776 Blood, Vol. 53, No. 4 (April), 1979
Congenital Deficiency of Blood Clotting FactorsII, VII, IX, and X
By Kuo-San Chung. Annie Bezeaud, Jonathan C. Goldsmith, Campbell W. McMillan,
Doris M#{233}nach#{233},and Harold R. Roberts
A patient congenitally deficient in factorsII. VII. IX. and X has been further investi-gated after a follow-up of 1 5 yr. At birth,these factors. when determined by clottingassays, were undetectable. Following ther-apy with vitamin K1. the clotting activity ofthese factors rose but never exceeded1 8 % of normal. Immunologic assaysrevealed much higher levels of thesefactors than did clotting assays, thussuggesting that the vitamin-K-dependentfactors were present in abnormal forms.Two-dimensional crossed immunoelectro-
phoresis showed that at least two forms ofprothrombin were present in the patient’s
plasma. One form was similar to normalprothrombin; the other had the same
mobility as acarboxyprothrombin. In addi-
tion. the majority of this fast-migrating
peak was not adsorbable onto insolublebarium salts. These observations sug-
gested that some molecules of the
patient’s prothrombin lacked the normalcomplement of gamma-carboxyglutamic
acid residues. This observation was con-
firmed by a specific assay for gamma-
carboxyglutamate. Since malabsorption ofvitamin K, warfarin intoxication, and
hepatic dysfunction were excluded as
causes of this patient’s syndrome, this rare
congenital abnormality could represent
either a defective gamma-carboxylation
mechanism within the hepatocyte or faulty
vitamin K transport.
I T HAS BEEN KNOWN for many years that vitamin K is necessary for the
complete synthesis of prothrombin (factor II) and factors VII, IX, and X.’
Vitamin K is also necessary for the production of protein C in the bovine species
and protein S in man.2’3 Recent investigations have shown that vitamin K acts at a
postribosomal level in the hepatocyte to modify precursor forms of these factors.4
Thus, vitamin K, in the presence of a carboxylating enzyme, hepatic microsomes,
02, and CO2. converts glutamic acid residues on the amino-terminal regions of
precursor forms of prothrombin and factors VII, IX, and X to gamma-carboxyglu-
tamic acid residues.5 These gamma-carboxyglutamic acid residues are necessary
for calcium-dependent phospholipid binding by the vitamin- K-dependent clotting
factors and are prerequisites for normal blood coagulation.6
Whereas defective gamma-carboxylation of glutamic acid residues occurs
following therapy with coumarin drugs,’ a congenital abnormality resulting in
defective gamma-carboxylation of the vitamin-K-dependent factors has hitherto
not been reported.
We have reevaluated a unique I 5-yr-old female who has congenital deficiency of
From the Departments of Medicine. Pediatrics. and Pathology. School of Medicine, Universit�’ of
North Carolina at Chapel Hill, NC., and the Service Central d’Immunologie et H#{234}matologie, Hbpital
Beaujon, Clichy, France.
Submitted October 8, 1978; accepted December 18. 1978.
Supported by NIH Grants HL-06350. HL-21975. HL-07149, and HL-20319 from the National
Heart, Lung. and Blood Institute. Dr. Goldsmith is a Fellow ofthe National Hemophilia Foundation.
Address reprint requests to Dr. K. S. Chung. Department of Medicine, University of North Carolina
School of Medicine, 415 Burnett-Womack Clinical Sciences Building 229-H, Chapel Hill, NC.
275 /4.(C) /979 by Grune & Stratton, Inc. JSSN 0006-4971/79/5304-0024$02.00/0
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
FACTORS II, VII, IX, AND X 777
the vitamin-K-dependent blood clotting factors.7 She was initially seen at the age of
3 mo with multiple bruises and a history of numerous hemorrhagic episodes.
Vitamin K deficiency due to malabsorption, liver disease, and warfarin intoxication
was excluded, both at the time of initial presentation and on several subsequent
occasions. Based on current studies, it appears that this patient may have a defect
in the gamma-carboxylation mechanism that results in the production of immuno-
logically recognizable factors II, VII, IX, and X that lack the full complement of
gamma-carboxyglutamic acid residues and therefore lack coagulant activity.
MATERIALS AND METHODS
Patient
The patient has been previously described.7 When she was initially seen it was determined by assay of
her plasma for bishydroxycoumarin and warfarin that she had received no anticoagulant drugs.
Subsequent assays for these drugs have also been negative. At the age of 6 mo the patient was observed
for 2 wk in the hospital, at which time it was ascertained that she was not receiving vitamin K
antagonists surreptitiously. In addition, studies at that time revealed no evidence for malabsorption or
liver disease.7 Subsequent to the initial report, the patient has been followed periodically for I 5 yr.
During this time she has required daily vitamin K supplements. Initially vitamin K was administered
parenterally, but subsequently oral doses have been shown to be efficacious. However, even with massive
intravenous doses of vitamin K1, the patient’s K-dependent factors never returned to normal, although
the prothrombin time decreased from more than 100 sec to 20-25 sec. Withdrawal of vitamin K1
resulted in prompt return of bleeding and a marked decrease in the clotting activities of prothrombin and
factors VII, IX, and X. Studies on parents ofthe patient showed normal levels offactors II, VII, IX, and
x.
Plasma
Normal human blood was collected from 10 normal donors into 3.2% trisodium citrate (8 parts blood
to I part citrate). After centrifugation the plasma was pooled and stored at -70#{176}C. The patient’s
plasma was prepared in the same manner.
Reagents
Sodium chloride, sodium citrate, kaolin, calcium lactate, diethylbarbituric acid, tris(hydroxyme-
thyl)aminomethane, sodium azide, potassium hydroxide, benzene, monobasic and dibasic sodium
phosphate, and barium chloride were purchased from Fisher Scientific, Fair Lawn, N.J. Imidazole,
acrylamide, and cyanogen bromide were purchased from Eastman Kodak, Rochester, N.Y. Calcium
chloride, methanol, glacial acetic acid, and formic acid were obtained from Mallinckrodt, St. Louis, Mo.
Sephadex G-100 and Sepharose 4B were obtained from Pharmacia Fine Chemicals, Piscataway, N.J.,
and Uppsala, Sweden. Alumina C--y gel was purchased from Calbiochem, La Jolla, Calif. Borosilicate
glass clotting tubes (10 X 75 mm) were purchased from Arthur H. Thomas, Philadelphia, Pa. Agarose
and Coomassie brilliant blue were from Bio-Rad Laboratories, Richmond, Calif. Simplastin was
purchased from General Diagnostics, Morris Plains, N.J. Thrombofax was from Ortho Diagnostics,
Raritan, Ni. Dansyl chloride and the venom from Echis carinatus were obtained from Sigma
Chemical, St. Louis, Mo. Staphylocoagulase was purchased from Diagonostica Stago, Asnieres, France.
Canine factor V and fibrinogen were obtained from platelet-poor canine oxalated plasma that was
exhaustively treated with barium sulfate and then raised to 25% saturation with (NH4)2S04 (to obtain
fibrinogen-rich material) and then to 33% (to obtain factor-V-rich material).
Normal Human Prothrombin
Highly purified normal human prothrombin was a by-product ofour factor IX purification procedure
employing heparin agarose chromatography as the last step.t’9
Heterologous Antibodies to Normal Human Prothrombin
Antibodies to normal human prothrombin were raised in two rabbits by four weekly injections of 0.25
mg of pure prothrombin and an equal volume of complete Freund’s adjuvant. The rabbits were then bled
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
778 CHUNG ET AL.
at weekly intervals, beginning 1 wk after completion of the immunization. The rabbit serum was heated
at 56#{176}Cfor 30 mm and adsorbed with one-tenth volume of alumina C-’y gel. After centrifugation, the
supernatant was dialyzed against 0.0l-M sodium phosphate buffer (pH 7.0), and the dialysate was
applied to a prothrombin-Sepharose affinity column that was prepared by coupling normal prothrombin
to cyanogen-bromide-activated Sepharose 4B.’#{176}Antiprothrombin antibodies were eluted with 0.l-M
glycine-HCI (pH 2.8), then immediately brought to pH 7.0 by the dropwise addition ofdilute NaOH.
Purification ofthe Patient ‘s Prothrombin
Because there were only limited quantities of the patient’s plasma, the patient’s prothrombin was
prepared from small amounts of citrated plasma. Ten milliliters of the patient’s plasma was stirred with
1 ml of l-M BaCl2 and centrifuged. The supernatant containing nonadsorbable prothrombin was
dialyzed against 0.01-M sodium phosphate buffer (pH 7.0) and applied to an antiprothrombin antibody
affinity column ( I .5 X 4.0 cm) that was prepared by coupling the purified antibody to cyanogen-
bromide-activated Sepharose 4B. The prothrombin was eluted with 0.l-M glycine-HCI buffer (pH 2.8)
and immediately raised to pH 7.0 with dilute NaOl-1. The eluate was then concentrated and subjected to
gel filtration on a Sephadex G-I00 column (0.35 x 8.5 cm). The purification was monitored
immunologically by the Ouchterlony double-diffusion technique.#{176} The results showed a single precipitin
line when the preparation was tested against an antiprothrombin antibody, against anti-whole-
human-plasma, and against antiserum to the barium citrate eluate of normal human plasma.
Determination ofGamma-Carboxyglutamic Acid in Patient
and Normal Prothrombin
Gamma-carboxyglutamic acid residues were assessed in the purified prothrombin from both normal
plasma and the patient’s plasma using thin-layer polyamide chromatography.’2 Purified prothrombin
was hydrolyzed for 24 hr at 100#{176}Cin 2-M KOHl3 The resulting amino acids were labeled with dansyl
chloride and subjected to two-dimensional chromatography on thin-layer polyamide sheets using
H,0-formic acid (100:1.5 v/v) in the first dimension and benzene-acetic acid (9:1 v/v) in the second
dimension. Labeled amino acids could be detected under short-wave ultraviolet light (A = 254 nm).t2 A
synthetic monomeric gamma-carboxyglutamic acid was prepared according to the methods of Boggs et
al.’4 It was then dansylated, subjected to chromatography as described previously, and used as a control
to detect a similar amino acid in normal prothrombin and in the patient’s prothrombin.
Coagulation Assays
Blood clotting factors VII, IX, and X were assayed in one-stage tests using as substrate platelet-poor
plasma obtained from patients congenitally deficient in the respective factors.�#{176} Prothrombin coagu-
lant activity was measured with the one-stage method,’t the two-stage method,’9 staphylocoagulase,2#{176}
and venom from E. carinalus. For the latter method, normal plasma and the patient’s plasma diluted
1 :7, 1 : I 4, 1 :2 1 , and I :28 in imidazole-HCI buffer (pH 7.2) were tested. Two milligrams of E. carinatus
venom were dissolved in I ml of normal saline; 10 MI of this solution were added to 0.2 ml of the plasma
dilutions and incubated for 10 mm at 37#{176}C.Two-tenths milliliter ofcanine fibrinogen was then added, a
stop watch was started simultaneously, and the tubes were tilted and observed for the appearance of a
clot.
Immunologic Procedures
Antibody neutralization was monitored in clotting assays as follows: Factor IX neutralization assays
were conducted by a previously described method2’ with a well-characterized specific human inhibitor of
factor IX.22 Factor X neutralization was a modification of the factor IX neutralization assay using a
heterologous antibody to factor X. Prothrombin neutralization was conducted using a heterologous
antibody to factor II in a two-stage assay to detect residual prothrombin activity.
Radial immunodiffusion was performed in 1% agarose in 0.I-M phosphate-buffered saline at pH
7�23 Electroimmunoassays were performed with agarose containing a specific heterologous antibody to
prothrombin.24 Crossed immunoelectrophoresis was conducted in 0.075-M tris-barbital buffer (pH 8.6)
containing 2-mM calcium lactate at 4#{176}Cin both directions.2126
RESULTS
Clotting factors not dependent on vitamin K for synthesis have been normal
throughout the patient’s life, as shown in Table I . Factor V was initially 57% of
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
FACTORS II, VII, IX, AND X 779
Table 1 . Speci fic Clotting Factor Assays (Non-Vitamin-K-Dependent)
Patient Control
Fibrinogen (mg/mI)
FactorV(%)
Factor VIII (%)
FactorXl(%)
FactorXll(%)
3.27 2.95
57-100 100
108 100
100 100
100 100
Table 2. Patient: Vitamin-K-Dependent Clotting Factors
Before After
Vitamin K, Vitamin K,
Treatment Treatment)% of normal) )% of normal)
Factorll
FactorVll
FactorlX
FactorX
<1 7.0
<1 6.9
<1 18.2
<1 7.0
normal, but has subsequently been 100% of normal. The patient’s vitamin-
K-dependent clotting factor levels are shown in Table 2. Note that factors II, VII,
Ix, and X were undetectable before the patient was treated with vitamin K.
Administration of large doses of oral vitamin K, resulted in an increase in these
factors not exceeding 18% of normal. However, despite massive doses of parenteral
vitamin K,, the patient’s vitamin-K-dependent factors never reached normal levels.
To determine if the vitamin-K-dependent factors were actually as low as
indicated by clotting assays, factors II, IX, and X were measured by immunologic
assays using antibody neutralization tests. The results shown in Table 3 indicate
that prothrombin is present in a concentration 57% of normal, factor IX in a
Table 3. Patient: Comparison of Clotting and Immunologic Activities
of Vitamin-K-Dependent Factors
Antibody
ClottingActivity
(% of normal)
NeutralizationActivity
(% of normal)
Factor II 7.0 57
FactorVll 6.9 -a
FactorlX 18.2 100
Factor X 7.0 55
Appropriate monospecific antibody not available.
Table 4. Prothrombin Determinations
(Levels in Percent of Normal)
Patient’s Plasma
Warfarin PlasmaNative Adsorbed
One-stage assay 3.5 1 -Two-stage assay 7 0 57
E. carinatus 47 1 8.2 75
Staphylocoagulase 40 - -
Anti-Il neutralization 56.8 55 89
Electroimmunoassay 58.4 45 85
Radial immunodiffusion 51 39 81.5
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
.. .
. -: -#{176}�
B �e�
A.
I
j
c
780 CHUNG ET AL.
concentration 100% of normal, and factor X in a concentration 55% of normal.
Factor VII was not measured immunologically because an appropriate antibody
was not available for this purpose. These data strongly suggested that altered forms
of the vitamin-K-dependent factors were present in the patient’s plasma, forms that
retained immunologic activity but lacked coagulant activity.
To explore this possibility further, the prothrombin content of the patient’s
plasma was measured by several techniques and compared with that of normal
pooled human plasma and of plasma obtained from a patient on warfarin therapy.
Alp ., . . - � 4�#{149}C�.
:�‘� #{149}�,
.i .�‘�:“ � : #{231}..‘�‘: �
,�. :
I
., *4� � � �t
�1
:. �
): � “ � ‘k’.,.
.-.---,
-9
Fig. 1 . Crossed immunoelectrophoresis: (A) patient’s plasma. (B) plasma from a patient onwarfarin therapy. (C) normal pooled plasma. The first dimension was run for 3.5 hr at 1 Ov/cm. andthe second dimension was run overnight at 20 v/cm. Both dimensions were run in 1 % agarosecontaining 2-mM calcium lactate. In the second dimension the agarose contained a specific antibody
to prothrombin.
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
- . ....- .� �-
FACTORS II, VII, IX, AND X 781
As shown in Table 4, when using physiologic activators, i.e., one- and two-stage
clotting assays, the prothrombin clotting activity was found to be very low (less
than 10% of normal). In contrast, when using nonphysiologic activators such as E.
carinatus venom and staphylocoagulase, the levels of prothrombin were found to be
47% and 40% of normal, respectively. Immunologic techniques using a specific
heterologous antiprothrombin antibody gave results similar to those obtained using
nonphysiologic activators. Similar results were obtained when assaying plasma
from a patient receiving long-term anticoagulant therapy with warfarin. After
adsorption of the patient’s plasma by barium citrate, the supernatant exhibited
prothrombin activity when using nonphysiologic activators, and these results were
corroborated by immunologic assays.
When subjected to crossed immunoelectrophoresis (Fig. 1 ), the patient’s plasma
exhibited at least two peaks of prothrombin (Fig. la). The smaller of the two peaks
had the same electrophoretic mobility as normal prothrombin (Fig. ic), whereas
the major peak had the same electrophoretic mobility as the acarboxyprothrombin
found in plasma from a patient on warfarin therapy (Fig. ib). These data suggest
that most of the patient’s prothrombin existed in the acarboxy form.
The patient’s plasma was then adsorbed with barium citrate, which usually
removes all normal prothrombin, but not acarboxyprothrombin, in warfarin plas-
ma. Resulting supernatant plasma was tested by crossed immunoelectrophoresis for
residual prothrombin (Fig. 2). The peak corresponding to normal prothrombin was
A�� ..
� . #{149}:-‘‘
. .. �-:-t.�� � � .,
�4I�
Fig. 2. Crossed immuno-
electrophoresis: (A) patients
plasma after adsorption by bar-ium citrate. (B) normal pooled
plasma control. The conditions
are the same as those de-
scribed in Fig. 1.
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
782 CHUNG ET AL.
removed by barium citrate, whereas most of the peak with the same electrophoretic
mobility as acarboxyprothrombin remained. The barium citrate cake resulting
from adsorption of the patient’s native plasma was eluted and then subjected to
crossed immunoelectrophoresis (Fig. 3). A peak of prothrombin with the same
electrophoretic mobility as normal prothrombin could be found, but in addition a
faster-migrating peak was also found, thus indicating that at least some of the
abnormal prothrombin was adsorbed by barium salts.
A more direct method was used to determine the content of gamma-carboxyglu-
tamic acid residues in the patient’s purified nonadsorbable prothrombin. Alkaline
hydrolysis and two-dimensional thin-layer chromatography of the patient’s
prothrombin and normal prothrombin were performed. The dansylated amino acids
from the patient’s prothrombin and normal prothrombin were compared to each
other and to a standard synthetic monomeric gamma-carboxyglutamic acid. The
migration of the synthetic monomeric gamma-carboxyglutamic acid in this experi-
ment is shown in Fig. 4c. A fluorescent spot corresponding to the gamma-
carboxyglutamic acid control was found in normal prothrombin even when normal
prothrombin was diluted to about the same antigenic concentration as the patient’s
prothrombin (Figs. 4b and 4d). However, when the patient’s prothrombin was
chromatographed in this system (Fig. 4a), no fluorescent spot corresponding to
. . - S .
-� � . � �
�; - � . . ‘ . -.. . ��:-r� #{149}��-�Y
� A�’�”t� S ‘ � � � 1�
�
�- �
� .� ;.�_�_;� #{149};;t6��,� ‘. .. . � __________
C - . �. 1�
� . -:�.. �
� � .. ------;---,, �
B � :‘�
Fig. 3. Crossed immunoelectrophoresis: (A) eluate from barium citrate cake resulted from
adsorption of the patient’s native plasma, (B) normal pooled plasma control. The conditions are thesame as those described in Fig. 1 and 2.
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
FACTORS II, VII, IX, AND X 783
Fig. 4. Thin-layer chromatography of dansylated amino acids from alkaline hydrolysis of: (A)
patient�s purified nonadsorbable prothrombin. (B) normal purified prothrombin. (C) syntheticmonomeric gamma-carboxyglutamic aci ontrol. (D) normal purified prothrombin diluted to approxi-
mately the same antigenic concentrati the patient’s prothrombin. Arrows denote the location ofgamma-carboxyglutamic acid.
gamma-carboxyglutamic acid was seen. This technique is sensitive to a prothrom-
bin gamma-carboxyglutamic acid content 1% of normal.
DISCUSSION
A unique patient with congenital deficiency of factors II, VII, IX, and X has
been evaluated. Even after therapy with massive doses of vitamin K,, either
parenterally or orally, her plasma showed greatly reduced levels of the vitamin-
K-dependent factors when measured by clotting assays. However, when immuno-
logic techniques were employed to measure these factors, much greater quantities
of the vitamin-K-dependent clotting proteins were found. Since vitamin K
deficiency due to malabsorption, liver disease, and warfarin ingestion had been
excluded by appropriate biochemical and spectrophotometric tests, the patient’s
partial response to parenteral and oral vitamin K, suggested a defect in vitamin K
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
784 CHUNG ET AL.
metabolism. Defective vitamin K metabolism could lead to the production of blood
clotting factors deficient in gamma-carboxyglutamic acid. Gamma-caroboxyglu-
tamic acid residues are normally found on the amino-terminal end of molecules of
the vitamin-K-dependent factors and are necessary for these factors to bind not
only to calcium but also to insoluble barium salts. Calcium binding is necessary for
the subsequent binding of these factors to phospholipids or specific receptors on the
platelet surface. In the absence of calcium-dependent phospholipid binding, the
vitamin-K-dependent factors do not participate normally in blood coagulation.
The patient’s prothrombin after vitamin K treatment was found to be 3.5% and
7% of normal when measured using clotting assays dependent on calcium binding.
However, when the patient’s prothrombin was assessed by nonphysiologic activa-
tors, which do not depend on the presence of gamma-carboxyglutamic acid residues
for the generation of thrombin, the level of prothrombin was found to be 40% and
47% of normal. Similarly, immunologic assays of prothrombin, factor IX, and
factor X were 57%, 100%, and 55% of normal. These data strongly suggest that
factors II, IX, and X in the patient’s plasma lacked the full complement of
gamma-carboxyglutamic acid residues. This hypothesis is supported by the finding
that on crossed immunoelectrophoresis in the presence of calcium ions the patient
had at least two peaks of prothrombin, one corresponding in mobility to normal
prothrombin and the other corresponding to a more rapidly migrating peak found in
the plasma of patients treated with warfarin. The rapid peak in the patient’s plasma
is essentially not adsorbable onto barium salts, again suggesting that the patient’s
prothrombin contained a decreased complement of gamma-carboxyglutamic resi-
dues. This hypothesis was substantiated with the observation that amino acids from
the patient’s purified nonadsorbable prothrombin contained no detectable gamma-
carboxyglutamic acid residues when tested by a qualitative technique employing
thin-layer chromatography of dansylated amino acids.
In addition to normal prothrombin and acarboxyprothrombin, the patient’s
plasma contains another form of prothrombin that behaves on crossed immunoelec-
trophoresis, in the presence of calcium ons, as a fast-migrating peak but has the
ability to be adsorbed onto insoluble � ‘im salts. These results suggest that the
patient’s prothrombin is heterogenous ‘� respect to the content of gamma-
carboxyglutamic acid residues. This hypoth :� is supported by the findings of
Esnouf and Prowse,27 who reported different ue,, � �f carboxylation of prothrom-
bin molecules obtained from patients on warfarin t. py.
Since even with immunologic techniques the patt5 ��t’s plasma contained only
about 50% of the prothrombin found in normal plasma, the question arises why
prothrombin deficient in gamma-carboxyglutamic acid residues is not present in
amounts approaching 100% of normal. The answer to this question is not known,
but it is possible that acarboxyprothrombin has a shorter biologic half-life than
normal prothrombin, as has been suggested by Lavergne and Josso.28 Thompson29
has suggested a similar explanation for decreased immunologic levels of acarboxy
factor IX found in patients on warfarin therapy.
Gamma-carboxyglutamic acid as a constituent of the vitamin-K-dependent
clotting factors has been described by Stenflo,3#{176}Nelsestuen et al.,3’ and Magnusson
et al.32 Later, Hauschka et al.33’34 described the presence of these same residues in
bone and in the kidney. Their results suggest that the gamma-carboxyglutamic acid
in proteins other than the blood clotting factors is also apparently vitamin-
K-dependent. It should be noted, however, that our patient had normal renal
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
FACTORS II, VII, IX, AND X 785
function and normal skeletal development, even though her vitamin-K-dependent
clotting factors had been decreased since birth and during the subsequent I 5 yr of
follow-up. Thus if the patient’s defective vitamin K metabolism had affected her
skeletal or renal development, the defects were either obscured by vitamin K
therapy or not detected in these tissues, our end points being currently too crude. It
is also possible that defects in the renal or skeletal system will be manifest only after
the patient gets older.
Although vitamin K deficiency per se and vitamin K antagonism have been
excluded in this patient, the precise defect in vitamin K metabolism is yet to be
clarified. There could be an abnormality in transport of vitamin K in the blood.
However, a carrier protein for vitamin K has not been identified. A second
possibility is that vitamin K may not be handled normally in the liver. Suttie et al.’
as well as others,35’36 have suggested that vitamin K,, when it enters the hepatocyte,
is obligatorily converted to vitamin K, epoxide by an epoxidase system. Vitamin K,
epoxide is then converted to reduced vitamin K, by a specific reductase system in
the hepatocyte before it can act as a co-factor in the carboxylating mechanism. We
have been unable to exclude in this patient defects of the epoxidase or reductase
systems. Another possibility is that vitamin K is metabolized normally within the
hepatocyte but that it does not function as a co-factor for gamma-carboxylation
because of a defective carboxylating mechanism. Suttie et al.’ have demonstrated
that the gamma-carboxylation of glutamic acid residues in the hepatocyte requires
hepatic microsomes, reduced vitamin K,, a carboxylating enzyme, 02, and CO2. It
seems unlikely that the patient’s hepatic microsomes are defective, but it is possible
that the patient has a congenitally defective carboxylating mechanism for the
vitamin-K-dependent clotting factors. Evaluation of the patient’s own hepatic
microsomes would be necessary to test this hypothesis, but this experiment is not
possible at the present time.
Nevertheless, it is clear that this unique patient has congenital deficiency in the
clotting activity of factors II, VII, IX, and X, and to our knowledge she is the only
such patient reported with this defect. We have not yet tested this patient for the
presence of protein 5, which is anoth. :jtamin..K�dependent factor found in
humans. The function of this protei’ not yet known. The current studies
demonstrate that the defect in thic dtient is due to synthesis of clotting factors
lacking the normal compleme �n a’� gamma-carboxyglutamic acid residues. Since
carboxylation of glutamic a �sidues is dependent on vitamin K, it follows that
there is some type of interft �.nce in metabolism of vitamin K, either by defective
transport in the blood or by defective intrahepatic epoxidase, reductase, or
carboxylase systems. While the latter hypothesis may be more attractive in view of
recent knowledge, it is not yet possible to distinguish among these possibilities.
ACKNOWLEDGMENT
The authors acknowledge the helpful experiments of Su Chung and the technical assistance of P. B.
Soule. We wish to thank Dr. R. G. Hiskey for supplying synthetic monomeric gamma-carboxyglutamic
acid. We express our appreciation to T. H. Duncan for superb secretarial assistance.
REFERENCES
l . Suttie iW: Oral anticoagulant therapy: The protein purification from bovine plasma and
biosynthetic basis. Semin Hematol 14:365, 1977 preliminary characterization. i Biol Chem
2. Stenflo J: A new vitamin K-dependent 251:355, 1976
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
786 CHUNG ET AL.
3. Di Scipio 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, 1977
4. Suttie JW: Mechanism of action of vitamin
K: Demonstration of a liver precursor of
prothrombin.Science 179:192, 1973
5. Jones JP, Gardner EJ, Cooper TG, Olson
RE: Vitamin K-dependent carboxylation of
peptide-bound glutamate. J Biol Chem 252:7738,
I 977
6. Esmon CT, Suttie iW, Jackson CM: The
functional significance of vitamin K action.
Difference in phospholipid binding between
normal and abnormal prothrombin. J Biol Chem
250:6125, 1975
7. McMillan CW, Roberts HR: Congenital
combined deficiency of coagulation factors II,
VII, IX and X. N EngI i Med 274:1313, 1966
8. Chung KS, Reisner HM, Roberts HR:
Purification and characterization of human blood
clotting factor IX. Circulation 53-54 [Suppl Ill:
464, 1976 (abstract)
9. Chung KS, Madar DA, Goldsmith JC,
Kingdon HS, Roberts HR: Purification and char-
acterization of an abnormal factor IX (Christmas
factor) molecule: Factor IX Chapel Hill. J Clin
Invest 62:1078, 1978
10. Cuatrecasas P. Protein purification by
affinity chromatography. J Biol Chem 245:3059,
I970
1 1. Ouchterlony 0: Handbook of lmmunodif-
fusion and lmmunoelectrophoresis. Ann Arbor,
Mich. Arbor Science Publishers, 1968
12. Woods KR, Wang KT: Separation of
dansyl-amino acids by polyamide layer chroma-
tography. Biochim BiophysActa 133:369, 1967
13. Hauschka PV, Lian JB, Gallop PM: Direct
identification of the calcium-binding amino acid,
‘y-carboxyglutamate, in mineralized tissue. Proc
Natl Acad Sci USA 72:3925, 1975
14. Boggs NT, Gawley RE, Koehler KA,
Hiskey RG: Synthesis of Dl--y-carboxyglutamic
acid derivatives. i Org Chem 40:3850, I 975
15. Biggs R: Human Blood Coagulation,
Haemostasis and Thrombosis (ed 2). Oxford,
Blackwell Scientific, 1976, p 704
16. Barrow EM, Bullock WR, Graham JB: A
study of the carrier state for plasma thromboplas-
tin component (PTC, Christmas factor) deficien-
cy, utilizing a new assay procedure. i Lab Clin
Med 55:936, 1960
17. Hougie C, Barrow EM, Graham JB: Stuart
clotting defect. I. Segregation of an hereditary
hemorrhagic state from the heterogeneous group
heretofore called �‘stable factor” (SPCA, procon-
vertin, factor VII), deficiency. J Clin Invest
36:485, 1957
18. Soulier iP, Larrieu Mi: Etude analytique
des temps de Quick allong#{233}s. Dosages de
prothrombine, de proconvertine et de proacc#{233}l#{233}r-
inc. Vox Sang 23:459, 1952
19. Smith HP, Warner ED, Brinkhous KM:
Prothrombin deficiency and the bleeding tendency
in liver injury (chloroform intoxication). J Exp
Med 66:801, 1937
20. Soulier JP, Prou-Wartelle 0: Etude com-
parative des taux de cofacteur de Ia staphyloco-
agulase (CRF) et des taux de facteur II (pro-
thrombine) dans diverses conditions. Nouv Rev Fr
Hematol 6:623, 1966
21. Roberts HR. Grizzle iE, McLester WD,
Penick GD: Genetic variants of hemophilia B:
Detection by means of a specific PTC inhibitor. J
Clin Invest 47:360, 1968
22. Pike IM, Yount Wi, Puritz EM, Roberts
HR: Immunochemical characterization of a
monoclonal lgG4 human antibody to factor IX.
Blood4O:l, 1972
23. Fahey JL, McKelvey EM: Quantitative
determination of serum immunoglobulins in anti-
body agar plates. i Immunol 94:84, 1965
24. Laurell CB: Quantitative estimation of
proteins by electrophoresis in agarose gel contain-
ing antibodies. Anal Biochem 15:45, 1966
25. Weeke B: Crossed immunoelectrophoresis.
Scand i Immunol I [Suppl 2]:47, 1973
26. Ganrot P0: Crossed immunoelectrophore-
sis. Scand J Clin Lab Invest 29(Suppl l24J:39,
1972
27. Esnouf MP, Prowse CV: The gamma-
carboxyglutamic acid content of human and
bovine prothrombin following warfarin treatment.
Biochim Biophys Acta 490:47 1 , I 977
28. Lavergne iM, iosso F: Metabolism of
PIVKA II in man, in Hemker HC, Veltkamp JJ
(eds): Prothrombin and Related Coagulation
Factors. Leiden, Leiden University Press, 1975, p
I 83
29. Thompson AR: Factor IX antigen by
radioimmunoassay-abnormal factor IX protein
in patients on warfarin therapy and with hemo-
philia B. i Clin Invest 59:900, 1977
30. Stenflo i: Vitamin K and the biosynthesis
of prothrombin. II. Structural comparison of an
amino terminal fragment from normal and from
dicoumarol-induced bovine prothrombin. J Biol
Chem 248:6325, 1973
31 . Nelsestuen GL, Zytkovicz TH, Howard
JB: The mode of action of vitamin K. Identifica-
tion of ‘y-carboxyglutamic acid as a component of
prothrombin. i Biol Chem 249:6347, 1974
32. Magnusson 5, Sottrup-iensen L, Petersen
TE, Morris HR. Dell A: Primary structure of the
vitamin K-dependent part of prothrombin. FEBS
Lett44:l89, 1974
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
FACTORS II. VII, IX, AND X 787
33. Hauschka PV, Reid ML, Lian iB, Fried-
man PA, Gallop PM: Probable vitamin K-depen-
dence of ‘y-carboxyglutamate formation in bone.
Fed Proc 35:1354, 1976
34. Hauschka PV, Friedman PA, Traverso JP,
Gallop PM: Vitamin K-dependent ‘y-carboxyglu-
tamic acid formation by kidney microsomes in
vitro. Biochem Biophys Res Commun 71:1207,
I 976
35. Willingham AK, Matschiner iT: Changes
in phylloquinone epoxidase activity related to
prothrombin synthesis and microsomal clotting
activity in the rat. Biochem i 149:435, 1974
36. Bell BG, Caldwell PT, Holm EET:
Coumarins and the vitamin K-K epoxide cycle.
Lack of resistance to coumatetralyl in warfarin-
resistant rats. Biochem Pharmacol 25:1067, 1976
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom
1979 53: 776-787
KS Chung, A Bezeaud, JC Goldsmith, CW McMillan, D Menache and HR Roberts Congenital deficiency of blood clotting factors II, VII, IX, and X
http://www.bloodjournal.org/content/53/4/776.full.htmlUpdated information and services can be found at:
Articles on similar topics can be found in the following Blood collections
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:
Copyright 2011 by The American Society of Hematology; all rights reserved.Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of
For personal use only.on October 3, 2017. by guest www.bloodjournal.orgFrom