Production and characterization of polyclonal anti-S 20499 antibodies: Influence of the hapten structure on stereospecificity

Article abstract of DOI:10.1021/js9700107

Immunoassays were studied as an alternative to HPLC methods for the stereoselective determination of a chiral drug, S 20499, a new anxiolytic compound that is chemically related to buspirone. The production of highly stereospecific polyclonal antibodies was sought following the construction of appropriately optimized hapten-protein conjugates. This process involved the selection of the structure and the length of the spacer arm used to couple S 20499 to the carrier protein as well as deciding on the location of the coupling site with respect to the chiral center. Two haptens were prepared: one a derivative resembling the original structure of S 20499, with the effective addition of a carboxylic acid group, and a second with the effective addition of a butanoic acid moiety that is supposed to favor stereorecognition. Six stereospecific polyclonal antisera were obtained in rabbits with two groups of antibody families defined in terms of specificity. Both approaches gave high levels of stereospecificity (cross-reactivity towards the optical antipode of S 20499 ranged from 4.1% to <0.1%). Although it did not decrease the mean apparent affinity constant, the longer spacer improved antibody specificity by decreasing cross-reactions towards dealkylated S 20499 derivatives. Hence, the addition of a four carbon atom bridge should be a valuable tool for increasing antibody stereospecificity with no drawbacks in terms of specificity and affinity. It was also shown that long immunization periods appear to have no effect on the stereospecificity of the antibodies obtained.

Full text of DOI:10.1021/js9700107

Production and Characterization of Polyclonal Anti-S 20499 Antibodies:
Influence of the Hapten Structure on Stereospecificity
X
§
P
IERRE A. GOT* , GERALD
G
UILLAUMET‡, CLAIRE
BOURSIER-NEYRET*, AND
J
EAN-MICHEL
S
CHERRMANN
Received January 10, 1997, from the *Department de Bio-Analyse, Technologie SERVIER, 27 rue Euge`ne Vignat,
F-45000 Orle´ans, France, ‡Institut de CHIMIE ORGANIQUE et ANALYTIQUE associe´ au CNRS, Universite´ d’Orle´ans,
BP 6759, 45067 Orle´ans Cedex 2, France, and §INSERM Unite´ 26, 200 rue du Faubourg St. Denis, 75010 Paris, France.
Accepted for publication February 19, 1997X.
hapten on the specificity of the elicited antibodies, especially
Abstract
0 Immunoassays were studied as an alternative to HPLC
in terms of asymmetric center recognition. To develop this
approach, we describe herein the production and characteriza-
tion of polyclonal anti-S 20499 (Figure 1), the eutomer of
S 20244 and a potent agonist of 5HT1a receptors related to
buspirone.14 We also conducted a study to determine the
influence of the immunization period on the titre, the mean
apparent affinity, and the stereospecificity. The same proce-
dure was applied for the production of anti-S 20500 antibodies
(S 20500 being the distomer of S 20244). The question of
preferential recognition of the active configuration was ad-
dressed by comparing the stereospecificity obtained for anti-
S 20499 and anti-S 20500 antibodies.
methods for the stereoselective determination of a chiral drug, S 20499,
a new anxiolytic compound that is chemically related to buspirone. The
production of highly stereospecific polylclonal antibodies was sought
following the construction of appropriately optimized hapten−protein
conjugates. This process involved the selection of the structure and the
length of the spacer arm used to couple S 20499 to the carrier protein
as well as deciding on the location of the coupling site with respect to
the chiral center. Two haptens were prepared: one
a derivative
resembling the original structure of S 20499, with the effective addition
of a carboxylic acid group, and a second with the effective addition of a
butanoic acid moiety that is supposed to favor stereorecognition. Six
stereospecific polyclonal antisera were obtained in rabbits with two groups
of antibody families defined in terms of specificity. Both approaches gave
high levels of stereospecificity (cross-reactivity towards the optical antipode
of S 20499 ranged from 4.1% to <0.1%). Although it did not decrease
the mean apparent affinity constant, the longer spacer improved antibody
specificity by decreasing cross-reactions towards dealkylated S 20499
derivatives. Hence, the addition of a four carbon atom bridge should be
a valuable tool for increasing antibody stereospecificity with no drawbacks
in terms of specificity and affinity. It was also shown that long
immunization periods appear to have no effect on the stereospecificity of
the antibodies obtained.
Materials and Methods
Rea gen tss[3H]S 20499 (specific activity 49.6 Ci/mmol) was ob-
tained from CEA (Gif sur Yvette, France). S 20499 and all related
compounds, including compounds used for hapten synthesis and ethyl-
7-bromoheptanoate, were provided by Technologie Servier (Orle´ans,
France). Triethylamine, dimethylformamide (DMF), methanol, N-
methylmorpholine, isobutylchloroformate, ethyl-4-bromobutanoate,
potassium iodide, potassium carbonate, Freund’s complete and in-
complete adjuvants, and bovine serum albumin (BSA) were purchased
from Sigma Chemicals (St. Louis, MO). Ammonium sulfate, sodium
phosphate monobasic and dibasic, and citric acid were purchased from
Merck (Darmstadt, Germany).
P r ep a r a tion of th e Ha p ten sThe procedure for the hapten
synthesis is shown in Figure 2.
Introduction
(S)-8-[4-[N-(5-Met h oxy-3,4-d ih yd r o-2H -1-b en zop yr a n -5-yl)-
a m in o]bu tyl]-8-a za sp ir o[4.5]-d eca n e-7,9-d ion e (2)s5-Methoxy-
3,4-dihydro-3-amino-2H-1-benzopyran 1 (1.00 g, 5.59 mmol), 8-(4-
bromobutyl)-8-azaspiro[4.5]decane-7,9-dione (1.86 g, 6.15 mmol),
triethylamine (1.70 g, 16.76 mmol), and potassium iodide (catalytic
amount) were mixed in anhydrous DMF (15 mL). The solution was
stirred for 24 h at 70 °C under argon, then cooled to room temperature.
The solvent was removed under reduced pressure, and the residue
was taken up in 20 mL of methylene chloride and washed three times
with water. The organic layer was dried over anhydrous magnesium
sulfate, and the solvent was evaporated. The crude product was then
chromatographed on silica gel (eluent diethylether:methylene chloride
50:50) to give 1.52 g (68%) of pure 2 as a clear oil. IR (neat): 1720
and 1670 cm-1 (ν CdO); 1H NMR (CDCl3): δ 1.45-1.85 (m, 12H,
CH2), 2.42 (dd, J 1 ) 16.0 Hz, J 2 ) 7.1 Hz, 1H, CHAr), 2.56 (s, 4H,
CH2CO), 2.72 (t, J ) 7.0 Hz, 2H, CH2N), 2.8-3.0 (m, 1H, CHAr),
3.0-3.15 (m, 1H, CHN), 3.76 (t, J ) 7.0 Hz, 2H, CH2NCO), 3.80 (s,
3H, CH3O), 3.80-3.90 (m, 1H, CHO), 4.10-4.20 (m, 1H, CHO), 6.41
and 6.46 (2d, J ) 8.0 Hz, 2H, arom.), 7.04 (t, J ) 8.0 Hz, 1H, arom.).
(S)-Eth yl(4-{[(4-{7,9-d ioxo-8-a za sp ir o[4,5]d eca n -8-yl}bu tyl)-
(5-m et h oxy-3,4-d ih yd r o-2H -1-b en zop yr a n -5-yl)]a m in o}b u t a -
n oa te (3) a n d (S)-Eth yl(4-{[(4-{7,9-d ioxo-8-a za sp ir o[4,5]d eca n -
8-yl}b u t yl)(5-m e t h oxy-3,4-d ih yd r o-2H -1-b e n zop yr a n -5-yl)]-
a m in o}h ep ta n oa te (4)sCompound 2 (1.00 g, 2.50 mmol), ethyl-4-
bromobutanoate (1.95 g, 10.00 mmol), or ethyl-7-bromoheptanoate
(2.37 g, 10.00 mmol) for 3 or 4, respectively, triethylamine (0.76 g,
7.50 mmol), and potassium iodide (catalytic amount) were mixed in
anhydrous DMF (15 mL). The solution was stirred for 24 h at 70 °C
under argon, then cooled to room temperature. The solvent was
removed under reduced pressure, and the residue was taken up in
Despite the possible beneficial effects from the administra-
tion of drug substances containing various proportions of
stereoisomers,1 the development of single enantiomers, as
opposed to racemates, has been viewed preferential in the
light of possible differences in the pharmacokinetic and/or
pharmacodynamic properties of different configurations.2-4 In
all cases, stereospecific analytical methods must be developed
to measure all of the possible enantiomers present after
racemate dosing or to study chiral inversion, racemization,
or epimerization after the administration of single enanti-
omers. For these reasons, chiral chromatographic methods
are widely used.5-8 However, because the use of very potent
drugs leads to lowered doses, analytical methods have to be
sensitive enough to quantify extremely low concentrations of
drugs in biological fluids. Stereospecific immunoassays, which
combine sensitivity and selectivity, have recently been devel-
oped and applied in the domains of pharmacology9-11 and
toxicology.12 Several techniques allow the modulation of the
antibody-drug specificity.13 In terms of stereospecificity, the
preparation of an appropriate hapten conjugate appears to
be of major importance. The influence of the immunization
period on the stereospecificity could also be an important
factor to consider and study. The aim of the present work
was to examine the influence of the chemical structure of the
X Abstract published in Advance ACS Abstracts, April 15, 1997.
654 / Journal of Pharmaceutical Sciences
S0022-3549(97)00010-5 CCC: $14.00
© 1997, American Chemical Society and
American Pharmaceutical Association
Vol. 86, No. 6, June 1997

pressure, and the aqueous phase was washed twice with 28 mL of
ethyl ether. The aqueous phase was then adjusted to pH 6 with 1 M
hydrochloric acid. After three extractions with methylene chloride
(v/v), the organic layer was dried over anhydrous magnesium sulfate,
and the solvent was evaporated. The crude products were chromato-
graphed on silica gel (eluent, methylene chloride:methanol, 92:8) to
give 0.31 g (65%) or 0.26 g (55%) of pure 5 and 6, respectively, as
colorless oils. Compounds 5 and 6 correspond to the haptens of the
A4 and A7 series, respectively.
Figure 1
s
Structure of S 20499.
Compound 5: IR (neat): 1725 and 1670 cm-1 (ν CdO); 1H NMR
(CDCl3): δ 1.40-1.80 (m, 12H, CH2), 1.80-1.90 (m, 2H, CH2CH2-
COOH), 2.56 (t, J ) 7.0 Hz, 2H, CH2COOH), 2.58 (s, 4H, CH2CO),
2.60-2.70 (m, 1H, CHAr), 2.70-2.90 (m, 4H, CH2N), 2.90-3.10 (m,
1H, CHAr), 3.30-3.50 (m, 1H, CHN), 3.74 (t, J ) 7.0 Hz, 2H, CH2-
NCO), 3.78 (s, 3H, CH3O), 4.00 (t, J ) 9.2 Hz, 1H, CHO), 4.25-4.40
(m, 1H, CHO), 6.41 and 6.46 (d, J ) 8.0 Hz, 2H, aromatic), 7.06 (t, J
) 8.0 Hz, 1H, arom.).
Compound 6: IR (neat): 1725 and 1670 cm-1 (ν CdO); 1H NMR
(CDCl3): δ 1.20-1.80 (m, 20H, CH2), 2.30 (t, J ) 7.0 Hz, 2H, CH2-
COOH), 2.50-2.80 (m, 5H, CHAr and CH2N), 2.58 (s, 4H, CH2CO),
2.80-2.95 (m, 1H, CHAr), 3.10-3.30 (m, 1H, CHN), 3.70-4.00 (m,
3H, CHO and CH2NCO), 3.83 (s, 3H, CH3O), 4.20-4.35 (m, 1H,
CHO), 6.41 and 6.47 (2d, J ) 8.0 Hz, 2H, aromatic), 7.04 (t, J ) 8.0
Hz, 1H, aromatic).
P r ep a r a tion of Im m u n ogen ssThe reaction pathways used for
the synthesis of conjugates are shown in Figure 2. Haptens 5 and 6
(A4 and A7 series respectively) were coupled to BSA according to the
mixed anhydride method.15 A solution of each hapten (75 mg, 0.15
mmol) in 2 mL of DMF cooled to -15 °C was prepared, neutralized
with N-methylmorpholine (23 mg, 0.23 mmol in 1.5 mL of DMF), and
stirred for 10 min. Isobutyl chloroformate (25 mg, 0.18 mmol) in 1.5
mL of DMF was then added. After 10 min of stirring, the reaction
mixture was allowed to warm up to 0 °C, and a solution of BSA (100
mg, 0.15 µmol) in 5 mL of 50 mM phosphate buffer (pH 7.4) was added
in a dropwise manner. After filtration, the conjugates were exten-
sively dialysed against a buffer/DMF mixture followed by buffer alone
for 48 h. Following this procedure, partially soluble conjugates (7
and 8) were obtained and used for immunizations.
Ch a r a ct er iza t ion of t h e Cou p lin g b y Ma ss Sp ect r om -
etr ysLow-resolution electron-impact mass spectra (EI-MS, 70 eV)
were recorded on a VG Analytica 70 VS mass spectrometer. Samples
(1-10-µg immunogen aliquots) were introduced via the direct inser-
tion probe that was heated to 430 °C. The source was maintained at
180 °C, and an accelerating voltage of 8 kV was applied. Scanning
spectra were acquired over the range m/z 45-800. High-resolution
measurements were made on the same instrument by peak matching
at a resolution of 10 000 (M/∆M).
Im m u n iza tion P r oced u r esTwo groups of three white New
Zealand rabbits (IFFA CREDO, France) were immunized by the
administration of 10 intradermal dorsal injections (100 µL for each
injection). Primary immunizations were composed of 500 µL of 0.9%
NaCl containing 100 µg of immunogen emulsified in 500 µL of
Freund’s complete adjuvant. Subsequent immunizations, at 3-8-
week intervals, were of the same volume, with complete adjuvant
replaced by incomplete adjuvant. After the fourth immunization,
rabbits were bled from the ear vein 7-10 days after each injection.
The last bleedings were performed ∼10 months after the first
immunization.
Assa y P r oced u r e for th e Ch a r a cter iza tion of th e P olyclon a l
An tibod iessA liquid-phase radioimmunoassay (RIA) was used to
follow the antisera titre and to characterize the elicited antibodies.
For titration experiments, 100 µL of rabbit antiserum were incubated
at various dilutions in citrate buffer (0.03 M, pH 6.4) with 200 µL of
0.6 nM [3H]S 20499, 50 µL of human plasma, and 150 µL of a 5-mg/
mL BSA solution in citrate buffer. The titre is determined as the
antiserum dilution factor that corresponds to the binding of 50% of
[3H]S 20499.
Figure 2
s
Hapten synthesis and immunogen preparation.
20 mL of methylene chloride and washed three times with water. The
organic layer was dried over anhydrous magnesium sulfate, and the
solvent was evaporated. The crude products were then chromato-
graphed on silica gel column (eluent, ether:ethylacetate, 60:40) to give
0.70 g (55%) or 0.63 g (45%) of pure 3 or 4, respectively, as clear oils.
Compound 3: IR (neat): 1730 and 1670 cm-1 (ν CdO); 1H NMR
(CDCl3): δ 1.26 (t, J ) 7.0 Hz, 3H, CH2CH3) 1.40-1.80 (m, 14H, CH2),
2.33 (t, J ) 7.0 Hz, 2H, CH2COOEt), 2.40-2.50 (m, 1H, CHAr), 2.50-
2.70 (m, 4H, CH2N), 2.59 (s, 4H, CH2CO), 2.80-3.00 (m, 1H, CHAr),
3.00-3.20 (m, 1H, CHN), 3.70-3.80 (m, 1H, CHO), 3.70-3.80 (m,
2H, CH2NCO), 3.80 (s, 3H, CH3O), 4.12 (q, J ) 7.0 Hz, 2H, CH2-
CH3), 4.20-4.30 (m, 1H, CHO), 6.41 and 6.45 (2d, J ) 8.0 Hz, 2H,
aromatic), 7.03 (t,. J ) 8.0 Hz, 1H, aromatic).
Compound 4: IR (neat): 1730 and 1670 cm-1 (ν CdO); 1H NMR
(CDCl3): δ 1.25 (t, J ) 7.0 Hz, 3H, CH2CH3) 1.30-1.80 (m, 20H, CH2),
2.28 (t, J ) 7.0 Hz, 2H, CH2COOEt), 2.40-2.80 (2m, 5H, CHAr and
CH2N), 2.57 (s, 4H, CH2CO), 2.70-2.90 (m, 1H, CHAr), 3.00-3.20
(m, 1H, CHN), 3.60-3.80 (m, 1H, CHO), 3.60-3.80 (m, 2H, CH2NCO),
3.82 (s, 3H, CH3O), 4.11 (q, J ) 8.0 Hz, 2H, CH2Et), 4.20-4.30 (m,
1H, CHO), 6.40 and 6.45 (2d, J ) 8.0 Hz, 2H, aromatic), 7.03 (t, J )
8.0 Hz, 1H, aromatic).
(S)-4-{[(4-{7,9-Dioxo-8-a za sp ir o[4,5]d eca n -8-yl}bu tyl)(5-m eth -
oxy-3,4-d ih yd r o-2H-1-ben zop yr a n -5-yl)]a m in o}bu t a n oic a cid
(5) a n d (S)-4-{[(4-{7,9-Dioxo-8-a za sp ir o[4,5]d eca n -8-yl}bu t yl)-
(5-m e t h oxy-3,4-d ih yd r o-2H -1-b e n zop yr a n -5-yl)]a m in o}h e p -
ta n oic a cid (6)sCompound 3 (0.50 g, 0.97 mmol) or 4 (0.50 g, 0.90
mmol) was dissolved in 35 mL of methanol. A solution of K2CO3 (0.41
g, 3.00 mmol) dissolved in 15 mL of water was then added, and the
resulting solution was stirred for 3 days at room temperature to
liberate acidic function. The methanol was removed under reduced
In competitive experiments, increasing amounts of cold drugs in
BSA-citrate buffer (100 µL, up to 2000 ng/mL) or BSA-citrate buffer
(100 µL) were added to the incubation medium, while maintaining a
total reaction volume of 500 µL. Nonspecific binding was determined
by replacing antiserum with citrate buffer. In both experiments,
reactions were allowed to proceed for 1.5 h at room temperature.
Proteins were then precipitated with 500 µL of a saturated ammonium
sulfate solution. Bound [3H]S 20499 was separated from the free
ligand by centrifugation. A 500-µL aliquot of the supernatant was
Journal of Pharmaceutical Sciences / 655
Vol. 86, No. 6, June 1997

added to 4 mL of Picofluor (Packard, France) and the radioactivity
was counted in a Packard Tricarb counter (Packard, France).
Displacement curves and binding parameters were obtained and
calculated with the GraphPad InPlot program (GraphPad, San
Diego, California). The 50% inhibitory concentration (IC50) was
calculated as the ligand concentration required to displaced 50% of
the labeled ligand. In accordance with the criteria of Abraham,16
cross-reactivity was expressed as the ratio (percent) of the IC50
for S 20499 with respect to the IC50 for the test ligand. Mean
apparent affinity constants were estimated according to the method
of Mu¨ller.17
Figure 3sStructure of molecular ions including a peptidic link, characteristic of
the coupling between haptens and BSA (n
m/z 526 for A7 series).
) 3, m/z 484 for A4 series; n ) 6,
Results and Discussion
Con str u ction of th e Ha p ten sThe structurally important
features of S 20499 are an azaspiro moiety, attached to the
amine function of an aminochroman moiety through a four-
carbon-atom methylene bridge. An asymmetric center is
located in the 3-position of the chroman moiety, and a propyl
side chain is coupled to the amine nitrogen atom (Figure 1).
The objectives of the study were first to produce anti-S 20499
stereospecific antibodies and, second to make them as specific
as possible towards potential metabolites. As the molecular
weight of S 20499 is too low to induce an immunogenic
response in rabbits, a coupling between S 20499 and a carrier
protein was undertaken to produce an immunogen. To carry
out this coupling, the introduction of a functional group on
S 20499 to yield the desired hapten was required due to the
absence of derivatizable functionalities in this compound.
S 20499, like other structurally related molecules (buspirone,
tiospirone), can be hydroxylated on the azaspiro moiety during
metabolism.18,19 Dealkylation of the methoxy group and
hydroxylation(s) of the chroman moiety were also considered
as potential biotransformations, as was the loss of the propyl
side chain.
The pioneering studies of Landsteiner20 showed that anti-
bodies resulting from conjugates are generally more selective
for the part of the molecule not involved in protein linkage.
With this in mind, a coupling site on the propyl side-chain
was selected (Figure 2). In underlining this selection, it
should be pointed out that several studies concerning ste-
reospecific immunoassays have demonstrated that stereorec-
ognition may be increased when the asymmetric center is
farther away from the coupling site.21 Having chosen an
appropriate coupling site, two hapten series were selected for
synthesis. In the first of these series, a structure close to that
of S 20499 was chosen, with the effective introduction of a
carboxylic function at the end of the propyl chain (Figure 2,
A4 series). Again, in justifying this choice, some studies have
previously shown that the presence of long spacer arms, or
the use of haptens not closely related to the free drug, ap-
pears to decrease the emergence of high-affinity antibodies
towards the unchanged (i.e., nonderivatized drug).22,23 Thus,
in the second series, to study the influence of the distancing
of the coupling site from the asymmetric center, a hepta-
noic acid group was added in place of the propyl chain of
S 20499 (Figure 2, A7 series). In both series, methylene
spacer groups were chosen because of their low immunogenic
reactivity. In effect, highly immunogenic groups, sometimes
used as a spacer arm for the coupling of haptens to carriers,
may interact with the paratope of elicited antibodies. For
low molecular weight drugs, this increases the risk of not
obtaining the three minimum interactions between haptens
and antibodies necessary for stereospecificity. Finally, be-
cause enantiomerically pure immunogens favor the forma-
tion of enantioselective antibodies,24 the chemical structures
of the haptens were fully identified and the enantiomeric
purities for each of the hapten series were found to be
>99%.
Figure 4sAnti-S 20499 antibody titre from rabbits nos. 2 and 6 as a function of
time.
Im m u n ogen P r ep a r a tion sThe hapten-BSA ratio could
not be easily determined because radiolabeled hapten was
not available and the conjugates were not totally soluble.
The immunization procedure is time consuming, so it was
important to have at least some qualitative information
relating to the coupling. An original indirect characteriza-
tion of the coupling was performed by mass spectrometry
in the electronic ionization mode. The analytical conditions
did not allow us to record the full spectrum of the hapten-
BSA complex, nevertheless ion fragments compatible with a
modified hapten structure, including the peptidic link, were
found at m/z 484 (A4 series) and 526 (A7 series), (Figure 3).
These fragments were not found in the control in the pres-
ence of BSA and all other reagents used in the formation of
the conjugates. Significantly, high-resolution analyses indi-
cated that the difference between the theoretical masses
and the masses found were as little as 1 ppm. The im-
munological response obtained in all rabbits confirmed that
the coupling was efficient and the hapten-BSA ratio was
sufficient.
An tibod y p r od u ction sSix rabbits, three from each hap-
ten series (A4 and A7, Figure 2), were immunized against
S 20499, the pharmacologically active enantiomer, and the
elicited antibodies were checked as described. The six animals
showed significant antibody response from the first bleeding,
taken after the fourth conjugate administration. The produc-
tion of anti-S 20499 antibodies followed the pattern shown in
Figure 4, which describes the response kinetics for two rabbits
representative of both series of immunogens. The highest
titres were observed after the first and the third bleedings,
with a marked decrease for rabbit no. 2 after the second
bleeding and a slight decrease for rabbit no. 6 up until the
last bleeding. The third bleeding yielded a high titre for all
rabbits, so the characterization of antibodies (in particular
their specificity) was performed following this bleeding in all
six cases.
The IC50 values, obtained from S 20499 displacement curves
ranging from 4.0 to 15.5 nM (Table 1), led to a mean apparent
affinity constant (calculated according to the Mu¨ller method17)
of 1 × 108 to 5 × 109 L/mol, showing no clear-cut differences
between the immunized rabbits from both series. The evolu-
tion of the IC50 throughout the immunization process was
656 / Journal of Pharmaceutical Sciences
Vol. 86, No. 6, June 1997

Table 1sCross-Reactivity with Anti-S 20499 Antisera
* IC50, Concentration required to displace 50% of [3H]S 20499. **CR, Cross-reactivity expressed as the ratio of the IC50 for S 20499 to that of the test compound.
***L, denotes rabbits nos. 1 to 6.
followed for rabbits nos. 2 and 6. No apparent differences
were seen from the first to the last bleedings.
already been obtained with the shorter spacer arm, and no
apparent difference was found in the two series. In this case,
even if no conclusion could be drawn about the value of dis-
tancing the coupling site from the asymmetric center, we were
at least able to confirm that the methylene spacer allows this
distancing without decreasing affinity and could favorably be
used for the construction of other immunogens. Concerning
the evolution of the antibody stereospecificity throughout the
immunization process, Abraham’s index remained <1% for
the two fully studied antisera (rabbits nos. 2 and 6, Figure
5). This result indicates that a short immunization period is
Ser u m An tibod y Sp ecificitysSter eosp ecificitysCross-
reactivity was expressed as the concentration of the compound
required to inhibit 50% of the binding of the antibody to
tritiated S 20499 (IC50), according to Abraham’s index.16
Clear-cut stereoselectivity was found for both antibody series,
as shown in Table 1, with cross-reactivity ranging from 4.1
to <0.1% for S 20500. From these results, the methylene
spacer groups appear to be good enough to allow stereospecific
recognition. Moreover, a high level of stereospecificity had
Journal of Pharmaceutical Sciences / 657
Vol. 86, No. 6, June 1997

differences in cross-reactivities were observed as a conse-
quence of the immunogen series. Cross-reactivities towards
the N-depropylated S 20499 (Y 464) and S 20499 hydroxylated
in the 4-position of the chroman moiety (S 21556), two other
potential metabolites, were appreciably decreased with anti-
bodies obtained from the longer of the two spacer groups
(cross-reactivities towards Y 464 of 26.1 ( 7.5% compared with
12.4 ( 1.7% for A4 and A7 series, respectively; cross-
reactivities towards S 21556 of 15.8 ( 6.3% compared with
6.0 ( 1.1% for A4 and A7 series, respectively; Table 1). As
the ligands tested (Y 464 and S 21556) were chemically
modified near the coupling site, a higher specificity of antibod-
ies was expected as a consequence of the longer spacer arm
between the hapten and BSA.
Conclusion
Highly stereospecific antibodies were sought against S 20499,
notably a 5HT1a agonist chemically related to buspirone. The
choice of a hapten structurally resembling that of the enan-
tiomer proved to be a good means of obtaining stereospecific
antibodies with high affinities, because the asymmetric center
is located farther away from the coupling site. The distancing
of the asymmetric center by effectively lengthening the spacer
arm with a four carbon atom methylene bridge was carried
out as a means of increasing stereospecificity. Both ap-
proaches gave high levels of stereospecificity. Although it did
not decrease the mean apparent affinity constant, the longer
spacer improved antibody specificity. Hence, the additional
four carbon atom bridge should be a valuable tool for increas-
ing antibody stereospecificity, with no drawbacks in terms of
specificity and affinity. Although the coupling site was
relatively distant from the chroman and the azaspiro moieties,
which are vulnerable to hydroxylations, none of the antibodies
exhibited narrow specificities. This result may be explained
by the folding up of the molecule onto BSA due to the
flexibility of the spacer arm. To avoid this hypothetical folding
up in future studies, or at least limit it, one may consider
integrating a more rigid spacer arm, such as the crotonate
structure described for anti-pentobarbital antibodies.27
Figure 5sEvolution of the antibody stereospecificity from the first (1) to the last
(7) bleedings: (A) rabbit no. 2, A4 series; (B) rabbit no. 6, A7 series. Displacement
curves obtained with S 20499 and S 20500 are represented by solid and dashed
lines, respectively.
consistent with the induction of highly stereospecific antibod-
ies. Further improvement was not found after subsequent
immunization steps.
Lastly, four out of the six anti-S 20499 antibodies did not
cross-react with S 20500 (the distomer of S 20244) to any
significant extent, whereas the two others showed only slight
cross-reactivities. Antibodies raised in response to distomer-
conjugates (four rabbits) all cross-reacted in the range of 0.7
to 7.4% with S 20499 (data not published), indicating the
possible preferential recognition of S 20499, the more active
configuration. These results may be compared with earlier
studies concerning â-adrenergic antagonists, where the anti-
bodies produced recognized the pharmacologically active
configurations preferentially.25 Moreover, other work per-
formed with a nonchiral hapten led to the production of
antibodies that recognized an active configuration preferen-
tially when the ligand studied was chiral.26 All of these
results suggest that antibodies could exhibit higher affinities
for pharmacologically active enantiomers.
References and Notes
1. Kupfer, A.; Roberts, R. K.; Schenker, S.; Branch, R. A. J .
Pharmacol. Exp. Ther. 1981, 218, 193-199.
2. Campbell, D. B. Eur. J . Drug Metab. Pharmacokinet. 1990,
15(2), 109-125.
Sp ecificity Tow a r d s Str u ctu r a l An a logssPotential me-
tabolites and analogs of S 20499 were studied for concentra-
tions up to at least 2000 ng/mL to determine their cross-
reactivities. The results from the different compounds tested
(namely, cleavage compounds and those with modifications
to the chiral center, to the propyl side-chain, and to both the
chroman and azaspiro moieties) are given in Table 1. The
compounds not recognized by antibodies were those that had
lost the chroman or the azaspiro moieties. Clearly, these two
parts of S 20499 belong to the epitope recognized by the
antibodies. Hydroxylation or dealkylation led to various
decreases in cross-reactivity depending on the antibodies, and
none of the six antisera were highly specific of S 20499.
Overall, two antibody families can be defined that are
unrelated to the immunogen series. One family is directed
towards the chroman moiety (rabbits 2, 4, and 6), with a
higher level of cross-reactivity against S 21555 (the 2-hy-
droxyazaspiro derivative) and a lower cross-reactivity against
S 20580 (the 5-hydroxychroman derivative), whereas the other
family shows opposing cross-reactivity results for the same
potential metabolites (rabbits 1, 3, and 5) and is predomi-
nantly directed towards the azaspiro moiety. Profitable
3. Drayer, D. E. Clin. Pharmacol. Ther. 1986, 40, 125-132.
4. Lee, E. J . D.; Williams, K. M. Clin. Pharmacokinet. 1990,
18(5), 339-345.
5. Gal, J . In Drug Stereochemistry; Wainer, J . W., Ed.; M. Dek-
ker: New York, 1993; pp 65-106.
6. Srinivas, N. R.; Igwemezie, L. N. Biomed. Chromatogr. 1992,
6, 163-167.
7. Mehta, A. C. J . Chromatogr. 1988, 426, 1-13.
8. Wainer, I. W. In Drug Stereochemistry; Wainer, J . W., Ed.; M.
Dekker: New York, 1993; pp 139-182.
9. Sahui-Gnassi, A.; Pham-Huy, C.; Galons, H.; Warnet, J . M.;
Claude, J . R.; Duc, H. T. Chirality 1993, 5, 448-454.
10. Sidki, A. M.; Al-Abdulla, H.; Rowell, F. J . Clin. Chem. 1987,
33(4), 463-467.
11. Rominger, K. L.; Mentrup, A.; Stiasni, M. Drug Res. 1990,
40-II(8), 887-895.
12. Niwaguchi, T.; Kanda, Y.; Kishi, T.; Inoue, T. J . Forensic Sci.
1982, 27(3), 592-597.
13. Chappey, O.; Sandouk, P.; Scherrmann, J . M. S. J . Clin.
Immunoassay 1992, 15(1), 51-56.
14. Podona, T.; Guardiola-Lemaˆıtre, B.; Caignard, D. H.; Adam, G.;
Pfeiffer, B.; Renard, P.; Guillaumet, G. J . Med. Chem. 1994,
37(12), 1779-1793.
15. Bodansky, A. In The Practice of Peptide Synthesis-Reactivity and
Structure Concepts in Organic Chemistry, vol. 21; Hafner, K.;
658 / Journal of Pharmaceutical Sciences
Vol. 86, No. 6, June 1997

Lehn, J . M.; Rees, C. W.; von Rague´ Schleyer, P.; Trost, B. M.;
Zahradnik, R., Eds.; Springer-Verlag: Berlin, Heidelberg, New
York, Tokyo, 1984; pp 109-110.
22. Findlay, J . W. A.; Warren, J . T.; Hill, J . A.; Welch, R. M. J .
Pharm. Sci. 1981l 70(6), 624-631.
23. Morel, A.; Darmon, M.; Delaage, M. Mol. Immunol. 1990,
27(10), 995-1000.
16. Abraham, G. E. J . Clin. Endocrinol. 1969, 29, 866-870.
17. Mu¨ller, R. J . Immunol. Meth. 1980, 34, 345-352.
18. J ajoo, H. K.; Mayol, R. F.; Labudde, J . A.; Blair, I. A. Drug
Metab. Dispos. 1989, 17(6), 625-633.
24. Cook, C. E. In Drug Stereochemistry; Wainer, J . W., Ed.; M.
Dekker: New York, 1993; pp 35-64.
25. Sawutz, D. G.; Sylvestre, D.; Homcy, C. J . J . Immunol. 1985,
135(4), 2713-2718.
26. Chatenoud, L.; Villemain, F.; Hoebeke, J .; Garbarg, M.; Korner,
M.; Gros, C.; Ruat, M.; Cazenave, P. A.; Ganellin, C. R.; Bach,
J . F.; Schwartz, J . C. Mol. Pharmacol. 1988, 34, 136-144.
27. Cook, E. C.; Seltzman, T. B.; Tallent, C. R.; Lorenzo, B.; Drayer,
D. E. J . Pharmacol. Exp. Ther. 1987, 241(3), 779-785.
19. Cipollina, J . A.; Ruediger, E. H.; New, J . S.; Wire, M. E.;
Shepherd, T. A.; Smith, D. W.; Yevich, J . P. J . Med. Chem. 1991,
34, 3316-3328.
20. Landsteiner, K. In The Specificity of Serological Reactions;
Harvard University: Boston, 1945 (reprinted by Dover, New
York, 1962).
21. McGilliard, P. L.; Olsen, G. D. J . Pharm. Exp. Ther. 1980,
215(1), 205-212.
J S9700107
Journal of Pharmaceutical Sciences / 659
Vol. 86, No. 6, June 1997