A Monoclonal Antibody to Human Protein S Used ... - Semantic Scholar

CLIN. CHEM. 36/1, 43-46 (1990)

A Monoclonal Antibody to Human Protein S Used as the Capture Antibody for Measuring Total Protein S by Enzyme Immunoassay A. Krachmalnlcoft,’ S. Tombesi,2 C. Valsecchl,’

A. Albertlnl,2 and P. M. Mannuccl1

Measurement of Protein S in human plasma is clinically important because a deficiency of this protein, which functions as a cofactor of the naturally occurring anticoagulant activated Protein C, is a risk factor for venous thromboembolism. We describe a two-site, enzyme-linked immunosorbent assay (EusA) for measuring Protein S in which a monoclonal lgG directed against the calcium-independent conformation of Protein S is the capture antibody. The range of detection for the assay was 10 to 160 ng of Protein S per milliliter. The coefficients of variation were 4.6%-7.3% withinassay and 7.7%-1 0.1% between-assay. We compared this assay with an ELISA involving a polyclonal anti-Protein S rabbit lgG as capture antibody (I) and with Laurell’s electroimmunoassay (II) to measure ProteinS in plasmafrom32 normal subjects and 121 patients or indMduais expected to have low concentrations of total Protein S (full-term newborns, pregnant women after the 18th week of gestation, patients with disseminated intravascular coagulation or liver cirrhosis, patients receiving therapy with warfarin, and patients with congenital Protein S deficiency). In general, the results obtained with the monoclonal antibody-based ELISA correlated well with those from I (r = 0.94), less well with those from II (r = 0.86).

AdditionalKeyphrases:endogenous anticoagulants embolism

. p,’jfl

C

vitamin K

ease (14), during pregnancy (15, 16) or oral contraceptive intake (16,17), in the neonatal period (18-20), and during disseminated intravascular coagulation (DIC) (14). Because some of these clinical and physiological conditions are associated with an increased risk for thrombosis, information on the concentration of Protein S in plasma is important. Laurell’s electroinimunoassay (EIA) technique is not ideal for measuring this protein, which circulates in both free and complexed forms, even though with some methodological adjustments (longer electrophoretic runs at room temperature) the assay has been used for diagnostic purposes (11). We describe here an enzyme-linked immunosorbent assay (EU3A) for measuring total Protein S (free plus cornplexed)-i.e., the amount of gene product. In this assay we used as the capture antibody a monoclonal IgG that reacts both with free and complexed Protein S and is directed against the calcium-independent conformation of the protein; a polyclonal anti-Protein S IgG was used as detecting antibody. The concentrations in normal plasma and plasma from patients and individuals with congenital and acquired deficiencies of Protein S were compared with those obtained by another ELISA in which polyclonal anti-Protein S rabbit IgG is used both as capture and detecting antibodies, and with those obtained by Laurell’s EIA.

thrombo-

heritabledisorders

Protein S.is a single-chain, vitamin K-dependent protein (1) that functions as a cofactor of naturally occurring anticoagulant activated Protein C in proteolytic cleavage of coagulation Factors V and Vifi (2,3) and in enhancement of fibrinolysis (4). Protein S is synthesized and secreted by vascular endothelial cells (5), is contained in megakaryocytes and platelets (6), and circulates in plasma in two forms, which are in dynamic equilibrium. One is the free protein (approximately40% of total Protein S in normal plasma), which functions as the cofactor of activated Protein C. The other, with no cofactor activity, is the protein reversibly complexed to the C4b-binding protein (C4bBP), a component of the complement system (7,8). Complexing of Protein S with C4bBP is regulated by the law of mass action, and the interaction has high affinity (Kd = 70 mniollL). The physiological importance of Protein S as a protective mechanism against thrombosis is demonstrated by the increased risk of venous thromboembolism in individuals with a congenital deficiency of it (9-12). There are also acquired deficiencies of Protein S in patients being treated with warfarin (13, 14), in patients with liver dis-

‘A. Bianchi Bonomi Hemophilia and Thrombosis Center, and Institute of Internal Medicine, University of Milan, Milano, Italy. 2Depar’ent of Chemistry and Biochemistry, University of

Materials and Methods Plasma samples. Blood was sampled from 32 healthy volunteers (17 men, 15 women), 18 pregnant women (18-42 weeks of gestation), six patients with laboratory evidence of DIC secondary to malignancy or infections, 30 patients on stabilized treatment with warfarin (international normalized ratio, 2.5-3.5), and 19 patients with biopsy-proven hepatic cirrhosis. Cord-blood samples were collected from 31 healthy full-term newborns immediately after vaginal delivery. Blood samples were also collected from 17 patients with a lifelong history of venous thromboembolism and previous diagnoses of congenital deficiency of Protein S. To obtain these diagnoses, we measured concentrations of total Protein S antigen by EIA (see below for methodological details); free Protein S was measured by EIA in the supernatant fluid of plasma precipitated at 4#{176}C after 1 h of incubation with 35 g/L polyethylene glycol (PEG) 8000 (BDH, Poole, England); two-dimensional immunoelectrophoresis of Protein S was done exactly as described by Comp et al. (21); and C4bBP was measured by EIA with use of a commercial antiserum (Istituto Behring, Scoppito, Italy). Eight of these patients had the phenotype of Protein S deficiency characterized by low normal concentrations of total Protein S antigen and low concentrations of free Protein S antigen, as judged serniquantitatively by twodimensional immunoelectrophoresis and quantitatively by

Brescia. Address correspondence to: P. M. Mannucci, via Pace 9, 20122

Milano, Italy. Received June 27, 1989; accepted September 5, 1989.

3Nonstandard abbreviations: ELISA, enzyme-linked immunosorbent assay; EJA, electroimmunoassay; DIC, disseminated intravascular coagulation; and PEG, polyethylene glycol.

CLINICALCHEMISTRY, Vol. 36, No. 1, 1990 43

free Protein S antigen in PEG supernatant plasma (21). The remaining nine patients had the phenotype of Protein S deficiency characterized by the parallel reduction of free and total Protein S antigen (21). Platelet-poor plasma was obtained from blood anticoagulated with a tenth volume of 129 mmol/L trisodium citrate, then centrifuged at 4#{176}C for 15 mm at 2000 x g. Plasma was quick-frozen and stored at -70 #{176}C until used. The reference plasma consistedof a pool of 40 specimens of normal plasma stored as indicated above, with its Protein S content arbitrarily defined as 100%. Reagents. Protein S was purified from human plasma according to the method described by Bertina et al. (22). The monoclonal antibody to Protein S was an IgG directed towards the calcium-independent conformation of human Protein S, which reacts with similar affinity with both free and complexed Protein S. The monoclonal antibody recognizes thrombin-cleaved Protein S and acarboxylated or hypocarboxylated ProteinS, which is present in the plasma of patients who are being treated with oral anticoagulants (23). The polyclonal antibody to Protein S was obtained by immunizing New Zealand White rabbits with purified human Protein S (0.3-0.5 mg). For the first injection the antigen was emulsified in an equal volume of Freund’s complete adjuvant. Two booster injections with Freund’s incomplete adjuvant were administered at two-week intervals after the initial injection. Fifteen days after the second booster, blood was collected and the serum was obtained. Rabbit IgG was obtained by affinity chromatography of rabbit antiserum on Protein A-Sepharose (Pharmacia, Uppsala, Sweden). Affinity-purified anti-Protein S rabbit IgG was prepared by affinity chromatography on CNBractivated Sepharose 4B (Pharmacia) to which Protein S was coupled (final concentration, 3 mg/mL). IgG was also conjugated with horseradish peroxidase (EC 1.11.1.7; Sigma, St. Louis, MO) according to the method described by Nakane and Kawaoi (24) and stored in small aliquots at -20 #{176}C (final concentration, 200 g/mL). Protein S-depleted plasma was prepared by passing fresh normal plasma through a column of CNBr-activated Sepharose 4B (Pharmacia) to which anti-Protein S rabbit IgG (see below) was coupled at a final concentration of 3 mglmL. The Protein S-depleted plasma was immediately frozen and storedat -70 #{176}C until use. The percentage of residual Protein S antigen varied between 2% and 5%, as measured by polyclonal ELISA (see below). Protein S-depleted plasma contained about 20% of the original C4bBP content in untreated plasma and had normal quantities of other clotting factors. Methods. The ELA was done according to Laurell(25), with use of the rabbit antibody to ProteinS.Electrophoresis was run at a higher temperature(25 #{176}C) and longer than usual (24 h) to favor the dissociation of complexed Protein S (11). Polystyrene plates (Nunc, Roskilde, Denmark) were coated with either the monoclonal anti-Protein S IgG (monoclonal ELISA) or the polyclonal anti-Protein S rabbit IgG (polyclonal ELISA). We added, per well, 125 L of 50 mmol/L Tris hydrochloride, pH 9.0, containing 5 pg of IgG (monoclonal or polyclonal) per milliliter and incubated at 4#{176}C overnight. After washing three times with a 20 mmol/ L solution of Tris hydrochloride containing 1 mL of Tween 20 surfactant (Sigma) per liter, pH 7.5, we added 100 L of serial dilutions of test plasmas, in duplicate. The diluting buffer was the Ths hydrochloridefFween 20 mixture plus 1 g of bovine serum albumin per liter. This diluting buffer assaying

44

CLINICAL CHEMISTRY, Vol. 36, No. 1, 1990

was supplemented with 2 mmol of EDTA per liter for the monoclonal ELISA. A calibration curve was constructed from 100- to 3200-fold dilutions of the reference (pooled) plasma. After incubation overnight at 20 #{176}C and washing three times, 100 pL of horseradish peroxidase-conjugated anti-Protein S rabbit IgG (diluted 250-fold in 20 mmol/L Tris hydrochloride, pH 7.5, containing 1 g of bovine serum albumin per liter) was added. The plates were incubated for 30 mm at 20#{176}C and washed three times; then 100 /LL of a mixture containing 22 mmol of trisodium citrate and 50 mmol of sodium hydrogen orthophosphate per liter (pH 5.0) as well as 400 pg of the substrate, o-phenylenedianiine dihydrochloride (Sigma), and 400 L of concentrated (300 g/L) hydrogen peroxide per liter, was added to each well. The reaction was stopped after 10 mm by adding 100 L of 4 molIL sulfuric acid. Absorbance was measured at 486 nm with a microplate photometer (SLT 210; Kontron, Durham, NC). Results Figure 1 shows a typical dose-response curve for the monoclonal ELISA, with use of purified Protein S free of C4bBP. This curve was linear from 10 to 160 ng of Protein S per milliliter. Dose-response curves for normal plasma (Figure 1) or plasma from a patient receiving oral anticoagulant therapy (not shown) paralleled those for purified ProteinS, indicating that the assay is suitable for measurement of all forms of Protein S. To assess the degree of cross-reactivity of the monoclonal antibody with other vitamin-K-dependent proteins, several preparations of Protein S-immunodepleted plasma containing normal concentrations of other plasma proteins were tested. In these preparations, the Protein S concentration was 5% of normal plasma, indicating that the antibody did not crossreact with other vitamin-K-dependent proteins. To assess the reproducibility of the method, we assayed plasmas from a normal subject (A) and from a warfarin-treated patient with about half-normal Protein S concentrations (B), 10 times on different days and 10 times on the same day. The coefficients of variation were 7.3% and 4.6% within-assay; PLASMA

1:6400

1:3200

1:1600

DILUTION

1:800

1:400

1:200

1:100

-20

V

.7. l’Q

20

40 PROIEIN

V

.7

10

U

05

Z 4

0 025

80 S

160

-

320

6.0

CONCENTRAIION

Fig. 1. MonoclonalELISAdose-response curve for purified Protein S (lower horizontal axis, closed squares)and for normal plasma (upper horizontal axis, closed circles) Absorbance measured at 486 nm

10.1% and 7.7% between-assay (Table 1). Table 2 and Figure 2 show the mean (and SD) values and the distribution of Protein S concentrations for patients and individuals whose plasmas were tested in parallel with the three assays. Whatever the assay, the mean concentration for patients receiving oral anticoagulants, for those with congenital Protein S deficiency, for healthy newborns, and for pregnant women were significantly lower (P 0.05) fromresults by monoclonalandpolyclonal ELISA. C P = 0.002 vs rnonoclonaland polyclonalELISA (Student’s f-test).

A? a #{176}

-

Discussion In theory, functional assays for Protein S should be preferable to immunoassays for patient screening because of their potential ability to detect cases of dysfunctional Protein S. Their use, however, is hampered by poor standardization and susceptibility to artifacts (26). Hence, imniunoassays of Protein S are widely used for diagnostic purposes at present. Notwithstanding the fact that Protein S circulates in plasma both free and complexed to C4bBP, total Protein S antigen can be measured immunologically if the complexed form is dissociated from C4bBP. In the EIA, this is achieved by running the electrophoresis longer than usual and at room temperature, conditions favoring this dissociation. Despite these precautions, sometimes double

Table 1. WithIn-Assay and Between-Assay Reproducibility Mean (and Within assay A B Between assay

A B

SD), %a

115 (8.4) 54(2.5)

112(11.3) 49 (3.8)

c,

%

#{149}

C

0

1

F

S

Fig. 2. Distribution of Protein S concentrations (normal plasma = 100%) in (A) healthy subjects, ( subjectswithcongenitalProteinS deficiency, (C) pregnant women, (C) full-termnewborns,(E) patients with liver cirrhosis, (F) DIC, or (C) on stabilized warfarin treatment, studied in parallel by monoclonal ELISA(closed circles), polyclonal ELISA (closedsquares), and electroimmunoassay (closedtriangles) Horizontal bars Indicate lower limits (mean - 2 SD from the mean)for the different assays

“rockets” are seen after electrophoresis, indicating that Protein S is not completely dissociated from C4bBP. This hampers precise measurement of the height of the rockets and hence of total Protein S antigen. In more sensitive immunoassays such as the ELISA or the immunoradiometric assay, dissociation is achieved more consistently by highly diluting the samples and prolonging their incubation with the antibody (13, 16,22). The present ELISA is based on the use of a monoclonal IgG directed to the calcium-independent conformation of Pro-

46

tein S asto aallow capture antibody (23). The sensitive enough dilutions of plasma from assay 100-to is 1600-fold.

10.1

As can be calculated from the Kd for the binding of Protein S to C4bBP and from the concentrations of the two proteins in plasma (13), these dilutions are high enough to allow

#{149} Values are expressed as percentof meanvaluefora pooled specimenof normalplasma. A = plasma from a normal subject.B = plasma from a warfatin-treatedpatient.n = 10 each.

dissociation of complexed Protein S and to avoid interference in the assay caused by C4bBP binding. The use of the monoclonal IgG as capture antibody assured high speciCLINICAL CHEMISTRY, Vol.36, No. 1, 1990 45

and lack of cross-reactivity with other vitamin-Kdependent proteins, as was demonstrated by the very low concentrations of Protein S antigen found in immunodepleted plasma (in which all vitamin-K-dependent proteins except Protein S were about 100% of normal). Assay specificity is also demonstrated by the good correlation of values obtained for clinical samples by this method with those obtained by an ELISA in which a polyclonal IgG is used as capture antibody and with those obtained by ficity

Laurell’s

EIA technique.

In healthy subjects we found total Protein S antigen in percentages ranging from 53% to 152%, with a mean of 106%, in agreement with previously reported data (13). Also in agreement with previous data are the low concentrations of total Protein S antigen measured during the third trimester of pregnancy (15,16), in full-term newborns (18-20), in warfarin-treated individuals (13,14), and in patients with hepatic cirrhosis (14). There was a wide range in the small number (n = 6) of DIC patients (5%135%) who had different degrees of decompensation of the syndrome. However, the monoclonal ELISA was not much more accurate than EIA and was less accuratethan the polyclonal ELISA in identifying patients with congenital Protein S deficiency. With the monoclonal ELISA, only about two-thirds of deficient patients had concentrations of total Protein S antigen below the lower limit of the normal reference interval. This well-known limited diagnostic sensitivity of any assay for total Protein S antigen is related to the fact that some cases of congenital Protein S deficiency show normal or borderline concentrations of total Protein S antigen, contrasting with low values for free Protein S antigen, the Protein S complexed to C4bBP exceeding the free Protein S (21). Although the possibility was not directly explored in this study, our monoclonal ELISA assay of total Protein S can probably be used to measure free Protein S antigen, which remains in the supernate of plasma treated with PEG at 4#{176}C to precipitate complexed Protein S(27). Measurement of free Protein S antigen should circumvent the relatively poor diagnostic accuracy of immunoassays of total Protein S antigen. In conclusion, the monoclonal ELISA is suitable for measurement of total Protein S antigen, because it generally gives the same information as the polyclonal ELISA that is routinely used in our laboratory. A monoclonal capture antibody has advantages over the polyclonal antibody: invariability and unlimited availability. We thank Dr. Santica Marcovina (Hybridoma Laboratory, San Raffaele Hospital, Milano), who supplied the monoclonal antibody. This work was supported in part by a grant from CNR, Progetto Finalizzato Biotecnologie e Biostrumentazione; and by a grant from Programma Nazionale di Ricerca Ne! Settore dei Farmaci, Consorzio Antitrombotici.

References 1. DiScipio RG, Hermodson MA, Yates SG, Davie EW. A comparison of human prothrombin, Factor IX, FactorX and Protein S. Biochemistry 1977;16:698-706. 2. Walker FJ. The regulation of activated Protein C by a new protein: a possible function for bovine Protein S. J Biol Chem 1980255:5521-4. 3. Walker FJ. Protein S and the regulation of activated Protein C [Review]. Semin Thromb Haemoat 1984;10:131-8. 4. DeFouw NJ, Haverkate F, Bertina RM, Koopman T, Van

48

CLINICAL CHEMISTRY, Vol.36, No. 1, 1990

A. The cofactor roleof Protein Sin the acceleration of whole blood clot iysis. Blood 1986;67:1192-6. 5. Stern DM, Brett J, Harris K, Nawroth PP. Participation of endothelia! cellsin the Protein C-Protein S anticoagulant pathway: the synthesis and release of Protein S. J Cell Biol Wijngasrden

1986;102:1071-8. 6. Schwartz HP, Heeb MJ, Wencel-Drake J, Griffin JH. Identiilcation and quantitation of ProteinS in human platelets. Blood 1985;66:1452-5. 7. Dahlback B, Stenflo J. High molecular weight complex in human plasma between vitamin K-dependent Protein S and complement component C4b-binding protein. Proc Nat! Acad Sci USA 1981;78:2512-6. 8. Dahlback B. Purification of human C4b-binding protein and formation of its complex with vitamin K-dependent Protein S. Biochem J 1983;209:847-56. 9. CompPC, Nixon RR, Cooper MR, Esmon CT. Familial Protein S deficiency is associated with recurrent thrombosis. J Clin Invest 1984;74:2082-8. 10. Comp PC, Esmon CT. Recurrent venous thromboembolism in patients with a partial deficiency of Protein S. N Engl J Med 1984;311:1525-8. 11. Schwartz HP, Fischer P, Hopmeier P, Batard MA, Griffin JH. Plasma Protein S deficiency in familial thrombotic disease. Blood 1984;64:1297-300. 12. Broekmans AW, Bertina EM, Reinalda-Poot J, et al. Hereditary Protein S deficiency and venous thromboembolism. A study in three Dutch families. Thromb Haemost 1985;53:273-7. 13. Fair SD, Revak DJ. Quantitation of human Protein S in the plasma of normal and warfarin-treated individuals by radioimmunoassay. Thromb Res 1984;36:527-35. 14. D’Angelo A, ViganO-D’Angelo S, Esmon CT, Comp PC. Acquired deficiencies of Protein S. Protein S activity during oral anticoagulation, in liver disease, and in disseminated intravascular coagulation. J Clin Invest 1988;81:1445.-54. 15. Comp PC, Thurnau GR, Welsh J, Esmon CT. Functional and immunologic Protein S levels are decreased during pregnancy. Blood 1986;68:881-5. 16. MaIm J, Laurell M, Dahlback B. Changes in the plasma levels of vitamin K-dependent Protein C and S and of C4b-binding protein during pregnancy and oral contraception. Br J Haematol 1988;68:437-43. 17. Huisve!d IA, Hospers JEH, Meijers JCM, Starkenburg AE, Erich WBM, Bouma BN. Oral contraceptives reduce total Protein 5, but not free Protein S. Thromb Res 1987;45:109-14. 18. Maim J, Bennhagen R, Holmberg L, Dahlback B. Plasma concentrations of C4b-binding protein and vitamin K-dependent ProteinS in term and preterm infants: low levels of Protein S-C4b-binding protein complexes. Br J Haematol 1988;68:445-9. 19. Schwartz HP, Muntean W, Watzke H, Richter B, Griffin JH. Low total Protein S antigen but high Protein S activity due to decreased C4b-binding protein in neonates. Blood 1988;71:562-5. 20. Moalic P, Gruel Y, Body G, Foloppe P, Delahousse B, Leroy J. Levelsand plasma distribution of free and C4b-BP-bound Protein S human fetuses and full-term newborns. Thromb Res 1988;49:471-80. 21. Comp PC, Doray D, Patton D, Esmon CT. An abnormal plasma distribution of Protein S occurs in functional Protein S deficiency. Blood 1986;67:504-8. 22. Bertina RM, van Wijngaarden A, Reinalda-Poot J, Poort SE, Bom VJJ. Determination of plasma Protein S-the protein cofactor of activated Protein C. Thromb Haemost 1985;53:268-72. 23. Marcovina 5, Coppola R, Valsecchi C, Zoppo A, Geffi C, Mannucci PM. Monoc!onal antibodies directed to the calcium-free conformation of human Protein S. Thromb Haemost 1989, in press. 24. Nakane PK, Kawaoi A. Peroxidase-labeled antibody: a new method of conjugation. J Histochem Cytochem 1974;22:1084-91.

25. Laure!l CB. Quantitative estimation of proteins by electrophoresis in agarose gel containing antibodies. Anal Biochem 1966;15:45-52. 26. Mannucci PM, Tripodi A. Laboratory screening of inherited thrombotic syndromes [Review]. Thromb Haemost 1987;57:24751. 27. Woodhams BJ. The simultaneous measurements of total and free Protein S by ELISA. Thromb Res 1988;50:213-20.