On the specificity of reactions catalysed by the antibody H11

Article abstract of DOI:10.1016/S0040-4020(99)01037-6

The substrate specificity and the stereochemical course of the reactions catalysed by the antibody H11 (which was raised to a protein conjugated derivative of the adduct of 1-acetoxy-buta-1,3-diene 1) have been investigated. The antibody shows high selectivity for acetoxybutadiene which it hydrolyses to the corresponding dienol, the major diene component of the cycloaddition reactions observed. However, it tolerates a range of N- alkylmaleimides. The stereochemical course of cycloaddition is shown to produce a significant enantiomeric excess of the 3aR, 4S, 7aR-endo- diastereoisomer by analysis with Mosher's ester derivatives. This study also revealed that H11 is capable of slowly catalysing the hydrolysis of N- alkylmaleimide substrates. The implications for the mechanism of action of H11 are discussed. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(99)01037-6

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 489–495
On the Specificity of Reactions Catalysed by the Antibody H11
Abedawn I. Khalaf,a Sabin Linaza,a Andrew R. Pitt,a William H. Stimsonb
and Colin J. Sucklinga,
*
aDepartment of Pure and Applied Chemistry, University of Strathclyde, Glasgow, Scotland, UK
bDepartment of Immunology, University of Strathclyde, Glasgow, Scotland, UK
Received 26 August 1999; revised 8 November 1999; accepted 25 November 1999
Abstract—The substrate specificity and the stereochemical course of the reactions catalysed by the antibody H11 (which was raised to a
protein conjugated derivative of the adduct of 1-acetoxy-buta-1,3-diene 1) have been investigated. The antibody shows high selectivity for
acetoxybutadiene which it hydrolyses to the corresponding dienol, the major diene component of the cycloaddition reactions observed.
However, it tolerates a range of N-alkylmaleimides. The stereochemical course of cycloaddition is shown to produce a significant enantio-
meric excess of the 3aR, 4S, 7aR-endo-diastereoisomer by analysis with Mosher’s ester derivatives. This study also revealed that H11 is
capable of slowly catalysing the hydrolysis of N-alkylmaleimide substrates. The implications for the mechanism of action of H11 are
discussed. ᭧ 2000 Elsevier Science Ltd. All rights reserved.
In a previous paper we presented evidence consistent with
the course of cycloaddition catalysed by the antibody H11
being through initial rapid hydrolysis of the substrate
acetoxybutadiene followed by trapping of the dienol by an
N-alkylmaleimide bound to the antibody (Scheme 1).1
product, butan-1,3-dien-1-ol. Although the acetoxy adduct
1 was observed, the major cycloaddition product was the
hydroxy adduct 2a. It was possible to rule out cycloaddition
leading directly to a hydroxy adduct 2a away from the anti-
body binding site because the dienol tautomerised affording
Scheme 1. Reactions catalysed by H11 from previous study. 1
Information was based upon studies of chemical modifi-
cation of the antibody, which implied the importance of
carboxylate functional groups in the catalytic mechanism
together with the intrinsic reactivity of the hydrolysis
crotonaldehyde more rapidly under the reaction conditions
than cycloaddition took place. Crotonaldehyde itself does
not undergo cycloaddition with maleimides under these
conditions. With this information to hand, a number of
further questions were raised about the specificity and selec-
tivity of reactions catalysed by H11. In this paper, we
present evidence to show that 1-acetoxybuta-1,3-diene is a
Keywords: antibody H11; acetoxybutadiene; Mosher’s ester.
* Corresponding author. E-mail: c.j.suckling@strath.ac.uk
0040–4020/00/$ - see front matter ᭧ 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(99)01037-6

490
A. I. Khalaf et al. / Tetrahedron 56 (2000) 489–495
Scheme 2. Additional reactions examined in this study.
relatively specific substrate as a diene precursor whereas a
variety of N-alkylmaleimides are accepted as dienophiles.
We also show that the cycloaddition reaction occurs with
significant stereoselectivity. Evidence emerged from the
study of the stereochemical course of the reaction that
H11 is also able to hydrolyse maleimides slowly to
maleamic acids but does not hydrolyse the imide ring of
cycloadducts.
Reaction conditions were established in which all of these
compounds and H11 were soluble. Typically, millimolar
concentrations of substrates were compatible with micro-
molar concentrations of H11 in 3:1 phosphate buffered
saline (PBS)/acetonitrile. Initial investigations were carried
out using UV absorption analysis at 300 nm, but this was
found to be unsatisfactory for the cases in question due to
the low sensitivity of the method. Reactions were therefore
followed by removing samples from control reactions and
reactions containing H11 at set times, freezing the solutions
in liquid nitrogen, and subsequently measuring the concen-
tration of products by HPLC. In the case of ethoxybutadiene
3, no difference in the rate of reaction between control and
H11-containing samples was observed. To study 1-acetoxy-
hexa-2,4-diene 4, it was necessary to raise the reaction
temperature from 25 to 37ЊC in order to ensure solubility
of the reactants. No catalysis of cycloaddition was observed.
Moreover no evidence for other products such as hydrolysis
of the acetate ester was found. This result demonstrates that
H11 is not a non-specific esterase. With carbamate 5 as
substrate, initial experiments indicated that catalysis might
be occurring. However, it was not possible to quantify this
observation due to the small size of the effect. We therefore
Substrate Specificity
Previous studies have shown that acetoxybutadiene1 and
N-ethyl- and N-benzylmaleimide2 are substrates for
cycloaddition reactions catalysed by H11. In view of the
unexpected preponderance of the hydroxy adduct 2a in
the reaction products it was of interest to establish whether
any other dienes with electron donating substituents would
be accepted. Accordingly, we examined 2-ethoxybuta-1,3-
diene 3, 1-acetoxyhexa-2,4-diene 4, and 1-ethoxycarbonyl-
aminobuta-1,3-diene 5 (Scheme 2). The last of these three
has previously been used as a substrate in studies of
cycloaddition catalysed by antibodies.3
Scheme 3. Reactions and structures of compounds used to determine the stereochemical course of reactions mediated by H11.

A. I. Khalaf et al. / Tetrahedron 56 (2000) 489–495
491
conclude that H11 is specific within the range of substrates
tested for acetoxybutadiene presumably because of its size
and the increased reactivity in cycloaddition of the dienol
produced by hydrolysis (Scheme 1).
effective with H11 normally afford the endo adduct and thus
a pair of enantiomeric products at equal abundance would
be expected from the reaction of N-ethylmaleimide and
acetoxybutadiene in the absence of antibody. A change in
this ratio from 1:1 would imply some degree of stereoselec-
tivity in the reactions catalysed by H11. We were unable to
separate the isomers of the hydroxy adduct 2 by chiral chro-
matography. The determination of the stereochemical
course of the reactions catalysed by H11 therefore relied
upon the preparation of a derivative of the alcohol of the
hydroxy adduct 2. It was first necessary to establish reaction
conditions for hydrolysis of the acetoxy adduct and its
subsequent conversion into a derivative with a chiral
reagent.
The ability of H11 to accept both N-ethyl- and N-benzyl-
maleimides as substrates has already been noted. It might
therefore be anticipated that other N-alkylmaleimides would
be substrates. N-Phenylmaleimide was shown to undergo
cycloaddition with acetoxybutadiene in the presence of
H11 affording both the corresponding acetoxy and hydroxy
adducts (1b and 2b, RˆPh). Kinetic measurements were not
undertaken with this substrate combination but the for-
mation of the hydroxy adduct 2b is evidence that the
cycloaddition takes place when bound to the antibody.1
The N-alkyl substituent in the hapten used to generate
H11 was n-butanoyl.2 It was therefore possible that a
maleimide with a longer alkyl chain than ethyl might bind
more tightly to the antibody. The kinetics of the reaction
between N-butylmaleimide and acetoxybutadiene were
investigated by measuring the formation of products at
37ЊC using HPLC analysis of samples removed from the
reaction mixture and frozen immediately. Product was
quantified as the total of the respective acetoxy and hydroxy
adducts 1c and 2c (Rˆn-Bu) minus the quantity of acetoxy
adduct produced by the control reaction. Analysis of the
data gave the kinetic constants KMˆ6:6 mM;
vmaxˆ 9×10Ϫ6 minϪ1: The KM value of 6.6 mM is of the
same order as that obtained for N-ethylmaleimide
(8.3 mM) and no additional affinity can therefore be identi-
fied. It can be concluded that the N-alkyl substituent does
not play a major role in binding during catalytically signifi-
cant events mediated by H11. By comparison, the kinetic
constants for acetoxybutadiene1 ꢀkcat=ksolvˆ7000 and
KMˆ0:027 mM at 30ЊC) imply significant affinity of acet-
oxybutadiene for the antibody.
The development of a suitable method for analysis was not
straightforward because chemical hydrolysis of acetoxy
adducts 1 was always accompanied by decomposition,
presumably due to retro-aldol opening of the six-membered
ring. Chiral chromatographic separations and 1H NMR
experiments using chiral shift reagents were also unsuccess-
ful. Bearing in mind that the major product in reactions
catalysed by H11 is in any case a hydroxy adduct (e.g. 2),
it should be possible to determine the stereochemical course
of the reaction by quantitative derivatisation of the crude
reaction products. In order to clearly identify the relevant
products, 19F NMR on the Mosher’s ester derivatives was
selected as the analytical method.4 Mosher’s acid was
converted into the acid chloride (which contained a small
amount of the symmetrical anhydride) and the crude acid
chloride used to convert the hydroxy adducts into Mosher’s
esters. With the mixture produced by derivatising a sample
of the hydroxy adduct 2 prepared in the absence of H11, two
distinct 19F resonances were observed (Ϫ72.09 and
Ϫ72.40 ppm, Table 1). The constitution of the products of
this derivatisation was confirmed by mass spectrometry.
In the context of the scope of hydrolytic reactions, N-ethyl-
and N-phenylmaleimides were tested individually as
substrates for hydrolysis reactions catalysed by H11.
Surprisingly there was some measurable hydrolysis leading
to maleamic amides, 10a and 10b (Scheme 3), which were
identified by HPLC in comparison with a synthetic sample.
On the other hand, when any of the bicyclic imides was
incubated in the presence of H11, there was no evidence
for the hydrolysis of the imide ring.
A preparative incubation of acetoxybutadiene, N-ethyl-
maleimide, and H11 was carried out in the normal solvent
mixture (PBS pH 7.4:acetonitrile, 3:1 v/v), the organic
products extracted with dichloromethane and derivatised
with Mosher’s acid chloride. The 19F NMR of this sample
(Table 1) showed resonances corresponding with unreacted
starting material and the bicyclic adduct 2, the two reso-
nances of which were present (Ϫ72.08 and Ϫ72.40 ppm).
In addition there was a pair of resonances (Ϫ72.1 and
Ϫ72.26 ppm) and a single resonance (Ϫ71.92 ppm).
Bearing in mind the observation of hydrolysis of
maleimides by H11, a reasonable possibility was that
these resonances arose from derivatisation of monocyclic
adducts of maleamic acids (Scheme 2). In such a case,
two regioisomers of the maleamic acid might be expected,
Stereochemical Course of the Reactions Catalysed
by H11
Cycloadditions between maleimides and dienes of the type
Table 1. Observed resonances from Mosher’s ester derivatives
Bicyclic adduct
Monocyclic adduct
H11 Incubation
Relative intensity
Starting Mosher’s acid
Ϫ71.70
Ϫ71.70
Ϫ71.70
Ϫ71.87?
Ϫ71.95ء (minor)
Ϫ72.04ء (minor)
Ϫ71.92ء
Not well resolved
Ϫ72.0900
Ϫ72.4000
Ϫ72.0800
2.1
1.0
5.4
3.8
Ϫ72.13^ (major)
Ϫ72.25^ (major)
Ϫ72.10^
Ϫ72.26^
Ϫ72.4000

492
A. I. Khalaf et al. / Tetrahedron 56 (2000) 489–495
Scheme 4. Summary of reactions catalysed by H11 from this and previous study.1
each of which could exist as two enantiomers assuming the
normal cis-stereochemical course of a cycloaddition. To
identify these resonances conclusively, we undertook the
synthesis of the maleamic acids and their subsequent
derivatisation to form Mosher’s esters.
also the more intense. These data confirm the origin of the
resonances observed in the derivatives of products from
H11 and are consistent with the ability of H11 to act as an
imide hydrolase. From measurements of the intensity of the
resonances, an estimate of the enantiomeric excess of
approximately 30% can be estimated for the bicyclic adduct
2 present in the mixture of products and about 70% for the
major monocyclic adduct 10. The difference between the
two estimates of enantiomeric excess is consistent with
the interpretation shown in Scheme 4 that once bound to
H11, a substrate molecule follows only one of the possible
pathways leading to product. Either the bicyclic hydroxy
adduct 2 is formed or the monocyclic adduct 10 but
H11 apparently does not hydrolyse 2 to give 10. This is
consistent with the observations reported in the previous
section.
Acetoxybutadiene was reacted with maleic anhydride to
form the cycloadduct 11, the configuration of which was
unambiguously assigned as endo,cis by X-ray crystallo-
graphy.5 Treatment of the adduct with ethylamine afforded
a mixture of enantiomeric regioisomers 10a and 10b
without cleavage of the acetate ester. The acetate was
hydrolysed and the resulting mixture of maleamic acids
derivatised as before with Mosher’s acid chloride.
The 19F resonances observed in the spectrum are shown in
Table 1. The resonances assigned to the two regioisomers of
the monocyclic adduct with the amide derivative of each of
the two carboxylic acids are marked ء and ^. These reso-
nances correspond to those observed in the derivatives of
products of reactions in the presence of H11. As shown in
Table 1,00the bicyclic adduct showed one pair of resonances
marked in which the higher field resonance was also more
intense. Similarly the major monocyclic adduct showed a
pair of resonances in which the high field resonance was
Mosher’s ester derivatives also give the opportunity to
determine the absolute configuration of a compound.4 In
this case, the environment of the 19F nuclei can be assessed
by molecular modelling. Using the Cache molecular
modelling software (Oxford Molecular), the structures of
the enantiomers of the Mosher’s ester derivatives of 13a
and 13b were built and energy minimised (Fig. 1). These
calculations showed that the trifluoromethyl group of one
Figure 1. Calculated structures of the epimeric Mosher’s ester derivatives 13a and 13b. The alkene hydrogen atoms and the fluorine atoms are shaded.

A. I. Khalaf et al. / Tetrahedron 56 (2000) 489–495
493
enantiomer was within range of the anisotropic influence of
the alkene of the six-membered ring of the adduct. Having
used the R-(ϩ) isomer of Mosher’s acid to prepare the
derivatives, it can therefore be deduced that the major
isomer produced by H11 has the probable configuration
3aR, 4S, 7aR (13a) and the minor isomer the 3aS, 4R, 7aS
configuration (13b).
1.07 (3H, t, Jˆ7.2 Hz, CH2CH3), 1.46 (3H, d, Jˆ7.5 Hz,
C7–CH3), 2.09 (3H, s, CH3CO), 2.43 (1H, q, Jˆ7.5 Hz,
H7), 2.62 (1H, m, H4), 3.05 (1H, dd, Jˆ5.3 and 2.2 Hz,
H7a), 3.22 (1H, dd, Jˆ5.7 and 2.6 Hz, H3a), 3.31 (2H, q,
Jˆ7.2 Hz, CH2CH3), 4.51 (1H, dd, Jˆ11.1 and 8.2 Hz,
CH2OAc), 4.68 (1H, dd, Jˆ11.1 and 8.2 Hz, CH2OAc),
5.78 (2H, bd s, CHyCH). nmax/cmϪ1 [KBr] 1855, 1784,
1736, 1379, 1243, 1050. EI-MS Found: m/z 265.1309.
C14H19NO4 requires 265.1314.
Summary of Selectivity of H11
Rate of hydrolysis of N-ethylmaleimide catalysed by H11
The above results are consistent with the catalytic activity of
H11 as promoting cycloaddition chiefly through its ability to
act as a hydrolase. The hydrolysis products are evidently
partially captured whilst still bound to the protein leading
to a measurable degree of asymmetric induction. The obser-
vations presented here and in previous papers1,2 show that
hydrolysed products do not take part in cycloaddition
reactions under the conditions of the H11 mediated reac-
tions. It is ironic that H11 should turn out to be able to
catalyse the formation of chiral maleamic acid derivatives
directly because the formation of such products was
originally one of the grounds for selecting the target
molecule 1.2 The observation of specific hydrolysis at the
UV method. Solutions of N-ethylmaleimide in PBS–aceto-
nitrile (3 mL) at concentrations of 1.33, 2.66, 4.0 and
5.33 mM were prepared and incubated at 25ЊC in a UV
cuvette measuring the absorbance at 300 nm (substrate).
Concentrations of substrate were determined from a cali-
bration curve ꢀeˆ533; rˆ0:999†: The reaction was followed
by monitoring the loss of substrate at 300 nm. The pseudo-
first order rate constant was obtained from a plot of
initial rate against concentration and the rate constant
obtained from the slope, kˆ2:8×10Ϫ4 minϪ1 ꢀrˆ0:89†:
HPLC method. Solutions of N-ethylmaleimide were
prepared and incubated as for the UV method. Samples
were removed at intervals over 3 h and immediately frozen
in liquid nitrogen. Subsequently, the solutions were thawed
and 10 mL samples were analysed by HPLC (ODS2, eluting
with acetonitrile–water 1:1 v/v). Similar experiments were
carried out with N-phenylmaleimide.
6
surface of an antibody is well-precedented even in cases
where such an activity was not sought.7 It has been pointed
out that such reactions are to be expected if a reasonable
degree of binding exists because of the probability that a
charged amino acid side chain group will be found adjacent
to the binding site on any protein8 including antibodies.9
Such charged groups promote general acid or general base
catalysis within a water-accessible region. Both enol ether
hydrolysis10 and imide hydrolysis11 by antibodies have been
described previously. We thus summarise the principal
reactions catalysed by H11 in Scheme 4. The availability
of relevant catalytic functional groups implicated by this
reaction course and the chemical modification studies
reported previously12 can be seen in the CDR regions
identified in the sequence determined for the antibody
H11.10 The details of the determination of the sequence,
the construction of a computer model of the antibody
H11, and its relationship to the mechanisms of the
reactions catalysed will be described in a forthcoming
paper.
Investigation of reactions of diverse substrates mediated
by H11
Dienes. UV method: Solutions of H11 (5 mM in antibody),
N-ethylmaleimide (1.3 mM), and the appropriate diene
(1.2–2.4 mM) were prepared in PBS–acetonitrile (3 ml)
and the reactions followed by UV spectroscopy at 300 nm.
A control reaction containing all reactants other than H11
was also followed. HPLC method: For the control reaction
the appropriate diene (3.5 mmol) and N-ethylmaleimide
(4 mmol) were dissolved in PBS–acetonitrile (3 mL) at
37ЊC and samples (20 mL) removed at intervals over 2 h.
For the test reaction, H11 (0.27 mg) was also present in the
solution.
Experimental
N-Butylmaleimide. The HPLC method was used to
measure the rate of addition of 1-acetoxybuta-1,3-diene
and N-butylmaleimide in the presence and absence of H11
in PBS–acetonitrile as solvent at 37ЊC. The quantities used
were: diene (1.8 mg, 16 mmol); N-butylmaleimide (1 mg,
6.5 mmol, 2.2 mM; 0.75 mg, 4.9 mmol, 1.6 mM; 0.5 mg,
3.3 mmol, 1.1 mM); H11 (0.27 mg). The reactions were
followed over a 46 min period during which samples
(20 mL) were removed and immediately frozen in liquid
nitrogen. The eluent for HPLC analysis was water–
acetonitrile (13:7 v/v). The initial rates determined for
In experiments with H11, PBS–acetonitrile refers to a
solution of phosphate buffered saline, pH 7.4, 10 mM and
acetonitrile (3:1 v/v). For all kinetic experiments, cali-
bration curves for HPLC were obtained for the respective
substrates and products.
4-Acetoxymethyl-2-ethyl-7-methyl-3a,4,7,7a-tetrahydro-
isoindole-1,3-dione (7). 1-Acetoxyhexan-2,4-diene (89.6 mg,
0.8 mmol) and N-ethylmaleimide (115 mg, 1.0 mmol) were
dissolved in acetonitrile (25 mL) and the solution heated
under reflux for 24 h. After evaporation of the solvent the
crude product was purified by flash chromatography on
silica eluting with ethyl acetate/hexane (1:1 v/v); the
required adduct was obtained as a non-crystalline solid
with no distinct melting point (134 mg, 63%). dH (CDCl3)
each concentration were 2:36×10Ϫ6
;
1:67×10Ϫ6
;
1:31×10Ϫ6 M minϪ1; respectively, from which the kinetic
parameters KMˆ6:6 × 10Ϫ3 M; and kcatˆ15 minϪ1 were
determined by a computational model of the Lineweaver
Burk plot.

494
A. I. Khalaf et al. / Tetrahedron 56 (2000) 489–495
Synthesis of Mosher’s ester derivatives (14a and b) of the
bicyclic hydroxyadduct (2a). R(ϩ)-a-methoxy-a-trifluoro-
methylphenylacetic acid (Mosher’s acid, 0.812 g, 3.5 mmol)
was dissolved in thionyl chloride and the mixture heated
under reflux for 48 h. The excess thionyl chloride was
evaporated under reduced pressure and the residual oil
kept dry, cold, and under nitrogen until required for use.
The hydroxy adduct 2 (28 mg, 0.15 mmol) and the acid
chloride (38 mg, 0.15 mmol) were dissolved in dry
dichloromethane (5 drops) and pyridine (5 drops) and
allowed to stand under nitrogen for 12 h. The reaction
mixture was then diluted with ether and the solution washed
with dil. hydrochloric acid, aqueous sodium bicarbonate
solution and brine. The organic solution was dried
(MgSO4) and the solvent evaporated to afford the required
mixture of Mosher’s esters as a colourless non-crystalline
solid (41 mg, 67%). dH (CDCl3) 0.99–1.02 (overlapping t,
3H, NCH2CH3), 2.36 (1H, m, H4), 2.67 (1H, m, H6), 3.06
(1H, m, H1), 3.43–3.45 (3H, 2×s; OCH3), 3.5 (2H, m,
NCH2), 5.68–5.76 (1H, t Jˆ5.0 Hz), 6.10–6.35 (2H,
m, CHyCH), 7.38 (5H, s, C6H5). dc (CDCl3) 12.88
(NCH2CH3), 21.93 and 22.16 (NCH2), 33.77 and 33.88
(C5), 36.39 (C6), 43.45 and 43.55 (C1), 55.43 and 55.63
(OCH3), 67.57 and 68.05 (C2), 121.18 (C[CF3]OMe),
124.77, 125.43, 127.39 and 127.47, 128.55 and 128.60,
129.83 and 129.89, 131.28 (C6H5), 126.38 (C4), 134.68
(C3), 165.65 (CF3), 174.84 (CO2–), 178.73 and 178.83
(CON). dF (CDCl3) Ϫ72.09 and Ϫ72.40. LREI-MS
Found m/z 411, C20H21F3O5N requires 411, FABϩ
found 412.1.
product which was purified by column chromatography on
silica using methanol/ethyl acetate 1:3 as eluant. The
products were obtained as a colourless oil (242 mg, 39%).
dH (CDCl3) 1.13–1.19 (3H, t, Jˆ5.1 Hz, CH2CH3); 2.04 and
2.05 (3H, 2×s, CH3CO2); 2.14–2.20 (1H, m); 2.86–3.18
(3H, m); 3.29–3.40 (2H, m, CH2CH3); 5.59–6.18 (3H, m);
6.55 (1H, s, CONH). nmax/cmϪ1 3296, 1745, 1723, 1634,
1587. HREI-MS: Found: m/z 255.11135 C12H17NO5
requires 255.11067.
Methyl 3-hydroxy-5(or 4) (carboxamidoethyl)cyclohex-
1-ene-4(or 5)-carboxylate. 3-Acetoxy-5(or 4)(carboxami-
doethyl)cyclohex-1-ene-4(or 5)-carboxylic acid (204 mg,
0.800 mmol) was dissolved in methanol (10 mL) to which
was added conc. HCl (300 ml) at room temperature with
stirring. The stirring was continued at room temperature
overnight. The solvent was removed under reduced pressure
and the product purified by silica gel column chroma-
tography eluting with methanol/ethyl acetate (1:3 v/v) to
give the products as a colourless oil (181 mg, 99%). dH
(CDCl3) 1.10–1.22 (3H, t, Jˆ5.0 Hz CH2CH3); 2.38–2.47
(2H, m); 2.96–2.98(1H, m); 3.27–3.38 (2H, m, CH2CH3);
3.72 and 3.73 (3H, 2×s; CO2CH3); 4.32–4.39 (2H, m);
5.77–5.97 (3H, m); 7.10 (1H, s, CONH). nmax/cmϪ1 3311,
1740, 1651, 1631, 1546. HREI-MS: Found: m/z 227.11709
C11H17O4N requires 227.11576.
3-Hydroxy-5(or 4)(carboxamidoethyl)cyclohex-1-ene-4(or
5)-carboxylic acid (10a and b). Methyl 3-hydroxy-
5(or 4)(carboxamidoethyl)cyclohex-1-ene-4(or 5)-carboxy-
late (170 mg, 0.748 mmol) was suspended in aqueous
sodium hydroxide (5 mL, 0.1 M) for 1 h at room tempera-
ture then extracted with dichloromethane (25 mL). The
water layer was cooled to 0ЊC then acidified with conc.
HCl, followed by extraction with ethyl acetate ꢀ2×25 mL†:
The organic extract was dried (Na2SO4), and the solvent
removed under reduced pressure to give the required
products as a colourless oil (40 mg, 25%). dH (CDCl3)
1.09–1.18 (3H, t, Jˆ5.0 Hz CH2CH3); 2.31–2.58 (2H,
m); 2.97–3.13 (2H, m); 3.23–3.30 (2H, m, CH2CH3);
4.47–4.52 (1H, m); 5.76–5.88 (2H, m); 7.62–7.65
(1H, t, CONH). nmax/cmϪ1 3311, 1723, 1634, 1587.
HREI-MS: Found m/z 196.09631 C10H14O3N requires
196.09737 [MϪH].
Synthesis of adducts and derivatives of maleamic acids
Acetic acid 1,3-dioxo-1,3,3a,4,7,7a-hexahydro-isobenzo-
furan-4-yl ester (11). 1-Acetoxy-1,3-butadiene (2.187 g,
19.50 mmol) was added at room temperature to a suspen-
sion of maleic anhydride (1.750 g, 17.84 mmol) in benzene
(40 mL) with stirring. The stirring was continued at room
temperature overnight. Solvent was removed under reduced
pressure and the crude product was purified by chroma-
tography on silica gel using ethyl acetate/n-hexane (1:3 v/v).
The required cycloadduct was obtained as a white crys-
talline solid (2.135 g, 52%), mp 53–56ЊC. [lit.13a mp.56–
57.5ЊC, lit.13b mp 54–55ЊC, lit.13c mp 56–59ЊC] dH
(CDCl3) 2.05 (3H, s, CH3CO2); 2.45–2.53 (2H, m, H-6-a
and -b); 3.43–3.49 (1H, ddd, Jˆ10, 8, and 4 Hz, H-1);
3.57–3.61 (1H, dd, Jˆ10 and 5 Hz, H-2); 5.41–5.44 (1H,
m, H-3); 6.09–6.16 (2H, m, H-4 and H-5). nmax/cmϪ1 1855,
1784, 1736.
5(or 4)(Carboxamidoethyl)cyclohex-1-ene-4(or 5)-car-
boxylic acid [R(ϩ)-a-methoxy-a-trifluoromethyl phenyl-
acetic acid] ester (12a and b). Mosher’s acid (102 mg,
0.404 mmol) was dissolved in thionyl chloride (1 mL) and
the reaction mixture heated under reflux for 48 h. Excess
thionyl chloride was removed under reduced pressure at
40ЊC. The acid chloride was dissolved in dichloromethane
(1 mL, dry) then added to 3-hydroxy-5(or 4)(carboxamido-
ethyl)cyclohex-1-ene-4(or 5)-carboxylic acid at room
temperature. Pyridine (100 mL, dry) was added to the
reaction mixture with stirring at room temperature. The
reaction mixture was left stirring at room temperature for
48 h. Dilution with ether (25 mL) was then followed by
addition of dilute HCl (5 mL). The mixture was extracted
then washed with water, dried (Na2SO4) and ether was then
removed under reduced pressure at room temperature to
give the required products as a pale yellow oil (37 mg). dF
see Table 1.
3-Acetoxy-5(or 4)(carboxamidoethyl)cyclohex-1-ene-4(or
5)-carboxylic acid. 3-Acetoxy-cyclohex-4-ene-1,2-dicar-
boxylic acid anhydride (506 mg, 2.407 mmol) was
dissolved in THF (20 mL, dry) to which was added ethyl-
amine (2.5 mL, 2.0 M solution in THF) with stirring at room
temperature. Stirring was continued overnight at room
temperature and then the solvent was removed under
reduced pressure. The crude product was dissolved in
dichloromethane then extracted with sodium hydrogen car-
bonate. The aqueous layer was collected, cooled to 0ЊC and
then acidified with conc. HCl. The acidified solution was
extracted with dichloromethane, dried and then the solvent
was removed under reduced pressure to give the crude

A. I. Khalaf et al. / Tetrahedron 56 (2000) 489–495
495
3. Braisted, A. C.; Schultz, P. G. J. Am. Chem. Soc. 1990, 112,
7430–7431.
Determination of stereochemical course of reaction
catalysed by H11
4. Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34,
2543–2549.
5. Kennedy, A. R.; Khalf, A. I.; Suckling, C. J. Acta Crystallogr.
C55, IUC9900072.
6. (a) Wade, H.; Scanlan, T. S. Ann. Rev. Biophys. Biomol. Struct.
1997, 26, 461–493. (b) Gigant, B.; Charbonnier, J. B.; Golinelli-
pimpaneau, B.; Zemel, R. R.; Eshhar, Z.; Green, B. S.; Knossow,
M. Eur. J. Biochem. 1997, 246, 471–476.
7. (a) Khalaf, A. I.; Proctor, G. R.; Suckling, C. J.; Bence, L. M.;
Irvine, J. I.; Stimson, W. H. J. Chem. Soc., Perkin Trans. 1, 1992,
1475–1481. (b) Stimson, W. H.; Bence, L. M.; Irvine, J.; Khalaf,
A. I.; Proctor, G. R.; Tedford, M. C.; Suckling, C. J. J. Clin. Lab.
Immunol. 1994, 43, 167–175.
1-Acetoxybuta-1,3-diene (14.8 mg, 0.132 mmol), N-ethyl-
maleimide (8 mg, 0.064 mmol), and H11 (1.62 mg) were
dissolved in PBS–acetonitrile (24 mL) and the solution
stirred at 37ЊC for 24 h. A second experiment was run for
12 h. At the end of each reaction, the solution was frozen in
liquid nitrogen and the solvent lyophilised. The organic
product was extracted from the residue with dichloro-
methane (15.7 and 29.4 mg product) and converted into
the Mosher’s ester derivatives using the method described
above. The products were analysed by 19F NMR and the
observed spectra are reported in Table 1. Evidence for the
presence of the maleamic acid derivatives was obtained
from the mass spectra (FABϩ) found m/z 430 (Mϩϩ1)
and 452 (Mϩϩ23) as required by C20H23F3O6NϩNa.
8. Hollfelder, F.; Kirby, A. J.; Tawfik, D. S. Nature 1996, 383,
60–63.
9. Riva, S.; Mendozza, S.; Carrea, M.; Chattopadhyay, G.;
Tramontano, P. Appl. Biochem. Biotechnol. 1998, 75, 33–44.
10. Brooks, L.; Suckling, C. J.; Stimson, W. H. Biochem. Soc.
Trans. 1996, 24, 313S. EMBL/GENBANK L41672, L41689,
L41728.
Acknowledgements
We thank the Basque Government for financial support
(S.L.).
11. Redmond, J. L.; Jahangiri, G. K.; Stoudt, C.; Lerner, R. A.
J. Am. Chem. Soc. 1993, 115, 3909–3917.
12. Liotta, L. J.; Benkovic, P. A.; Miller, G. P.; Benkovic, S. J.
J. Am. Chem. Soc. 1993, 115, 350–351.
References
13. (a) McDonald, E.; Suksamran, A.; Wylie, R. D. J. Chem. Soc.,
Perkin Trans. 1, 1979, 1893–1900. (b) Dolby, L. J. J. Org. Chem.
1962, 27, 2971–2975. (c) Ubukata, M.; Sakamura, S. Tetrahedron
Lett. 1977, 3473–3476.
1. Pitt, A. R.; Stimson, W. H.; Suckling, C. J.; Tellado, J.-M.;
Vazzana, C. Isr. J. Chem. 1996, 36, 171–175.
2. Suckling, C. J.; Tedford, M. C.; Bence, L.; Irvine, J.; Stimson,
W. H. J. Chem. Soc., Perkin Trans. 1 1993, 1925–1929.