| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
CEA,
1
Laboratoire d’Etudes Radioimmunologiques, SPI/DRM/DSV, and
2
Service des Molécules Marquées, DBCM/DSV, CEA/Saclay, 91191 Gif-sur-Yvette Cedex, France.
3
SPI-BIO, 2 rue du Buisson aux Fraises, ZI de la Bonde,
91741 Massy Cedex, France.
4
bioMérieux, Département Immunoassays, Chemin
de l’Orme, 69280 Marcy-l’Etoile, France.
a Address correspondence to this author at: CEA, Laboratoire d’Etudes Radioimmunologiques, SPI/DRM/DSV, CEA/Saclay, 91191 Gif-sur-Yvette Cedex, France. Fax 33-01-69-86-77-49;
philippe.pradelles{at}cea.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We used an anti-estradiol monoclonal antibody as capture antibody and Fenton-like reagents to cross-link it to E2. The same antibody, labeled with acetylcholinesterase, was used for detection. Serum was diluted 10-fold before assay.
Results: After correction by the dilution factor, the detection limit was 5 ng/L for human serum and intra- and interassay CVs were <7% and 15%, respectively, at concentrations of 169-2845 ng/L. No cross-reactivity was seen with other natural steroids. In comparison with a competitive commercial RIA performed on 88 undiluted human sera, the slope (SD) of the regression line was 1.05 (± 0.02) and the intercept was 47 (±27) ng/L (Sy|x = 186 ng/L) at concentrations of 20–5000 ng/L (r2 = 0.97).
Conclusions: The use of Fenton-like chemistry in SPIE-IA technology allows a sensitive measurement of E2 in human serum and could be a new approach for the development of sensitive immunoassays.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
):
|
SPIE-IA was initially applied to the quantification of low-molecular weight peptides possessing an amino group that was present naturally in the molecule or introduced before the assay (5) and that was able to react with the homobifunctional reagents. With the aim of extending the procedure to haptens not possessing an amine moiety, we have investigated the feasibility of using direct ultraviolet (UV) irradiation to induce the cross-linking of the hapten to its mAb. Our earliest work was initiated for an L-thyroxine assay (8), and the procedure was named Photo-SPIE-IA. More recently, our group described (9) a Photo-SPIE-IA for 17ß-estradiol (E2) using a cross-link step via direct UV irradiation at 254 nm. During the course of this study, we showed that a similar cross-linking reaction was observed when 60Co gamma irradiation (instead of UV irradiation) was used (results not shown).
The photoconjugation of proteins and amino acids to DNA represents one of the major causes of UV-induced damage in biological systems (10). The mechanism of direct photocoupling is complex and may vary from system to system. However, it is generally believed to proceed through the generation of free radicals followed by fast reactions with its surroundings, thereby inducing a zero-length cross-link. Among putative radical species, the hydroxyl radical (HO·) is well known to be a powerful agent for hydrogen atom abstraction (11) and is involved in damage caused by photolysis or radiolysis of biological samples. To investigate the role of HO· during the covalent epitope immobilization step of E2 in the Photo-SPIE-IA, we performed the UV irradiation in the presence of H2O2 diluted in aqueous medium, a well-known procedure for the clean generation of HO· (12). The results showed a drastic increase in the rate of cross-linking of E2 to the solid phase (data not reported) compared with the rate we reported in the absence of H2O2 (9). This experiment supports the hypothesis of an HO·-promoted cross-linking reaction and has encouraged us to produce HO· radicals chemically with the help of Fenton-like reagents.
Fenton (13) described the powerful oxidizing properties of
the mixture of H2O2 and
ferrous salts (Fe2+). Several years later, Haber
and Weiss (14) established the oxidizing species as
HO· generated according to the following reaction:
 |
Other transition metals were used to generate radicals through Fenton-like chemistry (15), although it is still the subject of debate (16)(17).
The present work describes, for the first time, the use of Fenton-like reactions to cross-link E2 to its specific mAb during step 2 of the SPIE-IA, which will hereafter be designated SPIE-IA. Various factors affecting the experimental conditions were analyzed, and the sensitivity, precision, and specificity of the assay were determined.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
apparatus
Unless otherwise stated, SPIE-IA was performed with specialized
microtitration equipment (washer, dispenser, and plate reader) from
Labsystem. The 96-well microtiter plates (Maxisorp) were from
Nunc.
chemicals
Unless otherwise stated, all reagents were from Sigma, and
steroids were from Steraloids. Thiourea and trolox were from Aldrich,
and mannitol was from Prolabo. Human serum E2 calibrators and
samples were provided by bioMérieux (Marcy-L’Etoile, France) and
were used for the calibration and correlation studies, respectively.
Urea hydrogen peroxide (UHP) was chosen as oxidant instead of
H2O2 itself. This
hydrogen-bonded adduct is easy to handle and can be stored for long
periods without significant decomposition.
immunochemicals and enzyme labels
AChE (EC 3.1.1.7) was purified in our laboratory from electric eel
(Electrophorus electricus) as reported elsewhere
(18). Mouse monoclonal anti-E2 antibody (10G6D6) was
generously provided by bioMérieux (9). The globular
form of AChE was covalently coupled to Fab' fragments of mAbs as
described previously (7), which were used as tracer
antibodies in the SPIE-IA experiments. The 96-well microtiter plates
were coated with mAbs as described elsewhere (19). EIA
buffer was 0.1 mol/L potassium phosphate buffer (pH 7.4) containing
0.15 mol/L NaCl, 1 g/L bovine serum albumin (BSA), and 0.1 g/L
sodium azide. Saturation buffer was 0.075 mol/L sodium phosphate (pH
6.8) containing 2 g/L casein, 0.5 mL/L Tween 20, and 1 g/L BSA.
Before each step of the SPIE-IA procedure, the plates were washed with
the washing buffer [0.01 mol/L potassium phosphate (pH 7), containing
0.5 mL/L Tween 20]. The Ellman unit and Ellman reagent have
been described elsewhere (19).
spie-ia procedures
Unless otherwise stated, routine SPIE-IA of human serum E2 was
performed as follows:
Step 1. Immunological capture.
Human serum E2 calibrators (0,
41, 56, 81, 179, 359, 529, 922, 1763, and 2602 ng/L E2) or samples were
diluted to 1:10 in EIA buffer containing 100 µg/L mesterolone.
Calibrators and samples (100 µL of each) were then incubated for 15
min at room temperature in wells coated with anti-E2 mAbs.
Human serum E2 calibrators were constructed by the addition of an E2
calibrator to E2-free human serum kindly provided by bioMérieux.
Step 2. Epitope immobilization.
After the plate was washed, 50
µL of 20 mmol/L UHP was added, followed by 50 µL of 20 mmol/L
CuSO4 for a 2-min reaction at room temperature.
UHP and CuSO4 were dissolved in 0.1 mol/L sodium
carbonate buffer (pH 9).
Step 3. Epitope release.
After the plate was washed, 200 µL
of 0.5 mol/L NaOH was added for a 2-min reaction at room
temperature.
Step 4. Visualization.
After the plate was washed, 100 µL of
E2 mAbs labeled with AChE (
4 Ellman units/mL) was added, and the
mixture was incubated for 1 h at room temperature. After the plate
was washed, 200 µL of Ellman reagent was added and the absorbance at
414 nm, expressed as absorbance units (AU), was measured after a 10-min
enzymatic reaction at room temperature with gentle stirring.
calculations
The results are expressed in terms of AU as a function of hapten
concentration. Nonspecific binding was determined at zero concentration
of the calibrator (EIA buffer or E2-free human serum). The minimum
detectable concentration was taken as the concentration of
hapten or human serum E2 calibrator inducing a significant increase in
nonspecific bonding (3 SD). Unknown E2 human serum concentrations were
calculated from a calibration curve using a four-parameter logistic
function (Immunofit Software; Beckman). All calibrators and biological
samples were measured in duplicate.
precision
The imprecision of the calibration curve was determined by
performing all measurements eight times and is expressed in terms of
CV. Intraassay imprecision was determined from eight
measurements of five control sera from bioMérieux diluted 10-fold
before assay (quality-control samples, containing 40, 169, 558,
1226, and 2845 ng/L, respectively) in one assay, and interassay
imprecision was determined from seven repeated analyses of these
control sera in consecutive assays.
correlation
We compared the E2 concentrations in 88 human sera measured by a
commercial competitive RIA (coatRIA2; bioMérieux; analytical
range, 31–2609 ng/L), using undiluted human sera, with the values
obtained by SPIE-IA, using 10-fold diluted human sera. Sera with an E2
concentration higher than the upper limit of the calibration curve
(>260.9 ng/L) were reassayed by SPIE-IA after an additional twofold
dilution.
specificity
Specificity was studied by establishing a calibration curve for
each steroid, and the results were expressed in terms of percentage of
cross-reactivity, arbitrarily defined as the ratio of the concentration
of E2 (pmol/L) vs the steroid concentration producing 0.5 AU.
influence of fenton-like reagents on the cross-linking
reaction
For all of these experiments, the E2 concentration for the
immunological capture step was 500 ng/L in EIA buffer, and the SPIE-IA
signal was the AU value obtained after 10 min of enzymatic reaction.
Concentrations of metallic salts, EDTA, and UHP were expressed in terms
of final concentration in the wells. Unless otherwise stated,
cross-linking buffers were 0.1 mol/L sodium carbonate (pH 9) or 0.1
mol/L potassium phosphate (pH 7.4). Immunological capture and tracer
antibody incubations were carried out for 1 h at room temperature.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
These data show that in carbonate buffer (pH 9), CuSO4 appears to be very effective in inducing the highest SPIE-IA signal in the absence of EDTA and that this signal is totally suppressed in the presence of EDTA. Effects opposite those of EDTA were observed for FeSO4. In phosphate buffer (pH 7.4), similar data were observed, although the SPIE-IA signals were lower than in carbonate buffer. Interestingly, all of the curves had the same bell-shaped form. We could hypothesize that the decrease of the signal observed at high UHP concentrations is related to various phenomena such as (a) a dramatic effect of the Fenton reagents, including a loss of the coated material on the solid phase; (b) a chemical modification of E2 that decreases its recognition by the tracer antibody; and (c) competitive consumption of HO· by the excess of UHP through some complex chemical reactions.
effects of other metal species
We tested various other metallic salts (used at 10 mmol/L in 0.1
mol/L carbonate buffer, pH 9), including CoSO4,
NiSO4, TiCl3,
CdCl2, CrCl2,
CrK(SO4)2,
HgCl2, MgSO4,
MnCl2, VCl3, and
ZnCl2, in the absence or presence of 20 mmol/L
EDTA. Only CrCl2 and
CrK(SO4)2 were able to
induce low but significant SPIE-IA signals in the presence of EDTA
(and none in its absence; data not shown). No signals were observed
with other metallic salts (data not shown).
pH AND KINETIC STUDIES
The pH and the nature of the salts affected the kinetics of the
cross-linking reaction and the SPIE-IA signal, as illustrated in Fig. 3
. A maximum was quickly (1–2 min) reached in carbonate
buffer (pH 9), although 10 min were required in 0.1 mol/L borate
buffer (pH 9) and up to 30 min in 0.1 mol/L borate solution (pH 7)
or 0.1 mol/L phosphate buffer (pH 7.4). The highest SPIE-IA signal was
obtained in carbonate buffer (pH 9), using 10 mmol/L
CuSO4 and 10 mmol/L UHP in the absence of EDTA.
These conditions were therefore selected for routine SPIE-IA.
Moreover, identical results were obtained when
H2O2 (instead of UHP) was
used as oxidant (data not shown).
|
importance of the different steps of the spie-ia procedure
As described for all previous SPIE-IAs
(2)(3)(4)(5)(6)(8)(9), we demonstrated that
each step of the SPIE-IA procedure was necessary to induce a signal
(Table 1
), and we showed that UHP and CuSO4 alone
fail to induce a significant SPIE-IA signal.
|
effect of the fenton-like reaction on coated mAbs ANDE2
The main reaction that occurs in the presence of excess HO·
radicals is oxidation and/or cleavage of the free radical target. To
explain the above results, our hypothesis is that two partners (mAb and
E2 in our case) can be cross-linked by free radical species if they are
close spatially. However, taking into account the tremendous excess of
HO· generated in our experiments, we cannot exclude sidechain
reactions such as degradation of the partners involved.
To evaluate the deleterious effects of oxidizing agents induced
by the Fenton-like reaction, we either subjected the coated mAb to the
action of each Fenton-like reagent and then performed the SPIE-IA, or
performed the SPIE-IA until the epitope release step and then
subjected the solid phase to the action of each Fenton-like reagent.
The results are presented in Fig. 4
and show no significant effect of CuSO4
or UHP alone on coated mAbs (Fig. 4
, experiments A and B) or E2 linked
to mAb (Fig. 4
, experiments D and E) compared with the results obtained
without pretreatment (experiment G). When CuSO4
and UHP were added together, we observed a strong decrease in the
signal (Fig. 4
, experiments C and F).
|
In light of these results, we can postulate degradation of the mAb and E2 when they are not bound to each other, but an epitope-paratope protection against the deleterious effect of the cross-linking agents induced by the Fenton-like reaction when the mAb-E2 complex is formed (the paratope being defined as the part of the antibody molecule directly in contact with the epitope). We have described similar results for the Photo-SPIE-IAs for L-thyroxine (8) and E2 (9).
effect of antioxidants on the cross-linking reaction
We studied the effect of various potential radical scavengers such
as BSA (20), thiourea (21), trolox
(22), cimetidine (23), tyrosine (24),
ascorbic acid (25), mannitol (26), and dimethyl
sulfoxide (27) on the SPIE-IA signal. The results are
presented in Fig. 5
and show that the cross-linking reaction was indeed inhibited
by radical scavengers. Stronger inhibition was obtained with BSA and
thiourea than with the other compounds. However, in the concentration
range used, mannitol and dimethyl sulfoxide, both known as HO·
scavengers (26)(27), failed to inhibit the
SPIE-IA signal.
|
sensitivity and precision
Preliminary experiments showed that E2 SPIE-IA performed on
undiluted human sera lacked sensitivity and reproducibility. However,
when human E2 sera were diluted 10-fold, this assay had good
sensitivity and precision, as shown in Fig. 6
. A minimum detectable concentration of 0.5 ng/L was obtained
with a CV <5% at concentrations of 4–260 ng/L (eight replicates for
each calibrator). The intra- and interassay CVs, determined with
quality-control material corresponding to 16.9, 55.8, 122.6, and 284.5
ng/L estradiol after 10-fold dilution in EIA buffer prior the assay,
were <7% and 15%, respectively. For a quality-control material
corresponding to 4 ng/L estradiol, the intra- and interassay CVs were
6% and 24%, respectively.
|
correlation study
Good correlation was obtained for SPIE-IA and coatRIA2 E2
measurements in 88 human serum samples. Linear regression analysis gave
the equation: SPIE-IA = 1.05 (± 0.02) coatRIA2 + 47 (± 27);
Sy|x = 186; range, 20–5000 ng/L;
r2 = 0.97; n = 88 (Fig. 7
).
|
specificity
When applied to different steroids, SPIE-IA had
cross-reactivities of 0.01% and 0.004% for estrone and
ethinyl-estradiol, respectively, and <0.001% for other tested
steroids as shown in Table 2
.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As hypothesized in our last report (9), we assume that such a cross-link would occur between some part of E2 outside of the epitope structure and some part of the anti-E2 antibody outside the paratope structure, such as the sidechains of amino acids. Through the data reported here, we clearly showed that Fenton reagents (e.g., a mixture of Cu2+ and UHP) are useful tools in cross-linking E2 to its specific mAb and that they allow the development of a new sensitive and easy immunometric assay through SPIE-IA technology. In addition, we can imagine several extensions of this kind of study. For example, we think that our technology will cast light on several findings concerning Fenton chemistry. Our data are either in agreement with literature findings for the oxidation and degradation of proteins and amino acids, or have never been described. Our important findings include the following: (a) In our experimental conditions for the E2 cross-linking step, we found that Cu2+ induced a higher SPIE-IA signal than Fe2+ and other species. Copper salts have previously been reported to generate HO· at a greater rate than do ferrous salts (31). Co2+ and Mn2+ failed in the presence or absence of EDTA to induce a significant SPIE-IA signal, whereas they have been reported as being powerful oxidative species (32)(33) in the presence or absence of chelates. (b) The fact that at pH 7.4, EDTA prevents the formation of HO· in a Cu2+-catalyzed Fenton-type reaction but enhances these phenomenon in the Fe2+-catalyzed Fenton-type reaction has been reported elsewhere (34)(35). Nevertheless, these findings had not been described at pH 9 in carbonate buffer. (c) We clearly showed that a change in pH (7.4 to 9) has a drastic effect on the kinetics of the cross-linking reaction and therefore on the strength of the SPIE-IA signal. This effect could suggest that the nature of the reactive species involved in the cross-linking process depends on the pH in the Fenton reaction. (d) The effect of carbonate ions was not assessed by Dakin (36) in his pioneering work on the oxidation of amino acids, but was described elsewhere (37). One possibility is the formation of a secondary radical species such as carbonate anion radical (CO), a powerful reducing agent (38). (e) As observed during the E2 Photo-SPIE-IA (9), we demonstrated epitope-paratope protection of the E2 and mAb binding site against the deleterious effects of free radicals by showing that the SPIE-IA signal was dramatically decreased when mAb-coated solid phase or E2 linked to the solid phase was subjected to the action of Fenton-like chemistry. (f) SPIE-IA experiments conducted in the presence of various HO· scavengers confirmed the involvement of highly reactive radical species in the cross-linking reaction, but also showed that these scavengers displayed various inhibitory properties probably related to their capacity to trap the reactive oxygenated species. Taking into account the easy and rapid procedure used in the SPIE-IA, we think that this technique could be a new and simple method for screening antioxidant capacity.
In conclusion, we have shown that Fenton-like chemistry could be a useful method in the development of a sensitive immunometric assay of E2 and that it may be a good alternative to the use of an UV irradiation device (9) because the cross-linking step with the Fenton reagents requires only a 2-min reaction compared with the 45-min UV irradiation described previously (9). The application of this technology to other molecules is being investigated in our laboratory, in particular for haptens that cannot be assayed by classical SPIE-IA.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and -1ß. Development of two enzyme immunometric assays (EIA) using acetylcholinesterase and their application to biological media. J Immunol Methods 1989;123:193-210.[Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  F. Paris, N. Servant, B. Terouanne, P. Balaguer, J. C. Nicolas, and C. Sultan A New Recombinant Cell Bioassay for Ultrasensitive Determination of Serum Estrogenic Bioactivity in Children J. Clin. Endocrinol. Metab., February 1, 2002; 87(2): 791 - 797. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, 0.1 mol/L carbonate buffer (pH 9); •, 0.1 mol/L borate buffer
(pH 9); 
, 0.1 mol/L phosphate solution (pH 9); 
, 0.1 mol/L
phosphate buffer (pH 7.4); 
, 0.1 mol/L borate solution (pH 7). Each
experiment was performed in duplicate. Error bars, SD.
, thiourea; 
, trolox; 
, cimetidine; 
, tyrosine;
, uric acid; 
