Highly diastereoselective synthesis of 4-N-methylformamidino trinem (GV129606), a potent antibacterial agent

Article abstract of DOI:10.1016/S0040-4020(00)00414-2

In this paper a highly diastereoselective synthesis of 4-N-methylformamidino trinem 3 is reported. The route offers advantages compared to that previously used, i.e. the higher overall yield, the robustness, the avoidance of toxic reagents. Most of the compounds were isolated by precipitation, therefore reducing the number of chromatographic separations. The efficient conversion of 4-N-methylamino trinem 11 into GV129606 3, was obtained by a new methodology in which a scavenger resin was used. The route presented in this paper allowed the preparation of the material required for early development studies and demonstrates the versatility of cyclohexenyl azetidinone 12 in the synthesis of 4-substituted trinems. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00414-2

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 5649±5655
Highly Diastereoselective Synthesis of 4-N-Methylformamidino
Trinem (GV129606), a Potent Antibacterial Agent
*
Stefano Biondi, Angelo Pecunioso, Filippo Busi, Stefania A. Contini, Daniele Donati,
Micaela Maffeis, Domenica A. Pizzi, Luciana Rossi, Tino Rossi and Fabio M. Sabbatini
Glaxo Wellcome SpA, Medicines Research Centre, Via Fleming, 4-37135 Verona, Italy
Received 4 October 1999; accepted 21 December 1999
AbstractÐIn this paper a highly diastereoselective synthesis of 4-N-methylformamidino trinem 3 is reported. The route offers advantages
compared to that previously used, i.e. the higher overall yield, the robustness, the avoidance of toxic reagents. Most of the compounds were
isolated by precipitation, therefore reducing the number of chromatographic separations. The ef®cient conversion of 4-N-methylamino
trinem 11 into GV129606 3, was obtained by a new methodology in which a scavenger resin was used. The route presented in this paper
allowed the preparation of the material required for early development studies and demonstrates the versatility of cyclohexenyl azetidinone
12 in the synthesis of 4-substituted trinems. q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
C have been reported in the literature,7 most of the work
was directed towards the synthesis of compounds having a
6-membered carbocyclic system.8 For this class we dis-
covered that the stereochemistry of carbon atoms at position
4 and 8 is important in determining the biological pro®le of
trinems. Generally the relative con®guration 4a,8b repre-
sents the best compromise among the various properties
required for a molecule to be effective as an antibacterial
agent, e.g. activity, enzymatic and chemical stability, and
pharmacokinetic pro®le.
The rapid emergence of bacteria with increased resistance to
available drug therapies is becoming a serious worldwide
health problem1 that affects important classes of antibiotics
including b-lactams, quinolones, macrolides, glycopeptides,
aminoglycosides and tetracyclines.
Bacteria are exceedingly capable of evading antibacterials
through processes of mutation and/or acquisition of resis-
tance genes followed by selection and evolution in versatile
ways to resist almost every new agent.2 The major mecha-
nisms of resistance to b-lactams involve the reduction in
penetration through the outer membrane (for Gram-negative
bacteria), the production of Penicillin Binding Proteins with
a reduced af®nity for inactivating agents, and the hydrolysis
of the drug molecule by b-lactamases.3
Trinems are potent antibacterial agents, with a broad spec-
trum of activity against Gram positive and Gram negative
strains, either aerobes or anaerobes, stable to most clinically
relevant b-lactamases and to the human hydrolytic enzyme
dehydropeptidase (DHP-I). In particular, Sanfetrinem9 2a
and its biolabile ester cilexetil10 2b (see Fig. 2), are now
undergoing phase II clinical studies, particularly for the
treatment of infections caused by penicillin resistant
strains.11 As part of the research programme conducted in
our laboratories we were interested in the introduction of a
nitrogen containing functional group12 at C-4. In particular
we synthesized the 4-N-methylformamidino trinem 3 (see
Fig. 2), which includes in its spectrum of activity the
Trinems4 (Fig. 1), a class of b-lactam antibiotics discovered
in our laboratories, are promising substances that can be
effectively used to treat infections. The general structure
of trinems 1a is characterised by a tricyclic ring system in
which ring C may be 55, 6 or 75e-membered and may contain
heteroatoms. When ring C is 6-membered as in structure 1b,
the position 4 is generally substituted and an hydroxyethyl
side chain of de®ned absolute stereochemistry is present at
position 10. Recently, some compounds having an alkyl-
idene substituent on C-10 showing b-lactamase inhibitory
activity6 have also been published. Although several
compounds with heteroatoms at various positions of ring
Keywords: b-lactams; trinems; gram-negative bacteria; pseudomonads;
diastereoselective synthesis.
* Corresponding author.
Figure 1.
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00414-2

5650
S. Biondi et al. / Tetrahedron 56 (2000) 5649±5655
4-acetoxyazetidin-2-one 4a14 (Scheme 1), ®ve chromato-
graphic separations and a ®ltration through silica gel pad
of the intermediates with an approximate overall yield of
0.5%.
The main drawbacks in this synthesis are:
(i) the moderate diastereoselectivity in favour of the
desired isomer 5a in the SnCl4 catalysed addition of 1-
(trimethylsilyloxy)cyclohexene to 4b (5a:5bˆ4:1) result-
ing in the isolation of compound 5a in 56% yield;15
(ii) the use of highly toxic diethylchlorophosphite in the
preparation of compound 6;
Figure 2.
dif®cult strain P. aeruginosa, and has shown good ef®cacy
and pharmacokinetic pro®le in animal models. The decision
to better characterise the compound through preliminary phar-
macological development studies prompted us to evaluate a
more ef®cient synthesis, suitable for gram scale preparation.
(iii) the instability of some intermediates like the epoxy-
phosphate 7 and the a-aminoketone derivative 8a that
caused variability in the recovered yield of carbamoyl
derivative 8b;
(iv) the lack of solid intermediates throughout the route
that required extensive use of chromatographic sepa-
rations;
(v) the complexity of the mixture obtained from the
formimidoylation reaction of compound 11 and the
consequent need for a reverse phase chromatography
puri®cation and lyophilisation to recover GV129606 3.
Chemistry
The original synthetic route13 used to obtain GV129606 (3)
required 14 steps starting from the commercially available
Scheme 1. Reagents and conditions: (i) TMSCl, TEA, CH2Cl2, 08C; (ii) SnCl4, 1-(trimethylsilyloxy)cyclohexane, 2258C, CH3CN; (iii) SiO2, MeOH, TEA,
then crystallisation from petroleum; (iv) TBDMSCl, TEA, DMF, RT; (v) LHMDSA, ClP(O)(Et)2, THF, 2708C; (vi) KF, MeOH, RT; (vii) MCPBA, CH2Cl2,
08C; (viii) MeNH2, K2CO3, ethyl acetate; (ix) ClCO2Allyl, TEA, CH2Cl2, 08C; (x) ClCOCO2Allyl, TEA, CH2Cl2, RT (xi) P(OEt)3, xylene, 1208C; (xii) TBAF,
AcOH, THF, RT; (xiii) Pd(Ph3P)4, 5,5-dimethyl-1,3-cyclohexanedione; (xiv) Benzyloxyformimidate hydrochloride, pHˆ8 phosphate buffer, 08C, then reverse
phase chromatography on octadecylsilyl silica gel and lyophilisation.

S. Biondi et al. / Tetrahedron 56 (2000) 5649±5655
5651
Scheme 2. Reagents and conditions: (i) B-2-cyclohexenyl-(2-methylcyclohexyl)2, Et2Zn, 08C; (ii) Magnesium monoperoxyphthalate, CH2Cl2, H2O, 08C; (iii)
MeNH2 (g), LiClO4, CH3CN, 608C; (iv) ClCO2Bn, TEA, CH2Cl2; (v) ClCOCOCl, DMSO, TEA, CH2Cl2; (vi) ClCO2¯uorenyl, TEA, toluene, RT; (vii) P(OEt)3,
1118C; (viii) Bu4NBr, KF, AcOH, THF; (ix) H2 1 atm, 5% Pd/C, H2O/isopropanol; (x) benzyloxyformimidate hydrochloride, amberlyst IRA68, H2O/CH3CN,
then crystallization from H2O/acetone.
To by-pass these issues, we planned for a new synthetic
route (Scheme 2) in which one of the key intermediates,
the cyclohexenyl azetidinone 12, had been previously
studied for the synthesis16 of GV104326 2a. The greater
diastereoselectivity in the condensation reaction of 4-acet-
oxy azetidin-2-one 4a with B-2-cyclohexenyl-(2-methyl-
cyclohexyl)2 with respect to the trimethylsilylenol ether of
cyclohexanone, the avoidance of the highly toxic diethyl-
chlorophosphite and the overall reduction in the number of
steps encouraged us to progress further with such a synthetic
route.
This 95:5 diastereomeric mixture was then treated with
methylamine in acetonitrile at 608C using lithium per-
chlorate as catalyst to give the corresponding aminoalcohol
14 that was isolated after precipitation from the reaction
mixture as a single diastereoisomer in 78% yield.
At this point, the choice of suitable protecting groups that
would allow isolation of crystalline intermediates to be
removed simultaneously was of critical importance and
required a certain number of optimisation studies. A signi®-
cant improvement of the overall synthetic ef®cacy was
achieved by the combination of a benzyloxycarbonyl
group on the N-methyl amino moiety at C4 with a ¯uorenyl
ester18 at C2. Selective acylation of the secondary amino
group with benzyl chloroformate gave the intermediate 15
in 86% yield, which was oxidised under Swern reaction
conditions to give compound 16 as crystalline material
(79%). The ¯uorenyloxalyl chloride, prepared from oxalyl
chloride and ¯uorenyl alcohol, was reacted with 16 to form
the oxalimido19 intermediate 17 that was immediately
The crystalline cyclohexenyl azetidinone 1217 was dia-
stereoselectively oxidized under various reaction conditions
to give the (10R,20S)-epoxide 13a together with its (10S,20R)
diastereoisomer 13b. The use of m-chloroperoxybenzoic
acid resulted only in a moderate diastereoselectivity 13a/
13bˆ85:15 but this could be improved to 95:5 with the
use of the less reactive magnesium monoperoxyphthalate
without signi®cantly affecting the yield (.95%).

5652
S. Biondi et al. / Tetrahedron 56 (2000) 5649±5655
converted into the fully protected trinem derivative 18 in
about 88% yield. Desilylation of 18 was initially performed
with a large excess (14 equiv.) of tetrabutyl ammonium
¯uoride in THF either at room temperature or at 408C. A
convenient modi®cation of the ¯uoride source was made by
using a cheaper combination of tetrabutylammonium
bromide and potassium ¯uoride. The hydroxyester inter-
mediate 19, obtained in 61% yield, was hydrogenated at
atmospheric pressure using 5% palladium on charcoal.
Optimisation of reaction conditions (see Experimental)
allowed us to minimise the amount of residual starting
material, the formation of side products side products
deriving from the decomposition of reactive intermediates
and reduction of the C2±C3 tetrasubstituted double bond.
This resulted in an easier recovery and puri®cation of the 4-
N-methylamino trinem 11 by crystallisation from a mixture
of water/isopropanol in 60% yield.
example of the synthetic utility of compound 12 in the
synthesis of 4-substituted trinems.
The route described in Scheme 2 avoids the use of the highly
toxic diethylchlorophosphite and the formation of unstable
intermediates, such as epoxyphosphate 7 and a-amino
ketone 8a.
The appropriate choice of protecting groups for both the
amino and carboxylic functions, allowed the elimination
of several chromatographic steps with respect to the original
route and the ef®cient simultaneous removal of both protect-
ing groups by heterogeneous catalysis. In addition, the use
of the tertiary amine resin IRA68w as proton scavenger in
the last step of the synthesis greatly simpli®ed the work-up
improving the amount and quality of the recovered ®nal
compound.
Still, the conversion of the 4-N-methylamino trinem 11 into
the corresponding 4-N-methylformamidino trinem 3 repre-
sented a critical step due to the limited stability of both the
starting material and the product to the reaction conditions
used in the previous synthesis (Scheme 1). A similar formi-
midoylation reaction was also reported by Merck scientists
for the synthesis of Imipenem.20 Unfortunately, the pro-
cedures carefully optimised for the Imipenem synthesis
were not ef®cient in our case and did not allow the isolation
of the ®nal GV129606 3 with the required purity. The main
problem associated with this procedure was the contami-
nation of the product 3 with both inorganic salts and by-
products deriving from the decomposition of the trinem
derivatives.
The number of steps was reduced from 14 to 10 and the
number of puri®cations by silica gel chromatography was
limited to 2 instead of the 5 previously required. This
allowed us to recover GV129606 3 in about 7% overall
yield with respect to the 0.5% of the original route. The
synthesis described in Scheme 2 was practical enough to
allow the isolation of the 4-N-methylformamidino trinem
3 in gram quantities and with the required purity to support
early development studies.
Further studies aimed at ®nding a more ef®cient route to
GV129606 3 are required for larger scale production of
this compound.
Experimental
It was therefore attempted to perform the reaction in organic
solvents using highly lipophilic tertiary amines, which
could trap the hydrochloric acid from the benzyloxy
formimidate forming an ammonium salt that could be
removed from the aqueous solution. Satisfactory results
were obtained with N,N-dimethyldodecylamine, that gave
the desired compound 3, in good yield. The main drawback
of this protocol was the high number of extractions (.20)
required for a complete removal of the ammonium salt from
the aqueous solution. The use of amines having more
lipophilic chains was unsuccessful as we observed the
formation of emulsions that were hardly treatable. There-
fore, we switched our attention towards basic resins as
proton scavenger, suitable for use in aqueous solutions
and easily removable by ®ltration. The study has been
extended to several commercially available amine based
resins covering a range of basicity. Best results were
obtained by using tertiary amine resin IRA68w. This
allowed us to obtain a yield .90% in solution by carrying
out the reaction in a mixture of water and acetonitrile at 08C.
Removal of the organic solvent under reduced pressure and
addition of acetone at 08C allowed crystallisation of
GV129606 3 with a purity .95 and 42% yield.
Materials and apparatus
The IR spectra were collected on a Bruker IFS48 instrument
using a dispersion in nujol on KBr pellets, or in solution of
CDCl3. The NMR spectra were recorded using a UMX 300
Varian (300 MHz) instrument or an AC 200 Bruker
(200 MHz) instrument. The MS spectra were recorded
with a VG quattro Fison instrument using a FAB1ioni-
zation technique (70 eV). Melting points were performed
using a Buchi 530 and are uncorrected. The HPLC analysis
were executed on Jasco PU 980 instrument equipped with a
Jasco UV 975 Detector using a Hypersil BDS C18 reverse
phase column (20£0.46 cm) at room temperature.
Compounds were puri®ed by ¯ash chromatography using
Merck silica gel 60 (230±400 mesh). The reactions were
monitored by thin layer chromatography (TLC) on Merck
silica gel 60 F254 TLC plates. Reagents and solvents (sure
seal package) used in this work were purchased from
Aldrich, Fluka and Janssen and used directly without any
puri®cation.
Methods
Conclusions
Compound 13. Cyclohexenyl azetidinone 1217 (50 g,
162 mmol) was dissolved in dichloromethane (550 ml) at
08C, water (500 ml) and magnesium monoperoxyphthalate
hexahydrate (112 g, 226 mmol) were added and the mixture
vigorously stirred for 24 h. A solution of 2 M sodium
A new, highly diastereoselective synthetic route to
GV129606 3 was established using the cyclohexenyl azeti-
dinone 12 as key intermediate. This represents a further

S. Biondi et al. / Tetrahedron 56 (2000) 5649±5655
5653
carbonate (600 ml) was added and the two phase system was
stirred for 20 min. The two phases were separated and the
aqueous solution was extracted with dichloromethane
(650 ml), the organics were combined and washed with a
10% solution of sodium metabisulphite (500 ml). The
organic solution was dried over sodium sulphate and the
solvent removed under reduced pressure to give a white
str.), 1755 (CvO str.), 1691 (CvO str.). MS (m/z): 491
(MH1). White solid.
Compound 16. Oxalyl chloride (19.4 ml, 222 mmol) was
dissolved in dry dichloromethane (200 ml) under nitrogen
atmosphere and cooled at 2708C. A solution of dimethyl
sulphoxide (31.6 ml, 444 mmol) in dichloromethane
(50 ml) was added dropwise maintaining the temperature
below 2608C and the mixture stirred for further 20 min.
A solution of intermediate 15 (54.5 g, 111 mmol) in
dichloromethane (500 ml) was slowly added at 2708C and
allowed to react for 1 h. Triethylamine (124 ml, 888 mmol)
was added and the temperature was raised to 108C, a satu-
rated solution of ammonium chloride (1 l) was added and
extracted with dichloromethane. The organic phase was
washed with a 2% solution of sodium bicarbonate
(600 ml), water (2£600 ml), and dried over sodium
sulphate. The solvent was removed under reduced pressure
to give a solid material (66 g) which was redissolved in the
minimum amount of ethyl acetate and the ketoazetidinone
16 was crystallized by addition of cyclohexane (43 g, 79%).
1H NMR (300 MHz, dˆppm; Si(CH3)4 in CDCl3): 7.34±
7.30 (m, 5H), 5.99 (s, 1H), 5.11 (s, 2H), 4.56 (m, 1H),
4.19 (dd, 1H), 4.04 (m, 1H), 2.99 (dd, 1H), 2.87 (s, 3H),
2.66 (m, 1H), 2.20±1.80 (m, 6H), 1.10 (d, 3H), 0.87 (s, 9H),
0.065 (s, 6H). IR (CDCl3): 3490 (NH str.), 1763 (CvO str.),
1718 (CvO str.), 1691 (CvO str). MS (m/z): 489 (MH1).
White solid, melting point 147±1488C.
1
solid (51.5 g, 97.8%). 13a H NMR (300 MHz, dˆppm;
Si(CH3)4 in CDCl3): 5.97 (br.s, 1H), 4.22 (m, 1H), 3.77
(dd, 1H), 3.16 (t, 1H), 3.12 (m, 1H), 3.01 (m, 1H), 2.00
(m, 1H), 1.96 (m, 1H), 1.86 (m, 1H), 1.55 (m, 1H), 1.40
(m, 1H), 1.26 (d, 3H), 1.22 (m, 2H), 0.87 (s, 9H), 0.07 (s,
6H). IR (nmax/cm21): 3413 (NH, str.), 1757 (CvO str.). MS
(m/z): 326 (MH1). White solid, melting point 134±1368C.
1
[a]Dˆ35.18 cˆ0.61 in CH2Cl2. 13b H NMR (300 MHz,
dˆppm; Si(CH3)4 in CDCl3): 6.00 (br.s, 1H), 4.13 (m,
1H), 3.55 (dd, 1H), 3.16 (m, 1H), 3.08 (m, 1H), 2.93 (dd,
1H), 2.10 (m, 1H), 1.99 (m, 1H), 1.69 (m, 1H), 1.62 (m, 1H),
1.43 (m, 2H), 1.27 (d, 3H), 0.90 (m, 1H), 0.88 (s, 9H), 0.09
(s, 6H). IR (nmax/cm21): 3414 (NH, str.), 1757 (CvO str.).
MS (m/z): 326 (MH1). White solid, melting point 96±988C.
[a]Dˆ2208 cˆ0.59 in CH2Cl2.
Compound 14. Compound 13a (51 g, 156 mmol) was
placed in a three necks round bottom ¯ask equipped with
an acetone dry ice trap, dissolved in dry acetonitrile
(500 ml), then anhydrous lithium perchlorate (100 g,
940 mmol) was added and the mixture was heated at 608C
to obtain a clear solution. Methylamine was bubbled for
30 min and the solution was stirred for 3.5 h at the same
temperature. The solution was cooled at room temperature
and the volume reduced to 150 ml, poured into water (1 l)
and extracted with diethyl ether (3£500 ml). The organic
solution was dried over sodium sulphate and the solvent
removed under reduced pressure to afford 65 g of a crude
that was treated with cyclohexane to give the methylamino
Fluorenyloxalyl chloride. 9-Hydroxy¯uorene (8.4 g,
46.1 mmol) was dissolved in THF (60 ml) and added drop-
wise over 30 min to a solution of oxalyl chloride (7.6 ml,
92.2 mmol) in THF (40 ml) at 08C under nitrogen atmo-
sphere. The reaction mixture was stirred for 1 h, the solvent
and excess oxalyl chloride were removed under reduced
pressure to give a pale green solid that was readily used in
the following reaction. 1H NMR (300 MHz, dˆppm;
Si(CH3)4 in CDCl3): 7.68 (d, 2H), 7.57 (d, 2H), 7.46 (dt,
2H), 7.32 (dt, 2H), 6.84 (s, 1H). IR (CDCl3): 1778 (CvO
str.), 1749 (CvO str.). MS (m/z): 274±272 (M1) EI.
1
derivative 14 (43.5 g, 78%). H NMR (300 MHz, dˆppm;
Si(CH3)4 in CDCl3): 6.26 (s, 1H), 4.20 (m, 1H), 3.80 (m,
1H), 3.72 (dd, 1H), 3.13 (m, 1H), 2.67 (m, 1H), 2.49 (s, 3H),
2.02 (m, 2H), 1.70±1.12 (m, 5H), 1.31 (d, 3H), 0.91 (s, 9H),
0.12 (s, 6H). IR (CDCl3): 3416 (NH, OH str.), 1753 (CvO
str.). MS (m/z): 357 (MH1). White solid.
Compound 17. Intermediate 16 (10 g, 20.5 mmol) was
dissolved in dichloromethane (20 ml), anhydrous potassium
carbonate (5.6 g, 41 mmol) and freshly distilled pyridine
(6.6 ml, 82 mmol) were added and the resulting suspension
stirred at 08C. A freshly prepared solution of ¯uorenyloxalyl
chloride (see above) in dichloromethane (60 ml) was added
through a dropping funnel equipped with a sintered glass to
remove the ¯uorenyl oxalic acid. After the addition was
complete, the temperature was allowed to rise to 258C and
the reaction monitored by thin layer chromatography. After
disappearance of compound 8c, the solvent was reduced to
1/3 of initial volume under reduced pressure. The slurry so
obtained was triturated with diethyl ether and ®ltered over a
silica gel pad, washing several times with diethyl ether. The
organic solution was extracted with a saturated solution of
sodium bicarbonate, with a solution of 0.4 M HCl and brine.
The organic solution was dried over anhydrous sodium
sulphate and the solvent removed under reduced pressure
to give 14 g of the oxalimido intermediate 17 (94.6% yield).
1H NMR (300 MHz, dˆppm; Si(CH3)4 in CDCl3): 7.66 (m,
4H), 7.5±7.2 (m, 9H), 6.97 (s, 1H), 5.09 (s, 2H), 4.65 (dd,
1H), 4.50 (t, 1H), 4.27 (m, 1H), 3.21 (m, 1H), 2.86 (s, 3H),
Compound 15. Compound 14 (71.5 g, 200 mmol) was
dissolved in dichloromethane (400 ml) under a nitrogen
atmosphere at 08C, diisopropyl ethylamine (63 ml,
360 mmol) and benzylchloroformate (43 ml, 300 mmol)
were sequentially added and the reaction stirred for 2 h.
The reaction mixture was poured into a saturated solution
of ammonium chloride (2 l) and the organic phase sepa-
rated. The aqueous solution was extracted with dichloro-
methane and the organics were joined and washed with a
2% solution of sodium bicarbonate (3£600 ml). The solu-
tion was dried over sodium sulphate and the solvent
removed under reduced pressure to afford a crude product
(120 g) that treated with cyclohexane gave compound 15
(84 g, 86%). 1H NMR (300 MHz, dˆppm; Si(CH3)4 in
CDCl3): 7.40±7.20 (m, 5H), 6.02 (s, 1H), 5.20±5.00 (m,
2H), 4.21 (m, 1H), 4.10±4.00 (m, 1H), 3.84 (dd, 1H),
3.88±3.76 (m, 1H), 3.16 (m, 1H), 2.88 (s, 3H), 3.00±2.70
(m, 1H), 2.26 (m, 1H), 1.92±1.82 (m, 1H), 1.76±1.68 (m,
1H), 1.58 (m, 1H), 1.55±1.35 (m, 3H), 1.31 (d, 3H), 0.88 (s,
9H), 0.08 (s, 6H). IR (CDCl3): 3406(OH str.), 3150 (NH

5654
S. Biondi et al. / Tetrahedron 56 (2000) 5649±5655
2.84 (m, 1H), 2.20±1.70 (m, 6H), 1.09 (bs, 3H), 0.75 (s, 9H),
0.02 to (20.08) (ss, 6H). IR (CDCl3): 1807 (CvO str.),
1732 (CvO str.), 1699 (CvO str.). MS (m/z): 725
(MH1). White solid, melting point 50±608C.
Compound 3. Compound 11 (2 g, 7.14 mmol) was
dissolved in water (50 ml), acetonitrile (50 ml) was added
and the resulting solution was cooled at 08C. A suspension
of freshly washed IRA68w resin (66.7 ml, 105 mmol) in
water/acetonitrile 1:1, was added, then benzyloxyformi-
midate hydrochloride (4.3 g, 25 mmol) was slowly added
over 1 h. The reaction was monitored by HPLC and the
resin removed by ®ltration through ®lter paper, washed
with water (10 ml) and the collected fractions were washed
with diethyl ether. The aqueous layer was concentrated
under reduced pressure, cooled at 08C and acetone was
slowly added with stirring until a precipitate of GV129606
3 was obtained. The solid was then ®ltered through paper
and washed with ice cold acetone to give 920 mg of ®nal
compound (42%). 1H NMR (300 MHz, dˆppm; Si(CH3)4 in
D2O): 7.72 (s, 1H), 5.13 (t, 1H), 4.10 (m, 1H), 4.04 (dd, 1H),
3.30 (dd, 1H), 2.88 (m, 1H), 2.85 (s, 3H), 2.17 (m, 1H), 1.80
(m, 4H), 1.35 (m, 1H), 1.07 (d, 3H). IR (CDCl3): 3420±
3240 (NH, OH str.), 1745 (CvO str.), 1715 (CvO str.),
1670 (CvN str.). MS (m/z): 308 (MH1). White solid.
Compound 18. The oxalimido derivative 17 (8.9 g,
12.3 mmol) was dissolved in nonane (178 ml) at 508C and
triethyl phosphite (8.4 ml, 49.2 mmol) was added. The
resulting solution was heated at re¯ux for 5 h, cooled at
room temperature and treated with a 5% solution of hydro-
gen peroxide (104 ml). The mixture was stirred for 2 h and
extracted with ethyl acetate. The organic phase was
extracted twice with water, dried over sodium sulphate
and concentrated under reduced pressure. The crude product
was puri®ed with a silica gel pad using a mixture of cyclo-
hexane and ethyl acetate in a 3/1 ratio as eluant. After
removal of the solvent, compound 18 was obtained (7.3 g,
1
88.5% yield). H NMR (300 MHz, dˆppm; Si(CH3)4 in
CDCl3): 7.7±7.6 (m, 4H), 7.4±7.2 (m, 9H), 6.86 (s, 1H),
5.32 (t, 1H), 4.98 (dd, 2H), 4.16 (m, 1H), 4.07 (dd, 1H), 3.21
(dd, 1H), 3.17 (m, 1H), 2.94 (s, 3H), 2.11 (m, 1H), 1.85 (m,
1H), 1.7±1.3 (m, 4H), 1.20 (d, 3H), 0.82 (s, 9H), 0.04±0.01
(ss, 6H). IR (CDCl3): 1782 (CvO str.), 1711 (CvO str.).
MS (m/z): 725 (MH1). White solid, melting point 50±558C.
References
1. AA.VV. Advances in Experimental Medicine and Biology.
Antimicrobial Resistance: a Crisis in Health Care; Jungkind,
D. L., Mortensen, J. E., Fraimow, H. S., Calandra, G. B., Eds.;
Plenum: New York, 1995; Vol. 390.
Compound 19. Compound 18 (11.5 g, 16.6 mmol) was
dissolved in THF (57.5 ml) and acetic acid (6.64 ml,
116.2 mmol), tetrabutylammonium bromide (26.2 g,
81.34 mmol), cesium ¯uoride (12.4 g, 81.34 mmol) were
sequentially added. The reaction mixture was warmed to
408C, water (0.732 ml, 40.67 mmol) was added and stirring
was continued for 4 h. After cooling at room temperature,
the reaction was diluted with ethyl acetate and washed with
an ice cold solution of saturated sodium bicarbonate,
saturated ammonium chloride and brine. The organic solu-
tion was dried over sodium sulphate and the solvent
removed under reduced pressure. The crude product was
puri®ed by ¯ash chromatography, eluting with a mixture
of ethyl acetate and cyclohexane in a 3:1 ratio to give
2. Mitscher, L. A.; Pillai, S. P.; Gentry, E. J.; Shankel, D. M. Med.
Res. Rev. 1999, 19 (6), 477±496.
3. Bush, K.; Mobashery, S. In Resolving the antibiotic paradox;
Rosen, B. P., Mobashery, S., Eds.; Kluwer Academic/Plenum:
New York, 1998.
4. (a) Biondi, S. Trinems: synthesis and antibacterial activity of a
new generation of antibacterial b-lactams, Spec. Publ.ÐR. Soc.
Chem., 1997; Vol. 198 (Anti-infectives), pp 86±100. (b) Ghiron,
C.; Rossi, T. The Chemistry of Trinems, Targets Heterocycl. Syst.,
1997; Vol. 1, pp 161±186. (c) Gaviraghi, G. Tribactams: a new
prospect in b-lactam chemotherapy, Eur. J. Med. Chem. (Suppl.
Proceedings of the 13th International Symposium on Medicinal
Chemistry 1994), 1995, 467s±478s. (d) Upender, V.; Kelson,
A. B.; Ryan, K. J.; Webb, R. R.; Tracy, M. Synthesis and biological
activity of novel trinems; Book of Abstracts; 214th ACS National
Meeting, Las Vegas, NV, 7±11 September 1997; MEDI-151.
5. (a) Biondi, S.; Gaviraghi, G.; Rossi, T. Bioorg. Med. Chem.
Lett. 1996, 6 (5), 525±528. (b) Di Fabio, R.; Feriani, A.; Gaviraghi,
G.; Rossi, T. Bioorg. Med. Chem. Lett. 1995, 5 (12), 1235±1240.
(c) Copar, A.; Solmajer, T. Zb. Ref. Posvetovanja Slov. Kem. Dnevi
1997, 37±42; Chem. Abstr. 130, 124904. (d) Kawamoto, I.; Shitaji,
Y.; Tracy, M.; Kelson, A. B.; Ryan, K. J.; Upender, V.; Webb, R. R.
JP 11060576 A2, 1999; Chem. Abstr. 130, 252193. (e) Sendai, M.;
Miwa, T. EP 422596 A2; Chem. Abstr. 1991, 115, 279692.
6. (a) Copar, A.; Solmajer, T.; Anzic. B.; Kuzman, T.; Mesar, T.;
Kocjan, D. WO 9827094 A1; Chem. Abstr. 1998, 129, 95355.
7. (a) Biondi, S.; Piga, E.; Rossi, T.; Vigelli, G. Bioorg. Med.
Chem. Lett. 1997, 7 (15), 2061±2066. (b) Andreotti, D.; Rossi,
T.; Marchioro, C. Bioorg. Med. Chem. Lett. 1996, 6 (21), 2589±
2594. (c) Mori, M.; Oida, S. Jpn. Kokai Tokkyo Koho 1998, JP
10310582 A2; Chem. Abstr. 1998, 130, 52269. (d) Miwa, T.;
Nagai, K.; Okonogi, K. Jpn. Kokai Tokkyo Koho 1994 JP
06166688 A2 Application: JP 93-21232; Chem. Abstr. 1995,
122, 9771.
1
compound 19 as a white solid (5.86 g, 61% yield). H
NMR (300 MHz, dˆppm; Si(CH3)4 in CDCl3): 7.63 (d,
4H), 7.4±7.2 (m, 9H), 6.89 (s, 1H), 5.29 (t, 1H), 5.00 (m,
2H), 4.17 (m, 1H), 4.05 (dd, 1H), 3.22 (dd, 1H), 3.17 (m,
1H), 2.92 (s, 3H), 2.12 (m, 1H), 1.90 (m, 1H), 1.75±1.40 (m,
4H), 1.28 (d, 3H). IR (CDCl3): 3420 (OH str.), 1776 (CvO
str.), 1695 (CvO str.). MS (m/z): 579 (MH1). White solid,
melting point 77±808C.
Compound 11. Compound 19 (3.15 g, 5.4 mmol), dissolved
in isopropanol (90 ml) and water (90 ml), was slowly added
under vigorous stirring. The solution was ¯ushed with
nitrogen and Pd/C 20%, w/w (0.63 g) was added. The result-
ing suspension was hydrogenated at 1 atm for 2.5 h, ®ltered
through a Celite pad and extracted twice with diethyl ether.
The aqueous solution was concentrated under reduced
pressure to a small volume until a solid material started to
precipitate. Isopropanol was added dropwise with stirring
and a white precipitate was ®ltered (0.88 g, 60% yield). 1H
NMR (300 MHz, dˆppm; Si(CH3)4 in D2O): 4.81 (m, 1H),
4.13 (m, 1H), 4.10 (dd, 1H), 3.38 (dd, 1H), 3.04 (m, 1H),
2.48 (s, 3H), 1.97 (m, 1H), 1.84±1.20 (m, 5H), 1.12 (d, 3H),
1. IR (CDCl3): 3462±3412 (NH, OH str.), 1755 (CvO str.).
MS (m/z): 281 (MH1). White solid.
8. (a) Hanessian, S.; Grif®n, A. M.; Rozema, M. J. Bioorg. Med.

S. Biondi et al. / Tetrahedron 56 (2000) 5649±5655
5655
Chem. Lett. 1997, 7 (14), 1857±1862. (b) Hanessian, S.; Rozema,
M. J.; Reddy, G. B.; Braganza, J. F. Bioorg. Med. Chem. Lett. 1996,
5 (21), 2535±2540.
14. (3R,4R,10R)-(1)-4-Acetoxy-3-[10-(tert-butyldimethylsilyloxy)-
ethyl]azetidin-2-one 4 is commercially available from the Aldrich
Chemical Company Inc., Milwaukee, WI.
9. (a) Tamburini, B.; Perboni, A.; Rossi, T.; Donati, D.; Andreotti,
D.; Gaviraghi, G.; Carlesso, R.; Bismara, C. Eur. Pat. Appl.
EP0416953 A2, 1991; Chem. Abstr. 1992, 116, 235337t. (b)
Hanessian, S.; Rozema, M. J. J. Am. Chem. Soc. 1996, 118 (41),
9884±9891.
15. Marchioro, C.; Pentassuglia, G.; Perboni, A.; Donati, D.
J. Chem. Soc., Perkin Trans. 1 1997, (4), 463±468.
16. (a) Biondi, S.; Rossi, T.; Contini, S. A. Eur. Pat. Appl. 1994,
EP 617017 A1; Chem. Abstr. 1994, 121, 280531. (b) Biondi, S.;
Andreotti, D.; Rossi, T.; Carlesso, R.; Tarzia, G.; Perboni, A. Eur.
Pat. Appl., 1992, EP 502464 A1; Chem. Abstr. 1992, 117, 251135.
17. Rossi, T.; Biondi, S.; Contini, S.; Thomas, R. J.; Marchioro, C.
J. Am. Chem. Soc. 1995, 117 (37), 9604±9605.
10. Perboni, A.; Rossi, T.; Gaviraghi, G.; Ursini, A.; Tarzia, G.
WO 9203437, 1992; Chem. Abstr. 1992, 117, 7735m.
11. (a) Tamura, S.; Miyazaki, S.; Tateda, K.; Ohno, A. I. Y.;
Matsumoto, T.; Furuya, N.; Yamaguchi, K. Antimicrob. Agents
Chemother. 1998, 42 (7), 1858±1861. (b) Ngo, J.; Castaner, J.
Drugs Future 1996, 21 (12), 1238±1245. (c) Di Modugno, E.;
Erbetti, I.; Ferrari, L.; Galassi, G. L.; Hammond, S. M.; Xerri, L.
Antimicrob. Agents Chemother. 1994, 38 (10), 2362±2368.
12. Andreotti, D.; Biondi, S.; Donati, D. Chem. Heterocycl.
Compd. (NY) 1999, 34 (11), 1324±1330. (b) Gehanne, S.; Piga,
E.; Andreotti, D.; Biondi, S.; Pizzi, D. Bioorg. Med. Chem. Lett.
1996, 6 (22), 2791±2794. (c) Tranquillini, M. E.; Araldi, G. L.;
Donati, D.; Pentassuglia, G.; Pezzoli, A.; Ursini, A. Bioorg. Med.
Chem. Lett. 1996, 6 (14), 1683±1688.
18. (a) Bywood, R.; Gallagher, G.; Sharma, G. K.; Walker, D.
J. Chem. Soc., Perkin Trans. 1 1975, 20, 2019±2022. (b) Long,
A. G.; Walker, D.; Eastlick, D. T.; Stables, H. G. Ger. Offen 1974,
DE 2335083 1974; Chem. Abstr. 1974, 80, 108858. (c) Bywood,
R.; Gallagher, G.; Sharma, G. K.; Walker, D. Ger. Offen. 1972, DE
2201018; Chem. Abstr. 1972, 77, 126623.
19. In attempts to isolate compound 17 we obtained the product in
about 94%, but generally we used the crude intermediate for the
cyclisation step.
20. (a) Oberholtzer, E. R. Anal. Pro®les Drug Subst. 1988, 17, 73±
114. (b) Hazen, G. G.; Volante, R. P.; Wilson, K. E. US 4374772,
1983; Chem. Abstr. 98, 179101.
13. Perboni, A.; Donati, D.; Tarzia, G. Eur. Pat. Appl. EP 502468
A1, 1992; Chem. Abstr. 1993, 118, 80719.