CA1110165A - Double antibody immunoassay using enzyme-labelled antibody and solid phase antigen - Google Patents

Abstract

ABSTRACT
The invention relates to quantitative or qualitative assay of an antigen in a sample, in which one brings into contact in a liquid medium, an antigen which is selected from immunogens and haptens, and is insolubilized by being carried on an insoluble support or in the form of an insoluble aggregate, a first antibody which belongs to an immunoglobulin class and which is anti- to the insolubilized antigen, and an enzyme-labelled second antibody which is anti- to the immunoglobulin class of the first antibody and which is enzyme-labelled by a covalent linkage between the enzyme and the second antibody; separates the insoluble and soluble material; and determines the enzyme activity of the soluble or insoluble separated materials, thereby determining the presence or amount of antigen in the sample. The improvement conferred by the invention lies in purifying said first antibody to remove antibodies non-specific to the insolubilized antigen, purifying the second antibody to remove antibodies non-specific to antibodies of the immuno-globulin class to which the first antibody belongs, preparing a pre-formed complex of the purified first antibody and purified first antibody and purified enzyme-labelled second antibody, and reacting the insolubilized antigen and the sample with said pre-formed complex, whereby the second antibody becomes attached to the insoluble material in inverse relationship to the amount of the antigen in the sample.

Description

THIS INVENTION relates to enzyme-linked immunoassay and provides a novel method for the qualitative and quantitative determination of antigens, and reagents for use therein. The term "antigen" as used herei~ means not only substances which are capable by themselves of provoking the production of antibodies in animals, i.e. immunogens, but also substances, sometimes called haptens, which, after 'conjugation with a carrier molecule, become capable of provoking the production of and of reaction with specific antibodies.
There is a considerable amou~t of theoretical and practical interest in being able to detect and measure the presence of antigens. Since such suhstances are often of complex, and frequently unknown, chemical ~structure, and occur at very low concentrations in biological and other materials it is not possible to use conventional analytical techniques. As a result, in the past few years, a number of techniques have been developed based upon the antigen-antibody reaction, in which a component of the reaction is labelled with a radio-active tracer or, ~r example with an enzyme, so that the products o~ the reaction, or in some cases the unused reagent, can be detected and/or measured. While methods based on the use of radio-active labels have certain theoretical advantages, in partic~r the fact that with certain types of label, notably those utilislng tritium or 14C, the radio-active tracer it~elf does not interfere with the ~ .

: . . , . . : . ~ . .
.' :' ' , ~' '' " '''' " , , . ': ' ' ~, .. ~ .. ~ . ,. , . . ', .: . :.

antigen-antibody reaction, there are certain disadvantages .inherent in the use of this technique, in particula.r the need to provide the necessary apparatus to measure the radic-active tracer, to protect the worker from possible adverse effects from the ~adio-activity and to dispose of the radio-active waste after use. Many radioactive labels must be discarded at regular intervals due to the radiochemical decompositi~n of the radio-active isotope. Furthermore. it may be difficult to radio-label small molecular weight antigens and retain their essential antigenic properties. Considerable interest has, therefore, heen shown in methods in which the antigen-antibody reaction i.s followed by using a reagent in which an enzyme acts as tracer~ Such methods ar~ simple to operate using relatively simple equipment, and do not involve any health hazard. Nonetheless, many of the system~ proposed hitherto have suffered from the disadvantage tha-t it is necessary.foF their successful operation to prepare a con~ugate .
between the enzyme used as tracer and the antigen to be detected. The preparation of these conjugates i~ not always ea~y, because oE.the need to preserve the enzymic activity of the en2yme part of the conju~ate, and it is particularly inconvenient to have to prepare a separate conjugate for each individual antigen which it is proposed to detect. In the mekhod of the present invention, on the other hand, a single enzyme-containing conjugate can be used with a wide variety of antigens. Indeed, in principle, a single such conjugate ~ , ,. . .. .. .. . .
. ; . . ~ . . ..
' .' :: `'' :, ' "' ' ' . ' ' ' ' ,, : . ~.

. : , , : .

can be used with all antigens.
In one prior proposal (Engvall et al, J. Immunology 109 (1972) page 129~, antibodies have been determined by reactlon with antigens on an insoluble support followed by reaction of the supported antigen-antibody conjugate with an enzyme-labelled anti-immunoglobulin known to be capable of reacting with the antibody of the conjugate. In this method, -the amount of enzyme finally attached to the support depends upon the amount of anti-body in the sample under test, since the enzyme becomes linked to the support by a chain of which the antibody to be determined constitutes a distinct and essential link.
The assay method of the present invention is of a different type and is for determining an antigen. In a method of enzyme-linked immunoassay for determining the p~esence or amount of an antigen in a sample, in which one brings into contact in a li~uid medium, an antigen which is selected from immunogens and haptens, and is insolubilized by being carr~ed on an lnsoluble support or in the form of an insoluble aggregate, a first antibody which belongs to an immunoglobulin class and which is anti- to the insolubilized antigen, and an enzyme-labelled second antibody which is anti- to the immuno-globulin class of the first antibody and which is enzyme-labelled by a covalent llnkage between the enzyme and the second antibody; separates the insoluble and soluble material; and determines the enzyme activity of the soluble or . ,, ' ..
'~ '' , , ~
': ' , .

. :. .: : , - . . .. . .. . : , : . : , : . . :: ~

insoluble separated materials, thereby determining the presence or amount of antigen in the sample, the invention provides an improvement which comprises purifying said first antibody to remove antibodies non-specific to the insolubilized antigen, purifying the second antibody to remove antibodies non-specific to antibodies of the immunoglobulin class to which the first antibody belongs, preparing a pre-formed complex of the purified first antibody and purified enzyme-labelled second antibody, and reacting the insolubilized antigen and the sample with said pre-formed complex whereby the second antibody becomes attached -to the insoluble material in inverse relationship to the amount of the antigen in the sample.
The first antibody is able to bind to the antigen on the insoluble support. The assay method is only useful in relation to determinlng an antigen to which the first antibody can also be bound. For a quantitative assay the first antibody is preferably present in an amoun-t more than sufficient to react with all the antigen present in the sample but not more than suficient to react with all the antiyen present in the sample plus all the insolubilized antigen. This is essential in quantitative procedures involving reaction of the first antibody sequentially with the sample then ~ , S

. : : . , '. ` " ' : . ' ' ' .

'' ~' ' ' ''~' ' ' ' .

with the insolubi~æed antigen~ The second antibody is specific to antibodies of the immunoglobulin class to which the first antibody belongs e.g. all sheep IgG, rahbit IgG or rabbit IgM~ For example, if the first antibody is an immunoglobulin raised in a particular animal, e.g. a sheep, then the second antibody is an antibody against immunoglobulins of that species, regardless of their own antibody specificities, rai~ed in a different animal, e r g ~ a donkey.
The new method can easily be adapted for the detection and assay of a wide variety of antigens proper and haptens. These include serum and tissue proteins e.g.
a-~oetoprotein, ~2-haptoglobulin and carcinoembryonic antigen hormones, e.g. triiodothyronine~norepinephrine, thyroid stimulating hormone, follicle stimulating hormone, lutenizing hormone, insulin and thyroxine, steroids, e.g.
cortisol, progesterone, oestrone, oestradiol and testosterone, drugs, e.g. morphine, amphetamine, barbiturates~
diphenylhydantoin, methotrexate and gentamycin, vitamins and food contaminants, e.g. pyridoxal-5~phosphate and aflatoxins, herbicldes, insecticides and nitrosoureas can also be used.

..,,,i .
'. ~ : ' ' ' ' ' It will be noted that the nature of the enzyme labelled second antibody in no way depends upon the antigen to be detected. It is only necessary to raise the first antibody in an animal of the same species as that which r 5 provided the immunoglobulin used to provoke the produc~ion in another species of the second antibody~ There is, therefore, no need to provide a wide variety of enzyme labelled antibodies, and the difficulties inherent in linking enzymes to antibodies require to be solved only once.
10Any enzyme could be employed as a label on the second antibody. In the following description specific enz~mes are identified by reference to Florkin M. & Stotz, E.H. "Comprehensive Bi.ochemistry", Volume 13, 3rd edition, Elsevier Pub. Co., New York (1973). Oxidases such as horse 15radish peroxidase (1.11.1.7), hydrolytic enzymes such as alkaline phosphatase (3.1.3.1.) or dehydrogenases such as glucose 6-phosphate dehydrogenase (1.1.1,49) are particularly appropriate. Ideally an enzyme used as the label should have the following properties: (1) it is available relatively cheaply in high purity; (2~ it has a high activity per unit weight; (3~ it i8 easily water-soluble and stable in ordinary storage conditions, (4) it makes it possible to use an assay procedure which is simple, rapid, sensitive and cheap; (5~
it should be absent from biological fluids; and (6~ biological fluids should contain no substrates inhibitors, or activators ~,''''u~ .

. .
: . ': ' -.

u~

for it, (73 it should contain suitable, chemical groups for conjugation to a second antibody, enabling co~jugation to have only a minimal effect on both enzyme and antibody activity. ~-galactosidase (3.2.1.23) meets these requirements. Other possible but less preferable enzymes include glucose oxidase (1.1.3.4.), acetylcholineesterase (3.1.1.7.), glucoamylase (3.2.1.3.), lysozyme (3.2.1~17), glucose-6-phosphate dehydrogenase ~1.1.1.49), and malate dehydrogenase (1.1.1.37).
The enzyme used is prefexably coupled to the second antibody using a coupling procedure described below, which is applicable to any combination of antibody and enzyme provided that the latter contains a mercapto group or can be modified so as to contain such a group. However, other coupling methods can be used if required.
These include a) One step glutaraldehyde linkage in which glutaraldehyde is used to link proteins in a com~licated reaction to give heterogenous conjugates of hîgh molecular weight, (see Avrameas~ S~, Immunochemistry 6 ~3 (1969)).

-b) Two steps glutaraldehyde linkage, in which the enzyme horse20 radish peroxidase reacts with only one molecule of glutaraldehyde.
This limits enzyme-enzyme conjugation and is an improvement on the one step method~ EIowever, this method is probably not applicable to many enzymes besides horse radish peroxidase as these will react with more than one molecule glutaraldehyde. (see Avrameas. S and ~ernynck, Immunochemistry 8 1175 (1971)).
c~ Oxidation of saccharide residues in the enzyme followed by Schiff base formationO Horse radish peroxidase has several oligosaccharide groups,,and their oxidation to aldehyde yroups (using perDdate) that can react with amino groups is the basis of this method of linkage. Other enzymes .
.

., also contain oligosaccharide groups and can in principle also be used. However, large molecular weight conjugates are formed using this procedure. (Nakane P.K. ~ Kawaoki, A.
J. Hi~tochemO Cytochem. 22 1084 (1974)).
d) Dimaleimide linkage using N,N'-o-phenylenedimaleimide, after in-troduction of -SH groups into the antibody using

2-mercaptoethylamine which reduces pre-existing -S-S-, bridges to two -SH groups. (Kato et al European ~. Biochem.
_ 285 (1976)).
e) Other bifunctional reagents which may be used include toluene-2,4-diisocyanate, ~ -difluoro-m,m'-dinitrophenyl sulphone, l-cyclohexyl-3-(2-morpholinoethyl)-c~rbodiimide, but these generally give inferlor results to those described abo~e.
lS f) Linkage may also be effected with hetero-bifunctional reagents, i.e. reagents contalning two different functional groups in each molecule, e.g. m-maleimidobenzyl N-hydroxy-succinimide ester, which has been used in a two step reaction to couple insulin to ~-galactosidase. (T. Kitagawa -~
20 T. Aikawa J. Biochem~ 79 233-236 (1976)).

,~P, .
,~

.

. ~ .
~ . .

Because of the greater speed and convenience associated with the handling of relatively large quantities of antibodies, it is often convenient if the first and second antibodies are raised in relati~ely large animals such as S horse, donkeys or sheep, rather than in small animals, such as rats, mice, guinea pigs or hamsters though the use of these is not excluded. As already indicated, the first and the second antibody must be raised in different species of animals. It has been found convenient to raise the first antibody in sheep and the second antibody in a donkey, but any other pair of different species of animals can be used if preferred~
The possibility of interference by the second antibody of the antigen-first antibody reaction is lessened if the second antibody is raised not against the whole of the first antibody, but only against the Fc fragment of the immunoglobulin constituting the first antibody.
The second antibodies can be prepared essentially in known manner, but the preferred method is to pr:ime the animals, follow the production of antibodies .

, ' .: , '' " ~, ' .' . ~ ~ .' ' ' ',:
' . ' , ,,, , : ' '' :

~ ~ ~f~6 5 and to give booster dosages when serum antibody levels are past their peak. The blood usually contains maximum levels of antibodies shortly after boosting. When a suitably high level (or a maximum level) of anti-immunoglobulin has accumulated in the blood stream of the animal the latter is bled and the serum containing the anti-immunoglobulin is separated.
The second antibody is purified, preferably by incubating the donkey antiserum with an appropriate immuno-adsorbent, for example, Sepharose or other insoluble supportwhich has been activated by reaction first with a suitable bifunctional reagent , e.g~ bis~2,3-epoxypropoxy)butane, and then with a solution of the immunoglobulin ~or Fc fragment thereof) used in raising the second antihody. This gives an immuno-adsorbent which is specific for the second antibody, and the latter can be purified by adsorption on -the immuno-adsorbent followed by elution, before or after labelling with an enzyme.
To prepare the labellecl second antibody, it is pre~erred to treat the immuno-adsorbent having the seconcl antibody adsorbecl thereon with a reagent such as methyl-mercaptobutyrimidate hydrochloride so as to modify the amino groups in the antibody and introduce from 3 to 5 thiol groups in each antibody molecule. The second antibody having the incorporated mercapto groups is then eluted from the immuno-' ~ .
~ .

,: , . : ~ , .
- ~ ~ : :. .. .
. . ~ : .. .. , :
. : ... . :. .
. . , :

.

?~

absorbent and treated so as to bind the enzyme thereto.
Preferably it is treated with N,N'-o-phenylene-bis-maleimide and then reacted with purified ~-galactosidase normally derived from Escherichia coli. The enzyme-second antibody conjugate is finally purified by chromatography on a diethylaminoethylagarose gel or other suitable adsorbent so as to separate the conjugate from unreacted enzyme and unreacted second antibody.
The first antibody is raised in an appropriate animal using known techniques. Where, as is often the case, the antigen to be detected or assayed is a hapten of relatively simple chernical structure e.g. tri:iodothyronine, oestradiol, gent~mycin, phenytoin or rnorphine, it is not capable by itself of provoking the formation of the first antibody. It is therefore necessary to conjugate the antigen with a carrier molecule to produce a conjugated molecu1e capable of provoking the production of antibody. For this purpose, it is convenient to conjugate the antigen with a readily available protein foreign to the animal in which the first antibody is to be raised, e.g. bovine serum albumin (BSA). The conjugation may be carried out by reaction of the albumin or other protein with, for example, l-ethyl-3-(3-dimethylaminopropyl)-carbo-diimide, followed by reaction of the product with the antigen, which must, for this reaction, contain an amino or :

" ' "' ' 65i carboxylic group. Other known methods of conjugation can be used if desired, ideally about 3 to 30 molecules of the antigen should be incorporated into each molecule of the albumin or other high molecular weight protein.
The first antibody may be raised by injecting into a suitable animal, e.g. a sheep, the antigen or antigen conjugate, repeatedly over a period of weeks or months until a sufficiently high antibody titre with the required binding properties 'has accumulated in the animal' 8 blood stream. I'he animal is then hled and the anti serum separated.
The firs~ antibody is then purified, preferably by contact with the antigen against which the antibody was raised or a related compound on an insoluble support~ After absorption', the antibodies can be eluted, e.g. with guanidinium hydrochlo~ide solution at high or low pH with solutions of high salt concentration, or with non-aqueous solvents, and then dialysed against an appropriate buffer to r~nove the elutiny reagent~
The insoluble support used to carry the antigen or related compound both in the purification procedure just mentioned and in the method of the invention may be any suitable watex-insoluble material to which the antigen or related compou~d can conveniently be bound suEficiently to en~ure that it r~mains attached during the reaction~ involved in the purification or assay rnethod as the case may be~

_ 13 _ ': , ,. .. . ' . ~ . :
:

d~ ~ ,4 ~3A7 While it is po6sible to use a support which simply adsorbs the antigen, as in the system described by Engvall et al in the paper already mentioned, it i~ preferred to bind the antigen chemically -to the support, and this means that it is 5 necessary for the ~upport to contain functional groups through which the necessary reactions can take place. Supports including gla~s, nylon, polyacrylamide, cellulose or de~tran may be used. Certain types of inanunoadsor~ents containing hydroxyl groups can be reacted with 1,4-bis(2,3-epoxy-10 propoxy)butane and the product reacted with the antigen~Thi8 method has been found to be simple, safe, effective and . give a stable covalent linkage provided the antigen or related compound contains hydroxyl, 2mino or thiol groups. The requireme~ts for an immunoadsorbent for purifying antibodies 15 are not necessarlly tha same as those for the immunoadsorbent in the actual assay.
The method o~ the present invention can be carried out in a variety of different ways; The quantity of ~i~st antibody and hence label attached to the insoluble antige~
20 bears an inverse relationship to the amount of antigen in the sample to be assayed. The enzyme activity, either free or attached to the insoluble or supported antigen is measured. Some of t~e speci~ic ways of carrying out the new method are a~ ~ollows.

~ .
.~`~ ' .

~ ; , , .
' , In a one method, the first antibody is reacted with enzyme-labelled second antibody to form a comple~
which is then added to excess to the sample containing the antigen to be determined and incubated. The insoluble support carrying the antigen is then added, and, after incubation, the solid and liquid phases are separated.
Determination of the enzyme activity of the liquid phase gives a result which is directly proportional to the amount of antigen in the sample.
In another method, the complex of first and second antibodies, the sample containing the antigen to be determined, and the support carrying the antigen are incubated simultaneously and, after separation of the solid phase, the enzyme activity of the liquid phase is determined. Again, the result is directly proportional to the concentration of antigen in the sample.

- 15 ~
J'~
~' ~ ' ' .

.
. .

.

As already indicated, fox a ~uantitative assay by the first method above, th'e amount of first antibody used must be in excess of that required "' to react with all the antigen in the sample, and the support carrying the antigen must be added in a sufficien~ exce~s to ensure adsorpt:ion of all the first antibody which ha~ not reacted with antigen in the sample~
In practical l;erms, th~ ça~;iest, way o~ ensuring thi~ is by carrying out a se~iqs of detenminations with different conaent~atio~S of first antlbody in geometr.ic or arithmetic series. If ~hi~ i~ done, ~t is a simple mattex to find an a,ppropriate concentration of the fir3t antibody which enable~ an accurate a~say to be carried out.

C~ .
. . .

.

.

The words "Biogel" Sephadex", 'Sepharose" "Tween"
and "Whatman" used herein are trade marks.
The following Examples illustrate the invention. , Preparation of materialP, lo Producti n ar~ L~13llbL~lE_c~E~d ankibod~
(a) P eparation of Fc fragment of sheep immunoglobulin j ~he method employed for the preparation of sheep ~c fragment is based on that of Porter (RoR~ Porter, 1959, Biochem J. 73, ll9)o 320 mlO of sera were collected from 28 sheep. 150 ml~
of serum were incubated with stirring in an ice-bath with 30 gu of moist diethylaminoethyl (DEAE) - ~ 50 Sephadex previously equilibrated with 10 mM phosphate buffer at pH 6.5 After one hour the gel was filtered OIl a coarse sintered glasæ funnel and washed with 50 ml. of 10 mM phosphate buffer pH 6.5~ The filtrate plus washings were incubated once more with 30 gO of moist D~AE-Sephadex, filtered and washed with 100 ml~ of the phosphate bufferO q~e filtrate and washings (30~j mlO) were treated with 150 ml. of saturated (NH4)2 S4 soluti,on which were slowly added to the continuously stirred solution. The precipitate was collected and washed with 100 mlO of 20% ~NH4)2S04 solution~ e final precipitate was dissolved in 50 ml~ of 50 mM phosphate buffer of p~ 8~0 and dialy~ed ovarnight at 4Co, against 5 lo of distilled water. The solution was then freeze ~ried.
1~50 gu of ~reeze dried sheep Ig~ so obtained was ..
.: , . :, .
. :~ .
' . ' ' ' ~ '' ' -, ' ' : ' :

dissolved ln 50 ml. of O.lM Tris-HCl buffer at pH 7.2 containing 10 mM cysteine and 2mM EDTA. The bufer was prepared immediately before use. 0.6 ml. of a crystalline suspension of Papain (Sigma. twlce crystallised) in 50 mM-Acetate buffer at pH 4.5 was added. The solution wasincubated at 37C for 4 hours with occasional shaking~
0.222 g. of iodoacetamide was then added and the solution immediately applied to a column (80 x 5 cm.) of G.100 Sephadex equilibrated with 10 mM-Acetate buffer at pH 5.6 at 40C.
The second peak eluted from the column with 10 mM-Acetate buf~r at pH 5.6 contains a mixture of F~ and F b frac~ments. The first peak contains undigested IgG. The second peak was applied to a column (40 x 2.5 cm.) of carboxymethylcellulose (Whatman CM 32) equilibrated with 10 mM-Acetate buffer at pH 5.6 at 40C. A peak was eluted from the column on washing with 10 mM-Acetate buffer pH 5.6 and several minor peaks and a major peak were eluted when a gradient from 10 mM- to 500 mM-Acetate pH 5.6 was applied (1000 ml. in each compartment of a gradient mixer). The major peak was collected and rechromatographed on the carboxy-methylcellulose column. This time a gradient from 10 mM-to 300 I~M~Acetate buffer at pH 5.6 with 1000 ml. in a mixing vessel was applied. A single peak was eluted. This was collected, dialysed exhaustively against distilled water, and ~reeze-dried.

,.

'' , ' ~' .' .

b) Preparation of second antibody (donkey anti-sheep imrnunoqlobulin~ _ A stable emulsion was prepared by mixing 1 part of saline containing sheep Fc fragment (0.75 mg/ml) and Bacillus Calmette-Guerin (1.5 mg/ml) with 2 parts of Marcol 52 adjuvant. Six intramuscular injections (1 ml~ of the emulsion were given in the lower leg. The production of donkey anti-sheep antibodies was monitored and the animal was ready for boosting after two months. An emulsion consisting of 2 parts adjuvant to 1 part saline (sheep Fc 0.75 mg/ml) was injected intramuscularly into six sites (0.5 ml) in the lower legs. The bleeds ten days after and for the next month contained high level of antibodies.

c) Preparation of immunoadsorbent for purification of donkey anti-s eep irnmunoqlobu~lin _ _ Sepharose 4B was activated with 1,4-bis-(2,3-epoxy-propoxy)butane by incubating equal volumes of the gel, thebisepoxide and 0.6M NaOH containing 2 mg. per ml. sodium borohydride with continuous stirring at 25C., or 8 hours, and then washed. Sheep IgG was prepared from a pool of normal serum by precipitation with 33% ammonium sulphate solution. A solution of the sheep IgG in O.lM NaHC03 buffer at pH 10 was added using 10 mg. oE sheep IgG per ml.
of gel. The activated gel was incubated with the IgG for 48 hours and the imrnunoadsorbent finally washed and treated with O.lM ethanolarnine at pH 10 for 24 hours.

`- - 19 -.1 , , ~ , : . :
- . , , , : - :

.

d) Purificatlon and labelling of donkey anti-sheep second antibod~
Donkey anti-sheep immunoglobulin was purified by incubating donkey anti-serum (prepared as described above) with the immunoadsorbent. Thiol yroups were incorporated into the immunoglobulin while it was adsorbed in the immuno-adsorbent using methyl-4-mercapto-butyrimidate/HCl to modify the amino groups of the immunoglobulin. The extent of incorporation was limited to give between 3 and 5 thiol -~
groups per globulin molecule.
The i~nunoadsorbent antibody complex was washed and suspended in O.lM N-ethylmorpholine/~ICl ak pH 7.5 at a concentration of 1 ml~ immunoadsorbent in a total volume of ten mls. A solution of methyl-l~-mercaptubutyrimidate/HCl in 0.2M Na2C03 (10 mg./ml) was prepared immediately before use. 0.1 ml. of this solution was then added for every 10 ml.
of suspension~ The suspension was stirred for 30 minutes at room temperature and then washed with O.lM ~-ethylmorpholine/
HCl at pH 7.5 containing 1 mg. dithiothreitol per 10 ml. of buffer. The gel was then thoroughly washed with HCl containing O.lM NaCl ko pH 3.5.
The immunoglobulin containing incorporated thiol groups thus prepared was eluted with HCl containing O.lM NaCl at pH 2.5. ~he most convenient method for achieving this was to pack the immunoadsorbent in a small column and collect the ef~luent in tubes containing buffer. Since the next :~ :

.
-. ': ' , ' ' .
.
.

6~

stage of the preparation is performed in 0.lM Acetate buffer at pH 5.0, the effluent was collected in sufficient 0.1M Acetate at pH 6.0 to give a final pH of 5Ø
2 ml o-f 0.lM Acetate buffer at pH 5.0 containing 2.5 mg. of the immunoglobulin modifiecl with thiol groups were cooled on an ice-bath. This solution was then slowly added drop b~ drop to a 1 ml. saturated solution of N,~'-o-phenylene-bis-maleimide in 0.1M Acetate buffer at pH 5.0 on an ice-bath. The mixture was next incubated at 30C., for 20 minutes. The immunoglobulin was then separated from excess N,N-o-phenylene-bis-maleimide by passing the solution through a column ,(20 x 1.5 cm.) of G.25 Sephadex ecLuilibrated at room temperature with 0.01M phosphate buffer containing 0.01M Mg~12 at pH 6Ø
5.0 mg. of ammonium sulphate precipitated ~-galactosidase from Escherichia coli were dissolved in the purified activated immunoglobulin solution and incubated at 30C., for l~ hours. Sufficient ~ ~aOH to adjust the pH of the reaction mixture to 7.5 and ,sufficient 2M
2-mercapto-ethanol to give a final concentration of 10 mM
was added.
The immunoglobulin-enzyme conjugate solution was applied to a column (60 x 0.9 cm.) of DE~E-Agarose (Biogel) equilibrated at 4~C~, with 10 mM Tris-HCl buffer at pH 7.5 -containing 10 r~M 2-mercapto-ethanol, 10 mM MgC12 and 50 mM

~ .

-',' . ' .'' ' ~' '" ' ," :

sodium chloride. The solution was washed onto the col~n with the equilibrating buffer. A gradient from 50 mM to 200 mM sodium chloride in the same buffer was applied with 250 ml in the mixing vessels. Immunoglobulin unconjugated with enzyme fails to adsorb to the column and is eluted before the gradient is applied. When the gradient is applied immunoglobulin labelled with enzyme is eluted before free enzyme. The eluted fraction containing the labelled immunoglobulin is freeze dried to provide the enzyme-labelled second antibody for use in the invention.
2. Production of first antibodY
(a) Preparation of triiodothyronine (rr3) conjugate) 250 my. oE bovine serum albumin (BSA) were dissolved in 25 ml. of distilled water and 1000 mg. of 1-ethyl-3-(3-I5 dimethylaminopropyl)-carbodiimide (EDC) were dissolved in 50 ml. of distilled water. Both solutions were cooled to 40C., and two-third.s of the EDC solution slowly mixed with the BSA solution. 150 mg. of purified T3 were dissolved in 25 ml. of ethanol:2M-NH40H (25:1, v/v) and the solution, cooled to 40C, was added dropwise to the mixture with continuous stirring the pH being maintained at 7.0 by drop-wise addition of 0.25M-HCl. The reaction mixture was stirred for a further 10 minutes and the remaining EDC
solution added dropwise maintaining the pH at 7.0~
The mixture was stirred overnight at 40C. In the , . ~
.

- . :

~ ~v ~

morning the solution was centriuged and the supernatant dialysed against three changes of O.lM Na2C~3/NaHC03 at pH 11.0 and then distilled water. The preparation was finally freeze-dried. The incorporation of T3 into BSA
was determined to be 4.3 moles of T3 per mole of BSA.
(b) Preparation of first antibody (T3-antibodies) A stable emulsion was prepared by mixing 1 part of saline containing Tween 80 (50%), T3-bovine serum albumin (2 mg./ml) and Bacillus Calmette-Guerin (BCG 1 mg/ml) with 2 parts of marcol S2 adjuvant. Six doses (1 ml each) of the emulsion were injected subcutaneously in the back, and four intramuscular injections (0.75 ml) were given in the legs~
The antibody titres were monitored and after 36 weeks the animals were boosted~ For the boost, six intramuscular injections (0.5 mls) w~re given in the legs. The emulsion consisted of 2 parts of adjuvant to 1 part of saline containing T3-BSA conjugate (2 mg/ml). The bleeds ten days after boosting and for the next two months contained high levels of T3 antibodies.

(c) Production of Immunoadsorbent for the purification of first antibody Sepharose l~B was activated with 1,4-bis-(2,3-epoxy~
propoxy)butane as described previously. The activated Sepharose (~ ml~) was added to phosphate buf~er (pH 12.0 0.025M, 20 ml.) containing T3 (50 mg~). The buffer was . . . - - - . : : . -.

.

- , stirred for 45 hours at room temperature~ The gel was then washed with phosphate buffer ~pH 12.0, 0.025M, 250 ml.~, sodium chloride solution (150 ml, O.lM) and distilled water (100 ml.). 9 mg. o-f T3 were incorporated per ml of swollen gelO Any remaining oxirane groups were blocked by suspending the gel in four times its vol~e of aqueous ethanolamine (lM,pH 11 and shaking for 4 hours. The gel was then extensively washed with distilled water to remove excess ethanol~nine.
(d) Purification of first antibody Anti-T3 serum (prepared as described above, 20 ml) was mixed for one hour at room ternperature with T3-substituted Sepharose (1 ml.). Non-speclfically adsorbed proteins were removed from the imrnunoadsorbent by extensive washing with pho,sphate buffered saline (0.05M, pH 7.0, 0.2M NaCl). The gel was packed in a syringe as described previously and washed with hydrochloric acid (pH 3) to remove the phosphate buffer.
T3-antibodies were eluted by washing with guanidinium hydrochloride (4 ml., 6M, pH2) and collected into barbitone buffer ~1 ml, 0.05M, pH 8.6, 0.1% bovine serum albu~nin). The T3-antibodies were extensively dialysed against this bu-ffer to remove the guanidiniurn hydrochloride.

3. Production of supported antigen (T3) for use in the assay The procedure is as described in the production of immunoadsorbent for the purification of the first antibody (2 c above) except that the concentration of T3 is less.

1.

, .'" ' ~3.~i5 "Protocol 2b"
Purified T3 antibodies (1:1500 dilution) in barbitone buffer ~0.05M, pH 8.6, 0.1% w/v BSA, 8.5 ml) were mixed with the enzyme label (donkey anti-sheep IgG-galactosidase conjugate, 5.1 ml) and stored at 40C until required.
T3 standards in barbitone buffer (50 ~1) were mixed with the label-first antibody complex (200 u~l) and additional barbitone buffer (0.5 u~l), then held at 4OC for 4 hours. T3 i~nunoadsorbent (50 ug) in barbitone buffer containing Tween 80 (0.4%) was added to the tubes and incubated at 40C for 1 hour.
The i~nunoadsorbent was washed in turn wlth barbitone buffer containing 5% then 1% sodium chloride. Enz~ne activity bound to the immunoadsorbent was assayed by the usual methods.
ResultS

T3 concentration ~ Optical density Percentage o~
~nmoljl) (1 hour) æero binding 0 o.1~4 100 0.125 0.40 83 0.~5 0.36 69 0.5 0.37 7L~
1 0.29 l~6 2 0.2~ 27 L~ 0.20 15 8 0.19 12 NSB 0.16 Binding calculated àfter subtracting ~SB.

- ~5 -. ~ : : . :

.
. . : , 6~

This experiment was repeated with a donkey ant.i-sheep Fc-enzyme label. The results were similar to those shown above.

"Protocol 3"
T3 standards in barbitone buffer (100 u~1) were mixed with the label-first antibody complex (200 u~l) and T3 immunoadsorbent (25 ~g) in barbitone buffer (100 u~l) containing Tween 80 (0.4% w/v) and incubated at 4OC for 2 hours.
The immunoadsorbent was washed and the bound en~yme was assayed as described for Example 1.
Results T3 concentration~ Optical density Percentage o~
(nmol/l) ~1 hour) zero bi.nding 0 0.33 ~00 0.5 0.32 94 1 0.29 83 2 0.28 77

4 0.24 60 8 0.21 42 16 0.18 32 NSB = 0.114 Binding calculated after subtracting NSB.

, : : ' ' '' '. , ' , , " ''~ ,', . , , ' ' . .. . : , ~
.

This decreases the concentration of T3 substituted into the gel.
Assay for T3 using the reagents whose preparation is described above is carried out in the manner already described.
Several different methods can be used. These include:- 1 1st meth~LI~rotocol 2b"~ ¦
Incubating the first antibody with -the enzyme- '' labelled second antibody to produce a pre-formed co~plex and adding the pre-fonmed complex to the sample. This is followed after a suitable time by addition of the 1~
immunoadsorbent (insolubilized antigen)~
2nd method l"Protocol 3"~
Protocol 3 .involves the simultaneous addit~n of sample and immunoadsorbent,to the pre-formed complex of first antibody and enzyme-labelled antibody.
The following Examples describe the results of assays for triiodothyronine (T3) performed using the methods described above.

- 27 _ .. . . . . . .
:. . . , . . . ~ - .

~ . .: . . :
:: ~ - . . : . . . , .. , , , , -.. ' : , . :' "Protocol 2b"
The procedure was as described previously for Example 1 except that the standards were made up in stripped plasma. Thyroid binding globulin was destroyed by heating the plasma samples in a water bath at 700C for 1 hour. The results obtained were similar to that described previously in Example 1.

Gi 28 _ ',:

.
, The invention provides a kit of materials for use in performing the method accorcling to this invention, such a kit comprising (1) a pxe-formed complex as defined above of purified first antibody ~7"
and an enzyme-labelled second antibody and (2) an insolubilized antigen as defined above.
Preferably the kit includes means for ; determining the enzyme activity as required in performing the method, and preferably includes for calibration purposes a solution containing a known amount of the antigen which the kit is intended for detecting or assaying.
It may be advantageous to have the solid phase attached to a rod which can be inserted into and removed from the other components of the assay mixture. Thls would facilitate washing of the solid phase and increase the precision of the assay~
For example, a.kit for T3 assay could consist of the following components:
1. an insoluble support, either as a suspension in a bu-ffer (e~g. barbitone) or in a lyophilized form, having ~ound thereto the antigen T3.
2. a pre-formed complex of the first antibody (anti-T3 raised in sheep) toge~er with the second antibody (raised in donkey,anti-sheep IgG Fc), the second antibody ~i~ ' .

':: ' '.' ' " ' ' -', ' . ' ' ' ' ,' . '':

6~ `

being labelled with ~-galactosidase, this component preferably being in a lyoph;l;zed form.
3. a substrate (e.g. nitroph~enyl galactoside) at a final concentration of about 10 M in a buffer ~e.g. barbitone) preferably having a pH from 7 to 9 and a molarity of 0.010 to 0.200, this component preferably being in a lyophilized form.
4. a solution containing a known amount of antigen T3 (e.g. 10 nanomoles per litre), this component preferably being in a lyophilized form.

_ 30 -:-J l

Claims (10)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. In a method of enzyme-linked immunoassay for determining the presence or amount of an antigen in a sample, in which one brings into contact in a liquid medium, an antigen which is selected from immunogens and haptens, and is insolubilized by being carried on an insoluble support or in the form of an insoluble aggregate, a first antibody which belongs to an immunoglobulin class and which is anti- to the insolubilized antigen, and an enzyme-labelled second antibody which is anti- to the immunoglobulin class of the first antibody and which is enzyme-labelled by a covalent linkage between the enzyme and the second antibody; separates the insoluble and soluble material; and determines the enzyme activity of the soluble or insoluble separated materials, thereby determining the presence or amount of antigen in the sample, the improvement which comprises purifying said first antibody to remove antibodies non-specific to the insolubillzed antigen, purifying the second antibody to remove antibodies non-specific to antibodies of the immuno-globulin class to which the first antibody belongs, preparing a pre-formed complex of the purified first antibody and purified enzyme-labelled second antibody, and reacting the insolubilized antigen and the sample with said pre-formed complex, whereby the second antibody becomes attached to the insoluble material in inverse relationship to the amount of the antigen in the sample.

2. A method according to claim 1, wherein the pre-formed complex is reacted with the sample and thereafter with the insolubilized antigen, the amount of pre-formed complex being more than sufficient to react with all the antigen present in the sample but not more than sufficient to react with all the antigen present in the sample plus all the insolubilized antigen.

3. A method according to claim 1, wherein said pre-formed complex is reacted with the insolubilized antigen and the sample simultaneously.

4. A method according to claim 1, 2 or 3 whereln the antigen which is insolubilized is the same antigen as the antigen in the sample.

5. A method according to claim 1, 2 or 3 wherein the enzyme activity is determined in the liquid phase.

6. A method according to claim 1, 2 or 3 wherein the second antibody is raised against the Fc fragment of the immunoglobulin class to which the first antibody belongs.

7. A method according to claim 1, 2 or 3 wherein the second antibody is labelled with an enzyme selected from oxydases, hydrolytic enzymes and dehydrogenases.

8. A kit of materials for use in practising a method according to claim 1, which comprises a pre formed complex, as defined in claim 1, of purified first antibody and enzyme labelled second antibody and an insolubilized antigen as defined in claim 1.

9. A kit according to claim 8 which further comprises means for determining the enzyme activity of the soluble or insoluble separated material.

10. A kit according to claim 8 or 9, which also includes a solution containing a known amount of the same antigen as that which is insolubilized.