Expeditious synthesis of 4-(tert-butylcarbonyl)-7α-methoxy-3-inethyl-Δ3-cephem 1,1-dioxide. Convenient access to 7-substituted analogues

Article abstract of DOI:10.1055/s-1998-1621

A practical and efficient route leading to the synthesis of 4-(tert-butylcarbonyl)-7α-methoxy-3-methyl-Δ³-cephem 1,1-dioxide (2), a key-intermediate in the preparation of potent inhibitors of mammalian serine proteinases, is reported. The new synthetic pathway has allowed easy access to an array of 7-substituted cephem derivatives.

Full text of DOI:10.1055/s-1998-1621

322
LETTERS
SYNLETT
3
Expeditious Synthesis of 4-(tert-Butylcarbonyl)-7α-methoxy-3-methyl-∆ -cephem 1,1-Dioxide.
Convenient Access to 7-Substituted Analogues
Marco Alpegiani,* Pierluigi Bissolino, Matteo D’Anello, Massimiliano Palladino, Ettore Perrone
Pharmacia & Upjohn, 20014 Nerviano (Milano), Italy
E-mail: marco.alpegiani@eu.pnu.com
Received 10 November 1997
Abstract: A practical and efficient route leading to the synthesis of 4-
3
(tert-butylcarbonyl)-7α-methoxy-3-methyl-∆ -cephem 1,1-dioxide (2),
a key-intermediate in the preparation of potent inhibitors of mammalian
serine proteinases, is reported. The new synthetic pathway has allowed
easy access to an array of 7-substituted cephem derivatives.
1
In the preceding paper we pointed out the pivotal role of 4-benzoyl-7α-
3
methoxy-3-methyl-∆ -cephem 1,1-dioxide
1
and its 4-tert-
butylcarbonyl counterpart 2 in the preparation of potent inhibitors of
2,3
mammalian serine proteinases, and described an innovative synthesis
to obtain 1. An improved method of synthesising 2, and the exploitation
of versatile intermediates for the introduction of various substituents at
cephem C-7 position, are dealt with in the present paper.
Scheme 1
butylmagnesium chloride in the presence of copper(I) iodide and
lithium chloride (THF, -70 °C) affording the crude ketone 5, whose
In previous approaches, when the pivaloyl moiety of cephem 2 was
introduced by Grignard reaction of cephem-4-carbonyl chlorides
already bearing the methoxy group at C-7α, yields ranged from poor to
5
peracid oxidation (MCPBA/ CH Cl ) gave the crystalline sulfone 6 (ca.
2
2
70% yield from 7-ADCA).
2,4
moderate. Holding the electron-withdrawing substituent vicinal to the
At this point we faced the problem of replacing the 7β-BOC group of
cephem 6 with the 7α-methoxy group of the target cephem 2. This
transformation was initially realised in an efficient but lengthy way,
comprising i) removal of the amino protecting group under acidic
conditions, ii) diazotisation with butyl nitrite and iii) rhodium catalysed
insertion of methanol (Scheme 2).
β-lactam carbonyl responsible for the unsatisfactory result, we planned
to postpone the methoxylation step and perform the Grignard reaction
on cephem substrates that featured a less reactive β-lactam ring. This
plan worked very well indeed, and we were able to convert the cheap
and commercially available 7β-amino-3-deacetoxycephalosporanic acid
3
(7-ADCA)
into
4-(tert-butylcarbonyl)-7α-methoxy-3-methyl-∆ -
By applying an optimised procedure developed at Merck for large scale
cephem 1,1-dioxide
2
in good overall yield, without need for
6
preparation of analogous compounds, the methoxy-deamination step
chromatographic purifications (Scheme 1 and 2).
might be expected to proceed more efficiently in terms of yield and
stereoselectivity, yet unattractive features of that protocol are the need
to operate under strictly controlled conditions and to employ special
equipment (flow reactor).
tert-Butoxycarbonylation of the 7-ADCA amino group with tert-
butylpercarbonate efficiently produced the N-protected cephem-4-
carboxylic acid 3. Acid chloride 4, obtained from 3 under Vilsmeier
conditions, cleanly underwent the crucial reaction with tert-
Scheme 2

March 1998
SYNLETT
323
Seeking
a more straightforward and practical access to 2, we
endeavoured to perform deprotection, diazotisation and methoxylation
steps in one pot. When 6 was exposed to the action of excess boron
7
trifluoride etherate in methanol (24 h, r.t.) and then to sodium nitrite (2
h), the 7α-methoxy derivative 2 was the preponderant reaction product
(Scheme 2), moreover it could be isolated as a white powder after
13
aqueous work-up and crystallisation from diisopropyl ether.
Direct conversion of 6 to 2, in addition to constituting an impressive
short cut, does not entail the handling of potentially sensitive
compounds like the diazocephem 8.
12
Similarly to the procedure described by Wiering and Wynberg,
triethylborane reacted with 8 affording an epimeric mixture of 7-ethyl
compounds 18 and 19, the trans isomer being predominant (Scheme 5).
On the other hand, considering the possibility of exploiting the synthetic
potential of the diazo function, cephem 8 was judged a useful
intermediate for the preparation of novel C-7 substituted cephem
sulfones. We aimed at converting, directly or indirectly, the diazo
moiety of 8 into alkyl groups, that were expected to improve the
Scheme 5
From a mechanistic point of view, the quaternary boron species A might
be generated first as the result of Lewis acid - Lewis base interaction;
1,2 alkyl shift from boron to carbon with simultaneous displacement of
nitrogen would then give B, eventually hydrolysed, possibly through the
intermediacy of enolborinate C, to 18 and 19.
2
stability of the cephem β-lactam moiety.
Prepared in gram scale, 8 was stored for weeks in the fridge without
apparent decomposition. Addition of hydrogen bromide or bromine
smoothly took place affording 7α-bromo and 7,7-dibromo derivatives 9
and 10, respectively (Scheme 3). Upon free-radical allylation with
9
allyltributyltin according to the Hanessian methodology, compounds
11 and 12 were obtained. Finally, transfer hydrogenation gave the
propyl cephems 13 and 14.
We were intrigued by the possibility of intercepting the postulated
species C with electrophiles other than the proton. While no desired
reaction took place when methyl iodide or acetyl chloride were
employed, using phenylselenyl chloride the electrophilic phenylselenyl
group was selectively delivered to the C-7α position of the cephem
nucleus to give 20 in satisfactory yield. The stereochemical outcome,
probably originating from attack of the electrophile at the less hindered
α-face, was assigned on the basis of NOE experiments.
Scheme 3
10
When 8 was treated with N-bromosuccinimide in methanol, the 7α-
11
bromo-7β-methoxy derivative 15 was produced stereoselectively
Scheme 6
(Scheme 4). Its conversion to 16 and 17 proceeded uneventfully using
the aforementioned protocol.
Finally, following oxidation with MCPBA (-20 °C), 20 was shown to
undergo easy elimination of phenylselenenic acid, affording a 1:1
13
isomeric mixture of E and Z 7-ethylenecephems 21 in moderate yield.
Summing up, a practical and efficient route to obtain a useful cephem
synthon (tert-butyl ketone 2) has been implemented. A variety of
unprecedented C-7 substituted 4-ketocephems was synthesised.
Acknowledgements: We are indebted to Dr. (Mrs) Daniela Borghi
(NMR spectra) and Dr. Emanuele Arlandini (MS spectra), and to Mr.
Edgardo Poma and Mr. Antonio Fiumanò (technical assistance).
Scheme 4

324
LETTERS
SYNLETT
References and Notes
dryness, affording crude 4 as a brownish powder (ca 4.2 g, 12.6
mmol), which was dissolved in dry THF (40 ml) and cooled to -70
°C. Separately, under a nitrogen blanket, a solution of LiCl (1.31
g, 30.9 mmol) and CuI (2.95 g, 15.5 mmol) in dry THF (20 ml)
was prepared. At -70 °C 1M tert-butylmagnesium chloride in THF
(15.5 ml, 15.5 mmol) was added dropwise. Then, via cannula, the
cold (-70 °C) solution of acyl chloride 4 in THF was added. After
stirring for 30 min at -70 °C, the reaction mixture was poured into
1)
Alpegiani, M.; Bissolino, P.; D’Anello, M.; Perrone, E. Synlett
1998, 319.
2)
Alpegiani, M.; Bissolino, P.; Corigli, R.; Del Nero, S.; Perrone,
E.; Rizzo, V.; Sacchi, N.; Cassinelli, G.; Franceschi, G.; Baici, A.
J. Med. Chem. 1994, 37, 4003.
3)
Alpegiani, M.; Bissolino, P.; Corigli, R.; Rizzo, V.; Perrone, E.
BioMed. Chem. Lett. 1995, 5, 687; ibidem 1995, 5, 691.
Alpegiani, M.; Bissolino, P.; Borghi, D.; Corigli, R.; Del Nero, S.;
Perrone, E.; Razzano, G.; Rizzo, V. BioMed. Chem. Lett. 1992, 2,
1127.
diethyl ether and 20% NH Cl 1:1. The insoluble material was
4
filtered off. The organic layer was washed twice with brine then
dried over Na SO and rotoevaporated, yielding crude 5 as a
2
4
brownish foam (ca. 4.5 g, 12.6 mmol). This product was dissolved
in CH Cl (60 ml), cooled to -20 °C, and treated with 55% meta-
4)
5)
Alpegiani, M.; Bissolino, P.; Perrone, E.; Cassinelli, G.;
Franceschi, G. Tetrahedron Lett. 1991, 32, 6207.
2
2
chloroperbenzoic acid (9.3 g, 29.6 mmol). The resulting mixture
was stirred at r.t. for 6 h. After this time, the precipitated meta-
chlorobenzoic acid (ca. 3 g) was removed by filtration. The filtrate
For analytical purposes a sample was purified by silica gel
chromatography (eluting with n-hexane/EtOAc mixtures) to
afford pure cephem 5:
was sequentially washed with aqueous NaHSO and aqueous
3
-1
1
IR (KBr) ν
3320, 1780, 1725(sh), 1695 cm . H NMR (200
NaHCO . After drying (Na SO ), the solution was concentrated
max
3
2
4
MHz, CDCl ) δ 1.21 (9H, s), 1.45 (9H, s), 1.75 (3H, d, J= 0.5 Hz),
in vacuo. The waxy residue was treated with CH Cl -Et O to give
3
2
2
2
3.09 (1H, d, J= 17.5 Hz), 3.52 (1H, br d, J= 17.5 Hz), 4.97 (1H, d,
the title product as a white powder (3.9 g, 78% yield from 3). IR
-1
1
J= 4.6 Hz), 5.20 (1H, d, J= 9.1 Hz, exch. D O), 5.51 (1H, dd, J=
(KBr) ν
3400, 1780, 1700 cm . H NMR (200 MHz, CDCl )
3
2
max
4.6 and 9.2 Hz).
δ 1.23 (9H, s), 1.45 (9H, s), 1.72 (3H, s), 3.54 (1H, d, J= 18.2 Hz),
3.88 (1H, d, J= 18.2 Hz), 4.79 (1H, d, J= 3.7 Hz), 5.81 (2H, m).
4-tert-Butylcarbonyl-7α-methoxy-3-methyl-∆ -cephem
6)
7)
8)
9)
Blacklock, T.J.; Butcher, J.V.; Sohar, P.; Lamanec, T.R.;
Grabowski, E.J.J. J. Org. Chem. 1989, 54, 3907.
3
1,1-dioxide (2). Compound 6 (1.17 g) was suspended in CH OH
3
After this time complete conversion of 6 into 7 had occurred
(HPLC monitoring).
(35 ml) and treated with 50% boron trifluoride etherate (6 ml).
The suspension was stirred for 24 h at r.t. To the resulting clear
solution, cooled to 0 °C, sodium nitrite (1.44 g) was added and
stirring was continued for 6 h at r.t. The mixture was poured into
EtOAc/water and the organic layer was washed sequentially with
Mother liquor removed minor contaminants together with an
appreciable quantity of the C-7 epimer of 2.
Hanessian, S.; Alpegiani, M. Tetrahedron Lett. 1986, 27, 4857.
Hanessian, S.; Alpegiani, M. Tetrahedron 1989, 45, 941.
aqueous NaHCO and brine, then dried (Na SO ). Removal of the
3
2
4
10) Cama, L.D.; Leanza, W.J.; Beattie, T.R.; Christensen, B.G. J.
Amer. Chem. Soc. 1972, 94, 1408.
solvent left a waxy solid, whose crystallisation from diisopropyl
ether yielded the title product as a white powder (0.57 g, 63%
-1
1
yield). IR (KBr) ν
1780, 1690 cm . H NMR (200 MHz,
11) The stereochemical outcome is consistent with the attack of the
bromonium ion from the less hindered α-face of diazocephem 8,
and displacement of nitrogen by methanol, with inversion, of the
max
CDCl ) δ 1.26 (9H, s), 1.70 (3H, s), 3.51 (2H, d, J= 18.1 Hz), 3.56
3
(3H, s), 3.93 (2H, d, J= 18.1 Hz), 4.66 (1H, m), 5.16 (1H, d, J= 1.7
Hz).
10
intermediate bromodiazonium ion.
3
7β-Amino-4-(tert-butylcarbonyl)-3-methyl-∆ -cephem
12) Wiering, J.S.; Winberg, H. J. Org. Chem. 1976, 41, 1574.
1,1-dioxide (7). TFA (5 ml) was added to a suspension of 6 (1.2 g,
13) All new compounds were fully characterised by spectroscopic
means. Selected experimental and spectral data are given below:
7β-tert-(Butoxycarbonyl)amino-3-deacetoxycephalosporanic
acid (3). Triethylamine (5.6 ml, 40 mmol) was added dropwise to
3.1 mmol) in CH Cl (5 ml) and anisole (1 ml). After stirring for 1
2
2
h at r.t., the solvent was evaporated in vacuo. The residue was
taken up in EtOAc, washed with aqueous NaHCO and brine, then
3
dried (Na SO ). The solvent was rotoevaporated and the residue
2
4
a
mixture of 7β-amino-3-deacetoxycephalosporanic acid (7-
was treated with a mixture of CH Cl /diisopropyl ether. The title
2
2
ADCA, 4.28 g, 20 mmol) in dioxane (80 ml) and water (40 ml).
Di-tert-butyl dicarbonate (6.6 g, 30 mmol) was then added and the
resulting mixture was stirred at r.t. for 6 h before partitioning
between diethyl ether and water. The aqueous phase was acidified
with 8% HCl and extracted with EtOAc. Following drying over
product was thus obtained as a white powder (0.89 g, 100% yield).
-1
1
IR (KBr) ν
1785, 1690 cm . H NMR (200 MHz, CDCl )
3
max
δ 1.23 (9H, s), 1.7 l(3H, s), 3.48 (1H, d, J= 18.2 Hz), 3.87 (1H,
br.d, J= 18.2 Hz), 4.71 (1H, br.d, J= 4.7 Hz), 4.80 (1H, br.d. J= 4.7
Hz).
Na SO and evaporation of the solvent, a foamy solid was
2
4
3
4-tert-Butylcarbonyl-7-diazo-3-methyl-∆ -cephem
1,1-dioxide
obtained which upon treatment with petroleum ether turned into a
white powder (5.4 g, 86% yield). IR (KBr) ν 3600-2500, 1780,
(8). Acetic acid (60 µl) and tert-butylnitrite (0.53 ml, 4.5 mmol)
were sequentially added to a solution of of 7 (0.86 g, 3.0 mmol) in
max
-1
1
1710 cm . H NMR (200 MHz, DMSO-d ) δ 2.00 (3H, s), 3.33
6
CH Cl (40 ml). The reaction mixture was heated at reflux for 1 h,
2
2
(1H, d, J= 17.8 Hz), 3.49 (1H, d, J= 17.8 Hz), 4.98 (1H, d, J= 4.6
Hz), 5.34 (1H, dd, J= 4.6 and 8.9 Hz), 7.95 (1H, d, J= 8.9 Hz,
then poured into aqueous NaHCO . The organic layer was
3
separated, washed with water and dried over Na SO . Removal of
2
4
exch. D O).
2
the solvent in vacuo, left an oily residue which upon treatment
with petroleum ether turned into a yellow powder (0.8 g, 90%
yield).
7β-tert-Butoxycarbonylamino-4-(tert-butylcarbonyl)-3-methyl-
3
∆ -cephem 1,1-dioxide (6). A solution of 3 (4.08 g, 13 mmol) in
dry THF (40 ml) was cooled to 0 °C and treated with oxalyl
chloride (1.68 ml, 19.5 mmol) and DMF (70 µl, 0.9 mmol). The
resulting solution was stirred for 2 h at 0-5 °C, then it was
rotoevaporated to dryness (temperature of the bath <25 °C). The
residue was taken up with toluene and carefully evaporated to
-1
1
IR (CHCl ) ν
2100, 1790, 1700 cm . H NMR (200 MHz,
max
3
CDCl ) δ 1.25 (9H, s), 1.78 (3H, s), 3.66 (1H, d, J= 17.1 Hz), 3.86
3
(1H, br.d, J= 17.1 Hz), 5.50 (1H, s).