Synthesis and in vitro enzyme activity of aza, oxa and thia derivatives of bacterial cell wall biosynthesis intermediates

Article abstract of DOI:10.1039/b105117m

Mechanism based inhibitors of diaminopimelate aminotransferase (DAP-AT) were designed using knowledge of its substrate specificity and mechanism. Synthesis of thiolester and amide substrate analogues was achieved prior to in vitro inhibition studies, but ester analogues proved too unstable to isolate. Thia substrate analogues showed no inhibitory properties, but the aza substrate analogue 12a showed reversible inhibition vs. DAP-AT and time dependent inhibition in the absence of the natural substrate 4. Substrate analogue 12a is thefirst example of an amide inhibitor of PLP dependent enzymes. Antibiotic properties of 12a were also briefly assessed.

Full text of DOI:10.1039/b105117m

View Online / Journal Homepage / Table of Contents for this issue
Synthesis and in vitro enzyme activity of aza, oxa and thia
derivatives of bacterial cell wall biosynthesis intermediates†
Russell J. Cox* and Paul S. H. Wang
1
School of Chemistry, University of Bristol, Cantock’s Close, Clifton, Bristol, UK BS8 1TS.
E-mail: r.j.cox@bris.ac.uk
Received (in Cambridge, UK) 8th June 2001, Accepted 18th July 2001
First published as an Advance Article on the web 10th August 2001
Mechanism based inhibitors of diaminopimelate aminotransferase (DAP-AT) were designed using knowledge of
its substrate specificity and mechanism. Synthesis of thiolester and amide substrate analogues was achieved prior
to in vitro inhibition studies, but ester analogues proved too unstable to isolate. Thia substrate analogues showed
no inhibitory properties, but the aza substrate analogue 12a showed reversible inhibition vs. DAP-AT and time
dependent inhibition in the absence of the natural substrate 4. Substrate analogue 12a is the first example of an
amide inhibitor of PLP dependent enzymes. Antibiotic properties of 12a were also briefly assessed.
Introduction
Bacterial cell wall biosynthesis is an attractive target for
the design of new antibacterial compounds.1 Indeed many
microbial organisms biosynthesize compounds such as β-
lactams (e.g. penicillin, clavulanic acid) and glycopeptides
(e.g. vancomycin) which target bacterial cell wall biosynthesis
for evolutionary advantage over competing microorganisms. In
order to achieve this, the antibiotic producing organisms have
evolved parallel resistance mechanisms to avoid suicide. Genes
coding for these resistance mechanisms (e.g. β-lactamases) can
be passed between different bacterial species and in environ-
ments rich in antimicrobial compounds these mechanisms can
become prevalent through natural selection.2 Such processes
may have given rise to the development of antibiotic resistant
pathogens which pose a threat to public health.
We have been studying the biosynthesis of the key bac-
terial cell wall component diaminopimelic acid (DAP) 1 in
order to develop new classes of antibacterial compounds.
The bacterial biosynthesis of DAP 1 has been well studied
(Scheme 1).3 In brief, the reduction of -aspartate gives the
C4 -aspartate semialdehyde 2 which condenses with pyruvate
to give -dihydrodipicolinate 3 (Scheme 1). Reduction to -
tetrahydrodipicolinate followed by acylation then gives -α-
(N-succinylamino)-ε-oxopimelic acid 4 for most common
bacteria.‡ In Bacillus species, and some others, the acylating
group is acetyl. At this point the operation of N-acyl diamino-
pimelate aminotransferase (DAP-AT) yields -N-acyl-DAP 5
which is subjected to deacylation and epimerisation to afford
meso-DAP 1. In a small number of bacterial species such as
Bacillus sphaericus meso-DAP can be synthesised directly from
tetrahydrodipicolinate, catalysed by meso-DAP dehydrogenase.4
The final step of the pathway is the specific decarboxylation of
meso-DAP 1 at the -centre to give -lysine 6. Since -lysine 6 is
required for protein synthesis, inhibition of the pathway could
be expected to be detrimental to bacterial development. How-
ever -lysine 6 and/or DAP 1 are also required by bacteria
for the construction of the peptidoglycan layer of the cell wall.
In this polymeric structure the lysine or DAP diamine plays
Scheme 1 The bacterial synthesis of -lysine: (i) -dihydrodipicolinate
synthase; (ii) meso-diaminopimelate decarboxylase.
the crucial role of cross-linking short peptides attached to the
glc-NAc-mur-NAc polysaccharide.5 These cross-links give the
peptidoglycan structure the physical strength to resist lysis
caused by the very high internal osmotic pressures of bacteria.
Thus DAP pathway inhibitors could be promising new anti-
microbial compounds.
All of the DAP pathway enzymes have been studied as poten-
tial targets for antimicrobial action. One of the most promising
targets is the aminotransferase DAP-AT and we have syn-
thesised potent slow binding inhibitors (e.g. 7, 8) of this enzyme
with inhibition constants in the low nM range.6 These com-
pounds do show antimicrobial activity, although not at a
comparable level of potency with medically useful drugs such
† Electronic supplementary information (ESI) available: 13C NMR
data, mass spectra and IR spectra for 11a. See http://www.rsc.org/
suppdata/p1/b1/b105117m/
‡ In this article the term succinyl refers to the group HO2CCH2CH2-
CO–, and methylsuccinyl to MeO2CCH2CH2CO–.
2022
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b105117m

View Online
as carbenicillin.7 One reason for this could be that compounds
such as 7 and 8 are reversible inhibitors of the aminotransferase.
One potential way of increasing in vivo potency could be to
utilise irreversible inhibitors of DAP-AT.
Design of potential irreversible DAP-AT inhibitors
Pyridoxal 5Ј-phosphate (PLP) dependent aminotransferases,
such as DAP-AT, interconvert ketones and amines.8 In the
‘forward’ direction ketones are stereoselectively reductively
aminated and in the ‘reverse’ direction (and in order to
regenerate the cofactor) amines are oxidised to ketones. For
DAP-AT the substrate 4 is reductively aminated and in vitro
(and presumably in vivo) an excess of -glutamic acid 9 acts as
an amine source to regenerate the cofactor (Scheme 2). For
DAP-AT the catalytic cycle is proposed to involve attack by the
nucleophilic amine of the pyridoxamine-5Ј-phosphate (PMP)
at the substrate carbonyl to form an enzyme bound ketimine.
Rearrangement of this ketimine follows, via the quinonoid
intermediate to the aldimine. Subsequent aminolysis by an
active site amine then releases the product. -Glutamic acid 9
can then drive the reaction in reverse, yielding α-ketoglutarate
10 and the PMP form of the cofactor ready for another
catalytic cycle.
Compounds such as the potent DAP-AT inhibitors 7 and 8
interrupt this process by forming (reversibly) an enzyme bound
hydrazone with the electrophilic PLP form of the cofactor.6,9
We envisaged designing compounds which could react irre-
versibly with the nucleophilic cofactor. Such compounds could
be esters, thiolesters or possibly amides, in which the α-methyl-
ene of the substrate is replaced by oxygen, sulfur or nitrogen.
These compounds, possessing a nucleofuge might be expected
to acylate the PMP form of the cofactor, forming an enzyme
bound amide which would be unlikely to be rapidly hydrolysed
(Scheme 2). A similar strategy was recently demonstrated by
Leeper who showed that the thiolester and ester substrate
mimics inhibit 5-aminolaevulinic acid dehydratase – an enzyme
which also operates via an enzyme–substrate imine inter-
mediate.10
Previous substrate specificity experiments with DAP-AT
revealed that for adequate recognition substrates should possess
a seven-carbon backbone.9 It is clear however that DAP-
AT must also process the five-carbon skeleton of glutamate.
Restrictions of choice of N-acyl group are less severe, with
succinyl being preferred, but substitution by Cbz, for example,
is tolerated well (kcat/KM ca. 25% wrt 4). We sought to preserve
the seven atom backbone of DAP and utilise the N-acyl groups
succinyl and Cbz for inhibitor design. Thus our initial targets
were 11–13. Clearly 11a and 11b, containing sulfur instead of
a methylene linkage could be expected to be somewhat larger
than the natural substrate 4 and we also sought to synthesise
the one atom shorter thia analogue 14 in order to limit the
overall size of the potential inhibitor.
Scheme 2 The mechanism of DAP-AT. Proposed mode of action of
oxalyl inhibitors (boxed).
Results and discussion
Thiol esters
Succinylation of homocysteine 15 with succinic anhydride in
saturated aqueous bicarbonate followed by rapid acidification,
organic extraction and treatment with diethyl ether–diazo-
methane gave a 2 : 1 mixture of the methylsuccinyl thiolester
16 and the dimethyl ester thiol 17 in good yield (Scheme 3).
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034
2023

View Online
Treatment of 17 with methyl oxalyl chloride in CH2Cl2 in the
presence of one equivalent of 2,6-lutidine then gave the
expected oxalyl thiolester 18 in good yield. Final selective
methyl ester hydrolysis using 3.0 equivalents of LiOHؒH2O
gave the deprotected target compound 11a with no trace of
free thiol or lithium oxalate (1H NMR, 13C NMR, IR).
An analogous route starting with -cysteine 19 was also
followed. Succinylation and methylation yielded a mixture of
the required thiol 20 and its symmetrical disulfide 21. The
purified thiol was esterified using methyl oxalyl chloride and
the product 22 was separated from starting material 20 by an
aqueous CuSO4 wash. Final deprotection using LiOHؒH2O
was difficult, but addition of solid LiOH under conditions
favouring very slow dissolution (2 : 1 CH3CN : H2O) afforded
the trilithium salt 14. No evidence of thiolester cleavage to yield
the free thiol was observed by MS or IR analysis.
An attempted parallel synthesis of the N-Cbz protected
homocysteine analogue began with the treatment of homo-
cysteine 15 with CbzCl under standard Schotten–Baumann
conditions. However, after acidification and treatment with
diethyl ether–diazomethane only the cyclic thiocarbamate 2311
was isolated (Scheme 4). In order to circumvent problems
caused by the nucleophilic thiol, homocystine 24 was used
Scheme
3
Synthesis of thia substrate analogues. Reagents and
conditions: (i) succinic anhydride, NaHCO3 (aq), then CH2N2, 31%; (ii)
MeO2CCOCl, 2,6-lutidine, CH2Cl2, 49%; (iii) 3.0 eq. LiOH, CH3-
CN–H2O, quantitative; (iv) succinic anhydride, NaHCO3 (aq), then
CH2N2, 19%; (v) MeO2CCOCl, 2,6-lutidine, CH2Cl2, 72%; (vi) 3.0 eq.
LiOHؒH2O, CH3CN–H2O, quantitative.
Scheme
4
Synthesis of thia substrate analogues. Reagents and
conditions: (i) CBzCl, NaOH (aq), then H3Oϩ; (ii) 10% HCl–MeOH,
∆, quant.; (iii) CbzCl, CH2Cl2, Et3N, 27%; (iv) 3% Na–Hg, MeOH,
24%; (v) MeO2CCOCl, 2,6-lutidine, CH2Cl2, 88%; (vi) 2.0 eq. Li-
OHؒH2O CH3CN–H2O, quant.
2024
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034

View Online
Scheme 5 Synthesis of aza substrate analogues. Reagents and conditions: (i) 10% HCl–MeOH, ∆, 83%; (ii) MeO2CCOCl, CH2Cl2, HOBT, then 31,
53%; (iii) 2.0 eq. LiOHؒH2O, CH3CN–H2O, quant.; (iv) H2 1 atm, 10% Pd/C, MeOH, 93%; (v) succinic anhydride, CHCl3, Et3N, then CH2N2, 76%;
(vi) 3.0 eq. LiOHؒH2O, CH3CN–H2O, 97%.
instead. Treatment of this compound with CbzCl under
Schotten–Baumann conditions gave poor yields of the bis-Cbz
protected product. The alternative procedure of methylating
first using methanolic HCl to give the diester 2512,13 followed by
Cbz protection in organic solvent, yielding the fully protected
disulfide 26, was preferable (Scheme 4). Disulfide cleavage by
sodium borohydride in methanol was rapid, but afforded
only the N-protected thiolactone 27. Cleavage using sodium
amalgam in methanol was slower, but afforded a 2.5 : 1 mixture
of the desired thiol 28 and the thiolactone 27 which were easily
separated by chromatography. Coupling the thiol 28 with
methyl oxalyl chloride in the presence of 2,6-lutidine gave the
thiolester 29 in high yield and the target 11b was produced by
selective methyl ester cleavage as before.
Amides
Commercially available αN-Cbz protected -diaminobutyric
acid 30 was treated with methanolic HCl to give a quantitative
yield of the methyl ester hydrochloride 3114 (Scheme 5).
Attempted coupling of this amine with methyl oxalyl chloride
in the presence of various organic bases failed to give the
desired amide 32, yielding only the α-N-protected lactam 33.
The amide 32 was produced by pretreatment of methyl oxalyl
Scheme
6
Synthesis of oxa substrate analogues. Reagents and
chloride with hydroxybenzotriazole (HOBT) and excess base
in CH2Cl2 before slow addition of the amine hydrochloride.
Methyl ester hydrolysis then afforded the desired 12b. Cbz
protected 32 could also be deprotected under standard hydro-
genolysis conditions and the resulting amine hydrochloride 34
then acylated with succinic anhydride in saturated aqueous
NaHCO3 followed by a diazomethane quench to afford the
triester 35, albeit in poor yield. Better yields were achieved
by treating the hydrochloride salt 34 with succinic anhydride
in CH2Cl2 in the presence of Et3N before diazomethane
treatment. Following chromatography the trimethyl ester 35
was quantitatively converted to the trilithium salt 12a using
standard procedures.
conditions: (i) NaHCO3 (aq), CBzCl, then CH2N2, 95%; (ii) CH2Cl2,
pyridine, MeO2CCOCl, 63%; (iii) 2.0 eq. LiOHؒH2O, CH3CN–H2O.
Attempted selective methyl ester hydrolysis under a number of
reaction conditions (e.g. LiOH, various sources of IϪ) resulted
only in formation of mixtures of protected vinylglycine 40 and
lactone 38.
In order to avoid excessive lactone formation and to improve
deprotection at the end of the synthesis we decided to utilise
tert-butyl ester protecting groups and to utilise a masked γ-
alcohol. Thus -aspartic acid 41a was selectively methylated at
the β-carboxylate by treatment with one equivalent of thionyl
chloride in methanol at 0 ЊC (Scheme 7). Subsequent Cbz and
tert-butyl protection using standard procedures then gave the
protected -aspartate 43b.15,16 The chemistry was repeated to
produce the succinyl protected material 43a, by succinylation of
the amino-monoester with succinic anhydride before treatment
with isobutylene and sulfuric acid in CH2Cl2. Selective methyl
ester hydrolysis using aqueous NaOH to give the sodium salts
44a and 44b was followed by formation of the mixed anhydride
carbonates 45a and 45b and in situ reduction with aqueous
NaBH4 afforded the protected primary alcohols 46a and 46b.
These tert-butyl esters did not show the propensity for lactone
formation displayed by methyl ester 37.
Esters
Following the successful selective LiOH deprotection of the
methyl ester of 29 in the presence of the sensitive thiol ester
functionality we considered synthesizing oxoaminooxapimel-
ates by a parallel strategy to that used for the thia analogues.
Homoserine 36 was thus treated with CbzCl followed by rapid
acidification and treatment with excess diethyl ether–diazo-
methane (Scheme 6). In the case of homocysteine 15 the major
product of this reaction sequence was the thiocarbamate 23.
Here, however, a mixture of the protected desired alcohol 37
and lactone 38 was formed. Rapid purification of the alcohol
followed by coupling with methyl oxalyl chloride gave the
expected oxopimelate 39 as well as additional lactone 38.
In order to complete the synthesis we required tert-butyl
protected oxalate. Thus methyl oxalyl chloride 47 was added to
tert-butanol in the presence of base to form the mixed ester 4817
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034
2025

View Online
(Scheme 7). Selective methyl ester hydrolysis could be achieved
with a variety of bases, but use of 1.0 equivalent of KOH
afforded the potassium salt 49 which was conveniently soluble
in organic media. Coupling with 46a and 46b was achieved
by first forming the mixed anhydride carbonate 50 with ethyl
chloroformate before addition of the alcohols (Scheme 7). The
bis-tert-butyl oxooxapimelate analogues 51a and 51b were
formed in acceptable yield. Attempted acid catalysed deprotec-
tion of 51a and 51b using different concentrations of TFA
in CH2Cl2 at varied temperatures gave the lactone 38 as the
only identifiable product in low yields. It is clear that synthetic
approaches to these oxooxapimelates will have to be devised
which do not use acidic or basic deprotection conditions.
In vitro enzyme activity
DAP-AT was isolated from E. coli DH5α by a procedure
involving rapid sonication of whole cells, cation and anion
chromatography and ultrafiltration. The enzyme fraction
obtained showed high activity with the natural substrate 4 with
no detectable background activity in the absence of substrate.
In the standard assay DAP-AT converts the natural substrate 4
(ca. 1 mM) to the ε--amine 5 (Scheme 8).9 This generates the
Scheme 8 Assay for DAP-AT activity: (i) DAP-AT; (ii) Glutamate
dehydrogenase.
PLP form of the enzyme which then reacts in the reverse
direction with an excess of glutamate 9 (10 mM) to generate
α-ketoglutarate 10. A coupling enzyme, glutamate dehydro-
genase, then converts the α-ketoglutarate 10 rapidly back to -
glutamate 9 with the consumption of ammonium ions and
NADPH. The fall in NADPH concentration is conveniently
monitored at 340 nm in order to obtain rate data.
Substrate activity of 11a, 11b, 12a, 12b and 14
Addition of increasing concentrations of 11a, 11b, 12a, 12b
and 14 to assay solutions of DAP-AT caused no observable
depletion in NADPH concentration, even at elevated concen-
trations in the region of 20–50 mM.
Scheme
7
Synthesis of oxa substrate analogues. Reagents and
Reversible inhibition
conditions: (i) MeOH, SOCl2, quant.; (iia) succinic anhydride, NaHCO3
(aq), 35%; (iib) CbzCl, NaHCO3 (aq), 34%; (iii) CH2Cl2, Ϫ5 ЊC,
isobutene, H2SO4, 88%; (iv) NaOH, H2O–MeOH, 43–68%; (v) NMM,
In order to measure possible reversible inhibition of DAP-AT
by compounds 11a, 11b, 12a, 12b and 14, each was added to a
standard DAP-AT assay mixture simultaneously with substrate
4 at 1 mM. At concentrations up to 10 mM none of the thia-
compounds (11a, 11b or 14) caused any significant reduction of
t
CH2Cl2, EtOCOCl; (vi) NaBH4 (aq) 30–42% (two steps); (vii) BuOH,
pyridine, Et2O, 53%; (viii) 1.0 eq. KOH, H2O–CH3CN, 93%; (ix)
CH3CN, Ϫ10 ЊC, EtOCOCl; (x) 46a–46b, CH3CN, RT, 18–22%
(two steps).
2026
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034

View Online
In all the assays described here -glutamate, the co-substrate
of DAP-AT, was present at 10 mM. In the absence of the sub-
strate 4 it would be expected that DAP-AT would be present
almost wholly in the PMP form. Under these conditions slow
inhibition occurs, perhaps by slow reversible covalent bond
formation. When 4 is present it may be that this cannot happen
because 12a is displaced by substrate before bond formation
can occur.
Antimicrobial properties of 12a
Sterilised 5 mm circles of Whatman number 1 paper were
loaded with standard aqueous solutions of 12a (300, 30, 3.0
and 0.3 µg mLϪ1). The circles were placed on minimal agar
plates (containing no -lysine or DAP) which had been spread
with E. coli DH5α. DAP-AT inhibitors such as the hydrazines 7
and 8 cause significant growth inhibition zones.6 Compound
12a caused no growth inhibition however, commensurate with
its poor DAP-AT inhibitory properties.
Concluding remarks
Thiol ester substrate analogues showed no detectable inter-
action with DAP-AT, either as substrates or inhibitors. Aza
analogue 12a, however, showed reversible inhibition in the
presence of natural substrate 4 and (partial) irreversible
inhibition when 4 was absent. Thus 12a is a member of a new
class of inhibitors of PLP dependent enzymes. However KI vs.
DAP-AT for reversible inhibition was 2.6 mM and significant
irreversible inhibition was manifested only at 12a concen-
trations >2 mM. Kinetic analysis indicated that 12a may
inhibit DAP-AT as intended but the inhibition is much poorer
than that shown by compounds such as hydrazines 7 and 8
which react preferentially with the PLP form of the enzyme.
Unsurprisingly 12a showed no detectable antibacterial activity
vs. E. coli at the highest concentrations tested of 300 µg mLϪ1
(ca. 1 mM). Ester substrate analogues may yet show inhibitory
properties and our synthetic efforts now focus on the use
of alternative protective strategies such as the use of allyl
esters which can be deprotected under neutral Pd catalysed
conditions.
Fig. 1 A Reciprocal plot of inhibition of DAP-AT by 12a. B Time
dependent inhibition of DAP-AT at varying 12a concentrations: ᭺ 20
mM; ᮀ 10 mM; ᭛ 5 mM.
rate of turn-over in comparison with control experiments
lacking inhibitor. However the aza substrate analogue 12a (N-
succinyl) did show inhibition of DAP-AT with an estimated
KI of 2.6 0.5 mM, assuming competitive inhibition (Fig. 1A).
As expected the Cbz protected analogue 12b (N-Cbz) showed
much less activity and although inhibition was detected it was
too feeble to enable accurate inhibition parameters to be
determined.
Experimental
All reagents and solvents were obtained from the Sigma-
Aldrich chemical company and were of ACS grade and not
further purified unless otherwise stated. All anhydrous solvents
were purchased from Fluka and were transferred under dried
N2 gas. NMR spectra were obtained using JEOL ∆-300, ∆-270
and ∆-400 spectrometers operating at 300, 270 and 400 MHz
(1H) and 75.5, 67.9 and 100.7 MHz (13C) respectively. Chemical
shifts are quoted in ppm relative to TMS. Coupling constants
(J) are quoted in Hz. The use of succ in the NMR assignments
refers to the succinyl groups. IR spectra were obtained using
a Perkin Elmer 1600 FTIR spectrometer, using KBr discs for
solids and thin film between NaCl plates for oils. Melting
points were obtained using a Reichert hot-stage apparatus
equipped with microscope and Comark digital thermometer.
Mass spectra were obtained in the indicated mode using a VG
analytical autospec instrument (EI, CI, FAB, accurate mass) or
Fisons VG Quattro spectrometer (ESMS). Optical rotations
were obtained using a Perkin Elmer 141 polarimeter using a
1 dm cell of 1 mL capacity. Flash chromatography was per-
formed according to the method of Still20 or using an impro-
vised automatic system comprising a nitrogen constant pressure
head, column packed with Merck silica gel 60 (0.040–0.063
mm), Gilson Holochrome UV detector set at 254 nM and an
LKB fraction collector. TLC analysis was performed using
Merck glass backed 0.2 mm silica plates (F254) developed with
phosphomolybdic acid when necessary. Enzyme assay methods
Time dependent inhibition
DAP-AT was incubated with all assay components in the
presence of each inhibitor 11a, 11b, 12a, 12b and 14 in the
absence of substrate 4 for set lengths of time. Over this period
the NADPH concentration was also monitored. For all com-
pounds 11a, 11b, 12a, 12b and 14 no NADPH consumption
was observed before addition of 1 mM substrate 4. After
addition, the rate of NADPH consumption was unchanged
compared to control experiments for compounds 11a, 11b and
14. However the N-succinyl amide 12a did cause inhibition,
with the extent of inhibition increasing with pre-incubation
time (Fig 1B). Inhibition did not appear to reach 100% comple-
tion, however. This behaviour can be indicative of slow-binding
inhibition.18,19
In order to test for potential slow binding inhibition the
enzyme and all assay components were combined and the N-
succinyl amide inhibitor 12a included at varying concen-
trations. The reactions were then observed over the period of
60 min. No curvature of the plots was observed showing that
in the presence of substrate 4 there is no time-dependent
inhibition in these assays (data not shown). Overall it appears
that 12a can cause a time dependent inhibition of DAP-AT
only in the absence of the natural substrate 4.
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034
2027

View Online
have been reported elsewhere.6,7,9 Diethyl ether–diazomethane
colourless solid in quantitative yield. Mp 123–125 ЊC; νmax (KBr
was generated according to the procedure of Vogel.21
disc)/cmϪ1 3287, 2950, 1649, 1582, 1411; δH(300 MHz, D2O)
4.22 (1H, m, αCH), 2.70 (2H, m, γCH2), 2.43 (4H, m, 2 × succ
CH2), 2.15 (1H, m, βCH), 1.94 (1H, m, βCH); δC(75.5 MHz,
D2O ϩ CH3CN) 182.0 (COS), 179.9 (CO), 179.5 (CO), 176.1
(CO), 173.8 (CO), 54.8 (αCH), 35.1 (CH2), 33.8 (CH2), 33.1
(CH2), 32.1 (CH2); m/z (FAB)ϩ 307 [(M Ϫ 3Li ϩ 3H)ϩ, 40%],
329 [(M Ϫ 3Li ϩ 2H ϩ Na)ϩ, 30%], 289 [(M Ϫ H2O)ϩ, 15%];
m/z (FAB)Ϫ 328 [(M Ϫ 3Li ϩ H ϩ Na)Ϫ, 100%].
Purification of DAP-AT from E. coli
-Broth, consisting of Difco Bacto Tryptone 10 g, Difco yeast
extract 5 g, NaCl 5 g, glucose 1 g, deionised water to 1 L was
dispensed into 10 × 500 mL conical flasks. The flasks were
sealed with cotton wool and autoclaved at 120 ЊC for 20 min.
E. coli starter culture (10 × 1 mL) was added after cooling. The
flasks were shaken at 37 ЊC, 200 rpm for 8 h (or OD590 of 4–5).
Cells from 1 L were collected by centrifugation (8000 rpm,
20 min, 0 ЊC) and resuspended in 30 mM phosphate buffer
pH 7.5 (10 mL) containing 2 mg LϪ1 PLP at 0 ЊC. The resus-
pended cell pellet was disrupted by sonication at 0 ЊC (6 × 10 s,
with 1 minute cooling intervals at 0 ЊC). The lysed product was
centrifuged (16000 rpm, 10 min, 0 ЊC) and the pellet discarded.
Glycerol was added to the supernatant to 10% and the protein
solution frozen overnight. The frozen solution was thawed and
streptomycin sulfate was added to 2% w/v at 0 ЊC. The mixture
was stirred at 0 ЊC for 15 min, then centrifuged at 8000 rpm for
15 min; the pellet was discarded. The protein solution was
diluted to a final volume of 21 mL with 20 mM phosphate
buffer pH 7.0 and loaded onto a Q-sepharose column (5 × 20
cm). The column was eluted with a linear gradient of 20 mM
phosphate buffer pH 7.0 from 0 to 500 mM NaCl and fractions
were collected at 0 ЊC in polypropylene tubes cooled in ice.
Active fractions (standard assay) were combined, desalted (by
dialysis vs. 20 mM phosphate buffer pH 5.0, three times)
adjusted to pH 5.0 if necessary and loaded onto an S-sepharose
column. A linear pH gradient was run using 20 mM phosphate
buffer from pH 5.0 to pH 8.0 and fractions were collected at
0 ЊC in polypropylene tubes. Active fractions were combined,
glycerol added to 10%, and frozen.
Dilithium (2RS)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-thia-
pimelate monohydrate 11b
Dimethyl
(2RS)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-
thiapimelate (29, 77.2 mg, 209 µmol) was dissolved in a mixture
of CH3CN (0.5 mL) and water (0.5 mL). Lithium hydroxide
monohydrate (2.0 eq., 418 µmol, 17.5 mg) was added and
the mixture stirred at RT until all of the LiOH had dissolved
(90 min). The solvent was removed in vacuo and the residue
was lyophilized to afford dilithium (2RS)-2-[N-(benzyloxy-
carbonyl)amino]-6-oxo-5-thiapimelate monohydrate 11b as a
colourless solid in quantitative yield. Mp 87–89 ЊC; νmax (KBr
disc)/cmϪ1 3090, 2949, 1718, 1696, 1554, 1454, 1323, 1253, 1052;
δH(300 MHz, DMSO-d6) 7.67 (1H, d, J 8.6, NH), 7.33 (5H, m,
Ph), 5.35 (2H, s, CH2Ph), 4.37 (1H, ddd, J 19.5, J 15.6, J 8.6,
αH), 3.37 (1H, m, γCH), 3.26 (1H, m, γCH), 2.43 (1H, m, βH),
2.08 (1H, m, βH); δC(75.5 MHz, DMSO-d6) 205.6 (COS),
172.39 (CO2Li), 172.37 (CO2Li), 155.9 (CO2NH), 136.8 (Ph),
128.4 (Ph), 127.8 (Ph), 126.6 (Ph), 65.3 (OCH2), 59.9 (αCH),
29.8 (γCH2), 26.4 (βCH2); m/z (FAB)ϩ 353 [(M)ϩ, 100%]; Anal.
Calcd for C14H13NO7SLi2ؒH2O: C, 45.30%; H, 4.07%; N, 3.77%;
S, 8.64%. Found: C, 45.44%; H, 3.92%; N, 3.72%; S, 8.78%.
Dilithium (2S)-2-[N-(4-lithiooxy-4-oxobutanoyl)amino]-6-oxo-
5-azapimelate 12a
Inhibition experiments
Dimethyl (2S)-2-[N-(4-methoxy-4-oxobutanoyl)amino]-6-oxo-
5-azapimelate (35, 20.0 mg, 60.2 µmol) was dissolved in a
mixture of CH3CN (0.5 mL) and water (0.5 mL). Lithium
hydroxide monohydrate (3.0 eq., 181 µmol, 7.60 mg) was
added and the mixture stirred at RT until all of the LiOH had
dissolved (90 min). The solvent was removed in vacuo and
the residue was lyophilized to afford dilithium (2S)-2-[N-(4-
lithiooxy-4-oxobutanoyl)amino]-6-oxo-5-azapimelate 12a as a
colourless solid (18.0 mg, 58.3 µmol, 97%). Mp 240–242 ЊC;
νmax (KBr disc)/cmϪ1 3012 (NH), 1705 (CO); δH(300 MHz, D2O)
4.21 (1H, m, αCH), 3.36 (2H, m, γCH2), 2.86 (4H, m, succ
CH2), 2.00 (2H, m, βCH2); δC(67.9 MHz, D2O) 181.5 (CO),
178.8 (CO), 175.7 (CO), 166.5 (CO), 165.4 (CO), 53.5 (αCH),
36.8 (γCH2), 33.3 (succ CH2), 30.9 (succ CH2), 31.5 (βCH2);
m/z (ES)ϩ 308.99 [(MH)ϩ, 100%], 302.99 [(M Ϫ Li ϩ 2H)ϩ,
95%], 297.05 [(M Ϫ 2Li ϩ 3H)ϩ, 85%].
For reversible inhibition experiments the standard DAP-AT
activity assay was used containing 1 mM substrate 4 and all
other reaction components.6,7,9 The inhibitors (11a, 11b, 12a,
12b and 14) were added at increasing concentrations (1–20 mM)
at the same time as the assay was initiated by addition of 4.
The assays were carried out in a 1 mL quartz cuvette at
37 ЊC and the progress of reaction was observed at 340 nM
(Pharmacia LKB ultrospec III spectrophotometer). The KM
of 4 has previously been determined6 and the KI of 12a
was determined from graphical plots assuming competitive
inhibition. For preincubation experiments DAP-AT was
incubated with the inhibitor at fixed concentration in assay
buffer containing all components except 4 at 37 ЊC. Aliquots
were removed at time points and residual activity was deter-
mined using the standard DAP-AT assay.6,7,9 For the slow-
binding assays DAP-AT was added to a standard assay mixture
containing 4 at 1 mM and 12a at fixed concentrations. DAP-AT
was diluted prior to addition (10-fold) to ensure slow reaction
over 3600 s. For all assays NADPH concentration was moni-
tored at 340 nm every 2 s. Data were read into an Excel database
and initial rates were calculated by the use of a best-fit straight
line for the first 60 seconds of reaction.
Dilithium (2S)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-aza-
pimelate hydrate 12b
Dimethyl (2S)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-aza-
pimelate (32, 101 mg, 286 µmol) was dissolved in a mixture
of CH3CN (1 mL) and water (1 mL). Lithium hydroxide
monohydrate (2.0 eq., 571 µmol, 24.0 mg) was added to the
mixture with stirring at RT until all of the LiOH had dissolved
(90 min). The solvent was removed in vacuo and the residue
was lyophilized to afford dilithium (2S)-2-[N-(benzyloxy-
carbonyl)amino]-6-oxo-5-azapimelate 12b as a colourless solid
in quantitative yield. Mp 126–127 ЊC; [α]2D4 Ϫ6.11 (c 2.03 in
H2O); νmax (KBr)/cmϪ1 3397, 3067, 2950, 1664; δH(300 MHz,
D2O) 7.37 (5H, m, Ph), 5.04 (2H, m, OCH2Ph), 3.92 (1H, m,
αH), 3.25–3.15 (2H, m, γH), 1.87 (2H, m, βCH2); δC(75.5 MHz,
D2O) 179.3 (CO), 171.2 (CO), 166.5 (CO), 158.4 (CO), 135.0
(Ph), 129.2 (Ph), 128.8 (Ph), 128.1 (Ph), 67.4 (OCH2Ph), 54.3
(αCH), 37.0 (γCH2), 31.5 (βCH2); m/z (FAB)ϩ 337 [(MH)ϩ,
Dilithium (2RS)-2-[N-(4-lithiooxy-4-oxobutanoyl)amino]-6-oxo-
5-thiapimelate 11a
Dimethyl
(1RS)-1-[N-(4-methoxy-4-oxobutanoyl)amino]-6-
oxo-5-thiapimelate (18, 123 mg, 352 µmol) was dissolved in
a mixture of CH3CN (3 mL) and deionised water (3 mL).
Lithium hydroxide monohydrate (3.0 eq., 1.06 mmol, 44.3 mg)
was added and the mixture stirred at RT until all of the LiOH
had dissolved (90 min). The solvent was removed in vacuo and
the residue was lyophilized to afford dilithium (2RS)-2-[N-(4-
lithiooxy-4-oxobutanoyl)amino]-6-oxo-5-thiapimelate 11a as a
2028
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034

View Online
86%]; m/z (ES)ϩ 338 [(M)ϩ, 61%]; Anal. Calcd for (C14H14-
Li2N2O7)2ؒ5H2O: C, 44.11%; H, 5.02%; N, 7.35%. Found: C,
44.30%; H, 4.97%; N, 7.49%.
Dimethyl
(2RS)-2-[N-(4-methoxy-4-oxobutanoyl)amino]-6-
oxo-5-thiapimelate 18 was obtained by flash chromatography
(50 : 50 EtOAc : hexane, Rf 0.18) as a colourless oil (227 mg,
649 µmol, 49%); νmax (thin film)/cmϪ1 3400, 3340, 2995, 1765,
1710, 1550, 1460; δH(300 MHz, CDCl3) 6.44 (1H, d, J 7.9, NH),
4.69 (1H, ddd, J 4.9, J 7.7, J 12.6, αCH), 3.93 (3H, s, OCH3),
3.78 (3H, s, OCH3), 3.69 (3H, s, OCH3), 3.00 (2H, m, γCH2),
2.63 (4H, m, 2 × succ CH2), 2.21 (1H, m, βCH), 1.99 (1H, m,
βCH); δC(75.5 MHz, CDCl3) 185.1 (COS), 173.3 (CO), 171.9
(CO), 171.5 (CO), 159.3 (CON), 53.8 (αCH), 52.6 (OCH3),
51.9, (OCH3), 51.1 (OCH3), 32.0 (CH2), 30.8 (CH2), 29.9 (CH2),
25.2 (βCH2); m/z (EI)ϩ 350 [(MH)ϩ, 1%], 318 [(M Ϫ OMe)ϩ,
3%], 290 [(M Ϫ CO2Me)ϩ, 3%], 262 [(M Ϫ CO2MeCO)ϩ, 72%],
230 [(M Ϫ CO2MeCOS)ϩ, 40%], 115 [(COCH2CH2CO2Me)ϩ,
100%]; Anal. Calcd for C13H19NSO8: C, 44.69%; H, 5.48%;
N, 4.01%. Found: C, 44.76%; H, 5.63%; N, 4.31%.
Dilithium (2S)-2-[N-(4-lithiooxy-4-oxobutanoyl)amino]-5-oxo-
4-thiaadipate hydrate 14
Dimethyl (2S)-2-[N-(4-methoxy-4-oxobutanoyl)amino]-5-oxo-
4-thiaadipate 22 (23.5 mg, 70.2 µmol) was dissolved in CH3CN
(HPLC grade, 200 µL) in a 2 mL Wheaton vial. H2O (deionised,
100 µL) was added and the solution was stirred at RT.
LiOHؒH2O (8.50 mg, 3.0 eq., 210 µmol) was added in portions
over 2 h. A further 250 µL H2O was added in 50 µL portions
over the same time-span until LCMS analysis indicated
full substrate consumption. H2O (2 mL) was added and the
solution freeze-dried to afford dilithium (2S)-2-[N-(4-
lithiooxy-4-oxobutanoyl)amino]-5-oxo-4-thiaadipate hydrate 14
as a pale yellow solid (22.7 mg, 69.0 µmol, 98%). Mp >220 ЊC;
νmax (KBr)/cmϪ1 3356, 2955, 1720, 1696, 1645, 1395, 1323;
δH(300 MHz, D2O) 4.50 (1H, m, αH), 2.85 (2H, m, βCH2), 2.50
(4H, m, 2 × succ CH2); m/z (ES)ϩ 293.96, [(MH)ϩ, 100%].
Methyl (2S)-N-(4-methoxy-4-oxobutanoyl)cysteinate 20
-Cysteine hydrochloride (2.00 g, 12.7 mmol) was dissolved in
sat. aq. NaHCO3 and the mixture was treated with succinic
anhydride (1.50 eq., 1.90 g, 19.0 mmol). After 3 h of stirring,
the reaction was acidified with conc. aq. HCl and was immedi-
ately extracted with EtOAc (5 × 40 mL). The organic extracts
were dried over Na2SO4 and evaporated in vacuo to afford a
clear colourless oil which was treated with an excess of diethyl
ether–diazomethane. Finally, removal of solvent in vacuo gave a
crude residue which was purified by flash chromatography (50%
EtOAc in hexane, Rf 0.28) to afford methyl (2S)-N-(4-methoxy-
4-oxobutanoyl)cysteinate 20 as a colourless solid (580 mg, 2.30
mmol, 12%); mp 73–75 ЊC; [α]2D4 ϩ35.96 (c 1.09 in CH2Cl2);
δH(300 MHz, CDCl3) 6.68 (1H, d, J 7.0, NH), 4.89 (1H, m, αH),
3.80 (3H, s, CO2CH3), 3.70 (3H, s, CO2CH3), 3.02 (2H, m,
βCH2), 2.65 (4H, m, 2 × succ CH2), 1.44 (1H, t, J 9.0, SH);
δC(67.9 MHz, CDCl3) 173.2 (CO), 171.3 (CO), 170.6 (CO),
53.7 (αCH), 52.6 (OCH3), 51.7 (OCH3), 30.8 (succ CH2), 29.2
(succ CH2), 26.8 (βCH2); m/z (EI)ϩ 248 [(M Ϫ H)ϩ, 54%], 115
[(COCH2CH2CO2Me)ϩ, 100%] (249.0671 calcd for C9H15NO5S
found 249.0665); Anal. Calcd for C9H15NO5S: C, 43.36%;
H, 6.06%; N, 5.62%. Found: C, 43.43%; H, 5.94%; N, 5.41%.
Methyl (2RS)-2-[N-(4-methoxy-4-oxobutanoyl)amino]homo-
cysteinate 17
Racemic homocysteine 15 (1.80 g, 13.3 mmol) was dissolved
in saturated aqueous NaHCO3 (36 mL). Succinic anhydride
(1.33 g, 13.3 mmol) was added in portions over 1 h. After
a further 1 h the reaction was acidified by addition of 1 M
aqueous HCl and extracted quickly with EtOAc (3 × 30 mL).
The combined organic extracts were dried (MgSO4) and
evaporated in vacuo. The residue was treated with an excess of
an ethereal solution of diazomethane. Excess diazomethane
was destroyed by the dropwise addition of acetic acid and
solvent was removed in vacuo. Methyl (2RS)-2-[N-(4-methoxy-
4-oxobutanoyl)amino]homocysteinate 17 was obtained by flash
chromatography of the residue (50 : 50 EtOAc : hexane, Rf
0.16) as a colourless oil (1.10 g, 4.18 mmol, 31%). νmax (thin
film)/cmϪ1 2890, 2480, 1705, 1640, 1520, 1500; δH(300 MHz,
CDCl3) 6.69 (1H, d, J 8.1, NH), 4.76 (1H, m, αH), 3.76 (3H, s,
OCH3), 3.69 (3H, s, OCH3), 2.68 (2H, m, succCH2), 2.58 (4H,
m, succCH2 ϩ γCH2), 2.15 (1H, m, βCH), 2.00 (1H, m, βCH),
1.63 (1H, t, J 8.3, SH); δC(75.5 MHz, CDCl3) 173.3 (CO2Me),
172.5 (CO2Me), 171.6 (CONH), 52.6 (αCH), 51.9 (OCH3),
51.0 (OCH3), 36.7 (CH2S), 30.7 (succCH2), 29.2 (succCH2), 20.7
(βCH2); m/z (EI) 264 [(MH)ϩ, 3%], 263 [(M)ϩ, 8%], 262
[(M Ϫ H)ϩ, 18%], 230 [(M Ϫ SH)ϩ, 13%], 115 [(CO(CH2)2-
CO2Me)ϩ, 100%]; Anal. Calcd for C10H17NO5S: C, 45.62%;
H, 6.46%; N, 5.32%; S, 12.15%. Found: C, 45.61%; H, 6.35%;
N, 5.22%; S, 11.97%.
Dimethyl
(2S,7S)-2,7-bis[N-(4-methoxy-4-oxobutanoyl)-
amino]-4,5-dithiasuberate 21. Compound 21 was obtained as a
by-product from the column (50% EtOAc in hexane, Rf 0.25)
as a colourless solid (970 mg, 1.96 mmol, 31%); δH(300 MHz,
CDCl3) 6.79 (2H, d, J 7.7, NH), 4.81 (2H, m, αH), 3.75 (6H, s,
OCH3), 3.70 (6H, s, CO2CH3), 3.50–3.28 (4H, m, βCH2), 2.94–
2.52 (8H, m, 4 × succ CH2); m/z (ES)ϩ 496.99 [(MH)ϩ, 100%],
247.95 [(disulfide cleavage)ϩ, 15%].
Mixed fractions containing 16 were also eluted from the
column (50 : 50 EtOAc : hexane, Rf 0.13). Selected data: δH(300
MHz, CDCl3) 6.59 (1H, d, J 7.9, NH), 4.65 (1H, m, αH),
3.76 (3H, s, OCH3), 3.70 (6H, s, 2 × OCH3), 2.91 (4H, m), 2.67
(6H, m), 2.13 (1H, m, βCH), 1.96 (1H, m, βCH); m/z (EI)
378 [(MH)ϩ, 1%], 262 [(M Ϫ succinyl)ϩ, 26%], 230 [(M Ϫ
succinyl Ϫ SH)ϩ, 15%], 115 [(CO(CH2)2CO2Me)ϩ, 100%].
Dimethyl (2S)-2-[N-(4-methoxy-4-oxobutanoyl)amino]-5-oxo-4-
thiaadipate 22
To a solution of methyl (2S)-N-(4-methoxy-4-oxobutanoyl)-
cysteine 20 (500 mg, 2.01 mmol) stirred in anhydrous CH2Cl2
(20 mL) under N2 was added 2,6-lutidine (1.1 eq., 261 µL, 2.21
mmol), followed by the dropwise addition of methyl oxalyl
chloride (1.1 eq., 203 µL, 2.21 mmol). The solution became
bright yellow and was stirred at RT under N2 for 2 h. The
reaction mixture was extracted with H2O (3 × 40 mL) and sat.
CuSO4(aq) (3 × 40 mL), organic extracts were dried over
Na2SO4 and evaporated in vacuo. Dimethyl (2S)-2-[N-(4-
methoxy-4-oxobutanoyl)amino]-5-oxo-4-thiaadipate 22 was
obtained by flash chromatography of the residue (50% EtOAc
in hexane, Rf 0.15) as a colourless solid (483 mg, 1.44 mmol,
72%); mp 69–70 ЊC; [α]D24 ϩ35.48 (c 1.26 in CH2Cl2); δH(300
MHz, CDCl3) 6.70 (1H, d, J 7.7, NH), 4.88 (1H, m, αH), 3.94
(3H, s, OCH3), 3.77 (3H, s, OCH3), 3.68 (3H, s, OCH3), 3.61–
3.41 (2H, m, βCH2), 2.70–2.52 (4H, m, 2 × succ CH2); δC(67.9
MHz, CDCl3) 184.6 (CO), 173.2 (CO), 171.5 (CO), 170.3 (CO),
159.1 (CO), 54.0 (αCH), 53.0 (OCH3), 51.9 (OCH3), 51.5
Dimethyl (2RS)-2-[N-(4-methoxy-4-oxobutanoyl)amino]-6-oxo-
5-thiapimelate 18
Methyl (2RS)-2-[N-(4-methoxy-4-oxobutanoyl)amino]homo-
cysteinate (17, 347 mg, 1.32 mmol) was dissolved in anhydrous
CH2Cl2 (10 mL) and stirred at RT under dry N2. 2,6-Lutidine
(156 mg, 169 µL, 1.45 µmol) was added dropwise to give a
bright yellow solution. Methyl oxalyl chloride (1.58 µmol,
194 mg, 146 µL) was added dropwise and the colour faded.
After 30 min the reaction was extracted with a mixture of
CH2Cl2 (10 mL) and dilute aqueous CuSO4 (to remove excess
thiol which co-elutes with the product during chromatography).
The organic layer was dried (MgSO4) and evaporated in vacuo.
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034
2029

View Online
(OCH3), 31.0 (succ CH2), 30.7 (succ CH2), 29.2 (βCH2);
m/z (ES)ϩ 336.08 [(MH)ϩ, 100%], 303.95, [(M Ϫ MeOH)ϩ,
100%] (336.0753 calcd for C12H18NO8S, found 336.0767); Anal.
Calcd for C12H17NO8S: C, 43.00%; H, 5.07%; N, 4.18%. Found:
C, 43.22%; H, 5.26%; N, 4.22%.
(CH2S), 20.6 (βCH2); m/z (CI, NH3)ϩ 283 [(M)ϩ, 0.1%],
282 [(M Ϫ H)ϩ, 1%], 91 [(C7H7)ϩ, 100%]; Anal. Calcd for
C13H17NO4S: C, 55.11%; H, 6.05%; N, 4.94%; S, 11.31%.
Found: C, 55.28%; H, 6.21%; N, 4.72%; S, 11.18%.
(3RS)-3-[N-(Benzyloxycarbonyl)amino]-2-oxo-tetrahydrothio-
phene 27.24 Compound 27 was obtained as a byproduct from
the column as a colourless solid (140 mg, 558 µmol, 35%).
Mp 97–100 ЊC, (lit.25 106 ЊC); νmax (KBr disc)/cmϪ1 3321, 3061,
2947, 1694, 1547, 1454; δH(300 MHz, CDCl3) 7.36 (5H, m, Ph),
5.25 (1H, br s, NH), 5.12 (2H, s, OCH2), 4.33 (1H, m, αCH),
3.28 (m, 2H, γCH2), 2.87 (1H, m, βCH), 1.95 (1H, ddd, J 7.0,
J 12.3, J 24.8, βCH); δC(75.5 MHz, CDCl3) 204.7 (COS), 156.1
(CONH), 136.0 (Ph), 128.6 (Ph), 128.3 (Ph), 128.2 (Ph), 67.3
(OCH2), 60.8 (αCH), 32.0 (γCH2), 27.2 (βCH2); m/z (CI, CH4)ϩ
252 [(MH)ϩ, 35%], 91 [(C7H7)ϩ, 100%]; m/z (EI)ϩ 251.0617
(M)ϩ, calcd for C12H13NO3S 251.0616; Anal. Calcd for
C12H13NO3S: C, 57.35%; H, 5.21%; N, 5.57%; S, 12.76%.
Found: C, 57.18%; H, 4.95%; N, 5.42%; S, 12.72%.
Dimethyl (2RS, 2ЈRS)-homocystinate dihydrochloride dihydrate
2522
Racemic homocystine 24 (1.00 g, 3.73 mmol) was stirred with
10% HCl in anhydrous methanol (50 mL) at room temperature
for 16 h. Methanol and HCl were removed by evaporation in
vacuo to afford a clear oil. The oil was crystallised by repeated
evaporation from dry CH2Cl2 and then dry toluene. The di-
methyl ester dihydrochloride was obtained in quantitative yield
as a hydroscopic colourless powder after air drying. νmax (KBr
disc)/cmϪ1 3415, 2958, 1992, 1740, 1617, 1507, 1412; δH(300
MHz, D2O) 4.65 (1H, dd, J 6.1, J 6.2, αCH), 3.71 (3H, s,
OCH3), 2.72 (2H, t, J 7.1, γCH2), 2.25 (2H, m, βCH2); δC(75.5
MHz, D2O) 173.5 (CO2CH3), 54.4 (OCH3), 52.3 (αCH), 33.5
(γCH2), 29.5 (βCH2); Anal. Calcd for C10H11N2O4S2Cl2ؒ2H2O:
C, 29.63%; H, 6.47%. Found: C, 30.01%; H, 6.26%.
Dimethyl (2RS)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-
thiapimelate 29
Methyl (2RS)-N-(benzyloxycarbonyl)homocysteinate (28,
42.2 mg, 149 mmol) was dissolved in anhydrous CH2Cl2
(300 µL) and stirred under dry N2 in a sealed dry Wheaton vial.
2,6-Lutidine (17.3 µL, 164 µmol) was added to give a bright
yellow solution. Methyl oxalyl chloride (20 µL, 224 µmol) was
added and the yellow colour faded. The reaction mixture
was stirred at RT for 30 min and then applied directly to a flash
chromatography column. Dimethyl (2RS)-2-[N-(benzyloxy-
carbonyl)amino]-6-oxo-5-thiapimelate 29 was obtained by flash
chromatography (40 : 60 EtOAc : hexane, Rf 0.37) as a colour-
less oil (48.2 mg, 131 µmol, 88%); νmax (film)/cmϪ1 3358, 3034,
2955, 2852, 1740, 1686, 1522, 1437; δH(300 MHz, CDCl3) 7.35
(5H, m, Ph), 5.46 (1H, d, J 7.9, NH), 5.12 (2H, s, CH2Ph), 4.47
(1H, m, αCH), 3.92 (3H, s, OCH3), 3.77 (3H, s, OCH3), 3.01
(2H, m, γCH2), 2.20 (1H, m, βCH), 1.98 (1H, m, βCH); δC(75.5
MHz, CDCl3) 185.0 (COS), 171.9 (CO), 159.3 (CO), 157.5
(CO), 136.0 (Ph), 128.6 (Ph), 128.3 (Ph), 128.2 (Ph), 67.2
(OCH2), 53.9 (OCH3), 53.1 (αCH), 52.7 (OCH3), 32.1 (γCH2),
25.2 (βCH2); m/z (CI)ϩ 370 [(MH)ϩ, 30%], 91 [(C7H7)ϩ, 100%];
m/z (EI)ϩ 369.0900, 369.0882 calcd for C16H19NO7S.
Dimethyl (2RS, 2ЈRS)-N,NЈ-bis(benzyloxycarbonyl)dihomo-
cystinate 2623
Dimethyl (2RS, 2ЈRS)dihomocystinate dihydrochloride 25
(200 mg, 542 µmol) was dissolved in CH2Cl2 (10 mL) containing
triethylamine (4.00 eq., 2.17 mmol, 219 mg). The solution was
stirred at RT and benzyloxycarbonyl chloride (1.19 mmol,
203 mg, 170 µL) added dropwise over 2 min. After a further
90 min the solution was acidified by the addition of 1 M
aqueous HCl (20 mL) and extracted into CH2Cl2 (3 × 50 mL).
The organic extracts were combined, dried (MgSO4) and
evaporated in vacuo. Dimethyl (2RS, 2ЈRS)-N,NЈ-bis(benzyl-
oxycarbonyl)dihomocystinate 26 was obtained by flash
chromatography of the residue (50 : 50 EtOAc : hexane, Rf 0.4)
as an oil (81 mg, 145 µmol, 27%). νmax (thin film)/cmϪ1 3027,
2941, 1707, 1531, 1445, 1344; δH(300 MHz, CDCl3) 7.34 (5H,
m, Ph), 5.50 (1H, br s, NH), 5.11 (2H, s, OCH2), 4.51 (1H, m,
αCH), 3.75 (3H, s, OCH3), 2.68 (2H, m, γCH2), 2.24 (1H,
m, βCH), 2.05 (1H, m, βCH); δC(75.5 MHz, CDCl3) 172.3
(CO2Me), 156.0 (CO2NH), 136.1 (Ph), 128.6 (Ph), 128.2 (Ph),
126.98 (Ph), 67.1 (OCH2), 52.9 (αCH), 52.6 (OCH3), 34.4
(βCH2), 32.4 (γCH2); m/z (CI)ϩ 565 [(MH)ϩ, 0.3%], 457
[(M Ϫ BnOH ϩ H)ϩ, 3%], 282 [(disulfide cleavage), 10%], 91
[(C7H7)ϩ, 100%].
Methyl (2S)-2-[N-(benzyloxycarbonyl)amino]-4-aminobutyrate
hydrochloride 31
(2S)-2-[N-(Benzyloxycarbonyl)amino]-4-aminobutyric acid 30
(Fluka, 200 mg, 793 µmol) was dissolved in 10% HCl in
anhydrous methanol (20 mL). The solution was stirred for
16 h at RT and the solvent was then removed by evaporation in
vacuo to afford methyl (2S)-2-[N-(benzyloxycarbonyl)amino]-
4-aminobutyrate hydrochloride 31 as a highly hydroscopic
colourless solid (240 mg, 661 µmol, 83%). [α]2D4 Ϫ21.05 (c 1.2 in
CH2Cl2); νmax (film)/cmϪ1 3380, 1705, 1532, 1439; δH(270 MHz,
CDCl3) 8.07 (3H, br s, NH3Cl), 7.26 (5H, br s, Ph), 6.25 (1H, br
s, NH), 4.95 (2H, br s, OCH2Ph), 4.35 (1H, m, αH), 3.50 (3H, s,
OCH3), 3.10 (2H, m, γCH2), 2.25 (1H, m, βCH), 2.10 (1H, m,
βCH); δC(75.5 MHz, CDCl3) 172.2 (CO2Me), 156.6 (CO2NH),
136.1 (Ph), 128.5 (Ph), 128.1 (Ph), 128.0 (Ph), 67.1 (OCH2), 52.8
(OCH 3), 51.9 (αCH), 37.1 (γCH2), 29.4 (βCH2); m/z (CI)ϩ 268
[(MH)ϩ, 30%], 267 [(M)ϩ, 100%], 250 [(M Ϫ NH3)Hϩ, 15%],
249 [(M Ϫ NH3)ϩ, 50%], 91 [(C7H7)ϩ, 60%].
Methyl (2RS)-N-(benzyloxycarbonyl)homocysteinate 28
Dimethyl (2RS,2ЈRS)-N,NЈ-bis(benzyloxycarbonyl)dihomo-
cystinate (26, 450 mg, 798 µmol) was dissolved in anhydrous
methanol (10 mL). An excess of 3% sodium amalgam (1.80 g,
approx. 2.40 mmol Na) was added and the mixture stirred
vigorously under dry N2 at 0 ЊC until all the amalgam had
reacted. Mercury was removed by paper filtration and the
methanolic solution acidified by the addition of 1 M aqueous
HCl. The mixture was extracted with CH2Cl2 (3 × 30 mL). The
combined organic extracts were dried (MgSO4) and evaporated
in vacuo. Methyl (2RS)-N-(benzyloxycarbonyl)homocysteinate
28 was obtained by flash chromatography of the residue (30 : 70
EtOAc : hexane, Rf 0.32) as a colourless solid (111 mg, 389
µmol, 24%). Mp 62–63 ЊC; νmax (thin film from CH2Cl2)/cmϪ1
3027, 2953, 2849, 2568, 1715, 1525, 1438, 1344; δH(300 MHz,
CDCl3) 7.35 (5H, m, Ph), 5.41 (1H, br d, J 7.9, NH), 5.11 (2H,
s, OCH2), 4.75 (1H, ddd, J 13.2, J 8.2, J 4.9, αH), 3.75 (3H, s,
OCH3), 2.57 (2H, m, γCH2), 2.13 (1H, m, βCH), 1.98 (1H, m,
βCH), 1.55 (1H, t, J 8.2, SH); δC(75.5 MHz, CDCl3) 172.4
(CO2Me), 156.0 (CO2NH), 136.1 (Ph), 128.6 (Ph), 128.3 (Ph),
128.2 (Ph), 67.2 (OCH2), 52.7 (αCH), 52.6 (OCH3), 37.0
Dimethyl (2S)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-
azapimelate 32
Methyl oxalyl chloride (220 µL, 2.40 mmol) was added to a
solution of hydroxybenzotriazole (HOBT, 324 mg, 2.40 mmol)
in anhydrous CH2Cl2 (3 mL). The solution was stirred at
2030
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034

View Online
RT under dry N2 for 45 min. A solution of methyl (2S)-2-
[N-(benzyloxycarbonyl)amino]-4-aminobutyrate hydrochloride
(31, 242 mg, 798 µmol) in anhydrous CH2Cl2 (2 mL) was added
dropwise with stirring. After 1 h the reaction was quenched
by the addition of water (10 mL) and CH2Cl2 (10 mL). The
mixture was extracted with 1 M aqueous HCl (1 × 25 mL), then
1 M aqueous NaHCO3 (1 × 25 mL) and finally water (1 × 25
mL). The organic layer was dried (MgSO4) and evaporated
in vacuo. Dimethyl (2S)-2-[N-(benzyloxycarbonyl)amino]-6-
oxo-5-azapimelate 32 was obtained by flash chromatography
(20 : 80 Et2O–CH2Cl2, Rf 0.30) as a colourless oil (150 mg, 426
µmol, 53%). [α]D24 Ϫ21.05 (c 1.53 in CH2Cl2); νmax (film)/cmϪ1
3328, 1693, 1531; δH(300 MHz, CDCl3) 7.33 (5H, m, Ph), 5.76
(1H, d, J 8.04, NH), 5.10 (2H, m, OCH2Ph), 4.42 (1H, m, αH),
3.87 (3H, s, OCH3), 3.73 (3H, s, OCH3), 3.62–3.24 (2H,
m, γCH2), 2.00 (2H, m, βCH2); δC(75.5 MHz, CDCl3) 172.3
(CO), 160.8 (CO), 157.9 (CO), 156.7 (CO), 136.0 (Ph), 128.6
(Ph), 128.3 (Ph), 128.1 (Ph), 67.2 (OCH2), 53.6 (αCH), 53.5
(OCH3), 53.1 (OCH3), 36.0 (γCH2), 32.2 (βCH2); m/z (CI,
NH3)ϩ 353 [(MH)ϩ, 58%], 91 [(C7H7)ϩ, 100%]. Anal. Calcd. for
C16H20N2O7: C, 54.54%; H, 5.72%; N, 7.95%. Found: C,
54.29%; H, 5.98%; N, 7.89%.
in CH2Cl2); νmax (film)/cmϪ1 2961, 1733; δH(300 MHz, CDCl3)
8.00 (1H, br m, NH), 6.42 (1H, d, J 7.2, NH), 4.60 (1H, m, αH),
3.84 (3H, s, OCH3), 3.69 (3H, s, OCH3), 3.62 (3H, s, OCH3),
3.00 (2H, m, γCH2), 2.52 (4H, m, succ CH2), 2.15 (1H, m,
βCH2), 2.68 (1H, m, βCH2); δC(67.9 MHz, CDCl3) 174 (CO),
172.5 (CO), 171.4 (CO), 162.6 (CO), 157.2 (CO), 53.8 (αCH),
52.8 (OCH3), 51.9 (OCH3), 49.8 (OCH3), 35.8 (γCH2), 32.7
(succ CH2), 30.9 (succ CH2), 29.1 (βCH2); m/z (CI) 332 [(M)ϩ,
14%], 115 [(MeO2CCH2CH2CO)ϩ, 100%]; m/z (EI)ϩ 333
[(MH)ϩ, 2%], 115 [(MeO2CCH2CH2CO)ϩ, 100%] (318.1074
calcd for C12H17N2O7, found 318.1063).
Dimethyl (2RS)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-
oxapimelate 39
-Homoserine 36 (NovaBiochem, 201 mg, 1.69 mmol) was
added to a vigorously stirred aqueous solution of NaHCO3
(1 M, 8 mL) containing benzyloxycarbonyl chloride (291 µL,
2.04 mmol). The mixture was stirred for 16 h at RT. The
reaction mixture was cooled to 0 ЊC and acidified to pH 3 with
1 M aqueous HCl. The cold solution was extracted with ethyl
acetate (3 × 20 mL). The pH of the aqueous phase was
readjusted to 3 after each extraction. The combined organic
extracts were dried (MgSO4) and evaporated in vacuo. The
colourless solid residue was treated with an excess of an
ethereal solution of diazomethane and solvent was removed
in vacuo. The resultant unstable hydroxy ester 37 (428 mg, 1.60
mmol, 95%) was not routinely purified further, but used
immediately. Methyl (2S)-2-[N-(benzyloxycarbonyl)amino]-4-
hydroxybutyrate 37 selected data: νmax (thin film)/cmϪ1 3035,
1758, 1519, 1438; δH(300 MHz, CDCl3) 7.33 (5H, m, Ph), 5.80
(1H, br d, J 7.7, NH), 5.13 (1H, d, J 12.3, OCHHPh), 5.08 (1H,
d, J 12.3, OCHHPh), 4.54 (1H, m, αH), 3.74 (3H, s, OCH3),
3.69 (2H, m, γCH2), 3.03 (1H, br s, OH), 2.10 (1H, m, βCH),
1.75 (1H, m, βCH); δC(75.5 MHz, CDCl3) 173.0 (CO2Me),
156.7 (CO2NH), 136.1 (Ph), 128.6 (Ph), 128.3 (Ph), 128.1 (Ph),
67.2 (OCH2), 58.4 (γCH2), 54.1 (OCH3), 51.3 (αCH), 35.4
(βCH2); m/z (EI)ϩ 267 [(M)ϩ, 0.7%], 249 [(M Ϫ H2O)ϩ, 15%], 91
[(C7H7)ϩ, 100%].
Dimethyl (2S)-2-amino-6-oxo-5-azapimelate hydrochloride 34
Dimethyl (2S)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-aza-
pimelate (32, 400 mg, 1.14 mmol) was dissolved in 10% chloro-
form in methanol. 10% Palladium on carbon (40.0 mg) was
added and the mixture was stirred overnight under hydrogen (1
atm). The solution was filtered through Celite and evaporated
in vacuo to afford dimethyl (2S)-2-amino-6-oxo-5-azapimelate
hydrochloride (34, 270 mg, 93%) as a colourless hydroscopic
solid. [α]2D4 Ϫ2.62 (c 0.183 in CH2Cl2); νmax (film)/cmϪ1 3359,
3235, 2957, 1750, 1692, 1527, 1441; δH(300 MHz, CDCl3) 8.62
(3H, br s, NH3Cl), 4.37 (1H, m, αH), 3.96 (3H, s, OCH3), 3.87
(3H, s, OCH3), 3.67 (2H, m, γCH2), 2.43 (2H, m, βCH2); δC(67.9
MHz, CDCl3) 170.3 (CO), 160.9 (CO), 157.5 (CO), 53.7 (αCH),
51.2 (OCH3), 50.8 (OCH3), 36.0 (γCH2), 29.0 (βCH2); m/z (CI)ϩ
219 [(MH)ϩ, 29%], 201 [(M Ϫ NH3ϩH)ϩ, 70%]; m/z (EI) 219
[(MH)ϩ, 80%] (219.0981 calculated for C8H15N2O5, found
219.0981).
To a solution of the alcohol 37 as prepared above (60.0 mg,
225 µmol) in CH2Cl2 (2 mL) was added dry pyridine (4.0 eq.,
253 µmol, 13.2 µL, 20.0 mg) and methyl oxalyl chloride (3.0 eq.,
245µmol, 21.0µl, 30.0 mg). The solution was stirred for 2 h at
RT, then applied directly to a flash chromatography column.
Dimethyl (2S)-2-[N-(4-methoxy-4-oxobutanoyl)amino]-6-oxo-5-
azapimelate 35
Dimethyl
(2RS)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-
Dimethyl (2S)-2-amino-6-oxo-5-azapimelate hydrochloride
(34, 270 mg, 1.06 mmol) was dissolved in saturated aqueous
NaHCO3 (15 mL). Succinic anhydride (180 mg, 1.80 mmol) was
added in portions over 1 h. After a further 1 h the reaction was
acidified by addition of 1 M aqueous HCl and extracted
quickly with EtOAc (3 × 30 mL). The combined organic
extracts were dried (Na2SO4) and evaporated in vacuo. The
residue was treated with an excess of an ethereal solution of
diazomethane. Excess diazomethane was destroyed by the
dropwise addition of acetic acid and solvent was removed in
vacuo. Dimethyl (2S)-2-[N-(4-methoxy-4-oxobutanoyl)amino]-
6-oxo-5-azapimelate 35 was obtained by flash chromatography
of the residue (30 : 70 EtOAc–hexane, Rf 0.23) as a clear oil
(25.3 mg, 76.2 µmol, 7%).
oxapimelate 39 was obtained by flash chromatography (50 : 50
EtOAc–hexane, Rf 0.38) as a colourless oil (48.7 mg, 142 µmol,
63%); νmax (thin film)/cmϪ1 3034, 2957, 1746, 1523, 1439, 1319;
δH(300 MHz, CDCl3) 7.29 (5H, m, Ph), 5.52 (1H, br d, J 7.6,
NH), 5.04 (2H, s, OCH2Ph), 4.45 (1H, m, αH), 4.30 (2H, m,
γCH2), 3.80 (3H, s, OCH3), 3.70 (3H, s, OCH3), 2.26 (1H, m,
βCH), 2.14 (1H, m, βCH); δC(75.5 MHz, CDCl3) 171.8 (CO),
157.6 (CO), 157.0 (CO), 155.8 (CO), 136.0 (Ph), 128.5 (Ph),
128.2 (Ph), 128.1 (Ph), 67.1 (OCH2Ph), 62.9 (γCH2), 53.6
(OCH3), 52.7 (OCH3), 51.2 (αCH), 30.8 (βCH2); m/z (CI,
NH3)ϩ 354 [(MH)ϩ, 32%], 250 [(M Ϫ CH3OCOCO2H)ϩ, 40%],
91 [(C7H7)ϩ, 100%]; Anal. Calcd for C16H19NO8: C, 54.39%;
H, 5.42%; N, 3.96%. Found: C, 54.12%; H, 5.52%; N, 4.10%.
(3RS)-2-Oxo-3-[N-(benzyloxycarbonyl)amino]tetrahydrofuran
38.26,27 Obtained as a byproduct from the column (50 : 50
EtOAc : hexane, Rf 0.29) as a colourless solid (11.0 mg, 47.0
µmol, 21%). Mp 125.5–126 ЊC (lit.28 129–130 ЊC); νmax (KBr
disc)/cmϪ1 3306, 1779; δH(300 MHz, CDCl3) 7.26 (5H, m, Ph),
5.58 (1H, br d, J 5.4, NH), 5.04 (2H, s, OCH2Ph), 4.31 (2H, m,
γCH2), 4.12 (1H, m, αCH), 2.62 (1H, m, βCH), 2.14 (1H, m,
βCH); δC(75.5 MHz, CDCl3) 175.0 (CO2), 156.0 (CON), 135.8
(Ph), 128.5 (Ph), 128.3 (Ph), 128.1 (Ph), 67.3 (OCH2), 65.7
(γCH2), 50.4 (αCH), 30.3 (βCH2); m/z (CI, NH3)ϩ 236 [(MH)ϩ,
27%], 91 [(C7H7)ϩ, 100%].
Alternative route. Dimethyl (2S)-2-amino-6-oxo-5-aza-
pimelate hydrochloride 34 (138 mg, 543 µmol) was dissolved in
CHCl3 (10 mL). Et3N (1.0 eq., 76.0 µL, 543 µmol) was added
followed by succinic anhydride (2.0 eq., 109 mg, 1.09 mmol),
in portions, over 30 min. After a further 2 h the solvent was
removed in vacuo and the residue treated with an excess of an
ethereal solution of CH2N2. Dimethyl (2S)-2-[N-(4-methoxy-
4-oxobutanoyl)amino]-6-oxo-5-azapimelate 35 was obtained by
flash chromatography of the residue (10% CH3CN in EtOAc,
Rf 0.5) as a clear oil (137 mg, 413 µmol, 76%). [α]2D4 Ϫ37.3 (c 1.2
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034
2031

View Online
ꢀ-Methyl hydrogen (2S)-aspartate hydrochloride 41b29
ꢀ-Methyl ꢁ-tert-butyl (2S)-N-(benzyloxycarbonyl)aspartate 43b
-Aspartic acid 41a (10.0 g, 76.9 mmol) was added to methanol
(glass distilled, 50 mL) and cooled to Ϫ10 ЊC. Thionyl chloride
(7.75 mL) was added dropwise to the mixture, the cooling
bath was removed and the solution was slowly warmed to
room temperature. After standing for 25 minutes, diethyl ether
(50 mL) was added to the mixture and upon cooling and
shaking, the product 41b was precipitated as a colourless solid
which was filtered immediately, washed with ice cold diethyl
ether (20 mL) and collected as a colourless solid (9.83 g, 58.0
mmol, 75%); mp 183–184 ЊC (lit.30 186–188 ЊC); νmax (KBr disc)/
cmϪ1 3415 (NH), 1733 (CO); δH(300 MHz, D2O), 4.19 (1H, m,
αH), 3.57 (3H, s, OCH3), 3.00 (2H, m, βH); δC(67.9 MHz, D2O)
172.1 (CO), 171.0 (CO), 54.1 (OCH3), 49.6 (αCH), 34.1 (βCH2);
m/z (CI)ϩ 148 [(MH)ϩ, 100%].
A solution of β-methyl hydrogen (2S)-N-(benzyloxycarbonyl)-
aspartate (42b, 2.33 g, 8.29 mmol) in dichloromethane (20 mL)
was cooled to Ϫ5 ЊC and sulfuric acid (conc., 0.1 mL) was
added to the solution. Isobutylene (2.0 g, 35.6 mmol) was
passed into the solution which was stirred at room temperature
for 24 h. Dichloromethane (50 mL) was added to the organic
layer which was washed with saturated aqueous sodium
bicarbonate (2 × 50 mL) then water (1 × 50 mL), then dried
over sodium sulfate. The solvent was evaporated in vacuo to give
β-methyl α-tert-butyl (2S)-N-(benzyloxycarbonyl)aspartate as a
colourless oil (2.46 g, 7.31 mmol, 88%); νmax(film)/cmϪ1 3380,
3080, 3040, 3000, 2960, 1740, 1520; δH(300 MHz, CDCl3) 7.70
(5H, m, Ph), 6.43 (1H, d, J 8.00, NH), 5.56 (2H, s, OCH2Ph),
4.95 (1H, m, αH), 4.05 (3H, s, OCH3), 3.40–3.20 (2H, m, βH);
δC(67.9 MHz, CDCl3) 171.4 (CO), 169.8 (CO), 157.5 (CO),
138.1 (Ph), 130.7 (Ph), 128.5 (Ph), 128.1 (Ph), 82.6 (OC(CH3)3),
67.0 (OCH2), 51.8 (OCH3), 50.9 (αCH), 36.8 (βCH2), 27.8
(OC(CH3)3); m/z (CI)ϩ 338 [(MH)ϩ, 1%], 282 [(M Ϫ tBu ϩ H)ϩ,
ꢀ-Methyl hydrogen (2S)-N-(benzyloxycarbonyl)aspartate 42b31
β-Methyl hydrogen (2S)-aspartate hydrochloride 41b (4.78 g,
28.2 mmol) was dissolved at room temperature in a solution
made up of water (50 mL), sodium hydroxide (4 M, 6.25 mL)
and sodium bicarbonate (4.32 g). Cbz-chloride (4.38 mL,
30.7 mmol) was added dropwise over 10 minutes with vigorous
stirring. The mixture was then extracted with diethyl ether
(3 × 30 mL) and the ether extracts were discarded. The aqueous
phase was acidified to pH 2.0 with aqueous hydrochloric acid
and then extracted with toluene (2 × 30 mL). The combined
organic layers were washed with water (2 × 30 mL), dried
and evaporated to give the title compound as a colourless oil
(2.72 g, 9.70 mmol, 34%). νmax(film)/cmϪ1 3420, 3125, 2960,
1740, 1460, 1540; δH(300 MHz, CDCl3) 7.40 (5H, m, Ph), 6.24
(1H, d, J 8.00, NH), 5.32 (2H, s, OCH2Ph), 4.80 (1H, m, αCH),
3.78 (3H, s, OCH3), 3.20–2.90 (2H, m, βCH2); δC(67.9 MHz,
CDCl3) 175.3 (CO), 171.2 (CO), 156.2 (CO), 129.1 (Ph), 128.6
(Ph), 128.3 (Ph), 127.0 (Ph), 67.4 (OCH2), 52.2 (OCH3), 50.3
(αCH), 36.3 (βCH2); m/z (CI)ϩ 282 [(MH)ϩ, 54%], 91 [(C7H7)ϩ,
100%].
t
28%], 238 [(M Ϫ CO2 Bu ϩ H)ϩ, 45%], 91 [(C7H7)ϩ, 100%].
(2S)-N-(4-tert-Butyloxy-4-oxobutanoyl)aspartic acid ꢁ-tert-
butyl ester 44a
(2S)-N-(4-tert-Butyloxy-4-oxobutanoyl)aspartic acid α-tert-
butyl ester β-methyl ester (43a, 2.28 g, 6.35 mmol) was dis-
solved in methanol (HPLC grade, 10 mL) and sodium
hydroxide (2 M, 8 mL), the mixture was stirred at RT for 3 h.
Methanol was removed at reduced pressure and the remaining
aqueous layer was washed with diethyl ether (2 × 50 mL). The
ether washes were discarded and water (20 mL) was added. The
aqueous layer was acidified to pH 2 with hydrochloric acid
(conc.) and then was extracted with ethyl acetate (2 × 50 mL).
The combined organic extracts were washed with water
(50 mL), dried over sodium sulfate and evaporated in vacuo to
yield (2S)-N-(4-tert-butyloxy-4-oxobutanoyl)aspartic acid α-
tert-butyl ester 44a as a pale yellow oil (950 mg, 2.76 mmol,
43%). νmax(film)/cmϪ1 2982, 1721; δH(300 MHz, CDCl3) 9.70
(1H, s, CO2H), 7.02 (1H, d, J 7.80, NH), 4.73 (1H, m, αH),
3.00–2.95 (2H, m, βH), 2.65–2.49 (4H, m, 2 × succ CH2), 1.45
(18H, s, 2 × tBu); δC(75.45 MHz, CDCl3) 177.8 (CO), 175.8
(CO), 172.1 (CO), 170.0 (CO), 83.2 (2 × OC(CH3)3), 49.2
(αCH), 36.4 (succ CH2), 31.6 (succ CH2), 29.6 (βCH2), 28.0
(2 × OC(CH3)3); m/z (CI)ϩ 346 [(MH)ϩ, 0.07%], 234 [(M Ϫ
2 × tBu ϩ 3H)ϩ, 25%], 101 [(HO2CCH2CH2CO2H)ϩ, 100%], 57
[(tBu)ϩ, 78%]; m/z (EI) 345 [(M)ϩ, 2%] (346.1876 calcd for
C16H28NO7, found 346.1866).
ꢀ-Methyl ꢁ-tert-butyl (2S)-N-(4-tert-butyloxy-4-oxobutanoyl)-
aspartate 43a
β-Methyl hydrogen (2S)-aspartate hydrochloride (41a, 2.00 g,
11.1 mmol) was dissolved in saturated sodium bicarbonate
(50 mL). Succinic anhydride (1.66 g, 16.6 mmol) was added in
portions over a period of 1.5 h. When reaction was complete,
the solution was acidified with hydrochloric acid (conc.) to
pH 4 and extracted with ethyl acetate (4 × 30 mL). The organic
extract was dried (MgSO4) and evaporated in vacuo. The
resulting colourless solid (42a, 0.95 g) was dissolved in dichloro-
methane (10 mL) and cooled to Ϫ5 ЊC before addition of
sulfuric acid (conc. 0.1 mL). Isobutene (0.5 g) was dissolved
into the solution and the mixture was stirred for 24 h at RT by
which time all the solid had disappeared. Dichloromethane
(50 mL) was added, and the organic layer was washed with
saturated aqueous sodium bicarbonate (2 × 50 mL) then water
(1 × 50 mL) and dried over sodium sulfate. The solvent was
evaporated under reduced pressure and the residue was purified
by flash chromatography (20% : 80% Et2O–hexane, Rf 0.14)
to give β-methyl α-tert-butyl (2S)-N-(4-tert-butyloxy-4-oxo-
butanoyl)aspartate 43a as a colourless oil (410 mg, 3.30 mmol,
30%). νmax(film)/cmϪ1 2980, 1740; δH(300 MHz, CDCl3) 6.70
(1H, d, J 7.90, NH), 4.72 (1H, m, αCH), 3.68 (3H, s, OCH3),
2.94–2.74 (2H, m, βCH2), 2.64–2.46 (4H, m, 2 × succ CH2),
1.45 (18H, s, 2 × tBu); δC (75.45 MHz, CDCl3) 176.2 (CO),
176.0 (CO), 172.0 (CO), 169.7 (CO), 82.6 (OC(CH3)3), 80.8
(OC(CH3)3), 60.4 (OCH3), 52.0 (αCH), 36.4 (succ CH2), 31.6
(succ CH2), 29.1 (βCH2), 28.1 (2 × OC(CH3)3); m/z (CI)ϩ 360
[(MH)ϩ, 0.06%], 57 [(tBu)ϩ, 100%]; m/z (EI)ϩ 360 [(MH)ϩ,
8%], 57 [(tBu)ϩ, 100%] (360.2013 calcd for C17H30NO7, found
360.2022).
ꢁ-tert-Butyl sodium (S)-N-benzyloxycarbonylaspartate 44b32
α-tert-Butyl methyl (2S)-N-benzyloxycarbonylaspartate (43b,
2.46 g, 7.30 mmol) was dissolved in a mixture of methanol
(5.1 mL) and aqueous sodium hydroxide (2 M, 3.7 mL), the
mixture was stirred at room temperature for 3 h. Methanol was
removed at reduced pressure and the remaining aqueous layer
was washed with diethyl ether (2 × 30 mL), the ether washes
were discarded and water (30 mL) was added to the aqueous
layer. The aqueous layer was acidified to pH 2 with hydro-
chloric acid (conc.) and was extracted with ethyl acetate (2 ×
50 mL). The combined organic phases were washed with water
(50 mL), dried over sodium sulfate and the solvent was evapor-
ated in vacuo to give the desired acid 44b as a colourless oil
(1.61 g, 5.00 mmol, 68%). νmax(film)/cmϪ1 3360, 2950, 1730;
δH(300 MHz, CDCl3) 7.35 (5H, m, Ph), 5.85 (1H, d, J 8.00,
NH), 5.17 (2H, s, OCH2Ph), 4.78 (1H, m, αCH), 3.09–2.80 (2H,
m, βCH2); δC(75.45 MHz, CDCl3) 171.5 (CO), 169.5 (CO),
156.2 (CONH), 128.5 (Ph), 128.2 (Ph), 128.1 (Ph), 128.0 (Ph),
82.9 (OC(CH3)3), 67.3 (OCH2), 50.9 (αCH), 36.6 (βCH2), 27.7
(OC(CH3)3); m/z (EI)ϩ 324 [(MH)ϩ, 10%], 91 [(C7H7)ϩ, 100%].
2032
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034

View Online
tert-Butyl (2S)-N-(4-tert-butyloxy-4-oxobutanoyl)homoserinate
46a
3.80 (3H, s, OCH3), 3.80 (9H, s, OC(CH3)3); m/z (EI)ϩ 160
[(M)ϩ, 30%], 57 [tBuϩ 100%].
A solution of (S)-N-(4-tert-butyloxy-4-oxobutanoyl)aspartic
acid α-tert-butyl ester (44a, 0.95 g, 2.75 mmol) in anhydrous
THF (10 mL) was cooled to Ϫ10 ЊC and N-methylmorpholine
(0.418 g, 4.13 mmol) was added to the mixture. After 1 minute
at Ϫ10 ЊC, ethyl chloroformate (0.45 g, 4.13 mmol) was added
dropwise and the mixture was stirred at Ϫ10 ЊC for a further
15 minutes. N-Methylmorpholine hydrochloride was removed
by filtration, and the clear organic solution was added over
a period of 10 minutes to a vigorously stirred suspension
of sodium borohydride (300 mg) in water (3.0 mL) at 5 to
10 ЊC. The mixture was stirred at room temperature for
3.5 hours.
tert-Butyl potassium oxalate 49
Methyl tert-butyl oxalate 48 (2.91 g, 18.2 mmol) was dissolved
in a mixture of CH3CN (HPLC grade, 10 mL) and H2O
(10 mL), followed by the addition of ground solid KOH (1.02 g,
18.2 mmol). The mixture was stirred at RT for 1 h until all
KOH had dissolved. The solvent was evaporated in vacuo to
yield tert-butyl potassium oxalate 49 (3.13 g, 17.0 mmol, 93%) as
a colourless solid; m/z (ES)ϩ 185 [(MH)ϩ, 14%], 57 [(tBu)ϩ,
100%] (Anal. Calcd for C6H9KO4ؒH2O: C, 35.63%; H, 5.48%.
Found: C, 35.95%; H, 5.59%).
The mixture was then cooled to 5 ЊC, acidified to pH 2 with
hydrochloric acid (conc. aq.) and extracted with ethyl acetate
(2 × 50 mL). The organic extracts were washed with water
(2 × 50 mL), dried over sodium sulfate and the solvent was
evaporated in vacuo to yield the crude product which was
purified by flash chromatography (35% : 60% Et2O–hexane, Rf
0.31) to give tert-butyl (2S)-N-(4-tert-butyloxy-4-oxobutanoyl)-
homoserinate 46a as a colourless oil (385 mg, 1.16 mmol, 42%).
δH(300 MHz, CDCl3) 6.70 (1H, d, J 7.70, NH), 4.30 (1H,
m, αH), 3.78–3.67 (2H, m, γCH2), 3.08–2.96 (2H, m, βCH2),
2.75–2.50 (4H, m, 2 × succ CH2), 1.45 (18H, s, 2 × tBu);
δC(75.45 MHz, CDCl3) 176.9 (CO), 174.2 (CO), 172.1 (CO),
80.9 (2 × OC(CH3)3), 62.2 (γCH2), 52.5 (αCH), 32.4 (succ CH2),
31.1 (succ CH2), 30.9 (βCH2), 28.0 (2 × OC(CH3)3); m/z (CI)ϩ
332 [(MH)ϩ, 24%], 57 [(tBu)ϩ, 100%].
Di-tert-butyl (2S)-2-[N-(4-tert-butyloxy-4-oxobutanoyl)amino]-
6-oxo-5-oxapimelate 51a
tert-Butyl potassium oxalate (49, 74.0 mg, 403 µmol) was dis-
solved in acetonitrile (15 mL) in the presence of 4 Å molecular
sieves under dry nitrogen. The mixture was cooled to Ϫ10 ЊC
and ethyl choroformate (6 eq., 2.42 mmol, 260 mg) was added
dropwise and with stirring at RT for 0.5 h. tert-Butyl (2S)-N-
(tert-butylsuccinyl)homoserinate (46a, 1.5 eq.) was added to
the solution and the mixture was stirred at room temperature
for 36 h. Saturated aqueous sodium bicarbonate (50 mL) was
added to the mixture and the organic layer was extracted with
dichloromethane (3 × 30 mL), dried (Na2SO4), filtered, and
solvent was evaporated in vacuo. Di-tert-butyl (2S)-2-[N-(4-
tert-butyloxy-4-oxobutanoyl)amino]-6-oxo-5-oxapimelate 51a
was obtained by flash chromatography of the residue (35 : 65
EtOAc–hexane, Rf 0.4) as an oil (50 mg, 109µmol, 18%).
[α]2D4 Ϫ3.26 (c 1.9 in CH2Cl2); νmax (film)/cmϪ1 2979 (NH), 1737
(CO); δH(300 MHz, CDCl3) 6.48 (1H, d, J 7.7, NH), 4.27 (1H,
m, αH), 2.55 (2H, m, γCH2), 2.20 (2H, m, βCH2), 1.55 (9H, s,
tBu), 1.46 (9H, s, tBu), 1.42 (9H, s, tBu); δC(75.5 MHz, CDCl3)
172.0 (CO), 171.5 (CO), 170.5 (CO), 158.3 (CO), 156.3 (CO),
85.0 (OC(CH3)3), 82.69 (OC(CH3)3), 80.72 (OC(CH3)3), 63.1
(αCH), 60.4 (γCH2), 50.3 (βCH2), 31.1 (succ CH2), 30.9 (succ
CH2), 28.1 (3 × OC(CH3)3); m/z (CI)ϩ 460 [(MH)ϩ 3%], 404
[(M Ϫ tBu ϩ 2H)ϩ, 3%], 348 [(M Ϫ 2 × tBu ϩ 3H)ϩ, 6%], 292
[(M Ϫ 3 × tBu ϩ 4H)ϩ, 40%], 57 [(tBu)ϩ, 100%] (460.2565 calcd
for C22H37NO9, found 460.2545).
tert-Butyl (2S)-N-(benzyloxycarbonyl)homoserinate 46b29
A
solution of α-tert-butyl hydrogen (S)-N-benzyloxy-
carbonylasparate (44b, 2.22 g, 6.87 mmol) in dry THF (7.21
mL) was cooled to Ϫ10 ЊC and N-methylmorpholine (0.73 g,
7.22 mmol) was added to the mixture. After 1 minute at Ϫ10 ЊC,
ethyl chloroformate (0.78 g, 7.19 mmol) was added dropwise
and the mixture was stirred at Ϫ10 ЊC for a further 15 min. N-
Methylmorpholine hydrochloride was removed by filtration,
and the clear organic solution containing the mixed anhydride
45b was added over a period of 10 minutes to a vigorously
stirred suspension of sodium borohydride (600 mg) in water
(3.6 mL) at 5 to 10 ЊC. The mixture was stirred at room
temperature for 3.5 hours.
Di-tert-butyl (2S)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-
The mixture was then cooled to 5 ЊC, acidified to pH 2 with
aqueous hydrochloric acid (conc.) and extracted with ethyl
acetate (2 × 50 mL). The organic extracts were washed with
water (2 × 50 mL), dried over sodium sulfate and the solvent
was evaporated in vacuo to yield the crude product which was
purified by flash chromatography (35% : 60% Et2O–hexane,
Rf 0.25) to give tert-butyl (2S)-N-(benzyloxycarbonyl)-
homoserinate 46b as a colourless oil (640 mg, 2.07 mmol, 30%).
νmax(film)/cmϪ1 3480, 3069, 1700; δH(300 MHz, CDCl3) 7.35
(5H, m, Ph), 5.72 (1H, d, J 7.50, NH), 5.12 (2H, s, OCH2), 4.40
(1H, m, αCH), 3.70 (2H, m, γCH2), 3.00–2.70 (2H, m, βCH2);
δC(74.5 MHz, CDCl3) 171.6 (CO), 156.8 (CO), 136.1 (Ph), 128.5
(Ph), 128.4 (Ph), 128.2 (Ph), 82.4 (OC(CH3)3), 77.5 (OC(CH3)3),
67.1 (γCH2), 62.0 (OCH2), 53.4 (αCH), 35.9 (βCH2), 27.9
(OC(CH3)3); m/z (EI)ϩ 91 [(C7H7)ϩ, 100%].
oxapimelate 51b
18-Crown-6 (173 mg, 0.12 mmol) was added to a suspension
of tert-butyl potassium oxalate (49, 131 mg, 712 µmol) in
acetonitrile (HPLC grade, 15 mL) in the presence of 4 Å
molecular sieves under dry nitrogen. The whole mixture was
stirred until the oxalate had dissolved into the organic solvent.
Ethyl chloroformate (4.00 g, 4.06 mmol) was added dropwise
to the solution over a period of 5 min and then the mixture was
stirred at RT for 30 minutes. tert-Butyl (2S)-N-(benzyloxy-
carbonyl)homoserinate (46b, 200 mg, 1.07 mmol) was added
and the solution was left to stir for 3 h. Saturated aqueous
sodium bicarbonate (50 mL) was added to the mixture and the
organic layer was extracted with dichloromethane (4 × 30 mL),
dried over sodium sulfate and evaporated in vacuo to give
the crude product. The crude product was purified by flash
chromatography (20% : 80% Et2O–hexane, Rf 0.31) to give di-
tert-butyl (2S)-2-[N-(benzyloxycarbonyl)amino]-6-oxo-5-oxa-
pimelate 51b (63.5 mg, 23%) as a colourless oil. [α]2D4 Ϫ5.78 (c
2.23 in CH2Cl2); νmax(film)/cmϪ1 2979, 1739, 1523; δH(300 MHz,
CDCl3) 7.33 (5H, m, Ph), 5.48 (1H, d, J 8.30, NH), 5.30 (2H, s,
OCH2), 4.30 (1H, m, αH), 4.19 (2H, m, γCH2), 2.32–2.08 (2H,
tert-Butyl methyl oxalate 48
Methyl oxalyl chloride (1.00 g, 8.16 mmol) was added dropwise
to a mixture of anhydrous pyridine (1.29 g, 16.3 mmol), tert-
butyl alcohol (1.21 g, 16.3 mmol) and anhydrous Et2O (60 mL).
The mixture was stirred at RT for 1 h. The mixture was then
washed with H2O (20 mL), sat. NaHCO3(aq) (20 mL) and H2O
(20 mL), dried over Na2SO4 and concentrated in vacuo to yield
tert-butyl methyl oxalate 48 as a colourless oil (690 mg, 4.32
mmol, 53%); νmax(film)/cmϪ1 1750 (CO); δH(270 MHz, CDCl3)
t
t
m, βCH2), 1.55 (9H, s, Bu), 1.53 (9H, s, Bu); δC(75.45 MHz,
CDCl3) 170.4 (CO), 158.3 (CO), 156.6 (CO), 155.7 (CO), 136.2
(Ph), 130.1 (Ph), 128.5 (Ph), 128.2 (Ph), 85.1 (OC(CH3)3),
82.8 (OC(CH3)3), 67.0 (OCH2), 63.0 (γCH2), 52.0 (βCH),
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034
2033

View Online
31.1 (βCH2), 27.9 (OC(CH3)3); m/z (CI)ϩ 438 [(MH)ϩ, 44%], 91
[(C7H7)ϩ, 100%] (438.2128 calcd for C22H32NO8, found
438.2110).
12 J. Choi and N. M. Yoon, J. Org. Chem., 1995, 60, 3266.
13 E. Fischer and U. Suzuki, Z. Physiol. Chem., 1905, 45, 405.
14 M. Hoffmann, F. Burkhart, G. Hessler and H. Kessler, Helv. Chim.
Acta, 1996, 79, 1519.
15 J. E. Baldwin, M. G. Moloney and M. North, Tetrahedron, 1989, 45,
6309.
16 P. M. Hardy, J. C. Haycock and H. N. Rydon, J. Chem. Soc., Perkin
Trans. 1, 1972, 605.
Acknowledgements
We thank the School of Chemistry, University of Bristol for
Financial Assistance.
17 J. Voss, Tetrahedron, 1971, 27, 3753.
18 J. F. Morrison and C. T. Walsh, Adv. Enzymol., 1988, 61, 201.
19 J. F. Morrison and S. R. Stone, Comments Mol. Cell. Biophys., 1985,
2, 347.
20 W. C. Still, M. Kahn and A. Mitra, J. Org. Chem., 1978, 43, 2923.
21 B. S. Furniss, A. J. Hannaford, P. W. G. Smith and A. R. Tatchell,
Vogels Textbook of Organic Chemistry, 5th edn. Longman, New
York, 1989, pp. 430–433.
22 L. Zervas and I. Photaki, J. Am. Chem. Soc., 1962, 84, 3887.
23 E. L. Gustus, J. Org. Chem., 1967, 32, 3425.
References
1 R. J. Cox, A. Sutherland and J. C. Vederas, Bioorg. Med. Chem.,
2000, 8, 843.
2 R. M. Hall and C. M. Collis, Development of Novel Antimicrobial
Agents: Emerging Strategies, ed. K. Lohner, Horizon Scientific
Press, Wymondham, UK, 2001, pp. 17–30.
24 J. A. Robl, C.-Q. Sun, J. Stevenson, D. E. Ryono, L. M. Simpkins
and M. Ligaya, J. Med. Chem., 1997, 40, 1570.
3 R. J. Cox, Nat. Prod. Rep., 1996, 13, 29.
4 H. Misono, H. Togawa, T. Yamamoto and K. Soda, Biochem.
Biophys. Res. Commun., 1976, 72, 89.
5 T. D. H. Bugg and C. T. Walsh, Nat. Prod. Rep., 1992, 9, 199.
6 R. J. Cox, H. Jenkins, J. A. Schouten, R. A. Stentiford and
K. J. Wareing, J. Chem. Soc., Perkin Trans. 1, 2000, 13, 2023.
7 R. J. Cox, J. A. Schouten, R. A. Stentiford and K. J. Wareing,
Bioorg. Med. Chem. Lett., 1998, 8, 945.
8 Transaminases, ed. P. Christen and D. E. Metzler, 1984, Wiley, New
York.
9 R. J. Cox, W. A. Sherwin, L. Lam and J. C. Vederas, J. Am. Chem.
Soc., 1996, 118, 7449.
25 R. Laliberte, Y. Knobler and M. Frankel, J. Chem. Soc., 1963, 2756.
26 Y. Knobler and M. Frankel, J. Chem. Soc., 1958, 1629.
27 D. M. Walker, J. F. McDonald and E. W. Logusch, J. Chem. Soc.,
Chem. Commun., 1987, 1710.
28 A. J. Ozinskas and G. A. Rosenthal, J. Org. Chem., 1986, 51, 5047.
29 J. S. Tou and B. D. Vineyard, J. Org. Chem., 1985, 50, 4982.
30 F. K. Brown, P. J. Brown, D. M. Bickett, C. L. Chambers,
H. G. Davies, D. N. Deaton, D. Drewry, M. Foley, A. B. McElroy,
M. Gregson, G. M. McGeehan, P. C. Meyers, D. Norton,
J. L. Salovich, F. J. Schoenen and P. Ward, J. Med. Chem., 1994,
37, 674.
10 D. Appleton, A. B. Duguid, S.-K. Lee, Y.-J. Ha, H.-J. Ha and
F. J. Leeper, J. Chem. Soc., Perkin Trans. 1, 1998, 89.
11 L. R. Pohl, R. D. Schulick, R. J. Highet and J. W. George, Mol.
Pharmacol., 1984, 25, 318.
31 J. E. Baldwin, M. G. Moloney and M. North, J. Chem. Soc., Perkin
Trans. 1, 1989, 833.
32 R. M. Valerio, P. F. Alewood and R. B. Johns, Synthesis, 1988, 10,
786.
2034
J. Chem. Soc., Perkin Trans. 1, 2001, 2022–2034