|
|
||||||||
Articles |
Department of Chemistry and Biochemistry, University of Windsor, 401 Sunset Ave., Windsor, Ontario, N9B 3P4 Canada.
a Author for correspondence. Fax 519-973-7098; e-mail tkc{at}uwindsor.ca
| Abstract |
|---|
|
|
|---|
Methods: The DNA label consisted of a T7 RNA polymerase promoter, a firefly luciferase-coding sequence, and a poly(dA/dT) tail. The 3' end of the DNA label was biotinylated and complexed with streptavidin. A sandwich immunoassay for prostate-specific antigen (PSA) was developed in which the antigen was first bound to an immobilized monoclonal antibody and then reacted with a biotinylated polyclonal antibody. The streptavidin-luciferase-coding DNA complex was then bound to the immunocomplex. The DNA label was subsequently expressed in vitro by coupled transcription and translation. The generated luciferase was measured by its characteristic bioluminescent reaction.
Results: The bioluminescence was linearly related to the concentration of PSA in the sample. As low as 30 ng/L PSA was measured (12.5-µL sample) with a signal-to-background ratio of 2.3, and the linear range extended to 3 µg/L. The results obtained from the proposed assay agreed well to those determined by IMx immunoassay (y = 0.98x + 0.74 µg/L; r = 0.971; n = 44).
Conclusions: The use of the newly developed DNA label in a two-site immunoassay was demonstrated for the first time. The assay was applied successfully to the measurement of serum PSA.
| Introduction |
|---|
|
|
|---|
Luciferase is inactivated by the chemical reactions involved in direct labeling of antigens or antibodies (9). Consequently, its use as a label in immunoassay is limited. The luciferase-catalyzed reaction, however, has been exploited as an indicator reaction in immunoassay. In one assay configuration (10), acetate kinase is used as a label, and the generated ATP is monitored by the bioluminescent reaction of luciferase. The bioluminescence observed is a measure of the kinase concentration. A competitive immunoassay has also been reported (11) that uses ATP-labeled antigens (as tracers) prepared by reacting an amino group-containing antigen with adenosine-5'-trimetaphosphate. The immunoreaction is followed by the release of ATP in a mild acidic solution and its measurement by the luciferase reaction. Another immunoassay configuration uses antibodies labeled with alkaline phosphatase or ß-galactosidase. Luciferin phosphate or luciferin galactoside are used as substrates, respectively (12). The enzymatically released luciferin can be quantified by the luciferase reaction. The recent progress in DNA technology has allowed the preparation of recombinant fusion proteins in which luciferase is "genetically" conjugated to protein A and used as a detection reagent in immunoassay (13). In another attempt to avoid inactivation, luciferase was biotinylated in vivo and then complexed to streptavidin (SA)1 (14).
Recently, we developed a new analytical system named expression
immunoassay (ExIA) (15). In ExIA, an expressible DNA
fragment that encodes for an enzyme (firefly luciferase) is used as a
label instead of the enzyme itself. The immunoreaction is performed in
microtiter wells, and the DNA label bound to the immunocomplex is
subjected to a cell-free (in vitro) coupled transcription/translation
reaction that produces several active enzyme molecules in solution. We
have also extended this principle (16) by using a DNA label
that encodes for a relatively small polypeptide with no inherent
enzymatic activity (
-peptide of ß-galactosidase). Each peptide,
however, is able to trigger an enzymatic reaction by interacting with a
larger, inactive protein molecule (M15 protein isolated from lacZ
M15
bacterial strains) in a "complementation reaction" to form a fully
active enzyme molecule in solution.
To date, the above systems have been tested in a simple assay in which
various amounts of a monoclonal antibody were immobilized in microtiter
wells and detected by using a goat anti-mouse antibody
(15)(16). In the present study, we report the
development of the first two-site, "sandwich-type" ExIA. The
principle of the proposed assay is illustrated in Fig. 1
. The proposed assay uses an expressible firefly
luciferase-coding DNA fragment (FLucDNA) as a reporter molecule. The
biotin/SA interaction (17) is used as a linkage between the
antibody and FLucDNA. In vitro expression of the FLucDNA that is bound
to the immunocomplex produces several luciferase molecules in solution.
The activity of synthesized luciferase is measured by its
bioluminescent reaction with luciferin. The proposed system was
evaluated by analyzing serum prostate-specific antigen (PSA) as a model
assay. PSA is a 33-kDa glycoprotein expressed predominantly by the
epithelial cells of the prostate gland (18). Recent studies
have also shown that PSA is present at low concentrations in breast
tumors (19)(20) and other tissues
(21). Over the past decade, the determination of serum PSA
has been widely used for monitoring patients diagnosed with prostate
cancer (22).
|
| Materials and Methods |
|---|
|
|
|---|
A rabbit reticulocyte-based transcription/translation mixture (rabbit reticulocyte TNT system; Promega) was prepared according to the manufacturer's instructions. The complete mixture consisted of the reticulocyte extract (containing rNTPs, ribosomes, tRNAs, and other translation factors), T7 RNA polymerase, and amino acids in the appropriate buffer.
The two anti-PSA antibodies used throughout this study were as described in Ref. (23) and were obtained from Medix Biotech (MBP0405, a monoclonal anti-PSA antibody; PBP0101, a polyclonal anti-PSA antibody).
Solutions of various PSA concentrations were prepared by diluting PSA (cat. no. P0724; Scripps Laboratories) in 50 mmol/L Tris, pH 7.8, 60 g/L bovine serum albumin, 1 g/L NaN3 and storing the dilutions at 4 °C.
biotinylation of anti-psa detection antibody
The polyclonal anti-PSA (0.25 mg) was dialyzed overnight against
3.5 L of 0.1 mol/L sodium bicarbonate at 4 °C. The purified antibody
was diluted twofold with 0.5 mol/L carbonate buffer, pH 9.1. The
resulting protein concentration was 1 g/L. The N-hydroxysuccinimide
ester of biotin (0.5 mg; NHS-LC-biotin; Pierce) was dissolved in 25
µL of dimethyl sulfoxide and added to the antibody solution. The
mixture was incubated for 2 h at room temperature with mixing at
regular time intervals. The biotinylated antibody was stored at 4 °C
and used without further purification.
preparation of SA-FLucDNA COMPLEX
The SA-FLucDNA complex was prepared as described in Ref.
(15). In brief, a plasmid containing the firefly luciferase
gene downstream from a T7 RNA polymerase promoter was digested with
Alw44I into three fragments possessing 3' recessive ends.
One of these fragments contained the T7 promoter, the luciferase gene,
and a poly(dA/dT) tail. The fragments were subsequently labeled with
biotin, using the exonuclease-free Klenow fragment of Escherichia
coli DNA polymerase I to fill in the recessive ends with dGTP,
dCTP, dTTP, and biotin-14-dATP. This process biotinylated both termini
of each fragment. A 0.49-kbp fragment just upstream from the T7
promoter was then removed by digestion with PvuII, thus
leaving a 2.1-kbp fragment (FLucDNA) labeled with biotin at only one
terminus. The FLucDNA was separated from the other fragments by agarose
gel electrophoresis and purified by elution from the gel. A SA-FLucDNA
complex was prepared by mixing biotinylated FLucDNA with an excess of
SA. The concentration of SA was confirmed by measuring its absorbance
at 282 nm (molar absorptivity, 2.4 x 105
mol · L-1 · cm-1).
The SA-FLucDNA was purified by size exclusion HPLC and concentrated by
ultrafiltration. The concentration of the SA-FLucDNA was determined by
scanning densitometry.
coating of microtiter wells
Transparent U-bottom polystyrene microtiter wells (Maxisorp; Nunc,
obtained from Life Technologies) were coated by incubation overnight at
room temperature with 25 µL per well of 5 mg/L monoclonal anti-PSA
antibody diluted in 50 mmol/L Tris pH 7.8, 0.5 g/L sodium azide. Before
use, the wells were washed twice with wash solution and blocked for
1 h with 200 µL of blocking solution per well.
two-site immunoassay for psa
A 12.5-µL aliquot of the assay buffer was pipetted into each
well, followed by the addition of 12.5 µL of PSA sample. The antigen
was allowed to react with the immobilized antibody for 3 h at room
temperature with continuous mechanical shaking. At the end of this
period, the wells were washed six times with wash solution followed by
the addition of 25 µL per well of 0.23 mg/L biotinylated anti-PSA
detection antibody diluted in blocking solution. After a 1-h
incubation, the excess detection antibody was removed by washing the
wells six times as above. A 25-µL aliquot of 0.86 µg/L SA-FLucDNA
complex (diluted in 10 g/L blocking reagent, 0.1 mol/L maleate, pH 7.5,
0.15 mol/L NaCl, and 2 mmol/L EDTA) was then added into each well and
allowed to bind to the immunocomplex for 20 min. The unbound SA-FLucDNA
complex was removed by washing the wells five times with wash solution
and once with Tris-EDTA buffer. The luciferase-coding DNA that was
bound to the solid phase was expressed in vitro by an one-step coupled
transcription/translation reaction. A 25-µL aliquot of the
transcription/translation mixture was added into each well and
incubated at 30 °C for 90 min in an incubator shaker (Model G24; New
Brunswick Scientific). The activity of synthesized firefly luciferase
was then measured by mixing 10 µL of the transcription/translation
reaction mixture with 50 µL of substrate solution at room temperature
in a microcentrifuge tube. The tube was placed immediately in a glass
scintillation vial, and the bioluminescence was measured for 1 min in
the liquid scintillation counter (Model LS-6500; Beckman Instruments).
The coincidence photon detection of the counter was disabled to
facilitate the counting of single photon events (single photon
monitoring mode).
analysis of clinical specimens
Serum samples were diluted 30-fold in blocking solution before the
analysis. Samples with high turbidity were centrifuged at
12 000g for 30 min at room temperature before the analysis.
The PSA concentration of the serum samples was determined by using a
calibration curve constructed from PSA calibrators whose concentrations
were confirmed by the Abbott IMx PSA assay (Abbott Laboratories).
| Results |
|---|
|
|
|---|
To optimize the concentration of biotinylated anti-PSA antibody, we
prepared dilutions of the antibody in the range of 0.23.7 mg/L and
analyzed a sample with 0.5 µg/L PSA. In Fig. 2
, the luminescence and the signal-to-background (S/B) ratio are
plotted against the biotinylated antibody concentration. The background
is defined as the luminescence obtained when no antigen is present in
the well and is a measure of the nonspecific binding of the
biotinylated anti-PSA antibody and the SA-FLucDNA complex. The
luminescence increased with the antibody concentration, but the maximum
S/B ratio was observed at 0.23 mg/L, followed by a continuous decrease
as the antibody concentration increased (because of increased
nonspecific binding).
|
In the range of 0.23.4 mg/L, the luminescence increased with the
SA-FLucDNA concentration (Fig. 3
), but the S/B ratio reached a plateau at 0.43 mg/L. At higher
concentrations, the background concomitantly increased and the S/B
ratio remained practically constant.
|
The linear range of the assay was 303000 ng/L (Fig. 4
). The S/B ratio for 30 ng/L PSA was 2.3 with a sample volume of
12.5 µL.
|
To further evaluate the linearity of the method, we prepared dilutions (using female serum as a diluent) of two serum samples with added PSA. Each sample was diluted another sixfold in the blocking solution before analysis. The relationship between the concentration (y, in ng/L) and the dilution factor (x) was linear. The linear regression equations were: log(y) = 3.739 - 0.964 log(x); r = -0.999; and log(y) = 3.537 - 1.116 log(x); r = -1 (n = 5).
In a method-comparison study of 44 serum samples with PSA
concentrations ranging from 1.1 to 53 µg/L with the Abbott IMx method
(Fig. 5
), the linear regression equation was: (ExIA) = 0.98(IMx) +
0.74 µg/L; r = 0.971; Sy|x =
0.562.
|
The within- and between-day imprecision (CVs) for three different serum
samples was 810% (Table 1
).
|
| Discussion |
|---|
|
|
|---|
ExIA is complementary to the above efforts because instead of providing an alternative method for measuring enzyme activity, it entails an increase in the number of enzyme molecules, thus introducing additional amplification. Consequently, ExIA might be used in combination with the above systems to further enhance the sensitivity. For example, the ALP-coding DNA might be used as a reporter to provide several ALP molecules that then could be detected by a chemiluminogenic substrate or a cascade amplification system. Furthermore, by using an enzyme-coding DNA fragment as a label, the problem of enzyme inactivation after conjugation to antibodies is eliminated, and the enzyme remains free in solution. Moreover, in direct comparison with fluorescent and chemiluminescent immunoassays and hybridization assays (15)(16)(29), substantial improvements on the sensitivity were achieved by using the enzyme-coding DNA labels. Disadvantages of the ExIA that might be considered are the use of a transcription/translation cocktail, which requires storage at -70 °C, and the need for a 90-min incubation step for expression. The transcription/translation cocktail is added as one reagent in a single step and therefore does not complicate the protocol.
In conclusion, we have developed a two-site ExIA and demonstrated its clinical utility. Because of its high sensitivity, we anticipate that the developed assay will be useful for determining other clinical analytes that exist at low concentrations in biological fluids.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
-peptide of ß-galactosidase. Analyst 1998;123:1309-1314.
[Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
![]() |
L. MARTINIOVA, S. OHTA, P. GUION, D. SCHIMEL, E. W LAI, B. KLAUNBERG, E. JAGODA, and K. PACAK Anatomical and functional imaging of tumors in animal models: focus on pheochromocytoma. Ann. N.Y. Acad. Sci., August 1, 2006; 1073: 392 - 404. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. A. Tannous, E. Laios, and T. K. Christopoulos T7 RNA polymerase as a self-replicating label for antigen quantification Nucleic Acids Res., December 15, 2002; 30(24): e140 - e140. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |