TRANSFUSION MEDICINE
Hemodynamic responses to a hemoglobin bis-tetramer and its
polyethylene glycol conjugate
Francine E. Lui, Binglan Yu, David M. Baron, Chong Lei, Warren M. Zapol, and Ronald Kluger
BACKGROUND: The design of hemoglobin-based
oxygen carriers (HBOCs) poses a significant challenge
as clinical trials of many materials have reported
adverse side effects that may come from the scaveng-
ing of the vasodilator nitric oxide (NO). A compensating
reaction, reduction of endogenous nitrite by hemoglobin
(Hb) and its derivatives, generates NO. Polyethylene
glycol (PEG) conjugation of Hb enhances the rate of
the reaction.
STUDY DESIGN AND METHODS: Hemoglobin bis-
tetramers (BT) and their PEGylated derivative (BT-PEG)
bind oxygen with a degree of cooperativity and also
have significantly enhanced nitrite reductase activity
compared to the native protein. Circulatory evaluation
will test if the properties of BT and BT-PEG are
reflected in their effects in vivo. BT and BT-PEG were
evaluated as infusions into healthy wild-type (WT) and
diabetic (db/db) mouse models. The effects were com-
pared to infusions of murine Hb.
RESULTS: The materials were found not to cause sig-
nificant increases in systemic blood pressure in either
WT mice or db/db mice. The latter are highly sensitive
to NO scavenging. Further hemodynamic measure-
ments in WT mice indicate that while a slight increase
in systemic vascular resistance (SVR) was observed
after infusion of BT, the extent is not significant. No
change in SVR from baseline was observed after infu-
sion of BT-PEG.
CONCLUSION: The enlarged Hb derivatives do not
evoke unfavorable circulatory responses that have been
noted to result from infusion of Hb derivatives. These
results suggest that a compromise between the P50, n50,
and nitrite reductase activity of a Hb derivative can
serve as the basis for producing HBOCs that can be
tested for vasoactivity.
974 TRANSFUSION Volume 52, May 2012
_3421 974..982
V
arious chemically modified human and bovine
hemoglobin (Hb) derivatives have been pro-
duced and evaluated as alternatives to red
blood cells (RBCs) as oxygen carriers in
transfusions.1-4 However, studies of a variety of potential
Hb-based oxygen carrier (HBOCs) in clinical trials indi-
cate that serious problems remain. Notable problems
include increased blood pressure due to hypertension and
an elevated risk of heart attacks.5 It has been proposed
that the vasoactivity and subsequent increase in blood
pressure elicited by HBOC infusion is the result of extrava-
sation of low-molecular-weight Hb derivatives and scav-
enging of nitric oxide (NO), an endothelium-derived
signaling molecule that plays a crucial role in maintaining
vasodilator tone.2,6-8
It has been reported that the production of NO as
catalyzed by deoxyhemoglobin may be a contributing
factor in hypoxic vasodilation.9 It follows that modifica-
tions that enhance the nitrite reductase activity of Hb
ABBREVIATIONS: BT = bis-tetramers of hemoglobin;
BT-PEG = bis-tetramers of hemoglobin that are specifically
PEGylated; db/db = diabetic (mice); HBOC(s) = hemoglobin-
based oxygen carrier(s); LV = left ventricle; metHb = methemo-
globin; mTet = tetrameric Hb; NAC = N-acetylcysteine; SBP =
systolic blood pressure; SVR = systemic vascular resistance;
WT = wild type.
From the Davenport Chemical Research Laboratories, Depart-
ment of Chemistry, University of Toronto, Toronto, Ontario,
Canada; and the Anesthesia Center for Critical Care Research,
Department of Anesthesia and Critical Care, Massachusetts
General Hospital, Harvard Medical School, Boston,
Massachusetts.
Address correspondence to: Ronald Kluger, Department of
Chemistry, University of Toronto, Toronto, Ontario, Canada,
M3B 1J7; e-mail: rkluger@chem.utoronto.ca.
Supported by the Canadian Blood Services, Inc., through
the Canadian Institutes of Health Research.
Received for publication June 13, 2011; revision received
September 15, 2011, and accepted September 15, 2011.
doi: 10.1111/j.1537-2995.2011.03421.x
TRANSFUSION 2012;52:974-982.

derivatives should lead to an increased rate of reduction to
produce NO, which would serve to counteract NO scav-
enging by hemes. The maximal nitrite reduction rate is
linked to the allosteric structural transition of the Hb
tetramer from the oxygenated conformation (R state) to
the deoxygenated conformation (T state).10 Hb’s nitrite-
reductase activity is thus affected by covalent modifica-
tions to Hb.
Hb derivatives that are reported to have minimal clini-
cal vasoactivity11 also have an especially high-level heme-
nitrite reductase activity.12 In particular, Hb derivatives that
are modified with polyethylene glycol (PEG) chains by con-
jugation at b-Cys93 produce NO at a faster rate than does
the native protein.12 This suggests that modification at this
key residue results in a conformational change that stabi-
lizes the R-state, an effect that is especially important for
increased reductase activity.9 Thus, according to this
hypothesis, b-Cys93 PEG-Hbs are able to alleviate the
vasoconstrictive response to NO scavenging observed by
restoring the NO concentration at the endothelium.
Based on the above considerations, features of an
ideal HBOC should also include an increase in size to
prevent extravasation through endothelial junctions into
the interstitium where they can avidly scavenge NO.6,8 The
use of chemically well-defined materials presents a clear
advantage in evaluating the sources of various physiologi-
cal effects. Pioneering attempts to create Hb derivatives
with increased molecular weight typically involved oligo-
merization of Hb tetramers with the nonspecific reagents
such as glutaraldehyde.13,14 These reagents produced het-
erogeneous mixtures (typically consisting of cross-linked
monomers, dimers, and trimers) with a mean molecular
weight range of 130 to 500 kDa, along with unreacted Hb.4
Other chemically enlarged proteins such as PEGylated
Hbs11 are also typically heterogeneous collections, espe-
cially in the case of thiol-mediated reactions.15 The vari-
ability makes it difficult to correlate the infusion of such
HBOCs with clinical events.
In considering characteristics for an ideal HBOC, we
developed an alternative modification protocol based on
tetra-functional cross-linkers that react selectively within
Hb tetramers, while creating interprotein linkages
between tetramers.16-18 We refer to the resulting materials
as bis-tetramers of Hb (BT; Fig. 1). They are structurally
Fig. 1. BT are cross-linked tetramers with an interprotein linkage. PEGylation at b-cysteine residues produces BT-PEG with four
PEG chains (n = 110) attached at the protein surface.
PEGYLATED BIS-TETRAMERS OF Hb IN CIRCULATION
well-defined high-molecular-weight compounds with
altered geometry and increased oxygen-carrying ability.
The cross-linking reaction is specific to amino groups of
the 2,3-bisphosphoglycerate binding site, creating specific
cross-linked tetramers and excluding the presence of dis-
sociated ab-dimers.18,19
We also produced and studied bis-tetramers of
Hb that are specifically PEGylated (BT-PEG) through
maleimide-PEG conjugation at the thiol of b-Cys93
(Fig. 1).19 BT-PEG should be superior to PEG-Hb because
the covalent cross-link within the BT tetramers prevents
dissociation into dimers. PEGylation further increases the
size of the protein, increases oxygen affinity, and enhances
nitrite-reductase activity.19 Both BT and BT-PEG are pure
and well-defined compounds with high cooperativity in
oxygenation (n50 = 2.7 and n50 = 2.4, respectively), a desir-
able property for effective targeted oxygen delivery.
The systemic vasoconstrictor effects of both BT and
BT-PEG were studied in awake mice using noninvasive
measurements and in anesthetized mice with invasive
hemodynamic measurements. We evaluated effects in
both healthy wild-type (WT) mice and diabetic (db/db)
mice. The latter have endothelial dysfunction and are
extremely sensitive to the effects of NO scavenging by
extracellular Hb.20
Previous studies indicated that adverse complica-
tions from administration of HBOCs in healthy, inbred
young animals do not mimic the complications seen in
clinical trials.4 Many of the patients in clinical studies
needing transfusions have significant underlying vascular
disease and significant health disorders.4 A prominent
feature of many of these patients is the presence of endot-
helial dysfunction, including impaired NO release from
the endothelium.21,22 If a potential HBOC is released into
the circulation it may then scavenge what little NO is
available. Thus, our study of db/db mice with existing
endothelial dysfunction serves as a model for the adverse
effects of HBOCs in clinical situations.4
MATERIALS AND METHODS
Preparation of BT and BT-PEG
Hb derivatives were prepared according to the methods
described by Lui and Kluger.19 Briefly, native cell-free car-
Volume 52, May 2012 TRANSFUSION 975
LUI ET AL.
bonmonoxyhemoglobin (20 mL, 1.5 mmol/L, 0.03 mmol)
was exchanged into 0.05 mol/L sodium borate buffer,
pH 7.4. This Hb mixture (approx. 0.6 mmol/L) was oxy-
genated at 0°C for 3 hours and then deoxygenated with
humidified nitrogen at 37°C for 3 hours. The tetra-
functional cross-linker, 3,5-tetrabromosalicyl benzylsul-
fone19 (0.157 g, 0.09 mmol) was dissolved in 420 mL of
dimethyl sulfoxide and the reaction was allowed to
proceed overnight. The mixture was then flushed vigor-
ously with carbon monoxide for 10 minutes and the crude
mixture was exchanged into 0.1 mol/L MOPS (pH 8), con-
centrated, and stored at 4°C under CO.
Purification of BT was carried out using gel-filtration
chromatography (Sephadex G-100, 1000 ¥ 35 mm) with 25
mmol/L Tris-HCl buffer, pH 7.4, containing 0.5 mol/L
magnesium chloride. Fractions were collected, concen-
trated, and analyzed using Superdex G-200 size-exclusion
chromatography, C4 reverse-phase analytical high-
performance liquid chromatography (HPLC), and sodium
dodecyl sulfate–polyacrylamide gel electrophoresis. Only
fractions with pure BT were used for further experiments.
Conjugation with PEG was carried out in a similar method
as described by Lui and Kluger.19
The PEGylation material O-(2-maleimidoethyl)-O′-
methylpolyethylene glycol-(5000) (Mal-PEG5K, ⱖ90%
purity) was purchased from Fluka (Buchs, Switzerland).
Solutions of purified BT (40 mL, 0.5 mmol/L, 0.02 mmol)
were combined with 10 eq of Mal-PEG5K (1 g, 0.2 mmol)
in sodium phosphate buffer (0.1 mol/L, pH 7.4) and kept
at 37°C overnight to give products with two PEG chains per
tetramer: BT-(PEG5K)4. The resulting mixture was passed
through a Sephadex G-25 column equilibrated with
sodium phosphate buffer (0.1 mol/L, pH 7.4), concen-
trated, and stored at 4°C. Separation from excess PEG
reagents was carried out by dialysis in phosphate buffer
(I = 0.1 mol/L, pH 7.4) at 4°C. The buffer was replaced
three times, at 12-hour intervals. The PEGylated Hbs were
flushed with carbon monoxide and stored at 4°C. Analysis
of the modified Hbs was carried out with a prepara-
tive size-exclusion column: Superdex G-200HR (10 ¥
300 mm). Protein samples (0.5 mmol/L) were eluted
under partially dissociating conditions by the addition of
0.5 mol/L magnesium chloride in buffer (25 ¥ 10-3 mol/L
Tris-HCl, pH 7.4). The effluent was monitored at 280 nm.
Sterilization of Hb samples
The Hb derivatives were exchanged into a modified
phosphate-buffered saline (PBS) solution (0.066 g/dL
N-acetylcysteine [NAC], pH 7.1) using a Sephadex G-25
column. The antioxidant NAC was added to the buffer in
an effort to minimize rapid auto-oxidation of ferrous
heme to ferric heme and its concentration adjusted to be
in a sixfold ratio to Hb. The final concentration of both BT
and PEG-BT were adjusted to be 4 g/dL. The Hb samples
976 TRANSFUSION Volume 52, May 2012
were then oxygenated as according to the method previ-
ously described, bulk-filtered with a 0.45-mm filter fol-
lowed by two sequential 0.2-mm sterile filtration steps and
stored frozen at -20°C.
Animal preparation
This study was approved by the Subcommittee on
Research Animal Care of Massachusetts General Hospital
(Boston, MA). All mice were obtained from The Jackson
Laboratory (Bar Harbor, ME). The animals in the study
were 8- to 10-week-old (25-30 g) male C57BL/6J WT mice
and B6.Cg-m+/+Leprdb/J (C57BL/6J background) db/db
mice.
Preparation of murine tetrameric Hb solution
Whole blood was obtained from WT mice by through
puncture of the left ventricle (LV) into a heparinized
syringe.23 The collected blood was diluted with PBS in a 1:2
ratio and subject to a freeze-thaw cycle (-80°C to room
temperature) three times to lyse the cells. The mixture was
centrifuged at 21,600 ¥ g for 1.5 hours at 4°C to separate
Hb from the cell debris. The supernatant was collected
and filtered through a 0.2-mm Nalgene filter and dialyzed
against deoxygenated 0.9% saline solution overnight at
4°C. After purification with dialysis, the solution was con-
centrated using a filter (30,000 MWCO centrifugal filter,
Centricon, Millipore, Billerica, MA) to 4 g/dL.
Measurement of systolic blood pressure in awake mice
Systolic blood pressure (SBP) was measured with a nonin-
vasive blood pressure system (XBP 1000, Kent Scientific,
Torrington, CT) in awake mice. Briefly, the mouse was
placed in a restrainer (Kent Scientific) for 10 minutes three
times to allow the animal to acclimatize to the device until
the mice remain comfortable for prolonged periods. The
Hb derivatives (0.48 g Hb/kg) were then administered
over 1 minute via a tail vein injection to allow for an
approximately 16% top load infusion (e.g., 0.3 mL in a 25-g
mouse). For experiments with db/db mice, the amount of
samples infused was kept to be 0.48 g/kg, leading to a
higher top-load infusion (0.6 mL in a 50-g mouse). SBP
was measured every 10 minutes for 1 hour. The HBOCs of
study, BT (0.48 g/kg) and PEG-BT (0.48 g/kg), were admin-
istered at the same concentration. The animal studies also
included positive and negative control groups of murine
tetrameric Hb (mTet; 0.48 g/kg) and modified PBS
(0.066 g/dL NAC, pH 7.1).
Invasive hemodynamic measurements in
anesthetized mice
Invasive hemodynamic measurements were performed as
described previously.24 Briefly, mice were anesthetized by
intraperitoneal injection with ketamine (120 mg/kg), fen-
tanyl (50 mg/kg), and pancuronium (2 mg/kg). After intu-
bation, animals were mechanically ventilated (FiO2 100%,
 PEGYLATED BIS-TETRAMERS OF Hb IN CIRCULATION

10 mL/g, 120 breaths/min), and a fluid-filled catheter was
inserted into the left carotid artery for infusion of saline
(1.5 mL/hr). One-hundred percent O2 was used due to the
complex nature of the open chest model used in the
experiment; the high oxygen level is required to keep
the animal alive during the course of the experiment.
Breathing atmospheric oxygen (21% O2) can cause
unwanted hypoxia and lead to hypoxic pulmonary vaso-
constriction. A thoracotomy was then performed, and a
pressure-volume conductance catheter (Size 1F, Model
PVR-1030, Millar Instruments, Inc., Houston, TX) was then
inserted via the apex into the LV. Once stable hemody-
namic measurements were obtained, Hb solutions—BT,
BT-PEG, and mTet (all at 0.48 g/kg)—were infused
through a catheter placed in the right jugular vein at a rate
of 75 mL/min. Systemic vascular resistance (SVR) was cal-
culated based on mean arterial pressure (MAP), central
venous pressure (CVP), and cardiac output (CO). The fol-
lowing formula was used:
SVR = ([MAP − CVP]/CO).
described in Table 1. Both HBOCs are derived from
human Hb, and both are cross-linked between the b sub-
units, thereby preventing tetrameric dissociation. The
nature of the cross-linking reaction provides an interpro-
tein linkage that tethers two Hb tetramers via a set of cova-
lent bonds. After modification, the Hb derivatives are
transferred into a modified PBS buffer containing the anti-
oxidant NAC to minimize autooxidation of the ferrous
heme to ferric heme (metHb). The final protein concen-
tration used for animal infusions was adjusted to 4 g/dL.
As a result of the extensive purification procedure needed
to obtain pure BT, there were no unmodified tetrameric
species present in the final product that was infused
(Fig. 2). The minor peak in the BT solution with retention
time of approximately 33 minutes comes from cross-
linked Hb hydrolysis product,18 and the minor peak in the
BT-PEG solution with retention time of approximately 38
minutes appears to be residual unreacted Mal-PEG. The
process that produces BT is subject to competing hydroly-
sis of the active ester. The reaction of the functional cross-

Blood sampling TABLE 1. Chemical and physical properties of BT

After 2 hours of injection, the mice were anesthetized with and BT-PEG
an intraperitoneal injection of ketamine (120 mg/kg) and BT BT-PEG
xylazine (4 mg/kg) and whole blood was obtained through Source Human Human
Modification Cross-linked Cross-linked
open chest cardiac puncture with a heparinized syringe. Interprotein linked Interprotein linked
The mouse was then euthanized by cervical dislocation. PEGylated
The blood gas values and oximetry values were then ana- Buffer Modified PBS (NAC) Modified PBS (NAC)
pH 7.1 7.1
lyzed with a blood gas analyzer (ABL800, Radiometer, Concentration 4.0 ⫾ 0.1 4.0 ⫾ 0.1
Copenhagen, Denmark) flex system to determine plasma of Hb (g/dL)
methemoglobin (metHb) levels. Blood plasma was % Tetramer 0 0
P50 (mmHg) 9.3 4.1
obtained by centrifuging the collected whole blood at n50 2.7 2.4
4000 ¥ g for 8 minutes at 4°C and stored at -80°C. Hb and NiR (/mol/L/sec)* 0.7 ⫾ 0.05 1.8 ⫾ 0.05
metHb concentrations in plasma were determined by the Total metHb (%) 20.2 24.1
cyanomethemoglobin method measuring absorption at * Rate constants taken from Lui and Kluger.19
540 and 630 nm with a spectrophotometer.

Statistical analysis
All data are expressed as mean ⫾ SD, unless otherwise
noted. Data of SBP and heart rate in awake mice were
analyzed by a repeated measures two-way analysis of vari-
ance with interaction (SigmaStat 3.0.1, Systat Software,
Inc., San Jose, CA). A paired t test was applied in the inva-
sive hemodynamic measurements in anesthetized mice.
Group comparison analysis was carried out with com-
puter software (MATLAB, r2010b, The MathWorks, Natick,
MA) to perform the t test after analysis of variance for
plasma Hb and metHb levels. A p value less than 0.05 was
considered significant.

RESULTS
Preparation of BT and BT-PEG
BT and BT-PEG were prepared according to reported
methods.19 Their chemical and physical properties are
Fig. 2. Characterization of BT (- - -) and BT-PEG (—). Size-
exclusion G-200 HPLC under dissociating conditions indi-
cated that the high-molecular-weight BT-PEG was fully
modified with high sample homogeneity.

Volume 52, May 2012 TRANSFUSION 977

LUI ET AL.

linker with Hb gives 40% BT.18,19 The remainder of the (0.48 g/kg, n = 5 each) did not significantly increase
sample is composed of cross-linked Hb that is not con- blood pressure, with SBP remaining at 113 ⫾ 5 and 113 ⫾
nected through an interprotein linkage to other tetram- 4 mmHg, respectively. In comparison, administration of
ers.19 Thus, during the purification process that we use to murine mTet (4 g/dL, n = 4) as a positive control resulted
obtain BT, the only Hb contaminants (<5%) are cross- in an increase in SBP from 118 ⫾ 6 to 128 ⫾ 3 mmHg that
linked tetramers that do not dissociate. was sustained over 1 hour after infusion (p < 0.05). A
BT and BT-PEG were oxygenated before administra- further control study was conducted with mTet in PBS/
tion to animal subjects. To minimize auto-oxidation when NAC to test the effect of NAC on blood pressure. It was
the heme is exposed to oxygen, the samples were stored found that there was no effect of NAC on the blood
frozen after oxygenation. Upon thawing, up to 20% to 25% pressure; mTet in PBS/NAC still caused the increase in SBP
of each sample was oxidized to nonfunctional metHb. (p > 0.01; data not shown).
This oxidation of ferrous heme during freeze-thaw cycles Due to the high levels of metHb (approx. 20%) present
has previously been observed: auto-oxidation has been a in the samples used in the animal studies, we examined if
problem in the design of HBOCs25 that may be prevented the decreased vasoactivity of BT and BT-PEG could be the
by the use of CO derivatives. Studies of animals with mate- result of their high metHb levels. Since approximately 20%
rials containing high concentrations of metHb have of the samples injected were nonfunctional metHb that
shown that metHb infusion does not significantly alter cannot scavenge NO, the actual functional HBOC was
blood pressure levels. Hb derivatives that elicit blood pres- effectively given at a lower dosage (approx. 80%). We
sure increases continue to do the same even with up to injected a lower dose (0.18 mL instead of 0.3 mL in a 25-g
14.5% metHb.23,26 We therefore conducted hemodynamic mouse) of the positive control (mTet) and monitored
measurements based on these materials. blood pressure levels. Even at this lower dosage (60%
volume), 0.6xmTet (n = 6) elicited a marked increase of
SBP (from 115 ⫾ 4 to 130 ⫾ 5 mmHg) that was sustained
Hemodynamic effects of infusing Hb solutions in over the 1-hour course of the experiment (p < 0.05,
awake WT mice Fig. 3A). This indicates that the 20% lower effective dosage
Infusion of BT (4 g/dL) and BT-PEG (4 g/dL) did not cause of ferrous BT and BT-PEG was unlikely to be a contributing
systemic hypertension when injected into the tail vein of factor to their lack of hypertensive effects.
awake WT mice (Fig. 3A). The animals were not anesthe-
tized, minimizing confounding effects. Baseline control
experiments with injections of modified PBS (n = 4) Hemodynamic effects of Hb solutions in awake
resulted in an unchanged SBP of 110 ⫾ 3 mmHg. No db/db mice
change of blood pressure was observed over 1 hour. db/db mice exhibit endothelial dysfunction and have
Intravenous (IV) administration of either BT or BT-PEG reduced endothelial reactivity due to acetylcholine chal-
lenge.20 They provide an excellent in vivo
model to assess even minor levels of
oxyHb-mediated NO scavenging.20 We
measured the change in SBP induced by
IV infusion of BT and BT-PEG in awake
db/db mice and found that the SBP did
not differ from a modified PBS injection
(Fig. 3B). In contrast, it has previously
been demonstrated that infusion of
murine mTet into db/db mice induces a
27 mmHg increase in SBP.20 After a
transient increase, SBP had returned
to 126 ⫾ 5 mmHg for BT and 122 ⫾
9 mmHg for BT-PEG 30 minutes after
infusion and remained constant until
Fig. 3. SBP measurements in awake WT mice. (A) Infusion of BT and BT-PEG by tail the end of the experiment. The transient
vein injection into healthy awake WT mice did not cause significant differences in increase of blood pressure is likely due to
SBP compared to baseline injections of modified PBS, while injections of mTet the stimulation of a venipuncture and
resulted in a sustained increase in SBP. Even at a lower dosage (0.6xmTet), the large volume of fluids being infused as a
increase in SBP was p < 0.05. (B) SBP measurements in awake db/db mice with topload. The volume of fluid infused is
endothelial dysfunction. No significant change was observed between infusion of BT calculated based on the animal’s weight;
and BT-PEG compared to modified PBS control. however, since the animal is the same

978 TRANSFUSION Volume 52, May 2012


Fig. 4. Plasma Hb levels and metHb levels in WT and db/db mice. (A, B) After a
2-hour infusion in WT mice, 5% of mTet remained in the blood stream. Both BT and
BT-PEG indicated a longer retention time in the circulatory system, 15% retained in
WT mice and 20% retained in db/db mice. (C, D) Plasma metHb levels were
increased upon infusion of BT (5%) and BT-PEG (9%) for both WT and db/db mice.
age as the WT mice, the additional weight comes from
excess fat, making the db/db mice top load larger than that
for WT mice. It appears that even in db/db mice that are
extremely sensitive to NO scavenging by extracellular
Hb, both BT and BT-PEG do not disrupt important
NO-signaling processes and do not alter systemic blood
pressure.
Plasma Hb and metHb levels
Plasma Hb levels and metHb levels were analyzed (Fig. 4).
Infusion of modified PBS did not alter cell-free plasma Hb
levels and only a small amount of cell-free Hb was observed
after the infusion of PBS (1% in WT mice and 0.05% in
db/db mice). In contrast, administration of BT, BT-PEG,
and mTet increased the level of cell-free Hb in plasma.
Analysis of plasma metHb levels indicated that
metHb levels were highest after BT-PEG infusion, up to
10% in both WT mice and db/db mice. Although the origi-
nal solutions contained approximately 20% to 25% metHb
due to auto-oxidation, the low levels of plasma metHb (5%
in BT and 10% in BT-PEG) after infusion indicate no sig-
PEGYLATED BIS-TETRAMERS OF Hb IN CIRCULATION
nificant further oxidation of the heme
iron. It also appears that the healthy
mouse was able to either remove or
chemically reduce the ferric metHb
within the course of the experiment.
This may come from the addition of the
reducing agent, NAC, as an additive in
the PBS used to dissolve the Hb.
Invasive hemodynamic effects in
anesthetized WT mice
Invasive LV hemodynamic measure-
ments were obtained in anesthetized
WT mice before (baseline) and 5 min-
utes after the infusion of Hb solutions
(Table 2). Infusion of the positive control
murine tetramer at a concentration of
4 g/dL resulted in a significant increase
in LV end-systolic pressure (104 ⫾ 4
mmHg vs. 148 ⫾ 7 mmHg; p < 0.05)
and SVR (6 ⫾ 1 mmHg/min/mL vs.
10 ⫾ 1 mmHg/min/mL; p < 0.05). All
other measured variables did not differ.
The observed systemic vasoconstric-
tion is consistent with the previously
reported vasoactive effect of injecting
cell-free murine Hb tetramers.23
In contrast, administration of both
BT and BT-PEG at a concentration of
4 g/dL showed no significant increase in
either LV end-systolic pressure or SVR. A
slight increase in SVR is observed upon
infusion of BT, but this is insignificant from baseline infu-
sions. No change from baseline is observed upon admin-
istration of BT-PEG. This correlates with results in
noninvasive blood pressure experiments in awake mice,
suggesting that both BT and BT-PEG did not produce
vasoconstriction in healthy WT mice.
DISCUSSION
We have examined the systemic vasoconstrictor effects of
both BT and BT-PEG in awake and anesthetized mice and
find that systemic vasoconstriction is not produced by
these compounds (compared to injections of murine Hb).
Also, infusion into db/db mice exhibiting endothelial dys-
function demonstrate that infusion of either BT or BT-PEG
does not alter SBP.
The Hb derivatives BT and BT-PEG used in this study
were specifically designed to be chemically enlarged and
with high oxygen affinities. The high purity of the prepa-
rations may also contribute to their nonhypertensive
effects. Efficient oxygen delivery is governed by an HBOC’s
ability to acquire oxygen in the lungs and release oxygen at
Volume 52, May 2012 TRANSFUSION 979
LUI ET AL.

TABLE 2. Cardiac function and systemic hemodynamic measurements in anesthetized WT mice*

Murine tetramer (n = 5) BT-PEG (n = 6) BT (n = 6)
Baseline 5 minutes after Baseline 5 minutes after Baseline 5 minutes after
HR (bpm) 565 ⫾ 34 528 ⫾ 24 571 ⫾ 25 530 ⫾ 35 608 ⫾ 17 554 ⫾ 25
LVESP (mmHg) 104 ⫾ 4 148 ⫾ 7† 99 ⫾ 6 110 ⫾ 4 106 ⫾ 7 111 ⫾ 7
LVEDP (mmHg) 6⫾1 10 ⫾ 3 5⫾1 9⫾1 5⫾1 7⫾1
CVP (mmHg) 7⫾1 7⫾1 5⫾1 6⫾1 5⫾1 6⫾1
CO (mL/min) 16 ⫾ 3 15 ⫾ 2 15 ⫾ 2 19 ⫾ 3 17 ⫾ 1 14 ⫾ 1
t (msec) 6⫾0 6⫾1 6⫾0 6⫾0 6⫾0 6⫾0
SVR (mmHg/min/mL) 6⫾1 10 ⫾ 1† 7⫾1 7⫾1 6⫾1 8⫾1
* Values are mean ⫾ SEM.
† p < 0.05 differs vs. baseline in WT mice.
CO = cardiac output; CVP = central venous pressure; HR = heart rate; LVEDP = left ventricular end-diastolic pressure; LVESP = left ventricu-
lar end-systolic pressure; t = time constant of isovolumic relaxation.

tissue sites. RBCs in whole blood have a P50 of 26 to 31
torr.27 In contrast, the smaller materials may act as a type
of bucket brigade for transferring oxygen from RBCs to the
blood vessel walls, facilitating oxygen diffusion.28 Both BT
and BT-PEG have high affinities for oxygen (P50 = 9.3 and
4.1, respectively) and increased molecular size.19
Oxygen delivery from RBCs within the physiological
range is assisted by the sigmoidal nature of oxygen
binding and release.27 An HBOC with good cooperativity
(n50 = 2.5-3) will be able to offload in a sharply defined
region of uniform oxygen tension. Both BT (n50 = 2.7) and
BT-PEG (n50 = 2.4) retain excellent cooperativity.19 This is
likely to result from cross-linking the tetrameric protein at
sites that do not interfere with the R to T state conforma-
tional changes.29 BT and BT-PEG are thus expected to
deliver oxygen in a suitably sigmoidal manner that is
more efficient than simple, un-cross-linked PEG-Hb
(n50 = 1.8).12
The rate of reduction of nitrite to NO may also con-
tribute to lack of vasoconstriction produced by either
material. Both these compounds have increased rates of
nitrite reduction,19 producing NO at faster rates than
would native Hb in RBCs. While BT exhibits a slightly
faster rate of nitrite reduction, PEGylation at b-Cys93 to
give BT-PEG further enhances heme-nitrite reaction rates
(approx. seven times faster than native Hb).19 This hypoth-
esis would predict hypertension on the order mTet > BT >
BT-PEG. However, although a slight increase in SVR is
observed during invasive measurements of infusions of
BT compared to BT-PEG, no statistical difference in SBP
between infusions of BT or of BT-PEG were observed
during noninvasive hemodynamic measurements in
awake mice. It appears that the level of the nitrite reduc-
tase activity of BT-PEG is not critical for its nonhyperten-
sive effect: BT, which reduces nitrite more slowly, is also
able to maintain normal blood pressure levels. Instead,
the production of NO from nitrite at physiologically rel-
evant sites (sites with low pO2) could be one of the con-
tributing factors to the nonhypertensive effect of these
compounds above a threshold level. We have previously
found that the correlation between NiR and the oxygen
binding properties (P50 and n50) may play a significant role
in making these materials nonvasoactive.12 The site at
which NiR is maximal, and where the rate of NO produc-
tion is fastest, depends directly on P50 of the material.30,31
BT and BT-PEG both have high oxygen affinities (P50 = 9.3
and 4.1, respectively). Thus, maximal production of NO
from nitrite as catalyzed by Hb will occur at tissue sites
with low pO2, counteracting NO scavenging by HBOCs.
The enhanced level of nitrite reduction at physiologically
relevant sites may be a contributing factor to making
these materials nonhypertensive. Recent work also indi-
cates that N2O3 generated by the concerted reductase/
anhydrase activity of these Hbs can contribute to the
nonhypertensive effect observed by BT and BT-PEG.32 The
coupling of these two cycles allow for the metHb pro-
duced from the reductase reaction to be regenerated to
functional ferrous Hb, which is consistent with our
observed low plasma metHb levels.
CONCLUSIONS
We have evaluated the physiological responses elicited
by a Hb cross-linked bis-tetramer and its PEGylated
derivative. Administration of the materials into circula-
tion in mice does not increase either SBP or SVR with
either normal WT or db/db mice with endothelial
dysfunction. Although the exact mechanisms that con-
tribute to the nonhypertensive nature of BT and BT-PEG
have yet to be elucidated, the lack of vasoconstriction
could come from a combination of their high purity
and chemical specificity, as well as from the specific
oxygenation properties, size, and enhanced nitrite reduc-
tase activity of the Hb derivatives. The desirable physi-
ological characteristics these materials invoke suggests
that a compromise between the P50, n50, and nitrite
reductase activity of a Hb derivative can aid the develop-
ment of safe and effective HBOCs for use in transfusion
therapy.

980 TRANSFUSION Volume 52, May 2012


ACKNOWLEDGMENTS
Modified Hbs were prepared by FEL in RK’s laboratory at the
University of Toronto. Noninvasive hemodynamic SBP measure-
ments were carried out by FEL and BY, and invasive hemody-
namic SBP measurements were conducted by DMB in WMZ’s
laboratory at Massachusetts General Hospital. FEL and DMB
wrote the paper together with revisions from WMZ and RK.
CONFLICT OF INTEREST
There are no conflicts of interest associated with the content of
the paper and the authors.
REFERENCES
1. Mozzarelli A, Ronda L, Faggiano S, Bettati S, Bruno S.
Haemoglobin-based oxygen carriers: research and reality
towards an alternative to blood transfusions. Blood Trans-
fus 2010;8(Suppl 3):s59-68.
2. Kluger R. Red cell substitutes from hemoglobin—do we
start all over again? Curr Opin Chem Biol 2010;14:538-43.
3. Buehler PW, D’Agnillo F, Schaer DJ. Hemoglobin-based
oxygen carriers: from mechanisms of toxicity and clear-
ance to rational drug design. Trends Mol Med 2010;16:447-
57.
4. Silverman TA, Weiskopf RB. Hemoglobin-based oxygen
carriers: current status and future directions. Transfusion
2009;49:2495-515.
5. Natanson C, Kern SJ, Lurie P, Banks SM, Wolfe SM. Cell-
free hemoglobin-based blood substitutes and risk of myo-
cardial infarction and death: a meta-analysis. JAMA 2008;
299:2304-12.
6. Bian K, Murad F. Nitric oxide (NO)–biogeneration, regula-
tion, and relevance to human diseases. Front Biosci 2003;8:
d264-78.
7. Cabrales P, Tsai AG. Will fixing the vasoactivity caused by
hemoglobin-based oxygen carriers be enough? Crit Care
Med 2009;37:2125-6.
8. Marsh N, Marsh A. A short history of nitroglycerine and
nitric oxide in pharmacology and physiology. Clin Exp
Pharmacol Physiol 2000;27:313-9.
9. Kim-Shapiro DB, Schechter AN, Gladwin MT. Unraveling
the reactions of nitric oxide, nitrite, and hemoglobin in
physiology and therapeutics. Arterioscler Thromb Vasc Biol
2006;26:697-705.
10. Gladwin MT, Raat NJ, Shiva S, Dezfulian C, Hogg N, Kim-
Shapiro DB, Patel RP. Nitrite as a vascular endocrine nitric
oxide reservoir that contributes to hypoxic signaling, cyto-
protection, and vasodilation. Am J Physiol Heart Circ
Physiol 2006;291:H2026-35.
11. Vandegriff KD, Malavalli A, Wooldridge J, Lohman J,
Winslow RM. MP4, a new nonvasoactive PEG-Hb conju-
gate. Transfusion 2003;43:509-16.
12. Lui FE, Dong P, Kluger R. Polyethylene glycol conjugation
PEGYLATED BIS-TETRAMERS OF Hb IN CIRCULATION
enhances the nitrite reductase activity of native and cross-
linked hemoglobin. Biochemistry 2008;47:10773-80.
13. Vandegriff KD, Winslow RM. Hemospan: design principles
for a new class of oxygen therapeutic. Artif Organs 2009;33:
133-8.
14. Jahr JS, Moallempour M, Lim JC. HBOC-201, hemoglobin
glutamer-250 (bovine), Hemopure (Biopure Corporation).
Expert Opin Biol Ther 2008;8:1425-33.
15. Manjula BN, Tsai AG, Intaglietta M, Tsai CH, Ho C, Smith
PK, Perumalsamy K, Kanika ND, Friedman JM, Acharya SA.
Conjugation of multiple copies of polyethylene glycol to
hemoglobin facilitated through thiolation: influence on
hemoglobin structure and function. Protein J 2005;24:133-
46.
16. Gourianov N, Kluger R. Conjoined hemoglobins. Loss of
cooperativity and protein-protein interactions. Biochemis-
try 2005;44:14989-99.
17. Hu D, Kluger R. Efficient generation of dendritic arrays of
cross-linked hemoglobin: symmetry and redundancy. Org
Biomol Chem 2008;6:151-6.
18. Hu D, Kluger R. Functional cross-linked hemoglobin bis-
tetramers: geometry and cooperativity. Biochemistry 2008;
47:12551-61.
19. Lui FE, Kluger R. Enhancing nitrite reductase activity of
modified hemoglobin: bis-tetramers and their PEGylated
derivatives. Biochemistry 2009;48:11912-9.
20. Yu B, Shahid M, Egorina EM, Sovershaev MA, Raher MJ, Lei
C, Wu MX, Bloch KD, Zapol WM. Endothelial dysfunction
enhances vasoconstriction due to scavenging of nitric
oxide by a hemoglobin-based oxygen carrier. Anesthesiol-
ogy 2010;112:586-94.
21. Ding H, Howarth AG, Pannirselvam M, Anderson TJ, Sever-
son DL, Wiehler WB, Triggle CR, Tuana BS. Endothelial
dysfunction in Type 2 diabetes correlates with deregulated
expression of the tail-anchored membrane protein SLMAP.
Am J Physiol Heart Circ Physiol 2005;289:H206-11.
22. Esper RJ, Nordaby RA, Vilariño JO, Paragano A, Cacharrón
JL, Machado RA. Endothelial dysfunction: a comprehen-
sive appraisal. Cardiovasc Diabetol 2006;5:4.
23. Yu B, Raher MJ, Volpato GP, Bloch KD, Ichinose F, Zapol
WM. Inhaled nitric oxide enables artificial blood transfu-
sion without hypertension. Circulation 2008;117:1982-90.
24. Buys ES, Cauwels A, Raher MJ, Passeri JJ, Hobai I, Cawley
SM, Rauwerdink KM, Thibault H, Sips PY, Thoonen R,
Scherrer-Crosbie M, Ichinose F, Brouckaert P, Bloch KD.
sGC(alpha)1(beta)1 attenuates cardiac dysfunction and
mortality in murine inflammatory shock models. Am J
Physiol Heart Circ Physiol 2009;297:H654-63.
25. Hu T, Li D, Manjula BN, Acharya SA. Autoxidation of the
site-specifically PEGylated hemoglobins: role of the PEG
chains and the sites of PEGylation in the autoxidation. Bio-
chemistry 2008;47:10981-90.
26. Donadee C, Raat NJ, Kanias T, Tejero J, Lee JS, Kelley EE,
Zhao X, Liu C, Reynolds H, Azarov I, Frizzell S, Meyer EM,
Donnenberg AD, Qu L, Triulzi D, Kim-Shapiro DB,
Volume 52, May 2012 TRANSFUSION 981
LUI ET AL.
Gladwin MT. Nitric oxide scavenging by red blood cell
microparticles and cell-free hemoglobin as a mechanism
for the red cell storage lesion. Circulation 2011;124:
465-76.
27. Voet DJ, Voet JG. Biochemistry. 3rd ed. New York: Wiley;
2004.
28. Lowe KC. Engineering blood: synthetic substitutes from
fluorinated compounds. Tissue Eng 2003;9:389-99.
29. Schumacher MA, Dixon MM, Kluger R, Jones RT, Brennan
RG. Allosteric transition intermediates modelled by
crosslinked haemoglobins. Nature 1995;375:84-7.
30. Minneci PC, Deans KJ, Shiva S, Zhi H, Banks SM, Kern S,
Natanson C, Solomon SB, Gladwin MT. Nitrite reductase
activity of hemoglobin as a systemic nitric oxide generator
982 TRANSFUSION Volume 52, May 2012
mechanism to detoxify plasma hemoglobin produced
during hemolysis. Am J Physiol Heart Circ Physiol 2008;
295:H743-54.
31. Huang Z, Shiva S, Kim-Shapiro DB, Patel RP, Ringwood LA,
Irby CE, Huang KT, Ho C, Hogg N, Schechter AN, Gladwin
MT. Enzymatic function of hemoglobin as a nitrite reduc-
tase that produces NO under allosteric control. J Clin
Invest 2005;115:2099-107.
32. Basu S, Grubina R, Huang J, Conradie J, Huang Z, Jeffers A,
Jiang A, He X, Azarov I, Seibert R, Mehta A, Patel R, King
SB, Hogg N, Ghosh A, Gladwin MT, Kim-Shapiro DB. Cata-
lytic generation of N2O3 by the concerted nitrite reductase
and anhydrase activity of hemoglobin. Nat Chem Biol
2007;3:785-94.