ANALYST, AUGUST 1991, VOL. 116 787 Immunocomplex-immobilization Technique Sven Oscarsson Biochemical Separation Centre, Uppsala Biomedical Center, Uppsala University, Box 577, 751 23 Uppsala, Sweden Jan Carlsson Pharmacia Diagnostics AB, 75 1 82 Uppsala, Sweden An immunoassay technique is presented, which works equally well with either a radiaoctive isotope or an enzyme as the label. It is based on the competition between a labelled antigen of known concentration and an unlabelled antigen (concentration to be determined) for binding to an antibody, which previously has been thiolated by the heterobifunctional reagent N-[3-(2-pyridyldithio)propionyl]succinamide [marketed as N-succinimidyl-3-(2-pyridyldithio)propionate; SPDP)]. The soluble immunocomplex formed is then immobi- lized onto agarose beads containing reactive disulphide groups.After washing, to remove any unbound and non-specifically adsorbed material, the radioactivity or the enzymic activity on the agarose beads is determined and the concentration of the antigen in the test sample calculated according to the conventional procedure for a competitive immunoassay. In order to demonstrate the applicability of the technique a test procedure for the determination of total serum immunoglobulin E is presented. Keywords: lmmunoassay technique; immunocomplex immobilization; radioimmunoassay; enzyme immunoassay A large number of immunoassay methods based on the biospecific reaction of an antibody with its corresponding antigen have been developed. With these methods it has been possible to detect low concentrations of many types of antigens and antibodies in complex media such as plasma and urine.One of the most widely used immunochemical techniques is the solid phase method presented by Wide.' In this method one of the immuno-components is bound to a solid phase, e . g . , polysaccharide beads, which can be separated from the liquid phase when the immunochemical reaction has reached equi- librium. One disadvantage of this method is the slow rate of transport of the immuno-components from the solution onto the immuno-components in the solid phase owing to poor convection and steric hindrance in the solid phase, thus leading to long reaction times. A system in which the immunocomplex is formed in solution rather than on the solid phase is the double antibody solid phase (DASP) system described by van Weemen and Schuurs.2 In this system the soluble immunocomplex formed is subsequently immobilized via a second antibody, covalently coupled to the solid phase; the second antibody is directed against the first.Unlike conventional competitive immunoassay , the DASP system allows the use of a large excess of immobilized antibody. This partly compensates for the slow rate of migration of the soluble immunocomplex in the solid phase and affords relatively rapid formation of the immobilized immunocomplex. For the DASP system to work properly however, a second antibody, specific for the primary antibody must be available. Non-specific reactions can otherwise occur between the antibody on the solid phase and, for instance, immunoglobulin (Ig) G from the sample or labelled analyte.This leads firstly, to a decreased capacity of the solid phase to bind the soluble immunocomplex, and secondly, to a high background. In order to make the second antibody specific, the first antibody used for the immunization step must be pure, however, an additional immuno-adsorption step might also be necessary. In the proposed technique a competitive immunochemical reaction occurring in solution is performed first, as in the DASP method, but in contrast to DASP, the second step is not accomplished by a subsequent immunochemical reaction but rather through a chemical reaction involving the thiol groups in the first antibody (introduced prior to the immunochemical reaction) and the reactive disulphide groups on the solid phase.Experimental Materials Sephadex G-25M, Activated Thiol-Sepharose 4B, Sephacryl S-300, N-[3-(2-pyridyldithio)propionyl]succinamide [mar- keted as N-succinimidyl-3-(2-pyridyldithio)propionate; SPDP] and Dextran T-70 were obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). Anti-IgE (anti-DE1) of IgG- class (obtained from immunized-rabbit serum) purified by immuno-adsorption; 1251 labelled IgE (100 ng -1.6 @); 1251 labelled anti-IgE (0.4 pg - 5 pCi) labelled by the chloramine- T method; human IgE sera from patient Nils Danielson (ND); and components from the Prist IgE kit, namely, paper discs with covalently coupled IgE and the horse serum diluent, were kindly supplied from Pharmacia Diagnostics (Uppsala, Sweden). The polystyrene test-tubes were from Kebo AB (Stock- holm, Sweden).The ethanol (99.5%) used to dissolve the SPDP was purchased from Svensk Sprit AB (Stockholm, Sweden). The buffer salts: NaH2P04-2H20; KH2P04; and disodium dihydrogen ethylenediaminetetraacetate (EDTA) (C10H14N2Na208-2H20), were all purchased from Merck (Darmstadt , Germany) and were of analytical-reagent grade. The NgCl2.6H2O, NaN3 and Na2C03 used were also pur- chased from Merck. p-Galactosidase (G-5635 VIII), glutathione, dithiothreitol (DTT), human serum albumin (HSA) and o-nitrophenylp-D- galactopyranoside were purchased from Sigma (St. Louis, MO, USA), NaCl and Tween 20 from Kebo AB and methanol from May and Baker (Rhone Poulenc, Dagenham, Essex, UK) . Di-2-pyridyl disulphide (2,2'-dithiopyridine; 2-PDS), iodoacetamide and 2-mercaptoethanol were purchased from Fluka (Buchs, Switzerland).Methods Thiolation of IgG For the thiolation of IgG, the heterobifunctional reagent SPDP was used. The IgG (2.6 x 10-5 mol dm-3,4.3 mg ml-1)788 ANALYST, AUGUST 1991, VOL. 116 was dissolved in 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, and the SPDP in the ethanol (to give a 1-8 mmol dm-3 concentration range). A 30 p1 aliquot of SPDP solution (3 mmol dm-3) was rapidly added to 0.3 ml of the protein solution (2.6 x 10-5 mol dm-3) with stirring. The final ethanol concentration should be less than 10% v/v in order to avoid precipitation of the protein. The reaction was carried out for 30 rnin at room temperature (21-25 "C). Any excess of reagent and reaction products of low relative molecular mass (M,) were removed by gel filtration on a Sephadex G-25 column (16 x 52 mm), which had been equilibrated with 0.1 mol dm-3 sodium phosphate buffer, pH 6.0, containing 0.15 mol dm-3 NaCl and 5 mmol dm-3 EDTA.The void fractions, containing the modified antibody (with the 2-pyridyl disulphide groups) were collected and pooled. To the pool was added DTT dissolved in 0.1 mol dm-3 sodium phosphate buffer, pH 6.0, containing 0.15 mol dm-3 NaCl and 5 mmol dm-3 EDTA (5 mmol dm-3 final concentration) and the reduction was allowed to proceed for 15 rnin at room temperature. Excess of reducing agent and 2-thiopyridone were removed by passing the reaction mixture through a Sephadex G-25 column. The number of thiol groups introduced into the antibody was estimated by titration with 2-PDS as described by Grazetti and Murray.3 Derivatives with 3-6 thiol groups per IgG molecule were prepared.The concentration of SPDP was estimated by a previously described method.4 Preparation of freeze-dried thiolated IgG To 1 ml of thiolated IgG (1 X 10-6 mol dm-3,0.16 mg ml-I), prepared as described above, 9 ml of 0.1 mol dm-3 sodium phosphate buffer (pH 6.0) containing 0.1% Tween 20,0.2% HSA, 2% Dextran T-70, 5 mmol dm-3 EDTA and 1 x 10-6 mol dm-3 glutathione, was added. The solution obtained was transferred into a glass bottle and freeze-dried. The lyophi- lized material was stored at +4 "C until used. Solid phase Activated Thiol-Sepharose 4B has previously been used as the solid phase.5 It has a content of 1 pmol of reactive disulphide groups per millilitre of swollen gel. The agarose derivative is delivered as freeze-dried powder and, therefore, had to be re-swollen before use.This was done by suspending the powder in distilled water or in 0.1 mol dm-3 sodium phosphate, pH 7.0, on a glass filter-funnel. The gel thus formed was washed with 0.1 mol dm-3 sodium phosphate buffer containing 5 mmol dm-3 EDTA, pH 7.5, to remove the lyophilizing additives. After washing, the gel was dried on the glass filter-funnel by use of a vacuum for 5 min in order to eliminate interstitial liquid. A 250 mg portion of the dried gel was then suspended in 1 ml of 0.1 mol dm-3 sodium phosphate buffer containing 5 mmol dm-3 EDTA, pH 7.5. This suspension was used in the test procedure. Preliminary experiments have also been performed with agarose of small particle size (2-4 pm) with a substitution degree of 377 pmol g-1 of dried product (kindly supplied by Pharmacia Fine Chemicals).Preparation of the IgE-/3-galactosidase conjugate P-Galactosidase (50 mg) was dissolved in 2 ml of 0.02 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.1 rnol dm-3 mercaptoethanol. After reaction for 60 rnin at room temperature, the enzyme solution was passed through a Sephacryl S-300 gel-filtration column. The column was previously equilibrated with 0.1 mol dm-3 sodium phosphate buffer, pH 7.0, containing 2 mmol dm-3 and 0.02% NaN3. After separation, the fractions that showed 6-galactosidase activity were pooled and the enzyme solution was concentrated using an Amicon cell (Amicon, Danvers, MA) to a volume of 2 ml.The concentrated enzyme solution was passed through a Sephadex G-25 column equilibrated with 0.1 mol dm-3 sodium phosphate buffer containing 2 mmol dm-3 M g Q , pH 8.0. The thiol content of the P-galactosidase preparation obtained, estimated by titration with 2-PDS73 was 10 mol of SH-groups per mol of enzyme. A 0.2 ml aliquot of human IgE(ND) (10 mg ml-1) dissolved in 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, was mixed with 8 pl of SPDP (1.25 mmol dm-3, dissolved in ethanol). The reaction was carried out for 30 min at room temperature. The excess of reagent, and reaction products of low M , were removed by gel filtration on a Sephadex G-25 column (16 X 52 mm) equilibrated with 0.1 mol dm-3 sodium phosphate buffer, pH 6.0.The void fractions containing the modified IgE (substituted with 2-pyridyl disulphide groups) were collected and pooled. The degree of substitution was found to be 0.66 reactive disulphide structures per IgE molecule, as estimated photometrically by reaction with DTT.4 A 1.4 ml aliquot of modified IgE (3.48 pmol dm-3) was mixed with 0.3 ml of 6-galactosidase (prepared as described above) (8.04 pmol dm-3). The reaction was allowed to proceed for 5-6 d at 4 "C. A 100 mol dm-3 excess of iodoacetamide was added to the solution (in order to block any unreacted sulphydryl groups). The excess of reagent was removed by gel filtration on a Sephadex G-25 column (16 x 42 mm) that had been equilibrated with 0.1 mol dm-3 sodium phosphate buffer, pH 6.7. A 1.7 ml aliquot of 0.3 mol dm-3 sodium phosphate buffer (pH 6.7) containing 0.04% NaN3, 0.2% HSA and 0.2% Tween 20 was then added to the reaction mixture.This preparation of IgE-P-galactosidase conjugate was stable for at least one month at 4 "C. Application of the immunocomplex immobilization technique to the determination of IgE using a radioisotope (1251) as the label Thiolated anti-IgE (either lyophilized and reconstituted with 10 ml of distilled water or freshly prepared as described above) was diluted to a final concentration of 1 x 10-9 mol dm-3 (0.16 pg ml-1) with 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.1% Tween 20 and 5 mol dm-3 EDTA. A 50 pl aliquot of this solution and 50 pl of patient serum were added to polystyrene test-tubes. After 30 min 50 pl of IgE labelled with 1251 and having an IgE concentration of 1.4 x 10-10 mol dm-3 (27 ng ml-1) were added to the test-tubes and the reaction was carried out for 90 rnin at room temperature (21-25 "C).A 100 p1 portion of a suspension of Activated Thiol-Sepharose 4B (corresponding to 25 mg of dried gel per tube), was then added to each test-tube and the mixture agitated for 60 min on an Ika-Wibrax-WXR shaker (Janke & Kunkel, IKA-Labortechnik, Staufen, Germany) at 200 rev min-1. Two millilitres of 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.15 mol dm-3 NaCl and 0.1% Tween 20 was then added to each test-tube. The tubes were centrifuged and the supernatant was sucked off; the procedure was repeated twice. The radioactivity on the solid phase was then measured with a y-counter.All stages were performed at room temperature. A calibration graph was constructed using standards with known IgE concentrations, where 1 unit = 2.42 ng of IgE. The standard points were obtained by diluting the IgE in IgE-free horse serum diluent. This serum diluent was also used as the negative control. The count values obtained for the adsorbed immune complex with thiolated anti-IgE, diluted in IgE-free horse serum, and 1251-labelled IgE as the only constituents during the immuno- logical reaction, is called BO. The count values for the standard points of known IgE concentrations were expressed as percentages of Bo. Application of the immunocomplex-immobilization technique on IgE determination with P-galactosidase as a marker Thiolated anti-IgE (either lyophilized and then reconstituted with 10 ml of distilled water, or freshly prepared as described above) was diluted to a final concentration of 2 x 10-9ANALYST, AUGUST 1991, VOL.116 789 rnol dm-3 (0.32 pg ml-1) with 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.1% Tween 20 and 5 mmol dm-3 EDTA; 50 p1 of the diluted solution and 50 pl of patient serum were added to each test-tube. After 30 min, 50 pl of IgE-P-galactosidase conjugate with an IgE concentration of 2 x 10-9 mol dm-3 (0.32 pg ml-1) were added to the test-tubes and the reaction was carried out for 90 rnin at room temperature. A 100 p1 portion of the stirred suspension of Activated Thiol-Sepharose 4B was then added, which corre- sponds to 25 mg of dried gel per test-tube, and the tubes were incubated for 60 rnin on the shaker (200 rev min-1) at room temperature; 2 ml of 0.1 mol dm-3 sodium phosphate buffer containing 0.1% Tween 20 was then added to each test-tube.the tubes were centrifuged and the supernatant was sucked off, this procedure was repeated twice. After the last washing and rinsing step a residue of 500 pl remained in all of the tubes, 400 pl of the substrate solution was then added to each tube and the contents were mixed. The substrate solution consisted of 0.2 mol dm-3 sodium phosphate buffer containing 2 mmol dm-3 MgC12, 0.25% Tween 20 and 13.2 mmol dm-3 o-nitrophenyl-P-D-galactopyranoside at pH 7.2. After 30 min incubation at 37 "C the enzyme reaction was stopped by adding 0.8 ml of 1 rnol dm-3 Na2C03 to each test-tube.The liberated o-nitrophenolate was measured at 405 nm on an LKB Ultralab System 7400 filter-photometer (LKB, Bromma, Sweden). A calibration graph was obtained by diluting IgE in IgE-free horse serum diluent. This serum diluent was also used as the negative control. The count value obtained for the adsorbed immune complex with thiolated anti-IgE diluted in IgE-free horse serum and 1251 labelled IgE as the only constituents during the immunological reaction is called Bo. The count values for the standard points of known IgE concentrations were expressed as percentages of Bo. Phadebas IgE-Prist method One paper disc with covalently coupled anti-IgE was added to each of a series of test-tubes, together with 100 p1 of patient serum. Incubation was performed (without shaking) at room temperature for 3 h, then 2 ml of 0.1 rnol dm-3 sodium phosphate buffer, pH 7.5, containing 0.15 mol dm-3 NaCl and 0.1% Tween 20 were added.After 5 rnin the wash solution was quantitatively aspirated. This procedure was repeated twice. The 125I-labelled IgE (77 ng ml-1) was added to each test-tube and the mixture incubated without shaking for 17 h at room temperature, after which time 2 ml of 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.15 rnol dm-3 NaCl and 0.1% Tween 20 were added. After 5 rnin the wash solution was sucked off; the wash procedure was repeated twice. The radioactivity of the paper discs was then measured with a y-counter and the IgE concentration of the different samples was determined by using a calibration graph obtained with standards of known IgE concentration.Results and Discussion The principle of the immunocomplex-immobilization tech- nique is shown in Fig. 1. Thiolated antibody is mixed with a sample containing a known (standard) or unknown (patient sample) amount of corresponding antigen. Isotope or enzyme labelled antigen (Ag*) is then added and incubation is performed for a fixed period of time during which a mixture of labelled and unlabelled immunocomplexes is formed, both types containing thiol groups. The immunochemical reaction is carried out until equilibrium has been reached. A suspension of beaded agarose containing reactive disul- phide groups (Activated Thiol-Sepharose 4B) is then added and the final mixture is incubated for 60 min in order to effect the immobilization of the immunocomplexes onto the agarose beads.The unbound material, as free labelled antigen, is finally removed from the agarose gel and the amount of bound material is determined by conventional techniques (isotope measurement for radioimmunoassays and enzyme activity measurements for enzyme linked immunosorbent assays (see also under Methods). Step 1 Step 2 Step 3 Step 4 Ab-SH 2 Ag (Competitive reaction) %pg* I -S-S-Ab-Ag* + unbound Ag -S-S-Ab-Ag + unbound AS* I I I Wash solution I 3 x 2 m l t I Determination of solid phase bound radioactivitv or enzymic activity Fig. 1 Test principle (schematic representation. Step 1. (a) Thio- lated antibody (Ab-SH) and sample containing antigen (Ag) to be determined are added to each test-tube and incubated for 30 min.(b) Labelled antigen (Ag* = 1251 or P-galactosidase) is then added to each test-tube. The reaction must then proceed for another 60 min. Step 2. Activated agarose is added to each test-tube and the tubes are incubated for 60 rnin with mixing. Step 3. The solid phase is washed three times to remove unbound and non-specifically bound Ag*. Step 4. The tubes containing the washed agarose beads are placed in a y-counter to determine the radioactivity. The enzymic activity is determined by adding substrate to the test-tubes Step 1 Step 2 Step 3 Step 4 0. DDT (ox) DDT (red) B Anti body-N H-C-CH,-CH,-SH + A 0,H OF SH-CH2-CH-CH-CH2-S H DDT = Di t h i ot h rei to I Fig. 2 Scheme for thiolation of antibody with SPDP. Step 1. SPDP is added to the antibody solution.Step 2. After 30 min of reaction, the excess of reagent and products of low M, are removed by de-salting on a Sephadex G-25 column. Step 3. The modified antibody is treated with DIT. Step 4. Excess of reducing agent and 2-thiopyridone formed are removed by de-salting on a Sephadex G-25 column790 ANALYST, AUGUST 1991, VOL. 116 Immunoglobulins (of IgG type) normally do not contain free sulphydryl groups and thus have to be provided with such groups; this can be achieved using a number of methods. In this work the two-step procedure outlined in Fig. 2 has been used. Reactive disulphide groups (2-pyridyl disulphide groups) are first introduced using the heterobifunctional reagent, SPDP, which reacts with exposed amino groups on the antibodies. By varying the amount of the excess of reagent, the desired degree of substitution can be obtained.The protein-bound 2-pyridyl disulphide groups are then converted into thiol groups by reaction with a thiol of low M, such as DTT. As the 2-pyridyl disulphide groups are very reactive, this reduction can be performed by using an equimolar concentration of the thiol of low M,, thus avoiding simultaneous reduction of the native disulphide groups. It is important to provide the antibodies with a sufficient number of thiol groups so that binding of the immunocomplex formed to the agarose beads can occur even if some of the thiol groups become sterically hindered as a result of the formation of the immunocomplex. Antibodies have been prepared with 2.9, 3.5,6.0,12.9 and 18.2 rnol of thiol groups per rnol of IgG.The amount of immunocomplexes immobilized, with a substitu- tion degree of 18.2 mol of thiol groups per IgG molecule, was 98.6% of the amount immobilized when an antibody with a substitution degree of 2.9 was used. The highly substituted antibodies (18.2 mol of thiol groups per mol of IgG) were found to precipitate into the solution after 3 d of storage at 4 "C. It was also found that 3-6 mol of thiol groups per mol of IgG was sufficient to ensure effective immobilization. More- over, within this range the immunochemical reactivity of the thiolated antibody was the same as that for the native antibody. The thiolated antibody preparations (with a substitution degree of 3-6 rnol of thiol groups per mol of IgG), when kept in 0.1 rnol dm-3 sodium phosphate buffer, pH 7.5 (containing 5 mmol dm-3 EDTA), at 4 "C for 1 month, showed no decrease in immunochemical or chemical reactivity. After the addition of HSA, Dextran T-70, EDTA and glutathione (see under Experimental), it was also possible to freeze-dry the thiolated antibodies.Such preparations appeared to be stable for at least 2-3 months. The rate of the immunochemical reaction, i. e . , the competition between sample and labelled antigen for binding sites on the thiolated antibody, depends, among other things, on the relative concentrations of the participating immunocomponents and their affinities for each other. As the concentration in a 100 80 - s $ 60 >. > 1 -0 c .- .- c 2 40 20 0 1 1 I I l l 1 2.5 5 25 lY0 200400 IgE/U rnl- Fig. 3 Calibration graph for determination of IgE with 1251 as a marker.The calibration graph for determination of IgE in unknown samples is established according to the procedure presented under Methods. The count min-1 values for the different standard concen- trations are plotted as a percentage of the counts min-* value obtained for the negative control (IgE-free horse serum diluent) competitive system is fixed for the antibody, faster reaction can be obtained, for example, by exchanging the antibody for one with higher affinity. With the IgE used in this work, equilibrium in the immunochemical reaction was obtained within 90 min. For the immobilization of the immunocomplex, any solid phase can, in principle, be used, provided that it can be substituted with reactive disulphide groups.In this study Activated Thiol-Sepharose 4B, which contains 1 pmol of reactive disulphide groups per ml of swollen gel, was used. This agarose gel shows low non-specific adsorption and is commercially available. With the optimal mass of gel (12.5-25 mg of dried gel per test-tube) maximum binding of immunocomplex was reached after about 60 min of incubation, with mixing. The calibration graph for IgE with 1251 as a marker, with a total reaction time of 2.5 h (1.5 h of immunochemical reaction and 1 h of immobilization reaction) is shown in Fig. 3. With this calibration graph it is possible to estimate the concentra- tion of IgE in the interval between 8 and 400 U of IgE ml-1. The non-specific adsorption to the matrix was 4.5% of the total added activity and BdT was 22%.The calibration graph for IgE with P-galactosidase as the label is shown in Fig. 4. Under the conditions specified it would be possible to estimate the IgE concentration over the range 5400 U of IgE ml-1 and with a total reaction time of 3 h (1.5 h of immunochemical reaction, 1 h of immobilization 100 80 s 7J 5 60 0 n >. > '0 40 Q I c .- .- 0 1 2.5 5 25 100200400 IgE/U ml-' Fig. 4 Calibration graph for determination of IgE with the enzyme as marker. The calibration graph for determination of IgE in unknown samples is established according to the test procedure presented under Methods. Every absorbance value at 405 nm for the standard points with known IgE concentrations is taken as a percentage of the absorbance value at 405 nm for the negative control (IgE-free horse serum diluent) 3.5 3.0 2.5 2.0 3 ; 1.5 0, 0, 1.0 0.5 m .- I - I - I J 0 0.5 1.0 1.5 2.0 2.5 3 3.5 Log(lgE/U ml-1) ICI Fig.5 Scattergram obtained after a comparative study of the IgE content estimated in 41 serum samples using the commercial Phadebas IgE-Prist technique and the immunocomplex-immobilization (ICI) techniqueANALYST, AUGUST 1991, VOL. 116 791 reaction and 0.5 h of substrate reaction time). The non- specific adsorption to the matrix was 0.5% of the total added activity and BdT was 10%. When the commercial product Phadebas IgE-Prist and the immunocomplex-immobilization-technique with 1251 as the marker were compared by estimating the IgE content in serum samples from 41 persons, the correlation coefficient between the two techniques was 0.955 (see Fig.5 ) . The concentration of IgE in the 41 different patient samples was 2.8-2890 U of IgE ml-1. All of the serum samples with IgE values >400 U ml-1 were diluted in IgE-free horse serum diluent. In an attempt to increase the speed of the reaction in the immunocomplex-immobilization step, agarose derivatives of smaller particle sizes were investigated. Preliminary experi- ments demonstrated that the use of 2-pyridyl disulphide- containing agarose derivatives with small particle diameter (2-4 pm) instead of the commercially available Activated Thiol-Sepharose 4B with a particle diameter of 50-250 vm, shortened the time required for immobilization by at least 30 min. Moreover, the need to agitate the suspension during the immobilization reaction was also eliminated with this prepara- tion because the small agarose beads did not settle out. Preliminary experiments with thyroid stimulating hormone, nortriptylin, human IgG, insulin and 62-microglobulin indi- cate that the immunocomplex-immobilization technique is a fast and easy method with a wide range of applications and thus, should be an interesting alternative to the conventionally used immunoassay techniques. We thank Dr. R. Brandt for valuable discussions and Dr. D. Eaker for the linguistic revision. References Wide, L., in Radioimmunoassay Methods, eds., Kirkham, K. E . , and Hunter, M. M., Livingstone, London, 1970, p. 199. van Weemen, B. K., and Schuurs, A. H., FEBS Lett., 1971,15, 232. Grazetti, D. R., and Murray, J. F., Arch. Biochem. Biophys., 1967, 119, 41. SPDP; Heterobifunctional Reagent. Pharmacia Fine Chemicals AB, Uppsala, Sweden, 1978. Affinity Chromatography; Principles and Methods, Pharmacia Fine Chemicals AB, Uppsala, Sweden, 1988. Paper 1 I00491 C Received February 4th, 1991 Accepted March 26th, 1991