Preparation of a monomeric 2,4-dinitrophenyl Fab'-beta-D-galactosidase conjugate for immunoenzymometric assay

A method is described for the preparation of a monomeric Fab'-beta-D-galactosidase conjugate, which is required for the development of a sensitive immunoenzymometric assay. Anti-human IgG F(ab')2 was labeled with 2,4-dinitrophenyl groups, split into Fab' by reduction and reacted with excess maleimide groups which had been introduced into beta-D-galactosidase through thiol groups using N,N'-o-phenylenedimaleimide. The monomeric 2,4-dinitrophenyl Fab'-beta-D-galactosidase conjugate was subsequently separated from unconjugated beta-D-galactosidase by affinity chromatography on a column of (anti-2,4-dinitrophenyl) IgG-Sepharose 4B. In the monomeric conjugate preparation, 98% of beta-D-galactosidase activity was associated with Fab' and 90% was associated with specific (anti-human IgG) Fab'. This conjugate allowed the measurement of 0.1 fmol of human IgG by an immunoenzymometric assay technique.     

Improved procedure for the conjugation of rabbit IgG and Fab' antibodies with beta-D-galactosidase from Escherichia coli using N,N'-o-phenylenedimaleimide.

The procedures for the conjugation of rabbit IgG and Fab' antibodies with beta-D-galactosidase from Escherichia coli using N,N'-o-phenylenedimaleimide were improved in several respects as compared with the previous methods (Eur. J. Biochem. 62, 285--292, 1976; J. Immunol. 116, 1554--1560, 1976). Maleimide residues were efficiently introduced into antibodies under an atmosphere of nitrogen; the average number of maleimide residues introduced into IgG and Fab' antibodies were 0.78 (0.65--0.86) and 0.86 (0.80--0.95) per molecule, respectively. The conjugation with the enzyme was performed at 4 degrees C at pH 6.5 for 15 or more hours. The conjugates were almost completely separated from unreacted IgG and Fab' by gel filtration. When the recoveries of IgG, Fab', and beta-D-galactosidase in the conjugates were 23-29, 35-44, and 99%, respectively, the average numbers of IgG and Fab' molecules conjugated with the enzyme were 1.5-1.7 and 2.1-2.8 per molecule, respectively. There was no significant impairment of beta-D-galactosidase activity or the activity of anti-human IgG antibody to bind to human IgG upon conjugation. However, the conjugate preparation was heterogeneous, and one-third of each preparation consisted of aggregated conjugates less useful in sandwich enzymoimmunoassay than the remaining material. The conjugate with Fab' antibody gave lower control values in sandwich enzymoimmunoassay with silicone rubber as a solid phase than that with IgG antibody.     

Detection of one milliattomole of ferritin by novel and ultrasensitive enzyme immunoassay

A novel and ultrasensitive enzyme immunoassay (immune complex transfer two-site enzyme immunoassay) for ferritin is described. Ferritin was reacted simultaneously with affinity-purified dinitrophenyl biotinyl anti-ferritin IgG and affinity-purified anti-ferritin Fab'-beta-D-galactosidase conjugate. The complex formed of the three components was trapped onto affinity-purified (anti-dinitrophenyl group) IgG-coated polystyrene balls. After eliminating excess conjugate by washing, the complex was eluted from the polystyrene balls with an excess of epsilon N-dinitrophenyl-L-lysine and transferred to streptavidin-coated polystyrene balls. The beta-D-galactosidase activity bound to streptavidin-coated polystyrene balls was assayed by fluorometry. Nonspecifically bound beta-D-galactosidase activity was remarkably lowered but there was much less decrease in specifically bound beta-D-galactosidase activity. As a result, the detection limit of ferritin was lowered to 1 milliattomole (1 x 10(-21) mol, 600 molecules as calculated from Avogadro's number). This technique will be useful for measuring, for example, antigens in single cells.     

Receptor aggregation is necessary for activation of the soluble insulin receptor kinase

Purified polyclonal human antibodies (B-8) against the receptor for insulin (anti-R IgG), and their F(ab')2 and Fab' fragments, were used to study a possible role of receptor aggregation in the process that couples insulin binding with the activation of the insulin receptor kinase. Anti-R IgG, F(ab')2, and Fab' fragments were shown to inhibit insulin binding to solubilized partially purified receptor preparations from rat liver. This suggests that the antibodies and fragments bind near or at the insulin-binding site. Only anti-R IgG and its bivalent F(ab')2 fragments were capable of stimulating the receptor kinase activity. Monovalent Fab' fragments were completely devoid of such activity. Cross-linking of anti-R Fab' with goat anti-human Fab' restored the capability of the Fab' fragments to activate the receptor kinase. These data strongly suggest that receptor cross-linking or aggregation constitutes a sufficient trigger to activate the insulin-receptor kinase and could, therefore, be an important step in the transmembrane signaling process. This step presumably precedes the activation of the receptor kinase and the resulting phosphorylation of its protein substrates.     

Synthesis of bifunctional antibodies for immunoassays

The synthesis of bifunctional antibodies using the principle of solid-phase synthesis is described. Two Fab' fragments were chemically linked together via a bismaleimide crosslinking reagent. The F(ab')(2) fragments from intact immunoglobulin G (IgG) were prepared using an immobilized pepsin column. Goat, mouse, and human antibodies were digested completely within 4 h. The F(ab')(2) fragments thus produced did not contain any IgG impurities. Fab' fragments were produced by reducing the heavy interchain disulfide bonds using 2-mercaptoethylamine. Use of the solid-phase reactor in the preparation of the bifunctional antibodies eliminated many of the time-consuming separation steps between the fragmentation and conjugation steps. This procedure facilitates the automation of bifunctional antibody preparation and the rapid optimization of reaction conditions.     

Pretargeting for cancer radioimmunotherapy with bispecific antibodies: role of the bispecific antibody's valency for the tumor target antigen

The use of a divalent effector molecule improves bispecific antibody (bsMAb) pretargeting by enabling the cross-linking of monovalently bound bsMAb on the cell surface, thereby increasing the functional affinity of a bsMAb. In this work, it was determined if a bsMAb with divalency for the primary target antigen would improve bsMAb pretargeting of a divalent hapten. The pretargeting of a (99m)Tc-labeled divalent DTPA-peptide, IMP-192, using a bsMAb prepared by chemically coupling two Fab' fragments, one with monovalent specificity to the primary target antigen, carcinoembryonic antigen (CEA), and to indium-loaded DTPA [DTPA(In)], was compared to two other bsMAbs, both with divalency to CEA. One conjugate used the whole anti-CEA IgG, while the other used the anti-CEA F(ab')(2) fragment to make bsMAbs that had divalency to CEA, but with different molecular weights to affect their pharmacokinetic behavior. The rate of bsMAb blood clearance was a function of molecular weight (IgG x Fab' < F(ab')(2) x Fab' < Fab' x Fab' conjugate). The IgG x Fab' bsMAb conjugate had the highest uptake and longest retention in the tumor. However, when used for pretargeting, the F(ab')(2) x Fab' conjugate allowed for superior tumor accretion of the (99m)Tc-IMP-192 peptide, because its more rapid clearance from the blood enabled early intervention with the radiolabeled peptide when tumor uptake of the bsMAb was at its peak. Excellent peptide targeting was also seen with the Fab' x Fab' conjugate, albeit tumor uptake was lower than with the F(ab')(2) x Fab' conjugate. Because the IgG x Fab' bsMAb cleared from the blood so slowly, when the peptide was given at the time of its maximum tumor accretion, the peptide was captured predominantly by the bsMAb in the blood. Several strategies were explored to reduce the IgG x Fab' bsMAb remaining in the blood to take advantage of its 3-4-fold higher tumor accretion than the other bsMAb conjugates. A number of agents were tested, including those that could clear the bsMAb from the blood (e.g., galactosylated or nongalactosylated anti-id antibody) and those that could block the anti-DTPA(In) binding arm [e.g., DTPA(In), divalent-DTPA(In) peptide, and DTPA coupled to bovine serum albumin (BSA) or IgG]. When clearing agents were given 65 h after the IgG x Fab' conjugate (time of maximum tumor accretion for this bsMAb), (99m)Tc-IMP-192 levels in the blood were significantly reduced, but a majority of the peptide localized in the liver. Increasing the interval between the clearing agent and the time the peptide was given to allow for further processing of the bsMAb-clearing agent complex did not improve targeting. At the dose and level of substitution tested, galacosylated BSA-DTPA(In) was cleared too quickly to be an effective blocking agent, but BSA- and IgG-DTPA(In) conjugates were able to reduce the uptake of the (99m)Tc-IMP-192 in the blood and liver. Tumor/nontumor ratios compared favorably for the radiolabeled peptide using the IgG x Fab'/blocking agent combination and the F(ab')(2) x Fab' (no clearing/blocking agent), and peptide uptake 3 h after the blocking agent even exceeded that of the F(ab')(2) x Fab'. However, this higher level of peptide in the tumor was not sustained over 24 h, and actually decreased to levels lower than that seen with the F(ab')(2) x Fab' by this time. These results demonstrate that divalency of a bsMAb to its primary target antigen can lead to higher tumor accretion by a pretargeted divalent peptide, but that the pharmacokinetic behavior of the bsMAb also needs to be optimized to allow for its clearance from the blood. Otherwise, blocking agents will need to be developed to reduce unwanted peptide uptake in normal tissues.     

Asymmetrically glycosylated IgG isolated from non-immune human sera

When human IgG or its F(ab')2 fragment purified from a pool of non-immune sera was passed through a Con A-Sepharose column, 12% of the molecules bound to concanavalin A. While 44% of Fab' and 72% of Fd' fragments obtained from F(ab')2 retained by concanavalin A and eluted with methyl alpha-D-mannoside bound to concanavalin A, the Fab' and Fd' fragments obtained from non-retained F(ab')2 and the L chains and Fc fragments did not interact with the lectin. Only Fd' fragment obtained from the F(ab')2 retained by concanavalin A inhibited the fixation of guinea-pig erythrocytes to concanavalin A. These results are similar to those previously observed for IgG antibodies of different animal species and indicate that partial asymmetric glycosylation is a general phenomenon that is not restricted exclusively to IgG molecules with known specificity.     

F(ab')2 reagents are not required if goat, rather than rabbit, antibodies are used to detect human surface immunoglobulin.

The present report provides evidence that whole goat anti-human immunoglobulin, unlike similar reagents produced in the rabbit, binds both to the same number of and the same individual cells as the F(ab')2 fragments of rabbit or goat anti-human immunoglobulin. These results suggest that goat IgG has a lower affinity for the Fc receptors of human lymphocytes and monocytes than rabbit IgG. Because of this property, whole goat antibodies against human immunoglobulin can be used as simple, convenient relatively inexpensive reagents for the routine detection of immunoglobulin on cell surfaces by immunofluorescence microscopy. The preparation of F(ab')2 fragments of anti-immunoglobulin, which are necessary when rabbit antibodies are used, does not appear to -e required if goat antibodies can be empolyed. This observation has multiple practival applications in cellular immunology.     

Idiotypic determinants on human T cells and modulation of human T cell responses by anti-idiotypic antibodies

The role of idiotypic anti-idiotypic interactions in the regulation of the human T cell response to tetanus toxoid (TT) antigen was examined in three subjects. Rabbit anti-idiotypic (anti-Id) antisera were raised against IgG (Fab')2 anti-TT obtained 7 to 10 days after booster immunization with TT. F(ab')2 fragments of rabbit-anti-Id IgG were used in conjunction with fluorescein-conjugated goat anti-rabbit Ig in an indirect immunofluorescence assay to determine the frequency of Id-positive cells in T cell-enriched preparations. This frequency was 24, 29, and 38 per 10,000, respectively, in the three subjects studied. Significant contribution of contaminating B cell to fluorescence-staining was ruled out by capping experiments using goat anti-human Ig (GAHIG) and by double staining experiments using rhodamine-conjugated GAHIG. Absorption of anti-Id antisera with Epstein Barr virus (EBV)-transformed B cell lines from the IgG (Fab')2 anti-TT donor, but not with EBV-B cell lines from unrelated donors, removed their reactivity with the T cells. Rabbit anti-Id IgG caused minimal proliferation (two-threefold) of T cells and had no effect on T cell proliferation in response to TT antigen when added to the cultures. Preincubation of T cells for 48 hr with rabbit anti-Id IgG (Fab')2, but not with preimmune rabbit IgG (Fab')2, resulted in the generation of antigen-specific suppressor cells that inhibited T cell proliferation in response to TT, but not in response to diphtheria toxoid (DT). These cells also inhibited the synthesis of IgG anti-TT in response to in vitro stimulation with TT antigen, but not the synthesis of IgG anti-DT in response to DT antigen. Adsorption of T cells over plates coated with rabbit anti-Id IgG (Fab')2 enhanced the proliferative response of the T cells to TT, but not to DT antigen, and enhanced the helper activity of the T cells for the in vitro synthesis of IgG anti-TT but not of IgG anti-DT antibodies. These results suggest that idiotypic-anti-idiotypic interactions play a role in the human T cell response to antigen.     

Neutralizing activities of early and late IgG fragments from rabbits immunized with herpes simplex virus.

Rabbits immunized with herpes virus were bled periodically and bivalent and univalent fragments of IgG from each serum sample were prepared by enzymatic digestion. The 2-week F(ab')2 showed a low neutralizing activity only after addition of anti-IgG. F(ab')2 of the 4-week serum retained almost all of the neutralizing activity of IgG, while its univalent fragments demonstrated none even when tested with anti-IgG. In contrast to these early IgG fragments, univalent fragments of the 9-week and 20-week IgG neutralized the virus to considerable extents in the absence of anti-IgG; after addition of anti-IgG the activity equaled that of intact IgG in the cases of Fab' and Fab-II, though the activity of Fab-I was relatively low. Three three univalent fragments were all sensitive to heating at 70 C and to ultraviolet irradiation, whereas intact IgG resisted these treatments. F(ab')2 was resistant to the heating and less sensitive to ultraviolet irradiation than univalent fragments. Neutralization kinetic curve experiments to test blocking effects of IgG fragments against the neutralization by intact IgG suggested that the early Fab' did combine with the virus and that the late Fab' exerted a higher blocking effect than the early Fab'.     

Targeting weak antigens to CD64 elicits potent humoral responses in human CD64 transgenic mice

Previous studies have documented that targeting foreign Ags to IgG FcgammaR leads to enhanced Ag-specific responses in vitro and in vivo. However, the ability to overcome immunologic nonresponsiveness by targeting poorly immunogenic Ags to FcgammaR has not been investigated. To address this question in a simple model, we immunized transgenic mice expressing human CD64 (FcgammaRI) and their nontransgenic littermates with Fab' derived from the murine anti-human CD64 mAb m22. The m22 Fab' served as both the targeting molecule and the Ag. We found that only CD64-expressing mice developed anti-Id titers to m22. Furthermore, chemically linked multimers of m22 Fab', which mediated efficient internalization of the human CD64, were significantly more potent than monomeric m22 F(ab')(2) at inducing anti-Id responses. In all cases, the humoral responses were specific for m22 Id and did not react with other murine IgG1 Fab' fragments. Chemical addition of a second murine Fab' (520C9 anti-human HER2/neu) to m22 Fab' multimers demonstrated that IgG1 and IgG2a anti-Id titers could be generated to 520C9 only in the CD64-expressing mice. These results show that targeting to CD64 can overcome immunological nonresponsiveness to a weak immunogen. Therefore, targeting to CD64 may be an effective method to enhance the activity of nonimmunogenic tumor vaccines.     

The role of A-cells in affecting the immunostimulating effect of F(ab')2-fragments

The immunoregulatory effect of F(ab')2 fragments on normal rabbit IgG and that preincubated with A-cells from spleen have been compared. Both products were tested for their ability to enhance primary immune response of rabbit spleen cells to SRBC. It was demonstrated that low molecular mass product appeared after F(ab')2 fragments incubation with A-cells at 37 degrees C and possessed immunostimulating activity similar to that of initial F(ab')2 fragments. In addition, it was shown that F(ab')2 reduction to monovalent Fab' fragment with the following alkylation of SH-group abolished the ability of Fab' fragment to enhance the immune response. It may signify that half cystein Fab' fragment residue is essential for processing of the fragment in A-cells and (or) for immune response enhancement.     

DEAE-Affi-Gel Blue chromatography of human serum: use for purification of native transferrin

Human serum was subjected to chromatography on DEAE-Affi-Gel Blue which combines ion-exchange and pseudo-ligand-affinity chromatography in a 0.02 M phosphate buffer, pH 7.0. All serum proteins were bound with the exception of transferrin, IgG (immunoglobulin G) and trace amounts of IgA. After a second step of Sephadex G-100 gel chromatography, or affinity chromatography against goat anti-human IgG F(ab')2 coupled to AH-Sepharose 4B, IgG and IgA were removed. The transferrin obtained was homogeneous and of high yield (greater than 80%), and was unaltered as judged by analyses of molecular weight, isoelectric point, iron-binding capacity, antigenicity, and ability to bind to high-affinity specific cellular receptors. Thus, DEAE-Affi-Gel Blue chromatography may be used as the basis for a simple, rapid, two-step method for the purification of large amounts of native transferrin from serum.     

Cytotoxicity of chimeric (human-murine) monoclonal antibody BR96 IgG, F(ab')2, and Fab' conjugated to Pseudomonas exotoxin.

We have made antigen-specific cytotoxic reagents by conjugating the chimeric antibody BR96 (chiBR96) to Pseudomonas exotoxin A (PE), as either native PE or a truncated form (LysPE40) devoid of the cell-recognition region (domain I). PE kills cells by ADP-ribosylation of elongation factor 2, thereby inhibiting protein synthesis. Chimeric BR96 immunotoxins were constructed by chemical conjugation of the toxin to Fab', F(ab')2, and intact IgG and purified by anion-exchange and gel-filtration chromatography. Chimeric BR96 [IgG and F(ab')2] immunotoxins were cytotoxic against tumor cell lines displaying the BR96 antigen, with EC50 values ranging from 0.1 to 110 pM. Immunotoxins constructed with chiBR96 Fab' were 50-100-fold less cytotoxic. Competition analysis showed that the immunotoxins were specifically active through their BR96 antigen-binding ability. The binding of chiBR96-PE and chiBR96-LysPE40 to antigen was equivalent to that of BR96 itself and these immunotoxins were found to internalize very rapidly, displaying 90% of their cytotoxicity within 1 h. Binding assays determined that chiBR96 F(ab')2-LysPE40 bound as well as chiBR96-LysPE40; however, chiBR96 Fab'-LysPE40 bound 20-fold less efficiently. The chiBR96 Fab'-LysPE40 internalized similarly to the F(ab')2 or the IgG immunotoxins. Therefore, the chiBR96 Fab'-LysPE40 immunotoxin is less cytotoxic toward target cells because of reduced antigen binding. This is may be due to the monovalent nature of chiBR96 Fab'-LysPE40. This study shows that the monoclonal antibody chiBR96-Pseudomonas exotoxin A immunotoxins can be effective at inhibiting protein synthesis in target cells.     

抗CD20嵌合抗体片段F(ab′)2突变体在大肠杆菌中的高效分泌表达

构建抗CD20嵌合抗体片段F(ab′)₂ 突变体 ,研究其在大肠杆菌中的高效表达及其表达产物的生物学活性。采用PCR法构建抗CD20嵌合抗体片段F(ab′)₂ 突变体 ,并用双脱氧终止法测定DNA序列 ;采用 19L发酵罐高密度发酵抗CD20嵌合抗体片段F(ab′)₂ 突变体 ,采用亲和色谱和分子筛色谱法纯化表达产物 ,并用SDS-PAGE和薄层激光扫描鉴定纯化产物 ;采用活细胞间接免疫荧光法测定纯化产物与靶细胞的结合活性 ;MTT法测定纯化产物对Raji细胞的生长抑制作用 ,并研究其作用机理。DNA序列测定结果表明 ,抗CD20嵌合抗体片段F(ab′)₂ 突变体已成功构建 ,表达可溶性产物的产量达 360mg L ,具有与Raji细胞 (CD20⁺)结合的活性 ,并抑制Raji细胞的生长 ,其作用机理为诱导Raji细胞凋亡。此突变体有望成为治疗非何杰金氏B细胞淋巴瘤的药物。     

Enzyme-linked immunoassay. II. A simple method for synthesis of the rabbit antibody-beta-D-galactosidase complex and its general applicability.

Rabbit anti-human IgG antibody was mildly reduced in the presence of 10 mM 2-mercaptoethylamine and then coupled to beta-D-galactosidase [EC 3.2.1.23] from Escherichia coli using N,N'-o-phenylenedimaleimide. Human IgG and the anti-human IgG antibody-beta-D-galactosidase complex were successively adsorbed on Sepharose 4B binding rabbit anti-human IgG antibody. In this way amounts of human IgG as small as 5X10(-15) moles could be measured by determining the activity of beta-D-galactosidase bound to the Sepharose.     

Highly Sensitive Immunoassays for Three Forms of Rat Brain Enolase

Highly sensitive enzyme immunoassay systems for three forms (alpha alpha, alpha gamma, and gamma gamma) of rat brain enolase were prepared by use of specific antisera to two distinct subunits (alpha and gamma) of the isozymes and beta-D-galactosidase from Escherichia coli as label. Less than fmol-levels of the homologous dimer forms (alpha alpha and gamma gamma) could be determined with the corresponding antibody F(ab')2-bound solid-phase and the antibody Fab'-beta-D-galactosidase complex. The hybrid form (alpha gamma) could also be assayed specifically by use of the antibody to one subunit as solid-phase, and the antibody to another subunit as labelled complex with a minimum detectable sensitivity of 1 fmol.     

Prolonged in vivo residence times of antibody fragments associated with albumin

Antibody fragments can be expressed at a high level in microbial systems, but they may have limited therapeutic value because they are rapidly eliminated from the body. We demonstrate here that site-specific conjugation or binding of bacterially derived Fab' to the long-lived protein serum albumin allows full retention of the antibody's binding characteristics while imparting the albumin's longevity in vivo. In rats the area under the curve for Fab' conjugated to rat serum albumin was 17-fold greater than for the control of Fab' conjugated to cysteine. Again, a bispecific F(ab')(2) with specificity for rat serum albumin showed an area under the curve about 8-fold greater than did a F(ab')(2) without specificity to albumin. Genetic fusions of scFv to albumin were similarly long-lived and could be expressed in yeast to provide the basis of a cost-effective production system.     

Role of antigen-specific T cell help in the generation of in vivo antibody responses. II. Sustained antigen-specific T cell help is required to induce a specific antibody response.

The injection of mice with a goat or rabbit antibody to mouse IgD stimulates a large polyclonal IgG response, approximately 10% of which is specific for antigenic determinants on the anti-IgD antibody molecule. The large goat IgG (GIgG)-specific antibody response in mice injected with goat antibody to mouse IgD requires that GIgG-specific B cells undergo much greater clonal expansion than B cells specific for other Ag. One possible explanation for the greater clonal expansion of GIgG-specific B cells is that B cells that lack GIgG specificity can only be stimulated with GIgG-specific T help during the relatively short time that anti-IgD binds to, and is processed and presented by, these B cells before they cease to express membrane mIgD. In contrast, GIgG-specific B cells can continue to bind, process, and present GIgG through mIgM after they lose mIgD. To test the hypothesis that extended stimulation with Ag-specific T help is required to generate a specific antibody response, we determined time requirements for Ag-specific T cell help for the development of such a response. Mice were injected with rabbit antibody to mouse IgD plus one or more daily injections of FITC conjugated to a F(ab')2 fragment of rabbit IgG (FITC-(Fab')2), which has a short in vivo half-life, and IgG1 anti-FITC antibody production was analyzed. In this system, each additional injection of FITC-F(ab')2 extends the period during which FITC-specific B cells can process this Ag and present it to rabbit IgG-specific T cells. Each additional injection of FITC-F(ab')2 stimulated a several-fold increase in IgG1 anti-FITC antibody levels, and injections on 5 consecutive days were required to induce a maximal anti-FITC response. These observations provide evidence that sustained Ag-specific T cell help is required to stimulate the degree of B cell clonal expansion that characterizes a specific antibody response.     

A more sensitive enzyme immunoassay of anti-insulin IgG in guinea pig serum with less non-specific binding of normal guinea pig IgG

An enzyme immunoassay of anti-insulin IgG in guinea pig serum was improved in sensitivity by reducing the non-specific binding of normal guinea pig IgG and enhancing the specific binding of anti-insulin IgG. Silicone rubber pieces or polystyrene balls were coated with normal rabbit IgG, followed by coupling of insulin using glutaraldehyde. The insulin-normal rabbit IgG-coated silicone rubber pieces or polystyrene balls were incubated with normal rabbit IgG and then with diluted guinea pig anti-insulin serum in the presence of normal rabbit IgG at a lower temperature (20 degrees C). Finally, the solid phases were incubated with rabbit (anti-guinea pig IgG) Fab'-horseradish peroxidase conjugate to measure the amount of guinea pig IgG bound. The detection limit of anti-insulin IgG in guinea pig serum was improved 10 to 100-fold compared to that of enzyme immunoassay performed by incubating insulin-bovine serum albumin-coated solid phases with diluted guinea pig anti-insulin serum at 37 degrees C and then with rabbit (anti-guinea pig IgG) Fab' conjugated to beta-D-galactosidase from Escherichia coli, according to a previous report (Kato, K., et al. (1978) J. Biochem. 84, 93-102).