Journal of Experimental Botany, Vol. 51, No. 343, pp. 305–307, February 2000
The monoclonal antibody MAC252 does not react with
the (−) enantiomer of abscisic acid
Philippe Barrieu and Thierry Simonneau1
Ecophysiologie des Plantes sous Stress Environnementaux, INRA, 2 place Viala, F-34060 Montepellier cedex 1,
France
Received 14 June 1999; Accepted 30 September 1999
Abstract
An RIA procedure has been developed for ABA quantification using MAC62, a monoclonal antibody raised
against (+)-cis, trans -ABA. This widely used method
now relies on MAC252, a recloned version of the
exhausted MAC62. Recently, it has been suggested
that MAC252 was not able to discriminate between
the (+) and (−) enantiomers of ABA. As this can be
misleading when interpreting RIA results, it has been
carefully examined here whether MAC252 reacts with
(−)-ABA. MAC252 exhibited negligible cross-reactivity
with (−)-ABA, which was confirmed with commercial
mixtures of ABA isomers. It is concluded that the RIA
protocol can continue to be used with MAC252 as it
was with MAC62.
Key words: RIA, MAC252, ABA isomers.
Introduction
During the past few years, monoclonal antibodies have
become widely used to quantify endogenous ABA concentrations in plant extracts, enabling large-scale ABA analyses ( Tuberosa et al., 1994; Pekic et al., 1995).
Immunological methods (for a review, see WalkerSimmons and Abrams, 1991) include enzyme-linked
immunosorbent assays, fluoroimmuno assays and radioimmuno assays (RIA). Many groups ( Wang et al., 1987;
Saab et al., 1990; Tuberosa et al., 1994; Jackson et al.,
1995; Dodd and Davies, 1996; Cramer et al., 1998;
Simonneau et al., 1998) have adopted the RIA protocol
(proposed by Quarrie et al., 1988) using MAC62, a highly
specific monoclonal antibody raised against (+) 2 cis-4
trans-ABA (Quarrie and Galfre, 1985). MAC62 has an
affinity constant of c. 5×108 l mol−1 for (+)-ABA, and
does not bind (−)-ABA or the large majority of ABA
metabolites and derivatives (Quarrie et al., 1988).
In 1998, results were published (Cramer et al., 1998)
based on the procedure of Quarrie et al. (Quarrie et al.,
1988), but with two versions of the antibody: MAC62,
that was extensively characterized by Quarrie et al. and
MAC252, a recloned version later supplied by the John
Innes Centre, Norwich UK. Surprisingly, it was found
that, in contrast to MAC62, MAC252 could not discriminate between the natural (+)- and unnatural (−)-cis,
trans ABA enantiomers, though it did not react with the
trans, trans isomer (Cramer et al., 1998). It is clear that
the sensitivity for ABA isomers varies with the nature
of the antibody ( Walker-Simmons et al., 1991), but it
was not expected to change between MAC62 and its
recloned version MAC252 (SA Quarrie, personal communication). Since MAC252 is now substituted for the
exhausted MAC62 for use in RIA, the results of Cramer
et al. need to be carefully checked.
The specificity of the antibody to (+)-ABA compared
to (−)-ABA is particularly crucial where racemic mixtures
of (±) 2 cis-4 trans-ABA are prepared as internal standards. It has also important implications when drawing
dose–response curves with (±)-ABA-fed plants.
Therefore, a set of controls was performed with different ABA isomers, either purified or mixed, to check for
the specificity of MAC252. Special attention was paid to
(−)-ABA reactivity with this antibody.
Materials and methods
ABA and antibody preparations
Purified enantiomers ((+)-ABA, ref. A4906 and (−)-ABA,
ref. A8451), racemic mixture ((±)-ABA, ref. A1012) and mixed
isomers (ref. A7383) of ABA were obtained from Sigma (Sigma
Chemical Co, St Louis, MO, USA).
1 To whom correspondence should be addressed. Fax: +33 4 67 52 2116. E-mail: [email protected]
© Oxford University Press 2000
306 Barrieu and Simonneau
A stock of lyophilized MAC252 was purchased from Dr SA
Quarrie (John Innes Centre, Colney, Norwich, UK ). MAC252
was rehydrated with water (following indications provided with
the stock of MAC252) and diluted 1520 in phosphate buffer
saline (PBS, 100 mM sodium phosphate and 200 mM sodium
chloride) containing 5 mg ml−1 bovine serum albumin (BSA,
Sigma) and 4 mg ml−1 soluble polyvinylpyrrolidone (PVP, MW
40 000, Sigma). This working stock solution was stored at
−20 °C. For the assay, a further dilution of about 15120 in
buffered BSA/PVP was made, giving a total dilution of 152400.
Radioactive abscisic acid preparation
(±)-cis,trans-[G-3H ]-Abscisic acid ( TRK 644 Amersham
Pharmacia Biotech, UK ) in ethanol at an approximate
concentration of 1.85–3.7 1012 Bq mmol−1 was diluted 1510 in
water and frozen in 50 ml aliquots (working stock). For the
assay, working stock aliquots were further diluted at about
15250 in PBS containing 2.5 mg ml−1 bovine c-globulin (Sigma).
Radioimmunoassay protocol
ABA or water (for determination of maximum binding with
3H-ABA) were mixed with 100 ml [G-3H ]-abscisic acid, (approximately 310 Bq), 100 ml MAC252 and 200 ml 50% PBS in 2 ml
Eppendorf vials. Non-specific binding (revealed by the
co-precipitation of radioactive material into the pellets) was
determined by replacing water with ABA 100 mg ml−1, to have
the greatest binding of the MAC252 with unlabelled ABA. The
assay mixture was incubated at 4 °C in darkness for 45 min. A
saturated solution (500 ml ) of ammonium sulphate (Sigma) was
then added and, after brief shaking, the mixture was left at
room temperature for 30 min. Free antibodies and those that
bound to ABA in the reaction mixture were precipitated and
pelleted by centrifugation for 4 min at 8800 g, then the
supernatant was discarded. The pellet was washed by resuspending in 1 ml of a 151 mix of water and saturated ammonium
sulphate solution, and centrifuged again for 5 min at 8800 g.
The supernatant was discarded and the pellet finally resuspended
in 100 ml deionized water. After adding 1.2 ml scintillation
cocktail (Emulsifier Safe, Packard Instrument Company, USA),
radioactivity was quantified in a liquid-scintillation counter
(Tri-carb, Packard ). A logit transformation was applied to the
results as follows:
(B−B ) (B −B )
min max
min
)=ln
max
1−((B−B )/(B −B ))
min
max
min
where B, in cpm, quantifies the radioactivity that remained
bound to MAC252 when ABA of the sample was added to the
vial; B
quantifies the maximal bound radioactivity when
max
only 3H-ABA reacted with MAC252; and B
quantifies the
min
minimal bound radioactivity when a large excess of non-labelled
ABA was added to the vial. The calibration curve was achieved
by plotting logit-transformed counts for a range of (+)-cis,
trans ABA standards against the natural logarithm of the
unlabelled ABA concentration present per vial. This gave a
straight line for which a linear regression equation was
calculated. This equation was used to convert the logittransform of the sample counts into the calculated concentration
of (+)-ABA. Any immunoreactant in the sample was thus
considered in this conversion as (+)-ABA equivalent with
regard to MAC252.
LOGIT(B/B
Results and discussion
Figure 1 clearly shows that (−)-ABA hardly binds to
MAC252. With added (−)-ABA concentrations as high
Fig. 1. Comparison of mixed isomers, (+/−)-, (+)- and (−)-ABA
added per vial and ABA recovered by RIA using MAC252 antibody.
Lines were fitted by linear regression, forced through zero. The values
of the slopes are 1.00 ((+)-ABA), 0.57 ((+/−)-ABA) and 0.39 (mixed
isomers). Each datum point represents the mean of 4–10 replicates
assayed for each concentration. The bars indicate standard deviations.
as 8000 pg/50 ml, resulting counts did not exceed an
equivalent of 200 pg (+)-ABA/50 ml, indicating that, in
the absence of non-labelled (+)-ABA, less than 3% of the
added (−)-ABA reacted with MAC252 as (+)-ABA did.
This is very different from the results of Cramer et al.,
who found that both enantiomers bound to the antibody
equally (Cramer et al., 1998). The slight binding of
(−)-ABA with MAC252 reported in Fig. 1 was estimated
in the absence of competition with non-radioactive
(+)-ABA. This should have overestimated the binding of
(−)-ABA when the (+)-enantiomer was also added to
the vial, for example, as racemic standards. This gives
confidence that ignoring the cross-reactivity of MAC252
with (−)-ABA leads to negligible error.
The reactivity of the racemic ABA is consistent with a
negligible binding of (−)-ABA. Compared to purified
(+)-ABA, the percentage of (±)-ABA that bound
MAC252, was close to 50%, though slightly but significantly higher (57.4±2.5%). Furthermore, reactivity of
mixed isomers with MAC252 was even more reduced
(39.7±0.8%). The proportions of each isomer in the
mixture are not strictly equal (Sigma-Aldrich France,
personal communication). This can partly explain why
the cross-reactivity of the mixed isomers with MAC252
was greater than 25%. However, a weak binding of trans
ABA enantiomers with MAC252 cannot be excluded and
merits further checking, notably if mixed isomers are used
as internal standards or as a source of exogenous ABA
for feeding plants.
These results suggest that the recloned version of
MAC62 kept the same immunological properties as its
Antibody MAC252 specificity against ABA 307
initial version. It is therefore unlikely that ABA derivatives which did not bind MAC62 could react with
MAC252. However, as for MAC62, the present results
do not exclude cross-reactivity with immunoreactants
other than those mentioned here, particularly in crude
plant extracts. As an example, it was found by thin layer
chromatography that 52% of MAC252 reactivity with
crude leaf extracts of perennial forage grasses were with
immunoreactive compounds other than (+)-ABA (P
Barrieu, unpublished data). In the same way, crossreactivities up to 36.4% were found with some artificial
ABA analogs like the C-4 C-5 acetylenic acid ( WalkerSimmons et al., 1991).
Recloning of the cell line to get MAC252 was not
expected to modify the antibody specificity to (+)-ABA,
neither its cross-reactivity for ABA analogues. Notably,
it is unlikely that MAC252 had lost the very high specificity of MAC62 for ABA analogues having the same
absolute configuration at the chiral centre C-1∞ as
(+)-ABA ( Walker-Simmons et al., 1991). It is thus not
known how cross-reactivity for (−)-ABA could be
achieved with MAC252 as reported earlier (Cramer et al.,
1998). Optically pure enantiomers of ABA were obtained
by these authors from racemic methyl abscisate that was
resolved to pure methyl esters of ABA (Dunstan et al.,
1992). It would be helpful to know whether Cramer et al.
obtained their (−)-ABA as a crystalline solid, in which
case it is likely to be sufficiently pure (Cramer et al.,
1998). If it was not crystallized, then it may have contained solvent residues or other substances that could
have cross-reacted with MAC252. This is, however,
unlikely to explain the huge differences between the
present findings ( Fig. 1), and those reported previously
(Cramer et al., 1988).
In conclusion, ABA can be reliably assayed by RIA
with MAC252 as it was with MAC62. Notably, constructing standard curves with racemic (±)-ABA and,
considering that only half the (±)-ABA added binds
MAC252, yields similar results to those for purified
(+)-ABA standards. MAC252 can also be used to discriminate between (+)- and (−)-ABA in physiological studies
based on (±)-ABA-fed plants.
Acknowledgements
The authors wish to thank Dr SA Quarrie for continuing help
and advice concerning the use of MAC62 and MAC252
antibodies in RIA.
References
Cramer GR, Krishnan K, Abrams SR. 1998. Kinetics of maize
leaf elongation. IV. Effects of (+)- and (−)-abscisic acid.
Journal of Experimental Botany 49, 191–198.
Dodd LC, Davies WJ. 1996. The relationship between leaf
growth and ABA accumulation in the grass leaf elongation
zone. Plant, Cell and Environment 19, 1047–1056.
Dunstan DI, Bock CA, Abrams GD, Abrams SR. 1992.
Metabolism of (+)- and (−)-abscisic acid by somatic embryo
suspension cultures of white spruce. Phytochemistry 31,
1451–1454.
Jackson GE, Irvine J, Grace J, Khalil AAM. 1995. Abscisic
acid concentrations and fluxes in droughted conifer saplings.
Plant, Cell and Environment 18, 13–22.
Pekic S, Stikic R, Tomljanovic L, Andjelkovic V, Ivanovic M.,
Quarrie SA. 1995. Characterization of maize lines differing
in leaf abscisic acid content in the field. 1. Abscisic acid
physiology. Annals of Botany 75, 67–73.
Quarrie SA, Galfre G. 1985. Use of different hapten–protein
conjugates immobilized on nitrocellulose to screen monoclonal antibodies to abscisic acid. Analytical Biochemistry
151, 389–399.
Quarrie SA, Whitford PN, Appleford NEJ, Wang TL, Cook SK,
Henson IE, Loveys BR. 1988. A monoclonal antibody to (S)abscisic acid: its characterization and use in a radioimmunoassay for measuring abscisic acid in crude extracts of cereal
and lupin leaves. Planta 173, 330–339.
Saab IN, Sharp RE, Pritchard J, Voetberg GS. 1990. Increased
endogenous abscisic acid maintains primary root growth and
inhibits shoot growth of maize seedlings at low water
potentials. Plant Physiology 93, 1329–1336.
Simonneau T, Barrieu P, Tardieu F. 1998. Accumulation rate of
ABA in detached roots correlates with root water potential
regardless of age and branching order. Plant, Cell and
Environment 21, 1113–1122.
Tuberosa R, Sanguineti MC, Landi P. 1994. Abscisic acid
concentration in leaf and xylem sap, leaf water potential, and
stomatal conductance in maize. Crop Science 34, 1557–1563.
Walker-Simmons MK, Abrams SR. 1991. Use of ABA immunoassays. In: Davies WJ, Jones HG, eds. Abscisic acid:
physiology and biochemistry. Environmental Plant Biology
Series. Oxford: Bios-Scientific Publishers, 53–79.
Walker-Simmons MK, Reaney MJT, Quarrie SA, Perata P,
Vernieri P, Abrams SR. 1991. Monoclonal antibody recognition of abscisic acid analogs. Plant Physiology 95, 46–51.
Wang TL, Cook SK, Francis RJ, Ambrose MJ, Hedley CL.
1987. An analysis of seed development in Pisum sativum.
Journal of Experimental Botany 38, 1921–1932.