544
Views
2
CrossRef citations to date
0
Altmetric
Original Article

PEGylation of αα-Hb using succinimidyl propionic acid PEG 5K: Conjugation chemistry and PEG shell structure dictate respectively the oxygen affinity and resuscitation fluid like properties of PEG αα-Hbs

, , &
Pages 270-281 | Received 14 Oct 2013, Accepted 16 Jan 2014, Published online: 06 Mar 2014

Abstract

PEGylation of intramolecularly crosslinked Hb has been studied here to overcome the limitation of dissociation of Hb tetramers. New hexa and deca PEGylated low oxygen affinity PEG-ααHbs have been generated. Influence of PEG conjugation chemistry and the PEG shell structure on the functional properties as well as PEGylation induced plasma expander like properties of the protein has been delineated. The results have established that in the design of PEG-Hbs as oxygen therapeutics, the influence of conjugation chemistry and the PEG shell structure on the oxygen affinity of Hb needs to be optimized independently besides optimizing the PEG shell structure for inducing resuscitation fluid like properties.

Abbreviations
αα-Hb=

hemoglobin intramolecularly crosslinked at Lys-99(α)

COP=

colloidal osmotic pressure

Hb=

hemoglobin

HBOC=

hemoglobin based oxygen carrier

HexaPEGylation=

conjugation with six copies of PEG chains

P50=

oxygen pressure at half saturation

4-PDS=

4, 4′-dithiopyridine

PEG=

poly(ethylene glycol)

PEG5K2 Hb=

hemoglobin conjugated with two copies of PEG5K

PEG10K2 Hb=

hemoglobin conjugated with two copies of PEG10K

PEG5K6 Hb=

hemoglobin conjugated with six copies of PEG5K

PEG5K4 canine Hb=

canine hemoglobin conjugated with four copies of PEG5K

(Propionyl)6-αα-Hb=

αα-Hb conjugated with 6 propionyl groups

(Propionyl)6- Hb=

Hb conjugated with six propionyl groups

(Propionyl-PEG5K)6-αα-Hb=

propionylated PEG-αα-Hb conjugate with six copies of PEG5K chains

(Propionyl-PEG5K)6-Hb=

propionylated PEG-Hb conjugate with six copies of PEG5K chains

SC PEG5K=

succinimidyl carbamate PEG5K

SEC=

Size-exclusion chromatography

SPA PEG5K=

succinimidyl ester of propionic acid PEG5K

(Urethane-PEG5K)6-αα-Hb=

urethane PEG-αα-Hb conjugate with six copies of PEG5K chains

(Urethane-PEG5K)6-Hb=

urethane PEG-Hb conjugate with six copies of PEG5K chains

Introduction

One of the major obstacles for the development of hemoglobin based oxygen carrier (HBOC) has been the in vivo hypertensive activity of acellular Hb. In recent years, PEGylation has emerged as a novel chemical platform that attenuates the in vivo hypertensive activity of acellular Hb (CitationCole and Vandegriff 2011, CitationAcharya et al. 2005). This new concept has energized the field of blood substitutes and represents the beginning of new era in the design and development of nonhypertensive Hb based oxygen carriers (HBOCs). PEGylation induced plasma expander like properties make PEGylated Hbs a novel class of oxygen carrying resuscitation fluids. This is a unique and distinguishing property of this class of modified Hb molecules as HBOCs as compared to most of the previously designed Hb derivatives designed as potential blood substitutes. Sanguinate of Prolong Pharmaceutical, a PEGylated bovine Hb that carries eight copies of PEG 5K conjuaged through urethane linkage (CitationNho et al. 1992, CitationAnanthakrishnan et al. 2013) and MP4 of Sangart, a PEGylated Hb that carries six copies of PEG 5K conjugated through extension arm chemistry based thiosuccinimidyl linkage (CitationVandegriff et al. 2003, 2004, Citation2009, CitationYoung et al. 2005), represent the two major PEG-Hbs that are currently in clinical trial.

However, PEGylation of Hb still presents some challenges before we can harvest the clinical potential of Hb as oxygen therapeutic. There are two major limitations with the PEGylation of Hbs. PEGylation of uncrosslinked Hb weakens the interdimeric interactions of the tetramers and hence facilitates the dissociation of PEGylated tetramers into PEGylated dimers. The conjugation chemistry used for PEGylation of Hb also appears to dictate the extent to which a given level of PEGylation induces the weakening of the interdimeric interactions. The weakening of interdimeric interaction can increase the oxygen affinity of Hb and also rate of autoxidation as well as the subsequent heme oxidation product mediated toxicity. The reductive alkylation chemistry based PEGylation using PEG aldehydes has the maximum effect on the PEGylation induced dissociation of Hb (CitationHu et al. 2005). On the other hand acylation chemistry based PEGylation using the succinimidyl active esters of PEG propionic acid has the least impact on the tetramer stability of the hexaPEGylated uncrosslinked Hbs generated using PEG 5K (CitationLi et al. 2008). The extension arm chemistry synergized with maleimide chemistry for surface decoration of Hb with PEG chains has the unique advantage of attenuating the PEGylation induced weakening of the interdimeric interactions (CitationLi et al. 2006, Citation2009, CitationManjula et al. 2005). This advantage of EAF PEGylation is not an all or none phenomenon; this protective role is reduced beyond the stage of hexaPEGylation (using PEG 5K), even when the EAF hexaPEGylation is carried out with reversible protection of Cys-93(β) to prevent the PEGylation of the two intrinsic reactive thiols of Hb (CitationLi et al. 2009).

The second limitation that arises on PEGylating Hb is due to its influence on oxygen affinity since PEGylated Hbs are invariably generated high oxygen affinity species. The increase in the oxygen affinity resulting from PEGylation can be dissected out as emerging from a site specific modification of the PEG chains and that induced by the presence of PEG shell surrounding the central Hb core. The PEG shell around the Hb molecule is expected to increase viscosity on the molecular surface of Hb (microviscosity). The viscous PEG should slow down both the entry and escape of the oxygen to and from the heme center. The influence of the packing density of the PEG shell of PEG-Hbs on the tissue oxygenation, i.e., delivery of oxygen by PEG-Hbs, when the overall oxygen affinity of PEG-Hbs are comparable, is an area that has not been addressed so far. The studies available on the tissue oxygenation by PEG5K2 Hb, PEG10K2 Hb, PEG5K4 canine Hb, and PEG5K6 Hb, all of which have comparable oxygen affinities clearly reflect the role of packing density of PEG shell (microviscosity) on the oxygen delivery by PEGylated Hbs (CitationAcharya et al. 2007a, CitationManjula et al. 2003).

Winslow and his colleagues advanced the concept of a high O2 affinity for Hb based oxygen therapeutics is an essential property to attenuate the in vivo hypertensive activity of acellular Hb. They argued that designing blood substitutes with an oxygen affinity comparable to that of RBCs is flawed since this molecule will release all its oxygen on the arterial side of the circulation leading to autoregulation mediated vasoconstriction (CitationCole et al. 2008, CitationWinslow 2005, CitationWinslow 2008, CitationTsai et al. 2003). However, the oxygen affinity of the two PEGylated Hbs in clinical trials appears to be too high to deliver adequate levels of O2 to tissues under physiological conditions. The oxygen affinity of Hemospan of Sangart is even higher than that of Sanguinate of Prolong. Besides PEG-Hbs are used at significantly lower concentrations than the previously designed HBOC candidates. These aspects of PEG-Hb that makes them well suited for targeting tissue oxygenation of hypoxic areas in the body is establishing a new basis for the design oxygen therapeutics aimed at minimizing the risk of damaging organs that are potentially hypoxic (CitationVandegriff and Winslow 2009, CitationYoung et al. 2005, CitationVandegriff et al. 2003). Since solutions of PEG-Hbs are very viscous, the clinical application of the PEG-Hb as blood substitute requires the optimization of the PEGylation platform using the just amount of conjugated PEG mass needed to prevent in vivo hypertensive activity and the right increase in viscosity to maximize the amount of PEG-Hb that may be infused.

It should be noted that one of the consequences of designing a blood substitute with a very low oxygen affinity is that the patients have to be transfused with fresh blood (CitationTsai et al. 2007, Citation2010) before the oxygen saturation in the tissues of patients falls in the range of the P50 of the current versions of PEG-Hbs. The oxygen carrying capacity of Hb in RBC represents essentially most of the oxygen carrying capacity of the system even when as much as 40–50% of the blood volume is replaced by a 4 gm% solution of MP4 (Hemospan) or Sanguinate. Assuming the total initial circulating concentration of Hb (in RBC) is around 12 gm%, after a 50% blood volume exchange with the present versions of PEG-Hbs, the Hb in RBC will account for only 6 gm%, a value below the transfusion trigger. At this level PEG-Hb will represent about 2 gm%, around 25% of the total Hb in circulation. Thus, as long as the oxygen affinity of PEG-Hb is high and the PEG-Hb presents in the plasma in low concentrations relative to Hb in RBC, Hb in RBC is still the primary oxygen delivery material. The extent to which oxygen is be delivered by high oxygen affinity PEG-Hb in circulation is dictated by the oxygen tension in hypoxic regions in the tissues and is around that of the P50 of PEG-Hb. Therefore, the current versions of high oxygen affinity PEGylated Hbs are primarily active plasma expanders (CitationSriram et al. 2012, CitationCabrales et al. 2008), and these semisynthetic hybrid biopolymers perform a secondary function of targeted oxygen delivery to hypoxic regions, if there are any.

Accordingly, a nonhypertensive PEG-Hb with an oxygen affinity that is intermediate to that of current versions of PEG-Hbs and RBC has been considered as better choice for designing blood substitutes of general utility. This is expected to satisfy to a significant degree, the requirements of the designed primary function for blood substitutes, transporting and delivering oxygen from lungs to tissues (CitationAcharya et al. 2007b).

One of the early chemically modified hemoglobin designed to serve as blood substitute is diaspirin crosslinked hemoglobin (CitationWinslow 2000); this is an intramolecularly cross-linked Hb, the fumaryl crosslink being between the α chains (lysine 99α1-lysine 99α2) in the middle of the central cavity. Stabilization of the tetrameric hemoglobin against dissociation into dimers by engineering intramolecular crosslinking was successful in overcoming the nephrotoxicity. The oxygen affinity of this intramolecularly crosslinked Hb is close to that of RBC and also to that of HbA in the presence of diphosphate glycerate (DPG). The vasoactivity of this molecule was even higher than the parent molecule. PEGylation could attenuate the in vivo hypertensive activity of this molecule as well by inducing resuscitation fluid like properties. The oxygen affinity of this intramolecularly crosslinked Hb molecule will be lower than that of PEGylated uncrosslinked Hb and this could result in a desirable oxygen carrying resuscitation fluid. Accordingly, PEGylation of αα-Hb has been undertaken to engineer “active plasma expander” properties to αα-Hb that still maintains an oxygen affinity lower than the current versions of PEG-Hbs in clinical trials.

Direct PEGylation of αα-Hb, i.e., without the intervention of extension arm chemistry, has been undertaken here as the PEGylation induced dissociation is no longer a concern with the intramolecularly crosslinked molecule. Accordingly, succinimidyl ester of propionic acid PEG (SPA PEG) has been chosen to conjugate the PEG chains to modify the protein through the formation of stable isopeptide linkages (). In this acylation chemistry based PEG conjugation, the positive charge on the surface amino groups is neutralized on conjugation. HexaPEGylation and decaPEGylation of the αα-Hb with PEG 5K using this acylation based PEGylation platform will generate a more negatively charged PEGylated protein. Reaction of αα-Hb with succinimidyl propionate has also been carried out in an attempt to generate hexa and decapropionyl αα-Hb as conjugation chemistry control for the PEGylation reaction. Accordingly, an attempt is made here to delineate the influence of conjugation chemistry and of intramolecular crosslinking in Hb on the PEGylation induced enhancement in plasma expander like properties as well as the PEGylation induced increase in the oxygen affinity of the molecule. Deca and hexapropionyl αα-Hb has also been generated to segregate the contribution of the PEG shell from that of covalent modification of αα-Hb by propionylation (). Further, the influence of the succinimidyl chemistry mediated formation of urethane linkage between Hb and PEG () using succinimidyl carbamate PEG5K (SC PEG5K) has also been investigated here by generating the hexaPEGylated αα-Hb to map the influence of conjugation chemistry on both the functional and solution properties.

Figure 1. Schematic representation of the conjugation chemistry used for PEGylation of αα-fumaryl Hb. Succinimidyl esters are used to generate either isopeptide and of urethane linkage between Hb and PEG: (A) succinimidyl ester of propionic acid PEG (SPA PEG), (B) N-succinimidyl propionate, and (C) succinimidyl carbamate PEG (SC PEG).
Figure 1. Schematic representation of the conjugation chemistry used for PEGylation of αα-fumaryl Hb. Succinimidyl esters are used to generate either isopeptide and of urethane linkage between Hb and PEG: (A) succinimidyl ester of propionic acid PEG (SPA PEG), (B) N-succinimidyl propionate, and (C) succinimidyl carbamate PEG (SC PEG).

Materials and methods

Hemoglobin solutions

HbA was purified from the human erythrocyte lysate by DE-52 chromatography (CitationManjula and Acharya 2003). HbA was intramolecularly cross-linked with bis-(3,5-dibromosalicyl) fumarate (DBBF) as described previously (CitationChatterjee et al. 1986, CitationNacharaju et al. 2007). Briefly, HbA (1 mM) was incubated overnight with 8 mM sodium tripolyphosphate in 50 mM Bistris acetate buffer (pH 6.5) at 4°C to block modifications of the β chains by DBBF. Then the HbA solution was deoxygenated and incubated with 1.9 mM DBBF in 50 mM Bistris acetate buffer (pH 6.5) at 37°C for 3 h. The reaction was stopped by adding 20 mM Gly-Gly. The sample was dialyzed against 50 mM Tris acetate buffer, pH 8.0, at 4°C overnight and then applied to a Q-Sepherose high performance column (2.6 × 65 cm, 350 ml of column volume, Pharmacia). The column was equilibrated with 50 mM Tris acetate at pH 8.0 (buffer A) using an AKTA Purifier 10 System (Amersham Pharmacia Biotech.), and eluted with a linear gradient of 0–100% buffer B, 50 mM Tris acetate at pH 6.8, at 4°C (CitationManjula and Acharya 2003). The column was eluted at a flow rate of 2 ml/min. The effluent was monitored at 540 and 630 nm. The DBBF cross-linked αα-HbA was collected on a Frac-900 collector (Amersham Pharmacia Biotech.).

PEGylation of Hb

0.5 mM HbA or αα-Hb in PBS was incubated with 5 mM or 10 mM SPA PEG5K at 4°C for 3 h. The unreacted PEG reagent was removed by centrifugal filtration against PBS, using Centricon (50 KD cut off membrane, Millipore Corp.) at 5000 rpm for 30 min four times.

Oxygen affinity measurements

The oxygen equilibrium curves (OECs) of Hb solutions were obtained using a Hemox analyzer (TCS Scientific, New Hope, PA). This instrument measures the oxygen tension with a Clark oxygen electrode (Model 5331 oxygen Probe; Yellow Springs Instrument, Yellow Springs, OH) and simultaneously calculates the Hb oxygen saturation via dual-wavelength spectrophotometer. The concentration of Hb samples was 20 μM (heme). The measurement was carried out in PBS, pH 7.4 at 37°C. The dependence of the oxygen affinities of Hb samples on pH was carried out in 100 mM phosphate buffer at a range of pH 6.5–8.5 at 37°C (CitationFronticelli et al. 1984).

Analytical methods

Size-exclusion chromatography (SEC) was performed, as described previously (CitationManjula and Acharya 2003), at room temperature on two HR10/30 Superose 12 columns (Amersham Biosciences) connected in series using a fast performance liquid chromatography instrument (Pharmacia). The column was eluted using PBS at pH 7.4 and a flow rate of 0.5 ml/min, and the effluent was monitored at absorbance 540 nm. The reactivity and number of thiol groups on Hb samples were estimated by reaction with 4,4′-dithiopyridine (4-PDS) (Aldrich Chemical Co.) according to CitationAmpulski et al. (1969).

Identification of sites of PEGylation

PEGylation sites were established by tryptic peptide mapping as described previously (CitationHu et al. 2007). Briefly, the globin chains of Hb prepared by acid acetone precipitation were dissolved in 100 mM ammonium bicarbonate to a final concentration of 1 mg/ml, and digested for 3 h at 37°C with TPCK-trypsin (Sigma Chemical Co., St. Louis, MO) at an enzyme to substrate ratio of 1:100 (w/w). The resulting tryptic peptides were analyzed by RPHPL on a Vydac C18 column (10 × 25 mm) using a linear gradient of 5–50% acetonitrile containing 0.1% TFA in 160 min followed by a linear gradient of 50–70% acetonitrile containing 0.1% TFA in 20 min at a flow rate of 2 ml/min.

Solution properties of PEGylated Hb

The molecular radius of the PEGylated HSA was determined at 25°C at 1 mg/ml by dynamic light scattering measurements on a DynaPro NanoStar Instrument (Wyatt Technology, Goleta, CA). Each sample was measured four times and the average of the measurements is presented. The viscosity of the PEGylated Hb in PBS was measured with a cone and plate Rheometer (Brookfield, Middleboro, MA) at 37°C. The instrument was calibrated with water prior to measurements of the viscosity of the Hb samples. The colloid osmotic pressure (COP) of the PEGylated Hb in PBS was determined using a Wescor 4420 Colloidal Osmometer (Wescor, Logan, UT, USA) with a 30 KD Mw cut off membrane at room temperature. The instrument was calibrated with Osmocoll reference standard prior to measurements of the samples. Each sample was measured at least twice and the average of the measurements is presented.

Results

Influence of non-conservative PEGylation of αα-fumaryl-Hb on molecular and solution properties

The amount of PEG conjugated to αα-Hb in the two PEGylated samples was estimated by 1H-NMR method as discussed previously (CitationManjula et al. 2005) and the results are shown in . The adduct generated using a 10 fold molar excess of PEG reagent (over the protein) carries an average six copies of PEG-5K chains per tetramer while the adduct generated using a 20 fold molar excess of PEG reagent carries on an average ten copies of PEG-5K chains per tetramer. Accordingly the former conjugation product is referred to as hexaPEGylated derivative, i.e., (Propionyl-PEG5K)6-αα-Hb and the latter product is referred as decaPEGylated material, i.e., (Propionyl-PEG5K)10-αα-Hb.

Table I. Estimation of PEGylation αα-Hb.

FPLC patterns of hexa and decaPEGylated αα-Hb in Superose 12 columns has been compared with those of unmodified Hb and the crosslinked Hb (). Intramolecular crosslinking of Hb has essentially no influence on its hydrodynamic volume as reflected by FPLC pattern. HexaPEGylation results in a significant enhancement of the hydrodynamic molecular volume of αα-Hb (curve 3). The decaPEGylation increases the hydrodynamic volume further as reflected by the elution pattern (curve 4). This elution pattern is more asymmetrical by comparison with the hexaPEGylated material, suggesting a degree of heterogeniety in terms of the hydrodynamic volume.

Figure 2. SEC analysis of modified Hbs. 1, HbA control; 2, αα-Hb; 3, (Propionyl-PEG5K)6-αα-Hb; 4, (Propionyl-PEG5K)10-αα-Hb.
Figure 2. SEC analysis of modified Hbs. 1, HbA control; 2, αα-Hb; 3, (Propionyl-PEG5K)6-αα-Hb; 4, (Propionyl-PEG5K)10-αα-Hb.

PEGylation induced plasma expander like properties of Hb has been correlated primarily with the attenuation of the in vivo hypertensive activity of acellular Hb; these properties include the high hydrodynamic volume, viscosity and COP. The molecular radius of (Propionyl-PEG5K)6-αα-Hb and (Propionyl-PEG5K)10-αα-Hb as determined by dynamic light scattering measurements, is presented in . The molecular radius of control αα-Hb is 3.1 nm, same as the control uncrosslinked HbA. The intramolecular crosslinking has little influence on the molecular dimension of Hb. The radius of (Propionyl-PEG5K)6-αα-Hb is 6.0 nm, the conjugation of just 30 kD PEG mass (only half of the mass of Hb) nearly doubles the molecular radius of the molecule. The conjugation of another four copies of PEG-5K chains (another 20 kD PEG mass), however, increases the radius of the PEGylated protein only by another 0.7 nm.

Table II. Molecular and solution properties of PEGylated αα-Hb.

Thus, it is clear that the increase in the hydrodynamic volume of αα-Hb by the PEG shell engineered around the molecule is not a direct correlate of the PEG mass conjugated to Hb. As new chains are conjugated to PEG-Hb, they start filling the molecular space occupied by PEG chains previously conjugated, i.e., as the total mass of PEG chains conjugated to Hb is increased by increasing the number of PEG chains (using the same molecular size PEG chains). The PEG chains in the PEG shell pack tightly against one another. As shown in , the packing density of PEG shell of (Propionyl-PEG5K)6-αα-Hb is 39.8 Dal/nm3 and that of (Propionyl-PEG5K)10-αα-Hb is 46.7 Dal/nm3. For comparison, the packing density of the central protein core is around 513 Dal/nm3. Thus the PEG shell engineered around Hb can be considered as a loosely packed molecular envelop around a central hard sphere. The packing density of this molecular envelope increases as the number of PEG chains in the PEG shell increases. Accordingly, PEGylation of Hb results in generation of a novel semi-synthetic hybrid biopolymer with an inner protein domain that is rigid and an outer domain PEG shell that represents the flexible domain of the molecule.

The intramolecular crosslinking of Hb has essentially no influence on the viscosity and COP of a 4 gm% solution. The PEGylation increases the viscosity and COP of Hb; and the solution properties of αα-Hb are a function of extent of PEGylation (). The COP and viscosity of (Propionyl PEG 5K)6 αα-Hb at 4 gm% are 41.4 mmHg and 2.0 cP, respectively. On decaPEGylation, i.e., when the number of conjugated PEG chains are increased from six to ten, the COP and viscosity of the adduct increase to 100.6 mmHg and 3.1 cP, respectively. The increase in the COP appears to be more pronounced for the addition of four more copies of PEG 5K chain than that of the increase for viscosity.

Influence of PEG conjugation chemistry and of intramolecular crosslinking on the PEGylation induced resuscitation fluid like properties of Hb

Direct PEGylation of uncrosslinked Hb in general, and in particular the direct reductive PEGylation of Hb has been shown to weaken the interdimeric interactions of Hb resulting in the formation of essentially PEGylated dimers (CitationHu et al. 2007, Citation2009). As noted earlier, the presence of inter dimeric intramolecular crosslinking Hb completely overcomes the PEGylation induced dimerization of Hb. PEG reagents activated as succinimidyl esters have been used to introduce isopeptide linkage and urethane linkage between PEG and the amino groups of uncrosslinked and intramolecularly crosslinked Hb to establish whether this is true of these linkages as well. As shown in , SPA PEG introduces the isopeptide linkage while the SC PEG introduces urethane linkage. The FPLC patterns of Hb and αα-fumaryl Hb are compared with their hexaPEGylated derivatives, respectively, and presented in . The FPLC patterns of PEGylated uncrosslinked Hb with either urethane linkage or with isopeptide linkage are essentially comparable while the presence of the intramolecular crosslinking results in the generation of products that elute earlier from the SEC columns than their uncrosslinked counterparts, respectively. Thus, the use of undissociable derivatives of Hb generates PEGylated products with higher molecular size, i.e., the presence of intramolecular crosslink in Hb prevents the dimerization of PEGylated tetramers, irrespective of whether the PEG chains are conjugated to Hb either through isopeptide linkages or through urethane linkages.

Figure 3. SEC analysis of modified Hbs: 1, HbA; 2, αα-Hb; 3, (Propionyl-PEG5K)6-αα-Hb; 4, (Propionyl-PEG5K)10-αα-Hb; 5, (Urethane-PEG5K)6-αα-Hb; 6, (Urethane-PEG5K)10-αα-Hb.
Figure 3. SEC analysis of modified Hbs: 1, HbA; 2, αα-Hb; 3, (Propionyl-PEG5K)6-αα-Hb; 4, (Propionyl-PEG5K)10-αα-Hb; 5, (Urethane-PEG5K)6-αα-Hb; 6, (Urethane-PEG5K)10-αα-Hb.

PEGylation induced plasma expander like properties of these products are compared in . Influence of the presence of intramolecular crosslinking on the molecular radius of PEGylated products confirms the PEGylation induced dissociation of PEGylated tetrameric Hb suggested by the FPLC studies of the PEGylated Hbs. The COP of the hexaPEGylated Hb sample is also impacted by the presence of intramolecular crosslinking in Hb. COP is a colligative property, reflecting the number of particles in solution. The dissociation of Hb tetramers into dimers increases the number of particles. Being a colligative property, the increased COP reflects the presence of a significant amount of dissociated PEGylated Hb dimers in the preparation of hexaPEGylated uncrosslinked Hb. The intramolecular crosslinking in Hb appears to prevent the dissociation of Hb into dimers after PEGylation, hence the COP of this hexaPEGylated intramolecularly crosslinked Hb is lower.

Table III. Influence of conjugation chemistry on the plasma expander like solution properties of PEG-Hb conjugates.

The role of PEG conjugation chemistry and αα-fumaryl crosslink in modulating oxygen affinity of hexaPEGylated Hb

The oxygen affinity of Hb is reduced significantly by the presence of the intramolecular crosslinking. Direct hexaPEGylation of uncrosslinked Hb with an isopeptide linkage or a urethane linkage between PEG chains and protein results in an increase in oxygen affinity (). However, there is some degree of sensitivity to the PEG conjugation chemistry, the PEG conjugation through an isopeptide linkage does reduce the oxygen affinity of both PEGylated Hbs, as the conjugation through urethane linkage. The presence of intramolecular crosslinking reduces the oxygen affinity significantly. The increase in the oxygen affinity of the intramolecularly crosslinked Hb on hexaPEGylation is essentially independent of the chemistry of conjugation, a result that is very distinctly seen on the hexaPEGylation of uncrosslinked Hb.

Table IV. Influence of conjugation chemistry on O2 affinities of PEG-Hb conjugates.

Influence of the propionylation vs propionylation based PEGylation on the modulation of the oxygen affinity of intramolecularly crosslinked Hb

The increase in the oxygen affinity of αα-fumaryl Hb has also been examined as a function of extent of PEGylation using isopeptide chemistry PEGylation platform (). Upon hexaPEGylation of αα-Hb, the P50 is lowered from 32 mmHg to 22.8 mmHg. This represents a significant (30%) increase in the oxygen affinity of the molecule. Concomitant with this increase in oxygen affinity, the cooperativity (Hill coefficient) is decreased significantly, namely to 1.7 from 2.8. When the level of PEGylation is increased to decaPEGylation, the oxygen affinity increased further. The P50 of (Propionyl-PEG5K)10-αα-Hb is 17.6 mmHg (a 45% increase in the oxygen affinity). The influence of the increased PEGylation on the Hill coefficient is marginal, which drops from 1.7 to 1.6.

Table V. Influence of PEGylation of the function properties of αα-fumaryl Hb.

The increase in the oxygen affinity of αα-fumaryl Hb on PEGylation may be coming either from the chemical modification of the side chains of Hb by propionic acid, i.e., propionylation or the presence of PEG chains at the distal end of the propionyl chains on Hb. Both structural aspects may contribute to the increase, and these aspects may be additive or synergistic. Accordingly we have prepared hexapropionyl and decapropionyl αα-fumaryl Hb and have established their influence on the oxygen affinity of the αα-fumaryl Hb. Surprisingly, the oxygen affinity of both hexa and decapropionyl derivatives of αα-fumaryl Hb are higher than that of the corresponding PEGylated derivatives. Thus the PEG-shell, irrespective of whether it is made up of six copies of PEG 5K chains or of ten copies of PEG chains, attenuates the high oxygen affinity inducing propensity of propionylation of αα-fumaryl Hb.

Intramolecular crosslinking of Hb impacts the sensitivity of Hb to H+ (Bohr effects) significantly (). The Bohr effect dropped from 1.76 of native molecule Hb to 0.73 of αα-fumaryl Hb (expressed for the tetramer). The hexaPEGylation of the crosslinked Hb has only a marginal influence on the on Bohr effect, reducing the cooperativity noticeably. Further increase in the level of PEG conjugation to Hb decaPEGylation reduces the cooperativity, with a further slight reduction in the Bohr effects beyond that of hexaPEGylation.

Figure 4. The oxygen affinity of HbA control, αα-Hb, (Propionyl-PEG5K)6-αα-Hb, and (Propionyl-PEG5K)10-αα-Hb at different pH of 100 mM phosphate buffer.
Figure 4. The oxygen affinity of HbA control, αα-Hb, (Propionyl-PEG5K)6-αα-Hb, and (Propionyl-PEG5K)10-αα-Hb at different pH of 100 mM phosphate buffer.

The influence of propionylation of αα-Hb on its Bohr effect has been investigated. Again, the PEG shell of both hexaPEGylated and decaPEGylated αα-fumaryl Hb attenuates the influence of the respective level of the propionyl PEGylation of the protein.

Influence of the PEG-shell on the reactivity of thiol of Cys-93(β) of oxy αα-fumaryl Hb

The reactivity of Cys-93(β) of Hb, particularly to form mixed disulfides, is very sensitive to conformation states of Hb. The thiol of this residue is reactive only in the oxy state, and not accessible under deoxy conditions. This reactivity is determined by its surrounding microenvironment, and also appears to show a correlation with the O2 affinity of Hb (CitationHu et al. 2005, CitationMeng et al. 2009). The reactivity of reactive SH groups of Cys-93(β) of PEGylated Hbs in oxyconformation has been probed by the reaction of 4-PDS at room temperature for 30 min.

The numbers of reactive SH groups in oxy samples of HbA, αα-Hb, (Propionyl-PEG5K)6-αα-Hb, and (Propionyl-PEG5K)10-αα-Hb are 2.08, 2.01, 1.90, 1.90, respectively (). The pseudo first order plots are nearly linear for every sample (plots not shown). The apparent first-order rate constants obtained from the initial slope of each reaction are summarized in . The reactivity of SH groups of Hb (0.165 min−1) is reduced slightly on αα-fumaryl crosslinking (0.151 min−1), in spite of the fact that the oxygen affinity of the crosslinked Hb is considerably lower than that of HbA. On hexa and decaPEGylation of αα-fumaryl Hb, the reactivity of Cys-93 (β) to form the mixed disulfide with thiopyridine is decreased further, being 0.112 min−1 and 0.098 min−1, respectively. The oxygen affinity of the PEGylated crosslinked Hb is higher than the uncrosslinked Hb. However, the higher oxygen affinity, stabilization of oxy conformation by PEGylation, did not induce a higher chemical reactivity to the thiol of Cys-93(β) of the corresponding PEGylated samples.

Figure 5. Kinetics of the reactions of SH groups of HbA, αα-Hb, Propionyl6-αα-Hb, Propionyl10-αα-Hb, (Propionyl-PEG5K)6-αα-Hb, and (Propionyl-PEG5K)10-αα-Hb with 4,4-dipyridine disulfide (4-PDS).
Figure 5. Kinetics of the reactions of SH groups of HbA, αα-Hb, Propionyl6-αα-Hb, Propionyl10-αα-Hb, (Propionyl-PEG5K)6-αα-Hb, and (Propionyl-PEG5K)10-αα-Hb with 4,4-dipyridine disulfide (4-PDS).

Table VI. Influence of conjugation chemistry and of PEG shell on the reactivity of Cys-93(β).

The influence of the PEG shell on the chemical reactivity of Cys-93(β) has also been mapped by preparing hexapropionylated and decapropionylated αα-Hb and establishing the thiol reactivities of these derivatives () The reactivities of SH of Cys-93(β) of Propinyl6-αα-Hb and Propionyl10-αα-Hb are 0.118 and 0.103, which are essentially comparable to the respective PEGylated derivatives, i.e., (Propionyl-PEG5K)6-αα-Hb and (Propionyl-PEG5K)10-αα-Hb, respectively. The influence PEGylation on the microenvironments of Cys-93(β) is apparently a direct consequence of the chemistry of the conjugation, i.e., propylation, and the PEG shell by itself has little influence on the micro-circumstance of Cys-93(β).

Identification of the sites of PEGylation on αα-Hb

The site selectivity of PEGylation of αα-Hb using SPA PEG has been established by tryptic peptide mapping of the globin chains from the PEGylated proteins. The tryptic peptides in the maps of PEGylated products have been quantitated and compared with the unmodified Hb, and the results are summarized in . PEGylation is distributed over a number of the Lys residues, more than the number of PEG conjugated to the molecule, reflecting the chemical heterogeneity of the sample with respect to the sites, in spite of a reasonable symmetrical elution pattern of the materials in SEC on Superose 12 columns ().

Table VII. Site of PEGylation of αα-Hb.

The extent of PEGylation at any given site is not quantitative in the hexaPEGylated sample, i.e., it is significantly less than complete modification of that residue. The PEGylation is distributed over a number of amino groups. For a homogeneous hexaPEGylated Hb sample only three residues per αβ dimer should be modified. The tryptic maps indicate that the ε-amino groups of ten lysine residues Hb are involved in PEGylation. There are 24 amino groups per αβ dimer. The total number of individual sites of PEGylation on Hb do not vary significantly on increasing the extent of PEGylation from hexaPEGylation to decaPEGylation, only the extent to which each of these sites are PEGylated increases. The differences in the reactivity of multiple amino groups of Hb for acylation reaction (and hence for PEGylation) is not significant enough under these reaction conditions to exhibit high site selectivity. Accordingly hexa and the decaPEGylated doubly modified Hbs generated are not chemically homogeneous even though they exhibit molecular size homogeneity on FPLC analysis.

Among the 10 sites of HbA modified by SPA PEG, the ε-amino group of Lys-90(α) is the most reactive site in HbA with nearly 75 and 95% modification in hexa and decaPEGylated materials, respectively. Lys-120(β) is PEGylated to about 52 and 55% in the hexa and decaPEGylated Hbs, respectively. Lys-82(β) is modified to an extent of 42 and 44%, respectively, in the two products. The rest of the seven lysine residues, Lys-11(α), Lys-16(α), Lys-40(α), Lys-56(α), Lys-139(α), Lys-8(β), and Lys-17(β), are only lightly modified (12–25%). Except for Lys-139(α) that shows a significant increase in the extent of PEGylation on increasing the extent of PEGylation from hexa to decaPEGylation, the modification on the remaining residues in the decaPEGylate material (relative to the hexaPEGylated material), is only marginal.

The PEGylation extent of the protein has also been quantitated from the tryptic peptide mapping and compared with the quantitation determined by the NMR method (). The quantitation of PEGylation by the NMR method correlates well with the value calculated for the hexaPEGylated αα-Hb. However, the correlation is not good with the decaPEGylated material. The quantitation of the tryptic peptide maps for products with higher level of PEGylation, apparently, seems to be complicated by the incomplete digestion of the heavily PEGylated chains leading to a lower level of estimation of the extent of PEGylation.

Discussion

The early paradigms for the design of HBOCs has been that (i) oxygen affinity of HBOC should be comparable to that of Hb inside erythrocytes and (ii) the Hb should be stabilized by intramolecular crosslinking to prevent its dissociation into dimers in vivo and to attenuate the kidney filtration to avoid the nephrotoxicity. The recognition of the principles of autoregulation of circulation and that it is activated by the oxygen content on the arterial side of circulation has raised concerns on the potential activation of the autoregulatory principles by the low oxygen affinity of Hb placed in plasma. This can induce a level of vasoconstriction, and in particular the higher potency of intramolecular crosslinked Hb. Winslow and his colleagues have accordingly advanced the concept that the oxygen affinity of HBOC's should be high and noncooperative to attenuate the premature release of oxygen by blood substitutes on the arterial side and the activation of the autoregulatory mechanisms to induce vasoconstriction.

Another new paradigm advanced for the design of blood substitute is to develop it as oxygen carrying plasma expanders (CitationIntaglietta et al. 2006). This change in the design strategy came from the observation that decaPEGylated bovine Hb of Enzon is unique and distinct from other derivatives of Hb designed so far as HBOC in that is a conjugated protein and exhibits plasma expander like properties (CitationShorr et al. 1999, CitationSong et al. 1995). A new PEG conjugated Hb, EAF P5K6 Hb, was designed and developed at Albert Einstein College of Medicine using the new Extension Arm Chemistry based PEGylation platform. This is a high oxygen affinity PEG conjugated Hb, and is non-hypertensive even though it has only six copies of PEG-5K conjugated to it (CitationAcharya et al. 2005, CitationLi et al. 2006, CitationManjula et al. 2005). A prototype hexaPEGylated human Hb has been developed by Sangart based on the Einstein PEGylation platform. The choice of the hexaPEGylated Hb, MP4, for commercialization as blood substitute represents the marriage of two major conceptual aspects of the new paradigm for the design of blood substitutes, the high the oxygen affinity concept and the inducing plasma expander like properties to Hb in order to overcome the hypertensive activity of acellular Hb in vivo.

The plasma expander like properties of MP4 and reduced level of autoxidation mediated in vivo toxicity (CitationVandegriff et al. 2006, CitationHu et al. 2008) makes this PEG conjugated Hb excellent candidate for targeting delivery of oxygen to hypoxic regions of the body given its high oxygen affinity. Accordingly, the class of high oxygen affinity PEG conjugated Hbs are not ideally optimized molecules for delivering oxygen to all regions of the body. For oxygen delivery by high oxygen affinity Hbs, the oxygen tension in surrounding tissues should drop to the region corresponding to the P50 of these high oxygen affinity PEG conjugated Hbs. The oxygen affinity for MP4 is around 5 mmHg and around 8 mmHg for EAF P5K6 Hb. Accordingly, these molecules should essentially function as an effective plasma volume expander at the systemic level, and as oxygen delivering plasma expanders if there are regions of ischemia and hypoxia. Besides, the high oxygen affinity PEG conjugated Hbs under clinical trial are used only in small amounts. Accordingly, they represent only a small fraction of the total Hb in circulation (Hb present in RBC plus PEG conjugated Hb transfused and present in the plasma). Given the oxygen carrying capacity of Hb in RBCs in these situations is not limiting, the high oxygen affinity PEG-Hb present in plasma is unlikely to deliver much oxygen to normoxic regions. An oxygen affinity for conjugated Hb that is in the range of 15 mmHg is considered as an optimum oxygen tension to deliver quantifiable amounts of oxygen when it presents as an oxygen carrier in the plasma in the presence of red blood cells and the amount of Hb in RBC becomes limiting in terms of oxygen carrying capacity, close to “transfusion trigger”.

The αα-fumaryl Hb, a low oxygen affinity intramolecular crosslinked Hb, is considered as the choice Hb molecule for PEGylation to generate hexaPEGylated Hbs with an oxygen affinity intermediate to that of RBC and the current versions of PEG conjugated Hbs that are in clinical trial. Reversible protection of the thiol of Cys-93(β) of αα-fumaryl Hb is necessary to generate PEG conjugated Hb with an oxygen affinity in this range if EAF PEGylation is the chosen PEGylation platform (CitationLi et al. 2009). The chemical manipulation of reversible protection of Cys-93(β) is needed to generate an EAF hexaPEGylated Hb with a lower oxygen affinity molecule, i.e., it is necessary to target the EAF PEGylation exclusively to the ε-amino residues of Lys residues. The reversible protection protocol also attenuated the increase in the PEGylation induced autoxidation (CitationLi et al. 2009). Accordingly, other PEGylation platforms that do not PEGylate at Cys-93(β) and can generate hexaPEGylated products should be as good as the product generated using EAF PEGylation (with reversible protection of Cys-93(β)). These may be preferred simpler platforms to design second generation low oxygen affinity Hbs using αα-Hb, especially since PEGylation induced dissociation of tetramer will not be a problem with αα-fumaryl Hb.

The direct PEGylation of Hb by the active ester chemistry platform, does not modify the Cys-93(β). Accordingly, the PEGylation of Hb can be carried out without the reversible protection of Cys-93(β). The presence of the αα-fumaryl intramolecular crosslinking overcomes limitation of the PEGylation induced dissociation of Hb into PEGylated αβ dimers, when this direct PEGylation platform is used to PEGylate this crosslinked Hb. Accordingly direct hexaPEGylation and decaPEGylation of αα-fumaryl Hb have been undertaken here to generate molecules with an oxygen affinity lower than the two PEG-Hbs under clinical trials, i.e., hexaPEGylated molecule (MP4) generated by EAF PEGylation platform using maleimidopropyl PEG-5K and decaPEGylated molecule (Sanguinate) generated using succinimidyl carbamate of PEG-5K.

A very significant observation here is that the increase in molecular radius of hexaPEGylated αα-Hb seen on increasing the number of PEG 5K chains in the PEG shell from six to ten copies is very small. Consequently, the packing density of PEG within the PEG-shell encasing the central protein core, αα-Hb is increased considerably as the level of PEGylation is increased. This observation is consistent with the results on the EAF PEGylation of Albumin (CitationAnanda et al. 2012). The biochemical, biophysical and the microcirculatory consequences of the increased rigidity of the PEG-shell as the extent of PEGylation is increased have not been explored as of now. The PEGylated Hb with multiple copies of PEG chains consists of structures with distinct molecular regions of different packing densities. The protein core of high packing density is covered with a soft (spongy) cover of PEG, which gives these semisynthetic hybrid biopolymers a degree of compressibility (CitationLi et al. 2006) or ability to change their molecular shapes when subjected to stress situations (for example, shear stress). These properties of PEGylated proteins are expected to be of considerable merit in dictating the vasodilatory and microcirculatory responses of PEGylated Hb and PEGylated albumin (CitationSriram et al. 2012).

The increase in the number of PEG chains in the PEG shell is also accompanied by an increase in COP. The COP of PEGylated Hb is more than doubled as the numbers of PEG chains increased to ten from six. The increase in the COP of the PEGylated proteins is an exponential correlate of the number of the PEG chains in the PEG shell (CitationAnanda et al. 2012). The viscosity of the PEGylated Hb and PEGylated Albumin at 4 gm% is not a direct correlate of the number of PEG chains; however, it is an exponential functional of number of PEG chains just as the increase of COP.

One of the significant microvascular properties of P5K6 Albumin is its ability to mimic the physiological consequence of dextran-500, a high viscosity plasma expander. This mimicry involves the dextran 500 mediated enhanced endothelial NO production and supra perfusion (CitationSriram et al. 2012). How does the low viscosity plasma expander like P5K6 αα-Hb or EAF P5K6 albumin mimic the in vivo physiological reactions of a high viscosity plasma expander? The viscosity of dextran 500 (a 6 gm% solution) as well as the molecular radius is nearly three times that of hexa-PEGylated αα-Hb (or hexaPEGylated albumin, viscosity measured at 4 gm% solution). But the COP of the hexaPEGylated albumin (of hexaPEGylated αα-Hb) is significantly higher than that of dextran 500. Accordingly, we speculate that the uniqueness of the structure of hexaPEGylated albumin and of hexaPEGylated Hb in terms of having regions in the molecules with different packing densities endows these PEGylated molecules with novel resuscitation fluid properties.

It may be noted that the molecular radius of the decaPEGylated αα-fumaryl Hb generated here is considerably smaller than the molecular radius reported for Enzon decaPEGylated bovine Hb by CitationVandegriff et al. (1997) as well as MP4. The molecular radius has been determined here for a decaPEGylated intramolecularly crosslinked Hb by light scattering. The Enzon decaPEGylated Hb does not have an intramolecular crosslinking, and approach for the determination of molecular radius of decaPEGylated bovine Hb and of MP4 was based on the COP of the solution, a colligative property. This calculation of the molecular radius from COP of the protein solution makes an intrinsic assumption that PEGylated bovine Hb does not dissociate into PEGylated dimers. Accordingly, further studies of decaPEGylated intramolecularly crosslinked bovine Hb will be needed to understand this difference in the molecular radius of decaPEGylated bovine Hb versus decaPEGylated αα-fumaryl Hb.

Direct PEGylation of αα-fumaryl Hb using succinimidyl ester of PEG-5K propionic acid also increases the oxygen affinity of the molecule just as EAF PEGylation (CitationLi et al. 2008). The oxygen affinity increase seen with hexaPEGylation is significantly lower than that seen with EAF hexaPEGylation of αα-fumaryl Hb carried out without reversible protection of thiol of Cys-93(β). But when the EAF hexaPEGylation of αα-fumaryl Hb is carried out targeting exclusively to the ε-amino groups with a reversible protection of Cys-93(β), the increase in oxygen affinity is comparable or only slightly higher than the direct nonconservative hexaPEGylation that involves the formation of isopeptide linkages between PEG and protein. Thus the overall oxygen affinity increasing influence of the PEG shell appears to be independent of the chemistry of PEGylation, conservative versus nonconservative. The very high oxygen affinity of EAF PEGylated αα-fumaryl Hb when carried out without reversible protection of Cys-93(β) is an additive (or synergistic) effect of EAF hexaPEGylation targeted to the ε-amino groups and of the maleimide modification of Cys-93(β) (CitationLi et al. 2009).

The increase in the oxygen affinity of αα-Hb on the nonconservative direct PEGylation appears to be a function of the level of PEGylation. The oxygen affinity of the decaPEGylated intramolecularly crosslinked Hb is higher as compared to the hexaPEGylated sample. When PEGylation is increased to a higher level, an increase in the packing density of PEG shell also takes place. However, the sites modified by the PEGylation did not change as a function of increase in the extent of PEGylation, the extent to which each site is functionalized is only increased. Besides, the increase in the oxygen affinity of the crosslinked Hb by direct hexaPEGylation does not appear to be a function of the chemistry of the linkage between Hb and PEG chain. PEG conjugation by urethane linkage versus isopeptide linkage has comparable influence on oxygen affinity of crosslinked Hb when the PEGylation levels are comparable.

The increase in oxygen affinity of Hb on PEGylation has been suggested to be a consequence of PEGylation resulting in a preferential stabilization of the oxy conformation of Hb versus deoxy conformation (CitationKhan et al. 2001). This observation suggests that the only way to overcome this effect of PEGylated Hb and generate lower oxygen affinity PEGylated Hbs is to use a lower oxygen affinity Hb as the starting material for PEGylation. In the present study, the structure of Hb has been reengineered by introducing deoxy conformation stabilizing αα-fumaryl intramolecular cross bridge to lower the oxygen affinity of Hb (relative to unmodified Hb). However, when the Hb molecule has been reengineered to lower its P50, it may be necessary to consider this as a new molecule since the thermodynamics of the reversible equilibrium of oxy to deoxy conformational transition is likely to be distinct as compared to the parent molecule.

The oxygen carrying capacity of blood is dictated by the amount of total Hb present. This capacity is unlikely to be influenced by chemical modification introduced to lower the oxygen affinity of Hb to be placed in the plasma. The saturation of the molecule with oxygen takes place in the presence of abundant oxygen in the lungs. Accordingly, it can be assumed that if there are any differences in the tissue oxygenation pattern among PEG-Hbs with comparable oxygen affinities, such a difference should emerge from the influence of PEGylation pattern on propensity of the molecules to deliver its oxygen to the tissue. Crosslinking of Hb, besides reducing oxygen affinity, slightly lowers cooperativity and lowers significantly the Bohr effect, and PEGylation enhances this molecular effect. Each of these will have more influence in terms of oxygen delivery than on oxygenation of the molecule, as the tissue oxygenation is a controlled environment of oxygen tension of the tissues, very different from the environment in the lungs.

The lowering of oxygen affinity of Hb by the αα-fumaryl crossbridge at neutral pH is accompanied by a reduction in the Bohr effect. The change in the Bohr effect is essentially the dependence of affinity of oxygen to the heme of Hb as a function of pH, and does not have any influence on the oxygen carrying capacity of the molecule. The significance of the molecular properties of the designed blood substitutes have not been fully appreciated so far. In fact it has been argued by Winslow that the only molecular property relevant to be a blood substitute is high oxygen affinity (CitationCole et al. 2008, CitationWinslow 2005, Citation2008). Many enzymes have been shown to change their pH optima for the catalysis on the chemical modification of their functional groups. This is generally associated with the change in the pKa of ionization behavior of functional groups that has a pKa around neutral pH, generally His residue in the microenvironment of active site of enzymes. In the case of Hb, these represent the acid Bohr groups, the surface His residues, carboxyl groups and the alpha amino groups. It may be noted that the isoelectric focusing pattern of αα-fumaryl Hb is indistinguishable as compared to that of unmodified Hb, in spite of the fact the introduction of αα-fumaryl Hb results in the loss of two positive charges, suggesting the change in the pKa of some other amino acid at neutral pH. However, we have no information at this stage on the mapping of Bohr groups of αα-Hb, but this has been well mapped for uncrosslinked Hb by NMR by Chien Ho and his colleagues (CitationBusch and Ho 1990). The oxygen tension of the tissues, coupled with the cooperativity and the pH of the blood (Bohr effect) dictates how much of the oxy Hb within the RBC will remain oxygenated and how much of the modified Hb placed in plasma will remain as oxy Hb, i.e., how much of the oxygen will be unloaded. As long as the oxygen tension in the tissue is higher than the P50 of modified Hb molecule, modified Hbs in the plasma are inert oxygen storing plasma expanders, and the oxygen delivering task is fully executed by Hb inside the red blood cells. The level of tissue oxygenation will function as a trigger to activate the oxygen delivering capacity of the oxygen carrying plasma expanders.

The changes in the molecular properties of Hb resulting as consequences of PEGylation can be dissected out as originating from the modification of the protein by the conjugation chemistry and coupled with that from the PEG-shell. The studies presented here have established that hexa and decapropionylation have essentially no influence on the molecular radius, viscosity and COP of Hb, i.e., the plasma expander like properties are essentially endowed to the molecule by the PEG-shell engineered on the molecular surface. On the other hand hexa and decapropionylation of αα-fumaryl Hb increases the oxygen affinity. Interestingly, the increase in the oxygen affinity is higher by comparison to the increase seen on the respective PEGylation. The PEG shell attenuates the increase induced by the conjugation chemistry.

This dissection of the contribution of the conjugate chemistry versus the PEG-shell () on the oxygen affinity of αα-Hb suggests that the earlier conclusion that PEGylation stabilizes the oxy conformation of Hb (CitationKhan et al. 2001) only represents an integrated picture of the conjugation chemistry and PEG shell on Hb. The influence predominantly comes from the conjugation chemistry. Therefore, it follows that the intrinsic effect of the PEG shell engineered around Hb is to reduce the oxygen affinity of Hb, or favor the release of oxygen from heme. Thus the conjugation chemistry and PEG shell appears to achieve very different and distinct influences on the molecular property of Hb, i.e., ability to reversibly bind oxygen. The same result is seen in terms of Bohr effect. PEG-shell attenuates the impact of conjugation chemistry induced influence on the Bohr effect of crosslinked Hb.

Figure 6. Schematic representation of the PEGylated αα-Hb molecule.
Figure 6. Schematic representation of the PEGylated αα-Hb molecule.

The earlier conclusions that PEGylation stabilizes the oxy conformational state of Hb were based on the EAF PEGylated uncrosslinked Hb (CitationKhan et al. 2001). In a more recent study (CitationLi et al. 2006), we have carried out a similar chemical dissection of the functional consequence of EAF PEGylation of Hb and have established that the Extension Arm chemistry increase in the oxygen affinity of Hb. When these extension arms are decorated with PEG shell, some attenuation of the EA chemistry induced increase in oxygen affinity of Hb is accomplished. Accordingly, we conclude, that the influence of PEGylation of Hb on the molecular properties is direct correlate of the conjugate chemistry, and PEG-shell by itself has only marginal influence. On the other hand all the plasma expansion like properties of PEG-Hb are a direct consequence of the PEG shell. This gives us a better handle to modulate the functional properties of Hb, once we have optimized the PEGylation platform to achieve the active plasma expansion function of Hb.

The results presented here along with our earlier studies establish that every one of the PEGylated Hbs designed should be considered as very distinct molecular species in terms of the functional properties. On the other hand in terms of the PEGylation induced solution properties, as a first approximation, we can still continue with the working hypothesis, as when the pattern of PEGylation is conserved (number and size of PEG chains conjugated) their solution properties should be comparable. Accordingly, detailed correlation of microcirculatory properties of PEGylated Hbs with PEGylation platform and the functional properties with the conjugation chemistry is necessary in designing the best oxygen delivering plasma expanders.

Acknowledgements

We thank Dr Abraham Abuchowski (Prolong Pharmaceuticals) for providing us with the succinimidyl carbamate PEG5K reagent.

Declaration of interest

The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.

This research is partially supported by the United States Army Medical Research Acquisition Activity Grant (W81XWH-11-2-0012) to AGT.

References

  • Acharya SA, Friedman JM, Manjula BN, Intaglietta M, Tsai AG, Winslow RM, et al. 2005. Enhanced molecular volume of conservatively PEGylated Hb: (SP-PEG5K)6-HbA is non-hypertensive. Artif Cells Blood Substit Immobil Biotechnol. 33:239–255.
  • Acharya SA, Acharya VN, Kanika ND, Tsai AG, Intaglietta M, Manjula BN. 2007a. Non-hypertensive tetraPEGylated canine haemoglobin: correlation between PEGylation, O2 affinity and tissue oxygenation. Biochem J. 405:503–511.
  • Acharya SA, Intaglietta M, Tsai AG. 2007b. Modulation of the polyethylene glycol-hemoglobin structure to increase the efficiency of plasma expansion and O2 carrying capacity. Transfus Altern Transfus Med. 9:254–264.
  • Ampulski RS, Ayers VE, Morell SA. 1969. Determination of the reactive sulfhydryl groups in heme proteins with 4, 4′-dipyridinedisulfide. Anal Biochem. 32:163–169.
  • Ananda K, Manjula BN, Meng F, Acharya VN, Intaglietta M, Acharya SA. 2012. Packing density of the PEG-shell in PEG-albumins: PEGylation induced viscosity and COP are inverse correlate of packing density. Artif Cells Blood Substit Immobil Biotechnol. 40:14–27.
  • Ananthakrishnan R, Li Q, O’Shea KM, Quadri N, Wang L, Abuchowski A, Schmidt AM, Ramasamy R. 2013. Carbon monoxide form of PEGylated hemoglobin protects myocardium against ischemia/reperfusion injury in diabetic and normal mice. Artif Cells Nanomed Biotechnol. 41:428–36 [Epub ahead of print].
  • Busch MR, Ho CE. 1990. Effects of anions on the molecular basis of the Bohr effect of hemoglobin. Biophys Chem. 37:313–322.
  • Cabrales P, Tsai AG, Ananda K, Acharya SA, Intaglietta M. 2008. Volume resuscitation from hemorrhagic shock with albumin and hexaPEGylated human serum albumin. Resuscitation. 79:139–146.
  • Chatterjee R, Welty EV, Walder RY, Pruitt SL, Rogers PH, Arnone A, Walder JA. 1986. Isolation and characterization of a new hemoglobin derivative cross-linked between the α chains (lysine 99α1→lysine 99α2). J Biol Chem. 261:9929–9937.
  • Cole, RH, Vandegriff KD. 2011. MP4, a vasodilatory PEGylated hemoglobin. Adv Exp Med Biol. 701:85–90.
  • Cole R, Vandegriff K, Szeri A, Savas O, Winslow R. 2008. Targeted O2 delivery by blood substitutes: in vitro arteriolar simulations of first- and second-generation products. Microvasc Res. 76:169–179.
  • Fronticelli C, Bucci E, Orth C. 1984. Solvent regulation of oxygen affinity in hemoglobin. J Biol Chem. 259:10841–10844.
  • Hu T, Prabhakaran M, Acharya SA, Manjula BN. 2005. Influence of the chemistry of conjugation of poly(ethylene glycol) to Hb on the oxygen-binding and solution properties of the PEG-Hb conjugate. Biochem J. 392:555–564.
  • Hu T, Manjula B, Brenowitz M, Acharya SA. 2007. Influence of intramolecular cross-links on the molecular, structural and functional properties of PEGylated haemoglobin. Biochem J. 402:143–151.
  • Hu T, Li D, Manjula, BN, Acharya SA. 2008. Autoxidation of the site-specifically PEGylated hemoglobins: role of the PEG chains and the sites of PEGylation in the autoxidation. Biochemistry. 47:10981–10990.
  • Hu T, Li D, Manjula BN, Brenowitz M, Prabhakaran M, Acharya SA. 2009. PEGylation of Val-1(alpha) destabilizes the tetrameric structure of hemoglobin. Biochemistry. 48:608–616.
  • Intaglietta M, Cabrales P, Tsai AG. 2006. Microvascular perspective of oxygen-carrying and -noncarrying blood substitutes. Ann Rev Biomed Eng. 8:289–321.
  • Khan I, Dantsker D, Samuni U, Friedman AJ, Bonaventura C, Manjula B, Acharya SA, Friedman JM. 2001. Beta 93 modified hemoglobin: kinetic and conformational consequences. Biochemistry. 40:7581–7592.
  • Li D, Manjula BN, Acharya AS. 2006. Extension arm facilitated PEGylation of hemoglobin: correlation of the properties with the extent of PEGylation. Protein J. 25:263–274.
  • Li D, Hu T, Manjula BN, Acharya SA. 2008. Non-conservative surface decoration of hemoglobin: influence of neutralization of positive charges at PEGylation sites on molecular and functional properties of PEGylated hemoglobin. Biochim Biophys Acta. 1784:1395–1401.
  • Li D, Hu T, Manjula BN, Acharya SA. 2009. Extension arm facilitated pegylation of alphaalpha-hemoglobin with modifications targeted exclusively to amino groups: functional and structural advantages of free Cys-93(beta) in the PEG-Hb adduct. Bioconjug Chem. 20:2062–2070.
  • Manjula BN, Acharya AS. 2003. Purification and molecular analysis of hemoglobin by high-performance liquid chromatography. Methods Mol Med: Hemoglobin Disord. 82:31–47.
  • Manjula BN, Tsai A, Upadhya R, Perumalsamy K, Smith PK, Malavalli A, et al. 2003. Site-specific PEGylation of hemoglobin at Cys-93(beta): correlation between the colligative properties of the PEGylated protein and the length of the conjugated PEG chain. Bioconjug Chem. 14:464–472.
  • Manjula BN, Tsai AG, Intaglietta M, Tsai CH, Ho C, Smith PK, et al. 2005. Conjugation of multiple copies of polyethylene glycol to hemoglobin facilitated through thiolation: influence on hemoglobin structure and function. Protein J. 24:133–146.
  • Meng F, Manjula BN, Tsai AG, Cabrales P, Intaglietta M, Smith PK, Prabhakaran M, Acharya SA. 2009. Hexa-thiocarbamoyl phenyl PEG5K Hb: vasoactivity and structure: influence of rigidity of the conjugation linkage on the PEGylation induced plasma expander-like solution properties of PEG-Hb adducts. Protein J. 28:199–212.
  • Nacharaju P, Friedman JM, Prabhakaran M, Acharya SA, Manjula BN. 2007. Combining the influence of two low O2 affinity-inducing chemical modifications of the central cavity of hemoglobin. Biochemistry. 46:4554–4564.
  • Nho K, Glower D, Bredehoeft S, Shankar H, Shorr R, Abuchowski A. 1992. PEG-bovine hemoglobin: safety in a canine dehydrated hypovolemic-hemorrhagic shock model. Biomater Artif Cell Immobil Biotechnol. 20:511–524.
  • Shorr RG, Kwong S, Gilbert C, Benesch RE. 1999Changes in the functional properties of bovine hemoglobin induced by covalent modification with polyethylene glycol. Artif Cells Blood Substit Immobil Biotechnol. 27:185–202.
  • Song D, Olano M, Wilson DF, Pastuszko A, Tammela O, Nho K, Shorr RG. 1995. Comparison of the efficacy of blood and polyethylene glycol-hemoglobin in recovery of newborn piglets from hemorrhagic hypotension: effect on blood pressure, cortical oxygen, and extracellular dopamine in the brain. Transfusion. 35:552–558.
  • Sriram K, Tsai AG, Cabrales P, Meng F, Acharya SA, Tartakovsky DM, Intaglietta M. 2012. PEG-albumin supraplasma expansion is due to increased vessel wall shear stress induced by blood viscosity shear thinning. Am J Physiol Heart Circ Physiol. 302:H2489–2497.
  • Tsai AG, Vandegriff KD, Intaglietta M, Winslow RM. 2003. Targeted O2 delivery by low-P50 hemoglobin: a new basis for O2 therapeutics. Am J Physiol Heart Circ Physiol. 285:H1411–1419.
  • Tsai AG, Johnson PC, Intaglietta M. 2007. Is the distribution of tissue pO(2) homogeneous?Antioxid Redox Signal. 9:979–984.
  • Tsai AG, Cabrales P, Intaglietta M. 2010. The physics of oxygen delivery: facts and controversies. Antioxid Redox Signal. 12:683–691.
  • Vandegriff KD, Winslow RM. 2009. Hemospan: design principles for a new class of oxygen therapeutic. Artif Organs. 33:133–138.
  • Vandegriff KD, McCarthy M, Rohlfs RJ, Winslow RM. 1997. Colloid osmotic properties of modified hemoglobins: chemically cross-linked versus polyethylene glycol surface-conjugated. Biophys Chem. 69:23–30.
  • Vandegriff KD, Malavalli A, Wooldridge J, Lohman J, Winslow RM. 2003. MP4, a new nonvasoactive PEG-Hb conjugate. Transfusion. 43:509–516.
  • Vandegriff KD, Malavalli A, Minn C, Jiang E, Lohman J, Young MA, Samaja M, Winslow RM. 2006. Oxidation and haem loss kinetics of poly(ethylene glycol)-conjugated haemoglobin (MP4): dissociation between in vitro and in vivo oxidation rates. Biochem J. 399:463–471.
  • Vandegriff KD, Malavalli A, Mkrtchyan GM, Spann SN, Baker DA, Winslow RM. 2008. Sites of modification of hemospan, a poly (ethylene glycol)-modified human hemoglobin for use as an oxygen therapeutic. Bioconjug Chem. 19:2163–2170.
  • Winslow RM. 2000. Alpha-alpha-crosslinked hemoglobin: was failure predicted by preclinical testing?Vox Sang. 79:1–20.
  • Winslow RM. 2005. Targeted O2 delivery by low-p50 hemoglobin: a new basis for hemoglobin-based oxygen carriers. Artif Cell Blood Substit Immobil Biotechnol. 33:1–12.
  • Winslow RM. 2008. Cell-free oxygen carriers: scientific foundations, clinical development, and new directions. Biochim Biophys Acta. 1784:1382–1386.
  • Young MA, Riddez L, Kjellström BT, Bursell J, Winslow F, Lohman J, Winslow RM. 2005. MalPEG-hemoglobin (MP4) improves hemodynamics, acid-base status, and survival after uncontrolled hemorrhage in anesthetized swine. Crit Care Med. 33:1794–1804.

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.