7.5
53
ARTIFICIAL RED BLOOD CELL SUBSTITUTES
Wilhelm, C. R., Ristich, J., Knepper, L. E., Holubkov, R., Wisniewski,
S. R., Kormos, R. L., and Wagner, W. R. (1999). Measurement of
hemostatic indices in conjunction with transcranial doppler sonography in ventricular assist device patients. Stroke 30: 2554–2561.
HEMOGLOBIN IN RBC
2-3-DPG
2-3-DPG
PG
7.5 ARTIFICIAL RED BLOOD CELL
SUBSTITUTES
PG
-D
2-3
D
-3-
2
RBC membrane
2-3-DPG
Thomas Ming Swi Chang
INTRODUCTION
G
DP
-
2-3
The potential presence of HIV in donor blood in the 1980s
was the main stimulus leading to serious development of the
idea of modified hemoglobin started many years ago (Chang,
1957, 1964, 1965; Bunn and Jandl, 1968). Much progress
has been made in the past 10 years leading to the final phase
of phase III clinical trials before routine clinical application.
The present modified hemoglobin can replace the plasma component and red blood cell component of blood. However, it
cannot replace the platelets or the leukocytes of blood. During
this period, much experience has been gained and much has
been learned (Chang, 1997a). This chapter is a brief overview
of this very large area. The readers are referred to much detailed
description in the references (Chang, 1997a, 2000, 2002),
which is updated on the Web site (www.artcell.mcgill.ca).
2-3-DPG
Artificial membrane
for artificial rbc
FIG. 1. Inside the red blood cell, hemoglobin stays as a tetramer.
The red blood cell membrane retains 2,3DPG in the cell to bind to
hemoglobin. This allows hemoglobin to release oxygen as needed by
the tissues with P50 of about 28 mm Hg. In 1957 Chang prepared
artificial red blood cells by removing the red blood cell membrane
and replacing it with artificial membrane. Reprinted with permission from Chang, T. M. S. (1992). Biomaterials, Artificial Cells
and Immobilization Biotechnology 20: 154–174, courtesy of Marcel
Dekker, Inc.
MODIFIED HEMOGLOBIN
Polyhemoglobin
HEMOGLOBIN
Hemoglobin is the protein in red blood cells that is responsible for the transport of oxygen from the lung to the other
tissues. When hemoglobin is extracted from red blood cells, it
can be ultrafiltered and sterilized to remove and inactivate bacteria, viruses and other microorganisms. Thus unlike red blood
cells, potential problems related to HIV, hepatitis viruses, parasites, and others could be eliminated. However, hemoglobin
extracted from red blood cells cannot be used as a blood substitute for the following reasons. Each hemoglobin molecule is a
tetramer consisting of four subunits—two alpha subunits and
two beta subunits (Perutz, 1980) (Fig. 1). There is a requirement for 2,3-diphosphoglycerate (2,3-DPG) to be present in
the 2,3-DPG pocket of the hemoglobin molecule. This is to
decrease the affinity of hemoglobin for oxygen, thus allowing
it to release oxygen properly. When infused into the body, there
is little or no 2,3-DPG in the plasma to bind to the 2,3-DPG
pocket. This results in hemoglobin with high affinity for oxygen, preventing oxygen from being easily released as required
to the tissue. What is even more of a problem is that when
free in the circulating blood, each hemoglobin molecule breaks
down into potentially toxic half-molecules, dimers. The dimers
are highly toxic especially their renal toxicity(Savitsky et al.,
1978). There are also other problems related to hemoglobin in
free solution. The challenge is therefore to modify the basic
biomaterial, hemoglobin, before it can be used as a blood
substitute.
[15:40 2/3/03 ch-07.tex]
The first modified hemoglobin was formed by removing
the biological cell membranes of red blood cells and replacing them with artificial membranes (Fig. 1) (Chang, 1957,
1964, 1965). These early artificial red blood cells can readily bind and release oxygen since 2,3-DPG is retained inside
the artificial cells (Chang, 1957). However, although they are
about the same size as red blood cells, they are not sufficiently flexible to allow them to survive for a long time after
infusion. A number of approaches were used to change the
surface properties of these artificial red blood cells (Chang,
1964, 1965, 1972). This includes surface modification to result
in changes in surface charge, surface composition and membrane composition. Since hemoglobin has many surface and
internal amino groups, Chang (1964, 1965, 1972) studied
the use of a bifunctional agent, sebacyl chloride, to cross-link
hemoglobin molecules into a cross-linked hemoglobin artificial
red blood cell membrane. By decreasing the diameter, all the
hemoglobin in the artificial red blood cells could be cross-linked
into polyhemoglobin (Fig. 2). The reaction is as follows:
Cl–CO–(CH2 )8 –CO–Cl + HB–NH2
Sebacyl chloride
Hemoglobin
= HB–NH–CO–(CH2 )8 –CO–NH–HB
Cross-linked hemoglobin
In 1971, Chang reported the use of another bifunctional agent,
glutaraldehyde, to cross-link hemoglobin and a red blood cell
RATNER: Biomaterials Science
Page: 53
54
7
APPLICATION OF MATERIALS IN MEDICINE AND DENTISTRY
Sebacyl chloride (Chang, 1964)
Glutaraldehye (Chang, 1971)
Pyridoxalation (Benesch, 1975)
PLP
PLP
PLP
CROSSLINKED HB: POLYHEMOGLOBIN
PLP
FIG. 2. Cross-linking the hemoglobin molecules together presents
their breakdown into dimers. Following the use of sebacyl chloride
(Chang, 1964, 1965) the same author reported (Chang, 1971) the
use of glutaraldehyde as another cross-linker. Since then, many other
investigators have extended and improved this, resulting in glutaraldehyde cross-linked polyhemoglobin, now in phase III clinical trial.
Benesch uses pyrodoxal phosphate to replace the 2-3-DPG required
for human polyhemoglobin. Reprinted with permission from Chang,
T. M. S. (1992). Biomaterials, Artificial Cells and Immobilization
Biotechnology 20: 154–174, courtesy of Marcel Dekker, Inc.
enzyme, catalase, into soluble polyhemoglobin (Fig. 2). The
reaction is as follows:
its present refined state. At present, Gould’s group has completed phase II clinical trial and showed in control studies that
this is effective in replacing blood loss in surgical patients.
They are now in phase III clinical trial in trauma surgery using
replacement of up to 10,000 ml which is equal to twice the
total blood volume of a patient (Gould et al., 1998). This
approach has also been developed extensively by the Biopure
group (Pearce and Gawryl, 1998) using bovine hemoglobin
to form bovine polyhemoglobin. Since hemoglobin from cows
does not need 2,3-DPG to decrease the affinity for oxygen, pyridoxal phosphate is not required. They have completed their
phase II clinical trial in patients and have started phase III clinical trial using as much as 11,000 ml in individual patients’.
Biopure’s veterinary product has been approved by the FDA
for routine use in canine veterinary medicine. Clinical trials on
the Biopure Polyhemoglobin (Sprung et al., 2002; LaMuraglia
et al., 2000) have led to its being approved for routine clinical
uses in South Africa (Lok, 2001) and awaiting approval in the
United States. Hsia (1991) originated a dialdehyde prepared
from open ring sugars to form polyhemoglobin. This crosslinker also modified the 2,3-DPG pocket so that the resulting
polyhemoglobin does not require 2,3-DPG or pyridoxal phosphate. This has been developed and extended by Hemosol
(Adamson and Moore, 1998), which has completed phase II
clinical trial and is proceeding to phase III clinical trial.
H–CO–(CH2 )3 –CO–H + HB–NH2
Glutaraldehyde
Other Types of Soluble Modified Hemoglobin
Hemoglobin
= HB–NH–CO–(CH2 )3 –CO–NH–HB
Cross-linked hemoglobin
Cross-linking hemoglobin into soluble hemoglobin prevents
the breakdown of hemoglobin into dimers. Furthermore, when
infusing modified hemoglobin, their colloid osmotic pressure
should normally be about the same as that of plasma proteins. This means that only about 7 g/dl of hemoglobin can be
infused. Now, colloid osmotic pressure depends on the number of solute particles more than on the total amount of protein
in solution. Polyhemoglobin is formed from the cross-linking
of an average of four or five hemoglobin molecules together
into a soluble macromolecular complex. This way, 15 g/dl of
polyhemoglobin would have about the same colloid osmotic
pressure as 7 g/dl of hemoglobin (Chang, 1997a). This means
that a much larger amount of hemoglobin can be used when
in the form of polyhemoglobin. However, there is no 2,3-DPG
in the polyhemoglobin to allow them to have lower affinity
for oxygen so that they can release oxygen to the tissue as
required. Benesch et al. in 1975 made the important finding
that the addition of a 2,3DPG analog, pyridoxal phosphate,
to hemoglobin can also decrease oxygen affinity (Fig. 2). This
is therefore added to cross-linked hemoglobin hemoglobin.
The 1971 glutaraldehyde approach of Chang has been combined with Benesch’s 1975 use of pyridoxal phosphate and
developed into pyridoxalated human polyhemoglobin blood
substitutes by Dudziak and Bonhard in 1976; Moss’ group
(Sehgal et al.) in 1980; DeVenuto and Zegna in 1982; and
Chang’s group (Keipert et al.) in 1982. The Northfield group
of Gould et al. (1998) has developed this very extensively to
[15:40 2/3/03 ch-07.tex]
Bunn and Jandl (1968) cross-linked hemoglobin intramolecularly and found that this prevents the loss of hemoglobin
through the kidney and serves to decrease renal toxicity.
Walder et al. in 1979 reported the use of a 2,3-DPG
pocket modifier, bis (3,5-dibromosalicyl) fumarate (DBBF)
to intramolecularly cross-link the two α subunits of the
hemoglobin molecule forming tetrameric hemoglobin. This
prevents dimer formation and improves P50 . P50 is the ease
with which oxyhemoglobin releases its oxygen and is measured by the oxygen tension at which 50% of the oxygen is
released from oxyhemoglobin. Unlike polyhemoglobin, crosslinked tetrameric hemoglobin is a single hemoglobin molecule.
Although its molecular weight is comparable to that of albumin, hemoglobin has much less surface negative charge. As a
result, it can cross the intercellular junction of the endothelial
cells lining the vasculature much more readily (Fig. 3). This
may have resulted in the removal of nitric oxide. Nitric oxide
decreases the contraction of smooth muscles including those
in vasculature. This may explain the increase in esophageal
and intestinal contraction and increase in blood pressure when
using these types of cross-linked tetrameric hemoglobin. Baxter
has extensively developed this (Nelson, 1998), but the results of
their clinical trials have led them to discontinue this approach
and to look at another approach.
Another very exciting area is the use of recombinant
technology to produce recombinant human hemoglobin from
Escherichia coli (Hoffman et al., 1990). This results in the
fusing of the alpha subunits of the hemoglobin molecule and
preventing the breakdown of hemoglobin into dimers. Furthermore, this also resulted in the modification of the 2,3-DPG
RATNER: Biomaterials Science
Page: 54
7.5
ARTIFICIAL RED BLOOD CELL SUBSTITUTES
CONJUGATED HB
POLY HB
endothelial cell
VASCULAR WALL
NITRIC
OXIDE (NO)
NO
TETRAMERIC HB
NO
NO
VASOCONSTRICTION
GI & OTHER EFFECTS
NO
ENCAP HB
NITRIC
OXIDE (NO)
FIG. 3. Different types of modified hemoglobin (Hb). Tetrameric Hb
being smaller can move across the intercellular junction of the endothelial cells. This Hb binds nitric oxide (NO) in the interstitial space and
lowers the NO concentration. This results in vasoconstriction. Other
types of modified hemoglobin contain a varying amount of tetrameric
hemoglobin that can also act similarly. Removal of these smaller
tetrameric hemoglobin from polyhemoglobin, conjugated hemoglobin
and microencapsulated hemoglobin would prevent this. Reprinted with
permission from Chang, T. M. S. (1997). Artificial Cells, Blood Substitutes and Immobilization Biotechnology 25: 1–24. Courtesy of Marcel
Dekker, Inc.
pocket in such a way that there is no requirement for 2,3DPG or pyridoxal phosphate. This has been developed by
Somatogen and used in a phase II clinical trial (Freytag and
Templeton, 1997). Being a single tetrameric hemoglobin, it can
also cause the same effect on smooth muscle as described above
for cross-linked tetrameric hemoglobin. They have now prepared a new generation of recombinant human hemoglobin
in which they have blocked the binding site for nitric oxide
(Doherty et al., 1998). The resulting recombinant human
hemoglobin no longer causes vasoconstriction. This is being
developed by Baxter toward a clinical trial.
Chang (1964, 1965, 1972) has used interfacial polymerization to form a conjugated membrane consisting of polyamides
and hemoglobin around hemoglobin. By decreasing the diameter of the conjugated membrane, all the hemoglobin can
be cross-linked with polymer into conjugated hemoglobin.
Wong (1988) cross-linked hemoglobin to dextran to form
soluble conjugated hemoglobin. A PEG-conjugated bovine
hemoglobin is now in phase II clinical trial (Shorr et al., 1996).
A conjugated human hemoglobin (Iwashita et al., 1996) is also
in phase II clinical trial.
Molecular Weight Distribution
Molecular weight (MW) distribution varies widely among
the different types of cross-linked hemoglobin blood substitutes
(Chang, 1997a). For intramolecularly cross-linked hemoglobin
or recombinant hemoglobin, they are in the tetrameric form
with a molecular weight of about 68,000. In polyhemoglobin
the molecular weight distribution can vary widely. In polyhemoglobin, the molecular weights follow a distribution curve
from tetrameric to much larger molecular weight. Some groups
use a larger mean molecular weight. Other groups prefer
[15:40 2/3/03 ch-07.tex]
55
using a smaller mean molecular weight. Those with lower
mean molecular weight also have more tetrameric hemoglobin.
Those with higher mean molecular weight have less tetrameric
hemoglobin because of the higher degree of polymerization.
The pyridoxalated polyhemoglobin prepared from our laboratory procedure can be used as an example. The molecular
weight distribution can be varied at will by changing the reaction time, ratio of glutaraldehyde to hemoglobin, and other
reaction parameters. To measure the MW distribution, pyridoxalated polyhemoglobin (PP-PolyHb) was run on Sephadex
G-200 (1.6 × 70 cm column) in 0.1 M Tris-HC1 (pH 7.5) at
12 ml/hr. The fractions of polymerized hemoglobin were in the
molecular weight ranges of 750,000 (10%); 470,000 (23%);
260,000 (39%); 130,000 (21%); and 66,000 (7%). Thus, 60%
falls in the 130,000–350,000 range and 33% ranges in molecular weight from 350,000 to 750,000. Only 7% was in the
tetrameric form. Gould’s group (1998) has emphasized the
need to remove as much tetrameric hemoglobin as possible.
After their preparation of polyhemoglobin, they carried out
another step to remove much of the tetrameric hemoglobin
resulting in a preparation with less than 1% hemoglobin
in the 68,000 MW range. Biopure has also removed much
of the tetrameric hemoglobin resulting in preparation with
less than 5% 68,000 MW hemoglobin (Pearce and Gawryl,
1998). These two preparations do not cause vasoconstriction
or smooth muscle contraction because insignificant amounts of
tetrameric hemoglobin crossing the endodthelial cell junctions
(Fig. 3).
In Vitro Biocompatibility Screening Test Using
Human Plasma/Blood
Detailed in vitro and in vivo tests are required by the
FDA before approval of using the blood substitutes for clinical trials (Fratantoni, 1991). However, even when the blood
substitute is shown safe in animals, the response in animals
is not always the same as for humans. This is especially
true in tests for hypersensitivity, complement activation, and
immunology. How do we bridge the gap between animal
safety studies and use in humans? Complement activation is
important in many adverse reactions of humans to modified
hemoglobin (Chang and Lister, 1990, 1993a,b, 1994). Modified hemoglobin may be contaminated with trace amounts
of blood-group antigen that can form antigen–antibody complexes. Other materials can potentially cause complement
activation. These include endotoxin, microorganisms, insoluble immune complexes, chemicals, polymers, organic solvents,
and others.
We have devised an in vitro test tube screening tests using
human plasma or blood (Chang and Lister, 1990, 1993a, b,
1994). The use of human plasma or blood gives the closest
response to human testing next to actual injection. If we want
to be more specific, we can use the plasma of the same patient
who is to receive the blood substitute. Many components of
human blood or plasma can be used for this in vitro screening test. If one were to select only one test, perhaps the most
useful one would be the effect of modified hemoglobin on complement activation (C3a) when added to human plasma or
blood. This simple test consists of adding 0.1 ml of modified
RATNER: Biomaterials Science
Page: 55
56
7
APPLICATION OF MATERIALS IN MEDICINE AND DENTISTRY
TABLE 1 Screening Test Based on Complement Activation
(Chang and Lister, 1990, 1993a, b, 1994)
1 Immediately before use, the plasma sample is thawed.
2 400 µl of the plasma is pipetted into 4-ml sterile polypropylene
tubes.
3 100 µl of pyrogen-free saline (or Ringer’s lactate) for injection
is added to the 400 µl of human plasma as control; 100 µl of
hemoglobin or modified hemoglobin is added to each of the other
tubes containing 400 µl of human plasma.
4 The reaction mixtures are incubated at 37◦ C at 60 RPM for 1 hour
in a Lab-Line Orbit Environ Shaker (Fisher Scientific, Montreal,
Canada).
5 After 1 hour the reaction is quenched by adding 0.4 ml of this
solution into a 2 ml EDTA sterile tube containing 1.6 ml of sterile
saline.
6 The samples are immediately stored at −70◦ C until analysis.
7 The analytical kit for human complement C3a is purchased from
Amersham Canada. The method of analysis is the same as the
instructions in the kit with two minor modifications. Centrifugation
is carried out at 10,000 g for 20 min. After the final step of inversion,
the inside wall of the tubes is carefully blotted with Q-Tips.
hemoglobin to a test tube containing 0.4 ml of human plasma
or blood. Then, the plasma is analyzed for complement activation after incubation for 1 hour. Blood preparation is described
in Table 1.
Blood is obtained by clean venous puncture from human
volunteers into 50-ml polypropylene (Sastedt) heparinized
tubes (10 IU heparin/ml of blood). Immediately separate
plasma by centrifugation at 5500 G at 2◦ C for 20 min and
freeze the plasma in separate portions at −70◦ C. Do not use
serum because coagulation initiates complement activation.
EDTA should not be used as an anticoagulant, because it
interferes with complement activation.
The exact procedure and precise timing described in Table 1
are important in obtaining reproducible results. The baseline
control level of C3a complement activation will vary with
the source and procedure for obtaining the plasma. Therefore, a control baseline level must be used for each analysis.
Furthermore, all control and test studies should be carried
out in triplicate. Much practice is needed to establish this
procedure when used for the first time. Reproducibility must
be established before this test can be used.
Instead of using plasma and the need to withdraw blood
with a syringe, a simpler procedure involves obtaining blood
from finger pricks (Chang and Lister, 1993a). Sterile methods
are used to prick a finger. Blood is collected in heparinized
microhematocrit tubes. The tubes are kept at 4◦ C, then used
immediately. Each blood sample used in testing contains 80 µl
of whole blood and 20 µl of saline. Each test solution is
added to a blood sample. Test solutions include saline (negative control); Zymosan (positive control); or hemoglobin. This
is incubated at 37◦ C at 60 rpm. After 1 hour of incubation,
EDTA solution is added to the sample to stop the reaction.
The analysis for C3a is then carried out as described. The test
kit is based on the ELISA C3a enzyme immunoassay (Quidel
Co., San Diego).
[15:40 2/3/03 ch-07.tex]
In research, development, or industrial production of blood
substitute, different chemicals, reagents, and organic solvents
are used. This includes cross-linkers, lipids, solvents, chemicals,
polymers, and other materials. Some of these can potentially
result in complement activation and other reactions in humans.
Other potential sources of problems include trace contaminants from ultrafilters, dialyzers, and chromatography. The
test described in Table 1 is a simple in vitro test based on
using human plasma or blood. For example, this screening test
can detect several problems related to potential hypersensitivity
reactions, and anaphylactic reactions, effects due to antibody–
antigen complexes, and others. These potential problems may
be related to contamination with trace blood group antigens,
polymeric extracts, organic solvents, emulsifiers, and others.
This is useful for preclinical trial studies and for screening
before clinical use. This is also useful for screening batches of
modified hemoglobin blood substitutes for industrial production to rule out potential problems. It is also useful in research
and development. What is very exciting is that this approach
may be the basis of a large-scale in vitro “clinical trial” in a
large number of patients without infusing the product. By doing
this, we can analyze the percentage of patients who may have
adverse reactions without having to introduce the product
into patients. Of course, ultimately, human clinical trials are
required.
Second-Generation Modified Hemoglobin
Blood Substitutes
The present, first-generation blood substitutes described
above are effective for short-term uses especially in elective surgery (Winslow, 1996; Chang, 1997a). On the other
hand, these first-generation blood substitutes can stay in the
body’s circulation only for about 12–24 hours (a typical red
blood cell lasts about 100 days). Thus, their present role
is restricted to short-term applications. For example, substitutes are being tested in humans for replacing blood lost
during some cardiac, cancer, orthopedic, and trauma surgeries. Furthermore, new generations of blood substitutes may
be needed for other conditions such as sustained severe hemorrhagic shock, stroke, or other conditions with sustained
ischemia (low oxygen). When oxygen from blood substitute
is reperfused to ischemic organs or tissues, oxygen radicals
will be produced (Fig. 4). The present generation of blood
substitutes do not have the enzymes needed to protect the
body against oxidants such as oxygen radicals. Unchecked,
oxygen radicals may cause reperfusion injuries and other problems (Alayash et al., 1998). Furthermore, peroxide formed
can break down hemoglobin resulting in further increase in
oxygen radicals (Fenton reaction). Researchers are studying
ways to solve this problem, including cross-linking the required
enzymes to the hemoglobin or further modifying the molecular
structure of the hemoglobin. For example, we have reported
the cross-linking of superoxide dismutase (SOD) and catalase to hemoglobin to result in polyhemoglobin–catalase–SOD
(Fig. 4) (D’Agnillo and Chang, 1998a,b; Chang et al., 1998).
In a rat intestinal ischemia-reperfusion model, we showed that
polyhemoglobin–catalase-SOD can markedly decrease the level
RATNER: Biomaterials Science
Page: 56
7.5
MEMBRANE MATERIAL IN 100 ml SUSPENSION
7
OXYGEN REPERFUSION
Xanthine
Oxidase
6
POLYHB-SOD-CAT
g/dl
ISCHEMIA
increased
Hypoxanthine
SUPEROXIDE
SOD
OH+
OXYGEN
RADICALS
5
lipid
4
biodegradable
polymer
3
2
H2O2
Catalase
1
TISSUE INJURY
0
H2O
LIPIDVES1
FIG. 4. Ischemic reperfusion injuries. Ischemia leads to accumulation of hypoxanthine and activation of xanthine oxidase. Reperfusion bringing oxygen results in superoxide formation. This and
other resulting oxidants and oxygen radicals can cause tissue injury.
First-generation modified hemoglobin is prepared from ultrapure
hemoglobin and therefore does not contain the required deoxidant
enzymes. We showed that cross-linking polyhemoglobin with superoxide dismutase and catalase can significantly decrease the formation
of oxygen radicals (D’Agnillo and Chang, 1998). Reprinted with
permission from Chang, T. M. S. (1997). Artificial Cells, Blood Substitutes and Immobilization Biotechnology 25: 1–24. Courtesy of Marcel
Dekker, Inc.
of oxygen radicals when using polyhemoglobin with the crosslinked enzymes (Razack et al., 1997). Unlike polyhemoglobin,
cross-linked polyhemoglobin–superoxide dismutase–catalase
supplies oxygen without causing blood–brain barrier disruption or brain edema in a rat model of transient global brain
ischemia–reperfusion (Powanda and Chang, 2002).
Third-Generation Blood Substitutes
Second-generation substitutes, like the first-generation
blood substitutes, only circulate with a half-time of 12–24
hours. They are therefore more suitable for shorter term applications. Furthermore, the hemoglobin molecules will not be
protected by a red blood cell membrane and highly purified hemoglobin is needed. Thus, researchers are working on
more complicated, third-generation blood substitutes that will
encapsulate hemoglobin and the required enzymes inside artificial red blood cells. Encapsulated hemoglobin is the first
modified hemoglobin studied (Fig. 1) (Chang, 1957, 1964,
1972), but being more complicated, it has only been seriously developed more recently. One method is to encapsulate
hemoglobin inside lipid vesicles about 0.2 µm in diameter
(Djordjevich and Miller, 1980; Rudolph et al., 1997; Tsuchida,
1995). This technique also increases the circulation time.
A more recent approach is to use nanotechnology methods
to encapsulate hemoglobin and enzymes inside biodegradable polylactic acid membrane nanocapsules some 0.15 µm
in diameter (Yu and Chang, 1996; Chang and Yu, 1997,
1998) (Fig. 5). Nano-dimension artificial red blood cells with
ulthrathin membranes of PEG-PLA [poly(ethylene glycol)–
polylactide] copolymer stay in the circulation twice as long as
polyhemoglobin (Chang and Yu, 2001).
[15:40 2/3/03 ch-07.tex]
57
ARTIFICIAL RED BLOOD CELL SUBSTITUTES
LIPIDVES2
NANOCAPS
FIG. 5. Amount of membrane materials in each 100-ml suspension.
Hemoglobin lipid vesicles(LIPIDVES) compared to that of biodegradable polymeric hemoglobin nanocapsules (NANOCAPS). Reprinted
with permission from Chang, T. M. S. (1997). Artificial Cells, Blood
Substitutes and Immobilization Biotechnology 25: 1–24. Courtesy of
Marcel Dekker, Inc.
PERFLUOROCHEMICALS
Introduction
Silicone and fluorocarbon are known for their ability to
carry oxygen. Thus, Clark and Gollan (1966) showed that
mice immersed in oxygenated silicone oil or liquid fluorocarbon could breathe in the liquid. In the same year, Chang
(1966) showed that artificial cells formed from a hybrid of
silicone rubber and hemolysate were very efficient in carrying
and releasing oxygen. However, these solid elastic siliconerubber artificial cells were removed rapidly from the circulation. Sloviter and Kamimoto (1967) showed that perfusion
using finely emulsified fluorocarbon could maintain rat brain
function for several hours. Geyer et al. (1968) showed that
finely emulsified fluorocarbon could replace essentially all the
blood of rats with the rats surviving and recovering. This
was extended and developed in Japan to produce Fluosol-DA
20 suitable for clinical testing (Naito and Yokoyama, 1976;
Mitsuno and Naito, 1979).
Earlier Perfluorocarbons: Fluosol-DA
Fluosol-DA is a 20% (w/v) mixture of seven parts of
perfluorodecalin and three parts perfluorotripropylamine, with
2.7% Pluronic F-68 as an emulsifier and 0.4% egg yolk phospholipids to form a membrane coating on the emulsion. The
average particle size of the emulsion is 0.118 µm. Because of
the viscosity of the fluorocarbon emulsion at high concentrations, the maximum amount used is only 20%. Patients have
to breathe 70% to 90% oxygen for this amount of fluosolDA to carry enough oxygen. The amount of oxygen dissolved
in fluorocarbon determines the amount of available oxygen.
When oxygen is increased to 70–90% from the atmospheric
oxygen of 20%, more oxygen is dissolved in the fluorocarbon.
CO2 also dissolves in the fluorocarbon, which carries it to the
lung for elimination.
Mitsuno and Ohyanagi (1985) reviewed the Japanese experience with 401 patients who received Fluosol-DA (20%).
RATNER: Biomaterials Science
Page: 57
58
7
APPLICATION OF MATERIALS IN MEDICINE AND DENTISTRY
Complement activation was the major clinical problem.
Adverse effects were observed in some patients due to complement activation caused by the Pluronic surfactant used in
Fluosol. The FDA approved the clinical use of Fluosol-DA only
for coronary artery balloon angioplasties in 1989.
Modern Perfluorochemicals
Modern perfluorochemicals been discussed in detail (Riess,
1998; Keipert, 1998). Two new types of preparations have
been developed. One type is based on perfluoroctyl bromide
(C8 F17 Br) and perfluorodichoroctane (C8 F16 Cl2 ). Both types
allow the use of higher concentrations of PFC. Oxygent from
Alliance Pharmaceutical Corp., San Diego, is prepared from
perfluoroctyl bromide (C8 F17 Br) with egg yolk lecithin as the
surfactant. The use of egg yolk lecithin instead of Pluronic
surfactant has solved the problem of complement activation.
Another approach, Oxyfluor from HemoGen, St. Louis, is
based on the use of perfluorodichloroctane (C8 F16 Cl2 ) with
triglyceride and egg yolk lecithin.
Safety clinical trials using Oxygent shows that doses up
to 1.8 g PFC/kg could be given without side effects (Keipert,
1998). Further increase in dosages may result in transient mild
febrile response (38–39◦ C), a transient decrease in platelets.
Oxygent is being used in phase II clinical trials in surgical
patients. The use of 0.9 g/kg of Oxygent can avoid the need for
2 units of blood. This is also being combined with autologous
blood predeposition in which about 1000 ml of each patient’s
blood is removed and stored before surgery. During surgery,
Oxygent is given as needed. After surgery, the patient’s blood
is reinfused. There are several other potential applications for
perfluorochemicals. One of the most important potential use
is in patients who because of religious beliefs cannot use red
blood cells or hemoglobin from animal sources.
GENERAL DISCUSSION
The basic ideas of cross-linked hemoglobin and encapsulated hemoglobin date back to before the 1960s. Concentrated
efforts to develop blood substitutes for human use only seriously started after 1986 because of public concerns regarding
HIV in donor blood. Unfortunately, a product must undergo
years of research and development followed by clinical trials
before it is ready for use in patients. It will take at least another
one to two years for the first generation blood substitutes to
be available for routine use. Had there been a serious development effort in the 1960s, blood substitutes would have already
been available in 1986. As it is, the public has continued to
be exposed to the potential, though extremely rare, hazard of
HIV in donor blood. It is hoped that investigators will have
the chance to develop new generations of blood substitutes
discussed here long before another urgent need arrives. The
principle of artificial cells has also been developed for other
areas of medical applications including enzyme therapy, cell
therapy, and other applications (Chang, 1997b; Chang and
Prakash, 1998, 2001) (www.artcell.mcgill.ca).
[15:40 2/3/03 ch-07.tex]
Bibliography
Adamson, J. G., and Moore, C. (1998). Hemolink, an o-Raffinose
crosslinked hemoglobin-based oxygen carrier. in Blood Substitutes:
Principles, Methods, Products and Clinical Trials, T. M. S. Chang,
ed., Vol. 2. Basel, Karger, pp. 62–79.
Alayash, A. I., Brockner-Ryan, B. A., McLeod, L. L., Goldman, D. W.,
and Cashon, R. E. (1998). Cell-free hemoglobin and tissue
oxidants: probing the mechanisms of hemoglobin cytotoxicity.
in Blood Substitutes: Principles, Methods, Products and Clinical
Trials, Vol. 2, T. M. S. Chang, ed. Basel, Karger, pp. 157–174.
Benesch, R., Benesch, R. E., Yung, S., and Edalji, R. (1975).
Hemoglobin covalently bridged across the polyphosphate binding
site. Biochem. Biophys. Res. Commun. 63: 1123.
Bunn, H. F., and Jandl, J. H. (1968). The renal handling of hemoglobin.
Trans. Assoc. Am. Physicians 81: 147.
Chang, T. M. S. (1957). Hemoglobin corpuscles. Report of a research,
McGill University, 1–25, 1957. Medical Library, McIntyre Building, McGill University, 1957 (reprinted in 30th anniversary in
artificial red blood cell research, Biomater. Artifi. Cells Artif.
Organs 16: 1–9, 1988).
Chang, T. M. S. (1964). Semipermeable microcapsules. Science
146(3643): 524–525.
Chang, T. M. S. (1965). Ph.D. Thesis. McGill University Medical
Library.
Chang, T. M. S. (1966). Semipermeable aqueous microcapsules (“artificial cells”): with emphasis on experiments in an extracorporeal
shunt system. Trans. Am. Soc. Artif. Int. Organs 12: 13–19.
Chang, T. M. S. (1971). Stabilization of enzyme by microencapsulation with a concentrated protein solution or by crosslinking
with glutaraldehyde. Biochem. Biophys. Res. Commun. 44: 1531–
1533.
Chang, T. M. S. (1972). Artificial Cells. Charles C. Thomas,
Springfield, IL.
Chang, T. M. S. (1997a). Blood Substitutes: Principles, Methods,
Products and Clinical Trials, Vol. I. Karger, Basel.
Chang, T. M. S. (1997b). Artificial cells. in Encyclopedia of Human
Biology, 2nd ed. Academic Press, San Diego, pp. 457–463.
Chang, T. M. S. (2000). Red blood cell substitutes. Best Pract. Res.
Clin. Haematol. 13(4): 651–668.
Chang, T. M. S. (2002). Oxygen carriers. Curr. Opin. Invest. Drugs
3(8): 1187–1190.
Chang, T. M. S., and Lister, C. (1990). A screening test of modified
hemoglobin blood substitute before clinical use in patients—based
on complement activation of human plasma. Biomater. Artif. Cells
Artif. Organs 18(5): 693–702.
Chang, T. M. S., and Lister, C. W. (1993a). Screening test for modified hemoglobin blood substitute before use in human. U.S. Patent
No. 5,200,323, April 6, 1993.
Chang, T. M. S., and Lister, C. W. (1993b). Use of finger-prick human
blood samples as a more convenient way for in vitro screening
of modified hemoglobin blood substitutes for complement activation: a preliminary report. Biomater. Artif. Cells Immobilization
Biotechnol. 21: 685–690, 1993.
Chang, T. M. S., and Lister, C. (1994). Assessment of blood substitutes: II. In vitro complement activation of human plasma and blood
for safety studies in research, development, industrial production
and preclinical analysis. Artif. Cells Blood Subs. Immobilization
Biotechnol. 22: 159–170.
Chang, T. M. S., and Prahash, S. (1998). Microencapsulated genetically engineered cells: comparison with other strategies and recent
progress. Mol. Med. Today 4: 221–227.
Chang, T. M. S., and Prakash, S. (2001). Procedure for microencapsulation of enzymes, cells and genetically engineered microorganisms.
Mol. Biotechnol. 17: 249–260.
RATNER: Biomaterials Science
Page: 58
7.5
ARTIFICIAL RED BLOOD CELL SUBSTITUTES
Chang, T. M. S., and Yu, W. P. (1997). Biodegradable polymer membrane containing hemoglobin for blood substitutes. U.S. Patent
No. 5,670,173, September 23, 1997.
Chang, T. M. S., and Yu, W. P. (1998). Nanoencapsulation of
hemoglobin and red blood cell enzymes based on nanotechnology and biodegradable polymer. in Blood Substitutes: Principles,
Methods, Products and Clinical Trials, Vol. 2, T. M. S. Chang, ed.
Basel, Karger, pp. 216–231.
Chang, T. M. S., and Yu, W. P. (2001). Biodegradable polymeric
nanocapsules and uses thereof. U.S. Provisional Patent Application
No. 60/316,001 (August 31, 2001).
Chang, T. M. S., D’Agnillo, F., and Razack, S. (1998). A second
generation hemoglobin based blood substitute with antioxidant
activities. in Blood Substitutes: Principles, Methods, Products
and Clinical Trials, Vol. 2, T. M. S. Chang, ed. Karger, Basel,
pp. 178–186.
Clark, L. C., Jr., and Gollan, F. (1966). Survival of mammals breathing
organic liquids equilibrated with oxygen at atmospheric pressure.
Science 152: 1755.
D’Agnillo, F., and Chang, T. M. S. (1997). Modified hemoglobin blood
substitute from Cross linked hemoglobin–superoxide dismutase–
catalase. U.S. Patent No. 5,606,025, February, 1997.
D’Agnillo, F., and Chang, T. M. S. (1998a). Polyhemoglobin–
superoxide dismutase–catalase as a blood substitute with antioxidant properties. Nat. Biotechnol. 16(7): 667–671.
D’Agnillo, F., and Chang, T. M. S. (1998b). Absence of hemoproteinassociated free radical events following oxidant challenge of crosslinked hemoglobin–superoxide dismutase–catalase. Free Radical
Biol. Med. 24(6): 906–912.
DeVenuto, F., and Zegna, A. I. (1982). Blood exchange with
pyridoxalated-polymerized hemoglobin. Surg. Gynecol. Obstet.
155: 342–346.
Djordjevich, L., and Miller, I. F. (1980). Synthetic erythrocytes from
lipid encapsulated hemoglobin. Exp. Hematol. 8: 584.
Doherty, D. H., Doyle, M. P., Curry, S. R., Vali, R. J., Fattor, T. J.,
Olson, J. S., and Lemon, D. D. (1998). Rate of reaction with nitric
oxide determines the hypertensive effect of cell-free hemoglobin.
Nat. Biotechnol. 16: 672–676.
Dudziak, R., and Bonhard, K. (1980). The development of hemoglobin
preparations for various indications. Anesthesist 29: 181–187.
Frantantoni, J. C. (1991). Points to consider in the safety evaluation of
hemoglobin based oxygen carriers. Transfusion 31(4): 369–371.
Freytag, J. W., and Templeton, D. (1997). Optro (Recombinant
Human Hemoglobin): a therapeutic for the delivery of oxygen and the restoration of blood volume in the treatment of
acute blood loss in trauma and surgery. in Red Cell Substitutes; Basic Principles and Clinical Application, A. S. Rudolph,
R. Rabinovici, and G. Z. Feuerstein, eds. Marcel Dekker,
New York, pp. 325–334.
Geyer, R. P., Monroe, R. G., and Taylor, K. (1968). Survival of
rats totally perfused with a fluorocarbon-detergent preparation.
in Organ Perfusion and Preservation, J. C. Norman, J. Folkman,
W. G. Hardison, L. E. Rudolf, and F. J. Veith, eds. Appleton
Century Crofts, New York, pp. 85–96.
Gould, S. A., Sehgal, L. R., Sehgal, H. L., DeWoskin, R., and Moss,
G. S. (1998). The clinical development of human polymerized
hemoglobin. in Blood Substitutes: Principles, Methods, Products
and Clinical Trials, Vol. 2, T. M. S. Chang, ed., Vol. 2, Karger,
Basel, pp. 12–28.
Hoffman, S. J., Looker, D. L., and Roehrich, J. M. (1990). Expression
of fully functional tetrameric human hemoglobin in Escherichia
coli. Proc. Natl. Acad. Sci. USA 87: 8521–8525.
Hsia, J. C. (1991). o-Raffinose polymerized hemoglobin as red blood
cell substitute. Biomater. Artif. Cells Immobilization Biotechnol.
19: 402.
[15:40 2/3/03 ch-07.tex]
59
Iwashita, Y., Yabuki, A., Yamaji, K., Iwasaki, K., Okami, T., Hirati, C.,
and Kosaka, K. (1996). A new resuscitation fluid “stabilized
hemoglobin” preparation and characteristics. Biomater. Artif.
Cells Artif. Organs 16: 271–280.
Keipert, P., Minkowitz, J., and Chang, T. M. S. (1982). Cross-linked
stroma-free polyhemoglobin as a potential blood substitute. Int.
J. Artif. Organs 5(6): 383–385.
Keipert, P. E. (1998). Perfluorochemical emulsions: future alternatives
to transfusion, development, pool. in Blood Substitutes: Principles,
Methods, Products and Clinical Trials, Vol. 2, T. M. S. Chang, ed.
Karger/Landes, Basel, Georgetown, Texas, pp. 101–121.
LaMuraglia, G. M., O’Hara, P. J., Baker, W. H., Naslund, T. C.,
Norris, M. W., Li, J., O’Hara, P. J., and Vandemeersch, E. (2000).
The reduction of the allogenic transfusion requirement in aortic
surgery with a hemoglobin-based solution. J. Vasc. Surg. 31(2):
299–308.
Lok, C. (2001). Blood product from cattle wins approval for use in
humans. Nature 410(6831): 855.
Mitsuno, T., and Naito, R. (eds.) (1979). Perfluorochemical Blood
Substitutes. Excerpta Medica, Amsterdam.
Mitsuno, T., and Ohyanagi, H. (1985). Present status of clinical
studies of fluosol-DA (20%) in Japan. in Perfluorochemical Oxygen Transport, K. K. Tremper, ed. Little Brown & Co, Boston,
pp. 169–184.
Naito, R., and Yokoyama, K. (1978). An improved perfluorodecalin emulsion. in Blood Substitutes and Plasma Expanders, G. A.
Jamieson and T. J. Greenwalt, eds. Alan R. Liss, New York, p. 81.
Nelson, D. J. (1998). Blood and HemAssist (DCLHb): potentially a
complementary therapeutic team. in Blood Substitutes: Principles,
Methods, Products and Clinical Trials, Vol. 2, T. M. S. Chang, ed.
Karger, Basel, pp. 39–57.
Pearce, L. B., and Gawryl, M. S. (1998). Overview of preclinical
and clinical efficacy of Biopure’s HBOCs. in Blood Substitutes:
Principles, Methods, Products and Clinical Trials, Vol. 2, T. M. S.
Chang, ed. Karger, Basel, pp. 82–98.
Perutz, M. F. (1980). Stereochemical mechanism of oxygen transport
by hemoglobin. Proc. R. Soc. Lond. B 208: 135–147.
Powanda, D., and Chang, T. M. S. (2002). Cross-linked
polyhemoglobin–superoxide dismutase–catalase supplies oxygen
without causing blood brain barrier disruption or brain edema in
a rat model of transient global brain ischemia-reperfusion. Artif.
Cells Blood Subst. Immob. Biotechnol. 30: 25–42.
Razack, S., D’Agnillo, F., and Chang, T. M. S. (1997). Effects of
polyhemoglobin–catalase–superoxide dismutase on oxygen radicals in an ischemia–reperfusion rat intestinal model. Artif. Cells
Blood Subst. Immobilization Biotechnol. 25: 181–192.
Riess, J. G. (1998). Fluorocarbon-based oxygen-delivery: basic
principles and product development, pool. in Blood Substitutes:
Principles, Methods, Products and Clinical Trials, Vol. 2, T. M. S.
Chang, ed. Karger, Basel, pp. 101–121.
Rudolph, A. S., Rabinovici, R., and Feuerstein, G. Z. (eds.) (1997).
Red Blood Cell Substitutes. Marcel Dekker, New York.
Savitsky, J. P., Doozi, J., Black, J., Arnold, J. D. (1978). A clinical safety trial of stroma free hemoglobin. Clin. Pharm. Ther.
23: 73.
Sehgal, L. R., Rosen, A. L., Gould, S. A., Sehgal, H. L., Dalton, L.,
Mayoral, J., and Moss, G. S. (1980). In vitro and in vivo characteristics of polymerized pyridoxalated hemoglobin solution. Fed.
Proc. 39: 2383.
Shorr, R. G., Viau, A. T., and Abuchowski, A. (1996). Phase 1B safety
evaluation of PEG hemoglobin as an adjuvant to radiation therapy
in human cancer patients. Artif. Cells Blood Subst. Immobilization
Biotechnol. 24(abstracts issue): 407.
Sloviter, H., and Kamimoto, T. (1967). Erythrocyte substitute for
perfusion of brain. Nature 216: 458.
RATNER: Biomaterials Science
Page: 59
60
7
APPLICATION OF MATERIALS IN MEDICINE AND DENTISTRY
Sprung, J., Kindscher, J. D., Wahr, J. A., Levy, J. H., Monk, T. G.,
Moritz, M. W., and O’Hara, P. J. (2002). The use of bovine
hemoglobin glutamer-250 (Hemopure) in surgical patients: results
of a multicenter, randomized, single-blinded trial. Anesth. Analg.
94(4): 799–808.
Tsuchida, E. (ed.) (1995). Artificial Red Cells. John Wiley & Sons,
New York.
Walder, J. A., Zaugg, R. H., Walder, R. Y., Steele, J. M., and
Klotz, I. M. (1979). Diaspirins that crosslink alpha chains
of hemoglobin: bis(3,5-dibromosalicyl)succinate and bis(3,5dibromosalicyl)fumarate. Biochemistry 18: 4265–4270.
Winslow, R. M. (1996). Blood Substitutes in Development. Ashley
Publications, pp. 27–37.
Wong, J. T. (1988). Rightshifted dextran hemoglobin as blood substitute. Biomater. Artif. Cells Artif. Organs 16: 237–245.
Yu, W. P., and Chang, T. M. S. (1996). Submicron polymer membrane
hemoglobin nanocapsules as potential blood substitutes: preparation and characterization. Artif. Cells Blood Subst. Immobilization
Biotechnol. 24: 169–184.
Peritubular capillaries
Glomerulus
Bowman's
capsule
H2O
Electrolytes
Glucose
Os
mo
(AT
Proximal
tubule
sis
Distal tubule
H2O
) NA+
K+
H2O Osmosis
(AT)
Active
Na+ transport
Activ
NH3
trans e
port
H+
Active
transport
Glucose
7.6 EXTRACORPOREAL ARTIFICIAL ORGANS
Paul S. Malchesky
Historically, the term “extracorporeal artificial organs” has
been reserved for life-support techniques requiring the on-line
processing of blood outside the patient’s body. The substitution, support, or replacement of organ functions is performed
when the need is only temporary or intermittent support may be
sufficient. The category of extracorporeal artificial organs does
not include various other techniques that may justifiably be
considered as such, such as infusion pumps or dermal patches
for drug delivery, artificial hearts used extracorporeally, eyeglasses and contact lenses for vision, and orthotic devices and
manipulators operated by neural signals to control motion.
This chapter focuses on artificial organ technologies
that perform mass-transfer operations to support failing or
impaired organ systems. The discussion begins with the oldest and most widely employed kidney substitute, hemodialysis,
and outlines other renal assist systems such as hemofiltration
for the treatment of chronic renal failure and fluid overload
and peritoneal dialysis. The blood treatment process of hemoperfusion, and apheresis technologies, which include plasma
exchange, plasma treatment, and cytapheresis—used to treat
metabolic and immunologic diseases—are also discussed. In
addition, blood-gas exchangers, as required for heart–lung
bypass procedures, and bioartificial devices that employ living
tissue in an extracorporeal circuit are addressed. Significant
concerns and associated technological considerations regarding these technologies, including blood access, anticoagulation,
the effects of the extracorporeal circulation, including blood
cell and humoral changes, and the biomodulation effects of
the procedure and materials of blood contact are also briefly
discussed.
KIDNEY ASSIST
The kidney’s function is to maintain the chemical and water
balance of the body by removing waste materials from the
[15:40 2/3/03 ch-07.tex]
FIG. 1. Schematic representation of the nephron.
blood. The kidneys do this by sophisticated mechanisms of filtration and active and passive transport that take place within
the nephron, the single major functioning unit in the kidney, of which there are about 1 million per kidney (Fig. 1).
In the glomerulus, the blood entry portion of the nephron, a
blood filtration process occurs in which solutes up to 60,000
Da are filtered. As this filtrate passes through the nephron’s
tubule system, its composition is adjusted to the exact chemical requirements of the individual’s body to produce urine. In
addition to these functions, the kidneys support various other
physiological processes by performing secretory functions that
aid red blood cell production, bone metabolism, and bloodpressure control. Renal failure occurs when the kidneys are
damaged to the extent that they can no longer function to
detoxify the body. The failure may be acute or chronic. In
chronic renal failure or in the absence of a successful transplant, the patient must be maintained on dialysis. More than
300,000 people in the United States and more than one million
worldwide are on dialysis.
Dialysis
Dialysis is the process of separating substances in solution
by means of their unequal diffusion through a semipermeable
membrane. The essentials for dialysis are (1) a solution containing the substance to be removed (blood); (2) a semipermeable
membrane permeable to the substances to be removed and
impermeable to substances to be retained (synthetic membrane
RATNER: Biomaterials Science
Page: 60