Acyloxymethyl as a drug protecting group. Part 7: Tertiary sulfonamidomethyl ester prodrugs of benzylpenicillin: Chemical hydrolysis and anti-bacterial activity

Article abstract of DOI:10.1016/S0968-0896(00)00099-7

Tertiary sulfonamidomethyl esters of benzylpenicillin (4) were synthesised and evaluated as a new class of potential prodrugs for β-lactam antibiotics. Their hydrolysis in aqueous buffers was studied by HPLC and reveal a U-shaped pH-rate profile with a pH-independent process extending from ca. pH 2 to ca. pH 10. This pathway is characterised by kinetic data that are consistent with a unimolecular mechanism involving rate-limiting iminium ion formation and penicillinoate expulsion. Benzylpenicillin and the corresponding sulfonamide are the ultimate products detected and isolated, indicating that β-lactam ring opening is much slower than ester hydrolysis. As expected from the high reactivity, benzylpenicillin esters (4) displayed similar in vitro antibacterial activity to benzylpenicillin itself. Compared to the benzylpenicillin derivatives, sulfonamidomethyl esters of benzoic, clofibric and valproic acids display a much higher stability, giving rise to a Bronsted β(lg) value of -0.96 and suggesting that tertiary sulfonamidomethyl esters may be useful prodrugs for carboxylic acid drugs with pK(a)>4. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00099-7

Bioorganic & Medicinal Chemistry 8 (2000) 1629±1636
Acyloxymethyl as a Drug Protecting Group.
Part 7: Tertiary Sulfonamidomethyl Ester Prodrugs of
Benzylpenicillin: Chemical Hydrolysis and Anti-Bacterial Activity
Jim Iley,a,* Helena Barroso,b Rui Moreira,c Francisca Lopesc and Teresa Calheirosc
aChemistry Department, The Open University, Milton Keynes, MK7 6AA, UK
bDepartment of Microbiology, Faculty of Pharmacy, University of Lisbon, 1600-83 Lisbon, Portugal
cCECF, Faculty of Pharmacy, University of Lisbon, Av. ForcËas Armadas, 1600-83 Lisbon, Portugal
Received 8 December 1999; accepted 7 March 2000
AbstractÐTertiary sulfonamidomethyl esters of benzylpenicillin (4) were synthesised and evaluated as a new class of potential
prodrugs for b-lactam antibiotics. Their hydrolysis in aqueous bu€ers was studied by HPLC and reveal a U-shaped pH±rate pro®le
with a pH-independent process extending from ca. pH 2 to ca. pH 10. This pathway is characterised by kinetic data that are con-
sistent with a unimolecular mechanism involving rate-limiting iminium ion formation and penicillinoate expulsion. Benzylpenicillin
and the corresponding sulfonamide are the ultimate products detected and isolated, indicating that b-lactam ring opening is much
slower than ester hydrolysis. As expected from the high reactivity, benzylpenicillin esters (4) displayed similar in vitro antibacterial
activity to benzylpenicillin itself. Compared to the benzylpenicillin derivatives, sulfonamidomethyl esters of benzoic, clo®bric and
valproic acids display a much higher stability, giving rise to a Brùnsted blg value of 0.96 and suggesting that tertiary sulfonami-
domethyl esters may be useful prodrugs for carboxylic acid drugs with pKa>4. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
hydrolysis,10 therefore contributing to the reduction of
in vitro stability.
Many drugs bearing a carboxylic acid group, including
penicillins and cephalosporins, display limited oral
bioavailability.1 Transformation of the drug molecule
into a prodrug, the main requisite of which is the in vivo
quantitative release either chemically or enzymatically
of the parent drug, has proved a useful approach for the
improvement of physicochemical properties and oral
drug delivery.2 Simple esters of b-lactam antibiotics are
too stable in vivo to be used as prodrugs.3,4 A more
suitable strategy is to prepare (acyloxy)alkyl derivatives
1, for which hydrolysis to the parent drug is enzymati-
cally triggered.5,6 However, some (acyloxy)alkyl esters
of neutral b-lactam antibiotics are also poorly active
orally because of their low aqueous solubility.7 The
introduction of a basic a-amino group into the acyl pro-
moiety increases the water-solubility but generally
reduces in vitro stability.8,9 Unfortunately, ester-
i®cation at the C-3 carboxylic acid group of peni-
cillins generally increases the rate of alkaline b-lactam
Recently, we reported that amidomethylation to yield
tertiary N-acyloxymethylamides 2, is a potential
approach for prodrug development of drugs that con-
13
tain free carboxylic acid groups.11
Speci®cally, ben-
zylpenicilloates 2 (R3=benzylpenicillin, 3) undergo
spontaneous hydrolysis without su€ering b-lactam ring
opening to any detectable extent.12 This is a clear
advantage over the majority of a-acyloxyalkyl double
prodrugs 1 of b-lactam antibiotics, which can su€er b-
lactam ring opening depending on the substrate and the
pH of the reaction.14 However, due to the fact that
compounds 2 contain a good carboxylate leaving group
of pKa ca. 3, they hydrolyse very rapidly12 making
pharmaceutical formulation a more dicult task. As the
reactivity of amidomethyl esters 2 is largely depressed
by electron withdrawing substitutents in the amide
pro-moiety, we anticipated that using a sulfonamide
pro-moiety would lead to substantially more stable
prodrugs. We now report upon the e€ect of the sulfo-
namide pro-moiety on the reactivity and mechanism
of chemical hydrolysis of tertiary sulfonamidomethyl
benzylpenicilloates 4, as well as on the extent of
*Corresponding author. Tel.: +44-908-652713; fax: +44-908-653744;
e-mail: j.n.iley@open.ac.uk
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00099-7

1630
J. Iley et al. / Bioorg. Med. Chem. 8 (2000) 1629±1636
competitive b-lactam ring opening. This study is direc-
ted toward evaluating the suitability of sulfonamido-
methyl esters as prodrugs for b-lactam antibiotics and
the data are compared with the sulfonamidomethyl
esters of benzoic, clo®bric, and valproic acids.
exhibit a characteristic singlet at ca. d 5.5 ppm due to
the NCH2O group. For the penicilloates 4a±g, the
NCH2O signal appears as two doublets, re¯ecting the
diastereotopic nature of the methylene protons (2J=11±
12 Hz).
Chemistry
Results and Discussion
The strategy for the synthesis of tertiary N-acyloxy-
methylsulfonamides 4a±k involves the preparation of N-
chloromethylsulfonamide intermediates 5 by reaction of
the appropriate secondary sulfonamide with chloro-
trimethylsilane and paraformaldehyde (Scheme 1).
This reaction allows the synthesis of compounds 5 in
excellent yield, even when the starting material is an
N-arylsulfonamide. Reaction of intermediates 5 with the
sodium salt of the corresponding carboxylic acids
a€orded 4 in ca. 30% yield. The same approach was
used with cefotaxime (6). However, in this case, the
reaction of N-chloromethyl-N-phenylmethanesulfon-
amide with 6 yielded a mixture of the Á3 and Á2 iso-
mers, 7a and 7b, respectively (Scheme 2). These two
Kinetics of hydrolysis and pH±rate pro®les
The pH±rate pro®les for the hydrolysis of sulfonamido-
methyl esters of benzylpenicillin, e.g., 4g, exhibit a
broad U-shape (Fig. 1), indicative of the presence of
acid-catalysed, base-catalysed and pH-independent
processes (eq (1)). These pro®les are analogous to those
of simple acyloxymethylsulfonamides,17 e.g., 4k, and
similar pH±rate pro®les have been previously described
for their amidomethyl counterparts.11 13
‡
‡
kobs ˆ ko ‡ kH ‰H Š ‡ kOH ‰OH Š
ꢀ1†
A striking feature of these pH±rate pro®les is the unu-
sually broad plateau extending from ca. pH 2 to ca. pH
10, indicating that the pH-independent hydrolysis is the
dominant process at all physiologically relevant pH
values. This contrasts markedly with simple ester
hydrolysis, for which speci®c-base catalysis usually
starts at ca. pH 6±7.18 Moreover, the pH±rate pro®les of
the penicilloates observed here di€er signi®cantly from
those of the hydrolysis of penicillins, which either have
no signi®cant spontaneous reaction (e.g., benzylpeni-
cillin19) or have only small plateaus around neutral
pH (e.g., phenethicillin20). The apparent ®rst-order rate
1
isomers could be easily distinguished in the H NMR
spectra.15,16 While the Á3 isomer presented two doublets
at d 3.47 and 3.59 ppm (2J=18 Hz) corresponding to
the C2-Hs, the Á2 isomer presented two singlets at 5.32
and 5.33 ppm and a broad singlet at 6.55 ppm, corre-
sponding to the C4-H and C2-H, respectively. All
attempts to separate the two isomers have so far resul-
ted in the decomposition of products.
The structure of compounds 4 follows from their spec-
troscopic and analytical data. The spectra of esters 4h±k

J. Iley et al. / Bioorg. Med. Chem. 8 (2000) 1629±1636
1631
Scheme 1.
Scheme 2.
penicillin are too reactive to study in the acid- and base-
catalysed regions of the pH±rate pro®les. For the more
stable compound, 4g, the best computer ®t (solid lines)
to the experimental data (individual points), using eq (1),
gave values for the catalytic second-order rate con-
+
1
stants, kH and kOH , of 0.0214 and 0.187 dm3 mol
s
1, respectively.
pH-Independent region
HPLC analysis of the hydrolyses of the benzylpenicillin
esters 4a±g in the pH-independent region reveals that
the parent drug and the corresponding secondary sulfo-
namide are released in quantitative yield. No lag time
was observed for the formation of the sulfonamide and
benzylpenicillin; neither was the corresponding N-
hydroxymethylsulfonamide detected at any pH value,
consistent with its extremely rapid decomposition to the
parent sulfonamide. The rate of decomposition of N-
hydroxymethyl derivatives of NH-acidic compounds
depends on the pKa of the parent NH-compound, with
a Brùnsted blg value of 0.8.21 For an N-hydroxy-
methylsulfonamide derived from a parent sulfonamide
that has a pKa of 11 a half-life of 80 s can be calculated.
Over the timescale of the hydrolysis reactions, no
decomposition of benzylpenicillin to penicilloic acid,
through b-lactam ring opening, was detected.
Figure 1. pH±rate pro®les for penicilloate 4g (&) at 37 ꢀC, and for
benzoate 4k (&) at 25 ꢀC.
constants for the pH-independent reactions for these
6
latter penicillins are ca. 10 7±10
s
at 30±35 ꢀC,
1
which are ca. 103 smaller than the values obtained for
the sulfonamidomethyl esters 4a±g at 37 ꢀC. The appar-
ent ®rst-order rate constants, ko, for the pH-indepen-
dent hydrolyses of compounds 4 are listed in Table 2.
Unfortunately, sulfonamidomethyl esters of benzyl-
The pH±rate pro®les with broad plateaus, the high
values of the apparent ®rst-order rate constants for the
pH-independent process and the quantitative formation
of benzylpenicillin clearly indicate that the source of
reactivity of the sulfonamidomethyl derivatives 4a±g is

1632
J. Iley et al. / Bioorg. Med. Chem. 8 (2000) 1629±1636
Table 1. Structures of compounds 4
value of 4.5 associated with the hydrolysis of the b-lac-
tam ring in penicillins.10 Fourth, there is no nucleophilic
or general base-catalysed hydrolysis, as indicated by the
absence of a bu€er e€ect (monochloroacetate, acetate
and phosphate) on the decomposition of 4g (Table 3).
The absence of bu€er catalysis also contrasts with the
hydrolysis of benzylpenicillin itself, which is bu€er-cat-
alysed and characterised by a Brùnsted b value of 0.39
that is consistent with general base catalysis.10 Bu€er
catalysis has also been reported for the hydrolysis of
other penicillins (e.g., pheneticillin20), as well as for the
decomposition of both talampicillin and bacampicillin,
which are acyloxymethyl prodrugs of ampicillin.14
Compound
R1
R2
R3CO2H
4a
4b
4c
4d
4e
4f
4g
4h
4i
4-MeC6H4
4-BrC6H4
4-ClC6H4
4-NO2C6H4
Me
Me
Me
Me
Me
Benzylpenicillin
Benzylpenicillin
Benzylpenicillin
Benzylpenicillin
Benzylpenicillin
Benzylpenicillin
Benzylpenicillin
OCOC(Me2)C6H4-4-Cla
OCOC6H5
C6H5
4-MeC6H4
C6H5
Me
Me
4-MeC6H4
4-NO2C6H4
4-NO2C6H4
4-NO2C6H4
4-MeC6H4
Me
Me
Prn
b
4j
4k
OCOCHPr2
OCOC6H5
aClo®brate derivative.
bValproate derivative.
These data provide strong support for a dissociative
mechanism for the pH-independent hydrolysis of esters
4a±g to generate both the penicilloate anion and an N-
sulfonyliminium ion (Scheme 3), in which the positive
charge is stabilised by electron-donating substituents in
the sulfonamide pro-moiety. The greater reactivity of
esters of benzylpenicillin over those of simple carboxylic
acids can also be accommodated by the mechanism
depicted in Scheme 3. The observed rate constants for
compounds 4d,h±j follow a Brùnsted relationship with
the pKa of the carboxylate leaving group, yielding a blg
value of 0.92 (eq (3)).
neither b-lactam ring opening nor conventional ester
hydrolysis. From inspection of the ko values in Table 2
the following observations may be made. First, the pH-
independent pathway is subject to the electronic e€ect
of the substituents in the benzenesulfonamide moiety.
Thus, for compounds 4a±d a plot of log ko against the
Hammett ꢀ values yields a ꢁ value of 0.93 (eq (2)).
This value indicates positive charge development in the
sulfonamide moiety in the transition state.
log ko ˆ 0:92ꢀÆ0:11†pKa 0:06ꢀÆ0:44†
log ko ˆ 0:93ꢀÆ0:11†ꢀ 2:06ꢀÆ0:05†
ꢀ3†
n ˆ 4; r2 0:97; s ˆ 0:19
ꢀ2†
n ˆ 4; r2 ˆ 0:97; s ˆ 0:073
Second, the temperature dependence of the reaction for
compound 4b (Table 2) gives rise to a value of ÁS of
43Æ5 J K 1 mol 1. This value is well within the range
observed for unimolecular ionisation reactions,22
including those reported for other sulfonamidomethyl
benzoates.17 Third, the solvolysis of 4b proceeds slightly
faster in H2O than in D2O, giving a solvent kinetic iso-
tope e€ect, kH O=kD O, of 1.21. Again, this value is con-
The Brùnsted blg value of ca. 1.0 indicates that the
reactivity of these esters is highly dependent on the
nucleofugacity of the carboxylate, a trend expected for
the SN1 mechanism of solvolysis of tertiary N-acyloxy-
methylsulfonamides via iminium ion formation. Fur-
thermore, the fact that the highly sterically hindered
esters of benzylpenicilloic, clo®bric and valproic acids
are included in this correlation clearly demonstrates that
neither conventional ester hydrolysis via nucleophilic
attack at the ester carbonyl carbon atom nor b-lactam
ring opening is a major pathway for the hydrolysis of
compounds 4a±g.
2
2
sistent with a unimolecular ionisation process being the
rate-limiting step,23,24 and contrasts sharply with the
Table 2. Lipophilicity, molar refractivity and pseudo-®rst-order rate
constants at 37 ꢀC for the pH-independent pathway for compounds
4a±k, together with the pKas of the parent sulfonamides
Table 3. E€ect of bu€er concentration on the pseudo-®rst order rate
constants, ko, for the pH-independent hydrolysis of 4g at 37 ꢀC and
Compound pKNa H
Log P MR/cm3
k0/10
s
a
4
1
3
m=0.50 mol dm
2
3
4
1
4a
4b
11.63
11.06
3.47
4.04
137.85
140.94
126
62.5; 12.8c; 9.97d;
Bu€er
[Bu€er]/10 mol dm
pH
kobs/10
s
3.40e; 2.80f; 2.07g; 1.51h
Monochloroacetate
2.00
4.00
8.00
10.0
3.75
7.50
15.0
22.5
1.67
3.33
5.00
10.0
15.0
2.13
2.02
1.97
1.95
3.96
3.96
3.97
3.99
6.97
7.17
7.26
7.36
7.35
9.94
4c
4d
4e
4f
4g
4h
4i
10.94
10.26
8.98b
9.23
8.55b
10.26
10.26
10.26
12.35
3.87
3.21
2.05
2.51
4.12
4.84
4.00
4.98
5.32
138.05
139.26
133.22
137.85
157.93
106.41
85.94
42.9
16.4
38.3
38.8
7.76
7.79
1.62
0.130
185
8.71
8.45
9.34
8.23
8.69
8.53
8.08
6.90
7.15
6.61
6.60
6.70
Acetate
4j
4k
93.62
93.79
Phosphate
apKa of the parent secondary sulfonamide. Unless otherwise stated,
these pKa values were calculated.31
bFrom ref 32. c30 ꢀC; d26 ꢀC; e20 ꢀC; f20 ꢀC in D2O; g13.6 ꢀC; h10 ꢀC.

J. Iley et al. / Bioorg. Med. Chem. 8 (2000) 1629±1636
1633
the only factor contributing to the stability of N-acyl-
oxymethyl prodrugs.
Also worth noting, the rate constants for the sulfona-
mides 4 are very similar to those of the corresponding
acyloxymethylamides 2: at 37 ꢀC the value of 4.3Â10
3
1
3
1
s
for 4c compares with 5.8Â10
s
obtained for its
amide analogue.29 Similarly, the rates of the pH-inde-
pendent hydrolysis of esters 4g±j correlate with the cor-
responding values for the equivalent esters based on the
ethyl hippurate carrier13 2, R1=Ph, R2=CH2CO2Et,
according to eq (4). This correlation clearly indicates
that the same order of reactivity is obtained with both
types of carrier and that similar factors must a€ect the
hydrolysis of both series of prodrugs. This surprising
®nding has been supported by semiempirical SCF-MO
calculations performed using the PM3 method.17
log ksoulf ˆ 0:94ꢀÆ0:13† log kaomide ‡ 0:48ꢀÆ0:57†
ꢀ4†
n ˆ 4; r2 ˆ 0:96; s ˆ 0:21
In vitro antibiotic activity
The in vitro antibiotic activities of compounds 4c±e,g
and benzylpenicillin against a range of Gram positive
and Gram negative bacteria (Table 4) show that, in
general, the sulfonamidomethyl esters of benzylpeni-
cillin have a similar activity pro®le to the parent drug.
This is particularly true of compounds 4c±e. Compound
4g, however, appears to be at least 16 times less active
than benzylpenicillin and the other derivatives. Although
this prodrug is chemically more stable than the others at
pH 7, it is unlikely that this e€ect alone is responsible
for the diminished antibiotic activity since such activity
was measured over 24 h, whereas the half-life of 4g is 15
min. Currently, it is unclear why 4g should display
reduced biological activity. Moreover, there is no
obvious correlation between either antibiotic activity or
chemical reactivity with the log P values of these com-
pounds (Table 2).
Scheme 3.
Conclusions
It has recently been suggested that a key factor deter-
mining the chemical stability, and thus the design, of N-
acyloxymethyl prodrugs is the pKa of NH-acidic parent
compounds.25 For example, the high stability of N-acyl-
oxymethyl derivatives of 5-¯uorouracil, 8 (R=Me),26
and phenytoin, 9 (R=CH2NMe2 or CH2CH2NEt2),27
at pH 4±8, when compared to the reactivity of N-acyl-
oxymethylamides 2,11 13 has been ascribed to the much
lower pKa values of the imide functionality (ca. 8±10)
when compared to the amide group (ca. 15).25 However,
in the present study the calculated pKa values of the
sulfonamide pro-moiety range from 8.5 to 12.3, and yet
compounds 4 are ca. 103 times more reactive than either
Sulfonamidomethylation of the carboxyl group of
benzylpenicillin with the appropriate N-chloro-
methylsulfonamide a€ords the corresponding sulfon-
amidomethyl esters in reasonable yields. Sulfonamido-
methyl esters of benzylpenicillin hydrolyse at pH 2±10
via cleavage of the ester group to release the parent drug
quantitatively. This is a clear advantage over the
majority of a-acyloxyalkyl double prodrugs of b-lactam
antibiotics, which can su€er b-lactam ring opening
depending on the substrate and the pH.14 For example,
at 25 ꢀC and pH 7.5 hydrolysis of bacampicillin involves
22% b-lactam ring opening and 78% ester hydrolysis,
whereas at pH 3.5, b-lactam hydrolysis accounts for
74% of the total hydrolysis reaction.14 Even so, simple
sulfonamidomethyl esters 4a±f are clearly too unstable
to be of general use as prodrugs. However, compound
4g, which contains an N-phenylbenzenesulfonamide pro-
moiety, is a promising candidate, presenting a half-life
5
8 or 9. We have also recently reported a ko value of 10
1
s
for the N-acyloxymethyl derivative 1028 (the pKa
value for the parent sulfonamide is 7.38), which is ca.
102 times more reactive than 8 or 9. Clearly, our results
indicate that the pKa of the parent NH compound is not

1634
J. Iley et al. / Bioorg. Med. Chem. 8 (2000) 1629±1636
of 15 min at pH 2±10. This value compares to the half-
life of 15±20 min at pH 4±8 and 35 ꢀC reported for the
hydrolysis of hetacillin, a 4-imidazolidinone prodrug of
ampicillin.30 Improved stability might be expected by
introducing electron withdrawing substituents either in
the benzenesulfonyl N-aryl moieties of the carrier. The
sulfonamidomethyl derivatives of other acid moieties,
4h±j, are much more stable at pH 7.4, with half-lives
ranging from 15 min to 15 h. Thus tertiary sulfonami-
domethyl esters may be useful prodrugs for carboxylic
acid drugs with pKa>4, as well as for secondary sulfo-
namide drugs.
Benzypenicillin was purchased commercially. All chem-
icals were reagent grade except those for kinetic studies
and HPLC, which were analytical or LiChrosolv
(Merck) grade. Column chromatography was per-
formed using silica gel 60 mesh 70-230 (Merck). Com-
pounds 4h±j were obtained from a previous study.17
(N-Methyl-4-toluenesulfonamido)methyl (2S,5R,6R)-1-
aza-3,3-dimethyl-7-oxo-6-phenylacetamido-4-thiabicyclo-
[3.2.0]heptane-2-carboxylate 4a. A solution of N-methyl-
4-toluenesulfonamide (6 mmol) and paraformaldehyde
(0.3 g) in chlorotrimethylsilane (20 mL) was re¯uxed
during 2 h. Removal of the solvent yielded N-chloro-
methyl-N-methyl-4-methylbenzenesulfonamide 5a; 99%;
mp 83±87 ꢀC; dH/ppm: 2.38 (3H, s), 2.85 (3H, s), 5.34
(2H, s), 7.12±7.74 (4H, AA0BB0). A solution of 5a (1.1
equiv) in dry THF (1 mL) was added to a suspension of
sodium benzylpenicilloate (5 mmol) in THF (5 mL) at
room temperature. Upon completion of the reaction,
the solvent was evaporated, and the residue treated with
ice-water and extracted with ethyl acetate. The organic
extracts were washed with ice-water, sodium hydrogen
carbonate, brine and dried over magnesium sulfate.
Evaporation of the solvent gave pure 4a as a cream gum
(68%); ꢂmax (cm 1): 3300, 1770, 1720, 1650; dH/ppm:
1.32 (3H, s, C3Me), 1.34 (3H, s, C3Me), 2.64 (3H, s,
ArMe), 2.92 (3H, s, NMe), 3.63 (2H, s, PhCH2), 4.17
(1H, s, C2H), 5.15 (1H, d, J=4.0 Hz, C5H), 5.42 (1H, d,
J=11.4 Hz, NCH2O), 5.52 (1H, dd, J=4.0, 9.0 Hz,
C6H), 5.54 (1H, d, J=11.4 Hz, NCH2O), 6.07 (1H, d,
J=9.0 Hz, NH), 7.25±7.72 (9H, m, 2ÂAr). Found C,
56.3; H, 5.8; N, 7.6; C25H29N3O6S2 requires C, 56.5; H,
5.5; N, 7.9.
Experimental
Elemental analyses were obtained from ITQB (Oeiras,
Portugal) and from Medac (Engle®eld Green, UK)
laboratories. 1H and 13C NMR spectra were recorded as
CDCl3 solutions on General Electric QE-300 or Jeol
JNM-EX400 spectrometers using Me4Si as internal
standard; coupling constants, J, are quoted in Hz. FTIR
spectra were recorded on a Nicolet Impact 400 spectro-
photometer. Electron-impact ionisation (EI) and fast-
atom bombardment (FAB) mass spectra were recorded
on a VG Mass Lab 20-250 spectrometer. Melting points
were determined using a Buchi 510 instrument and are
uncorrected. HPLC was performed using a system
comprising a Shimadzu LC-9A pump coupled to a
variable wavelength Shimadzu SPD-6AV UV±Vis
detector, a 20 mL loop injection valve and a Merck
LiChrospher1 100 RP-8 5 mm 125Â4 mm column.
Table 4. In vitro antibacterial activity of sulfonamidomethyl esters
4c±e,g and of benzylpenicillin (Pen G)
(N-Methyl-4-bromobenzenesulfonamido)methyl (2S,5R,
6R)-1-aza-3,3-dimethyl-7-oxo-6-phenylacetamido-4-thia-
bicyclo[3.2.0]heptane-2-carboxylate 4b. Synthesised as
4a starting from N-methyl-4-bromobenzenesulfona-
mide: cream gum (66%); ꢂmax (cm 1): 3300, 1775, 1725,
1650; dH/ppm: 1.34 (3H, s, C3Me), 1.35 (3H, s, C3Me),
2.95 (3H, s, NMe), 3.65 (2H, s, PhCH2), 4.20 (1H, s,
C2H), 5.28 (1H, d, J=4.0 Hz, C5H), 5.45 (1H, d, J=12.0
Hz, NCH2O), 5.52 (1H, d, J=12.0 Hz, NCH2O), 5.60
(1H, dd, J=4.0, 9.0 Hz, C6H), 6.06 (1H, d, J=9.0 Hz,
NH), 7.27±7.71 (9H, m, 2ÂAr). Found C, 47.9; H, 4.7;
N, 6.9; C24H26BrN3O6S2 requires C, 48.3; H, 4.4; N, 7.0.
3
MIC 10 3/mmol dm
Bacteria
4c
4d
4e
4g
Pen G
P. aeruginosa ATCC 27853
K. pneumoniaa
>500 >500 >500 >500 >500
>500 >500 >500 >500 >500
K. pneumonia ATCC 10031
E. colia
62.5
125
62.5
250
>500 >500 >500 >500 >500
62.5
E. coli ATCC 25922
S. typhimurium ATCC 43971
S. enteritidis
62.5
3.9
7.8
31.2
15.6
125
7.8
62.5
31.2
31.2
62.5 >500 62.5
3.9
7.8
250
62.5
62.5
125
3.9
7.8
S. ¯exneri
31.2
15.6
31.2
15.6
S. dysenteriae ATCC 13313
S. marcescens NCTC 1377
P. mirabilis
S. aureus ATCC 25923
S. faecalis ATCC 10541
>500 >500 >500 >500 >500
61.2 125 61.2 250 7.8
<0.24 0.48 <0.24 31.2 <0.24
15.6 125 15.6 >500 15.6
(N-Methyl-4-chlorobenzenesulfonamido)methyl (2S,5R,
6R)-1-aza-3,3-dimethyl-7-oxo-6-phenylacetamido-4-thia-
bicyclo[3.2.0]heptane-2-carboxylate 4c. Synthesised as
4a starting from N-methyl-4-chlorobenzenesulfonamide:
ab-Lactamase producing bacteria.

J. Iley et al. / Bioorg. Med. Chem. 8 (2000) 1629±1636
1635
cream gum (66%); ꢂmax (cm 1): 3300, 1775, 1725, 1640;
dH/ppm: 1.34 (6H, s, C3Me2), 2.93 (3H, s, NMe), 3.64
(2H, s, PhCH2), 4.19 (1H, s, C2H), 5.28 (1H, d, J=4.2
Hz, C5H), 5.44 (1H, d, J=11.4 Hz, NCH2O), 5.51 (1H,
d, J=11.4 Hz, NCH2O), 5.60 (1H, dd, J=4.2, 9.0 Hz,
C6H), 6.13 (1H, d, J=9.0 Hz, NH), 7.27±7.42 (9H, m,
2ÂAr). Found C, 52.6; H, 5.0; N, 7.3; C24H26 ClN3O6S2
requires C, 52.2; H, 4.7; N, 7.6.
with an ionic strength maintained at 0.5 mol dm 3 using
3
NaClO4. A 20 mL aliquot of a 10 2 mol dm stock
solution of substrate in acetonitrile was added to 10 mL
of the appropriate thermostatted bu€er solution. At
regular intervals, 50 mL samples of the reaction mixture
were quenched with ice-cold acetonitrile (400 mL) and
the resulting solution analysed by HPLC using the fol-
lowing conditions: detector wavelength, 230 nm (290 nm
for 4d and 4h±j); mobile phase, methanol±water con-
3
(N-Methyl-4-nitrobenzenesulfonamido)methyl (2S,5R, 6R)-
1-aza-3,3-dimethyl-7-oxo-6-phenylacetamido-4-thiabicy-
clo[3.2.0]heptane-2-carboxylate 4d. Synthesised as 4a
starting from N-methyl-4-nitrobenzenesulfonamide:
slightly yellow gum (83%); ꢂmax (cm 1): 3300, 1775,
1725, 1630; dH/ppm: 1.34 (6H, s, C3Me2), 2.99 (3H, s,
NMe), 3.64 (2H, s, PhCH2), 4.16 (1H, s, C2H), 5.34 (1H,
d, J=4.0 Hz, C5H), 5.48 (1H, d, J=11.0 Hz, NCH2O),
5.52 (1H, d, J=11.0 Hz, NCH2O), 5.59 (1H, dd, J=4.0,
9.0 Hz, C6H), 6.09 (1H, d, J=9.0 Hz, NH), 7.25±7.37
(5H, m, Ar), 7.95-8.42 (4H, AA0BB0, ArSO2). Found C,
51.1; H, 4.8; N, 9.7; C24H26N4O8S2 requires C, 51.2; H,
4.6; N, 10.0.
taining 0.1 mol dm tetrabutylammonium phosphate
(55:45 to 60:40%). Compound 4k was studied by UV
spectroscopy using a previously described protocol.11
Computations
Molecular refractivities (MR) were calculated with the
software ChemSketch v. 4.01 from Advanced Chemistry
Development (ACD) Inc. Octanol±water partition coef-
®cients (log P) were obtained using the ACD/I-lab ser-
vice (log P v. 3.6). No reliable experimental log P values
could be obtained as result of extensive hydrolysis of
penicillin derivatives when the shake-¯ask method was
used. Calculation of the kinetic parameters in eq (1) was
performed using the Solver option of the Microsoft
Excel1 software package. The regression analysis of
kinetic data in Table 1 was performed using the statis-
tical package of SigmaPlot1 5.0 from SSPS Inc.
(N-Phenylmethanesulfonamido)methyl (2S,5R,6R)-1-aza-
3,3-dimethyl-7-oxo-6-phenylacetamido-4-thiabicyclo[3.2.0]-
heptane-2-carboxylate 4e. Synthesised as 4a starting
from N-phenylmethanesulfonamide: cream gum (56%);
ꢂmax (cm 1): 3320, 1778, 1740, 1630; dH/ppm: 1.48 (6H,
s, C3Me2), 3.02 (3H, s, MeSO2), 3.65 (2H, s, PhCH2),
4.41 (1H, s, C2H), 5.48 (1H, d, J=4.0 Hz, C5H), 5.67±
5.71 (3H, m, NCH2O and C6H), 6.15 (1H, d, J=9.0 Hz,
NH), 7.27±7.42 (10H, m, 2ÂAr). Found C, 51.1; H, 5.4;
N, 7.9; C24H27N3O6S2 requires C, 51.7; H, 5.2; N, 8.1.
Antimicrobial susceptibility tests
The minimum inhibitory concentrations (MICs) of
benzylpenicillin and its sulfonamidomethyl esters were
determined by the Agar Dilution Method (Mueller-
Hinton) using a multipoint inoculator. The plates were
incubated in aerobic atmospheres at 37 ꢀC for 24 h.
MICs were recorded as the lowest concentration com-
pletely inhibiting visible bacterial growth. The range of
concentrations used for all the compounds was
(N-4-Tolylmethanesulfonamido)methyl (2S,5R,6R)-1-aza-
3,3-dimethyl-7-oxo-6-phenylacetamido-4-thiabicyclo[3.2.0]-
heptane-2-carboxylate 4f. Synthesised as 4a starting
from N-4-tolylmethanesulfonamide: cream gum (27%);
ꢂmax (cm 1): 3300, 1770, 1730, 1625; dH/ppm: 1.48 (3H,
s, C3Me), 1.49 (3H, s, C3Me), 2.39 (3H, s, ArMe), 3.00
(3H, s, MeSO2), 3.65 (2H, s, PhCH2), 4.40 (1H, s, C2H),
5.49 (1H, d, J=4.2 Hz, C5H), 5.64±5.70 (3H, m,
NCH2O and C6H), 6.15 (1H, d, J=9.0 Hz, NH), 7.21±
7.40 (9H, m, 2ÂAr). Found C, 56.7; H, 5.6; N, 7.8;
C25H29N3O6S2 requires C, 56.5; H, 5.5; N, 7.9.
10
6
2.24Â10
0.50Â10 mol ml 1. The following refer-
ence bacterial organisms were used: Pseudomonas aeru-
ginosa ATCC 27853, Klebsiella pneumoniae ATCC
10031, Escherichia coli ATCC 25992, Salmonella typhi-
murium ATCC 43971, Shigella dysenteriae ATCC
13313, Serratia marcescens NCTC 1377, Staphylococcus
aureus ATCC 25923 and Streptococcus faecalis ATCC
10541. The following bacterial organisms belong to the
collection of the Faculty of Pharmacy: b-lactamase
producing Klebsiella pneumoniae, b-lactamase produ-
cing Escherichia coli, Salmonella enteritidis, Shigella
¯exneri, and Proteus mirabilis.
(N-Phenyl-4-toluenesulfonamido)methyl (2S,5R,6R)-1-
aza-3,3-dimethyl-7-oxo-6-phenylacetamido-4-thiabicyclo-
[3.2.0]heptane-2-carboxylate 4g. Synthesised as 4a start-
ing from N-phenyl-4-toluenesulfonamide: cream gum
(66%); ꢂmax (cm 1): 3360, 1780, 1720, 1680; dH/ppm:
1.41 (6H, s, C3Me2), 2.44 (3H, s, ArMe), 3.64 (2H, s,
PhCH2), 4.27 (1H, s, C2H), 5.20 (1H, d, J=4.0 Hz,
C5H), 5.56 (1H, dd, J=4.0, 9.0 Hz, C6H), 5.65 (1H, d,
J=11.0 Hz, NCH2O), 5.81 (1H, d, J=11.0, NCH2O),
6.08 (1H, d, J=9.0, NH), 7.26±7.54 (14H, m, 3ÂAr).
Found C, 60.3; H, 5.3; N 6.9; C30H31N3O6S2 requires C,
60.7; H, 5.2; N 7.1.
Acknowledgements
This work was supported by Junta Nacional de Investi-
gacao Cientõ®ca e Tecnologica (Portugal) under the
contract PBIC/SAU/1546/92.
References
Kinetic studies
All kinetic experiments were carried out in aqueous
bu€ers (borate, phosphate, acetate and chloroacetate)
1. Jack, D. B. Handbook of Clinical Pharmacokinetic Data;
Macmillan: Basingstoke, 1994; pp 89±94 and 113±115.

1636
J. Iley et al. / Bioorg. Med. Chem. 8 (2000) 1629±1636
2. Bundgaard, H. In Bioreversible Carriers in Drug Design:
Theory and Application; Roche, E. B., Ed.; Pergamon Press:
Oxford, 1987; pp 13±94.
3. Agersborg, H. P.; Batchelor, A.; Cambridge, G. W.; Rule,
A. W. Br. J. Pharmacol. 1966, 26, 649.
4. Holysz, R. P.; Stavely, H. E. J. Am. Chem. Soc. 1950, 72,
4760.
5. Durckheimer, W.; Adam, F.; Fischer, G.; Kirrstetter, R. In
Advances in Drug Research, Vol. 17; Testa, B., Ed.; Academic
Press: London, 1988; pp 62±234.
17. Lopes, F.; Moreira, R.; Iley, J. J. Chem. Soc., Perkin
Trans. 2 1999, 431.
18. Euranto, E. K. In The Chemistry of Carboxylic Acids and
Esters; Patai, S., Ed.; Wiley: Chichester, 1969; pp 505±588.
19. Gensmantel, N. P.; Gowling, E. W.; Page, M. I. J. Chem.
Soc., Perkin Trans. 2 1978, 335.
20. Schwartz, M. A.; Granatek, A. P.; Buckwalter, F. H. J.
Pharm. Sci. 1962, 51, 523.
21. Johansen, M.; Bundgaard, H. Arch. Pharm. Chemi, Sci.
Ed. 1979, 7, 175.
6. Mizen, L.; Burton, G. In Integration of Pharmaceutical
Discovery and Development; Borchardt, R. T.; Freidinger, R.
M.; Sawyer, T. K.; Smith, P. L., Eds.; Plenum Press: New
York, 1998; pp 345±365.
7. Wheeler, W. J.; Preston, D. A.; Wright, W. E.; Hu€man, G.
W.; Osborne, H. E.; Howard, D. P. J. Med. Chem. 1979, 22,
657.
8. Wheeler, W. J.; Wright, W. E.; Line, V. D.; Frogge, J. A. J.
Med. Chem. 1977, 20, 1159.
9. Bundgaard, H. Drugs of the Future 1991, 16, 443.
10. Page, M. Adv. Phys. Org. Chem. 1987, 23, 165.
11. Iley, J.; Moreira, R.; Rosa, E. J. Chem. Soc., Perkin Trans.
2 1991, 563.
12. Moreira, R.; Calheiros, T.; Cabrita, J.; Mendes, E.;
Pimentel, M.; Iley, J. Pharm. Res. 1996, 13, 70.
13. Iley, J.; Moreira, R.; Calheiros, T.; Mendes, E. Pharm.
Res. 1997, 14, 1634.
14. Nguyen, N. A. T.; Mortada, L. M.; Notari, R. E. Pharm.
Res. 1988, 5, 288.
15. Miyauchi, M.; Sasahara, K.; Fujimoto, K.; Kawamoto, I.;
Ide, J.; Nakao, H. Chem. Pharm. Bull. 1989, 37, 2369.
16. Saab, A. N.; Hussain, A. A.; Patel, I. H.; Dittert, L. W. J.
Pharm. Sci. 1990, 79, 802.
22. Winstein, S.; Fainberg, A. H. J. Am. Chem. Soc. 1957, 79,
5937.
23. Thornton, E. K.; Thornton, E. R. In Isotope E€ects In
Chemical Reactions, eds. Collins, C. J.; Bowman, N. S., Eds.;
Van Nostrand Reinhold: New York, 1970; pp 213±285.
24. Johnson, S. L. Adv. Phys. Org. Chem. 1967, 5, 273.
25. Bundgaard, H.; Rasmussen, G. J. Pharm. Res. 1991, 8,
1238.
26. Buur, A.; Bundgaard, H.; Falch, E. Int. J. Pharm. 1985,
24, 43.
27. Varia, S. A.; Schuller, S.; Stella, V. J. J. Pharm. Sci. 1984,
73, 1074.
28. Lopes, F.; Moreira, R.; Iley, J. Bioorg. Med. Chem. 2000,
8, 707.
29. Moreira, R.; Mendes, E.; Calheiros, T.; Bacelo, M. J.; Iley,
J. Tetrahedron Lett. 1994, 35, 7107.
30. Tsuji, A.; Yamana, T. Chem. Pharm. Bull. 1974, 22, 2432.
31. Perrin, D. D.; Dempsey, B.; Serjeant, E. P. pKa Prediction
for Organic Acids and Bases; Chapman and Hall: London,
1981; pp 109±132.
32. King, J. F. In The Chemistry of Sulphonic Acids, Esters and
their Derivatives; Patai, S.; Rappoport, Z., Eds.; John Wiley:
Chichester, 1991; pp 249±259.