American Journal of Hematology 7: 119-125 (1979)

A New Treatment Strategy for Hemophilia

B: Incorporation of Factor IX into Red Cell
Ghosts
Jonathan C. Goldsmith, Marjorie E.S. Roer, and Eugene P. Orringer
Division of Hematology, Departments o f Medicine and Laboratory Medicine, University
of North Carolina School of Medicine, Chapel Hill, North Carolina

Although patients with severe hemophilia B have lifelong spontaneous hemorrhage and crippling hemarthroses, patients with 0.0 1-0.03 U/ml Factor IX
activity have a milder disease state. The clinical condition of severely affected
individuals could potentially be improved by prolonging the half-life of transfused Factor IX. The feasibility of incorporating Factor IX into red cell ghosts
was suggested by resealing experiments with similar sized molecules such as
albumin. We have prepared resealed red cell ghosts containing human Factor
IX and X. Human red cells were subjected t o hypotonic lysis at O"C, pH 6.0.
Commercial prothrombin complex concentrate was dissolved in the lysing
medium immediately prior to the addition of the red cells. After being returned to isotonicity, the red cell ghosts were annealed at 37OC for 30 minutes and
then washed extensively. When intact red cell ghosts were tested, no Factor
IX or X activity could be demonstrated. After disruption of the red cell ghost
membranes with 3M urea or 2% Triton X-100, the procoagulants could be
quantitatively recovered. Similar recovery of the clotting factors could be
demonstrated from the lysate and the early wash samples. Red cells from Factor
IX and X deficient patients served equally well as those from normal subjects.
Red cell ghosts prepared in similar fashion but not exposed t o the procoagulants
had negligible clotting activity. We have demonstrated that human clotting
factors can be incorporated into red cell ghosts. The ability of this system t o prolong the biological half-life of Factor 1X is under investigation.
Key words: hemophilia B, Factor IX,red cell ghosts

INTRODUCTION

Hemophilia B is an X-linked recessive disorder characterized by a decreased circulating level of Factor IX coagulant activity and lifelong clinical hemorrhage. Patients with
severe hemophilia have undetectable plasma levels of Factor IX (< 0.01 U/ml) and suffer
from recurrent spontaneous hemarthroses and ultimately crippling joint disease. Those
affected individuals with low but measurable circulating levels of Factor IX (0.01-0.03
A portion of this work has been published in abstract form: [Clin Res 27:295A, 19791.
Received for publication May 1, 1979;accepted July 1, 1979.
Address reprint requests to Dr. Eugene P. Orringer, Division of Hematology, Department of Medicine,
School of Medicine, The University of North Carolina, Chapel Hill, NC 27514.

0361-8609/79/0702-0119$01.70 0 1979 Alan R. Liss, Inc.

120

Goldsmith, Roer, and Orringer

U/ml) have a milder clinical course. Methods prolonging the half-life of infused Factor
IX beyond its expected 20-24 hours could potentially ameliorate the hemorrhagic state
in severely affected individuals. In an attempt to create a preparation which might accomplish this goal, resealed red cell ghosts were prepared which contained commercially
available prothrombin complex concentrate as a source of Factor IX. The feasibility
of incorporating Factor IX (MW 57,000) [ l ] into red cell ghosts was suggested by earlier
resealing experiments with molecules of similar size such as albumin and hemoglobin [2,3].
At the point of hypotonic lysis, the procoagulant molecules gain access to the cell interior. After isotonicity is restored, the red cell membrane recovers its normal permeability
characteristics and the Factor IX molecules are sealed within the ghosts. When prepared in
this manner, the red cell ghost represents a physiologic vehicle for the infusion of Factor
IX which could potentially extend the survival of the procoagulant in vivo.
METHODS AND MATERIALS

Resealed red cell ghosts were prepared by a modification of previously published

methods [ 2 , 3 ] in which hypotonic lysis was carried out at 0C in a thermostatted beaker
with a magnetic flea stirrer. After obtaining informed, written consent, heparinized blood
samples (25 units/ml) were obtained from normal human volunteers as well as subjects
congenitally deficient in Factors IX and X. The freshly drawn red blood cells were then
washed twice in cold unbuffered NaCl 150 mM, resuspended in this same medium at an
hematocrit of 305'4, and placed in the thermostatted beaker. An aluminum foil cap was
placed over the beaker, and the red cell suspension was gassed with hydrated 100%C 0 2
until the pH was reduced to 6.0. The suspension was removed and washed once with cold
NaCl(l50 mM), The red cells were then packed and cooled to 0C on salted ice. The
hemolyzing solution containing MgS04 4 mM, Tris 1 mM, and ATP 1 mM was next
transferred to the thermostatted beaker and titrated to pH 6.0 at 0C using 100 mM acetic
acid. At this point a predetermined amount of Factor IX concentrate (see Table I legend)
was added to the ice cold hemolyzing solution. As soon as the procoagulant had dissolved,
one volume of red cell suspension was injected as rapidly as possible into the beaker which
contained five volumes of cold hemolyzing solution. The pH of the lysate remained between
6.0 and 6.2. Five minutes later, one volume of reconstitution medium (KCI 1160 mM,
Tris 4 3 mM) was added, after which the pH rose to 7.0-7.3. Following a further 10
minute incubation at O"C, the suspension was transferred to a polystyrene Erlenmeyer
flask and incubated for 30 minutes at 37C in a water bath shaker. At the end of this
period of annealing, the suspension was removed and the ghosts were washed vigorously
four times to remove any procoagulant material which might have become adsorbed
to the outer ghost membrane. Solution A (the ghost wash) contained NaCl 150 mh4 and
HEPES 5 mM (pH 7.4 at 25C). The hemoglobin concentration in the resealed ghosts was
in all cases the same as that found in the hemolysate, indicating that 100% hemolysis
had been achieved.
Factor IX activity was quantitated in a one-stage assay using either human [4] or
canine [5] congenitally Factor IX deficient plasma as substrate. All samples were assayed
at four dilutions in Human Diluent which contained NaC162.5 mM, sodium citrate 18.2
mM, and Imidazole-HC1 104 mM (pH 7.2). Clotting assays were performed immediately
on samples maintained at 4C. A pooled platelet-poor normal plasma served as a standard.
Factor IX activity in red cell ghost suspensions was assayed before and after lysis with 3 M
urea. Similar data could be obtained if lysis were accomplished with 2% Triton X-100.

Factor IX in Red Cell Ghosts

121

Samples were also assayed in the non-activated partial thromboplastin time test (NAPTT)
[ 6 ] .Ghosts were quantitated using a model Zf Coulter counter.
Thrombofax was obtained from Ortho Diagnostics (Raritan, New Jersey) and ATP was
purchased from Sigma Chemical Co. (St. Louis, Missouri). All chemicals were of reagent
grade or better.
RESULTS

The data from a representative experiment are presented on Table I. In this study 4 ml
of packed red cells were added to 20 ml of hemolyzing solution. The subsequent addition
of 4 ml of reconstitution medium brought the volume in the thermostatted beaker to
28 ml. Prior to the addition of the intact red cells, 70 mg of prothrombin complex concentrate had been dissolved in the hemolyzing solution. Thus, at the end of the study there
were 2.5 mg of concentrate per ml in the beaker. As can be seen from the table, this concentration of prothrombin complex yielded a Factor IX activity of 1.61 units per ml (U/ml)
Because Factor IX is thermolabile, we examined the effect of a 30-minute incubation at
37C on the activity of the starting material. As noted on the table, such an incubation resulted in only minimal deterioration of Factor IX procoagulant activity.
After the 30-minute annealing process, the ghosts were spun from the hemolysate
and washed four times with Solution A. Each wash cycle had a 20: 1 ratio of Solution
A:ghosts. Factor IX activity in the hemolysate was 1.2 U/ml, a value which approached
the predicted level of 1.50 U/ml. The Factor IX activity in the initial wash of 0.32 U/ml
indicated that some procoagulant might have been adsorbed to the outside of the ghosts,
thus accounting for the slightly lower than expected concentration of Factor IX present
in the hemolysate. The intact red cell ghosts demonstrated no procoagulant activity in
our assay system. Factor IX activity could only be recovered from the ghosts after they
had been lysed with either urea or Triton. Sonication and fyeeze-thaw techniques failed to
disrupt the red cell ghost membranes. The presence of 0.8 units of Factor 1X per ml of
packed ghosts represents a yield of approximately 60% of the predicted value.
The following control experiments were performed, the results of several of which
are also included in Table I. The lysing agent (either Triton or urea) failed to inhibit Factor
1X activity in our assay system. Red cell ghosts prepared with bovine serum albumin rather
TABLE I. Factor IX Procoagulant Activity* Recoverable From a Variety of Solutions?
and Ghost Systemst
Test material
Human diluent
Human diluent + 3 M urea
Human diluent, incubated at 37C for 30 niin
If ern oly sa t e
Ghost wash No. 1
Ghost wash No. 4
Intact ghosts (containing Factor IX)
Urea-lysed ghosts (containing Factor IX)
Urea-lysed ghosts (containing bovine serum albumin)

Factor 1X activity
1.61

1.56
1S O
1.25
0.32
0.00
0.00
0.80
0.00

*Expressed as units of Factor IX/ml

tAll of the appropriate samples contained 2.5 mg of prothrombin coniplex/nii

t 10" ghosts/ml

122

Goldsmith, Roer, and Orringer

than prothrombin complex concentrate had no procoagulant activity. In addition Factor IX

coagulant activity was unaffected when albumin containing ghosts were mixed with a Factor
IX-containing solution. When the Tris buffered hemolysis and resealing solutions were replaced by phosphate buffered media of similar osmolality and pH, identical procoagulant
activity was recovered. The NAPTT was not shortened by: a) the prothrombin complex concentrate used in these experiments; b) the hemolysate obtained after annealing at 37C for
30 minutes; or c) the lysed or intact red cell ghosts.
The effect of increasing the amount of prothrombin complex concentrate added to
the hemolyzing solution on the Factor IX procoagulant activity subsequently recovered
from both the hemolysate and the resealed ghosts is shown in Figure 1. A direct relationship is evident in both instances. In panel A the material recovered from the hemolysate is
virtually identical to that added to the hemolyzing solution. In panel B a linear correlation
between the amount of material added to the hemolyzing solution and that recovered from
the red cell ghosts is shown. In this study the percentage incorporation of Factor IX is
approximately 35%.
In the last group of studies heparinized blood samples were obtained from patients
congenitally deficient in either Factor IX or Factor X. Using these erythrocytes, resealed
ghosts containing prothrombin complex concentrate were prepared, and the activity of
the deficient clotting factor was then measured. As seen in Table I1 successful incorporation of Factor IX into ghosts from Factor IX-deficient patients and Factor X into ghosts
from Factor X-deficient patients could be achieved.
DISCUSSION
Hemophilia B (Christmas disease) was initially described in 1952 [7-9]. Whole blood
and fresh frozen plasma were the primary sources of Factor IX available for replacement
therapy in these patients. More recently, however, concentrates of the vitamin K-dependent
blood clotting factors (IT, VII, IX, X) have become commercially available. Despite reported
complications such as thromboembolism [ 101 and hepatitis [ I I ] , the use of Factor IX
concentrates has risen dramatically over the past 10 years. This increase can be directly attributed to the inherent advantages which the concentrates possess when compared to conB.

r=0.97

/ ,

!
i
W
[L

HEMOLYSIS MEDIUM

I.

Factor

IX ( u / m l )

Fig. 1. Factor IX procoagulant activity recoverable from: A ) the hemolysate; and B) the resealed
ghosts; as a function of the amount of prothrombin complex concentrate present in the hemolyzing solution.

Factor IX in Red Cell Ghosts

123

TABLE 11. Incorporation of Procoagulants Into Erythrocytes From Patients Congenitally

Deficient in Factor IX or Factor X

Source of erythrocytes

Factor IX activitya

Factor X activitya

Hemolysate Ghostsb

Hemolysate Ghostsb

Factor IX deficient patient'

1.25

0.8

1.5

0.32

Factor X deficient patient'

NT

NT

1.25

0.46

Normal control

1.43

0.53

0.70

0.23

"Expressed as U/ml

b l O ' O erthrocyte ghosts/ml

CBoth subjects had no detectable immunologic activity of their deficient factor (CRM-) when
tested with monospecific heterologous antibody.

NT, not tested.

ventional plasma therapy. These advantages include: 1) the ability to provide factor replacement in a concentrated form capable of achieving high circulating levels of Factor IX procoagulant without the hazard of fluid overload; '2) the suitability of these preparatioiis for
home administration since they are stable for prolonged periods of time at 4C and can be
readily reconstituted and infused by syringe and needle; and 3) the cost per unit administered which is similar t o that of plasma [ 121 .
The in vivo kinetics of infused Factor IX differ considerably from those of certain
other clotting proteins. Immediately following an infusion of Factor IX the amount of
procoagulant demonstrable within the circulation is 40% or less of that which would be
predicted, assuming the plasma volume as the total space of distribution [I31 . By way
of contrast the recovery of Factor VllI approaches 100% [14j. These observations have
been interpreted as evidence that Factor IX is distributed not only throughout the intravascular space but also into extravascular sites. This could result from the smaller
molecular weight of Factor IX (57,000) compared with that of Factor VIII (> 1,000,000)
[14]. In addition to the poorer recovery, the disappearance of Factor IX exhibits a
biphasic survival curve within the circulation of both hemophiliac men [13] and dogs
[15]. The first component has a T% between 3 and 6 hours and the second component
has a T%of 20 to 30 hours.
Theoretically, one might be able to alter the course of severe hemophilia B by administering Factor IX procoagulant in a fashion which could either prevent its rapid distribution to extravascular sites or could preserve the function and extend the half-life of
the molecule. Liposomes have previously been employed as a physical means for enzyme
replacement therapy in certain human deficiency states [16] . For example, patients with
Gaucher's disease have been given liposomes containing glucocerebroside ~-glucosidase
activity [17]. Since Factor IX is a serine esterase, these phospholipid suspensions could
potentially serve as carriers of the procoagulant. We have chosen red cell ghosts rather
than liposomes for several reasons. First, they offer the opportunity to use a patient's own
cells as the modality for replacement therapy rather than a synthesized, foreign material.
Second, our goal is to increase the plasma activity of Factor IX for a prolonged time interval. This might be accomplished through the normal senescence of the red cell ghosts
with the subsequent intravascular discharge of their contents. By contrast liposomes have
been used to increase the intracellular content of a particular deficient enzyme. Third,

124

Goldsmith, Roer, and Orringer

liposomes which are phospholipid suspensions could accelerate blood clotting and potentially threaten their recipients with disseminated intravascular coagulation. Theoretically,
red cell stroma could pose a similar problem. Previous workers have suggested that this
material can serve as a phospholipid source for coagulation assays [18]. Under our experimental conditions however, disrupted or intact red cell ghosts showed no intrinsic
procoagulant activity.
Our experiments have demonstrated that human Factor IX activity can be incorporated into red cell ghosts. Semiquantitative recovery of the procoagulant can be shown
after disruption of the red cell ghost membrane with urea or detergent. Red cells from
normal individuals as well as Factor IX- and X-deficient patients serve equally well as vehicles for the encapsulation of Factor IX from commercially available prothrombin complex concentrates. Factor X activity could also be demonstrated in these red cell ghosts.
The red cell ghost represents a promising vehicle for prolongation of the half-life of Factor IX activity. The achievement over an extended period of time of low but measurable
circulating levels of Factor IX in severely affected hemophilia B patients might ameliorate
the disease and prevent the development of destructive arthropathy. The in vivo application of this technology is currently under investigation.
ACKNOWLEDGMENTS

This work was supported in part by USPHS grants: AM1 1356 and HL06350. Dr.
Goldsmith was supported by a National Hemophilia Foundation Fellowship.
The authors acknowledge the generous support of Cutter Laboratories, Berkeley,
California, and Ms Debby Clark, their local representative who made the prothrombin
complex concentrate available for these studies. We also recognize the invaluable secretarial assistance of Fay Weaver.
REFERENCES
I . DiScipio RG, Hermodson MA, Yates SG, Davie EW: A comparison of human prothrombin, factor
IX (Christmas factor), factor X (Stuart factor), and protein S. Biochemistry 16:698-706, 1977.
2. Schwoch G, Passow H: Preparations and properties of human erythrocyte ghosts. Mol Cell Biochem
2:197-218, 1973.
3. Orringer EP, Roer ME: An ascorbate-mediated transniembrane-reducing system of tlic human
erythrocyte. J Clin Invest 63:53-58, 1979.
4. Barrow EM, Bullock W, Graham JB: A study of the carrier state for plasma thromboplastin component
(PTC, Christmas factor) deficiency, utilizing a new assay procedure. J Lab Clin Med 55:936-945,
1960.
5. Goldsmith JC, Chung KS, Roberts HR: A simple assay for human factor IX: Use of canine hemophilia
B plasma as substrate. Thromb Res 12:497-502, 1978.
6. Kingdon HS, Lundblad RL, Veltkamp J J , Aronson DL: Potentially thrombogenic materials in factor
IX concentrate. Thromb Diath Haeinorrh 33:6 17-629, 1975.
7. Aggeler PM, White SG, Glendening MB, Page EW, Leake TB, Bates G : Plasma thromboplastin component (PTC) deficiency. A new disease resembling hemophilia. Proc Soc Exp Biol Med 79:692694,1952.
8. Biggs R , Douglas AS, Macfarlane RG, Dacie JV, Pitney WR, Merskey C, OBrien J R : Christmas
disease. A condition previously mistaken for haemophilia. Br Med J ii: 1378-1382, 1952.
9. Schulman I, Smith CH. Hemorrhagic disease in an infant due to deficiency of a previously undescribed clotting factor. Blood 7:794-807, 1952.
10. Blatt PM, Lundblad RL, Kingdon HS, McLean G , Roberts HR: Thrombogenic inaterials in prothrombin complex concentrates. Ann Intern Med 81:766-770, 1974.

Factor IX in Red Cell Ghosts

125

11. Roberts HR, Blatt PM: Post-transfusion hepatitis following the use of prothrombin complex concentrates. Thromb Diath Haemorrh 33:610-616, 1975.
12. White GC 11, Lundblad RL, Kingdon HS: Prothrombin complex concentrates. Preparation, properties, and clinical uses. Current Topics in Hematology, Vol. 11, in press.
13. Zauber NP, Levin J : Factor IX levels in patients with hemophilia B (Christmas disease) following
transfusion with concentrates of factor IX or fresh frozen plasma (FFP). Medicine 56:213-224, 1977.
14. Hoyer LW: Von Willebrands disease. Prog Heniostasis Thromb 3:231-287, 1976.
15. Goldsmith JC, Orringer EP: Unpublished data.
16. Gregoriadis G: Liposoines in therapeutic and preventive medicine. The development of the drugcarrier concept. Ann NY Acad Sci 308:343-370,1978.
17. Belchetz PE, Braidman IP, Crawley JCW, Gregoriadis G : Treatment of Gauchers disease with
liposome-entrapped g1ucocerebroside:Pglucosidase. Lancet i: 1 16-1 17, 1977.
18. Quick AJ, Georgatsos JG, Hussey CV. The clotting activity of human erythrocytes: theoretical
and clinical implications. Am J Med Sci 228:207-213, 1954.