Download Functional Activityof Protein S Determined with

CLIN. CHEM. 35/8, 1644-1648 (1989)
Functional Activity of Protein S Determined with Use of Protein C Activated by Venom
Activator
Iaao Kobayashl,’ Norlhlko Amemlya,2Takeehl Endo,2 KohJlOkuyama,2 KohjI Tamura,1 and ShojI Kume2
A new screening procedure, an easy and specific assay for
determining functional Protein S activity, has been developed, with use of Protein C activated by venom activator
(Protac). Purified Protein C (100% amidolytic activity) was
activated by venom activator (6 units/mL). To a mixture of
100 L of Protein S-deficient plasma, 20 L of sample
plasma, 100 L of cephalin (Actin), and 20 L of activated
Protein C we added 100 L of 25 mmol/L CaCI2 solution and
measured the clotting time with a KC 10 coagulometer. The
functional Protein S activity correlated well with the concentrations of Protein S antigen measured by enzyme immunoassay and the Laurell rocket technique (r = 0.810 and
0.850, respectively) in normal subjects, patients with myocardial infarction undergoing warfarin therapy, and patients
with liver cirrhosis.
Addftlonal Keyphrasea: screening
. anticoagulant
therapy
liver cirrhosis . enzyme immunoassay
and electroimmunoassay compared
coagulation assays
myocardial infarction
The naturally
occurring anticoagulant
proteins
antithrombin Ill, Protein C, and Protein S function to inhibit
blood clotting. Several hereditary deficiencies and abnormalities among these proteins have been linked topredisposition to recurrent thrombosis (1). Protein S, a vitamin
K-dependent plasma protein, functions to inhibit the procoagulant cofactors Va and Villa by serving as a cofactor
for another plasma protein, activated Protein C (2-4).
Patients deficient in Protein C are subject to severe, recurrent venous thrombosis (5-7) because of their inability to
regulate intravascular
clot formation. Because Protein S is
required for the expression of the anticoagulant activity of
activated Protein C, a hereditary deficiency of it has also
been associated with recurrent thrombosis (8-10).
Several specific methods for determining
Protein S antigen have already been established (11-14). Two types of
Protein S deficiency are recognized. In type I, the patient
has little or no free Protein S but has significant quantities
of Protein S associated with C4b-binding protein. In type H,
the patient has no free Protein S and very little Protein S
associated with C4b-binding protein (11). In some cases of
Protein S deficiency the presence of an abnormal Protein S
molecule with decreased functional activity has been recognized (8, 15). Therefore,
a functional activity assay
method for Protein S that could be performed in a clinical
laboratory
would be useful.
Several assay methods specific for protein S functional
activity have been reported (15-17) but are unsuitable for
use as a screening assay. For example, they require specific
purified coagulation factors or are subject to nonspecific
‘2nd Department of Internal Medicine and 2Depent
of
Hematology of the Central Clinical Laboratory, Yamanashi Medical College, Shimokato 1110,Tainahocho, Nakakoxnagun, Yamanashi 409-38, Japan.
Received February 3, 1989; accepted May 8, 1989.
1644
CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989
activation of plasma by venom activator, as we have
already reported (18).
Now we have developed a new, simple, specific assay for
determining
Protein S functional activity by using Protein
C activated by venom activator (Protac), without interference from the nonspecific activation of plasma.
MaterIals and Methods
Materials
Purified
venom activator (“Protac,” one vial contains 3
goat IgG against human Protein S, and
Protein C were obtained from American
Diagnostica, Greenwich, CT. One unit of Protac activates
all the Protein C in 1 mL of normal plasma. Protein
S-deficient human plasma was obtained from Diagnostica
Stago, Asnieres, France. The concentration of Protein S
antigen in Protein S-deficient plasma was less than 2% of
the normal concentration in plasma as determined by
enzyme-linked immunosorbent assay (EusA). Cephalin
(Actin) and Owren’s barbital buffer (50 mmolJL, pH 7.35)
were purchased from American Dade, Miami, FL. Blood
was collected into 38 g/L trisodium citrate solution, in a 9:1
ratio of blood to anticoagulant. Platelet-poor plasma (ppp)
was prepared by centrifugation
at 4#{176}C
(2000 x g for 20
mm), then stored at -70 #{176}C
until use. Normal plasma
pooled from 26 ostensibly healthy normal subjects who had
not been taking any drugs was used as standard plasma.
Because there is yet no generally
purified
reference
standard for Protein 5, the activity in normal plasma is
usually used in defining the unit of biological and immunological activity of ProteinS, i.e., the amount of activity in
1 mL of average normal plasma (usually a pooled specimen
of 20 normal plasmas) (19). The advantage of this system of
units is that the severity of Protein S deficiency in a patient
can be easily related to normal on a percentage scale. The
Protein S activity in normal pooled plasma is stable for a
month at -70 #{176}C,
and the coefficient of variation (CV) for
Protein S activity under these conditions is <5%. We also
used citrated plasma samples from patients with myocardial infarction who were being treated with warfarin and
from patients with liver cirrhosis.
Substances used for interference experiments were as
follows. Biirubin (161 mgldL) was an aqueous standard
solution (Nippon Shoji Co., Ltd., Tokyo, Japan), hemoglobin was a hemolysate of normal erythrocytes washed with
isotonic saline solution and then lysed with distilled water,
lipid was a fat emulsion used for intravenous administration (Intrafat; Takeda Co., Ltd., Tokyo, Japan). Vitamin C
(L-ascorbic acid), heparin (heparin sodium salt, 198.3 units/
mg), and dipotassium
ethylenediaminetetraacetic
acid
(EDTA) were from Wako Pure Chemical Industries Co.,
Ltd., Tokyo, Japan.
units of activity),
purified human
Methods
Activated Protein C was prepared by mixing 100 L of
Protein C (diluted with Owren’s barbital buffer, pH 7.35, to
100% amidolytic
activity)
with 20 zL of venom activator
(diluted with distilled water to 5 unita/mL), then incubating at 37#{176}C
for 15 mm. We used an automated KC 10
coagulometer (Heinrich Amelung GmbH, Lieme, F.R.G.) to
measure the clotting time.
To prepare the standard curve, we assayed standards
having 0%, 6.25%, 12.5%, 25%, 50%, 100%, and 200% of the
amount of Protein S in pooled normal plasma. The pooled
normal plasma and the sample plasma were diluted with
Owren’s barbital buffer, pH 7.35, or Protein S-deficient
plasma. When sample plasma was diluted twofold, pooled
normal plasma without dilution thus contained twice the
concentration of Protein S activity, comparatively speaking.
The assay was performed as follows. To the well of the
KC 10 coagulometer add 20 pL of the standard or sample
plasma, and 100 zL of Protein S-deficient plasma. After a
1-min incubation at 37#{176}C,
add 100 p.L of cephalin. After 1
mm, add 10 zL of activated Protein C (69.7 g/mL).
Exactly 2 mm later, add 100 1zL of 25 mmol/L CaC12
solution and measure the clotting time.
We also assayed total Protein S antigen with an ELISA
method (Asserachrom ProteinS; Diagnostica Stago, Francoville, France) and the Laurell rocket technique (11),
using sample plasma. Protein S complexed to C4b-binding
protein was precipitated
with 3.75% polyethylene
glycol
8000 according to the report of Comp et al. (11) and free
Protein S antigen in the supernate was also assayed by the
ELISA method and the Laurell rocket technique. Protein C
amidolytic activity was measured by our method (18).
Interference experiments were done as follows: To one
volume of test substance add nine volumes of normal
plasma and measure Protein S functional activity of the
mixture after 30 mm of incubation at room temperature.
We computed the linear-regression analysis as the principal standardized component, using the method proposed by
Feldmann et al. (20).
Results
Figure 1 shows the influence of nonspecific activity of
venom activator on coagulation time in the assay system.
The clotting time of normal plasma was markedly prolonged in proportion to the concentration of venom activator present. In Protein S-deficient plasma the clotting time
was also slightly prolonged at high concentrations of
venom activator (2-4 units/mL). There is ordinarily some
#{149}
Normal
o
80
Protaim
plasma
S depleted
nonspecific activation of Protein S-deficient plasma by
venom activator. In our assay system the final concentration of venom activator was <1 unit/mL. Therefore, there
may have been no influence of nonspecific activation of
Protein S-deficient plasma by venom activator in this assay
system.
After mixing 10 pL of venom activator (6 units/mL) with
100 ,uL of Protein C (100% amidolytic
activity),
we studied
the activation of Protein C kinetically at 37#{176}C
(Figure 2).
The amidolytic activity of Protein C increased quickly,
reaching a maximum after 15 miii, and after 45 mm it
gradually
decreased. Therefore, we measured Protein S
activity from 15 to 45 miii after the activation of Protein C
by venom activator.
To determine the effect of incubation time, after adding
activated Protein C, on clotting time, we assayed normal
plasma, using incubation periods from 1 to 5 mm. Clotting
time was longest after a 2-mm incubation. Therefore, we
incubated the assay mixtures for 2 miii after adding activated Protein C.
The standard curve from 6.25% to 200% is shown in
Figure 3. Pooled normal plasma was diluted with Owren’s
barbital buffer or with Protein S-deficient plasma to obtain
the various Protein S concentrations.Both standard curves
are linear on a semilogarithm scale, but the slope is steeper
for the standard curve incorporating
data for standard
diluted with Protein S-deficient plasma. Therefore, we
usually used the latter standard curve but measured sample plasma without dilution.
We measured total Protein S antigen, using the standard
curve for pooled normal plasma, and the value was the
(LI Afmjn)
0
30
60
90
120
(mm)
Fig.2. Kinetics forvenom activator
actionon ProteinC activity
After
100 pL ofProtein
c (100% amidolytlc
activity)
was mixedwIth10 p1.of
venom activator (6 unfts/mL),
theamidolytic activity
of Protein C was measuredserially by a Idnetic
assay,at 405 nm
plasma
#{149}
Diluted
with
Buffer
o
with
Protein
(S)
300
Diluted
depleted
S
plasma
#{149}2O0
‘I,
01
20
100
Q
U
0
1
2
3
L
Protac (unlts/mL)
FIg.1. Effect of venom activator
(Protac)on coagulationtime of
normal plasma (#{149})
and ProteinS-depletedplasma(0)
To themixture of 100 1L of normal plasmaorProtein
S-deficient
plasma,100
uL of cephalin, and 20 pL of venom activator in variousconcentrationswe
addediOOpL of 25mmol/L.CaCI2and measured
theclotting
time(In seconds)
..-/25
6.25 12.5
50
100
Protein S activity
200(0/0)
Fig. 3. Standardcurvefor ProteinS activity from 6.25% to 200% of
normal
(S) = seconds
CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989 1645
Table 1. PrecisIon of the Assay
Mean
SD
Sampi.
Within-run
23.5
95.2
2.49
154.6
5.48
7.83
2.67
3.54
19.2
104.7
164.3
3.11
16.25
5.46
6.22
5.21
Low
Normal
High
Between-run
Low
Normal
High
1.84
3.79
n = 20 each.
percentage of normal for total ProteinS antigen in pooled
normal plasma. Free Protein S antigen was measured by
using the standard curve for free Protein S of poolednormal
plasma, and again the value was in terms of the percentage
of free Protein S antigen in pooled normal plasma.
Table 1 shows the precision, within-run and betweenrun.
Resultsof interference experiments are shown in Table
2. A biirubin final concentration of 26 mg/dL, >0.1 unit of
heparin per milliliter,
and EDTA all interfered with the
assay of Protein S functionalactivity.
Figure 4 summarizes assay results for plasma samples
from 31 normal subjects, 25 patients with liver cirrhosis,
and 15 warfarun-treated
patients with myocardial
infarction. Free Protein S antigen as measured by the Laurell
rocket technique
or by ELISA exhibited good correlation
with the Protein S functionalactivity (r = 0.850 and 0.810,
respectively). Values for Protein S functional activity and
antigen (total and free) and results for two casesof Protein
S deficiency are shown in Table 3.
DIscussIon
The naturally
occurring
anticoagulantproteins-Protein
C, ProteinS, and antithrombin Ill-inhibit
blood clotting,
and several hereditary
deficiencies and abnormalities
among these proteins have been linked to predisposition to
recurrent thrombosis (5, 8,21). The frequencies
of heterozygous Protein C and Protein S deficiencies detected by
total antigen concentrations in plasma of young patients
with venous thrombosis are 4% for type I ProteinC deficiency and 5% for type I Protein S deficiency (22). Many
previously reported Protein S deficiencies were detected
because of the decreased concentration of antigen. Protein
in plasma is present both in the free form, which is the
active Protein C cofactor, and in a functionally
inactive
form that is complexed to complement compound C4bbinding protein (23). These two forms of Protein S complicate diagnosis of Protein S deficiencies and abnormalities,
because ordinary immunoassays do not distinguish
between the free and bound forms (11). Most homozygous
Protein S-deficient patients lack the free Protein S form but
may have significant quantities of the bound inactive form
(11). But most heterozygotes have concentrations of total
Protein S antigen that are below the lower limit of the
normal range (24). Therefore, methods for assaying functional Protein S activity are needed, but existing ones are
not yet feasible in most laboratories.
Our assay system of functional Protein S activity
is
simple and specific, because measurement of Protein C
activity activated by venom activator (Protac) is easy, the
standard curve is practicably steep, and in addition the
functional activity
is not influenced by the nonspecific
activity of venom activator and shows good correlation with
Protein S antigen as measured by ELISA and the Laurell
rocket technique. Therefore, this assay method is useful in
clinical
laboratories
for screening assay of functional Protein S activity.
In normal subjects, functional Protein S activity was
102% (SD 21%) and values for total and free Protein S
antigen were 106% (SD 25%) and 106% (SD 20%), respectively. The ratio of Protein S functional activity
to the
Table 2. Interference ExperImentsa
Measured
Rscovsry
Substances
Bilirubin,mg/dL
Hemoglobin,
mg/dL
Lipid, %
Added
2
5
7
10
26
(%)
rate (%)
78
105
72
97
101
98
37
10
50
100
300
72
72
72
72
500
75
75
73
28
0.5
74
1.0
77
1.5
75
2.0
2.5
74
78
97
Materials
VIt C, mg/dL
Heparin,unltlmL
2
5
7
Measured
(%)
63
75
76
Recovery
rate (%)
85
101
102
10
30
75
74
101
Added
100
97
0.01
0.03
70
75
94
101
97
97
101
0.05
0.07
0.1
75
77
100
101
104
135
10
20
39
49
52
50
70
100
50
53
40
67
71
54
100
104
101
100
105
EDTA, mg/dL
66
100
Control
0
74
100
0
74
Control
Theunit of each substance Is the final concentrationof the mixture. Isotonic saline solution was added to the mixture insteadofsubstance
Incontrol.
1646
CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989
Table 3. Plasma Content of Protein S (PS) Functional ActIvity and Antigen In VarIous Subjects
Total PS
PS activity
n
31
Normal
antigen
Free PS antigen
102 ± 21
Free PSI
total PS
S
106 ± 25
Warfarintherapy
15
31
±
16
59 ± 13
Livercirrhosis
25
52
±
22
66 ± 29
106 ± 20 (E)
110±18(L)
51 ± 12(E)
48±17(L)
64 ± 18 (E)
1.00
0.85
0.97
63 ± 25 (L)
Protein S deficiency
a
2
37
15
±
78 ±
2
61 ±
52±
8(E)
4(L)
0.78
PSactivity!
free PS antigen
0.96
0.93
0.61
0.65
0.81
0.83
0.60
0.72
E, ELISA; L Laurell’s rockettechnique.
Activity
#{149}
Liver
150
0
o Normal
A Wararin
therapy
disease
0
0
0
0
0
100
%
c%
#{149}
a0
000.
‘S
‘
n-li
r-0
:9
#{149}I.
I
I
#{149} #{149}#{149}
50
n-li
A
.850
#{149}r-0.8i0
Pathol Lab Med 1988;112:28-36.
2. Walker FJ. Regulation of activated Protein C by a new protein.
J Biol Chem 1980;255:5521-4.
3. Walker FJ. Regulation of activated Protein C by Protein S. the
role of phospholipid in factor Va inactivation. J Biol Chem
1981;256:11128-31.
4. Suzuki K, Nishioka J, Hashimoto S. Regulation of activated
Pi-otein C by thrombin modified Protein S. J Biochem (Tokyo)
1983;94:699-705.
5. Griffin
JH, Evatt B, Zimmerman TS, Wideman C. Deficiencyof
Protein C in congenital thrombotic disease. J Clin Invest
1981;68:1370-3.
6. Comp PC, Nixon ER, Eamon CT. Determination of functional
levels of Protein C, an antithrombotic protein, using thrombin
150
100
50
100
150
0
(()
Laurell 50
thrombomodulin complex.Blood 1984;63:15-21.
ELISA
(‘Ia)
7. Broekmans AW, Veltkamp JJ, Bertina EM. Congenital Protein
Fig. 4. Correlationbetween ProteinS functional activity
and Protein
C
deficiency and venous thromboembolism: study of three Dutch
S antigen (free form) measured by Laurell rocket techniqueand
families. N Engl J Med 1983;309:340-4.
ELISA in normal subjects (0), patIents with myocardlal Infarction
undergoingwarfarintherapy(Lx),and patientswithlivercirrhosis(#{149})8. Comp PC, Esmon CT. Recurrent venous thromboembolism in
patients with a partial deficiencyof Protein S. N Engi J Med
1984;311:1525-8.
9. Broekmans AW, Bertina RW, Reinalda-Poot J, et al. Herediconcentrations of total or free forms of Protein S assayed by
taly ProteinS deficiencyand venousthromboembolism.A studyin
ELISA in normal subjects is nearly 1.0. In our patients with
three Dutch families. Thromb Haemost 1985;53:273-7.
myocardial
infarction
undergoing therapy with warfarin or
10. Kamiya T, Sugthara J, Ogata K, et al. Inherited deficiencyof
in those with liver cirrhosis, the decreases in Protein S
Protein Sin a Japanese family with recurrentvenous thrombosis:
functional activity and antigen concentration (total and
study of three generations. Blood 1986;67:406-10.
11. Comp PC, Doray D, Patton D, Esmon CT. An abnormal
free forms) are prominent. And the ratio of functional
plasma distribution of Protein S occurs in functional Protein S
activity to the free form of Protein S as assayed by ELISA
deficiency. Blood 1986;67:504-8.
decreased more prominently in patients undergoing ther12. BertinaEM, Von Wijngaarden A, Reinalden-PoolJ, Poort SR,
apy with warfarin than in patients with liver cirrhosis. But
BornVJJ. Determination of plasma Protein S, the proteincofactor
the ratio of the concentrations of free Protein S antigen to
of activated Protein C. Thromb Haemost 1985;53:268-72.
total Protein S antigen as assayed by ELISA is also nearly
13. Mannucci PM, Tripodi A, Bertina EM. ProteinS deficiency
associated
with juvenile arterial and venousthrombosis.Thromb
1.0. Therefore, in patients being treated with warfarin
Haemost 1986;55:440.
there would be decreased synthesis of the total and free
14. Bennhagen MJ, Holmberg R, DahlbOck B. Plasma concentraforms of Protein S antigen accompanying the production of
tionsof C4b-binding protein and vitamin K-dependentProtein Sin
incompletely y-carboxylatedand dysfunctional free form of
term and preterm infants: low levels of protein S-C4b-binding
Protein S. These discrepancies between Protein S funcprotein complexes.Br J Haematol 1988;68:445-9.
tional activity and Protein S antigen were also recognized
15. CompPC, Nixon RB, Cooper MB, EsrnonCT. Familial Protein
S deficiency is associatedwith recurrent thrombosis. J ChinInvest
in other reports (15, 17).
1984;74:2082-8.
During a screening examination of healthy subjects, we
16. Van de Waart P, PreissnerKT, Bechtold JR, Muller Berghans
found two cases of Protein S deficiency. Both cases were
G. A functional test for Protein S activity in plasma. Thromb Res
type ll, as based on a report by Comp et al. (11), and the
1987;48:427-37.
patients themselves had no thrombotic episodes nor did
17. Suzuki K, Nishioka J. Plasma Protein S activitymeasured
their families.
using Protac,a snake venom derived activator of Protein C.
Thromb Res 1988;49:241-51.
Evaluation of Protein S functional
activity
by a simple
18. Kobayashi I, Amemiya N, Endo T, et al. Amidolytic kinetic
and specific screening assay method, both in healthy subassayof Protein C by selectivespectrophotometryin a centrifugal
jects and patients with thrombosis, is quite useful and is
analyzer. Clin Chem 1988;34:2260-3.
urgently needed to treat such patients and to analyze the
19. Barrowcliffe TW. Standards and controls in assay of blood
mechanism and pathophysiology of thrombosis.
coagulation factors. In: Biggs R, Rizza CR, eds. Human blood
coagulation, haemost.asis and thrombosis, 3rd ed. Oxford-LondonReferences
Edinburgh-Boston-Palo Alto-Melbourne: Blackwell Scientific
1. High KA. Antithrornbin ifi, Protein C, and Protein S. Arch
1984:563-82.
p<0.001
p(0
I
.001
AS
CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989 1647
20. Feldmann U, Schneider B, KlinkersH, HaeckelR.A multivariate approach for the biometric comparison of analytical methods in
clinical chemistry. J Chin Chem Clin Biochem1981;19:121-30.
21. Thaler E, Lechner K. Antithrombin ifi deficiencyand thromboembolisni.Clin Hematol 1981:369-90.
22. Gladson CL, Scharrer I, Hach V, Beck K!!, Griffin JH. The
frequency oftype I heterozygousProtein Sand ProteinC deficiency
in 141 unrelated young patients with venous thrombosis. Thromb
1648
CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989
Haemost 1988;59:18-22.
23. DahlbOck B, Stenflo J. High molecular weight complex in
human plasma between vitamin K-dependentProtein S and complement component C4b-binding protein. Proc Natl Acad Sci USA
1981;78:2512-6.
24. Engesser L, Broekmans AW, Briet E, Brommer EJP, Bertina
EM. HereditaryProtein S deficiency:clinical manifestations. Ann
Intern Med 1987;106:677-82.