetHb levels greater than 35% result in mild exertional dyspnea, headaches and dizziness, and it will become fatal when blood MetHb levels are 70% or more [[3]]. In normal status, enzyme mechanism within RBC protects against the effects of MetHb by cycling hemoglobin from its oxidized to reduced state and thus maintaining MetHb levels less than 3% [[4]]. In contrast, hemoglobin oxidizes at a rate of approximately 4% per hour in physiological salt solutions at 37°C [[5]]; so, discommodious for use and costly, it's necessary to store the HBOCs at − 80°C to avoid oxidation of the products during longer storage time. Short–term storage of hemoglobin solutions (< 14 days) can be done at 4°C [[6]]. PEG-bHb also has to face the storage problem.

As mentioned above, MetHb is a key factor for HBOC's quality, and it's practical to investigate how much methemoglobin is safe for effective usage of HBOCs. Linberg R. et al. showed that PEG-methemoglobin levels at or below 10% did not significantly alter the ability of solutions to deliver oxygen [[4]]. Hemolink contains less than 8% methemoglobin [[7]] and less than 15% methemoglobin exists in HBOC-201 [[8]]. The aim of this article is to study how much methemoglobin is safe and will not lower the oxygen delivery capability of PEG-bHb (Kaizheng, Beijing) based on our previous results using 50% exchange transfusion model [[1]].

MATERIALS AND METHODS

Animals

Chinese White Rabbits (male, 2.5–3.0 kg) provided by Experimental Animals' Center of the Military Medical Institute of China.

Test Solutions

Polyethylene glycol conjugated bovine hemoglobin (PEG-bHb) (5.8 g/dl solution, endotoxin < 1.0 EU/ml, osmolality 330 ± 30 mOsm, pH 7.4 ± 0.2) was formulated in buffer consisting of 150 mmol/L NaCl, 5 mmol/L NaHCO₃, 4 mmol/L Na₂HPO₄ and 1mmol/L NaH₂PO₄ at Beijing Kaizheng Biotech Co., Ltd. (Beijing, China). PEG-bHb, which contained methemoglobin < 5%, was incubated at 37°C to adjust methemoglobin to 8%, 15%, and 25%, respectively. Ten percent dextran 40 was used as control (Beijing Double-Crane Pharmaceutical Co., Ltd., China), and 0.9% sodium chloride solution containing 5% glucose was purchased from Otsuka Pharmaceutical Co. Ltd., China.

Animal Models

Rabbits were anesthetized using sodium pentobarbital (30 mg/kg, i.v.) and placed on a heating pad with their body temperature maintained at approximately 38°C. After 50% of their estimated blood volume (weight of animal (kg) × (60 ml/kg) × 50% = dosage (ml)) was withdrawn through carotid intubatiton, different resuscitation fluids were infused to rabbits of different groups immediately via ear marginal vein at a rate of 3 ml/min (Table 1). Rabbits in group A were resuscitated by isovolume heparinized autologous whole blood, while half hemorrhage volume of PEG-bHb (containing different levels of MetHb) or dextran was infused to groups B, C, D, E, and F according to our pharmcodynamic study [[1]], that is, 25% estimated blood volume of fluids were transfused to rabbits, and another 25% blood loss was complemented by 0.9% sodium chloride solution.

Table 1. Experimental groups

					Transfusion fluids PEG-bHb
Groups	Rabbits No.	Blood withdrawal (ml/kg)	Whole blood (ml/kg)	Dextran (ml/kg)	MetHb levels (%)	Transfusion volume (ml/kg)	Glucose N.S (ml/kg)
A	5	30	30	–	–	–	–
B	5	30	–	–	4.8	15	15
C	5	30	–	–	8.0	15	15
D	5	30	–	–	15.0	15	15
E	5	30	–	–	25.0	15	15
F	5	30	–	15	–	–	15

Physiological Measurements

Mean arterial blood pressure, electrocardiogram and respiratory frequency were continuously monitored through the acute experiment (from baseline to the 120th min after exchange transfusion) using a polygraph system (M6001, Nihon Kohden, Japan), and the data of the baseline, 0 min after controlled bleeding, 0 min after transfusion, 30 min after transfusion and 120 min after transfusion were recorded. The rabbits were allowed to recover from anesthesia, then extubated and returned to their cages to be observed every day before final evaluation (the 8th day after resuscitation).

Blood Analysis

Blood samples were taken before experiment, after bleeding, and 0 h, 2 h, 4 h, 8 h, 24 h, 48 h, and 8d after transfusion. Arterial blood gases (pH, PCO₂, HCO₃⁻), red blood cell count, reticular red blood cell count, plasma Hb and MetHb were measured. Plasma samples were stored at − 20°C for blood chemistry testing including the measurement of alkaline aminotransferase (ALT), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), creatine kinase (CK), blood urea nitrogen (BUN) and creatinine (Cr) using automatic biochemistry analyzer (Hitachi 7170, Japan).

Pathological Examinations

The survival rabbits of six groups were killed at 8 days after exchange transfusion and organs (heart, liver, spleen, lung, kidney, intestine, stomach, brain, and skeletal muscle) were harvested for histological examination by optic microscope. Liver and kidney tissue were cut to 2 mm³ and fixed in 3% glutaraldehyde for electron microscopy.

Statistics

Data are presented as means ± SD (standard deviation of mean). The differences among treatment groups were assessed by one-way ANOVA. Multiple comparisons, when significant differences existed, were determined by least significant differences techniques. Statistical significance was defined as P < 0.05 to reject a null hypothesis.

The mean circulating half-life of PEG-bHb in plasma was determined by regression curve analysis.

RESULTS

Survival and Physiological Measurements

All rabbits in six groups were resuscitated 0.5 h to 1 h after extubation. The rabbits in group A, B, C, D, and E, which were transfused by autologous whole blood (group A) or PEG-bHb (group B to E), respectively, were in good status after resuscitation and survived until the 8th day before being killed. The rabbits of group F, resuscitated by dextran 40 (as control), were weak and two rabbits died within 24 h, and another one died in the third day after acute experiment; therefore the survival rate in group F was 2/5.

The mean arterial pressure (MAP) decreased to low level when 50% estimated blood volume of rabbits was removed. Rabbits in group A infused with autologous whole blood had their MAP and heart rate (HR) raised to normal. Rabbits in groups B, C, D, and E displayed significant increase of MAP levels close to baseline when treated with PEG-bHb containing 5%, 8%, 15%, and 25% PEG-MetHb, respectively. In contrast, MAP in group F was restored to 97 ± 11 mmHg at 0 min after dextran transfusion, and decreased gradually to 59 ± 5 mmHg at 0.5 h, holding the line until the end of acute experiment (Figure 1). HR changed at the same trend with the MAP (Figure 2). ST-segment elevation/depression was seen on ECG after bleeding; the situation was mended with fluids infusion, especially in group A rabbits. During the acute experiment, all rabbits lived with smooth respiratory.

Figure 1 Mean arterial pressure changes of rabbits in groups A, B, C, D, E, and F during experiments. MAP decreased after blood withdrawl and increased to normal level when rabbits transfused by autologous whole blood (group A), PEG-bHb containing different concentration of MetHb (groups B, C, D, and E) and dextran (group F), and MAP of rabbits maintained at normal levels until the acute experiment finish except group F.

Figure 2 Heart rate changes with hemorrhage transfusion among six group rabbits. Heart rate decreased after rabbits were withdrew 50% volume blood, and recruited when whole blood (group A), PEG-bHb (5% MetHb in group B, 8% MetHb in group C, 15% MetHb in group D, and 25% MetHb in group E) and dextran (group F) were transfused.

Pharmacokinetics

Rabbits that received PEG-bHb containing 5%, 8%, 15%, and 25% PEG-MetHb showed the maximum serum concentrations of PEG-bHb of 1.7 ± 0.1 g%, 1.6 ± 0.2 g%, 1.5 ± 0.1 g% and 1.5 ± 0.1 g%, respectively, with the mean circulatory half-life of 27.5 ± 3.8 h (Figure 3). Concentrations of MetHb in plasma were significantly different among rabbits of group B/C, group D and group E (P < 0.01) immediately after infusion of PEG-bHb containing 5%, 8%, 15%, and 25% PEG-MetHb; however, no significant differences were found among the four groups 4 hrs after infusion (Figure 4).

Figure 3 Shown is plasma concentration-time curve of PEG-bHb obtained after rabbits were transfused with PEG-bHb containing 5% MetHb (group B), 8% MetHb (group C), 15% MetHb (group D), and 25% MetHb (group E). Half time of PEG-bHb in four groups was similar.

Figure 4 Plasma MetHb concentration changes of rabbits in groups B, C, D, and E transfused by PEG-bHb containing MetHb of 5%, 8%, 15%, and 25%. MetHb concentration decreased quickly during 4 hours after PEG-bHb infusion.

Hematology

Whole blood Hb concentrations of rabbits in the six groups decreased significantly after exchange transfusion. Hb concentrations of the survival rabbits in group B, C, D, E and F increased after acute experiments, and recovered to 80% of baseline values at the 8th day. Autologous whole blood infusion recovered the Hb concentration of group A rabbits to normal level immediately (Figure 5). Reticulocyte count showed significant difference between rabbits infused by PEG-bHb and autologous whole blood, that is, red blood cells proliferation was more active in groups B, C, D, and E than in group A (Table 2).

Figure 5 Whole blood Hb concentrations of group B, C, D, E and F rabbits decreased significantly after exchanged transfusion and recovered to about 80% of baseline values at the 8th day.

Table 2. Reticular red blood cell counts of six group rabbits during experiments

	Ret (%)
Time	A	B	C	D	E	F
0	2.4 ± 1.7	2.1 ± 0.4	1.5 ± 0.4	1.3 ± 0.6	1.1 ± 0.4	1.1 ± 0.3
24 h	2.8 ± 2.3	3.2 ± 0.5	4.9 ± 1.7	3.2 ± 0.8	3.7 ± 1.9	3.3 ± 2.1
8 d	2.6 ± 1.6	5.9 ± 1.8	5.9 ± 1.0	6.6 ± 0.7	10.9 ± 1.5	8.6**

**One of two live rabbits was measured.

Arterial Blood Gas Analysis

Arterial pH values decreased progressively during fluids infusion in groups A, B, C, D, and E, to the lowest values immediately after infusion, and then increased and leveled off at 48 h after transfusion. In group F rabbits, arterial pH did not increase until 2 h after dextran infusion (Figure 6a). Arterial PCO₂ decreased during hemorrhage and then changed at the contrary direction as arterial pH (Figure 6b). Arterial HCO₃⁻ fluctuated during experiment (Figure 6c). There were significant differences of PCO₂ and HCO₃⁻ at 8 hr after transfusion between F group and the other five groups (P < 0.05).

Figure 6 a) Arterial pH decreased after exchange transfusion and was upregulated to about normal level since 2 hours after fluids of whole blood and PEG-bHb containing different MetHb. Transfusion of dextran did not correct low pH of experiment rabbits quickly. b) Arterial PCO₂ decreased during hemorrhage and then changed at the contrary direction as arterial pH. c) Shown is aterial [HCO3–] changes of six group rabbits. It implied dextran transfusion could not improve rabbits ischemia status that low HCO3-level in group F.

Blood Chemistry Assay

All rabbits' blood chemistry profiles of CK, LDH, AST, ALT, BUN, and Cr were monitored at baseline, 0 h, 8 h, 24 h, 48 h and 8 d after exchange transfusion. CK, LDH, and AST values, which indicate cordis damage, were elevated in all rabbits, but the peak value appeared at different time among rabbits that infused with autologous whole blood (group A), dextran (group F) and PEG-bHb (groups B, C, D, and E) at 8 h, 48 h, and 24 h, respectively. ALT level, which was used to assess hepatic toxicity results, raised transiently in all six groups of animals. Renal function, assessed by BUN and Cr, remained normal in group A. BUN value was raised markedly in group F and remained in normal range in groups B, C, D, and E. All these biochemistry markers (CK, LDH, AST, ALT, BUN, and Cr) recovered at the 8th day after exchange transfusion (Figure 7).

Figure 7 Blood biochemical markers of CK, ALT, and BUN were used to evaluate the heart, liver, and kidney damage. a): circulation CK increased in rabbits after exchange transfusion, peaked at different time owning to different resuscitation fluids infusion and restored to normal level until 8 d. b): ALT increased in six group rabbits after exchange transfusion and returned to normal at 8 d. c): There is no significant increase in circulating blood urea concentration in groups that exchange transfused by autologous whole blood and PEG-bHb (containing MetHb of 5%, 8%, 15%, and 25%) and recovered at 8 d.

Histology

The survival rabbits of the six groups were killed at the 8th day after resuscitation. Histopathological slides of tissues of heart, liver, spleen, lung, kidney, intestine, stomach, brain, and skeletal muscle were examined by optic microscopy. No histology changes were observed in heart, intestine, skeletal muscle and brain of all groups. There was slight turgidity in the spleen. Lung showed approximately normal configuration except for slight mononuclear cells infiltration in one or two rabbits. Hepatocytes showed cloudy swelling in the cytoplasm of groups A and B, C and F animals. Slight hydropic change and a few of eosinophil granules were present in group D, and moderate hydropic change and more eosinophil granules were observed in group E rabbits (Figure 8a). The glomerulus maintained normal configuration in all group rabbits. Slight vacuolar degeneration occurred in renal tubule epithelia cytoplasma in groups B, C and D, and moderate vacuolar degeneration was in tubule epithelia cytoplasm of group E rabbits (Figure 8b).

Figure 8 a) Moderate hydropic change and eosinophil granules were observed in group E rabbit liver (× 100). b) moderate vacuolar degeneration was in tubule epthelia cytoplasma of group E rabbits transfused by PEG-bHb containing 25% MetHb (× 100).

Based on the pathological changes in hepatocytes and renal cells of the rabbits in groups B, C, D, and E, electron microscopy was used to clarify ultrastructural changes. Endoplasmic reticulum dilatation, fat denaturalization and high electron density granules were observed in hepatocytes, and there were endoplasmic reticulum dilatation and increase of lysosome in renal tubule epithelial cells of group B, C and D animals (Figures 9a and 9b). In addition, partial cell organelle collapse occurred in hepatocytes and renal tubule epithelial cells of group E rabbits infused with PEG-bHb containing 25% MetHb (Figures 9c and 9d).

Figure 9 a) Ultrastructure was basically normal except that endoplasmic reticulum dilatation, fat denaturalization and high electron density granules were observed in hepatocytes of rabbits exchange transfused by PEG-bHb containing MetHb of 5%, 8% and 15% (group B, C, and D) (× 4000). b) There're endoplasmic reticulum dilatation and increase of lysosome in renal tululae epithelial cells of group B, C, and D rabbits, other organelles were normal (× 4000). c) Except changes mentioned in group B, C, and D, partial cell organelle collapse occurred in hepatocytes of group E (× 4000). d) Partial cell organelle collapse were also observed in renal tubule epithelial cells of group E rabbits (× 4000).

DISCUSSION

A 50% exchange transfusion rabbit model was used to evaluate the effect of PEG-bHb (Kaizheng Biotech Ltd., Beijing, China) containing 5%, 8%, 15%, and 25% PEG-MetHb. Experimental data showed that PEG-bHb infusion (containing different levels of MetHb) restored MAP of hemorrhage rabbits just as autologous whole blood transfusion did, and was superior to dextran infusion. During our experiment, three of five (3/5) rabbits in group F (dextran infused) died within 72 h after resuscitation, whereas the rabbits of groups B, C, D and E survived until the 8th day as group A resuscitated by autologous whole blood. The results confirmed the validity of PEG-bHb again [[1]] and suggested that the PEG-bHb containing higher concentration of MetHb (for example, 15%, so much as 25%) still have favorable resuscitation effect.

Hematology and blood biochemistry tests demonstrated that rabbits in groups B, C, D, and E restored evidently at the 8th day after transfusion. The whole Hb concentration was raised to the level closely to baseline. The reticulocyte count indicated that red blood cells proliferated actively after hemorrhage. There was no other evidence implying that MetHb in PEG-bHb, not excessive 25% at least, could hinder experimental rabbits' hematosis. CK is a sensitive marker for myocardium ischemia [[9]]. During experiments, CK increased remarkably in rabbits of groups B, C, D, and E, and the enzyme activity peak appeared at 24 h after transfusion and decreased to the normal range until 8 days after infusion. The change of CK showed that PEG-bHb infusion could correct transient ischemia in cardiac myocytes of rabbits in time. A similar process was observed in rabbits that were exchange transfused with autologous whole blood. The results suggested that hemorrhage might be the important cause of transient ischemia of myocardium and PEG-bHb (MetHb less than 25%) infusion could ameliorate blood supply and deliver oxygen to heart as autologous whole blood. In contrast, dextran didn't improved cardiac ischemia in good time so that the concentration of CK peaked at higher level at 48 h after transfusion. Especially, the absolute majority of rabbits transfused with PEG-bHb showed normal renal function, as indicated by BUN, although BUN increased significantly in one rabbit in group B and another rabbit in group C. However, the phenomena might not be correlated with MetHb concentration of PEG-bHb, as none of rabbits in group D and E, which were infused with higher MetHb concentrations of PEG-bHb than in group B and C, was measured at the higher BUN level. Minim free bHb in PEG-bHb might be responsible for the renal toxicity [[10]].

MetHb concentrations in plasma were different in groups B, C, D and E at 0 min after transfusion (Figure 4). They were 0.09 g%, 0.13 g%, 0.21 g%, and 0.3 g%, respectively, corresponding to PEG-bHb containing 5%, 8%, 15%, and 25% MetHb transfused. MetHb decreased markedly, however, during 4 h after PEG-bHb infusion, and the similar plasma level of MetHb was about 0.1 g% at 18 h after transfusion in the four groups' animals. The results implied that a certain extent MetHb could be reduced by intrinsic reducing systems of rabbits [[11]]. Normal plasma contains reducing components, such as ascorbic acid and glutathione, which can afford protection to acellular HBOCs through electro-transfer mediated process [[12], [13]].

Transient ischemia due to exchange transfusion can bring some reversible cell injury to rabbits [[14], [15]], such as CK increase in plasma resulted from myocardium ischemia, and these changes may be recovered quickly when autologous whole blood or PEG-bHb is transfused, so no distinct pathological changes could be observed by optic microscopy in heart, lung, spleen, and some other tissues at the 8th day after acute experiment.

Cloudy swelling and hydropic change in hepatocytes and renal tubule epithelia can be induced by ischemia or hypoxaemia owing to 50% exchanged transfusion; these changes were reversible if the damage factors can be eliminated in time [[14]]. In our study vacuoles were observed in cytoplasm of hepatocytes and renal tubule epithelia by optic microscopy, which were corresponding to lysosome increase and endoplasmic reticulum dilation confirmed by electron microscopy. Liver and kidney are important metabolic organs of animals and they are primary paths of medicament metabolism; in general, lysosomes increase with large molecules loading increase [[14]]. So, it's reasonable that transfusion of PEG-bHb (about 100 kD) led to upregulate lysosome numbers in cytoplasm of liver and kidney as observed. The vacuolar changes were reversible, as proved by Conover et al. [[16]], and confirmed by another research work on dogs in our lab.

The risk of pathological changes in rabbits' livers and kidneys increases when the MetHb concentration of the infused PEG-bHb rises. In our study, the most remarkable pathological changes in liver and kidney were observed in group E rabbits among the four groups infused with PEG-bHb. There were no significant differences nearly in organelles pathological changes of liver and kidney among groups B, C, and D, as confirmed by electron microscopy, while more injuries, such as partial organelles collapse, were visualized in group E. The lower capacity of oxygen delivery of PEG-bHb containing 25% MetHb may not remedy ischemia/hypoxaemia of rabbits after hemorrhage in time, resulting in more pathological changes. In addition, toxicity of MetHb itself would be involved in the pathological process [[2]].

CONCLUSIONS

Based on our previous pharmacodynamics study on PEG-bHb, we evaluated the effects of PEG-bHb containing 5%, 8%, 15%, and 25% MetHb in 50% exchange transfusion rabbit model with many indices of physiology, biochemistry and histology. The results show that PEG-bHb is effective and safe when the concentration of PEG-MetHb in the product is less than 15%.

This project is supported by Chinese National High-Tech Program (No: 2002AA205101).

The authors wish to thank Mr. Jianping Ren and Mr. Zirong Qi for their help on the exchange transfusion model, Mr. Deyong Zou for biochemistry tests, Ms. Yabing Gao and Mr. Dingrong Zhong for histological examination, and Ms. Renlan Wang and Mr. Wanzong Zou for electron microscopy.