Phosphinate inhibitors of UDP-N-acetylmuramoyl-L-alanyl-D-glutamate: L-lysine ligase (MurE)

Article abstract of DOI:10.1002/ardp.200600191

The increasing emergence of pathogenic bacterial strains with high resistance to antibiotic therapy has created an urgent need for the development of new antibacterial agents that are directed towards novel targets. We have focused our attention on the Mu

Full text of DOI:10.1002/ardp.200600191

ˇ
Arch. Pharm. Chem. Life Sci. 2007, 340, 127–134
K. Strancar et al.
127
Full Paper
Phosphinate Inhibitors of UDP-N-Acetylmuramoyl-L-Alanyl-D-
Glutamate: L-Lysine Ligase (MurE)
Katja ꢀtrancar1, Audrey Boniface2, Didier Blanot2, and Stanislav Gobec1
1 Faculty of Pharmacy, University of Ljubljana, Ljubljana, Slovenia
2 Enveloppes Bactꢀriennes et Antibiotiques, IBBMC, UMR 8619 CNRS, Universitꢀ de Paris-Sud, Orsay,
France
The increasing emergence of pathogenic bacterial strains with high resistance to antibiotic the-
rapy has created an urgent need for the development of new antibacterial agents that are direc-
ted towards novel targets. We have focused our attention on the Mur ligases (MurC-F), which
catalyze the early steps of bacterial peptidoglycan biosynthesis, and which to date represent
under-exploited targets for antibacterial drug design. We show that some of our phosphinate
inhibitors of UDP-N-acetylmuramoyl-L-alanyl:D-glutamate ligase (MurD) also inhibits UDP-N-ace-
tylmuramoyl-L-alanyl-D-glutamate:L-lysine ligase (MurE). To obtain new information on their
structure-activity relationships, three new, structurally related phosphinates were synthesized
and evaluated for inhibition of MurD and MurE.
Keywords: Antibacterials / Inhibitors / MurD / MurE /
Received: November 16, 2006; accepted: December 19, 2006
DOI 10.1002/ardp.200600191
acetylglucosamine (GlcNAc) and N-acetylmuramic acid
(MurNAc). The carboxyl group of MurNAc residues is sub-
stituted in most bacteria by a peptide unit, L-alanyl-c-D-
glutamyl-meso-diaminopimeloyl(or L-lysyl)-D-alanine [3].
The enzyme MurD (UDP-MurNAc-L-alanine:D-glutamate
ligase), which catalyses the addition of D-glutamate to
the cytoplasmic peptidoglycan precursor UDP-MurNAc-L-
Ala, is highly specific for this amino acid [4]. In contrast,
the enzyme MurE has specificities that can vary from one
bacterial species to another. In Escherichia coli and Staphy-
lococcus aureus MurE catalyses the addition of meso-A2pm
(UDP-N-acetylmuramoyl-L-alanyl-D-glutamate:meso-diami-
nopimelate ligase) or L-lysine (UDP-N-acetylmuramoyl-L-
alanyl-D-glutamate:L-lysine ligase), respectively, to the
cytoplasmic peptidoglycan precursor UDP-MurNAc-L-Ala-
D-Glu (Scheme 1). Through their e-amino group, A2pm
and L-lysine are involved in cross-linking between the
peptide units linked to the glycan chains, and thus they
have a key role in the solidity of the macromolecule and
the integrity of the bacterial cell.
Introduction
Over the last decade, the emergence of bacterial resis-
tance to antibiotic therapy has resulted in a global health
threat [1]. One of the attractive strategies to overcome
this problem is to find new antibacterial agents that are
directed towards novel targets. The best known and most
validated target for antibacterial therapy is the system of
enzymes that are involved in peptidoglycan biosynthesis
[2, 3]. The late, extracellular stages of bacterial peptido-
glycan biosynthesis are inhibited by b-lactam and glyco-
peptide antibiotics. In contrast, the early intracellular
biosynthetic steps have so far received only marginal
attention as potential drug targets.
Peptidoglycan is an essential macromolecular compo-
nent of the cell wall of both Gram-positive and Gram-
negative bacteria. Its main function is to preserve cell
integrity by withstanding the internal osmotic pressure.
The glycan chains are composed of alternating units of N-
MurE ligases from both species are very promising but
under-exploited targets for antibacterial drug design.
There have been several attempts to target MurE from E.
coli with inhibitors. While several analogues of diamino-
Correspondence: Stanislav Gobec, Faculty of Pharmacy, University of
ˇ
Ljubljana, AÐkerceva 7, 1000 Ljubljana, Slovenia
E-mail: gobecs@ffa.uni-lj.si
Fax: +386 1 425 8031
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

ˇ
128
K. Strancar et al.
Arch. Pharm. Chem. Life Sci. 2007, 340, 127–134
Scheme 1. Reaction catalysed by MurE and the general structure of phosphinate substrate-like inhibitors.
the case for the b-sulfonamidopeptides, which were mod-
erate inhibitors of MurE from S. aureus [11]. Here, we
report on the inhibitory potencies of the phosphinate
compounds on MurE from S. aureus, which represents
one of the major pathogenic species. We also describe the
detailed synthesis and the biochemical evaluation of
three new phosphinate inhibitors. Comparative struc-
ture-activity relationships towards MurD and MurE are
deduced.
Results and discussion
Figure 1. Structures of the potent inhibitors of MurE from E.coli.
Among the compounds we synthesized previously, we
have identified several promising inhibitors of MurE
from S. aureus. Compounds 1a and 1c inhibited MurE to a
significant extent (Table 1). Compound 1b, the positional
isomer of 1a with the nitro group on the meta-position, is
a less potent inhibitor. The inhibitory potencies of com-
pounds 1d and 1e on MurE are also significantly lower.
Compounds 1d and 1e possess the same 1,3-benzodioxo-
lyl fragment as compound 1c; however, it is connected to
the phosphinodipeptide by two shorter linkers, and this,
presumably, explains the weaker inhibition observed. In
compound 1f, introduction of m-coumaric acid to the key
phosphinodipeptide resulted in promising inhibition of
MurE. Its analogue 1g, without the hydroxyl substituent
on the aromatic ring, was less effective. The introduction
of a more flexible oxymethyl linker to compound 1h
resulted in a slight improvement in its biological activity,
as compared to compound 1g (Table 1).
pimelic acid exhibited moderate inhibitory activities [5–
7], phosphinate derivatives designed as transition-state
analogues of UDP-N-acetylmuramoyl-L-alanyl-D-glutama-
te:meso-diaminopimelate ligase showed inhibition of the
E. coli enzyme in the micromolar range (Fig. 1) [8]; the
best of these, I, had an IC50 of 1.1 lM, whereas the deriv-
ative devoid of the UMP part, II, was a poorer inhibitor
(IC50 = 700 lM). However, until recently no inhibitors of
MurE from S. aureus (a UDP-N-acetylmuramoyl-L-alanyl-D-
glutamate:L-lysine ligase) have been reported.
Lately, we reported on the synthesis and structure-
activity relationships of two types of transition-state ana-
logue inhibitors of MurD: i) a series of phosphinate com-
pounds of general formula R-(L,D)-Ala-W(PO2-CH2)-(L,D)-Glu
were shown to be good inhibitors of the E. coli enzyme
[9, 10]; ii) in contrast, a series of b-sulfonamidopeptides of
general formula R-L-Ala-W(CH2-SO2-NH)-D-Glu were very
poor MurD inhibitors [11]. As MurE follows MurD in the
cascade of biosynthetic amino acid addition, the transi-
tion-state analogue inhibitors of MurD show structural
resemblance to the substrate for MurE (Scheme 1). It is
thus reasonable to evaluate their inhibition of MurE, for
which they could act as substrate analogues [12]. This was
It should be noted that the best inhibitor of MurD 1a is
also one of the best inhibitors of MurE (Table 1). As far as
the other compounds are concerned, there does not
appear to be a straightforward correlation between the
inhibitory properties towards these two enzymes (cf. 1b,
the positional isomer of 1a, which is a good MurD inhibi-
tor, but a poorer MurE one).
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2007, 340, 127–134
New Inhibitors of MurE
129
Table 1. Phosphinate inhibitors of MurE and MurD.
Table 2. MurE and MurD inhibitory activities of truncated com-
pounds 2a–c and comparison with their non-truncated analo-
gues.
Compounds
R2
RA (%)
MurE
RA (%)
MurDa)
Compounds
1a
R1
RA (%)
MurE
RA (%)
MurDa)
1a
2a
13
46
8
13
42
12
8
100
1b
1c
19
44
1d
2b
52
35
17
91
1h
2c
52
50
40
92
1d
1e
60
52
36
Data represent means of two experiments. Standard deviations
17
are within l 10% of given values.
a)
From reference [9].
for the inhibition of MurE. However, compound 2a, the
truncated analogue of 1a, which showed absolutely no
inhibition of MurD, was a moderate inhibitor of MurE
(Table 2). The same trend was seen with two other trun-
cated analogues, 2b and 2c. From these results, we can
conclude that the c-carboxylate of D-glutamic acid in this
type of compounds is not of as great importance for their
binding to MurE as it is for their binding to MurD.
In addition to the compounds described above, some
sulfonamide-substituted phosphinate inhibitors of MurD
were screened against MurE (compounds 3a-c; Table 3)
and they were seen to be promising inhibitors. Encour-
aged by these results, we initiated the synthesis of some
new compounds based on compounds 3b and 3c, which
were also good inhibitors of MurD. To study the influence
of the phosphinoalanine residue on inhibition of MurE,
we prepared analogues of compounds 3b and 3c in which
this was replaced by phosphinovaline (14a, Scheme 2 and
Table 3) and phosphinoisoleucine 14b and 15a. Since
these compounds are new, they were also evaluated for
inhibition of MurD using the procedure we described
previously [9].
1f
21
50
35
23
28
41
1g
1h
Data represent means of two experiments. Standard deviations
are within € 10% of given values.
a)
From reference [9]. IC50 values for compounds 1a, 1e and 1g
were 78 lM, 95 lM and 432 lM, respectively.
Previously, we demonstrated the essential importance
of the integrity of the glutamate residues in these types
of compounds for their inhibitory activities towards
MurD by the synthesis and biological evaluation of their
truncated analogues [9]. We expected the same to be true
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

ˇ
130
K. Strancar et al.
Arch. Pharm. Chem. Life Sci. 2007, 340, 127–134
Scheme 2. Synthesis of phosphinate inhibitors 14a–b and 15a.
The synthesis of compounds 14a–b and 15a is pre- with benzyl chloroformate. In the next step, the phos-
sented in Scheme 2. The starting 1-(benzhydrylami- phonous acids 6a–b were esterified with methanol using
no)alkyl phosphinic acids 4a–b were readily synthesized EDC, as we have described previously [9]. To prevent their
from benzhydrylamine, proper aldehyde and phosphinic decomposition, methyl phosphinates 7a–b were imme-
acid in aqueous medium (i.e. in a 50% aqueous solution diately used in the next reaction step. Sodium methano-
of H3PO2), although Baylis et al. described the general pro- late was added, followed by dimethyl 2-methylenepenta-
cedure using absolute ethanol as solvent [13]. Our modi- nedioate, to yield the protected phosphinodipeptides
fied procedure is advantageous over that in absolute 8a–b. The Cbz group was removed by catalytic hydroge-
ethanol since anhydrous H3PO2 is not required; anhy- nation, and the free amines 9a–b were coupled to substi-
drous H3PO2 can be obtained from the commercial 50% tuted benzylsulfonyl chlorides 10 and 11, to obtain the
aqueous solution only by time-consuming high-vacuum sulfonamides 12a–b and 13a, respectively. The latter
distillation [15]. The yields for the synthesis of 1-(benzhy- ones were converted by alkaline hydrolysis into the tar-
drylamino)alkylphosphinic acids 4a–b in aqueous med- get compounds 14a–b and 15a, respectively.
ium were similar to those reported for the reaction in
The phosphinoisoleucine derivatives 14b and 15a
ethanol [13]. Compounds 4a–b were converted in two inhibited MurE less than their parent compounds 3c and
steps into Cbz-protected a-aminoalkylphosphonous acids 3b, respectively. Conversely, the phosphinovaline deriv-
6a–b using concentrated HCl, followed by their reaction ative 14a showed a relatively good inhibitory potency,
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2007, 340, 127–134
New Inhibitors of MurE
131
Table 3. MurE and MurD inhibitory activities of sulfonamide-sub-
stituted phosphinates 3a–c, 14a–b and 15a.
Com-
pounds
R3
R4
RA (%)
MurE
RA (%)
MurD
3a
3b
3c
–CH3
–CH3
–CH3
39
52
44
52a)
22a)
47a)
Figure 2. Superimposition of the computer model of compound
1c (in red) on the X-ray structure of UMT (in yellow) bound to
MurE. Only the surface of the protein is shown for clarity.
the UDP-binding pocket, and the phosphinodipeptide
part partially overlaps with the peptide part of UMT.
To conclude, in the present study, we report on the
synthesis and activity of inhibitors of MurE from S. aur-
eus. Although they possess the N-substituted phosphino-
dipeptide (L,D)-Ala-c(PO2-CH2)-(L,D)-Glu, our compounds
are not transition-state analogues of MurE, but analogues
of the substrate UDP-MurNAc-dipeptide. Their inhibitory
character is therefore not surprising. However, if we com-
pare them to the simple substrate analogue MurNAc-L-
Ala-D-Glu, for which we determined a residual activity of
61%, some of the inhibitory potencies presented e.g. for
1a, 1c are unexpected. Furthermore, some of these com-
pounds e.g. 1a, 1f also inhibit MurD; these constitute pro-
mising starting points for further development as anti-
bacterial agents.
14a
14b
–CH(CH3)2
29
54
49
–CH(CH3)CH2CH3 54
–CH(CH3)CH2CH3 71
15a
46
Data represent means of two experiments. Standard deviations
are within l 10% of given values.
a)
From reference [9].
better than its parent compound 3c (Table 3). For MurD,
all of the compounds formed by the replacement of phos-
phinoalanine with the branched aminophosphinic acids
resulted in weaker or unchanged inhibitory potencies.
To investigate the possible binding mode of our new
MurE inhibitors, compound 1c was docked into the MurE
active site (pdb code 1E8C) [16] using the eHiTS software
[17]. In this crystal structure, MurE from E. coli was co-
crystallized with the product of the reaction, UDP-Mur-
We thank Dr. Chris Berrie for critical reading of the manu-
script and SimBioSys, Inc., for a free academic license for
eHiTS. This work was financially supported by the Ministry of
Education, Science and Sport of the Republic of Slovenia, the
Centre National de la Recherche Scientifique, the Proteus pro-
gramme of integrated actions, and Lek Pharmaceuticals d.d.
NAc-L-Ala-c-D-Glu-meso-A2pm
(UDP-MurNAc-tripeptide,
Experimental
UMT) [16]. Until now, no crystal structure of MurE from S.
aureus has been published, so for our docking we used
the coordinates of the structurally related enzyme from
E. coli. According to eHiTS, compound 1c binds to the
active site of MurE in a similar manner as the product
UMT (Fig. 2). The 1,3-benzodioxole ring points towards
Chemistry
All of the reactions were carried out under dry conditions and
with magnetic stirring. Chemicals were purchased from Acros
(Acros Organics, Geel, Belgium) and used without further purifi-
cation. Solvents were used without purification or drying,
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

ˇ
132
K. Strancar et al.
Arch. Pharm. Chem. Life Sci. 2007, 340, 127–134
unless otherwise stated. Reactions were monitored using analy-
tical TLC plates (Merck, silica gel 60 F254) with sulphuric acid
staining. Silica gel grade 60 (70–230 mesh, Merck, Darmstadt,
Germany) was used for column chromatography. NMR spectra
were obtained on a Bruker Avance DPX 300 instrument (Bruker
Bioscience, Billerica, MA, USA). 1H-NMR spectra were recorded at
300.13 MHz with tetramethylsilane as an internal standard. 31P-
NMR spectra were recorded at 121 MHz with H3PO4 as an exter-
nal standard. Mass spectra were obtained with a VG-Analytical
Autospec Q mass spectrometer with EI or FAB ionization (MS
Centre, Joˇzef Stefan Institute, Ljubljana). IR spectra were
recorded on a Perkin-Elmer FTIR 1600 spectrometer (Perkin
Elmer, Norwalk, CT, USA). Elemental analyses were performed
by the Department of Organic Chemistry, Faculty of Chemistry
and Chemical Technology, Ljubljana, on a Perkin Elmer elemen-
tal analyzer 240 C. Melting points were determined using a
Reichert hot-stage microscope (C. Reichert, Vienna, Austria) and
they are uncorrected.
1 h, the ice bath was removed and the solution was stirred at r.t.
for 2.5 h. Then, the reaction mixture was washed with Et2O
(3630 mL), concentrated in vacuo and acidified to pH = 1-2 at
08C. The separated crystals were filtered off, washed and dried.
1-{[(Benzyloxy)carbonyl]amino}-2-
methylpropylphosphinic acid 6a
Yield: 59.3%; m.p. 97–998C (lit. m.p. 110–1118C) [13].
1-{[(Benzyloxy)carbonyl]amino}-2-
methylbuthylphosphinic acid 6b
Yield: 72.2%; m.p. 96–998C [14].
General procedure for the synthesis of methyl 1-
{[(benzyloxy)carbonyl]amino}alkylphosphinates 7a–b
1-{[(Benzyloxy)carbonyl]amino}alkylphosphinic acid (30 mmol)
was dissolved in dry methanol (100 mL) and EDC (6.33 g,
33 mmol) was added, and the reaction was stirred in an argon
atmosphere for 12 h. The solvent was evaporated in vacuo, and
the residue was dissolved in CH2Cl2 and washed with 10% citric
acid (3650 mL), saturated NaHCO3 (3650 mL) and brine
(2650 mL). The organic phase was dried (anh. Na2SO4), filtered
and evaporated. The crude product was immediately used in the
next reaction step due to its instability.
General procedure for the synthesis of 1-
(benzhydrylamino)alkylphosphinic acids 4a–b
Diphenylmethylammonium chloride (21.97 g, 100 mmol) and
H2O (50 mL) were added to a 50% aqueous solution of phospho-
nous acid (10.4 mL, 100 mmol), and the mixture was heated to
reflux. Water solutions of the appropriate aldehydes (100 mmol
in 25 mL H2O) were added and the solutions were heated at their
reflux temperature for 2 h. The solutions were then cooled and
the white crystals were filtered off, washed with cold water and
dried.
Methyl 1-{[(benzyloxy)carbonyl]amino}-2-
methylpropylphosphinate 7a
Yield: 84%; Oil;1H-NMR (300 MHz, DMSO-d6): d (ppm) = 0.85–1.14
(m, 6H, –(CH3)2), 1.95-2.35 (m, 8H, –CH(CH3)2), 3.34–3.78 (m, 4H,
–POCH3 and –PCH-), 5.08 (s, 2H, ArCH2 –), 6.91 (d, 1JPH = 545.9 Hz,
1H, –PH), 7.24–7.43 (m, 5H, Ar-H); MS (FAB) m/z: 286 [M+H]+.
1-(Benzhydrylamino)-2-methylpropylphosphinic acid 4a
Yield: 65%; m.p. 186–1898C (lit. m.p. 1868C) [13].
1-(Benzhydrylamino)-2-methylbuthylphosphinic acid 4b
Yield: 28.3%; m.p. 164–1678C (lit. m.p. 1758C) [13].
Methyl 1-{[(benzyloxy)carbonyl]amino}-2-
methylbuthylphosphinate 7b
Yield: 84.7%; Oil; 1H-NMR (300 MHz, DMSO-d6): d (ppm) = 0.75–
1.10 (m, 6H, –CH(CH3)–CH2 –CH3), 1.11–2.18 (m, 3H, –CH(CH3)–
CH2 –CH3), 3.62–3.98 (m, 4H, –POCH3 and –PCH-), 5.03–5.15 (m,
2H, ArCH2 –), 7.26–7.43 (m, 5H, Ar–H); MS (FAB) m/z: 300 [M+H]+.
General procedure for the synthesis of 1-
aminoalkylphosphinic acids 5a–b
1-(Benzhydrylamino)alkylphosphinic acid (100 mmol) was dis-
solved in a concentrated solution of HCl (115 mL) and refluxed
for 12 h. The reaction mixture was cooled, washed with Et2O
(4660 mL) and evaporated in vacuo. The residue was dissolved in
ethanol (10 mL for 1 g of residue) and propyleneoxide was added
drop-wise until precipitation was seen. The suspension was fil-
tered and the white crystals were washed with the mother liquor
and cold ethanol, and dried.
General procedure for the synthesis of protected
phosphinodipeptides 8a-b
Methyl
1-{[(benzyloxy)carbonyl]amino}alkylphosphinate
(20 mmol) and dimethyl 2-methylpentanedioate (22.6 mmol)
were dissolved in dry methanol (25 mL), and a stream of dry
argon gas was bubbled through this solution. The mixture was
cooled to 08C and 2 M NaOMe (11 mL) was added, followed by a
balloon filled with Ar. The reaction was then stirred at r.t. over-
night, after which 120 mL 1 M HCl was added and the solution
extracted with CH2Cl2 (5650 mL). The organic phases were col-
lected and washed with brine (50 mL), dried (anh. Na2SO4), fil-
tered and evaporated in vacuo.
1-Amino-2-methylpropylphosphinic acid 5a
Yield: 89.3%; m.p. 205–2068C (lit. m.p. 201–2028C) [13].
1-Amino-2-methylbuthylphosphinic acid 5b
Yield: 28.3%; m.p. 211–2128C (lit. m.p. 2038C) [13].
Dimethyl 2-{[(1-{[(benzyloxy)carbonyl]amino}-2-
methylpropyl)(methoxy)phosphoryl]methyl}
General procedure for the synthesis of 1-
{[(benzyloxy)carbonyl]amino}alkylphosphinic acids 6a–b
1-Aminoalkyphosphinic acid (50 mmol) was dissolved in diox-
ane (20 mL) and 0.1 M NaOH (20 mL) and cooled to 08C. Benzyl
chloroformate (7.8 mL, 55 mmol) was added and the pH of the
resultant solution was adjusted to 9–10 with 1 M NaOH. After
pentanedioate 8a
Yield: 39.2%; Oil; 1H-NMR (300 MHz, CDCl3): d (ppm) = 0.96–1.12
(m, 6H, –(CH3)2), 1.73–2.38 (m, 8H, –CH(CH(CH3)2)–P–CH2 –
CH(COOCH3)–CH2 –CH2 –COOCH3), 3.60–3.80 (m, 9H, –POCH3
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2007, 340, 127–134
New Inhibitors of MurE
133
and 26 –COOCH3), 3.81–4.05 (m, 1H, –PCH– ), 5.09–5.22 (m,
2H, ArCH2 –), 7.29–7.43 (m, 5H, ArH); 31P-NMR (121 MHz, CDCl3):
d (ppm) = 52.48, 52.64, 52.85, 53.67; IR (NaCl, cm– 1): 3236, 2956,
1737, 1540, 1456, 1210, 1040, 826; MS (FAB) m/z: 458 [M+H]+;
Anal. Calcd. for C21H32N1O8P1 (%): C, 55.14; H, 7.05; N, 3.06. Found:
C, 55.04; H, 6.93; N, 3.23.
Dimethyl 2-{[methoxy(2-methyl-1-{[(4-
nitrobenzyl)sulfonyl]amino}butyl)phosphoryl]methyl}
pentanedioate 12b
Yield: 54.3% (two steps); Oil; 1H-NMR (300 MHz, CDCl3): d (ppm) =
0.87–1.12 (m, 6H, –CH(CH3)–CH2 –CH3), 1.13–2.47 (m, 9H,
–CH(CH(CH3)–CH2 –CH3)-P-CH2 –CH(COOCH3)–CH2 –CH2 –
COOCH3), 2.80–3.05 (m, 1H, –CH2 –CH(COOCH3)–CH2 –) 3.63–
4.03 (m, 10H, –POCH3, 26 –COOCH3 in -PCH-), 4.45–4.63 (m, 2H,
ArCH2), 7.65–7.75 (m, 2H, ArH-2 in ArH-6), 8.253 + 8.246 + 8.249 +
8.263 (4 d (diastereoisomers), Jortho = 8.9 Hz, 2H, ArH-5 and ArH-3);
IR (NaCl, cm– 1): 2957, 1736, 1607, 1523, 1438, 1349, 1202, 1039,
858, 696; MS (FAB) m/z: 537 [M+H]+; this compound being highly
hygroscopic, we were unable to obtain elemental analysis.
Dimethyl 2-{[(1-{[(benzyloxy)carbonyl]amino}-2-
methylbutyl)(methoxy)phosphoryl]methyl} pentanedioate
8b
Yield: 34.2%; Oil; 1H-NMR (300 MHz, DMSO-d6): d (ppm) = 0.75–
0.89 (m, 3H, –CH2 –CH3), 0.89–1.02 (m, 3H, –CH(CH3)–), 1.07–
2.37 (m, 9H, –CH(CH(CH3)–CH2 –CH3)–P–CH2 –CH(COOCH3)–
CH2 –CH2 –COOCH3), 2.60–2.80 (m, 1H, –CH2 –CH(COOCH3)–
CH2 –), 3.49–3.63 (m, 9H, –POCH3 and 26COOCH3), 3.64–4.02
(m, 1H, –PCH-), 5.01–5.17 (m, 2H, ArCH2), 7.26–7.40 (m, 5H,
ArH), 7.41–7.74 (m, 1H, –OCONH–); 31P-NMR (121 MHz, DMSO-
d6): d (ppm) = 52.97, 53.19, 53.42, 53.59, 53.83, 53.85, 54.17,
54.30; IR (NaCl, cm– 1): 3478. 1724, 1546, 1199, 1035; MS (FAB)
m/z: 472 [M+H]+; Anal. Calcd. for C22H34N1O8P1 (%): C, 56.04; H,
7.27; N, 2.97. Found: C, 56.26; H, 7.59; N, 3.27.
Dimethyl 2-{[methoxy(2-methyl-1-{[(3-
nitrobenzyl)sulfonyl]amino}butyl)phosphoryl]methyl}
pentanedioate 13a
1
Yield: 58.9%; Oil; H-NMR (300 MHz, DMSO-d6): d (ppm) = 0.81–
1.11 (m, 6H, –CH(CH3)–CH2 –CH3), 1.12–2.50 (m, 9H,
–CH(CH(CH3)–CH2 –CH3)-P-CH2 –CH(COOCH3)–CH2 –CH2 –
COOCH3), 2.81–3.07 (m, 1H, -CH2 –CH(COOCH3)–CH2 –), 3.59–
4.04 (m, 10H, –POCH3 and 26 –COOCH3), 4.44–4.67 (m, 2H,
ArCH2), 7.54–7.64 (m, 1H, ArH-5), 7.81–7.90 (m, 1H, ArH-6), 8.25
(d, Jortho = 8,3 Hz, 1H, ArH-4), 8.35–8.45 (m, 1H, ArH-2); IR (NaCl,
cm– 1): 2959, 2878, 1736, 1659, 1531, 1438, 1353, 1329, 1203,
1040, 815, 690; MS (FAB) m/z: 537 [M+H]+; Anal. Calcd. for
C21H33N2O10P1S161.5H2O (%): C, 44.76; H, 6.44; N, 4.97. Found: C,
44.76; H, 6.23; N, 4.70.
General procedure for the synthesis of dimethyl 2-{[(1-
aminoalkyl)(methoxy)phosphoryl]methyl} pentanedioates
9a–b
The Cbz-protected phosphinodipeptide (4 mmol) was dissolved
in methanol (15 mL) and hydrogenated in a Parr hydrogenator
(25 psi) over Pd/C (10%) overnight. The catalyst was filtered off
and the solvent was evaporated in vacuo. The crude product was
immediately used in the next reaction step.
General procedure for the alkaline hydrolysis of methyl
esters. Synthesis of target phosphinates 14a–b and 15a
Triester (0.35 mmol) was dissolved in a mixture of dioxane
(6 mL) and 1 M LiOH (6 mL) and stirred at r.t. until the reaction
was completed according to thin-layer chromatography anal-
ysis. The solvent was evaporated in vacuo, water (25 mL) was
added, and the solution washed with CH2Cl2 (4615 mL). The
aqueous phase was acidified to pH 1 with conc. HCl and
extracted with EtOAc (8615 mL). The organic phases were col-
lected, dried (anh. Na2SO4), filtered and evaporated in vacuo.
General procedure for the synthesis of sulfonamides
12a–b and 13a
The free amine (2 mmol) was dissolved in CH2Cl2 (30 mL), cooled
to 08C, and then
a solution of proper sulfonylchloride
(1.1 mmol) in CH2Cl2 (5 mL) was added, followed by Et3N
(0.56 mL, 4 mmol). Afterwards, another fraction of sulfo-
nylchloride (1.1 mmol) was added, and the reaction was stirred
in an ice bath for 1 h and then for 1 h at r.t.; this was then
washed with 10% NaHSO4 (3620 mL) and brine (20 mL), and
dried (anh. Na2SO4), filtered and evaporated in vacuo.
2-{[Hydroxy(2-methyl-1-{[(4-
nitrobenzyl)sulfonyl]amino}propyl)phosphoryl]methyl}
pentanedioic acid 14a
Yield: 19.0%; m.p. 82–848C; 1H-NMR (300 MHz, DMSO-d6): d
(ppm)
= 0.91–1.01 (m, 6H, –(CH3)2), 1.60–2.31 (m, 7H,
–CH(CH(CH3)2)-P-CH2 –CH(COOH)–CH2 –CH2 –COOH), 2.62–2.81
(m, 1H, –CH2 –CH(COOH)–CH2 –), 3.33–3.48 (m, 1H, PCH), 4.58–
4.70 (m, 2H, ArCH2), 7.47 (m, 1 H, –SO2NH–), 7.75 (d, Jortho = 8.7
Hz, 2H, ArH-2 and ArH-6), 8.24 (d, Jortho = 9.0 Hz, 2H, ArH-5 and
ArH-3); 31P-NMR (121 MHz, DMSO-d6): d (ppm) = 45.82, 46.02; IR
(KBr, cm– 1): 2930, 1731, 1522, 1350, 1272, 1125, 857, 695; MS
(FAB) m/z: 481 [M+H]+; Anal. Calcd. for C17H25N2O10P1S1 60.7H2O
(%): C, 41.41; H, 5.40; N, 5.68. Found: C, 41.85; H, 5.55; N, 5.10.
Dimethyl 2-{[methoxy(2-methyl-1-{[(4-
nitrobenzyl)sulfonyl]amino}propyl)phosphoryl]methyl}
pentanedioate 12a
Yield: 26.6% (two steps); Oil;1H-NMR (300 MHz, DMSO-d6): d (ppm)
= 0.91–1.03 (m, 6H, –(CH3)2), 1.70–2.42 (m, 7H, –CH(CH(CH3)2)–
P–CH2 –CH(COOCH3)–CH2 –CH2 –COOCH3), 2.67–2.86 (m, 1H,
–CH2 –CH(COOCH3)–CH2 –), 3.50-3.70 (m, 10H, –POCH3, 26
–COOCH3 and –PCH), 4.59 (d, 2JH – H = 11.3 Hz, 1H, Ar-CHaHb-), 4.67
(d, 2JH-H = 12.8 Hz, 1H, Ar-CHaHb-), 7.74 (dd, Jortho = 9.4 Hz, 2H, ArH-2
and ArH-6), 8.25 (d, Jortho = 8.3 Hz, 2H, ArH-5 and ArH-3); 31P-NMR
(121 MHz, DMSO-d6): d (ppm) = 53.22, 53.86, 54.01; IR (NaCl,
cm–1): 2956, 1735, 1523, 1437, 1349, 1205, 1042, 858, 696; MS
(FAB) m/z: 523 [M+H]+; Anal. Calcd. for C20H31N2O10P160.5H2O (%):
C, 45.19; H, 6.07; N, 5.27. Found: C, 45.18; H, 6.23; N, 5.56.
2-{[Hydroxy(2-methyl-1-{[(4-nitrobenzyl)sulfonyl]-
amino}butyl)phosphoryl]methyl} pentanedioic acid 14b
Yield: 76.2%; m.p. 69–708C; 1H-NMR (300 MHz, DMSO-d6): d
(ppm) = 0.78–1.02 (m, 6H, –CH(CH3)–CH2-CH3), 1.14–2.35 (m,
9H, –CH(CH(CH3)–CH2 –CH3)–P–CH2 –CH(COOH)–CH2 –CH2 –
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

ˇ
134
K. Strancar et al.
Arch. Pharm. Chem. Life Sci. 2007, 340, 127–134
COOH), 2.60–2.76 (m, 1H, –CH2 –CH(COOH)–CH2 –), 3.48–3.58
(m, 1H, -PCH-), 4.51–4.77 (m, 2H, ArCH2), 7.713 + 7.734 (2 d (dia-
stereoisomers), Jortho = 7.9 Hz, 2H, ArH-2 in ArH-6), 8.235 + 8.256 (2
d (diastereoisomers), Jortho = 8.3 Hz, 2H, ArH-5 and ArH-3); 31P-NMR
(121 MHz, DMSO-d6): d (ppm) = 45.60; IR (KBr, cm– 1): 3476, 1718,
1638, 1523, 1350, 1151, 858, 696; MS (FAB) m/z: 495 [M+H]+; Anal.
Calcd. for C18H27N2O10P1S1 (%): C, 43.72; H, 5.50; N, 5.67. Found: C,
43.50; H, 5.84; N, 5.34.
USA). The crystal structure of MurE was retrieved from the RCSB
protein data base (PDB entry 1E8C). Docking was performed on a
PC (1.5 MHz Intel processor) running on Linux with the eHiTS
6.1 programme [17] (SimBioSys, Inc., Toronto, Canada, 2006)
using the default parameter settings.
References
2-{[Hydroxy(2-methyl-1-{[(3-nitrobenzyl)sulfonyl]-
[1] P. Courvalin, J. Davies, Curr. Opin. Microbiol. 2003, 6, 425–
amino}butyl)phosphoryl]methyl}pentanedioic acid 15a
Yield: 78.8%; m.p. 73–758C;1H-NMR (300 MHz, DMSO-d6): d (ppm)
= 0.76–0.97 (m, 6H, –CH(CH3)-CH2 –CH3), 1.00–2.33 (m, 9H,
–CH(CH(CH3)–CH2 –CH3)–P–CH2 –CH(COOH)–CH2 –CH2 –COOH),
2.56–2.77 (m, 1H, –CH2 –CH(COOH)–CH2-), 3.48–3.68 (m, 1H,
–PCH-), 4.57–4.74 (m, 2H, ArCH2-), 7.62–7.76 (m, 1H, ArH-5),
7.82–7.97 (m, 1H, ArH-6), 8.18–8.27 (m, 1H, ArH-4), 8.29–8.39
(m, 1H, ArH-2); 31P-NMR (121 MHz, DMSO-d6): d (ppm) = 45.98,
46.17; IR (NaCl, cm– 1): 3429, 1713, 1531, 1456, 1354, 1127, 905,
690, 548; MS (FAB) m/z: 495 [M+H]+; Anal. Calcd. for
C18H27N2O10P1S1 (%): C, 43.72; H, 5.50; N, 5.67. Found: C, 43.84; H,
5.74; N, 5.36.
426.
[2] H. Labischinski, H. Maidhof in Bacterial Cell Wall (Eds.: J.-
M. Ghuysen, R. Hakenbeck), Elsevier Science B.V., Amster-
dam, 1994, p. 23.
[3] J. van Heijenoort, Nat. Prod. Rep. 2001, 18, 503–519.
[4] F. Pratviel-Sosa, F. Acher, F. Trigalo, D. Blanot, et al., FEMS
Microbiol. Lett. 1994, 115, 223–228.
[5] C. Michaud, D. Mengin-Lecreulx, J. van Heijenoort, D. Bla-
not, Eur. J. Biochem. 1990, 194, 853–861.
[6] I. van Assche, M. Soroka, A. Haemers, M. Hooper, et al.,
Eur. J. Med. Chem. 1991, 26, 505–515.
[7] G. Auger, J. van Heijenoort, J. C. Vederas, D. Blanot, FEBS
Lett. 1996, 391, 171–174.
Enzymatic assays
The compounds were tested for their ability to inhibit the addi-
tion of L-[14C]Lys to UDP-MurNAc-L-Ala-D-Glu in a mixture (final
volume: 50 lL) containing 0.1 M Tris-HCl, pH 8.6, 15 mM MgCl2,
100 lM UDP-MurNAc-L-Ala-D-Glu, 200 lM L-[14C]Lys (50 000 cpm),
5% (v/v) DMSO, 4 nM MurE from S. aureus [18] (diluted with
20 mM potassium phosphate, pH 7.2, 1 mM dithiothreitol) and
1 mM tested compound (at this concentration, all compounds
were soluble in the assay mixture containing 5% DMSO). The
mixture was incubated for 30 min at 308C, and the reaction was
stopped by adding 10 lL glacial acetic acid. The mixture was
then lyophilized and taken up in HPLC elution buffer. The radio-
active substrate and product were separated by reverse-phase
HPLC with a Nucleosil 5C18 column (15064.6 mm) as stationary
phase, and isocratic elution at a flow rate of 0.6 ml/min with
50 mM ammonium formate, pH 4.7. The compounds were
detected and quantified with a HPLC radioactivity monitor LB
506 C-1 (Berthold France, Thoiry, France) using Quickszint Flow
2 scintillator (Zinsser Analytic, Maidenhead, UK) at 0.6 mL/min.
Residual activity was calculated with respect to a similar assay
without the inhibitors. Data are expressed as means of duplicate
determinations. Standard deviations were within l10% of given
values.
[8] B. Zeng, K. K. Wong, D. L. Pompliano, S. Reddy, M. E. Tan-
ner, J. Org. Chem. 1998, 63, 10081–10086.
[9] K. ꢀtrancar, D. Blanot, S. Gobec, Bioorg. Med. Chem. Lett.
2006, 16, 343–348.
[10] K. ꢀtrancar, S. Gobec, D. Blanot, U. Urleb (Lek Pharmaceu-
ticals, d.d.), Slovene Patent SI 21851, 2006.
[11] J. Humljan, M. Kotnik, A. Boniface, T. ꢀolmajer, et al., Tet-
rahedron 2006, 67, 10980–10988.
[12] A. El Zoeiby, F. Sanschagrin, R. C. Levesque, Mol. Microbiol.
2003, 47, 1–12.
[13] E. K. Baylis, C. D. Campbell, J. G. Dingwall, J. Chem. Soc.
Perkin Trans. I 1984, 2845–2853.
[14] K. ꢀtrancar, S. Gobec, Synthesis (Stuttg) 2004, 3, 359–362.
[15] S. J. Fitch, J. Am. Chem. Soc. 1964, 86, 61–64.
[16] E. Gordon, B. Flouret, L. Chantalat, J. van Heijenoort, et
al., J. Biol. Chem. 2001, 276, 10999–11006.
[17] Z. Zsoldos, D. Reid, A. Simon, S. B. Sadjad, A. P. Johnson,
J. Mol. Graph. Model. 2007, in press.
[18] The enzyme (C-terminal His-tagged form) was obtained
from E. coli cells over-expressing the murE gene from S.
aureus. It was purified by affinity chromatography (Boni-
face, A., Dementin, S, Blanot, D., unpublished results).
Docking studies
The inhibitor was prepared with HyperChem 7.5 (HyperChem,
version 7.5 for Windows, 2002. Hypercube, Inc., Gainesville, FL,
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com