Clinical Chemistry AACC Online Job Center
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 46: 658-666, 2000;
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (25)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Eriksson, S.
Right arrow Articles by Lövgren, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Eriksson, S.
Right arrow Articles by Lövgren, T.
Related Collections
Right arrow Proteomics and Protein Markers
(Clinical Chemistry. 2000;46:658-666.)
© 2000 American Association for Clinical Chemistry, Inc.

Articles

Dual-Label Time-resolved Immunofluorometric Assay of Free and Total Prostate-specific Antigen Based on Recombinant Fab Fragments

Susann Erikssona,1, Markus Vehniäinen1, Tove Jansén1, Ville Meretoja1, Petri Saviranta1, Kim Pettersson1 and Timo Lövgren1

1 Department of Biotechnology, University of Turku, Tykistökatu 6, FIN-20520 Turku, Finland.
a Author for correspondence. Fax 358-2-333 8050; e-mail susann.eriksson{at}utu.fi


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Recombinant Fab fragments are attractive as reagents for novel sandwich immunoassays, but no such assays have been described. We developed a dual-label two-site immunoassay based entirely on recombinant Fab fragments and compared it to the same assay with intact monoclonal antibodies.

Methods: The capture Fab fragment, which binds free prostate-specific antigen (PSA) and PSA in complex with {alpha}1-antichymotrypsin on an equimolar basis, is site-specifically biotinylated and attached to the solid phase in streptavidin-coated microtitration wells. The Fab fragment that detects only free PSA is site-specifically labeled with a fluorescent europium chelate, and the Fab fragment that detects both free and serpin-complexed PSA in an equimolar fashion is labeled with a fluorescent terbium chelate. Time-resolved fluorescence is used to measure both europium and terbium signals in one well.

Results: The detection limits of the assay (mean + 3 SD of zero calibrator) were 0.043 and 0.28 µg/L, respectively, for free and total PSA. The within-run and day-to-day CVs were 2–11% and 4–10%, respectively. Mean recoveries were 93% and 98% in female and male sera, respectively. Compared with the commercial ProStatus PSA Free/Total Assay, the intercepts of the regression equations (r >0.99) were not significantly different from zero, and the slopes were 0.95–1.01. In one female serum sample, PSA was falsely increased with the monoclonal assay but was undetectable with the recombinant assay.

Conclusions: The performance of this novel assay based on recombinant components is comparable to a conventional assay based on monoclonal antibodies. The more complete control of the essential characteristics of site-specifically derivatized recombinant Fab fragments will be valuable for the design of miniaturized and multianalyte assay concepts where correct antibody orientation, density, and capacity as well as uncompromised binding affinity are required.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Monoclonal antibodies have been used in immunoassays since the beginning of the 1980s. Lately, antibody fragments such as Fab and single-chain Fv fragments expressed in Escherichia coli have become potential substitutes for the polyclonal sera and monoclonal antibodies used in research and therapy (1). These recombinant fragments have several potential advantages that can make them valuable tools for future diagnostic concepts. The production of recombinant antibody fragments is less time-consuming and less expensive than the production of polyclonal sera and monoclonal antibodies. Because antibody fragments have been produced as soluble, correctly processed, and active proteins in E. coli in titers of grams per liter (2)(3), the reagent requirements of the diagnostic industry can be easily satisfied. In addition, antibody fragments with different binding specificities can be isolated from large naïve phage display libraries, thus bypassing immunization of animals (4). Recombinant antibody fragments can also be derived by cloning from hybridoma cultures (5). The antigen recognition can be improved in vitro by site-directed or random mutagenesis (6)(7). Peptide tails and amino acid modifications can be introduced by genetic engineering to simplify the purification and labeling of recombinant proteins (8)(9)(10).

The recent progress within the field of immunodiagnostics toward miniaturized assay systems calls for carefully controlled properties of the antibodies used. The simple, passive, and random immobilization of antibodies to plastic surfaces, although successfully used in most of today’s immunoassay systems, is highly inefficient because only a minor fraction of the antibodies remains fully functional (11). Especially in cases where the surface area is limited, efficient immobilization of the antibody molecules in the correct orientation is required. At the same time, the tracer antibody should ideally have a high specific activity without negatively affecting the recognition of the antigen. Because random chemical labeling may easily disturb the antigen binding site (12), more specific and controllable methods for derivatization of antibodies are needed. One approach to solve this problem is the introduction of cysteine residues by genetic engineering of the antibody fragment, which then can be labeled by thiol-specific reagents (10)(13)(14). Immobilization can as well be performed by site-specific biotinylation of the capture antibody [Ref. (13); and Meretoja et al., manuscript in preparation] and attachment to streptavidin-coated wells. A small albumin-binding domain has also been used for the directed attachment of recombinant antibody fragments to microtiter plates coated with human serum albumin (15).

Anti-animal antibodies, complement, or rheumatoid factors present in the sera of some patients can react with the antibodies used in immunoassays, causing either false-positive or -negative test results [reviewed in (16)(17)]. In two-site assays, human anti-mouse antibodies and rheumatoid factors can react with the capture and detection antibodies, linking them together even in the absence of analyte (18). On the other hand, complement factors reacting with solid-phase antibodies may interfere with antigen binding, causing spuriously decreased values (19)(20)(21). A false-positive or -negative result may lead to unnecessary or harmful further investigations and interventions, which can be psychologically stressful for the patient. In large-scale screening tests, a reduction of false results could also be of considerable financial importance. Recombinant antibodies are produced as fragments (Fab or scFv) lacking the Fc part of the antibody, which is a major source of unwanted interferences.

Our aim was to develop an immunoassay based entirely on recombinant Fab fragments produced in E. coli and to compare this assay with an assay based on monoclonal antibodies produced in hybridoma cell culture as whole IgG molecules. Prostate-specific antigen (PSA)1 was chosen as a model analyte that provides an additional challenge because it exists both as free PSA (F-PSA) and in complex with {alpha}1-antichymotrypsin (ACT) (22). The determination of serum PSA concentrations has been proposed as a screening test for prostate cancer (CAP) (23), and the discrimination between benign prostatic hyperplasia and CAP can be improved by determination of the ratio of free to total PSA (24). Another purpose of this study was to investigate the feasibility and potential advantages of using site-specifically derivatized Fab fragments in a dual-label two-site immunoassay. As the target for our study, we selected the antibodies of the commercial ProStatus PSA Free/Total (F/T) assay, which was also used as our reference assay. The detection technology used was that of the ProStatus F/T assay, namely the use of time-resolved fluorometry of lanthanide-based chelates (25).


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
recombinant Fab FRAGMENTS
The recombinant anti-PSA Fab fragments were cloned from the hybridoma cell lines H117, H50, and 5A10 by reverse transcription-PCR. Of these, H117 and H50 recognize both free and complexed PSA, whereas 5A10 recognizes only F-PSA (26). H117, which was used as the capture antibody, was genetically engineered to increase the bacterial expression (Jansén et al., manuscript in preparation) and to insert a free cysteine residue at the C-terminal end of the Fd chain (Meretoja et al., manuscript in preparation). H50 and 5A10 were engineered by PCR to contain a Thr-Ser-Cys-Pro-His6 tag at the C-terminal end of the Fd-chain, using the forward primer 5'-TAC TGG GGC CAA GG-3' and the reverse primer 5'-ACA TGC GAG CTC TTA ATG GTG ATG GTG ATG GTG AGG ACA ACT AGT ACA ATC CCT-3'. The product was digested with BamHI and SacI (SacI site is underlined in the reverse primer) and ligated into the similarly treated vector containing the original sequences. The DNA sequences of the final constructs were verified by sequencing the modified regions. The Fab fragments were produced in E. coli, and a periplasmic extract was prepared and purified with a cation-exchange chromatography column (SP STREAMLINE; Pharmacia Biotech), a Ni-NTA Superflow column (Qiagen), and a HiTrap Protein G column (Pharmacia Biotech; Meretoja et al., manuscript in preparation).

labeling with lanthanide chelates and biotinylation
The europium and terbium chelates used for labeling of the detection antibodies were kindly provided by Wallac Labeling Service (Turku, Finland). Two different chelates were used for labeling: a europium(III) chelate of 2,2',2'',2'''-[[4-[4-(iodoacetamido)phenylethynyl]pyridine-2,6-diyl]bis(methylenenitrilo)]-tetrakis(acetic acid) (27) and the terbium(III) chelate of N1 -(p-iodoacetamidobenzyl)-diethylenetriamine-N1 ,N2,N3,N3-tetraacetic acid (28). Before labeling, the purified Fab fragments were stored at -20 °C in 50 mmol/L Tris-HCl (pH 7.75), 15 mmol/L NaCl, 0.5 g/L NaN3 supplemented with 1 mmol/L dithiothreitol and 2 mmol/L EDTA. Immediately before labeling, the buffer was changed to 50 mmol/L carbonate buffer, pH 8.3, using a NAP-10 column (Pharmacia Biotech), and the protein was concentrated to ~1 g/L using a Centricon-10 concentrator (Amicon).

The labeling of Fab fragments was performed with a 30-fold molar excess of label reagent at 4 °C for 4 h. The labeled Fab fragment was separated from excess free label on a Superdex 200 HR 10/30 gel filtration column (Pharmacia Biotech) equilibrated and run with 50 mmol/L Tris-HCl (pH 7.75), 15 mmol/L NaCl, 0.5 g/L NaN3 at 25 mL/h, and 0.5-mL fractions were collected. The fractions containing labeled protein were pooled, the protein concentration was measured by absorbance, and the degree of labeling was determined using a europium or terbium calibration solution (Wallac). Bovine serum albumin was added to a final concentration of 1 g/L, and the solution was filtered through a 0.22 µm pore size filter (Millipore) and stored at 4 °C. The capture Fab fragment (0.2 g/L) was biotinylated with 50 µmol/L 3-(N-maleimido-propionyl)-biocytin (Sigma) (29) as described (Meretoja et al., manuscript in preparation).

commercial immunoassay reagents and equipment
All immunoassays were performed using the DELFIA® technology (Wallac). The following DELFIA reagents and equipment were used: streptavidin-coated strips, Assay Buffer, Wash Concentrate, Enhancement Solution, Enhancer for Tb and Dy measurement, ProStatus PSA F/T kit, Plate Wash, Plateshake, Plate Dispenser, and a 1234 Research Fluorometer.

assay procedures
The principle of the dual-label assay based on recombinant Fab fragments for measurement of free and total PSA (T-PSA) is shown in Fig. 1 , and it will hereafter be referred to as the "rFab F/T PSA assay". The biotinylated capture Fab was immobilized to the streptavidin-coated wells. After a prewash, 160 ng of biotinylated capture Fab per well was added in Assay Buffer supplemented with 2 nmol/L free d-biotin and incubated for 30 min at room temperature with slow shaking. After the wells were washed four times, 25 µL of the calibrators (PSA calibrators from the ProStatus PSA F/T kit) or serum samples in duplicate and 100 ng of each tracer Fab fragment in 100 µL of ProStatus PSA Assay Buffer were added. The plate was incubated for 30 min at room temperature with slow shaking. After the wells were washed six times, 200 µL of Enhancement Solution was added to each well, the plate was incubated for 30 min at room temperature with slow shaking, and the europium signal was measured with the fluorometer. In the dual-label assay, 50 µL of Enhancer for Tb and Dy measurement was then added, the plate was incubated for 5 min, and the terbium signal was measured. The MultiCalc immunoassay program created a spline-fitted calibration curve and calculated the unknown PSA concentrations. The assay used as reference, ProStatus PSA F/T Assay, was performed according to the instructions of the manufacturer.



View larger version (39K):
[in this window]
[in a new window]
  		Figure 1. Principle of the dual-label assay measuring both free and total PSA.

The biotinylated capture H117 Fab fragment recognizes an epitope accessible on both the free and complexed forms of PSA. The Tb3+-labeled H50 Fab fragment recognizes a nonoverlapping epitope accessible on both the free and complexed forms of PSA and measures the T-PSA in the sample. The Eu3+-labeled 5A10 Fab fragment recognizes an epitope present only on F-PSA, thus measuring only the F-PSA in the sample.

labeling of psa
PSA purified from seminal plasma (30) was labeled with europium using the DELFIA Eu-Labeling kit. Briefly, 0.5 mg of PSA in 50 mmol/L carbonate buffer, pH 9.8, was incubated with a 20-fold molar excess of label reagent in a final volume of 500 µL overnight at 4 °C. The labeled PSA was purified similarly to the labeled Fab fragments. The labeling degree was 2.8 Eu3+/PSA molecule.

determination of affinity constants
The affinity of the lanthanide-labeled Fab fragments was determined according to the method of Scatchard (31) and as described previously (32). To prewashed H117-coated strips from the ProStatus Kit, 25 µL of 40 µg/L PSA or calibration diluent (0 µg/L PSA) was added together with 200 µL of six dilutions of labeled Fab (5A10 or H50) between 3.9 and 125 µg/L. The strips were incubated for 3 h and then washed, and the fluorescence was measured after development in Enhancement Solution. The affinity of H117 Fab was determined by incubating 200 µL of 0 or 5 µg/L biotinylated H117 Fab in anti-mouse IgG-coated strips for 2 h, followed by washes. Six dilutions of Eu3+-labeled PSA (1.5–60 µg/L; 200 µL) were added and incubated for 2 h; the strips were then washed, and the fluorescence was measured after development in Enhancement Solution. The signals obtained after subtracting the nonspecific binding from the total binding were used to calculate the affinity of the labeled Fab fragment.

samples
Serum samples were kindly provided by Dr. Franz Recker at the Clinic of Urology, Kantonsspital Aarau, Aarau, Switzerland. Fifteen of the serum samples were from healthy men, 22 were from patients with benign prostatic hyperplasia, and 35 were from patients diagnosed with CAP. Additional serum samples from patients with unspecified or suspected urological disorders were kindly provided by Dr. Hans Lilja at the Department of Clinical Chemistry, Lund University, Malmö, Sweden. Ten serum samples used for analytical recovery tests were collected from female volunteers at the Department of Biotechnology, University of Turku, Turku, Finland. One serum sample, known to cause false-positive results in several assays, was obtained from Wallac, Turku, Finland. The procedures followed were in accordance with the Helsinki Declaration of 1975. The samples were stored at -20 °C and thawed just before use.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
optimization of assay
Labeling degree and amount of antibodies.
The labeling degree of the Eu3+-labeled Fab fragments was 0.7–0.8 Eu3+/Fab, and that of the Tb3+-labeled Fab was 0.3 Tb3+/Fab. The assay optimization (e.g., amount of Fab fragment, and kinetics) and initial serum sample tests were done with both tracer Fab fragments labeled with Eu3+ and assaying the free and total PSA in separate wells. With 160 ng/well of capture Fab, the streptavidin surface was essentially saturated. The optimal amount of both tracer Fabs was 100 ng/well added in ProStatus PSA Assay Buffer. The volume of calibrators and samples was invariably 25 µL, and no other sample volumes were tested.

Kinetic experiments.
The results from the kinetic experiment showed that a 30-min incubation of sample and tracer Fab at room temperature gave >90% of the maximum (1 h) signal with concentrations of 2–250 µg/L PSA (Fig. 2 ). The 30-min incubation time at room temperature was chosen for further assays.



View larger version (23K):
[in this window]
[in a new window]
  		Figure 2. Kinetics of the rFab F/T PSA assay using three different concentrations of F-PSA (A) and T-PSA (B).

The maximum signal was defined as the signal obtained after 1 h of incubation.

assay characteristics
Calibration curve.
A typical calibration curve based on 12 replicates of the calibrators is shown in Fig. 3 . The signal from the Tb3+-labeled Fab was ~10-fold lower than that of the Eu3+-labeled Fab, as a consequence of both the lower labeling degree (0.3 compared with 0.7) and the lower fluorescence intensity (specific activity) for the terbium chelate. The linear response using the rFab F/T PSA assay extended beyond 1000 µg/L for both the free and total PSA assay, even if the highest ProStatus PSA F/T kit calibrator was 250 µg/L.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 3. Calibration curves (filled symbols) and within-assay precision profiles (open symbols) for the dual-label assay of F-PSA ({blacktriangleup} and {triangleup}) and T-PSA (• and {circ}).

Detection limit.
The lower limit of detection was defined as the concentration corresponding to a signal 3 SD above the mean of 12 replicates of the zero calibrator (mean values, 200 and 150 cps, respectively, for the free and total PSA assays). The detection limit of the dual-label assay was 0.043 µg/L for F-PSA and 0.280 µg/L for T-PSA. When Eu3+-labeled Fab was used for measurement of T-PSA (single-label assay), the detection limit was 0.070 µg/L.

Within-assay variability.
The within-assay variability was determined by performing the assay according to the routine protocol both with calibrators (n = 12) and with serum samples (n = 12). The CVs based on calculated concentration for the calibrators were between 15% and 1.5% for F-PSA (F-PSA, 0.09–249 µg/L) and between 22% and 2.6% for T-PSA (T-PSA, 0.51–257 µg/L) as shown in Fig. 3Up . The CVs for serum samples varied between 1.9% and 11% as shown in Table 1 .


View this table:
[in this window]
[in a new window]
  		Table 1. Assay variability.

Between-assay variability.
The between-assay variability was studied by performing the same assay with calibrators and samples in duplicate for 10 days. The CVs were 4.2–10% for the serum samples. The results are shown in Table 1Up .

Analytical recovery.
The analytical recovery was studied by adding purified PSA-ACT to male and female serum samples to a final concentration of 10 µg/L, and performing the assay to determine the T-PSA concentration in the samples with and without added PSA-ACT. The results are shown in Table 2 . The average recovery of added analyte was 93% in 10 female serum samples and 98% in 9 male serum samples. One male and one female sample gave low recoveries, 68% and 59%, respectively, and the same results were obtained with the ProStatus F/T PSA assay. The recovery test was not performed with F-PSA because free, enzymatically active PSA is captured by serine protease inhibitors, mainly {alpha}2-macroglobulin, thus escaping immunodetection (21).


View this table:
[in this window]
[in a new window]
  		Table 2. Analytical recovery of PSA-ACT complex in sera.

Effect of dilution.
To study the effect of dilution, three serum samples with high PSA concentrations were diluted in serum samples with low PSA concentrations, and the results were compared with the calculated results. The measured results correlated well with the calculated results, as shown in Fig. 4 .



View larger version (19K):
[in this window]
[in a new window]
  		Figure 4. Effect of dilution.

The calculated PSA value was compared with the measured PSA value when serum samples with high PSA concentrations were diluted with serum samples with low PSA concentrations. The points represent the measured values, and the lines represent the calculated values for the three different dilution series (e.g., Serum H1/Serum L1, where H stands for high PSA and L for low PSA concentrations). (A), F-PSA; (B), T-PSA.

affinity constants
The affinity constants of the labeled Fab fragments were 5.3 x 109 L/mol for H117 Fab, 4.4 x 109 L/mol for 5A10 Fab, and 1.1 x 109 L/mol for H50 Fab. The affinity constants of the monoclonal antibodies were 1.0 x 1010 L/mol for H117, 7.4 x 109 L/mol for 5A10, and 1.2 x 109 L/mol for H50.

correlation
The correlation between the values obtained by the rFab F/T PSA assay and the ProStatus PSA F/T assay was studied by analyzing 95 serum samples on the same day. The results from both assays were compared and analyzed using linear regression plots (Fig. 5 ). The correlation coefficients were excellent for both the whole PSA range, giving r values of 1.000 for F-PSA and 0.999 for T-PSA, and at PSA concentrations of F-PSA <1 µg/L and T-PSA <5 µg/L, with r values of 0.994 and 0.992, respectively. The y-intercepts were insignificant, and the slopes were between 0.95 and 1.01. One sample from a female, known to give false-positive signals in several different immunoassays regardless of the analyte measured, gave no signal in the rFab F/T PSA assay but gave PSA values of 5.00 and 46.33 µg/L for free and total PSA, respectively, in the ProStatus PSA F/T assay. This sample was not included in calculations of linear regression and correlation.



View larger version (22K):
[in this window]
[in a new window]
  		Figure 5. Linear regression analysis of the results of PSA measurements.

Comparison of PSA (F-PSA in A and T-PSA in B) in sera measured with the ProStatus PSA F/T assay (on the x axis) and the dual-label rFab F/T PSA assay (on the y axis). The equations of the lines of regression and the correlation coefficients are indicated. The female problem sample is indicated with • and was not included in the regression analysis. The insets show the correlation at low PSA concentrations, <1 µg/L for F-PSA and <5 µg/L for T-PSA. For F-PSA (A): y = 0.953x + 0.038 (Sy|x = 0.133; 95% confidence interval for slope, 0.953 ± 0.005; 95% confidence interval for intercept, 0.038 ± 0.030; n = 94). For F-PSA <1 µg/L (inset in A): y = 0.999x - 0.028 (Sy|x = 0.029; 95% confidence interval for slope, 0.999 ± 0.031; 95% confidence interval for intercept, -0.028 ± 0.015; n = 55). For T-PSA (B): y = 1.009x - 0.368 (Sy|x = 1.077; 95% confidence interval for slope, 1.009 ± 0.008; 95% confidence interval for intercept, -0.368 ± 0.267; n = 94). For T-PSA <5 µg/L (inset in B): y = 0.953x - 0.108 (Sy|x = 0.1656; 95% confidence interval for slope, 0.953 ± 0.035; 95% confidence interval for intercept, -0.108 ± 0.089; n = 53).

stability
The stabilities of the biotinylated and lanthanide-labeled Fab fragments were studied by storing each Fab fragment (at concentrations between 126 and 243 mg/L) at 4 and 35 °C for 1 week, 3 weeks, and 3 months before they were used in the assay. The calibration curves obtained were compared with the calibration curve obtained before the start of the storage experiment. No decrease in activity was observed; even the Fab fragments stored for 3 months at 35 °C were indistinguishable from the control.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
To our best knowledge, this is the first report of a sandwich-type immunoassay based completely on the use of recombinant Fab fragments. A recombinant ELISA system based on single-chain Fv fusion proteins as coating and detecting reagents has been described previously for the detection of the plant pathogen, beet necrotic yellow vein virus (33). Our assay is a time-resolved immunofluorometric assay that utilizes the DELFIA technology for the simultaneous detection of free and total PSA in one well, using recombinantly produced Fab fragments of the monoclonal antibodies used in the commercial ProStatus PSA F/T kit. Three different Fab fragments were used: one for capture, and two for detection of the two measured forms of PSA. All of the Fab fragments were site-specifically modified at a free cysteine residue, which had been introduced by genetic engineering. The capture fragment was biotinylated with a thiol-specific biotinylating reagent, whereas the detection fragments were labeled with iodoacetate derivatives of two different lanthanide chelates. Fab fragments were chosen instead of single-chain antibodies because there have been reports of problems associated with the folding and stability of scFv fragments (34), and because Fab fragments were considered a more reliable alternative when this work was started. In addition, the detection of Fab fragments is easier than that of scFv because anti-mouse antibodies can recognize Fab fragments but not scFv. In the future, it would still be interesting to use scFv for comparison because it may further improve the solid-phase capacity.

The rFab F/T PSA assay correlated excellently with the ProStatus PSA F/T Assay, with correlation coefficients >0.99. However, the lower labeling degree of the Fab fragments compared with the parental intact monoclonal antibodies produced higher limits of detection. The higher detection limit in dual-label assays for T-PSA (0.280 vs 0.043 µg/L) is partly attributable to the known lower fluorescence yield of the Tb3+ chelate (35) and a lower degree of incorporation of the iodoacetate derivative. The reason for the latter is not known but may have been attributable to decreased reactivity of the thiol group. The single-label assay gave a lower detection limit (0.070 µg/L) for T-PSA with an Eu3+-labeled Fab fragment. The detection limits of the ProStatus PSA F/T assay are reported to be better than 0.04 and 0.10 µg/L for F-PSA and T-PSA, respectively, based on a detection limit of 2 SD above the mean value for the zero calibrator (ProStatus PSA F/T kit performance data). Despite the higher detection limits compared with some ultrasensitive assays, the present dual-label assay would be adequate for screening of CAP because the clinical cutoff is usually set at 4 µg/L T-PSA (36) or in some instances as low as 2.5 µg/L (37). However, the assay could not be used for monitoring of CAP patients for relapse after radical prostatectomy because the detection limit required in these cases is at least 0.1 µg/L (38)(39), but this is predominantly a consequence of the chelate used.

The expressed aim of the present study was to site-specifically derivatize the Fab fragments with biotin or lanthanide chelate in a 1:1 manner. This naturally presents a limitation in pursuing lower detection limits. This problem can be solved, however, by the development of branched structures of low molecular weight for introducing several label molecules (40). An alternative approach has been reported with streptavidin coupled to a labeled high-molecular weight protein to improve assay sensitivity (41). Another alternative could be the introduction of additional cysteine residues.

Immunoassays using recombinant antibody fragments, such as the present assay, might overcome some problems inherent in assays based on intact polyclonal or monoclonal antibodies. Unwanted reactions caused by human anti-mouse antibodies, complement, or rheumatoid factors may give rise to false-positive or -negative signals. Although such samples may not be very frequent, they can require substantial additional testing and cause psychological stress for the patient. Moreover, they introduce a certain element of uncertainty into the interpretation of all results generated with assays prone to these interferences. In this study, one female sample, known to give false-positive signals in several other two-site assays, gave an expected nondetectable value in the assay based on Fab fragments. Despite the inclusion of a large excess of neutralizing mouse antibodies into the assay buffer, the ProStatus PSA F/T assay based on monoclonal antibodies gave highly increased concentrations of both free and total PSA. Although high PSA concentrations have been reported previously in some female sera (42), the data we have for this sample indisputably show that it causes a false signal in immunoassays based on intact mouse immunoglobulins. Because the only difference between the assays is the absence or presence of the Fc part, the interfering factor is probably either human anti-mouse antibodies or rheumatoid factor that cross-link(s) the capture and detector by reacting with the Fc part of the intact antibody. The Fc part of the antibody molecule is generally known to cause nonspecific interactions, and the interference has been eliminated or substantially reduced by the use of proteolytic Fab or F(ab')2 fragmentation (43)(44). Proteolytic fragmentation of whole IgG molecules is not always straightforward and can be a tedious procedure compared with recombinant techniques, which directly provide a constant source of antibody fragments. Another approach to reduce interference was to use human/mouse chimeric antibodies, which was more effective in reducing interference than the addition of neutralizing mouse antibodies (45).

The low recovery seen with two serum samples is conceivably a result of the presence of autoantibodies to PSA present in these sera. Patients with benign prostatic hyperplasia have been reported to have autoantibodies to PSA (46), and women also could develop antibodies against PSA. Because the ProStatus PSA F/T assay with the same antibody-binding specificities gave equally low recoveries of PSA-ACT in these sera, this cannot be regarded as a problem introduced by the Fab fragments.

Because the Fab fragment accounts for only one-third of the IgG molecule, the coating of wells can be more efficient, and the site-specific biotinylation ensures that the fragments are in the correct orientation and able to bind antigen. This can be a great advantage in the future because the trend in immunodiagnostics is going toward miniaturized assay systems. A more dense capture antibody surface would be especially advantageous in situations where the signal is measured directly from the surface (47) compared with situations where the measurement is performed through a separate development step that integrates the signal from larger area into a homogeneous liquid phase. Such situations would be encountered in multianalyte assays where the different analytes are spatially separated on the solid-phase carrier (48). In the present study, we used microtiter wells passively coated with streptavidin to bind the biotinylated Fab fragments, and it is conceivable that the capacity of the immobilized streptavidin is the limiting factor for obtaining a denser antibody surface. Preliminary results indicate that the binding capacity for the anti-PSA Fab fragment can be increased severalfold compared with the intact monoclonal antibody by the use of high-capacity streptavidin surfaces (unpublished observations).

The small size of the Fab fragment may potentially give faster kinetics compared with the whole IgG molecule. Such an advantage would partly stem from the higher, more dense solid-phase capacity and the optimal antibody site orientation after the site-specific derivatization. In part, it would stem from the smaller size of the tracer and its potentially higher degree of access to the epitope on the antigen bound to the capture antibody. Efforts to systematically evaluate the effects of the capture and tracer Fab fragments on the kinetics of the assay were not carried out in this study. A 30-min incubation time was sufficient for the rFab F/T PSA assay, which used a one-step protocol after the immobilization of the biotinylated Fab fragment, compared with the commercial ProStatus PSA F/T kit, which is based on a sequential 1- plus 2-h incubation protocol. However, in a similar PSA assay using monoclonal antibodies, Mitrunen et al. (49) reported that maximum signals for both free and complexed PSA were obtained after a 30-min incubation at room temperature. By precoating the streptavidin wells with the capture Fab and storing the dry coated wells, the total assay time and number of steps can be reduced. The concept of the rFab F/T PSA assay is an ideal candidate for applying the all-in-one dry reagent described earlier, providing for an ultimately simplified and rapid assay protocol. Development of such an assay is currently being carried out.

A justified concern was the stability of the recombinant Fab fragments compared with the parental antibodies. This study has shown that the stability of Fab fragments, produced in E. coli and subsequently purified to homogeneity, is as good as that of monoclonal antibodies and, more importantly, that the stability is not affected by the site-specific labeling.

The small size of the Fab fragment compared with the IgG molecule makes it more likely that nonspecific conjugation of the Fab fragments will affect the antigen-binding site (50). The site-specific labeling with biotin and lanthanide chelates of the antibody fragments was fully successful in reducing this risk. We compared the affinities of the three labeled Fab fragments with the labeled monoclonal antibodies to reveal any alteration of the antigen binding. Some minor (twofold or less) differences could be seen, but these are not likely to be of any practical importance; rather they are probably a consequence of the somewhat inaccurate method used for the determination of affinity constants. The approximately twofold decrease in affinity of H117 Fab is presumably attributable to the genetic engineering performed to improve the bacterial expression (Jansén et al., manuscript in preparation).

In conclusion, we have demonstrated that the performance of an assay based entirely on recombinant Fab fragments is at least comparable to that of an assay based on monoclonal antibodies. Recombinant bacterial techniques for the establishment of new immunoreagents and the exact tailor-made modifications of these are likely to overcome some of the problems present in assays that use monoclonal antibodies or polyclonal sera, such as some sample-derived interferences, negative effects from antibody conjugations, inadequate capacities, and slow kinetics. The advantage of the carefully controlled qualities of recombinant antibodies in immunoassays will most probably increase during the years to come. This will especially be the case in the design of future multianalyte diagnostic systems where miniaturization of assays is combined with uncompromised analytical performance.


   Acknowledgments
 
We thank the Technology Development Centre of Finland (TEKES) for financial support.


   Footnotes
 
A portion of this work has been communicated previously as Eriksson S, Vehniäinen M, Jansén T, Meretoja V, Saviranta P, Lövgren T, Pettersson K. Dual-label time-resolved immunofluorometric assay of free and total PSA based on recombinant Fab fragments [Abstract]. Clin Chem Lab Med 1999;37(Special Supplement):S224.

1 Nonstandard abbreviations: PSA, prostate-specific antigen; F-PSA and T-PSA, free and total PSA; ACT, {alpha}1-antichymotrypsin; CAP, prostate cancer; and F/T, ratio of F-PSA to T-PSA.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Hayden MS, Gilliland LK, Ledbetter JA. Antibody engineering. Curr Opin Immunol 1997;9:201-212. [ISI][Medline] [Order article via Infotrieve]
  2. Carter P, Kelley RF, Rodrigues ML, Snedecor B, Covarrubias M, Velligan MD, et al. High level Escherichia coli expression and production of a bivalent humanized antibody fragment. Biotechnology 1992;10:163-167. [Medline] [Order article via Infotrieve]
  3. Horn U, Strittmatter W, Krebber A, Knüpfer U, Kujau M, Wenderoth R, et al. High volumetric yields of functional dimeric miniantibodies in Escherichia coli, using an optimized expression vector and high-cell-density fermentation under non-limited growth conditions. Appl Microbiol Biotechnol 1996;46:524-532. [ISI][Medline] [Order article via Infotrieve]
  4. Winter G, Griffiths AD, Hawkins RE, Hoogenboom HR. Making antibodies by phage display technology. Annu Rev Immunol 1994;12:433-455. [ISI][Medline] [Order article via Infotrieve]
  5. Winter G, Milstein C. Man-made antibodies [Review]. Nature 1991;349:293-299. [Medline] [Order article via Infotrieve]
  6. Saviranta P, Pajunen M, Jauria P, Karp M, Pettersson K, Mäntsälä P, Lövgren T. Engineering the steroid-specificity of an anti-17ß-estradiol Fab by random mutagenesis and competitive panning. Protein Eng 1998;11:143-152. [Abstract/Free Full Text]
  7. Chowdhury PS, Pastan I. Improving antibody affinity by mimicking somatic hypermutation in vitro. Nat Biotechnol 1999;17:568-572. [ISI][Medline] [Order article via Infotrieve]
  8. Lindner P, Bauer K, Krebber A, Nieba L, Kremmer E, Krebber C, et al. Specific detection of His-tagged proteins with recombinant anti-His tag scFv-phosphatase or scFv-phage fusions. Biotechniques 1997;22:140-149. [ISI][Medline] [Order article via Infotrieve]
  9. Knappik A, Plückthun A. An improved affinity tag based on the FLAG peptide for the detection and purification of recombinant antibody fragments. Biotechniques 1994;17:754-761. [ISI][Medline] [Order article via Infotrieve]
  10. Lyons A, King DJ, Owens RJ, Yarranton GT, Millican A, Whittle NR, Adair JR. Site-specific attachment to recombinant antibodies via introduced surface cysteine residues. Protein Eng 1990;3:703-708. [Abstract/Free Full Text]
  11. Butler JE, Ni L, Nessler R, Joshi KS, Suter M, Rosenberg B, et al. The physical and functional behavior of capture antibodies adsorbed on polystyrene. J Immunol Methods 1992;150:77-90. [ISI][Medline] [Order article via Infotrieve]
  12. Rodwell JD, Alvarez VL, Lee C, Lopes AD, Goers JW, King HD, et al. Site-specific covalent modification of monoclonal antibodies: in vitro and in vivo evaluations. Proc Natl Acad Sci U S A 1986;83:2632-2636. [Abstract/Free Full Text]
  13. Kipriyanov SM, Dübel S, Breitling F, Kontermann RE, Little M. Recombinant single-chain Fv fragments carrying C-terminal cysteine residues: production of bivalent and biotinylated miniantibodies. Mol Immunol 1994;31:1047-1058. [ISI][Medline] [Order article via Infotrieve]
  14. Neri D, Natali PG, Petrul H, Soldani P, Nicotra MR, Vola R, et al. Recombinant anti-human melanoma antibodies are versatile molecules. J Investig Dermatol 1996;107:164-170. [ISI][Medline] [Order article via Infotrieve]
  15. König T, Skerra A. Use of an albumin-binding domain for the selective immobilisation of recombinant capture antibody fragments on ELISA plates. J Immunol Methods 1998;218:73-83. [ISI][Medline] [Order article via Infotrieve]
  16. Weber TH, Käpyaho KI, Tanner P. Endogenous interference in immunoassays in clinical chemistry. A review. Scand J Clin Lab Investig 1990;50(Suppl 201):77-82. [ISI][Medline] [Order article via Infotrieve]
  17. Kricka LJ. Human anti-animal antibody interferences in immunological assays [Review]. Clin Chem 1999;45:942-956. [Abstract/Free Full Text]
  18. Kricka LJ, Schmerfeld-Pruss D, Senior M, Goodman DB, Kaladas P. Interference by human anti-mouse antibody in two-site immunoassays. Clin Chem 1990;36:892-894. [Abstract/Free Full Text]
  19. Käpyaho K, Tanner P, Weber T. Effect of complement binding on a solid-phase immunometric TSH assay. Scand J Clin Lab Investig 1989;49:211-215. [ISI][Medline] [Order article via Infotrieve]
  20. Vanham G, Bloemmen FJ, Ceuppens JL, Stevens EA. Influence of serum complement and rheumatoid factor on detection of immune complexes by the C1q and monoclonal rheumatoid factor solid-phase assay. J Immunol Methods 1984;73:301-311. [ISI][Medline] [Order article via Infotrieve]
  21. Börmer OP. Interference of complement with the binding of carcinoembryonic antigen to solid-phase monoclonal antibodies. J Immunol Methods 1989;121:85-93. [ISI][Medline] [Order article via Infotrieve]
  22. Lilja H, Christensson A, Dahlén U, Matikainen MT, Nilsson O, Pettersson K, Lövgren T. Prostate-specific antigen in serum occurs predominantly in complex with {alpha}1-antichymotrypsin. Clin Chem 1991;37:1618-1625. [Abstract/Free Full Text]
  23. Catalona WJ, Smith DS, Ratliff TL, Basler JW. Detection of organ-confined prostate cancer is increased through prostate-specific antigen-based screening. JAMA 1993;270:948-954. [Abstract]
  24. Christensson A, Björk T, Nilsson O, Dahlén U, Matikainen MT, Cockett AT, et al. Serum prostate specific antigen complexed to {alpha}1-antichymotrypsin as an indicator of prostate cancer. J Urol 1993;150:100-105. [ISI][Medline] [Order article via Infotrieve]
  25. Lövgren T, Pettersson K. Time-resolved fluoroimmunoassay, advantages and limitations. Van Dyke K Van Dyke R eds. Luminescence immunoassay and molecular applications 1990:233-253 CRC Press Boca Raton, FL. .
  26. Lövgren J, Piironen T, Övermo C, Dowell B, Karp M, Pettersson K, et al. Production of recombinant PSA and hK2 and analysis of their immunologic cross-reactivity. Biochem Biophys Res Commun 1995;213:888-895. [ISI][Medline] [Order article via Infotrieve]
  27. Takalo H, Mukkala VM, Mikola H, Liitti P, Hemmilä I. Synthesis of europium (III) chelates suitable for labeling of bioactive molecules. Bioconjug Chem 1994;5:278-282. [ISI][Medline] [Order article via Infotrieve]
  28. Mukkala VM, Mikola H, Hemmilä I. The synthesis and use of activated N-benzyl derivatives of diethylenetriaminetetraacetic acids: alternative reagents for labeling of antibodies with metal ions. Anal Biochem 1989;176:319-325. [ISI][Medline] [Order article via Infotrieve]
  29. Bayer EA, Zalis MG, Wilchek M. 3-(N-Maleimido-propionyl)biocytin: a versatile thiol-specific biotinylating reagent. Anal Biochem 1985;149:529-536. [ISI][Medline] [Order article via Infotrieve]
  30. Christensson A, Laurell CB, Lilja H. Enzymatic activity of prostate-specific antigen and its reactions with extracellular serine proteinase inhibitors. Eur J Biochem 1990;194:755-763. [ISI][Medline] [Order article via Infotrieve]
  31. Scatchard G. The attractions of proteins for small molecules and ions. Ann N Y Acad Sci 1949;51:660-672. [ISI]
  32. Pettersson KS, Söderholm JR. Individual differences in lutropin immunoreactivity revealed by monoclonal antibodies. Clin Chem 1991;37:333-340. [Abstract/Free Full Text]
  33. Kerschbaumer RJ, Hirschl S, Kaufmann A, Ibl M, Koenig R, Himmler G. Single-chain Fv fusion proteins suitable as coating and detecting reagents in a double antibody sandwich enzyme-linked immunosorbent assay. Anal Biochem 1997;249:219-227. [ISI][Medline] [Order article via Infotrieve]
  34. Raag R, Whitlow M. Single-chain Fvs [Review]. FASEB J 1995;9:73-80. [Abstract]
  35. Hemmilä I, Mukkala VM, Latva M, Kiilholma P. Di- and tetracarboxylate derivatives of pyridines, bipyridines and terpyridines as luminogenic reagents for time-resolved fluorometric determination of terbium and dysprosium. J Biochem Biophys Methods 1993;26:283-290. [ISI][Medline] [Order article via Infotrieve]
  36. Catalona WJ, Richie JP, Ahmann FR, Hudson MA, Scardino PT, Flanigan RC, et al. Comparison of digital rectal examination and serum prostate specific antigen in the early detection of prostate cancer: results of a multicenter clinical trial of 6,630 men. J Urol 1994;151:1283-1290. [ISI][Medline] [Order article via Infotrieve]
  37. Smith DS, Carvalhal GF, Mager DE, Bullock AD, Catalona WJ. Use of lower prostate specific antigen cutoffs for prostate cancer screening in black and white men. J Urol 1998;160:1734-1738. [ISI][Medline] [Order article via Infotrieve]
  38. Yu H, Diamandis EP, Prestigiacomo AF, Stamey TA. Ultrasensitive assay of prostate-specific antigen used for early detection of prostate cancer relapse and estimation of tumor-doubling time after radical prostatectomy. Clin Chem 1995;41:430-434. [Abstract/Free Full Text]
  39. Yu H, Diamandis EP, Wong PY, Nam R, Trachtenberg J. Detection of prostate cancer relapse with prostate specific antigen monitoring at levels of 0.001 to 0. 1 µg/L. J Urol 1997;157:913-918.
  40. Takalo H, Mukkala VM, Hurskainen P. Chelate clusters—new type of luminescent lanthanide labels [Short Oral Presentation]. Second International Conference of f-Elements, August 1–6, 1994, Helsinki, Finland, OG2:126..
  41. Diamandis EP, Morton RC, Reichstein E, Khosravi MJ. Multiple fluorescence labeling with europium chelators. Application to time-resolved fluoroimmunoassays. Anal Chem 1989;61:48-53. [Medline] [Order article via Infotrieve]
  42. Borchert GH, Giai M, Diamandis EP. Elevated levels of prostate-specific antigen in serum of women with fibroadenomas and breast cysts [Letter]. J Natl Cancer Inst 1997;89:587-588. [Free Full Text]
  43. Kato K, Umeda U, Suzuki F, Hayashi D, Kosaka A. Use of antibody Fab' fragments to remove interference by rheumatoid factors with the enzyme-linked sandwich immunoassay. FEBS Lett 1979;102:253-256. [ISI][Medline] [Order article via Infotrieve]
  44. Vaidya HC, Beatty BG. Eliminating interference from heterophilic antibodies in a two-site immunoassay for creatine kinase MB by using F(ab')2 conjugate and polyclonal mouse IgG. Clin Chem 1992;38:1737-1742. [Abstract/Free Full Text]
  45. Kuroki M, Matsumoto Y, Arakawa F, Haruno M, Murakami M, Kuwahara M, et al. Reducing interference from heterophilic antibodies in a two-site immunoassay for carcinoembryonic antigen (CEA) by using a human/mouse chimeric antibody to CEA as the tracer. J Immunol Methods 1995;180:81-91. [ISI][Medline] [Order article via Infotrieve]
  46. Zisman A, Zisman E, Lindner A, Velikanov S, Siegel YI, Mozes E. Autoantibodies to prostate specific antigen in patients with benign prostatic hyperplasia. J Urol 1995;154:1052-1055. [ISI][Medline] [Order article via Infotrieve]
  47. Christopoulus TK, Lianidou ES, Diamandis EP. Ultrasensitive time-resolved fluorescence method for {alpha}-fetoprotein. Clin Chem 1990;36:1497-1502. [Abstract/Free Full Text]
  48. Ekins RP, Chu FW. Multianalyte microspot immunoassay—microanalytical "compact disk" of the future [Review]. Clin Chem 1991;37:1955-1967. [Abstract/Free Full Text]
  49. Mitrunen K, Pettersson K, Piironen T, Björk T, Lilja H, Lövgren T. Dual-label one-step immunoassay for simultaneous measurement of free and total prostate-specific antigen concentrations and ratios in serum. Clin Chem 1995;41:1115-1120. [Abstract/Free Full Text]
  50. Hemminki A, Hoffrén AM, Takkinen K, Vehniäinen M, Mäkinen ML, Pettersson K, et al. Introduction of lysine residues on the light chain constant domain improves the labelling properties of a recombinant Fab fragment. Protein Eng 1995;8:185-191. [Abstract/Free Full Text]



The following articles in journals at HighWire Press have cited this article:


Home page
 Clin. Chem.Home page
Y. Xu and Q. Li
Multiple Fluorescent Labeling of Silica Nanoparticles with Lanthanide Chelates for Highly Sensitive Time-Resolved Immunofluorometric Assays
Clin. Chem., August 1, 2007; 53(8): 1503 - 1510.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
S. Zhang, C. Zhang, Z. Xing, and X. Zhang
Simultaneous Determination of {alpha}-Fetoprotein and Free {beta}-Human Chorionic Gonadotropin by Element-Tagged Immunoassay with Detection by Inductively Coupled Plasma Mass Spectrometry
Clin. Chem., July 1, 2004; 50(7): 1214 - 1221.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
P. Tarkkinen, T. Palenius, and T. Lovgren
Ultrarapid, Ultrasensitive One-Step Kinetic Immunoassay for C-Reactive Protein (CRP) in Whole Blood Samples: Measurement of the Entire CRP Concentration Range with a Single Sample Dilution
Clin. Chem., February 1, 2002; 48(2): 269 - 277.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
H. Harma, P. Tarkkinen, T. Soukka, and T. Lovgren
Miniature Single-Particle Immunoassay for Prostate-specific Antigen in Serum Using Recombinant Fab Fragments
Clin. Chem., November 1, 2000; 46(11): 1755 - 1761.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
A. Scorilas, A. Bjartell, H. Lilja, C. Moller, and E. P. Diamandis
Streptavidin-Polyvinylamine Conjugates Labeled with a Europium Chelate: Applications in Immunoassay, Immunohistochemistry, and Microarrays
Clin. Chem., September 1, 2000; 46(9): 1450 - 1455.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (25)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Eriksson, S.
Right arrow Articles by Lövgren, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Eriksson, S.
Right arrow Articles by Lövgren, T.
Related Collections
Right arrow Proteomics and Protein Markers