Disulfide symmetric dimers as stable pre-hapten forms for bioconjugation: A strategy to prepare immunoreagents for the detection of sulfophenyl carboxylate residues in environmental samples

Article abstract of DOI:10.1002/chem.200701232

A convenient, generic synthesis of bioconjugates from haptens with a thiol group has been established. The corresponding haptens are synthesized as stable symmetric dimmers through a disulfide bond that is reduced immediately before conjugation with the a

Full text of DOI:10.1002/chem.200701232

DOI: 10.1002/chem.200701232
Disulfide Symmetric Dimers as Stable Pre-Hapten Forms for Bioconjugation:
A Strategy to Prepare Immunoreagents for the Detection of Sulfophenyl
Carboxylate Residues in Environmental Samples
M.-Carmen EstØvez, Roger Galve, Francisco Sµnchez-Baeza, and M.-Pilar Marco*[a]
Abstract: A convenient, generic syn-
thesis of bioconjugates from haptens
with a thiol group has been established.
The corresponding haptens are synthe-
sized as stable symmetric dimmers
through a disulfide bond that is re-
duced immediately before conjugation
tion of the SPC family. With this pur-
pose, two types of immunizing haptens
have been synthesized and used to pre-
pare bioconjugates and raise antibod-
ies. Type-A bioconjugates (SPCA–pro-
tein) were prepared by synthesizing
type-A haptens as stable symmetric
dimers, generically 2,2’-dithiobis[5-{4-
(N-ethylsulfamoyl)}phenylalkanoic
acids] (X-Cz-S-SPC). On the other
hand, type-B bioconjugates (SPCB–pro-
tein) were prepared by treating the car-
boxylic groups of the corresponding 4-
sulfophenylalkanoic acids (X-Cz-SPC)
with the amino groups of the lysine res-
idues by using classical carbodiimide
procedures. Type-A haptens produced
antibodies with a much higher avidity
for the target analyte. Under competi-
tive immunochemical configurations
(As112/2-C5–ovalbumin), these anti-
bodies can reach a limit of detection
(LOD) of 40 ngLÀ1 with an IC50 value
of 200 ngLÀ1 for 3-C6-SPC, which
opens up the possibility of trace con-
tamination of edible waters by surfac-
tants with 3-C6-SPC as a marker of
LAS pollution. A comparative study of
the properties of the three families of
polyclonal antibodies produced re-
vealed that antibodies raised through
with the aid of a di(n-butyl)phenyl-
N
phosphine polystyrene (DBPP) resin.
This strategy was used to prepare hap-
tenized biomolecules and to raise anti-
bodies against short-alkyl-chain sulfo-
phenyl carboxylates (X-Cz-SPCs; X is
the position of the benzylic group and
z is the alkyl-chain length) formed
after degradation of the widely used
domestic and industrial linear alkylben-
zene sulfonates (LASs) surfactants. Be-
cause of the complexity of the LASs
technical mixture, homologous and
pseudo-heterologous
immunization
strategies produced antibodies with a
broader specificity versus the SPC
family. These results indicate that this
approach could be useful in avoiding
synthetic difficulties associated with
preparing haptens that preserve all the
most important chemical functionalities
of the molecule.
Keywords: antibodies · immuniza-
tion
· immunoassays · resins ·
pseudo-heterologous
immunization
solid-phase synthesis · sulfophenyl
carboxylates
strategies have been studied with the
aim of broadening antibody recogni-
Introduction
these particular groups while introducing other chemical
functionalities that allow covalent attachment to the desired
biomolecule. However, the reactivity of such new chemical
functionalities must be orthogonal to the reactivities of the
other chemical groups, which may give additional difficulties
in hapten synthesis and further covalent coupling of highly
functionalized small organic molecules to biomacromole-
cules. Such a circumstance was confronted within the Euro-
pean Project aimed at developing an immunosensor for the
detection of industrial pollutants, such as the sulfophenyl
carboxylates (SPCs).[3]
The preparation of bioconjugates with small organic mole-
cules often faces the problem that important functional
(and/or antigenic) chemical groups can be blocked by the
biomolecule if used as conjugation sites.[1,2] Hapten deriva-
tives can be designed and synthetically prepared to respect
[a] Dr. M.-C. EstØvez, Dr. R. Galve, Dr. F. Sµnchez-Baeza, M.-P. Marco
Applied Molecular Receptors Group (AMRg)
CSIC, CIBER of Bioengineering, Biomaterials and Nanomedicine
IIQAB-CSIC, Jorge Girona, 18—26, 08034-Barcelona (Spain)
Fax : (+34)93-204-5904
SPCs are the main metabolites of the degradation of
linear alkylbenzene sulfonates (LASs),[4,5] which are used as
surfactants in domestic and industrial purposes in high
amounts worldwide. Their frequent detection in the environ-
E-mail: mpmqob@iiqab.csic.es
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
1906
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1906 – 1917

FULL PAPER
ment at significant concentration levels[6–8] has raised con-
cern over their potential synergistic role, which favors the
permeation of other toxicants through biological membranes
thanks to their amphoteric and surfactant properties.[9,10] Mi-
crobial degradation[11] is very fast and can be considered to
be the major elimination pathway, which involves the oxida-
tion of the terminal methyl group of the alkyl chain (w oxi-
dation) followed by the successive loss of two carbon units
as a result of b-oxidation processes.[4] The complexity of the
LAS mixture (i.e., alkyl chains of 10–14 carbon units and
different positional isomers at the benzylic position) deter-
mines the consequent complexity of the SPC mixture
formed after degradation of LASs. Thus, the first com-
pounds generated after the degradation of LAS are SPCs
with long alkyl chains (>10 carbon units), but the relatively
rapid b-oxidation processes lead to SPCs with short alkyl
chains (see Scheme 1). Particularly, those SPC isomeric mix-
tures with alkyl chains of five and six carbon units (X-C5-
SPC and X-C6-SPC, X indicates that the aromatic ring can
be placed at any position) have been considered by many
authors as the more likely key intermediates of the degrada-
tion pathway.[12–14] Nevertheless, certain analytical limitations
may have prevented the proper detection of shorter-alkyl-
chain SPC congeners. Thus, the high polarity of these short-
alkyl-chain SPCs, as a result of the presence of the carboxyl-
ic acid and sulfonate groups, causes significant difficulties in
extracting them efficiently from the aqueous media before
chromatographic analysis.[12,15–17]
The preparation of immunoreagents able to detect a wide
range of short-alkyl-chain SPCs was one of the goals of this
investigation. However, the production of antibodies with
class-directed specificity is an uncertain goal that has been
persecuted by many research groups with the purpose of de-
veloping immunochemical screening procedures able to
detect, to a similar extent, different congeners of structurally
related families of substances (i.e., sulfonamide antibiot-
ics,[18–20] sulfonyl urea herbicides,[21,22] and so forth). The
strategies employed frequently consist of preparing hapten–
protein immunoconjugates that expose common features of
the chemical structures to the immune system. However,
this approach, which may require long and intricate synthet-
ic procedures to keep important antigenic epitopes free, has
often failed.
We recently reported our attempts to produce generic an-
tibodies for SPCs through using an equimolar mixture of im-
munogens prepared by coupling six short-alkyl-chain SPC
haptens, thus maximizing the recognition of the common
sulfonic group (type-B haptens) to horseshoe crab hemocya-
nin (HC; SPCmix-B; see Scheme 1).[23] However, these anti-
bodies only recognized X-C5-SPCs and X-C6-SPCs preferen-
tially, while other SPCs were poorly recognized, thus indi-
cating that the different lengths and branching of the alkyl-
Scheme 1. Chemical structures of the six representative SPCs and haptens.
Chem. Eur. J. 2008, 14, 1906 – 1917
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1907

M.-P. Marco et al.
chain patterns played a decisive role in the recognition pro-
cess. More success has recently been achieved from produc-
ing generic antibodies by using antibody engineering meth-
odologies.[24–26] Alternatively, the heterologous immunization
approach has been reported a few times with the aim of elic-
iting antibody responses against more than one epitope,
mainly in the context of catalytic antibodies.[27–29] This idea
prompted us to investigate the possibility of expanding anti-
body recognition of the short-alkyl-chain SPC family
through this strategy.
tibody features. Moreover, we present a simple and generic
strategy to prepare bioconjugates with highly functionalized
small organic molecules.
Results
Haptendesignand synthesis of the SPC dimers : We planned
to synthesize the two types of haptens for each of the six
short-alkyl-chain SPC represen-
The heterologous immunization approach envisaged by
Masamune and co-workers[27,28,30] uses a combination of two
different, but structurally related, haptens thus maximizing
distinct epitopes. With this purpose, we addressed the syn-
thesis of two haptens maximizing recognition of the alkyl
carboxylic chains (type-A haptens) and of the sulfonic
groups (type-B haptens) to use them in a heterologous im-
munization protocol. For the type-B haptens, the carboxylic
groups of the alkyl chain could be directly used for conjuga-
tion. However, the preparation of type-A haptens required
the introduction of a spacer arm with an orthogonal func-
tionality. Thus, the SPC molecules could be generically
tatives. The carboxylic acid
groups could directly be used
for bioconjugation of the six
type-B haptens, although it was
necessary to introduce a linker
with a thiol group for the type-
A haptens (Scheme 2).
To prevent oxidation or
avoid intra/intermolecular reac-
Scheme 2. Type-A hapten.
À
tions, free SH (thiol) groups
are commonly protected as thi-
oethers or thioesters, although
treated with a H2N
G
the formation of thiazolidines, unsymmetrical disulfides, and
S-sulfenyl derivatives has also been used to a more limited
extent.[31] Since oxidation to form the corresponding sym-
metrical disulfides is also a possible strategy, we envisaged
the synthesis of the haptens as disulfide pre-haptens. Finally,
cleavage of the disulfide bond could be approached immedi-
ately before bioconjugation to a conveniently derivatized
protein.
The synthesis of the dimers was performed by preparing
the corresponding chlorosulfonic derivatives of phenylcar-
boxylic acids 1–7 followed by the formation of two sulfona-
mide bonds using cystamine dihydrochloride (Scheme 3). In
the first step, it was important to always have an excess of
the chlorosulfonic acid to prevent the formation of the cor-
responding sulfones as by-products.[33] Moreover, the phe-
nylcarboxylic acids were first protected as esters to avoid in-
tramolecular cyclization through Friedel–Crafts acylation re-
actions. For phenylcarboxylic acids 1–5, the use of a bulky
protecting group, such as the butyl group, was required to
a sulfonamide, where X would be the necessary orthogonal
chemical group for bioconjugation. Another carboxylic
group would compete with the same group on the SPC alkyl
chain at the conjugation step. Similarly, the use of linkers to
introduce a free amine group was ruled out since the forma-
tion of (a SPC–CONH )n species could take place in addi-
tion to SPC-NHCO-protein conjugates. The same applied if
an alcohol group was introduced to form ester bonds with
the protein. Alternatively, thiol groups provided an appro-
priate reactivity when carboxylic, aldehyde, amine or halo-
gen groups, present in the same small organic molecule,
want to be preserved. Moreover, the conjugation reaction
could take place through a variety of commercially available
bifunctional cross-linkers chosen to react with the amino
groups of the lysine residues on one side and with the thiol
group on the other side.[32]
We report herein the investigation into the potential of
pseudo-heterologous immunization procedures to tailor an-
Scheme 3. Synthesis of SPC–SH.
1908
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1906 – 1917

Hapten–Protein Bioconjugates as Immunoreagents
FULL PAPER
ensure unique chlorosulfonation at the para position, thus
preventing reactivity of the ortho positions that would lead
to polysubstituted derivatives. Protection of the carboxylic
acids as methyl esters was sufficient for the case of phenyl-
carboxylic acids 6 and 7 with longer alkyl chains. The mod-
erate yield of 1a--7a was attributed to the instability of the
chlorosulfonate groups in the aqueous media used to treat
the reaction.
Preparationof the hapten–biomolecule conjugates : For the
type-A haptens, N-succinimidyl-3-maleimidyl propanoate
(N-SMP) and N-succinimidyl bromoacetate (N-SBrA) were
used as heterobifunctional cross-linkers to prepare hemocya-
nin (HC), bovine serum albumin (BSA), ovalbumin (OVA),
and conalbumin (CONA) conjugates. Hemocyanin conju-
gates are selected for use as immunoagents because of the
high immunogenic potential of this protein, while the prepa-
ration of bioconjugates with albumins was carried out to
evaluate them as competitors in the immunoassay. The prep-
aration of the haptenized biomolecules comprised three
steps: 1) cleavage of the dimer under reducing conditions,
2) coupling of the linker N-SMP (or N-SBrA) to the protein,
and 3) conjugation of the hapten to the maleidimidylpropa-
noyl- or bromoacetyl-derivatized proteins (MP–protein and
BrA–protein, respectively; Scheme 4). The whole procedure
was initially evaluated using 5-C5-S-SPC as a model hapten,
BSA as the protein, and N-SMP as the cross-linker. An im-
portant aspect of the synthesis was the use of the correct
hapten/cross-linker/protein ratio to avoid free epitopes of
the cross-linker in the final immunogen conjugate.
Surprisingly, the formation of meta-monosubstituted prod-
ucts as by-products in a low yield (6–12%) was also ob-
served for the phenylcarboxylates 1–3 containing the phenyl
group at the a position to the carboxylic group, as con-
1
firmed by H and 13C NMR spectroscopic analysis (see the
analytical data for 1b--3b in the Supporting Information).
Finally, the formation of the symmetric dimers was achieved
in one step in good yields by directly treating two molecules
of chlorosulfonyl derivatives 1a–7a with one molecule of
cystamine dihydrochloride through the formation of two sul-
fonamide bonds. The hydrolysis of the esters led to the final
desired products: the dimers 2-C3-S-SPC, 2-C4-S-SPC, 2-C5-
S-SPC, 3-C4-S-SPC, 3-C5-S-SPC, 3-C6-S-SPC, and 5-C5-S-
SPC.
Scheme 4. Strategy used to prepare type-A hapten–protein bioconjugates.
Chem. Eur. J. 2008, 14, 1906 – 1917
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1909

M.-P. Marco et al.
Regarding the reduction of the dimer, the efficiency of
different reducing agents was evaluated using the Ellman
test[34] to follow the formation of the thiol groups.
NaCNBH3 was initially chosen with the intention of per-
forming this reaction in aqueous media and in the presence
of the protein; however, the degree of conversion was very
poor, even with a high excess of
haptens, except for 5-C5-S-SPC, for which a yield of 70%
was reached.
For step 2, the efficiency of the coupling reaction between
the cross-linker and the protein was evaluated with BSA as
the model and N-SMP at different molar ratios (1:2, 1:5,
and 1:10 of lysine residues/N-SMP). Table 1 shows the yield
NaCNBH3 (the use of 10 equiv
Table 1. Efficiency of the two steps involved in the conjugation strategy used to prepare the immunogens.[a]
of NaCNBH3 for 30 h yielded
only a yield of 1.5% of free
thiol groups; data not shown).
A higher yield was obtained
Step 2Step 3
Global conditions
d hapten[c]
Lys/N-SMP[b]
d N-SMP[c]
Yield
[%][d]
MP–protein/
Yield
[%][f]
Lys/N-SMP/
global
yield [%][d]
[g]
hapten[e]
hapten[b]
when
a stronger, albeit less
1:214
1:5
–
–
1:10
40–47
18
–
–
22
–
–
1:3
1:8
1:16
–
–
–
nt
51–60
–
–
10
11
1266
–
55 9–33 1:5:22
stable in aqueous media, reduc-
ing agent, such as NaBH4, was
used, but the results were not
61
1:5:10
–
1:5:5
34–40
nt
31–37
63–73
satisfactory enough (only
a
[a] Conjugation studies related to the ratio of the reagents were carried out with 5-C5-S-SPC as the model
hapten and BSA; the results have been extracted from the analysis of the conjugates by MALDI-TOF mass-
spectrometric analysis. [b] Molar ratio used in the conjugation reaction with respect to the lysine residues of
yield of 14% was produced
over 5 h by using 5 equiv of
NaBH4). In contrast, tributyl- the BSA. [c] Number of residues (N-SMP or hapten) covalently attached to the BSA. [d] The degree of conju-
phosphine (Bu3P),[35] which re-
quires the presence of water as
a coreagent, allowed the quanti-
gation calculated on the basis that that BSA has 30–35 free accessible lysine moieties. [e] Molar ratio used in
the conjugation reaction with respect to the maleimido residues of the N-SMP derivatized protein. [f] The con-
jugation yield calculated on the basis of the available maleimido residues incorporated in step 2. [g] nt=not
tested.
tative reduction of the dimers
by using 1.5 equivalents over
only 2h.
When using N-SMP as the cross-linker, the formation of
the corresponding hapten-MP-protein conjugates was dem-
onstrated by matrix-assisted laser desorption ionization
time-of-flight mass spectrometry (MALDI-TOF-MS); how-
ever, when N-SBrA was employed, the phosphorous atom
of the tributylphosphine, which remained in the mixture
after the reduction step, interfered as a result of the forma-
tion of the phosphonium salt through the presence of the
bromine atom in the BrA-derivatized protein. Complete re-
moval of this reagent, prior to adding the thiol group to the
protein, required the introduction of a purification step in
which concomitant partial oxidation of the thiol moiety
could occur, thus leading again to the dimer. Hence, with
the aim of avoiding contact of the bromine atom in the pro-
tein with the phosphine group, a polystyrene resin with im-
of these conjugation reactions as measured by MALDI-TOF
mass spectrometric analysis. A 1:5 ratio was chosen since 18
maleimido residues (around the 50% of the free lysines de-
rivatized with the linker) were considered enough active
points for hapten coupling. Different maleimido residue-
s:hapten molar ratios were evaluated; however, as can be
seen in Table 1, increasing the molar ratio only produced a
slight augment of the number of hapten residues attached to
the protein. The yield of the conjugation products in this
step did not ensure free maleimido/Br epitopes, even at a
higher hapten/MP/protein molar ratio. In light of these re-
sults, we decided to proceed with the conjugation of the X-
Cz-S-SPC haptens, thus decreasing the lys/N-SMP (or N-
SBrA) molar ratio to 1:2(around 14 residues; Table 1) and
keeping a medium MP (or BrA)–protein/hapten ratio of
about 1:8. Thus, the final lys/N-SMP (or N-SBrA)/hapten
molar ratio chosen was 1:2:4. Table 2 shows the degree of
conjugation of the different X-Cz-S-SPC protein conjugates
under these conditions. It can be observed that in spite of
the decreased lys/N-SMP (or N-SBrA) ratio some MP and
BrA residues remained. A potential explanation of these re-
sults is the possible undesired reaction of the hydroxy
groups from the aqueous media with the maleimido or bro-
mide groups, although this supposition has not been proved.
Thus, we proceeded to use these conjugates to raise antibod-
ies. Employing a different cross-linker for the coating anti-
gens used in the competitive immunoassay would avoid the
interference caused by the polyclonal fraction of antibodies
against the cross-linker. Similarly, while HC conjugates were
used to raise antibodies, distinct proteins, such as OVA,
BSA, and CONA, were used for the preparation of the
mobilized di
(n-butyl)phenylphosphine (0.5 mmolgresinÀ1) was
A
evaluated as an alternative methodology for reduction, thus
using the advantages of solid-phase synthesis. The evolution
of the reaction was again followed with the Ellman test. In
this case, contrary to the methodology employed before, the
reduction had to be performed in the absence of water, oth-
erwise the rate of the reaction proceeded very slowly and
several hours were needed for complete reduction to occur,
even with an excess of reducing agent (up to 5 equiv). In
contrast, an almost complete conversion was accomplished
after 2hours using 1.25 equivalents of the reducing agent
only if anhydrous N,N-dimethylformamide (DMF) was used
as the solvent. After this time, water was added and the re-
action mixture was stirred for 1 h further. Under these con-
ditions, high yields, which ranged between 95 and 100%, of
the reduction products could be accomplished for all the
1910
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1906 – 1917

Hapten–Protein Bioconjugates as Immunoreagents
FULL PAPER
Table 2. Hapten density and degree of conjugation of the type-A protein
conjugates.[a]
molar mixture of the BSA conjugates (SPCmix-AB–BSA). The
type-AB antisera obtained were named As118, As119, and
As120.
Hapten
N-SMP[b]
N-SBrA[b]
d Resi-
Conjugation
d Resi-
Conjugation
dues[c]
[%][e]
dues[c]
[%][e]
Antibody characterization: Antibody characterization was
accomplished through the development of competitive
enzyme-linked immunosorbent assays for each type of anti-
sera raised. With this aim, the avidity of the antisera raised
versus all type-A and -B bioconjugates was tested through
noncompetitive indirect ELISAs. As expected, homologous
immunoassay combinations (type-A antibody/type-A anti-
gen or type-B antibody/type-B antigen) gave higher anti-
body titers when the antisera was raised through homolo-
gous immunization procedures. In contrast, heterologous
type-AB antibodies (As118–120) showed a higher affinity
for the type-B antigens than the type-A antigens (see the-
Supporting Information).
The ability of these antiserums to recognize the SPCmix (a
mixture of the six short-chain SPCs 2-C3-SPC, 2-C4-SPC, 2-
C5-SPC, 3-C4-SPC, 3-C5-SPC, and 3-C6-SPC) was determined
through competitive assays by screening 225 antisera/coating
antigen combinations. Heterologous immunoassay combina-
tions (type-A antibody/type-B antigen or type-B antibody/
type-A antigen) allowed better detectability to be reached
than homologous combinations (see the Supporting Infor-
mation). Antisera raised against type-A haptens provided
better detectability values than type-B and -AB antibodies.
Thus, the IC50 values recorded were 17.1Æ7.1 (N=11, type-
A antibodies and type-B antigens), 166.7Æ36.3 (N=14,
type-B antibodies and type-A antigens), and 1143Æ461
(N=42, type-AB antibodies and type-B antigens) mgLÀ1.
One-way analysis of variance (ANOVA) and t-tests provid-
ed p values below 0.0001, thus demonstrating the signifi-
cance of this result. Table 3 shows the features for the best
indirect competitive ELISAs obtained for each family of an-
tibody in terms of detectability and reproducibility.
N-SMP or N-SBrA
2-C3-S-SPC
2-C4-S-SPC
2-C5-S-SPC
3-C4-S-SPC
3-C5-S-SPC
3-C6-S-SPC
5-C5-S-SPC
15[d]
8
6
3
7
43–50
23–27
17–20
9–10
20–23
17–20
11–13
nc
14[d]
6
5
8
6
4
nc
7
40–47
17–20
14–17
23–27
17–20
11–13
nc
6
4
nc[f]
20–23
[a] The molar ratio between lysine residues, the heterobifunctional
linker, and free thiol groups was 1:2:4; the data shown correspond to the
BSA derivatives of each hapten. [b] N-SMP was the cross-linker used to
prepare the immunogens, while N-SBrA was the cross-linker used to pre-
pare the rest of the antigens. [c] Estimated number of residues (cross-
linker or hapten) covalently bound for each molecule of protein, as deter-
mined by MALDI-TOF mass-spectrometric analysis. [d] Estimated
number of cross-linker residues covalently bound to each molecule of
protein determined after step 2. [e] The degree of conjugation calculated
on the basis that BSA has 30–35 free accessible lysine residues. [f] nc=
not conjugated.
other immunoreagents used in the enzyme-linked immuno-
sorbent assay (ELISA) procedures.
Bioconjugates with type-B haptens could easily be pre-
pared using the water-soluble carbodiimide 1-(3-dimethyla-
minopropyl)-3-ethylcarbodiimide (EDC). The high polarity
of the SPCs allowed the conjugation to be performed in
buffer without the need for an organic co-solvent, thus ach-
ieving a high degree of conjugation. A hapten density within
16–18 mol of hapten per mol of protein was accomplished
for all the cases, except 9-C9-SPC which had 5 mol of hapt-
ens per mol of protein.
Antibodies against SPCs: Previous attempts to use a single
SPC hapten congener failed to produce generic antibodies
for the short-alkyl-chain SPC family. For this reason, we de-
cided to use equimolar mixtures of the prepared haptens.
Hence, white New Zealand rabbits were immunized with
type-A haptens (SPCmix-A-MP-HC) and type-B haptens
(SPCmix-B–HC)[23] and the evolution of the antibody titer was
followed by indirect noncompetitive experiments under ho-
mologous conditions by using the analogous mixture conju-
gated to BSA as the coating antigen (SPCmix-A-MP-BSA and
SPCmix-B–HC, respectively). The polyclonal antibodies ob-
tained were named As112, As113, and As114 (type A) and
As115, As116, and As117 (type B). Following the plan of
heterologous immunization, three more rabbits were inocu-
lated by combining SPCmix-A-MP-HC and SPCmix-B–HC
(SPCmix-AB–HC). Although in most of the examples reported
heterologous-immunization protocols are performed by al-
ternating two immunogens on each boosting injection (i.e.,
AABAAB or ABABAB, and so forth), we used an equimo-
lar combination of both mixtures of immunogens on each
boosting injection (pseudo-heterologous). Similarly, the evo-
lution of the titer was followed with the homologous equi-
Assays As112/2-C5–OVA, As115/2-C5-S-SPC-CH2-OVA, and
As119/2-C3–BSA were further evaluated to determine the
detectability and specificity of the antisera raised through
homologous (type A and B) and pseudo-heterologous (type
AB) immunization approaches, respectively (see the Sup-
porting Information). Attending to these studies, three
ELISA protocols were established. Figure 1 and Table 4
show, respectively, the standard curves and the immunoassay
parameters of the three indirect ELISAs obtained by using
each group of antibodies (open symbols in the graph).
As112in combination with 2-C 5–OVA as the coating antigen
allowed detection of the short-alkyl-chain SPCmix com-
pounds with
a limit of detection (LOD) of 0.46 nm
(0.11 mgLÀ1, based on an average of molecular weight of
250) and an IC50 value of 3.05 nm (0.76 mgLÀ1).
Antibody specificity was evaluated to find out the effect
of the immunization strategies on the capability of the anti-
bodies to recognize the different SPCs congeners formed
during the degradation process. Hence, individual recogni-
tion of these SPCs and other related compounds was as-
Chem. Eur. J. 2008, 14, 1906 – 1917
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1911

M.-P. Marco et al.
Table 3. Features of the best competitive immunoassays obtained for the three families of antisera raised against the three types of SPC immunizing hap-
tens.[a]
As
Coating antigen
CA
[mgmLÀ1
As[c]
Amax
Amin
IC50
[mgLÀ1
]
Slope
R2
[b]
[d]
G
]
N
Type A (As112)[e]
2-C5-OVA
3-C4–OVA
3-C5–OVA
3-C6–BSA
0.019
0.039
0.019
0.039
1:500
1:500
1:500
1:500
0.865Æ0.126
0.755Æ0.129
0.779Æ0.154
0.688Æ0.186
0.109Æ0.046
0.095Æ0.027
0.099Æ0.024
0.041Æ0.003
0.70Æ0.05
0.52Æ0.10
0.95Æ0.17
2.08Æ0.27
À0.99Æ0.120.997
À1.13Æ0.20
Æ0.002
0.995Æ0.002
0.965Æ0.021
0.966Æ0.016
À1.05Æ0.17
À1.49Æ0.48
Type B (As115)[e]
2-C5-S-BSA
2-C5-S-OVA
3-C5-S-CONA
0.63
1.25
1.25
1:4000
1:2000
1:1000
1.012Æ0.4320.032
Æ0.034
0.035Æ0.025
0.044Æ0.037
42.4Æ20.4
35.6Æ10.9
31.7Æ9.9
À1.27Æ0.21
À1.42Æ0.26
À1.28Æ0.25
0.958Æ0.026
0.960Æ0.017
0.951Æ0.035
0.795Æ0.155
0.881Æ0.280
Type AB (As119)[e]
2-C3–BSA
9-C9–BSA
0.3121:4000
0.3121:1000
0.640
0.851
Æ0.243
Æ0.137
0.028Æ0.036
0.047Æ0.047
29.8Æ12.5
33.1Æ13.7
À0.82Æ0.19
À1.05Æ0.30
0.961Æ0.055
0.943Æ0.068
[a] The parameters were extracted from the four-parameter equation used to fit the standard curves; the data presented correspond to the average of
five calibration curves run on 5 different days; each curve was built using two-well replicates. [b] Coating antigen. [c] Dilution factor of the antisera.
[d] SPCmix composed of an equimolar mixture of the six short-alkyl-chain SPCs was used as the standard analyte; an average molecular weight of 250
was used to calculate the IC50 value. [e] As selected for each type of immunization protocol.
sessed for the three types of antibody. The antisera raised
through homologous or pseudo-heterologous immunization
procedures showed a clearly different recognition pattern.
Thus, antibodies obtained through pseudo-heterologous pro-
cedures showed a wider recognition pattern (see Table 5).
Whereas the long-alkyl-chain SPCs tested (9-C9-SPC and
12-C12-SPC) and LASs were not recognized at all by the
As112/2C5–OVA (type-A antibodies) and As115/2-C5-S-
SPC-CH2-OVA (type-B antibodies) immunoassays, these
substances cross-reacted significantly (45, 36, and 15%, re-
spectively) in the As119/2-C3–BSA (type-AB pseudo-heter-
ologous antibodies) assay. Regarding individual recognition
Figure 1. Standard calibration curves obtained with the antibodies pro-
of the short-alkyl-chain SPCs, type-A and -B antibodies
duced through homologous (type A and B) and pseudo-heterologous
showed almost the same recognition pattern (3-C6-SPC@3-
strategies (type AB). Open symbols indicate that an equimolar mixture
C5-SPC>2-C5-SPC>3-C4-SPC>2-C4-SPCꢀ2-C3-SPC) with
of the six SPCs was used as the standard analyte. Solid symbols indicate
a superior recognition of the congeners with longer alkyl
chains, such as 3-C6-SPC (more than 300%) and 3-C5-SPC
(around 100%), while the shorter-alkyl-chain SPCs 2-C3-
SPC and 2-C4-SPC were barely
that 3-C6-SPC was used as the analyte. The curves shown correspond to
assays performed over different days. See Table 4 for the immunoassay
features.
recognized. In contrast, in the
case of the type-AB antibodies,
although 3-C5-SPC and 3-C6-
Table 4. Features of the optimized immunoassays.[a]
Antibody type:
As/CA combination:
Analyte:
Type A
As112/2-C5–OVA
mixture of SPCs
Type B
As115/2-C5-S-OVA
mixture of SPCs
Type AB
As119/2-C3–BSA
mixture of SPCs
SPC were also better recog-
nized (>200 and ꢀ50%, re-
spectively), the rest of the
short-chain SPCs showed signif-
icant cross-reactivity values
(30–35%). Other structurally
related environmental contami-
nants, such as naphthalene and
benzene sulfonates, were not
recognized by any of the anti-
bodies, thus indicating that the
selectivity was only addressed
versus the SPC family.
3-C6-SPC
As dilution
1:500
0.0201.0.2
1:500
1:2000
1:4000
CA [mgmLÀ1
Amax
]
5
0.31
0.898Æ0.221
0.114Æ0.041
0.76Æ0.11
0.23Æ0.06
2.97Æ053
0.11Æ0.04
À1.13Æ0.13
0.996Æ 0.003
16
0.785Æ0.131
0.079Æ0.008
0.20Æ0.04
0.06Æ0.02
0.58Æ0.04
0.03Æ0.01
À1.27Æ0.24
0.995Æ 0.003
6
0.914Æ0.272
0.023Æ0.020
23.3Æ7.5
0.907Æ0.358
0.032Æ0.028
29.7Æ13.6
3.61Æ2.30
171.72Æ92.41
0.95Æ0.92
À0.80Æ0.16
0.969Æ0.030
12
Amin
IC50 [mgLÀ1
]
dynamic
5.60Æ3.80
76.61Æ22.40
1.5Æ 1.4
range[b] [mgLÀ1
]
LOD [nm; mgLÀ1
]
slope
À1.21Æ0.39
0.956Æ0.030
15
R2
N[c]
[a] Three immunoassays have been optimized, one for each family of antibody raised (type-A, -B, and -AB an-
tibodies); the parameters are extracted from the four-parameter equation are used to fit the standard curve.
[b] Upper and lower limits of the dynamic range, thus corresponding to the IC20 and IC80 values. [c] Number of
assays used to calculate the data are presented as an average and the standard deviation; the assays were run
on different days and each curve was built using two-well replicates.
Since some researchers con-
sider X-C6-SPC to be the more
1912
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1906 – 1917

Hapten–Protein Bioconjugates as Immunoreagents
FULL PAPER
Table 5. SPC pattern of recognition observed using the three types of antibodies.[a]
pointed to a significant contri-
bution of the different sizes and
geometries of the alkyl carbox-
ylic moieties to the recognition
event. In fact, it has been
known since the early 1990s
that electrostatic and hydrogen-
bonding interactions contribute
only a small fraction (ca. 20%)
to the free energy of the recog-
Antibody type:
As/CA combination:
Compound[a]
Type A
As112/2-C5-OVA
IC50 [nm]
Type B
As115/2-C5-S-OVA
IC50 [nm]
Type AB
As119/2-C3-BSA
IC50 [nm]
CR [%]
CR [%]
CR [%]
SPCs
2-C3-SPC
2-C4-SPC
3.05Æ0.42100
7420.4
887
93.2
Æ30.0
100
<0.01
2330
118.7Æ54.2100
409
>30000
4015
29
0.3
36
2-C5-SPC
20
15
849
11
400
409
55
220
–
289
330
766
30
29
216
54
–
41
36
15
3-C4-SPC
215
1
1277
7
3-C5-SPC
2.76
110
65.7
142
3-C6-SPC
0.84
363
26.8
348
5-C5-SPC
9-C9-SPC
12-C12-SPC
LAS
NP
p-TS
EBS
p-XS
BDS
17.5
17
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
nition
event
(for
Ka =
>30000
>30000
80
<0.01
<0.01
4
ꢀ1010 mÀ1, DG= ꢀ68 kJmolÀ1),
while the so-called hydrophobic
interactions contribute to the
rest of the total energy (see
ref. [42–44] and references cited
therein), with an energy of
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
3613
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.1
>30000
>30000
2000
<0.01
<0.01
6
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
>30000
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
-2
SDS
about 100 JŠ of the contact
zone.[45] Thus, the main role of
the electrostatic and hydrogen-
bonding interactions seems to
be to keep the two molecules in
the correct orientation to allow
short-range hydrophobic inter-
actions to be established.
1-NaphSO3
1,5-NaphdSO3
1,3,5-NaphtSO3
2-phepropionic acid
2-phebutyric acid
2-phepentanoic acid
3-phebutyric acid
3-phepentanoic acid
3-phehexanoic acid
969
>30000
227
0.3
<0.01
1
57
5
Attending to these prece-
dents, we decided to investigate
the potential of the heterolo-
gous immunization strategy de-
vised by Masamune and co-
workers[27,28,30] to raise antibod-
ies with class-directed specifici-
ty against the SPC family. Con-
trary to homologous immuniza-
[a] The percentage of recognition is expressed as cross-reactivity (CR) according to the expression [IC50
A
IC50(cross-reactant)]”100; 2-C3, 2 -C, 2 -C, 3-C4, 3-C5, and 3-C6 are the SPCs used as immunogens. LAS=
4
5
linear alkylbenzenesulfonates, NP=nonylphenol, p-TS=para-toluensulfonic acid, EBS=para-ethyl benzene-
sulfonic acid, p-XS=para-xilenesulfonic acid, BDS=1,3-benzenedisulfonic acid, SDS=sodium docecyl sulfate,
1-naphSO3 =1-naphthalenesulfonate, 1,5-naphdSO3 =1,5-naphthalenedisulfonate, 1,3,5-naphtSO3 =1,3,6-naph-
thalenetrisulfonate, 2-Phepropionic acid=2-phenylpropionic acid, 2-Phebutyric acid=2-phenylbutyric acid, 2-
Phepentanoic acid=2-phenylpentanoic acid, 3-Phebutyric acid=3-phenylbutyric acid, 3-Phepentanoic acid=
3-phenylpentanoic acid, 3-Phehexanoic acid=3-phenylhexanoic acid.
abundant short-alkyl-chain SPC found in the environment,
assay As112/2C5–OVA was characterized for 3-C6-SPC as a
potential indicator of the contamination of water samples
with SPCs. 3-C6-SPC could be detected with a LOD of
0.13 nm (35 ngLÀ1) and an IC50 value of 0.67 nm (181 ngLÀ1).
The calibration curve is shown in Figure 1 (solid symbol).
tion procedures, in which only one type of immunizing
hapten is used in each boosting injection, this strategy con-
sists of successively immunizing the animal with two differ-
ent but structurally related haptens. This approach has
mainly been used in the catalytic antibody field for zwitter-
ionic transition-state analogues with positive and negative
charges, in which the chemical synthesis of the correspond-
ing hapten could be problematic. Instead, heterologous im-
munization with two individual haptens, containing a differ-
ent charge, provides the opportunity to simultaneously gen-
erate acidic and basic catalytic residues in the antibody-com-
bining site. Thus, Ersoy et al.[46,47] designed three haptens to
produce by heterelogous immunization nucleophile-mediat-
ed (phenol) amide bond-cleaving catalytic antibodies with a
binding pocket with 1) a hydrophobic area, 2) an acidic resi-
due complementary to the oxyanionic transition state, and
3) a basic residue to aid deprotonation of a phenol nucleo-
phile and protonation of the departing amine. These exam-
ples point to the possibility of modulating the selectivity and
affinity of the immunoresponse by combining immunoconju-
gates prepared by using two or more haptens without the
need to invest time on complex synthetic procedures to pre-
serve all the epitopes on the same hapten.
Discussion
Generic recognition of a chemical family of substances using
analytical methods based on biomolecular interactions is an
uncertain goal. The possibility of tailor-made bioreceptors
has been a matter of investigation mainly by using theoreti-
cal chemistry tools or molecular bioengineering methods.
Thus, immunochemical methods have often been exploited
for their capability to direct antibody specificity through ap-
propriate hapten design by using theoretical methods and
models[20,36–39] or recombinant antibody methodologies.[40,41]
However, as reported previously, addressing antibody recog-
nition towards the common sulfonic group of the SPCs
failed to produce generic antibodies.[23] The great differences
in the antibody recognition of the different SPC congeners
Chem. Eur. J. 2008, 14, 1906 – 1917
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1913

M.-P. Marco et al.
Hence, from a synthetic point of view, the preparation of
SPC type-A and -B haptens was easier than trying to syn-
thesize a single hapten that preserved both epitopes. The in-
troduction of a linker, as in the case of type-A haptens, was
easily possible using the corresponding para-chlorosulfonic
acid derivatives. On the other hand, type-B haptens were
conjugated through the carboxylic group of the target mole-
cule. The only challenge was to select a bifunctional linker
able to react with the chlorosulfonic group while introducing
a chemical functionality that did not interfere with this reac-
tion and was orthogonal to the functionality of the carboxyl-
ic group of the SPC. The thiol group complied with these re-
quirements, although it could spontaneously oxidize into the
disulfide form. For this reason, we envisaged the synthesis
of type-A haptens as disulfide symmetric dimmers to keep
this possibility under control. The formation of the symmet-
ric dimers (2-C3-S-SPC, 2-C4-S-SPC, 2-C5-S-SPC, 3-C4-S-
SPC, 3-C5-S-SPC, 3-C6-S-SPC, and 5-C5-S-SPC) was ach-
ieved in acceptable yields following the synthetic strategy
shown in Scheme 3. The synthesis of haptens as dimers, in-
stead of protecting the thiol group, proved to be a good
strategy to prepare hapten bioconjugates. The formation of
a dimer allows the oxidation of the thiol group to be kept
under control. Furthermore, release of the thiol group can
take place under very mild conditions by using a DBPP
polystyrene resin immediately before the bioconjugation,
without needing to introduce a purification step. Although
reduction did also occur in a very high yield using tributyl-
phosphine in aqueous media, the use of the resin avoided
the potential interference of the reducing agent in the conju-
gation step with the derivatized protein (BrA–protein). A
lys/N-SMP (or N-SBrA)/hapten molar ratio of 1:2:4 pro-
duced the X-Cz-S-SPC protein conjugates with 3–8 hapten
residues, although some MP and BrA residues remained.
The strategy of synthesizing disulfide symmetric dimers as
pre-haptens to further derivatize surfaces or biomolecules,
as in described herein, can be of general applicability, espe-
cially if highly functionalized small organic molecules need
to be coupled. Moreover, thiol groups can frequently be the
functionality of choice for a biosensor or in nanobiotechno-
logical applications that involve the use of silver or gold flat
or nanostrutured surfaces.
with some limitations, pseudo-heterologous immunization
procedures may offer some advantages, thus overcoming the
synthetic difficulties derived from preparing a complex
hapten and preserving all the most important antigenic epi-
topes of a molecule. Moreover, it could be the right choice
to produce generic (class-selective) antibodies, although the
general applicability of this hypothesis should be proved
with other antibodies against different target analytes.
Finally, the excellent features of the immunochemical
method developed using As112/2-C5-OVA, with a LOD of
35 ngLÀ1 and an IC50 value of 181 ngLÀ1 for 3-C6-SPC, sug-
gest the potential applicability of these antibodies as screen-
ing tools to detect the contamination of water samples by
anionic surfactants using SPCs as indicators.
Experimental Section
Organic chemistry
Chemicals and instruments: Thin layer chromatography (TLC) was per-
formed on 0.25-mm precoated silica gel. Unless otherwise indicated pu-
rification of the reaction mixtures was accomplished by flash chromatog-
raphy using silica gel as the stationary phase. 1H and 13C NMR spectra
were recorded at 300 and 75 MHz or at 500 and 125 MHz, respectively,
using trimethylsilane (TMS) as the internal reference. IR spectra were
measured on a Bomen MB 120 FTIR spectrophotometer (Hartmann &
Braun, QuØbec, Canada). High-resolution mass spectrometry (HRMS)
by electronic impact (EI) was performed on a Micromass Autospec spec-
trometer (Unity of Mass Spectrometry, Universidad de Santiago de Com-
postela, Spain). The chemical reagents used for the synthesis of the hapt-
ens were purchased from Aldrich Chemical Co. (Milwaukee, WI, USA).
The six short-chain SPCs used as standards were prepared as previously
described.[23]
Synthesis of haptens
General: The six short-chain SPCs were directly used as type-B haptens.
The synthesis of type-A haptens is reported herein (see Scheme 3). The
precursors 1, 2, 4, and 7 were obtained from commercial sources, whereas
3, 5, and 6 were synthesized as previously described.[23] The spectroscopic
characterization of the intermediates 1’–7’ and 1a–7a can be found in the
Supporting Information section.
Phenyl carboxylate esters (1’–7’): The phenyl carboxylic acids 1–5 were
protected as butyl esters and phenyl carboxylic acids 6 and 7 as methyl
esters. Briefly, the acid was dissolved in BuOH (10 equiv) or MeOH
(10 equiv) in a round-bottom flask and few drops of H2SO4 were subse-
quently added. The reaction mixtures were left to stir at room tempera-
ture until the total disappearance of the starting material was observed
by TLC. The reaction mixtures were treated with a solution of saturated
NaHCO3, and the products were extracted with Et2O. The organic phases
were dried over MgSO4, filtered, and evaporated to dryness. The ob-
tained products were purified by flash chromatography on silica gel elut-
ing with a hexane/Et2O gradient to obtain them in a pure form.
Antibodies raised against SPCmix-A-MP-HC provided the
best immunochemical detectability, which indicated the
great importance of the alkanoic moieties as antigenic deter-
minants. From this point of view, pseudo-heterologous im-
munization procedures did not enhance the avidity of the
antisera for the target analytes. In contrast, this strategy pro-
duced antibodies with a much broader recognition pattern.
Thus, although there was better recognition of short-alkyl-
chain SPCs of five and six carbon atoms, those with three
and four carbon atoms were also recognized with 30–35%
cross-reactivity. Similarly, long-alkyl-chain SPCs were recog-
nized to a significant extent (45–40%). In contrast, antibod-
ies raised against only type-A or -B haptens recognized 3-
C5- and 3-C6-SPCs almost exclusively, while other short-
alkyl-chain SPCs were only barely recognized. Although
General procedure for the preparation of the chlorosulfonyl derivatives
(1a–7a): Chlorosulfonic acid (3 equiv) was added dropwise to a round-
bottom flask fitted with a trap containing a 1m NaOH solution and kept
under argon. The ester 1’–7’ (1 equiv) was then slowly added to change
the color of the solution from yellow to brown. The reaction mixture was
left to stir at room temperature until the total disappearance of the start-
ing material was observed by TLC. The crude product was poured onto
ice/water and a yellow precipitate formed, which was extracted with
hexane. The organic phase was washed with a saturated solution of
NaHCO3, dried over MgSO4, filtered, and evaporated to dryness. The ob-
tained yellow oil was then purified by flash chromatography on silica gel
eluting with a hexane/Et2O gradient to yield the corresponding 4-chloro-
sulfonylphenyl derivatives.
1914
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1906 – 1917

Hapten–Protein Bioconjugates as Immunoreagents
FULL PAPER
1159 ( SO2N st si) cmÀ1
;
HR-MS (ESI-TOF-MS): m/z calcd for
À
General procedure for the synthesis of the dimers 2-C3-S-SPC, 2-C4-S-
SPC, 2-C5-S-SPC, 3-C4-S-SPC, 3-C5-S-SPC, 3-C6-S-SPC, and 5-C5-S-SPC:
Cystamine hydrochloride (0.5 equiv) was added to a solution of the cor-
responding chlorosulfonyl phenylalkanoic esters (1a–7a; 1 equiv) in an-
hydrous CH2Cl2 (final concentration: 0.05–0.08m) in a round-bottom
flask and stirred for few minutes. Triethylamine (2.1 equiv) was added,
which initially produced an off-white suspension that rapidly became
clear. As the reaction progressed, a white precipitate corresponding to
HNEt3Cl appeared. When complete disappearance of the starting materi-
al was observed by TLC, the reaction mixture was dissolved on pouring
into water and extracted with CH2Cl2, dried over anhydrous MgSO4, fil-
tered, and evaporated to dryness. The dimers obtained were purified by
flash chromatography on silica gel eluting with a hexane/EtOAc gradient.
Subsequently, the ester compounds were hydrolyzed in MeOH with 1m
KOH at room temperature until the total disappearance of the starting
materials was observed by TLC. The MeOH was evaporated and the
crude product was washed with Et2O. The aqueous phase was acidified
with concentrated HCl to pH 5.5, and a white precipitate was formed
that was extracted with EtOAc. The organic phase was dried over
MgSO4, filtered, and evaporated to isolate the desired acids.
C26H36N2NaO8S4 [M+Na+]: 655.1252; found: 655.1242.
2,2’-Dithiobis[5-{4-(N-ethylsulfamoyl)}-3-phenyl hexanoic acid] (3-C6-S-
SPC): Obtained as 64 mg, 25% yield. 1H NMR (300 MHz, CD3OD): d=
0.87 (t, J=7 Hz, 6H), 1.17 (tq, J=7, 7 Hz, 4H), 1.66 (m, 4H), 2.58 (dd,
J=16, 8.5 Hz, 2H), 2.65 (t, J=6 Hz, 4H), 2.70 (dd, J=16, 6.5 Hz, 2H),
3.14 (t, J=6.5 Hz, 4H), 3.19 (tt, J=8.5, 6 Hz, 2H), 7.44 (d, J=8.5 Hz,
4H), 7.79 (d, J=8.5 Hz, 4H) ppm; 13C NMR (75 MHz, CD3OD): d=
14.3, 21.5, 38.6, 39.3, 42.0, 43.0, 43.1, 128.2, 129.6, 139.7, 151.2, 175.7 ppm;
À
IR (KBr): n˜ =3526 (OH st), 3271 (NH st), 2963–2873 (C H st), 1710 (C=
O st), 1598 (ArC C), 1324 ( SO2N st as), 1160 ( SO2N st si) cmÀ1; HR-
MS (ESI-TOF-MS): m/z calcd for C28H40N2NaO8S4 [M+Na+]: 683.1565;
found: 683.1585.
À
À
À
2,2’-Dithiobis[5-{4-(N-ethylsulfamoyl)}phenyl pentanoic acid] (5-C5-S-
SPC): Obtained as 102mg, 37%yield. 1H NMR (300 MHz, CD3OD): d=
1.67 (m, 8H), 2.33 (t, J=7 Hz, 4H), 2.64 (t, J=7 Hz, 4H), 2.73 (t, J=
7 Hz, 2H), 3.12 (t, J=6.5 Hz, 4H), 7.40 (d, J=8.5 Hz, 4H), 7.76 (d, J=
8.5 Hz, 4H) ppm; 13C NMR (75 MHz, CD3OD): d=25.3, 31.7, 34.7, 36.3,
38.7, 43.1, 128.2, 130.3, 139.2, 149.2, 179.7 ppm; IR (methyl ester) (KBr):
À
À
n˜ =3280 (-NH- st), 2948–2864 (C H st), 1731 (C=O st), 1598 (ArC C),
1328 ( SO2N st as), 1159 ( SO2N st si) cmÀ1; HR-MS (ESI-TOF-MS):
2,2’-Dithiobis[5-{4-(N-ethylsulfamoyl)}-2-phenyl propanoic acid] (2-C3-S-
SPC): Obtained in 40 mg, 70% yield. 1H NMR (300 MHz, CD3OD): d=
1.49 (d, J=7 Hz, 6H), 2.65 (t, J=7 Hz, 4H), 3.13 (t, J=6.5 Hz, 4H), 3.84
(q, J=7 Hz, 2H), 7.53 (d, J=8 Hz, 4H), 7.81 (d, J=8 Hz, 4H) ppm;
13C NMR (75 MHz, CD3OD): d=18.5, 38.7, 43.1, 46.5, 128.3, 129.6, 140.5,
À
À
m/z calcd for C26H37N2O8S4 [M+H+]: 633.1433; found: 633.1415.
Immunochemistry
Chemicals and immunochemicals: Di(n-butyl)phenylphosphine polystyr-
G
ene resin (DBPP; with a derivatization of 0.5 mmolgresinÀ1) was purchased
from NovaBiochem (Läufelfingen, Switzerland). Specific immunore-
agents (antibodies and protein and enzyme conjugates) were prepared as
described below. The SPC standards and the phenyl carboxylic acids used
in the cross-reactivity studies were synthesized as described previously.[23]
LAS standards were a gift from PETRESA (Cµdiz, Spain). The heterobi-
functional cross-linkers N-succinimidyl-3-maleimidyl propanoate (N-
SMP) and N-succinimidyl bromoacetate (N-SBrA) used to prepare the
protein conjugates were synthesized following the procedures described
by Nielsen and Buchardt[48] and Bernatowicz and Matsueda,[49] respec-
tively. Short-alkyl-chain SPCs and the corresponding immunoreagents
were used as mixtures in some experiments. Thus, screening experiments
to find out the best immunoassay conditions were carried out using
SPCmix as an equimolar mixture of 2-C3-SPC, 2-C4-SPC, 2-C5-SPC, 3-C4-
SPC, 3-C5-SPC, and 3-C6-SPC. Immunization protocols were performed
with mixtures of hapten–protein conjugates. The term SPCmix-A-MP (or
-CH2)-protein designates an equimolar mixture of six immunogens pre-
pared by covalent attachment of the type-A haptens (2-C3-S-SPC, 2-C4-S-
SPC, 2-C5-S-SPC, 3-C4-S-SPC, 3-C5-S-SPC, and 3-C6-S-SPC) to the pro-
tein, through the corresponding N-SMP or N-SBrA cross linkers, respec-
tively. Similarly, the term SPCmix-B–protein defines an equimolar mixture
of six immunogens prepared by covalent attachment of type-B haptens
(2-C3-SPC, 2-C4-SPC, 2-C5-SPC, 3-C4-SPC, 3-C5-SPC, and 3-C6-SPC) to
the protein. Finally, an equimolar mixture of the twelve immunogens pre-
pared with both types of haptens have been designated as SPCmix-AB–pro-
tein. The buffers used can be found in the Supporting Information.
À
147.7, 165.4 ppm; IR (KBr): n˜ =3273 (NH and OH st), 2981–2937 (C H
À
À
À
st), 1714 (C=O st), 1598 (ArC C), 1319 ( SO2N st as), 1157 ( SO2N st
si) cmÀ1; HR-MS (ESI-TOF-MS): m/z calcd for C22H28N2NaO8S4 [M+
Na+]: 599.0626; found: 599.0634.
2,2’-Dithiobis[5-{4-(N-ethylsulfamoyl)}-2-phenyl butanoic acid] (2-C4-S-
SPC): Obtained in 337 mg, 62% yield. 1H NMR (300 MHz, CD3OD): d=
0.91 (t, J=7 Hz, 6H), 1.80 (ddq, J=17.5, 7.5, 7 Hz, 2H), 2.11 (ddq, J=
15, 7.5, 7 Hz, 2H), 2.65 (t, J=6 Hz, 4H), 3.14 (t, J=6.5 Hz, 4H), 3.58 (t,
J=8 Hz, 2H), 7.53 (d, J=8.5 Hz, 4H), 7.82(d, J=8 Hz, 4H) ppm;
13C NMR (75 MHz, CD3OD): d=12.5, 27.6, 38.7, 43.1, 54.4, 128.3, 130.0,
140.6, 146.1, 176.7 ppm; IR (KBr): n˜ =3272 (NH and OH st), 2968–2877
À
À
À
(C H st), 1706 (C=O st), 1597 (ArC C), 1322 ( SO2N st as), 1157
( SO2N st si) cmÀ1
;
HR-MS (ESI-TOF-MS): m/z calcd for
À
C24H32N2NaO8S4 [M+Na+]: 627.0939; found: 627.0935.
2,2’-Dithiobis[5-{4-(N-ethylsulfamoyl}-]2-phenyl pentanoic acid] (2-C5-S-
SPC): Obtained as 600 mg, 68% yield. H NMR (300 MHz, CD3OD): d=
1
0.92(t, J=7 Hz, 6H), 1.29 (ddq, J=14.5, 8.5, 8 Hz, 4H), 1.75 (ddt, J=
15.5, 7.5, 6 Hz, 2H), 2.04 (ddt, J=15.5, 7.5, 6 Hz, 2H), 2.65 (t, J=6 Hz,
4H), 3.14 (t, J=6.5 Hz, 4H), 3.68 (t, J=8 Hz, 2H), 7.53 (d, J=8.5 Hz,
4H), 7.82(d, J=8 Hz, 4H) ppm; 13C NMR (75 MHz, CD3OD): d=14.1,
21.7, 36.6, 38.6, 43.0, 52.4, 128.2, 130.0, 140.4, 146.2, 176.7 ppm; IR (KBr):
À
n˜ =3274 (NH and OH st), 2960–2873 (C H st), 1708 (C=O st), 1596
(ArC C), 1322 ( SO2N st as), 1158 ( SO2N st si) cmÀ1; HR-MS (ESI-
TOF-MS): m/z calcd for C26H37N2O8S4 [M+H+]: 633.1433; found:
633.1444.
À
À
À
2,2’-Dithiobis[5-[4-(N-ethylsulfamoyl)]-3-phenyl butanoic acid] (3-C4-S-
SPC): (680 mg, 76% yield). 1H NMR (300 MHz, CD3OD): d=1.32(d,
J=7 Hz, 6H), 2.63 (d, J=7.5 Hz, 4H), 2.66 (t, J=6 Hz, 4H), 3.13 (t, J=
6.5 Hz, 4H), 3.31 (tq, J=7 Hz, J=7 Hz, 2H), 7.48 (d, J=8 Hz, 4H), 7.78
(d, J=8 Hz, 4H) ppm; 13C NMR (75 MHz, CD3OD): d=22.4, 37.6, 38.7,
43.0, 43.5, 128.3, 128.9, 139.7, 152.7, 175.7 ppm; IR (KBr): n˜ =3276 (NH
Preparation of the immunoreagents using type-A haptens: Haptens 2-C3-
S-SPC, 2-C4-S-SPC, 2-C5-S-SPC, 3-C4-S-SPC, 3-C5-S-SPC, and 3-C6-S-SPC
were conjugated to HC, BSA, CONA, and OVA with N-SMP or N-SBrA
through a three-step procedure (see Scheme 4). All the bioconjugates
(hapten-MP-protein or hapten-CH2-protein and MP–protein or BrA–pro-
tein) were purified by dialysis against 0.5 mm phosphate-buffered saline
(PBS; 4”5 L) and MilliQ water (1”5 L) and stored freeze dried at
À408C. Stock solutions of 1 mgmLÀ1 were prepared in PBS buffer and
stored in aliquots at À408C. Working aliquots were stored at 48C in
À
À
and OH st), 2968–2873 (C H st), 1708 (C=O st), 1603 (ArC C), 1322
( SO2N st as), 1157 ( SO2N st si) cmÀ1; HR-MS (ESI-TOF-MS): m/z
calcd for C24H32N2NaO8S4 [M+Na+]: 627.0939; found: 627.0933.
À
À
10 mm PBS at 1 mgmLÀ1
.
2,2’-Dithiobis[5-{4-(N-ethylsulfamoyl)}-3-phenyl pentanoic acid] (3-C5-S-
1
SPC): Obtained as 630 mg, 67% yield. H NMR (300 MHz, CD3OD): d=
Step 1: Reductionof the dimer to generate the thiol group : The DBPP
resin (25 mg, 12.5 mmol of reducing agent, 1.25 equiv) was added to a so-
lution of the dimers (10 mmol) in anhydrous DMF (160 mL) and kept
under argon. The reaction mixture was gently stirred at room tempera-
ture for 2h. MilliQ water (40 mL) was then added to the suspension,
which was stirred for 1 h more until complete reduction of the dimer was
observed using the Ellman test.[34] The suspension was then filtered
through a syringe provided with a filter (0.45-mm porous size). The re-
0.78 (t, J=7 Hz, 6H), 1.63 (ddq, J=13.5, 7.5, 6.5 Hz, 2H), 1.77 (ddq, J=
13.5, 7.5, 6.5 Hz, 2H), 2.60 (dd, J=15.5, 9 Hz, 2H), 2.66 (t, J=6 Hz, 4H),
2.72 (dd, J=15.5, 6.5 Hz, 2H), 3.09 (tt, J=9, 6 Hz, 2H), 3.15 (t, J=
6.5 Hz, 4H), 7.44 (d, J=8.5 Hz, 4H), 7.80 (d, J=8.5 Hz, 4H) ppm;
13C NMR (75 MHz, CD3OD): d=12.2, 30.0, 38.7, 41.5, 42.9, 44.9, 128.1,
129.6, 139.5, 150.9, 175.8 ppm; IR (KBr): n˜ =3525 (OH st), 3270 (NH st),
À
À
À
2963–2875 (C H st), 1710 (C=O st), 1598 (ArC C), 1324 ( SO2N st as),
Chem. Eur. J. 2008, 14, 1906 – 1917
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1915

M.-P. Marco et al.
tained resin was washed with water (2”120 mL). Finally, the collected
fractions were combined and used immediately for conjugation (step 3).
imum absorbance, C is the concentration producing 50% of the maximal
absorbance, and D is the slope at the inflection point of the sigmoid
curve.
Step 2: Activationof the protein : Simultaneously to step 1, the cross-
linkers N-SMP (or N-SBrA, 20 mmol) in anhydrous DMF (100 mL) were
added dropwise to a solution of the protein (HC, BSA, OVA, or CONA;
20 mg) in PBS (1.5 mL). The reaction mixtures were stirred at room tem-
perature for 2h and then purified by passing the solution through a Se-
phadex G-25 desalting column and eluting with PBS buffer. The fractions
containing the protein conjugate (MP–protein or BrA–protein, depend-
ing on the cross-linker used) were collected, combined, and then split in
two equal volumes. One of them was used for conjugation with the
hapten and the other one was stored separately for further inspection of
this reaction step by MALDI-TOF-MS.
Optimized indirect ELISAs: Microtiter plate assays were established for
each type of antibody using the best As/coating antigen (CA) combina-
tions. Thus, for the antibodies raised against SPCmix-A-MP-HC, As112
(1:500) was used on plates coated with 2-C5-OVA (0.02 mgmLÀ1). For the
case of the antibodies prepared using SPCmix-B–HC, As115 (1:2000) was
used in combination with 2-C5-S-OVA (1.25 mgmLÀ1). Finally, for the an-
tibodies raised using the whole mixture SPCmix-AB–HC, the As119
(1:4000) combined with 2-C3–BSA (0.31 mgmLÀ1) gave the best results.
The microplates were processed as described in the general protocol. For
the combination As112/2-C5-OVA, PBST used at the competitive step
contained 1% BSA. The SPCmix was used as the standard analyte. In
some experiments performed using ELISA with As112/2-C5-OVA, only
3-C6-SPC was used as the standard analyte.
Step 3: Coupling of the hapten to the protein: The solutions of the re-
duced haptens (ca. 20 mmol) obtained in step 1 were added to the derivat-
ized proteins (10 mg) prepared in step 2. The reaction mixtures were
stirred for 2h at room temperature to obtain the corresponding hapten-
MP-protein or hapten-CH2-protein conjugates depending on the cross-
linker used.
Specificity studies: Stock solutions of each SPC and several structurally
related compounds were prepared in H2O or dimethyl sulfoxide
(DMSO)/MeOH (6 mm) and stored at 48C. Standard curves were pre-
pared in PBST by serial dilution (30 mm–64 pm). Each IC50 value was de-
termined in the competitive experiments following the optimized proto-
col. The cross-reactivity values were calculated according to the equation
Preparation of the immunoreagents using type-B haptens: Haptens 2-C3-
SPC, 2-C4-SPC, 2-C5-SPC, 3-C4-SPC, 3-C5-SPC, 3-C6-SPC, 5-C5-SPC, and
9-C9-SPC were coupled to OVA through the carboxylic group by using
the active ester method following a standard methodology[50] by activat-
ing the haptens (10 mmol) with N,N’-dicyclohexylcarbodiimide (DCC;
50 mmol) and N-hydroxysuccinimide ester (NHS; 25 mmol) in anhydrous
[IC50
(SPC)/IC50(cross-reactant)]”100.[53]
A
DMF (200 mL) and adding the protein (10 mg) in
a borate buffer
(1.8 mL). A second batch of type-B BSA conjugates was prepared using
EDC[51] (see the Supporting Information).
Acknowledgements
Hapten density analysis: The hapten densities were calculated by
MALDI-TOF mass spectrometric analysis by determining the molecular
weight (see the Supporting Information).
This study was supported by the Ministry of Science and Technology
(contract numbers: AGL2004–20341-E and NAN2004–09415-C05–02)
and by the European Community (NMP2-CT2003–505485). We thank
PETRESA for providing samples of the technical LASs. The AMR
group is a Grup de Recerca de la Generalitat de Catalunya and receives
support from the Departament d’Universitats, Recerca i Societat de la In-
formació la Generalitat de Catalunya (expedient 2005SGR 00207).
Polyclonal antisera: As112, As113, and As114 (type-A antibodies) were
obtained by immunizing female white New Zealand rabbits weighting 1–
2kg with SPC mix-A-MP-HC following an already described protocol.[52]
The preparation of As115, As 116, and As117 (type-B antibodies) using
SPCmix-B–HC (conjugated by the mixed anhydride (MA) method) has
been described previously.[23] As118, As119, and As120 (type-AB anti-
bodies) were obtained in the same way but combining type-A and -B
haptens in an equimolar mixture of the twelve immunogens (SPCmix-AB
–
[1] A. OubiÇa, B. Ballesteros, P. Bou, R. Galve, J. Gascón, F. Iglesias, N.
Sanvicens and M.-P. Marco in Immunoassays for Environmental
Analysis, Vol. 21 (Ed.: D. Barceló), Elsevier, Amsterdam, The Neth-
erlands, 2000, pp. 289–340.
[2] Y. Xu, N. Yamamoto, K. D. Janda, Bioorg. Med. Chem. 2004, 12,
5247–5268.
[3] J. Emneus in Integrated Immuno Extraction Sampling and Portable
Biosensor Prototype for In-field Monitoring, EC, Environmental and
Climate Program ENV4-CT97–0476, Lund 1997–2000.
[4] P. Schçberl, Tenside Surfactants Deterg. 1989, 26, 86–94.
[5] T. P. Knepper, M. Kruse, Tenside Surfactants Deterg. 2000, 37, 41–
47.
[6] A. Marcomini, G. Pojana, A. Sfriso, J. M. Quiroga, Environ. Toxicol.
Chem. 2003, 19, 2000–2007.
[7] V. Andreu, Y. Picó, Anal. Chem. 2004, 76, 2878–2885.
[8] V. M. Leon, A. Gomez-Parra, E. Gonzalez-Mazo, Environ. Sci.
Technol. 2004, 38, 2359–2367.
HC): six of type A (SPCmix-A-MP-HC) and six of type B (SPCmix-B–HC;
MA method). The evolution of the antibody titer was assessed by meas-
uring the binding of serial dilutions of the antisera to microtiter plates
coated with an equimolar mixture of the corresponding BSA conjugates
(SPCmix-A-MP-BSA and SPCmix-AB–BSA). After an acceptable antibody
titer was observed, the animals were exsanguinated and the blood was
collected on vacutainer tubes provided with a serum separation gel. Anti-
sera were obtained by centrifugation and stored at À408C in the pres-
ence of 0.02% NaN3.
Evaluationof the antibody features : The detectability and specificity of
the antibodies raised by homologous and heterologous immunization
procedures were assessed by determining the IC50 values, the LOD val-
ues, and the selectivities of the competitive ELISAs developed for each
antibody type.
ELISA general protocol: The plates were coated with the antigens
(100 mLwellÀ1 in coating buffer) overnight at 48C and covered with adhe-
sive plate sealers. The next day, the plates were washed four times with
phosphate-buffered saline containing Tween-20 (PBST; 300 mLwellÀ1),
and solutions of the analyte (50 mLwellÀ1 in PBST; only PBST for zero
analyte) and the antisera (50 mLwellÀ1 in PBST) were added and incubat-
ed for 30 min at room temperature. The plates were washed as before,
and a solution of antiIgG–HRP (1:6000 in PBST) was added to the wells
(100 mLwellÀ1) and incubated for a further 30 min at room temperature.
The plates were washed again, and the substrate solution was added
(100 mLwellÀ1). Color development was stopped after 30 min at room
temperature with 4n H2SO4 (50 mLwellÀ1), and the absorbances were
read at 450 nm. The standard curve obtained was fitted to a four-parame-
ter logistic equation according to the following formula: Y=
[9] J. A. Perales, M. A. Manzano, D. Sales, J. M. Quiroga, Int. Biodeteri-
or. Biodegrad. 2003, 51, 187–194.
[10] M. Mosche, U. Meyer, Water Res. 2002, 36, 3253–3260.
[11] J. Vives-Vego, R. López-Amorós, T. Guindulain, M. T. Garcꢁa, J.
Comas, J. Sµnchez-Leal, J. SurfactantsDeterg. 2000, 3, 303–308.
[12] P. Eichhorn, S. V. Rodrigues, W. Baumann, T. P. Knepper, Sci. Total
Environ. 2002, 284, 123–134.
[13] P. Eichhorn, M. E. Flavier, M. L. Paje, T. P. Knepper, Sci. Total En-
viron. 2001, 269, 75–85.
[14] E. Gonzalez-Mazo, V. M. Leon, M. Saez, A. Gomez-Parra, Sci.
Total Environ. 2002, 288, 215–226.
[15] W.-H. Ding, J.-H. Lo, S.-H. Tzing, J. Chromatogr. A 1998, 818, 270–
279.
A
G
1916
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1906 – 1917

Hapten–Protein Bioconjugates as Immunoreagents
FULL PAPER
[16] P. Eichhorn, T. P. Knepper, F. Ventura, A. Diaz, Water Res. 2002, 36,
2179–2186.
[17] E. Gonzalez-Mazo, M. Honing, D. Barcelo, A. Gomez-Parra, Envi-
ron. Sci. Technol. 1997, 31, 504–510.
[18] C. A. Spinks, G. M. Wyatt, H. A. Lee, M. R. A. Morgan, Bioconju-
gate Chem. 1999, 10, 583–588.
[19] P. Cliquet, E. Cox, W. Haasnoot, E. Schacht, B. M. Goddeeris, J.
Agric. Food Chem. 2003, 51, 5835–5842.
[20] C. A. Spinks, G. M. Wyatt, S. Everest, R. Jackman, M. R. A.
Morgan, J. Sci. Food Agric. 2002, 82, 428–434.
[21] P. Degelmann, J. Wenger, R. Niessner, D. Knopp, Environ. Sci.
Technol. 2004, 38, 6795–6802.
[35] J. T. Ayers, S. R. Anderson, Synth. Commun. 1999, 29, 351–358.
[36] B. Ballesteros, D. Barcelo, F. Sanchez-Baeza, F. Camps, M. P. Marco,
Anal. Chem. 1998, 70, 4004–4014.
[37] R. Galve, F. Camps, F. Sanchez-Baeza, M. P. Marco, Anal. Chem.
2000, 72, 2237–2246.
[38] N. Sanvicens, F. Sanchez-Baeza, M. P. Marco, J. Agric. Food Chem.
2003, 51, 3924–3931.
[39] M. C. Estevez, M. Kreuzer, F. Sanchez-Baeza, M. P. Marco, Environ.
Sci. Technol. 2006, 40, 559–568.
[40] J. Brichta, M. Hnilova, T. Viskovic, Vet. Med. 2005, 50, 231–252.
[41] W. Kusharyoto, J. Pleiss, T. T. Bachmann, R. D. Schmid, Environ.
Prot. Eng. 2002, 15, 233–241.
[22] V. Kolar, A. P. Deng, M. Franek, Food Agric. Immunol. 2002, 14,
91–105.
[42] V. G. Omelyaneko, W. Jiskoot, J. N. Herron, Biochemistry 1993, 32,
10423–10429.
[23] M.-C. Estevez, R. Galve, F. Sanchez-Baeza, M.-P. Marco, Anal.
Chem. 2005, 77, 5283–5293.
[24] T. Korpimaki, V. Hagren, E. C. Brockmann, M. Tuomola, Anal.
Chem. 2004, 76, 3091–3098.
[25] T. Korpimaki, J. Rosenberg, P. Virtanen, T. Karskela, U. Lammin-
maki, M. Tuomola, M. Vehniainen, P. Saviranta, J. Agric. Food
Chem. 2002, 50, 4194–4201.
[43] C. J. van Oss, Mol. Immunol. 1995, 32, 199–211.
[44] J. N. Herron, D. M. Kranz, D. M. Jameson, E. W. J. Voss, Biochem-
istry 1986, 25, 4602–4609.
[45] G. Zeder-Lutz, N. Rauffer, D. Altschuh, M. H. V. Van Regenmortel,
J. Immunol. Methods 1995, 189, 131–140.
[46] O. Ersoy, R. Fleck, M. J. Blanco, S. Masamune, Bioorg. Med. Chem.
1999, 7, 279–286.
[26] T. Korpimaki, J. Rosenberg, P. Virtanen, U. Lamminmaki, M. Tuo-
mola, P. Saviranta, Protein Eng. 2003, 16, 37–46.
[27] T. Tsumuraya, H. Suga, S. Meguro, A. Tsunakawa, S. Masamune, J.
Am. Chem. Soc. 1995, 117, 11390–11396.
[28] H. Suga, O. Ersoy, S. F. Williams, T. Tsumuraya, M. N. Margolies,
A. J. Sinskey, S. Masamune, J. Am. Chem. Soc. 1994, 116, 6025–
6026.
[29] O. Ersoy, R. Fleck, A. Sinskey, S. Masamune, J. Am. Chem. Soc.
1996, 118, 13077–13078.
[30] T. Tsumuraya, N. Takazawa, A. Tsunakawa, R. Fleck, S. Masamune,
Chem. Eur. J. 2001, 7, 3748–3755.
[47] O. Ersoy, R. Fleck, A. Sinskey, S. Masamune, J. Am. Chem. Soc.
1998, 120, 817–818.
[48] O. Nielsen, O. Buchardt, Synthesis 1991, 819–821.
[49] M. S. Bernatowicz, G. R. Matsueda, Anal. Biochem. 1986, 155, 95–
102.
[50] J. Gascón, A. OubiÇa, B. Ballesteros, D. Barceló, F. Camps, M.-P.
Marco, M.-A. Gonzµlez-Martinez, S. Morais, R. Puchades, A. Ma-
quiera, Anal. Chim. Acta 1997, 347, 149–162.
[51] C. Wittmann, B. Hock, J. Agric. Food Chem. 1991, 39, 1194–1200.
[52] B. Ballesteros, D. Barceló, F. Camps, M.-P. Marco, Anal. Chim. Acta
1997, 347, 139–147.
[31] T. W. Green, P. G. M. Wuts, Protective Groupsin Organic Synthesis
Wiley, Canada, 1999.
[32] S. S. Wong, Chemistry of Protein Conjugation and Cross-linking,
CRC Press, Boca Raton, Fl, 1993.
[33] H. T. Clarke, G. S. Babcock, T. F. Murray, Organic Synthesis, Vol. I,
,
[53] Spectroscopic data (1H and 13C NMR, IR, and high-resolution mass
spectra) of most of the intermediates obtained during the SPCs syn-
thesis, results from the screening of the different antisera raised re-
garding the recognition of SPCmix, and the immunochemical perfor-
mance of the antibodies under different physicochemical conditions
are given in the Supporting Information.
1932.
[34] P. W. Riddles, R. L. Blakeley, B. Zerner, Anal. Biochem. 1979, 94,
Received: August 7, 2007
75–81.
Published online: December 10, 2007
Chem. Eur. J. 2008, 14, 1906 – 1917
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1917