An efficient and concise entry to (-)-4,5-dihydroxy-d-threo-l norvaline. Formal synthesis of clavalanine

Article abstract of DOI:10.1016/0040-4039(91)85018-Z

The title amino acid (2) has been synthesized for the first time in a seven-step sequence from D-ribonolactone, in 20% overall yield. Since the N-carbamoyl derivative of (2) in its y-lactone form has been used to prepare clavalanine, a formal total synthe

Full text of DOI:10.1016/0040-4039(91)85018-Z

Tetrahedron Letters, Vol.32, No.17, pp 1979 1981 1991
Prhlted inGreatBritain
0040-4039/91 $3.00
Pergamon Press plc
+ .00
AN EFFICIENT AND CONCISE E N T R Y TO (-)-4,5-DIRYDROXY-D-threo-i-NORVA[INE.
FORMAl SYNTHESIS OF CiAVALANIHE.
JESUS ARIZA, JOSEP FONT, * and ROSA M. ORTU~O*
Uniter de091mlcaOr~nica, Departament de Qulmica, UniversitatAuthnomadeBarcelona,
08193Bellaterra (Barcelona), Spain.
SUmmary: The title amino acid (2) has been synthesized for the first time in
a seven-
step sequence from D-ribonolactone, in 20% overall yield. Since the _N-carbamoyl derivative
of (2) in its Y-lactone form has been used to prepare clavalanine,
synthesis of this antibiotic is derived.
a formal total
Clavalanine
(No 22-5417)
(I) is
a
clavam antibiotic isolated from
St reptomyces clavuligerus in 1983.1 In contrast to the previously isolated
~-laetams of this microorganism in all of w h i c h the carbon atom joining the
bieyclic ring systems has the
R
configuration, the new congener possesses
S
stereochemlstry at the ring juncture. Ic Presumably as consequence of this
stereochemical difference, clavalanine (I) does not exert its antimicrobial
activity, like the other B-lactams, via inhibition of bacterial cell wall
synthesis; it is also neither
Moreover, (I) is unique
a
substrate nor an inhibitor of ~lactamases.
that it is an antimetabolite of O-
in
sueeinylhomoserine and intervenes in methionine biosynthesis, whereas most
B-lactam antibiotics inhibit peptidoglycan biosynthesis, la
A
total synthesis of (I) has been reported by
group that involved the multistep preparation of the 4~5-dihydroxy-D-threo-
[-norvaline derivative (10), as key intermediate, from D-xylose. 2 later,
Williams e t a l . 3 have prepared (I0) using an eleetrophilic glycine template
a
Hoffmann-la Roche
a
obtained through resolution of
a
raeemic suitable precursor.
H
H2N, H
OH
NH2
I
2
c l a v a l a n i n e
1979

1980
We report herein the first enantioselective synthesis of the free
amino acid (2), that has been also characterized in this work as the
hydrochloride salt (9) and the N-carbamoyl derivative (I0), using available
D-ribonolactone as
(Scheme I).
a
ehiral precursor in
a
short synthetic
sequence.
HO
0
a
~
,
0
c
•
d
O ~
b *
0~:~ 0
3
4
S
0
f
)
0
g
)
h
)
2
~ )
I~ H C b z
N H C D z
NH 2
Ph
6
7
8
Reagents. (a): PhCHO, HCI. (b): (CF3SO2)20 , pyr,
0
°C. (c): NAN3, DMF,
OC, 3:1 H20-THF.
2 arm, EtOH. (h): Dowex 50W-
r.t. (d): H2, I0% Pd/C, EtAcO. (e) PhCH2OCOCI , NaHCO3
0
(f): NaH, THF, -20 °C; IN HCI. (g): H2, Ra-Ni,
x2 (H+).
Scheme
easily
i
D-Ribonolactone
(3)
was
converted
into
the
known
azido
derivative (5) through the three-step sequence described by Fleet et el. 4
Thus, (3) reacted with benzaldehyde in concentrated hydrochloric acid to
give the 3,4-O-benzylidene derivative (4). 5 Reaction of the alcohol (4)
with trifluoromethane sulfonic anhydride in pyridine gave the corresponding
triflate; subsequent treatment with sodium azide formed the azido lactone
(5) (57~ yield from (3)), with overall retention of configuration at C-2. 4
Hydrogenation of the azldo group (10% palladium on charcoal) and
reaction of the resultant amine with benzyloxycarbonyl
ehloroformate, CbzCl) and excess NaHCO3 in 3:1 H20-THF, at
new carbamate (6), m.p. 129-131 °C, {~} D -2060 (c 3.8 in chloroform) (84~
yield). 6 Butenollde (7), m.p. 141-143 °C, {~} -3"8o (c 2.66 in acetone),
was obtained from (6) in 65% yield via base (NaH) induced elimination of
chloride (benzyl
0
°C, gave the
D
a

1981
benzaldehyde,
followed by ring rearrangement to give the 1,4-1aetone (7).
This size ring contraction probably takes place during the acid hydrolysis
of the reaction mixture. 7 The elimination reaction performed directly on
the
azlde
(5)
was
not
satisfactory
enough,
obtaining
the
expected
butenolide (ll) in only 15~ yield. This fact has been attributed to the
lability of these azldes in the reaction and work-up conditions.
H y d r o g e n a t i o n of the C-C double bond in (7) (Ra-Ni,
afforded dlastereospeciflcally the amino lactone (8), immediate precursor
of (2), as single isomer (13C NMR). This strategy to induce chirallty at
2
arm. pressure)
a
C-2 in D-rlbonolactone derivatives, from C-4 as stereogenlc center, has been
previously used in our laboratory in the s t e r e o c o n t r o l l e d syntheses of both
(-)-erythro-
hydrogenation of butenolldes prior or later to the introduction of
nitrogen group provides an easy and useful method to control the relative
and absolute configuration of homochiral 2,4-dIsubstltuted amino 1,4-
lactones.
Compound (8) was treated with Dowex 50W-X2 (H+) resin to give the
and
(-)-threo-- y - hydroxynorvallnes. 8
Stereospeciflc
a
amino acid (2), m.p. 205-210 °C,
overall yield from (3).
{e} D -19"5° (c 1.7 in water), in 20~
The derivatives (9) and (I0) were also easily prepared from the amino
lactone (8). Thus, treatment of (8) with hydrogen chloride gave the
hydrochlorlde (9), m.p. 195-200 °C (dec.), {~}D +18"0° (c 1.6 in water). On
the other hand, reaction of (8) with CbzCl in the standard conditions
described above yielded the carbamate (lO), used as precursor of
clavalanine, 2 whose physic c o n s t a n t s compare well with those previously
described for this compound: m.p. 114-115 °C, (~}D +3"60 (c 1.6 in
methanol) (llt. 2 m.p. 112-115 °C, (a}D +3.3
o
(c I.I in methanol)).
X
N3
9
N H 2 . H C l
11
10 N H C D z
As concluslon~ an efficient and concise method for the preparation of
(-)-4,5-dlhydroxy-D-threo-L-norvaline (2) and its derivatives (9) and (lO)
has been achieved, starting from D-ribonolactone. With these products in

1982
our hand,
a
formal total synthesis of the antibiotic clavalanine (I) has
been accomplished.
Acknowledgements.- J. A. thanks the Ministerio de Educaci6n
y
Ciencia for
a
grant.
Financial
support
from DGICYT, project
PB86-0320
and PB89-0287
is gratefully acknowledged.
R E F E R E N C E S AND N O T E S
i. (a) Pruess, D. I.; Kellet, M. J. Antibiot. 1983, 36, 208. (b) Evans, R.
H.; Ax, H.; Jacoby, A.; Williams, T. ~.; Jenkins, E.; Scannel, J. J.
Antibiot. 1983, 36, 213.
(c) Muller, J. C. Yoome, V.; Pruess, D. L.
Blount, J. F.; Weigele, M. J. Antibiot. 1983, 36, 217.
2. De Bernardo, S.; Tenglj J. P.; Sasso, G.; Weigele, M. J._~. Or S. Chem.
1985, 50, 3457.
3. Williams, R. M.; Sinclair, P. J.; Zhai, D.; Chen, D. J. Am. Chem. Soc.
1988, Ii0, 1547.
4. Dho, J. C.; Fleet, G. W. J.; Peach, J. M.; Prout, K.; Smith, P. W.
Tetrahedron left. 1986, 27, 3203.
5. (a) Chen, S. Y.;
Joulli4, M. M. J-_t" Ors. Chem. 1984, 49, 2168. (b)
Baggett, M.; Bucanan, J. G.; Fatah, M. Y.; lachut, C. H.; McCullough, K.
J.; Webber, J. M. J. Chem. Soc.~ Chem. Commun. 1985, 1826.
6. All
new
compounds
in
this
paper,
(2),
(6),
(7),
(8),
(9),
have
satisfactory analytical data and spectra (IR, IH and 13C NMR) consistent
with the structures reported.
7. Details
on
this
elimination-rearrangement
process,
its
scope
and
limitations will be reported elsewhere.
8. Ariza, J.; Font, J.; 0rtu~o, R. M. Tetrahedron, 1990, 46, 1931.
(Receivedin UK 11Febmary1991)