Synthetic approach to gain insight into antigenic determinants of cephalosporins: In vitro studies of chemical structure-IgE molecular recognition relationships

Article abstract of DOI:10.1021/tx100446g

Cephalosporins, after penicillins, are the most widely used antibacterial agents in infectious diseases and the cause of adverse immune reactions in the world. Whether a patient with a suspected allergy to a β-lactam can safely take a cephalosporin is often a matter of debate. However, there are no tests with enough sensitivity to detect allergy to cephalosporins. Understanding the way in which the drug metabolizes after protein conjugation is important if we are to make advances in the diagnosis of clinical allergy. Structural studies of cephalosporin-protein adducts have never been addressed successfully and are difficult to investigate. Our approach to determine the requirements involved in antigenic determinant structures consisted of designing and synthesizing a proposed skeleton that remains linked to the carrier protein after chemical degradation in cephalosporin conjugated to carrier proteins. In this study, a series of proposed epitopes were efficiently synthesized following a versatile methodology, involving the condensation of the R¹ acyl side chains of native cephalosporins, with the nuclear fragment structures (derived from amino acids or other aminofunctionalized molecules). The final well-defined structures 1-4 (a-f), representing a fragment from the proposed cephalosporin-Lys(protein) adduct intermediate, consist of closely related low-molecular-weight molecules, differing only in the functional group at C-3 and the R¹ side chains. They were assessed with sera from patients allergic to cephalosporins to study structure-IgE molecular recognition relationships. These IgE showed an enhanced recognition to proposed new skeleton epitopes with adequate functionality at C-3, with the specifities mainly related to the R¹ acyl side chain. Thus, this study led us to refine the model haptenic structures of cephalosporins and gain insight into the chemical mechanism of epitope formation.

Full text of DOI:10.1021/tx100446g

ARTICLE
pubs.acs.org/crt
Synthetic Approach to Gain Insight into Antigenic Determinants of
Cephalosporins: In Vitro Studies of Chemical StructureꢀIgE Molecular
Recognition Relationships
Maria Isabel Monta~nez,†,‡ Cristobalina Mayorga,‡ Maria Jose Torres,§ Adriana Ariza,‡ Miguel Blanca,‡,§ and
Ezequiel Perez-Inestrosa*,†
†Department of Organic Chemistry, Faculty of Science, University of Malaga, 29071 Malaga, Spain
‡Research Laboratory, IMABIS Foundation-Carlos Haya Hospital, 29009 Malaga, Spain
§Allergy Service, Carlos Haya Hospital, 29009 Malaga, Spain
S Supporting Information
b
ABSTRACT: Cephalosporins, after penicillins, are the most
widely used antibacterial agents in infectious diseases and the
cause of adverse immune reactions in the world. Whether a
patient with a suspected allergy to a β-lactam can safely take a
cephalosporin is often a matter of debate. However, there are no
tests with enough sensitivity to detect allergy to cephalosporins.
Understanding the way in which the drug metabolizes after
protein conjugation is important if we are to make advances in
the diagnosis of clinical allergy. Structural studies of cephalos-
porinꢀprotein adducts have never been addressed successfully and are difficult to investigate. Our approach to determine the
requirements involved in antigenic determinant structures consisted of designing and synthesizing a proposed skeleton that remains
linked to the carrier protein after chemical degradation in cephalosporin conjugated to carrier proteins. In this study, a series of
proposed epitopes were efficiently synthesized following a versatile methodology, involving the condensation of the R1 acyl side
chains of native cephalosporins, with the nuclear fragment structures (derived from amino acids or other aminofunctionalized
molecules). The final well-defined structures 1ꢀ4 (aꢀf), representing a fragment from the proposed cephalosporinꢀLys(protein)
adduct intermediate, consist of closely related low-molecular-weight molecules, differing only in the functional group at C-3 and the
R1 side chains. They were assessed with sera from patients allergic to cephalosporins to study structureꢀIgE molecular recognition
relationships. These IgE showed an enhanced recognition to proposed new skeleton epitopes with adequate functionality at C-3,
with the specifities mainly related to the R1 acyl side chain. Thus, this study led us to refine the model haptenic structures of
cephalosporins and gain insight into the chemical mechanism of epitope formation.
’ INTRODUCTION
antibiotics and the study of molecular recognition by specific IgE
antibodies.5 As a consequence, attempts to develop in vivo and in
vitro tests to evaluate allergy to cephalosporins have been un-
successful, and skin testing including just the native drug has to
be used as the skin test agent to detect antibodies specific to these
drugs.6,7
Penicillins and cephalosporins share structural similarities;
they possess a β-lactam ring which has the antimicrobial activity
and diverse side chains (Scheme 1). However, they differ in their
immunologic behavior, determined by their intrinsic chemical
reactivity, which is related with the electrophilic properties of the
β-lactam carbonyl,8 and therefore the possibility to bind to
proteins with subsequent epitope formation. The main difference
between cephalosporins and penicillins is the ring to which the
β-lactam is fused. A five-membered thiazolidine ring in penicillins
Cephalosporins are widely prescribed antibiotics and repre-
sent, after penicillins, the most common cause of adverse drug
1,2
reactions
mediated by specific immunological mechanisms.
IgE-mediated hypersensitivity reactions to penicillins have been
extensively studied and used as a model to increase under-
standing of the immunological mechanisms involved in these
reactions. Penicillins are chemically reactive small allergenic
molecules, and according to the hypothesis of Landsteiner,3 only
when they covalently bind to protein, through reaction of the
β-lactam ring with the nucleophilic free amino groups on
proteins,4 can the so formed haptenꢀprotein conjugate induce
an immune response. The mechanism of cephalosporin-induced
allergic reactions fits the hapten hypothesis also due to its
β-lactam reactivity. However, unlike well-established structures
responsible for penicillin allergy, the metabolites involved in the
immunological responses to cephalosporins are still unknown.
This has limited adequate evaluation of allergic reactions to these
Received: December 21, 2010
Published: March 22, 2011
r
2011 American Chemical Society
706
dx.doi.org/10.1021/tx100446g Chem. Res. Toxicol. 2011, 24, 706–717
|

Chemical Research in Toxicology
ARTICLE
Scheme 1. Nucleophilic Ring-Opening of β-Lactams by
Proteins and Subsequent Formation of Antigenic
Determinantsa
Scheme 2. Proposed Aminolysis Pathway and Resulting De-
gradation Products for Cephalosporins
Whatever the case, there is clinical evidence that the R1 side chain
may contribute to IgE induction and cross-reactive responses. In
fact, some patients can react selectively to the cephalosporin
involved in the reaction, whereas others react to different
cephalosporins which have the same or a similar R1 side chain,17ꢀ19
and a third group of patients has cross-reactivity with other
β-lactams, especially penicillins, which have the same or a similar
R1.6,18,20ꢀ22 The study of allergy to cephalosporins has classically
focused on the assessment of tolerance in patients primarily
sensitized to penicillin derivatives.6,20ꢀ24 However, few studies
have analyzed the problem of cross-reactivity in subjects primar-
ily sensitized to cephalosporins and potentially allergic to
penicillins.17,18 Whether a patient with an allergic reaction to
penicillins can safely take other β-lactams such as cephalosporin
is a matter of debate and will depend on the β-lactam involved in
the reaction.17,18,25,26 However, cross-reactivity due to similari-
ties in R2 has only vaguely been described.27
Chemically, all attempts to study the reaction between cepha-
losporins with simple nitrogen nucleophiles have failed to
identify an exact chemical structure. Results with butylamine as
nucleophile suggest the cephalosporoyl conjugate formation and
further fragmentation of the dihydrothiazine ring, leading to
structures including at least the remaining β-lactam and the R1
side chain (Scheme 2).28
Taking into account both the evidence of the departure of the
R2 group as a rule and the important role of the R1 structure in
specific immunologic recognition, we proposed that after open-
ing the β-lactam ring by amino groups of proteins, the resulting
structure is unstable; thus, the dihydrothiazine ring undergoes
degradation and becomes lost. As a result, only R1 and the amino
acid residue included in the β-lactam moiety of the cephalosporin
remain linked to the protein, and this haptenꢀcarrier conjugate
is the final structure that interacts with the immunological system
(Scheme 2). Our approach to determine the structure of the
epitope consisted of synthesizing the proposed skeleton that
remains linked to the carrier protein after chemical degradation
in the cephalosporin conjugate. To carry out this synthesis,
butylamine was chosen as the nucleophilic amine model to
emulate the primary amino groups of the side chains of lysine
present in proteins. Thus, the proposed model structures, in
agreement with our hypothesis, would result from the nucleo-
philic reaction of butylamine to cephalosporins. A number of
a Unlike well-established structures responsible for allergy to penicillin
and penicilloyl, the equivalent chemical structures of cephalosporin
antigenic determinants are still unknown.
and a six-membered dihydrothiazine ring in cephalosporins pro-
vide different degradation patterns. In penicillins, the higher
tension within the β-lactam ring compared to that in cephalos-
porins results in a more increased chemical reactivity.9 Hapteni-
zation of proteins by penicillins takes place quickly and efficiently,
obtaining a conjugate, penicilloyl (PO; Scheme 1), which is
stable enough to be isolated and characterized by spectroscopic
techniques. In cephalosporins, the lower reactivity of the
β-lactam ring slows the haptenization process. The R2 chemical
structure modulates this reactivity depending on its ability to
polarize electronic binding. Most clinically relevant cephalospor-
ins have a good leaving group at the 30 position which increases
the reactivity of the β-lactam via the elimination of R2. The de-
parture of R2 has been interpreted in terms both of concerted10ꢀ14
and nonconcerted15,16 with the β-lactam ring-opening (Scheme 1).
Whatever the process, the opening of the β-lactam leads to a well-
evidenced departure of the good R2 leaving group. The resulting
conjugate is unstable and suffers degradation through dihy-
drothiazine ring rupture.
Despite the lack of knowledge regarding the chemical struc-
tures derived from cephalosporins which interact specifically with
the variable region of antibodies, it is accepted that the cepha-
losporin is conjugated to the carrier molecule. Evidently, un-
certainty exists about the epitope structure, which could involve
either the complete cephalosporin structure or just parts of it
after degradation. This unresolved question, as well as whether
R1 and/or R2 are part of the antigenic determinant, will have
important implications in the IgE response and, moreover, in
further clinical evaluation with both in vivo and in vitro assays.
707
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
Figure 1. General retrosynthetic analysis for proposed antigenic determinants and description of all synthesized compounds, combining different R1
side chains (aꢀf) and functional groups (FG) at C-3 (1ꢀ4).
chemically well-defined compounds (1aꢀf, Figure 1), proposed
as models, were synthesized, and their recognition by IgE from
cephalosporin allergic patients was evaluated by a radioallergo-
sorbent test (RAST).28 The immunological results indicated that
the IgE antibodies recognize the proposed metabolites and that
specificities were related mainly to the R1 acyl side chain contain-
ing the culprit cephalosporin. The functional group in the
3-position was a methyl group, the most simple version of the
proposed model structures, in order to establish the basis of the
structural requirements for a specific IgE interaction. However,
the C-3 atom in the model structure, derived from C-6 in the
native cephalosporin, is originally like a carbonyl group and, as
such, can suffer different reactions in the physiological media. We
expected C-3 in our model antigenic determinants to present a
higher oxidation state than that corresponding to a methyl group,
by C-3 covalent binding to either oxygen or sulfur atom.
To confirm this assumption, we have extended our preliminary
work on structural requirements for a cephalosporin epitope by
assessing the recognition of new more complex molecules. The aim
of the present study was to investigate how the functional group at
C-3 in the proposed antigenic determinant model structures can
modulate their recognition by specific IgE antibodies. The synthesis
of a series of proposed antigenic determinants (Figure 1), all
containing the same basic structure but with variations in both the
substituent R1 and different functional groups at the C-3 position,
including hydroxyl, thiol, and acetal derived, is described. Obtaining
pure and chemically well-defined products allows one to study the
specificity of IgE antibodies from patients with an immediate allergic
reaction to cephalosporin. In vitro clinical assays were performed to
establish the recognition of different synthesized structures by IgE
antibodies specific to cephalosporins.
anhydride conditions were needed.29 Column chromatography was carried
out on silica gel 60 (63ꢀ200 μm, Merck 7734). Organic solutions were
dried with MgSO4 and concentrated in vacuum.
Instrumentation and Mode of Analysis. TLC analyses were
performed on silica gel 60 F254 (Merck 5719) or aluminum oxide 60
F254 (Merck 5550). As general criteria of purity, TLC analysis of all
immunologically evaluated compounds was performed in two different
sets of conditions, and a single spot was always observed. Melting points
were determined on a Gallenkamp instrument and are given uncor-
rected. MS (EI) were recorded on an HP-MS 5988A spectrometer
operating at 70 eV or Thermo Finnigan Trace at 70 eV. HRMS were
recorded on an AutoSpecE, CACTI, University of Vigo (Spain). NMR
1
spectra were recorded on a Bruker ARX-400 at 400 MHz for H and
100.6 MHz for 13C. Chemical shifts are given relative to the residual
signal of solvents, δ H 7.24 ppm and δ C 77.0 ppm for deuteriochloro-
form or δ H 3.31 ppm and δ C 49.0 ppm for deuteriomethanol. Optical
rotations were determined by using a Perkin-Elmer 241 digital polari-
meter. Elemental analyses were carried out in the Microanalytical
Laboratory, SCAI, University of Malaga.
General Procedures for Amide Formation (Coupling Re-
actions). Generel Procedure A. To a stirred solution of N-(tert-
butoxycarbonyl)-L-amino acid (1 equiv.) in anhydrous CHCl3 at 0 °C,
N-methylmorpholine (1.1 equiv.) and isobutylchloroformate (1.1
equiv.) were sequentially added dropwise. After 1 h, butylamine (1.2
equiv.) was added, and the reaction was slowly warmed to room tempera-
ture. After 40 h of reaction, the mixture was washed with H2O, 5% HCl
aqueous solution, H2O, 10% NaHCO3 aqueous solution, and H2O.
After workup, pure compounds were obtained.
Generel Procedure B. To a solution of acid 16aꢀf (1 equiv.) in
anhydrous CH2Cl2 (2 mL per mmol) in an ice bath were added successively
recrystallized N-hydroxybenzotriazole (HOBT) (1.2 equiv.) in DMF/
CH2Cl2 1:1 (1 mL per mmol) and N,N0-dicyclohexylcarbodiimide
(DCC) (1 equiv) in anhydrous CH2Cl2 (1 mL per mmol). The mixture
was stirred for 2 h at room temperature under argon. Then, amine
(1 equiv. of 7 þ N-methylmorpholine (NMM), 10, or 15) solution in
CH2Cl2 (2 mL per mmol) was added dropwise, and the mixture was
stirred for 40 h at room temperature. Most of the dicyclohexylurea was
’ EXPERIMENTAL PROCEDURES
Chemicals. All materials were obtained from commercial suppliers
and used without further purification. Solvents were distilled when
708
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
removed by filtration and washed with a small amount of CH2Cl2. The
solvents were evaporated. The remaining syrup was dissolved in CH2Cl2,
washed with 10% NaHCO3 aqueous solution, dried (MgSO4), and
evaporated.
washed with H2O, and dried over MgSO4. Solvent evaporation yielded a
white solid (97%). Recrystallization was from EtOAcꢀHexane.
Compound 12. This was synthesized as previously described.30
Compound 13. An excess of n-butylamine (5.75 mL, 58.1 mmol) was
added dropwise to stirred bromo acetal 12 (6 g, 26.4 mmol) under argon
at 5 °C. After addition, the reaction mixture was allowed to reach room
temperature and stirred for 24 h. Then, the mixture was concentrated
under reduced pressure. The residue was dissolved in CHCl3, washed
with 1 N HCl, dried over MgSO4, and concentrated under reduced
pressure to give a yellow solid (6.38 g, 90%) corresponding to the
desired product.
Compound 14. A mixture of 13 (2 g, 7.5 mmol) and benzyl amine
(3.25 mL, 29.8 mmol) was stirred and heated to 110 °C for 48 h. After
the reaction, the mixture was allowed to reach room temperature, a
minimum volume of cold CHCl3 was added, and the mixture was
filtered. The filtrate was dissolved with more CHCl3 (50 mL), washed
with H2O (ꢁ3), dried (MgSO4), and concentrated under reduced
pressure. The crude product was purified by column chromatography on
silica gel (Hexane/EtOAc, 1:1) affording 14 as a yellow oil (2.1 g, 95%).
Compound 15. A suspension of 14 (0.8 g, 2.7 mmol) and Pearlman
catalyst (0.67 g) in MeOH (90 mL) was hydrogenized for 48 h at 90 psi at
room temperature. The mixture was filtered through Celite, and the solvent
was eliminated to afford the desired amine 15 as a yellow liquid (99%).
Synthesis/Protection of Side Chain Acid: Side Chain Fragment
Compound 16a. 2-Oxo-furanacetic acid (840 mg, 6 mmol) was
treated with methoxylamine hydrochloride (1 g, 12 mmol) and Na2CO3
(2.5 g, 24 mmol) in anhydrous MeOH. The mixture was refluxed for 90
min. When the crude reaction reached room temperature, concentrated
HCl was added until the mixture reached pH 3. The solvent was
concentrated, and the resulting residue was extracted with Et2O
(70 mL and 3 ꢁ 25 mL). The organic phases were dried (MgSO4) and
concentrated under reduced pressure affording a 1:4 E/Zmixture of 16a as a
brown solid (0.91 g, 84%). Recrystallization from benzene gave the pure
isomer Z-16a as a white crystalline solid (364 mg, 36% global yield).
Generel Procedure C. The same procedure as that described in
General Procedure B was followed using the following reagents and
workup: acid 16aꢀf (1 equiv.) in anhydrous CH2Cl2 (2 mL per mmol),
HOBT (1.1 equiv.) in DMF/CH2Cl2 1:1 (1 mL per mmol), 1--
(3-(dimethylamino)propyl)-3-ethyl-carbodiimide hydrochloride (EDCI)
(1 equiv.) in anhydrous CH2Cl2 (5 mL per mmol), and amine (1 equiv.
of 7 þ NMM, 10, or 15) solution in CH2Cl2 (2 mL per mmol). After
stirring for 24 h at room temperature, solvents were evaporated. The
remaining syrup was dissolved in CH2Cl2, washed with 10% HCl
aqueous solution, 10% NaHCO3 aqueous solution and H2O, dried
(MgSO4), and evaporated.
General Procedure for the Hydrolysis of the Trityl Group.
The N-tritylated compound was treated with a 50% formic acid aqueous
solution for 30 min at 70 °C. The mixture was then cooled to room
temperature, and water was added. Triphenylcarbinol was separated by
filtration and washed with water. The filtrate was concentrated to
dryness under reduced pressure, and the residue was triturated with
Et2O. The solid was filtered and dried to obtain a pure product.
General Procedure for the Hydrolysis of the tert-Butox-
ycarbonyl Group. General Procedure D. To a stirred solution of the
corresponding compound in EtOAc, at room temperature, a 3 M HCl
aqueous solution was added. After 5 h, the solvent was evaporated in
vacuum, and the oily residue was triturated with Et2O affording the
product as a white solid in quantitative yield.
General Procedure E. To a stirred solution of the corresponding
compound in CH2Cl2, at room temperature, trifluoroacetic acid was
added. After 4 h, the solvent was evaporated in vacuum, and the oily
residue was triturated with Et2O affording the product as a white solid in
quantitative yield.
General Procedure for Disulfide Reduction. General Proce-
dure F. A solution of the disulfide (1 equiv.), triethylamine (TEA)
(2 equiv.), and DTT (4 equiv.) in degassed solvent was stirred for 1 h at
room temperature under argon. CH2Cl2 was the solvent used for side
chains a and c; MeOH was used for b and e. When the solvent was
MeOH, it was evaporated, and the mixture was dissolved in EtOAc. The
organic phase was washed with water and dried over MgSO4.
General Procedure G. A solution of the disulfide (1 equiv.) and DTT
(4 equiv.) in degassed water was stirred for 10 h at room temperature
under argon. Then, several drops of concentrated HCl were added, and
the solution was stirred for 10 min. The aqueous phase was washed with
CH2Cl2 (ꢁ3) and evaporated in vacuum, and the resultant residue was
triturated with degassed ethyl ether.
Compound 16b.
A solution of 2-(2-amino-1,3-thiazol-4-yl)-
2-(methoxyimino) acetic acid (6 g, 29.8 mmol) and TEA (9.2 mL,
65.6 mmol) in CHCl3/DMF 2:1 (150 mL) was cooled to 0 °C. Then
trityl chloride (9.14 g, 32.8 mmol) was added in small portions for one
hour. The mixture was slowly warmed to room temperature overnight.
Afterward, 1 M HCl aqueous solution was added, and the mixture was
extracted with CHCl3. After drying (MgSO4) and solvent evaporation, a
white solid was obtained (10.7 g, 81%).
Compounds 16d and 16e. A solution of di-tert-butyldicarbonate (1.1
equiv.) in acetone was slowly added to a stirred solution of the
corresponding compound (1 equiv.) and TEA (1 equiv.) in acetone/
H2O 1:1 at room temperature. The mixture was stirred for 24 h, and the
acetone was evaporated. The aqueous phase was acidified to pH 2 with
10% HCl aqueous solution and extracted with Et2O (ꢁ2). The
combined ethereal extracts were dried over MgSO4, filtered, and
evaporated in vacuum yielding the corresponding pure compound.
Compounds 2ꢀ4 (aꢀf): Model Structures
Amine Synthesis: Nuclear Structure Fragments
Compound 6 (N-Boc-Ser-NHnBu). This was obtained following
general procedure A for amide formation starting from compound 5
(1.4 g, 6.9 mmol). It recrystallized from EtOAcꢀhexane as a white
crystalline solid (1.4 g, 77%).
Compound 7 (Ser-NHnBu). To a stirred solution of compound 6
(1.4 g, 5.4 mmol) in EtOAc (6 mL), at room temperature, a 3 M HCl
aqueous solution was added. After 3 h, the solvent was evaporated in
vacuum, and the oily residue was triturated with Et2O to obtain the
hydrochloride product as a colorless oil (0.94 g, 88%).
Compound 2a. This was obtained following general procedure B
described for amide formation starting from 16a (338 mg, mmol). The
crude product was purified by column chromatography on silica gel
(hexane/EtOAc, 4:6) affording 2a as a white solid (250 mg, 40%). It was
recrystallized in benzene.
Compound 9 (N-Boc-Cys-NHnBu)2. This was obtained following
general procedure A for amide formation starting from compound 8
(2.6 g, 6 mmol). Recrystallization from CHCl3ꢀEtOAc gave a white
solid (1.6 g, 99%).
Compound 17b. This was obtained following general procedure
B described for amide formation starting from 16b (887 mg, 2 mmol).
The crude product was purified by column chromatography on silica gel
(EtOAc) affording 17b as a yellow solid (730 mg, 63%). It was
recrystallized in EtOAcꢀhexane.
Compound 10 (Cys-NHnBu)2. To a stirred solution of compound
9 (2.4 g, 4.35 mmol) in CH2Cl2 (25 mL), at room temperature,
trifluoroacetic acid (2 equiv) was added. After 20 h, 75 mL of CH2Cl2
was added. The solution was neutralized with NaOH aqueous solution,
Compound 2b. This was obtained following the general proce-
dure described for hydrolysis of trityl group, starting from 17b (0.68 g,
1.2 mmol). Yellow solid (0.3 g, 75%).
709
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
Compound 2c. This was obtained following the general procedure B
described for amide formation starting from 16c (284 mg, 2 mmol). The
crude product was purified by column chromatography on silica gel
(hexane/EtOAc, 1:1) affording 2c as a white solid (250 mg, 45%). It was
recrystallized in CHCl3ꢀhexane.
Compound 18e. This was obtained following general procedure C
described for amide formation starting from 16e (801 mg, 3 mmol).
After flash chromatography (hexane/EtOAc, 1:1), the pure compound
was obtained as a white solid (535 mg, 42%).
Compound 19e. This was obtained following general procedure E
described for hydrolysis of tert-butoxycarbonyl group, starting from 18e;
white solid.
Compound 3e. This was obtained following disulfide reduction
procedure F as described, starting from 19e (0.16 g, 0.25 mmol). After
flash chromatography (CHCl3/MeOH, 10:0.3), the pure compound
was obtained. It was recrystallized in benzene to give white needles (78
mg, 50%).
Compound 17d. This was obtained following general procedure B
described for amide formation starting from 16d (948 mg, 7 mmol). The
crude product was purified by column chromatography on silica gel
(hexane/EtOAc, 4:6) affording 17d as a white solid (760 mg, 32%). It
was recrystallized in EtOAcꢀhexane.
Compound 2d. This was obtained following procedure D described
for the hydrolysis of the Boc group, starting from 16d; white solid.
Compound 17e. This was obtained following general procedure B
described for amide formation starting from 16e (512 mg, 1.9 mmol).
The crude product was purified by column chromatography on silica gel
(hexane/EtOAc, 1:1) affording 17e as a white solid (370 mg, 47%).
Compound 2e. This was obtained following procedure D described
for hydrolysis of the Boc group, starting from 17e; white solid.
Compound 2f. This was obtained following general procedure B
described for amide formation starting from 16f (220 mg, 1.4 mmol).
The crude product was purified by column chromatography on silica gel
(hexane/EtOAc, 1:1 and then increasing the polarity EtOAc/MeOH,
9.5:0.5) affording 2f as a white solid (100 mg, 23%).
Compound 18f. This was obtained following general procedure C
described for amide formation starting from 16f (229 mg, 1.5 mmol). A
white solid (150 mg, 32%) corresponding to compound 18f was
obtained after flash chromatography (hexane/EtOAc, 1:1) or recrystal-
lization in MeOHꢀCHCl3ꢀhexane.
Compound 3f. This was obtained following disulfide reduction
procedure F as described, starting from 18f (0.16 g, 0.25 mmol). The
pure compound was obtained after flash chromatography (CHCl3:
MeOH, 10:0.3); white solid (78 mg, 50%).
Compound 4a. This was obtained following general procedure B
described for amide formation, starting from a mixture of E/Z isomers
(relative proportion 1:4) of 16a (507 mg, 3 mmol). The crude product
was purified by column chromatography on silica gel (CHCl3) affording
an E/Z 1:4 mixture of 18a (910 mg, 85%). The isomers were separated
by column chromatography on silica gel (CHCl3/MeOH, 10:0.5) to
give the Z isomer as a white solid (665 mg, 62%).
Compound 20b. This was obtained following general procedure B
described for amide formation, starting from 16b (655 mg, 1.5 mmol).
After flash chromatography (CHCl3), the pure compound 20b was
obtained as a yellow solid (559 mg, 60%).
Compound 18a. This was obtained following general procedure C
described for amide formation starting from 16a (507 mg, 3 mmol). The
crude product was purified by flash chromatography (CHCl3/MeOH,
10:0.15) affording 18a as a white solid (570 mg, 58%). It was
recrystallized in CHCl3ꢀbenzeneꢀhexane.
Compound 3a. This was obtained following disulfide reduction
procedure F described, starting from 18a (0.3 g, 0.46 mmol). After
flash chromatography (CHCl3/MeOH, 10:1.5), the pure compound
was obtained. It was recrystallized in benzene to obtain the compound as
white needles (180 mg, 71%).
Compound 4b. This was obtained following the general procedure
described for hydrolysis of the trityl group, starting from 20b (0.75 g, 1.2
mmol); yellow solid (0.3 g, 60%).
Compound 18b. This was obtained following general procedure C
described for amide formation starting from 16b (1.33 g, 3 mmol). The crude
extract was purified by flash chromatography (hexane/EtOAc, 7:3) to give
18b as a yellow solid (1.04 g, 58%). It was recrystallized in CHCl3ꢀhexane.
Compound 19b. Compound 18b (0.37 g, 0.3 mmol) was treated
with a 1:1 mixture of acetone/50% aqueous formic acid solution for 90
min at 70 °C. After reaching room temperature, acetone was evaporated,
and then a yellow solid (0.17 g, 78%) was obtained, working-up the
general procedure described for the hydrolysis of the trityl group.
Compound 3b. This was obtained following disulfide reduction
procedure F as described, starting from 19b (165 mg, 0.23 mmol).
The product was purified by column chromatography (hexane/EtOAc,
1:1); white solid (130 mg, 79%).
Immunochemical Studies. Subjects. From the patients with a
confirmed diagnosis of an immediate reaction to cephalosporins, we
selected seven cases that showed high levels (greater than 7%) of specific
IgE antibodies to the cephalosporin involved in the reaction measured
by direct RAST. Data from these patients included in the study are
shown in Table 1. These patients were diagnosed following the pro-
cedure described in the European Network of Drug Allergy (ENDA)
protocol based on skin testing, in vitro tests, and a drug provocation test
if necessary.31 Three types of clinical manifestations had been experi-
enced: anaphylactic shock, anaphylaxis, or urticaria. Anaphylaxis was
considered to be the presence of several of the following symptoms:
pruritus on palms or soles, later becoming generalized, generalized
erythema, urticaria, dyspnea, difficulty speaking or swallowing, and/or
tachycardia. If hypotension and/or loss of consciousness also appeared,
an anaphylactic shock was considered. Urticaria was defined as rapidly
evolving and transient pruriginous wheals occurring at different
body sites.
The institutional review boards approved the study, and informed
consent for the diagnostic procedures was obtained from all patients.
Skin Tests. Skin testing was carried out,31,32 using 0.02 mL of solution
prepared daily. The reagents and maximum concentrations accepted
nowadays were benzylpenicilloyl-poly-L-lysine (PPL) (Diater Labora-
tories, Madrid, Spain) (5 ꢁ 10ꢀ5 M), minor determinant mixture
(MDM) (Diater Laboratories, Madrid, Spain) (2 ꢁ 10ꢀ2 M), benzyl-
penicillin (Normon SA, Madrid, Spain) (105 IU/mL), amoxicillin (SB
Smithkline Beecham, Madrid, Spain) (20 mg/mL), and ampicillin
(Normon SA) (20 mg/mL). The following cephalosporins (2 mg/
mL) were also used: cefaclor (Lilly, Madrid, Spain), cefuroxime
(GlaxoSmithKline S.A, Madrid), ceftazidime (GlaxoSmithKline S.A),
Compound 18c. This was obtained following general procedure C
described for amide formation starting from 16c (428 mg, 3 mmol).
Compound 18c was obtained as a white solid (661 mg, 73%) from the
crude product recrystallization with CHCl3ꢀhexane.
Compound 3c. This was obtained following disulfide reduction
procedure F as described, starting from 18c (0.42 g, 0.7 mmol). The
product was purified by column chromatography (CHCl3/MeOH,
10:0.5); white solid (0.3 g, 71%).
Compound 18d. This was obtained following general procedure C
described for amide formation starting from 16d (753 mg, 3 mmol). A
white solid (946 mg, 77%) corresponding to compound 18d precipi-
tated with CHCl3ꢀEt2O.
Compound 19d. This was obtained following general procedure B
described for the hydrolysis of the tert-butoxycarbonyl group, starting
from 18d; white solid.
Compound 3d. This was obtained following disulfide reduction
procedure G as described, starting from 19d (280 mg, 0.45 mmol);
white solid (0.19 g, 67%).
710
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
Table 1. Clinical Characteristics of Allergic Patients and Skin Test Results with Different Cephalosporins
skin tests
case age sex reaction
drug
INTa cefuroxime cefotaxime cefozidime ceftriaxone cefepime cefonicid cefaclor ceftazidime cefadroxil
1
2
3
4
5
6
7
57
38
42
36
22
42
32
F
F
M
F
F
F
F
anaph.
ceftazidime 120
ꢀ
þ
þ
þ
þ
þ
þ
ꢀ
þ
ꢀ
þ
ꢀ
þ
ꢀ
ꢀ
þ
ꢀ
þ
ꢀ
þ
ꢀ
ꢀ
þ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
þ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
urticaria cefotaxime
10
1
anaph.
anaph.
anaph.
anaph.
anaph.
cefuroxime
cefuroxime
6
cefuroxime 0.5
cefuroxime 0.5
cefuroxime
1
a INT, interval between the reaction and study (in months).
Scheme 3. Synthesis of (A,B) Nuclear Structure Fragments and (C) Model Haptenic Structures
ceftriaxone (Roche, Basel, Switzerland), cefotaxime (Roussel Iberica
Laboratorios, Madrid, Spain), and cefonicid (GlaxoSmithKline S.A).
Higher concentrations may cause nonspecific irritative reactions, even in
subjects with good tolerance or no exposure to β-lactams.31 In the skin
prick tests, a wheal >2 mm was considered positive. In the intradermal
tests, the wheal area was marked initially and 20 min after testing, and an
increase in diameter greater than 3 mm was considered positive.31,33
Radioimmunoassay for IgE Determination. These were done by
RAST to benzylpenicillin, amoxicillin, ampicillin, and to different
cephalosporins (cefuroxime, cefotaxime, cefoxidime, ceftriaxone, cefe-
pime, cefonicid, cefaclor, ceftazidime, and cefadroxil) conjugated to
poly-L-Lysine (PLL) (Sigma, St. Louis, MO), as previously described.32
Blood samples were obtained when patients were evaluated, and sera
were kept at ꢀ20 °C until assayed. Briefly, 30 μL of the patient’s sera was
incubated in the solid phase with different β-lactamꢀPLL conjugates.
After washing, radiolabeled anti-IgE antibody (Kindly provided by ALK-
Abello Laboratories, Madrid, Spain) was added and incubated overnight.
The discs were then washed, and their radioactivity was measured in a
gamma counter, Cobra II autogamma (Packard BioScience Company,
Frankfurt, Germany). Results were calculated as a percentage of the
maximum and considered positive if they were higher than 2.5%, which
was the mean þ2SD of a negative control group.32
Study of Immunochemical Recognition. In order to determine the
recognition of the different synthetic structures by specific IgE anti-
bodies, competitive inhibition immunoassays were carried out with
those sera that showed positive results in the RAST, higher than 7%,
using in the solid phase the culprit cephalosporin conjugated to PLL, as
described.34 The RAST inhibition test was undertaken using in the fluid
phase the sera from the patients incubated with 10-fold concentrations
(from 100 to 10 mM) of the compounds synthesized with different
711
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
Figure 2. Comparison of the recognition of structures 1 (aꢀf) and 2 (aꢀf) by RAST inhibition studies in sera from patients allergic to ceftazidime
(serum 1) (A), cefotaxime (serum 2) (B), and cefuroxime (serum 3) (C).
functionalities (H, OH, SH, and O(CH3)2) at the C-3 position in each
case with six side chains at the R1 position (Figure 1). Only two
concentrations were used in order to save sera since many determina-
tions were required because of the number of compounds to be tested.
On the basis of previous studies, these concentrations, although high,
were known not to induce nonspecific binding.34 After 18 h of
incubation at room temperature, discs sensitized with the culprit
cephalosporin conjugated to PLL were added. Results were expressed
as percentage inhibition with respect to the noninhibited serum.
Following the same methodology for model structures type 2
(FG = OH) and 3 (FG = SH), the nuclear structure fragments
could be derived from serine and cysteine, respectively. Serine
was used without any protection in the hydroxyl group and the
use of cystin, instead of cystein, provided a protected thiol group
during synthetic procedures. Both amino protected amino acids,
N-Boc-L-serine and N-Boc-L-cystin, were reacted with n-butyla-
mine via carboxyl group activation as a mixed anhydride, to
obtain amides 6 and 9, respectively (Scheme 3A). Their sub-
sequent amine deprotection under acidic conditions yielded the
nuclear structure fragments 7 and 10, with an overall good yield.
To obtain the new third nuclear structure fragment, necessary
to have model structures with a higher oxidation state at C-3 in
the model structure, compounds 4 (FG = (OCH3)2), the amine
nuclear structure fragment 15 was quantitatively obtained following
an analogous methodology used for the synthesis of similar
products,30 described in Scheme 3B. The methoxy acrylate 11
was converted to bromo acetal 12 with N-bromosuccinimide
(NBS)/MeOH by a bromo etherification reaction of the
alkene.30 Reaction of 12 with an excess of n-butylamine at room
temperature selectively afforded bromo amide 13, and the
nucleophilic displacement of bromide 13 took place with an
excess of benzyl amine at higher temperatures to give product 14.
The last step of hydrogenolysis with a Pearlman catalyst afforded
the free primary amine 15.
’ RESULTS
Synthesis. To synthesize the series of proposed antigenic
determinant structures, we followed the general retrosynthetic
analysis described in Figure 1. The first disconnection of the
amide bond leads to two synthons, the acid functional side chain,
and the amine functional nuclear structure. The amide bond of
the latter structure can be disconnected into a protected R-amino
acid and n-butylamine. Therefore, the suggested antigenic de-
terminant 1ꢀ3 (aꢀf) structures are proposed to be synthesized
from the corresponding R1 acyl side chain, plus a derived
R-amino acid, and n-butylamine (Figure 1). In addition, the
possibility of using different amino acids makes the synthetic
methodology versatile, leading to different functional groups
(FG) present in the nuclear structure fragment. The most basic
structure involving our proposed antigenic determinant model
(1, FG = H) presents a methyl group in the carbon correspond-
ing to the 6-position remaining from the cephalosporins. The
straightforward synthesis was carried out starting from an alanine
derivative and the different R1 acyl side chains.
Amino-functionalized R1 side chains (b,d,e) were employed as
0
protected, R1 , with either the tertbutoxicarbonyl or the triphe-
nylcarbinol group 16(b,d,e). A nonreported synthesis and isola-
tion of the pure Z isomer 16a acyl side chain, which involved a
712
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
Figure 3. RAST inhibition studies in sera from five patients allergic to
cefuroxime using compounds 1a (FG = H), 2a (FG = OH), and 3a (FG
= SH), at two concentrations (100 and 10 mM).
dehydrating condensation for imine formation and further re-
crystallization of the Z isomer in benzene, facilitated the pre-
paration of structures 1aꢀ4a, avoiding the extensive chromato-
graphic separations needed when the methoximine group was
formed in the last step of the synthesis and therefore yielding
higher overall conversions. The configuration of the 16a Z
isomer was determined by nuclear Overhouser effect spectros-
copy (NOESY) experiments.
Figure 4. RAST inhibition studies in sera from three patients allergic to
cephaloporins using compounds 1 (a,b) (FG = H), 2 (a,b) (FG = OH),
3 (a,b) (FG = SH), and 4 (a,b) (FG = (OCH3)2) at two concentrations
(100 and 10 mM).
The three amine nuclear structure fragments (7, 10, and 15)
were reacted with the acyl side chains 16aꢀf via a previous carboxyl
group activation with a carbodiimide reagent (Scheme 3C). An
additional deprotection step under acidic conditions was carried
out to molecules containing b, d, and e side chains. Furthermore,
to obtain structures 3, the disulfide bonds of symmetrical di-
sulfides were finally reduced with DTT,35 to liberate two
equivalents of the desired thiols 3aꢀf. The addition of TEA in
this reaction to reach high pH was not necessary for the synthesis
of products 3d and 3e.
All the products obtained, as well as the intermediates in their
synthesis, were conveniently purified and analyzed to confirm the
structure of the proposed molecules (Supporting Information).
Immunochemical Studies. The ability of IgE in sera from
patients allergic to selected cephalosporins to recognize com-
pounds 1ꢀ4 (aꢀf) was studied by competitive RAST inhibition
immunoassays.
specific antibody recognition, the inhibition of a new synthesized
series of compounds 2aꢀf was compared to those involving the
more basic structures 1aꢀf. For this, three sera from patients
with allergic reactions to ceftazidime, cefotaxime, and cefurox-
ime, respectively, were analyzed. Sera 1 (Figure 2A) and sera 2
(Figure 2B) achieved the maximal inhibition percentage with
compounds 1b and 2b, which contain the same R1 side chain (b)
as the cephalosporins involved in the reaction. Serum 2 also
showed a high recognition with compounds 1a and 2a, the side
chain chemical structure of which is similar to that in the
cephalosporin involved in the reaction. Serum 3 (Figure 2C),
specific to cefuroxime, showed a maximal inhibition percentage
with compounds 1a and 2a, which contain the same side chain as
cefuroxime. In addition, there was an important specific binding
to 1b, 2b, 1c, and 2c, showing as a consequence a high cross-
reactivity to cephalosporins including a 5-membered heterocyc-
lic structure in their R1 side chain.
Regarding the comparison between different functional
groups at C-3, (FG = H, OH), RAST inhibition studies showed
To study the effect of presenting a hydroxyl functional group
at C-3 in model structures, containing the six R1 side chains, on
713
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
Table 2. RAST Results with Different Cephalosporinsa
RAST
case
drug involved
cefuroxime
cefotaxime
cefozidime
ceftriaxone
cefepime
cefonicid
cefaclor
ceftazidime
cefadroxil
1
2
3
4
5
6
7
ceftazidime
cefotaxime
cefuroxime
cefuroxime
cefuroxime
cefuroxime
cefuroxime
1.72
4.52
3.09
19.75
0.89
2.02
0.35
0.15
2.22
0.12
0
2.27
2.62
0.16
1.99
0.2
1.99
13.34
0.11
1.51
0.36
0
0.43
0.49
0.18
1.08
0.3
0.46
1.03
0
23.31
0.25
0
0.35
0.11
0
9.43
20.76
2.61
13.9
0.53
0.73
0.4
0.06
0
0.67
0.28
0.15
1.36
0.03
0.02
0.14
0.02
3.46
0.18
0.29
1.02
0.29
0.45
13.15
1.45
0.89
0
a Bold faced values indicate positive results.
the same tendency in the IgE recognition of either the methyl
group in structures 1 or the hydroxymethyl group in structures 2,
with the chemical structure in the R1 position being critical
(Figure 2AꢀC). The extension of the study including the
functional group thiol (Figure 3) showed a specific recognition
of 1a and 2a molecules, with the better recognized in most cases
being those with hydroxyl group 2a, whereas there was poor
recognition for the molecule with thiol group 3a.
select β-lactams for penicillin/cephalosporin allergic patients
since many questions about cross-reactivities between cephalos-
porins and between cephalosporins and penicillins are still
unresolved. The lack of standard in vitro diagnostic methods
has led to the use of in vivo methods using the native cephalos-
porin, which can be risky. Therefore, the consumption of these
β-lactams is unsafe when the subject has had an adverse reaction
to just one β-lactam.
The results showed that the specificity of IgE antibodies from
all the patients evaluated recognized these synthetic structures
with different patterns that correlated with the in vivo and in vitro
patterns of response reported in previous studies.17,28 To sum-
marize the overall trend seen in most sera, the cases showed
optimal inhibition with some of the proposed model structures
containing the R1 side chain of the cephalosporin which induced
the reaction. Additionally, different degrees of cross-reactivity
were observed with structures containing the R1 side chain with a
similar chemical structure. This is in agreement with previous
studies where IgE recognition was evaluated with other ap-
proaches, such as skin test or RAST.17,18
Regarding the influence of the functionality at C-3, RAST
inhibition studies showed similar IgE recognition to the pro-
posed model structures when the C-3 formed part of a methyl
(1), hydroxymethyl (2), or dimethoxyacetal group (4). Further-
more, results indicated that molecules containing FG = OH and
FG = (O(CH3)2) at C-3, structures 2 and 4, respectively, slightly
enhanced the IgE recognition in most cases (sera 3ꢀ7). This
improvement in the IgE molecular interaction process indicates
that these structures mimic the real antigenic determinant in-
volved in the initial immunological recognition. The actual
antigenic determinants, as a consequence, likely contain a higher
oxidation state than that corresponding to a methyl group at C-3
in the model structure, which is in agreement with the fact that
the C-6 in the native cephalosporin presented an oxidation state
equivalent to a carbonyl functional group and therefore higher
than that corresponding to the methyl group.
To complete the study, structures presenting two R1 side
chains, a and b, corresponding to cefuroxime and cefotaxime
respectively, and four different functional groups (FG = H, OH,
SH, and (OCH3)2) were evaluated through inhibition studies by
using three sera from patients allergic to cephalosporins
(Figure 4). Data showed that serum 5, specific to cefuroxime,
showed higher recognition with three of the structures contain-
ing the cefuroxime side chain (a) and was similar with structures
1a, 2a, and 4a. The same serum also showed inhibition with
compounds containing the cefotaxime side chain (b), again with
similar recognition of 1b, 2b, and 4b. There was no recognition
of compounds containing the thiol functional group at C-3, 3a
and 3b. For serum 7, also specific to cefuroxime, the RAST
inhibition assay showed a strong binding with structure 2a,
followed by 4a and 1a, with no recognition of 3a. Among
compounds containing the cefotaxime side chain, only structure
4d showed a weak recognition. Serum 2, specific to cefotaxime,
showed a specific binding with the structures 1b, 2b, and 4b, and
no inhibition with 3b. This sera also showed high inhibition with
compounds (1ꢀ4)a, which included the cefuroxime side chain.
No recognition was obtained with the compounds 1ꢀ4 (dꢀe)
no matter the group present at C-3. This is because these
compounds have the side chain d of cefaclor or cephalexin, e
of cefadroxil or cefprozilo, and f of cefonicid or cefamandole.
Significantly, the different sera included in the study came from
patients allergic to cephalosporins in which the culprit cephalos-
porin was ceftazidime, cefotaxime, or cefuroxime, which have a
very different side chain at R1 position.
All these patterns of recognition of different compounds
depending on the R1 side chain correlated with the skin test
and direct RAST results (Table 2).
However, despite the fact that the thiol group provided C-3
with an oxidation grade equivalent to the hydroxyl group, the
simple presence of a sulfur atom on structures 3 drastically
decreased the inhibition and showed no inhibition in most cases.
The lack of IgE recognition obtained with structures 3, even
though they contained the same R1 side chain as the cephalos-
porin inducing the reaction, provides significant information regard-
ing the nuclear structure fragment of the antigenic determinant.
Thus, the sulfur atom may not remain bonded to the epitope.
These data help understand the generation of metabolites
which are later recognized by specific IgE, particularly relating to
’ DISCUSSION
Despite the uncertainty about structures comprising the epitopes
of cephalosporins, the same clinical behavior model has been
assumed for cephalosporins and penicillins, without considering
the different chemical reactivities of both β-lactams.27,36 Conse-
quently, clinicians have to make difficult decisions in order to
714
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
tested in C-3, hydroxymethyl, and acetal groups showed the same
pattern with improved inhibition compared to the methyl group.
From a practical point of view, structures 2 (FG = OH) have an
additional advantage when included in the test since they present
higher solubility in aqueous media. As these structures can be
detected with a high specificity, model structures 2 could be
useful in the development of in vitro methods for detection and
quantification of IgE specific to cephalosporins.
’ CONCLUSIONS
A series of well-defined synthetic chemical structures proved
to be elegant tools to investigate the antigenic determinants to
cephalosporin, which is otherwise difficult to study. The pro-
posed epitopes, 1ꢀ4 (aꢀf), consist of closely related low-
molecular-weight molecules, differing only in the functionality
at position C-3 and the R1 side chains. Fine structural selective
recognition was detected between IgE from patients allergic to
cephalosporins and our proposed skeleton epitopes involving the
appropriate functionality and R1 side chain. Thus, as a result of
this research we have been able to refine the earlier design of
model antigenic determinants which fulfill the structural implica-
tions needed for recognition by IgE antibodies to cephalosporin
and which are consequently responsible for the adverse reactions.
Besides the improvement obtained with the new structures (2
and 4) in IgE molecular recognition, the aqueous solubility
benefit of hydroxyl functionalized model structures (2) makes
them potential candidates for a useful in vitro diagnostic test. We
have shown here that minor alterations of the structure can result
in major deviations in the ability to be recognized by the IgEs. As
a result, the change of the oxygen atom in model compounds 2 by a
sulfur atom, such as compounds 3, as can be supposed from the
hypothetical pathways of fragmentation of the starting antibiotic,
produces a significant diminution in the possibility of being recog-
nized by IgEs. Other chemical modifications of the model structures,
as well as their conjugation to carrier proteins and solid phases, are
under way to extend the evaluation of structural implications in the
recognition process and develop clinically useful diagnostic tests.
Figure 5. Hypothetic degradation structures derived from cephalos-
porin conjugation depending on the broken bond. (A) Nonconcerted
process intermediate; (B) structure which eliminated its R2.
the broken bond in the cephalosporins during the formation of
their antigenic determinant. The hypothetical intermediate struc-
tures after conjugation of the cephalosporin to proteins are de-
scribed in Figure 5, indicating the bonds which could undergo
fragmentation. Since model structures 2 and 4, involving hydro-
xymethyl and acetal groups as C-3, were the best recognized by
specific IgE, the in vivo formation of cephalosporin antigenic de-
terminants could occur via fragmentation of the S1ꢀC6 bond.
These new well-defined structures help us understand epitope
formation after conjugation of the cephalosporin to carrier pro-
teins. The findings confirmed that besides recognition at the R1
acyl side chain structure, the rest of the nuclear structure is also
important for optimal molecular recognition.
Several toxicological effects can result from the covalent
binding of small drug molecules to proteins.37 Structural studies
of cephalosporinꢀprotein adducts have never successfully been
addressed and are difficult to investigate. The way in which the
drug metabolizes after protein conjugation is very important not
only from a chemical point of view but also from an applications
in the clinical diagnosis of allergy point of view. The use of well-
defined structures and their exact concentrations in these im-
munoassays provide precise information about the chemical re-
quirements of IgE recognition to cephalosporins. Whether a
monomeric conjugate permits a better or worse inhibition and
thus different recognition by specific IgE antibodies provides
information about the relevant role of the different parts of the
cephalosporin molecule at the hapten binding site and, therefore,
about the nature of the recognition process and epitope compo-
sition. As a rule, evaluated sera had optimal recognition with
structures containing the same R1 side chain as the cephalosporin
inducing the allergic reaction and, in cross-reactivity cases,
structures including a similar R1 structure. The functional groups
’ ASSOCIATED CONTENT
S
Supporting Information. Structural characterization in-
b
cluding MS, 1H, and 13C NMR are described for final structures
2ꢀ4 (aꢀf) and intermediate pure products needed for the
synthesis. This material is available free of charge via the Internet
at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: (þ34) 952137565. Fax: (þ34) 952131941. E-mail:
inestrosa@uma.es.
Funding Sources
Grants from Ministerio de Ciencia e Innovacion-Spain (CTQ2007-
60190; CTQ2010-20303), Junta de Andalucía (PI-0551/2009, PI-
0545-2010, CTS-06603), Fondo de Investigaciones Sanitarias-Spain
(FIS/PS09/01768), and FIS-Thematic Networks and Co-operative
Research Centres, RIRAAF (RD07/0064), are acknowledged.
’ DEDICATION
This article is dedicated to Professor Rafael Suau, who passed
away before the submission of this article. Dr. Suau was an
715
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
eminent scientist and a good friend, and he was highly involved
in this research.
(17) Antunez, C., Blanca-Lopez, N., Torres, M. J., Mayorga, C.,
Perez-Inestrosa, E., Monta~nez, M. I., Fernandez, T., and Blanca, M.
(2006) Immediate allergic reactions to cephalosporins: Evaluation of
cross-reactivity with a panel of penicillins and cephalosporins. J. Allergy
Clin. Immunol. 117, 404–410.
(18) Romano, A., Mayorga, C., Torres, M. J., Artesani, M. C., Suau,
R., Sanchez, F., Pꢀerez, E., Venuti, A., and Blanca, M. (2000) Immediate
allergic reactions to cephalosporins: Crossreactivity and selective re-
sponses. J. Allergy Clin. Immunol. 106, 1177–1183.
(19) Saenz de San Pedro, B., Mayorga, C., Torres, M. J., Florido, J. F.,
Quiralte, J., and Blanca, M. (2002) Boosted IgE response after anaphy-
laxis reaction to cefuroxime with cross-reactivity with cefotaxime. Ann.
Allergy Asthma Immunol. 89, 101–103.
(20) Miranda, A., Blanca, M., Vega, J. M., Moreno, F., Carmona,
M. J., García, J. J., Segurado, E., Justicia, J. L., and Juarez, C. (1996)
Cross-reactivity between a penicillin and a cephalosporin with the same
side chain. J. Allergy Clin. Immunol. 98, 671–677.
(21) Audicana, M., Bernaola, G., Urrutia, I., Echechipia, S., Gastaminza,
G., Mu~noz, D., Fernꢀandez, E., and Fernꢀandez de Corres, L. (1994) Allergic
reactions to betalactams: studies in a group of patients allergic to penicillin
and evaluation of cross-reactivity with cephalosporin. Allergy 49, 108–113.
(22) Novalbos, A., Sastre, J., Cuesta, J., De Las Heras, M., Lluch-Bernal,
M., Bombín, C., and Quirce, S. (2001) Lack of allergic cross-reactivity to
cephalosporins among patients allergic to penicillins. Clin. Exp. Allergy
31, 438–443.
(23) Romano, A., Piunti, E., Di Fonso, M., Viola, M., Venuti, A., and
Venemalm, L. (2000) Selective immediate hypersensitivity to ceftriaxone.
Allergy 55, 415–416.
(24) Blanca, M., Fernandez, J., Miranda, A., Terrados, S., Torres,
M. J., Vega, J. M., Avila, M. J., Perez, E., Garcia, J. J., and Suau, R. (1989)
Cross-reactivity between penicillins and cephalosporins: clinical and
immunologic studies. J. Allergy Clin. Immunol. 83, 381–385.
(25) Pichichero, M. E., and Casey, J. R. (2007) Safe use of selected
cephalosporins in penicillin-allergic patients: A meta-analysis. Otolaryngol.,
Head Neck Surg. 136, 340–347.
(26) Pichichero, M. E. (2005) A review of evidence supporting the
American Academy of Pediatrics recommendation for prescribing
cephalosporin antibiotics for penicillin-allergic patients. Pediatrics
115, 1048–1057.
(27) Baldo, B. A. (1999) Penicillins and cephalosporins as allergens:
structural aspects of recognition and cross-reactions. Clin. Exp. Allergy
29, 744–749.
(28) Sanchez-Sancho, F., Perez-Inestrosa, E., Suau, R., Monta~nez, M. I.,
Mayorga, C., Torres, M. J., Romano, A., and Blanca, M. (2003) Synthesis,
characterization and immunochemical evaluation of cephalosporin anti-
genic determinants. J. Mol. Recognit. 16, 148–156.
(29) Armarego, W. L. F., and Perrin, D. D. (1996) Purification of
Laboratory Chemicals, Butterworth-Heinemann, Oxford, UK.
(30) Swaminathan, S., Singh, A. K., Li, W.-S., Venit, J. J., Natalie, K. J.,
Jr., Simpson, J. H., Weaver, R. E., and Siverberg, L. J. (1998) A
chemoselective, acid mediated conversion of amide acetal to oxazole:
the key step in the synthesis of cardiovascular drug, ifetroban sodium.
Tetrahedron Lett. 39, 4769–4772.
’ ABBREVIATIONS
CPO, cephalosporoyl; DCC, N,N0-dicyclohexylcarbodiimide; FG,
functional group; EDCI, 1-(3-(dimethylamino)propyl)-3-ethyl-
carbodiimide hydrochloride; ENDA, European Network of Drug
Allergy; HOBt, N-hydroxybenzotriazole; MDM, minor determi-
nant mixture; NBS, N-bromosuccinimide; NMM, N-methylmor-
pholine; NOESY, nuclear Overhouser effect spectroscopy; PLL,
poly-L-lysine; PO, penicilloyl; RAST, radioallergosorbent test; 2
SD, standard deviation; TEA, triethylamine.
’ REFERENCES
(1) Uetrecht, J. (2009) Immune-mediated adverse drug reactions.
Chem. Res. Toxicol. 22, 24–34.
(2) Greenberger, P. A. (2006) Drug allergy. J. Allergy Clin. Immunol.
117, 464–470.
(3) Landsteiner, K., and Jacobs, J. (1935) Studies on the sensitiza-
tion of animals with simple chemical compounds. J. Exp. Med.
61, 643–656.
(4) Uetrecht, J. (2008) Idiosyncratic drug reactions: past, present,
and future. Chem. Res. Toxicol. 21, 84–92.
(5) Perez-Inestrosa, E., Suau, R., Monta~nez, M. I., Rodriguez, R.,
Mayorga, C., Torres, M. J., and Blanca, M. (2005) Cephalosporin
chemical reactivity and its immunological implications. Curr. Opin.
Allergy Clin. Immunol. 5, 323–330.
(6) Kelkar, P. S., and Li, J. T. (2001) Currents concepts: cephalosporin
allergy. N. Engl. J. Med. 345, 804–809.
(7) Somech, R., Weber, E. A., and Lavi, S. (2009) Evaluation of
immediate allergic reactions to cephalosporins in non-penicillin-allergic
patients. Int. Arch. Allergy Immunol. 150, 205–209.
(8) Batchelor, F. R., Dewdney, J. M., Weston, R. D., and Wheeler,
A. W. (1966) The immunogenity of cephalosporins derivatives and their
cross-reaction with penicillin. Immunology 10, 21–33.
(9) Lowe, G. (1979) β-Lactam Antibiotics, in Comprehensive Organic
Chemistry (Barton, D., and Ollis, D., Eds.) pp 289ꢀ320, Pergamon
Press, Oxford, UK.
(10) O’Callaghan, C. H., Kirby, S. M., Morris, A., Waller, R. E., and
Duncombe, R. E. (1972) Correlation between hydrolysis of the -lactam
bond of the cephalosporin nucleus and expulsion of the 3-substituent.
J. Bacteriol. 110, 988–991.
(11) Waller, R. E. (1973) A method for determining free azide ions
by automatic analysis in the presence of a covalent cephalosporin azide.
Analyst 98, 535–541.
(12) Bundgaard, H. (1975) Chemical studies related to cephalos-
porin allergy. I. Kinetics of aminolysis of cephalosporins and effect of C-3
substituents on β-lactam reactivity. Arch. Pharm. Chem., Sci. Ed.
3, 94–123.
(13) Boyd, D. B., Hermann, R. B., Presti, D. E., and Marsh, M. M.
(1975) Electronic structures of cephalosporins and penicillins. 4.
Modeling acylation by the beta-lactam ring. J. Med. Chem. 18, 408–417.
(14) Boyd, D. B., and Lunn, W. H. W. (1979) Electronic structures
of cephalosporins and penicillins. 9. Departure of a leaving group in
cephalosporins. J. Med. Chem. 22, 778–784.
(15) Faraci, W. S., and Pratt, R. F. (1984) Elimination of a good
leaving group from the 30-position of a cephalosporin need not be
concerted with b-lactam ring opening: TEM-2 β-lactamase-catalized
hydrolysis of pyridine-2-azo-40-(N0,N0-dimethylaniline) cephalosporin
(PADAC) and of cephaloridine. J. Am. Chem. Soc. 106, 1489–1490.
(16) Pratt, R. F., and Faraci, W. S. (1986) Direct observation by 1H-
NMR of cephalosporoate intermediates in aqueous solution during the
hydrazinolysis and β-lactamasecatalized hydrolysis of cephalosporin
with 30 leaving groups: kinetics and equilibria of the 30 elimination
reaction. J. Am. Chem. Soc. 108, 5328–5333.
(31) Torres, M. J., Blanca, M., Fernandez, J., Romano, A., de Weck,
A., Aberer, W., Brockow, K., Pichler, W. J., Demoly, P., ENDA, and
EAACI Interest Group on Drug Hypersensititvy (2003) Diagnosis of
immediate allergic reactions to beta-lactam antibiotics. Allergy, 58,
961ꢀ972.
(32) Blanca, M., Mayorga, C., Sanchez, F., Vega, J. M., Fernandez, J.,
Juarez, C., Suau, R., and Perez, E. (1991) Differences in serum IgE
antibody activity to benzylpenicillin and amoxicillin measured by RAST
in a group of penicillin allergic patients. Allergy 46, 632–638.
(33) Torres, J., Romano, A., Mayorga, C., Carmen, M., Guzman,
A. E., Reche, M., Juarez, C., and Blanca, M. (2001) Diagnostic evaluation
of a large group of patients with immediate allergy to penicillins: the role
of skin testing. Allergy 56, 850–856.
(34) Moreno, F., Blanca, M., Mayorga, C., Terrados, S., Moya, M.,
Pꢀerez, E., Suau, R., Vega, J. M., García, J., Miranda, A., and Carmona,
716
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717

Chemical Research in Toxicology
ARTICLE
M. J. (1995) Studies of the specificities of IgE antibodies found in sera
from subjects with allergic reactions to penicillins. Int. Arch. Allergy
Immunol. 108, 74–81.
(35) Kadereit, D., and Waldmann, H. (2000) Synthesis of character-
istic H-ras lipopeptides by employing noble-metal-, acid-, and reduction-
labile blocking groups. ChemBioChem 1, 200–203.
(36) Baldo, B. A., and Harle, D. G. (1990) Identification of penicillin
allergenic determinants that bind IgE in the sera of patients with
penicillin allergy. Mol. Immunol. 27, 1063–1071.
(37) Potashman, M. H., and Duggan, M. E. (2009) Covalent modifiers:
an orthogonal approach to drug design. J. Med. Chem. 52, 1231–1246.
717
dx.doi.org/10.1021/tx100446g |Chem. Res. Toxicol. 2011, 24, 706–717