(~) ELSEVIER Ann. Inst. Pasleur/Immunol. Paris 1985 1985, 136 D, 47-55

S A N D W I C H E N Z Y M E I M M U N O A S S A Y

OF H U M A N CHORIONIC G O N A D O T R O P I N

U S I N G P O L Y S T Y R E N E B E A D S AS SOLID S U P P O R T

by S. K. Gupta (1) (,), j. L. Guesdon (0, S. Avrameas e), and G. P. Talwar (1)

(1) National Institute o] Immunologg, Post Box No. 4922, New Dehli-llO 029, and

(8) Unit~ d'Immunocgtochimie, Institul Pasteur, 75724 Paris Cedex 15

SUMMARY

A solid-phase sandwich enzyme immunoassay for human chorionic gonadotropin (hCG) employing monoclonal antibodies directed against ~- and ~-subunits is described. Polystyrene beads used as solid support were coated with monoclonal anti-~-hCG antibody. Monoclonal anti-~-hCG antibody was labelled with the enzyme, horseradish peroxidase. The assay could be performed by acc one-step ~, or ~ two-step ~, procedure. The sensitivity of the two-step procedure was 7.8 ng hCG/ml with a linear range up to 500 ng hCG/ml. The one-step procedure had a measurement range between 15.6 and 125 ng hCG/ml. Within the limits of measuring ranges, the two procedures had similar accuracy, the correlation coefficient being 0.987. Twenty-one urine samples from pregnant women were ana- lysed by the two-step procedure and also by radioimmunoassay. The results were comparable, with r=0.956.

KEY-WORDS: Pregnancy, Chorionic gonadotropin, Monoclonal anti- body ; Sandwich EIA.

Human chorionic gonadotropin (hCG) is normally a product of tropho- blast and is detectable in circulation within 8 to 10 days post-fertiliza- tion [6, 12]. Studies on human embryos cultured in vitro demonstrated tha t the synthesis of hormone starts as early as 170 h after fertilization, e. g. in preimplantation embryos [7]. Its level during the first week after the expected menstrual period, i. e. the third week of pregnancy, is around

M a n u s c r i t re~u le 21 mar s 1985, aceept6 le 12 ju i l le t 1985.

(*) To w h o m cor respondence shou ld be addressed.

48 S . K . GUPTA AND COLL.

0.3-0.4 IU/ml of blood [5], with a doubling t ime of 1.7 to 2 days in early gestat ion [3]. I t has t radi t ional ly been used as an index of pregnancy and is the basis of all available bio- and immunoassays. Besides pregnancy, the synthesis of this hormone is observed in patients with trophoblast tumours and, to a variable degree, in a var ie ty of non-trophoblast tumours [2, 4, 9].

Radioimmunoassay (RIA) based on ~-hCG has generally been exten- sively employed and gives satisfactory sensitivity and specificity. In the quest for alternate markers to radiolabelled tracers avoiding radiation hazards and limited shelf life, a number of enzyme- and fluoroimmuno- assays have been reported [11, 13]. In this communication, we report a sandwich enzyme immunoassay for quant i ta t ion of hCG using polystyrene beads as solid support and two monoclonal antibodies. The method is amenable to both quali tat ive and quant i ta t ive use.

M A T E R I A L S AND METHODS

Reagents.

Highly purified hCG (13,000 IU/mg) calibrated against the second International Standard for hCG was made available by Dr Y. Y. Tsong of the Population Council, New York, USA. Horseradish peroxidase, grade I (RZ=3.0) was obtained from Boehringer (Mannheim, FRG). Tween-20 was from Merck-Schuchardt (Darmstadt, FRG), bovine serum albumin (BSA) from Povite (Amsterdam, the Netherlands), 0-phenylenediamine from Sigma, (St. Louis, MO), 25% aqueous solution of glutar- aldehyde from TAAB Laboratories (Reading, U. K.) and glycine, hydrogen peroxide (30%) and other chemicals were from Prolabo (Paris). Polystyrene beads (6.5 mm diameter) were purchased from Precision Plastic Balls, Inc., Chicago, IL.

Urine samples of pregnant women were obtained from Dr A. Funes, Laboratoire CERBA, 95005, Saint-Ouen l'Aumone, France and the laboratory of the (( Fonda- tion de Recherche en Hormonologie ~), 94260, Fresnes, France.

5lonoclonal antibodies.

Two different mouse monoclonal antibodies were used in the present studies. One, specific for hCG, was a product of the hybrid cell clone P3Wso grown as ascites in inbred BALB/c mice as reported earlier [8]. Monoclonal antibody (P3Wso) was an IgG1 kappa; it recognized a conformation common to the native hormone and the ~-subunit of hCG, but had very low reactivity (less than 1%) with the a-sub- unit of hCG and hLH (LER-960). It was devoid of a reaction with carboxy terminal peptides of ~-hCG ranging from 93 to 145 amino acid residues and other pituitary hormones, e. g. FSH and TSH [14]. It had high affinity for hCG with an association constant of K a = 3 • 101~ litres/mole [14]. The second monoclonal antibody which was used to prepare a conjugate with the enzyme was a product of the hybrid cell clone P~23. Monoclonal antibody P~23 recognized ~-hCG, hCG and pituitary gonado- tropins, i. e. hLH and FSH. However, no binding was observed with ~-hCG and carboxy terminal peptides of ~-hCG (amino acid 101-145). The association constant

BSA = bovine serum albumin. hCG ~ human chorionic gonadotropin. PBS = phosphate-buffered saline.

SANDWICH EIA OF CHORIONIC GONADOTROPIN 49

(Ka) of the product of hybrid cell clone P223 for binding with ~-hCG was 1.3 • 1091/ mole [10].

The hybrid cells were grown as ascites in the intraperitoneal cavity of BALB [c mice primed with Pristane (Aldrich Chemical Co., USA). Ascites fluid tapped from the intraperitoneal cavity was made cell-free by centrifugation at 800 g for 15 rain at 4 ~ C and then heat-inactivated at 56 ~ C for 30 rain and centrifuged at 15,000 y to remove debris. Immunoglobulins were obtained from the ascites by precipita- tion with ammonium sulphate at 40% saturation, followed by dialysis in phos- phate-buffered saline (PBS), pH 7.4 (0.01M phosphate, 0.15 M NaC1).

Preparation o/the enzyme-antibody conjugate. Monoclonal anti-~-hCG antibody (P92~) (5 mg) was coupled with 10 mg of

horseradish peroxidase by the one-step glutaraldehyde procedure [1]. The resulting conjugate was centrifuged to remove any precipitate, dialysed against PBS over- night at 4 ~ C, mixed with an equal volume of bidistilled glycerol and stored at --20 ~ C.

Coating o/polystyrene beads. Polystyrene beads were rinsed several times with double-distilled water and

covered with a known volmne of PBS. For coating, monoclonal anti-[~-hCG anti- body (P~Wso) diluted in PBS was added dropwise at a final concentration of 1, 5, 10 or 100 ~g immunoglobulin/ml and mixed immediately by continuous stirring of the vessels containing polystyrene beads. Coating was performed for 1 h at 37 ~ C and then for 18 h at 4 ~ C. Coated beads were stored at 4 ~ C in PBS containing 0.1% sodium azide.

Quantitation o/hCG by sandwich enzyme immunoassay. Polystyrene beads coated with monoclonal anti-[~-hCG antibody (P.~Wso) were

washed 3 times with PBS-Tween (PBS containing 0.1% Tween-20) before using. The assay was performed in duplicate in 12 • 75 mm glass tubes. Each tube received one antibody-coated bead and 0.2 ml of hCG standards prepared in PBS-BSA (PBS containing 0.1% Tween-20~-1% BSA) or urine samples diluted 1/1 in PBS- BSA. The tubes were incubated for 1 h at 37 ~ C and subsequently washed with PBS-Tween. The tubes were filled with 0.2 ml of the monoclonal anti-~-hCG anti- body/enzyme conjugate diluted in PBS-BSA to contain 1 ~g anti-~-hCG antibody per ml. After incubating for 1 h at 37 ~ C, the beads were washed three times by filling the tubes with PBS-Tween. Peroxidase activity was measured by adding 0.2 ml of substrate solution (1 mg 0-phenylenediamine per ml of 0.05 M citrate buffer, pH 5.5, containing 0.06% H~O~) per tube. The tubes were incubated in the dark for 1 to 10 rain. The enzyme reaction was stopped by transferring the reaction product into another set of tubes containing 50 ~1 of 6 N HC1 and absorbance was read at 492 nm.

To reduce the operational steps and time, the assay can also be performed in one step by co-incubating 100 ~1 of hCG standards or urine samples with 100 al of ant ibody/enzyme conjugate in one step for 1 h at 370 C followed by washing with PBS-Tween and subsequent quantitation of enzyme activity.

R E S U L T S

P o l y s t y r e n e beads were coa ted wi th d i f ferent concen t ra t ions of mono- clonal ant i -~-hCG a n t i b o d y as descr ibed in ~ Mater ia ls and Me thods ~. T h e r igh t c o n c e n t r a t i o n of the monoc lona l a n t i b o d y for coa t ing to ob ta in

Ann. Inst. Pasteur/Imrnunol., 136 D, n ~ 1, 1985. 4

50 S. K. GUPTA AND COLL.

beads usable in sandwich enzyme immunoassay for hCG was determined by experiments reported in figure 1. hCG bound to the solid surface was revealed in these assays by addit ion of monoclonal anti-~-hCG ant ibody/ enzyme conjugate. Polystyrene beads coated with 5 or 10 ~g monoclonal anti-~-hCG antibody per ml gave opt imum absorbance as a function of hCG concentrat ion above the detection limits of the assay (fig. 1). Polys ty- rene beads coated with either lower (1 ~g/ml) or higher (100 ~g/ml) amounts of monoclonal ant ibody did not give satisfactory absorbance.

. 6 - e ~ l pg/ml o ~ 5 . // II1~ 10 I t

�9 ~ 1 0 0 I I

E r

. ~ , 4 -

o

] i I' 1 10 1)00 2 5 0

CONCENT RAT ION OF hCG (ng/ml

FIG. 1. - - Optimization o/coaling o/the polyslyrene beads with the monoclonal anti-~-hCG antibodg (P~Wso).

Absorbance ob ta ined wi th hCG s t anda rds (1-250 ng hCG/ml) as a func t ion of monoc lona l a n t i b o d y c o n c e n t r a t i o n used in the coa t ing p rocedure is represented .

Figure 2 shows the typical dose-response curve obtained by using polystyrene beads coated with 10 ~g monoclonal anti-~-hCG ant ibody per millilitre. The detection limit of the assay was 7.8 ng hCG per ml in the case of the two-step method and 15.6 ng hCG per ml for the one-step method. The detection limit was defined as two s tandard deviations of the corresponding blank value. However, the shapes of the standard curves were different in the two cases. In the two-step assay, there was a gradual increase in absorbance from 7.8 ng to 500 ng hCG per ml to reach a plateau In the one-step procedure, there was an increase in absorbance from 7.8 to 125 ng hCG per ml, after which it decreased with increasing hCG concen- trations. Forty-seven urine samples obtained from pregnant women of different gestation age were analysed for quant i ta t ion of hCG by either

SANDWICH EIA OF CHORIONIC GONADOTROPIN 51

.6-.

~t

Ill 0 Z ,< r n-

O~ �9 2""~

:" ...- "~

�9 " . 4 . ~ %:

- : ",.

. . "..... " / x

�9 * : .

/ .~ ./,2,~ ~176

**~176

...o.;'." . . . . . . . A �9

~. .~176176

J I f 10 100 1000 CONCENTRATION OF hCG (ng/ml)

FIa. 2. - - Standard curves/or hCG in ttvo-step and one-step sandwich enzyme immunoassay.

Curve (e ) was for the two-s tep procedure where polys tyrene balls previously coaled wi th mono- clonal anti-~-hCG ant ibody were was first incubated wi th hCG standards followed by washing wi th PBS-Tween and subsequen t incubat ion wi th monoclonal anti-e-hCG ant ibody/enzyme eonjugate. Curve (&) was the one-step method where hCG standards and monoclonal anti-e hCG ant ibody/enzyme conjugate were eo-incubated at the same time.

two- or one-step sandwich enzyme immunoassay. A good agreement was found between the values obtained using the two assays, with a cor- relation coefficient of 0.987 (fig. 3). In addition, 21 urine samples were also analysed by two-step sandwich enzyme immunoassay and the values compared to those obtained independently by radioimmunoassay (RIA) i~ another laboratory, the ~ Fondation de Recherche en Hormonologie ~, 94260 Fresnes, France, using the Second International Standard (WHO) of hCG. Values of hCG obtained by enzyme immunoassay had a good cor- relation coefficient (r=0.956) with the values obtained by RIA (fig. 4).

D I S C U S S I O N

A sandwich enzyme immunoassay for hCG using polystyrene beads as solid support has been described. By using purified monoclonal anti-F- hCG antibody (P3Wso) for coating the solid phase, very high concentrations of the antibody can be adsorbed on polystyrene beads, which enables sufficient antibody-antigen binding in one hour. Monoelonal antibodies of uniform characteristics can be obtained in large amounts, thereby ensuring

52 S. K. GUPTA AND COLL.

u

100-

, f lo4

" " Z ' J "

.~ ,, g l

FIa. 3. - - Comparison o/ hCG concentration in urine samples o/ preg1~anl womeJz obtained bg hvo-step ( X ) and one-step ( Y ) sandwich enz!tme immunoassatj.

Values are expressed as ]U /ml .

the availability of monoclonal-antibody-based products with consistent properties on a large scale. Polystyrene beads coated with monoelonal antibody and stored at 40 C, when tested for up to 6 months, were found to be stable without any loss in the sensitivity of the assay. The advantage of using polystyrene beads over a 96-well microtitration plate is the ease in handling individual samples.

The sensitivity of the method is adequate. From the data of Braun- stein el al. [5], hCG in the amount of 7.8-15.6 ng/ml were observed at 4-4.5 weeks after the last menstrual period which, in a way, indicates the earliest period at which pregnancy will be detectable by this method. In the event that a suspected ease is found to be negative by this assay, a repeat after 3-7 days would provide confirmatory evidence, as the hor- mone levels rise rapidly during early gestation.

The assay can be performed either in one step or in two steps. The choice between the one-step and two-step assay depends on its application. The one-step method is more rapid, but the standard curve undergoes an inflection at high concentrations of hCG. This observation retleets a situa- tion in which the monoclonal anti-~-hCG antibody/enzyme conjugate increasingly binds to an excess of antigen (hCG), which itself cannot bind to the saturated solid phase. However, for qualitative purposes it may not make much difference and a single-step procedure would be largely

SANDWICH EIA OF CHORIONIC GONADOTROPIN 53

Y

100-

10-

n =21 ~ 4 ~ Y =.823 X +.254 r =.956

! 10

FIG. 4. - - Comparison o/ 21 urine hCG obtained bg two-step sandwich enzyme immunoassag (X ) and radioimmunoassag ( Y ).

Values are expressed as I U / m l .

~,x 100

utilisable as a correct index of pregnancy or hCG-synthesizing tumour up to a concentration of 1 000 ng hCG/ml. For quantitative purposes, however, it is advised to analyse a test sample suspected to contain high levels of hCG at 1/10 and 1/100 dilutions.

RESUME

D O S A G E I M M U N O E N Z Y M A T I Q U E

DE LA G O N A D O T R O P I N E C H O R I O N I Q U E H U M A I N E

U T I L I S A N T DES B I L L E S DE P O L Y S T Y R I ~ N E COMME S U P P O R T S O L I D E

Nous d~crivons un dosage enzymoimmunologique de la gonadotropine chorionique humaine (hCG) employant des anticorps monoclonaux dirig~s contre les sous-unit6s ~ et ~. Nous utilisons comme immunoadsorbant des billes de polystyrene recouvertes par un anticorps monoclonal anti-F- hCG coupl6 ~ la peroxydase. Le dosage peut ~tre r6alis~ en suivant ua proc~d6 en une ou deux 6tapes. Le seuil de d6tection du proc6d~ en deux 6tapes est ~gal ~ 7,8 ng hCG/ml avec une gamme de dosage efficace jus- qu'a 500 ng hCG/ml.

54 S . K . GUPTA AND COLL.

Le proc~d~ en une ~tape est utilisable pour des concentrations d 'hCG comprises entre 15,6 et 125 ng/ml. Entre ces limites, les deux proc6d~s ont une prScision similaire, et le coefficient de correlation est ~gal h 0,987. Des urines de femmes enceintes (21 pr~l~vements) ont 6t~ analys~es simul- tan~ment par le proc~d~ en deux ~tapes et par radioimmunologie quanti- t a t ive : le coefficient de corr61ation obtenu est ~gal h 0,956.

MOTS-CLI~S : Gravidit6, Gonadotropine chorionique, Anticorps mono- clonal ; IEA-sandwich.

ACKNOWLEDGEMENTS

This work benefitted from the collaborative programmes between the Institut Pasteur, Paris and the National Institute of Immunology, New Delhi.

REFERENCES

[1] AVRAMEAS, S., Coupling of enzymes to proteins with glutaraldehyde: use of the conjugate for detection of antigens and antibodies. Immunochem., 1969, 6, 43-53.

[2] BAGSHAWE, K. D., Marker proteins as indicator of tumour response to therapy, Brit. J. Cancer, 1980, 41, 186-190.

[3] BRAUNSTEIN, G. D., GRODIN, J. M., VAITUKAITIS, J. • ROSS, G. T., Secretory rates of human chorionic gonadotropin by normal trophoblast. Amer. d. Obstet. Gyn., 1973, 115, 447-457.

[4] BRAUNSTEIN, G. D., VAITUKAITIS, J. L., CARBONE, P. P. & Ross, G. T., Ectopic production of human chorionic gonadotropin by neoplasms. Ann. int. Med., 1973, 78, 39-45.

[5] BRAUNSTEIN, G. D., RASOR, J., ADLER, D., DANZER, H. • WADE, M. E., Serum human chorionic gonadotropin levels throughout normal pre- gnancy, Amer. J. Obstet. Gyn., 1976, 126, 678-681.

[6] CATT, K. J., DUFAU, M. L. & VAITUKAITIS, J. L., Appearance of hCG in pregnancy plasma following the initiation of implantation of the blasto- cyst. J. clin. Endocrinol. Metab., 1975, 40, 537-540.

[7] FISHEL, S. B., EDWARDS, R. G. & EVANS, C. J., Human chorionic gonadotro- pin secreted by preimplantation embryos cultured in vitro. Science (Wash.), 1984, 223, 816-818.

[8] GUPTA, S. K. ~r TALWAR, G. P., Development of hybridomas secreting anti- human chorionic gonadotropin antibodies. Indian J. exp. Biol., 1980, 18, 1361-1365.

[9] GUPTA, S. K., BUCKSHEE, K. & TALAWR, G. P., Ectopic secretion of human chorionic gonadotropin by gynaecological tumours. Indian J. reed. Res., 1984, 80, 189-195.

[10] GUPTA, S. K., SINGH, O., KAUR, I. & TALWAR, G. P., Characteristics of mono- clonal anti-alpha human chorionic gonadotropin antibody. Indian J. reed. Res., 1985, 81, 281-285.

[11] PETTERSSON, K., SIITARI, H., HEMMILA, I., SOINI, E., LOVGREN, T., HANNI- NEN, V., TANNER, P. & STENMAN, U. H., Time-resolved fluoroimmuno- assay of human choriogonadotropin. Clin. Chem., 1983, 29, 60-64.

[12] SAXENA, B. B., HASAN, S. H., HAoun, F. & COLLWITZER, S. M., Radio-

SANDWICH EIA OF CHORIONIC GONADOTROPIN 55

receptor assay of human chorionic gonadotropin: Detection of early pregnancy. Science (Wash.), 1974, 184, 793-795.

[13] TALWAR, G. P., GAuR, A., SINGH, A. K. • GUPTA, S. K., Two simple and sensitive methods for detection of pregnancy and hCG synthesizing tumours amenable to both qualitative and quantitative assays. Indian J . reed. Res., 1983, 77, 231-238.

[14] TALWAR, G. P., GAUR, A., GUPTA, S. K., SINGH, A. K. & PAUL, S., A highly sensitive method for detection of human pregnancy and hCG synthe- sizing tumours in laboratory and field. In : Non-isotopic immunoassays and their applications (Ed. G. P. Talwar), Vikas Publishing House, 1983, India, p. 188-196.