Synthesis of (-)-3-Amino-3-deoxyquinie Acid

Article abstract of DOI:10.1021/jo981291l

An efficient synthesis of 3-amino-3-deoxyquinic acid, starting from (-)-quinic acid, is reported and its importance for the biosynthesis of 3-amino-5-hydroxybenzoic acid, the precursor of mCvN-units in ansamycin and mitomycin antibiotics is discussed.

Full text of DOI:10.1021/jo981291l

J . Org. Chem. 1998, 63, 9753-9755
9753
Syn th esis of (-)-3-Am in o-3-d eoxyqu in ic Acid
Michael Mu¨ller,† Rolf Mu¨ller,‡ Tin-Wein Yu, and Heinz G. Floss*
Department of Chemistry, Box 351700, University of Washington, Seattle 98195-1700
Received J uly 2, 1998
An efficient synthesis of 3-amino-3-deoxyquinic acid, starting from (-)-quinic acid, is reported and
its importance for the biosynthesis of 3-amino-5-hydroxybenzoic acid, the precursor of mC7N-units
in ansamycin and mitomycin antibiotics is discussed.
A variety of antibiotics contain a biosynthetically
unique structural moiety, called an mC7N-unit, consisting
of a six-membered carbocycle carrying an extra carbon
and a nitrogen in a meta arrangement. This unit was
first recognized as a structural element of the ansamycin
antibiotic rifamycin B (1).1 It is present in all the
ansamycins2 and ansamitocins3 as well as in the struc-
turally quite different mitomycins.4
extracts of the rifamycin B producer, Amycolatopsis
mediterranei, consistent with the proposed9 pathway of
AHBA formation shown in Scheme 1. The conversion of
5-deoxy-5-aminoshikimic acid (aminoSA) into AHBA
under the same conditions is notably lower (0.9%) but
nevertheless clearly detectable.7 This is probably due to
some transformation of aminoSA into aminoDHS cata-
lyzed by a presumed aminoSA/aminoQA dehydrogenase
(RifI)10 present in A. mediterranei or by the normal
shikimate dehydrogenase of the organism.
The question then arose whether 3-deoxy-3-aminoqui-
nic acid (aminoQA) could be excluded as a biosynthetic
precursor. In view of the low incorporation of aminoSA,
we assumed that aminoQA would also be incorporated
at a very low level into AHBA, if at all. For further
investigations of the enzymatic dehydration step from
aminoDHQ to aminoDHS, additional amounts of the
former were needed as substrate. Our previous prepara-
tion of aminoDHQ proceeded by cyclization of amino-
DAHP, catalyzed by dehydroquinate synthase from Es-
cherichia coli.11 However, aminoDAHP itself is only
available through a multistep synthesis8 similar to the
DAHP synthesis described by Frost and Knowles.12
A
Tracer and genetic experiments have demonstrated the
shikimate pathway origin of this mC7N unit via 3-amino-
5-hydroxybenzoic acid (AHBA),5 although none of the
early pathway intermediates were incorporated.6 Kim et
al. have shown that labeled 5-deoxy-5-amino-3-dehy-
droshikimic acid (aminoDHS),7 5-deoxy-5-amino-3-dehy-
droquinic acid (aminoDHQ)6 as well as 3,4-dideoxy-
4-amino-D-arabino-heptulosonate 7-phosphate (amino-
DAHP),7,8 are converted efficiently (95, 41, and 45%,
respectively) into AHBA upon incubation with cell free
short nonenzymatic synthesis of aminoquinic acid would
make possible a more efficient preparation of aminoDHQ,
assuming that aminoQA can be oxidized by chemical
means or by the NADP+-dependent shikimate dehydro-
genase from E. coli.
Starting from (-)-quinic acid (1) the known quinide 2
was obtained in 74% yield (Scheme 2).13 Cleavage of the
lactone with NaOMe-MeOH, protection of the 3-hydroxy
group14 and removal of the isopropylidine group with CF3-
CO3H in THF-H2O gave diol 3. These three steps can
be carried out advantageously without purification of the
intermediates to give the diol 3 in 85% yield. The
elimination of methylsulfonic acid to the epoxide 4 could
not be performed with DBU in THF at ambient temper-
ature. After addition of a catalytic amount of LiBr,15 the
† Present address: Institut fu¨r Biotechnologie, Forschungszentrum
J u¨lich GmbH, D-52425 J u¨lich, Germany.
‡ Present address: Gesellschaft fu¨r Biotechnologische Forschung,
Mascheroder Weg 1, D-38124 Braunschweig, Germany.
(1) (a) Brufani, M.; Kluepfel, D.; Lancini, G. C.; Leitich, J .; Mesent-
sev, A. S.; Prelog, V.; Schmook, F. P.; Sensi, P. Helv. Chim. Acta 1973,
56, 2315. (b) White, R. J .; Martinelli, E.; Gallo, G. G.; Lancini, G.;
Beynon, P. Nature 1973, 243, 273.
(2) Rinehart, K. L., J r.; Shields, L. S. Fortschr. Chem. Org. Naturst.
1976, 33, 231.
(3) Higashide, E.; Asai, M.; Ootsu, K.; Tanida, S.; Kozai, Y.;
Hagesawa, T.; Kishi, T.; Sugino, Y.; Yoneda, M. Nature 1977, 270, 722.
(4) Hornemann, U. In Antibiotics; Corcoran, J . W., Hahn, F. E., Eds.;
Springer-Verlag: Berlin, 1981; Vol. 4, p 297.
(9) Floss, H. G.; Beale, J . M. Angew. Chem. 1989, 101, 147; Angew.
Chem., Int. Ed. Engl. 1989, 28, 146.
(10) August, P. R.; Tang, L.; Yoon, Y. J .; Ning, S.; Mu¨ller, R.; Yu,
T.-W.; Taylor, M.; Hoffmann, D.; Kim, C.-G.; Zhang, X.; Hutchinson,
C. R.; Floss, H. G. Chem. Biol. 1998, 5, 69.
(11) Medhi, S.; Frost, J . W.; Knowles, J . R. Methods Enzymol. 1987,
142, 306.
(12) Frost, J . W.; Knowles, J . R. Biochemistry 1984, 23, 4465.
(13) Billen, G.; Karl, U.; Scholl, T.; Stroech, K. D.; Steglich, W. In
Natural Product Chemistry III; Rahman, A., Le Quesne, P. W., Eds.;
Springer-Verlag: Berlin, 1988; p 305.
(5) Cf.: (a) Ghisalba, O. Chimia 1985, 39, 79. (b) Floss, H. G. Nat.
Prod. Rep. 1997, 14, 433.
(6) Kim, C.-G.; Kirschning, A.; Bergon, P.; Zhou, P.; Su, E.; Sauer-
brei, B.; Ning, S.; Ahn, Y.; Breuer, M.; Leistner, E.; Floss, H. G. J .
Am. Chem. Soc. 1996, 118, 7486 and references therein.
(7) Kim, C.-G.; Kirschning, A.; Bergon, P.; Ahn, Y.; Wang, J . J .;
Shibuya, M.; Floss, H. G. J . Am. Chem. Soc. 1992, 114, 4941.
(8) Kirschning, A.; Bergon, P.; Wang, J . J .; Breazeale, S. A.; Floss,
H. G. Carbohydr. Res. 1994, 256, 245.
(14) Sutherland, J . K.; Watkins, W. J .; Bailey, J . P.; Chapman, A.
K.; Davies, G. M. J . Chem. Soc., Chem. Commun. 1989, 1386.
(15) Seebach, D.; Thaler, A.; Blaser, S.; Ko, Y. Helv. Chim. Acta.
1991, 74, 1102.
10.1021/jo981291l CCC: $15.00 © 1998 American Chemical Society
Published on Web 12/04/1998

9754 J . Org. Chem., Vol. 63, No. 26, 1998
Mu¨ller et al.
Sch em e 1. P r op osed Biosyn th etic P a th w a y to
AHBA
Sch em e 3. En zym a tic F or m a tion of Am in oDHQ
of the diallylamine group17 could not be performed on a
100 mg scale so that for completion of the synthesis
dibenzylamine was chosen as the nucleophile. Compound
5 can easily be deprotected by catalytic transfer hydro-
genation with formic acid18 and subsequent hydrolysis
of the ester under classical conditions to give (-)-3-amino-
3-deoxyquinic acid (6) in 14% overall isolated yield over
nine steps.19
Although aminoquinic acid is a substrate for the NAD+-
dependent aminoquinate-shikimate dehydrogenase (RifI)10
from A. mediterranei21 and the NADP+-dependent shiki-
mate dehydrogenase from E. coli (Scheme 3),22 the
oxidation of aminoquinic acid to aminoDHQ could not be
performed on a preparative scale with either of these
enzymes. When about 5-10% of the NAD+ or NADP+,
respectively, have been consumed, the reaction gradually
comes to a halt. With the aminoquinate-shikimate de-
hydrogenase from A. mediterranei, after that amount of
NAD+ consumption in the oxidation of aminoquinic acid
no further product formation can be detected by UV
Sch em e 2. Syn th esis of 3-Deoxy-3-a m in oqu in ic
Acid
1
(formation of NADH, 340 nm) or by H NMR spectros-
copy. The same was observed by assay of NADPH
formation in the oxidation of aminoquinic acid with
shikimate dehydrogenase from E. coli. These observa-
tions are not due to the NADH (NADPH) concentration
or the NADH:NAD+ ratio. Addition of alcohol dehydro-
genase and acetaldehyde to the reaction mixture with
aminoquinate-shikimate dehydrogenase does not result
in higher product formation, suggesting that the enzyme
is not inhibited by NADH. Product inhibition by amino-
DHQ, however, cannot be ruled out.
Nevertheless, formation of aminoDHQ by enzymatic
oxidation of aminoquinic acid was unambiguously dem-
onstrated, directly by proton NMR spectroscopy and
indirectly after aromatization to protocatechuic acid and
silylation by GC-MS.
Incubations of aminoquinic acid (6) with cell-free
extracts of the rifamycin B producer, A. mediterranei,
revealed that 6 is converted into AHBA to a similar
extent as aminoSA (1-2%). This finding, as the earlier
one with aminoSA,6 does not answer the question whether
aminoquinic acid, and/or aminoSA, are biosynthetic
precursors of AHBA. Their conversions into AHBA are
clearly less efficient than those of aminoDAHP, amino-
DHQ, and aminoDHS. However, it cannot be ruled out
elimination proceeded smoothly at 0 °C. After 10 min at
0 °C, the reaction was almost complete. At higher
temperatures or longer reaction times, migration of the
1-O-benzoyl group to the hydroxy group at C-3 took place.
The epoxide 4 can be opened nucleophilically at C-3 in
the presence of Ti(OiPr)4 either with diallylamine or even
more advantageously with dibenzylamine.16 TLC analy-
ses of the reaction mixture showed that the reaction with
dibenzylamine is almost quantitative without the forma-
tion of any major side products. The rather low yield of
product 5 (40%) is due to the harsh workup conditions
(Kugelrohr distillation 120 °C/0.1 Torr), which are neces-
sary to remove excess dibenzylamine. The deprotection
(17) (a) Gigg, R.; Conant, R. J . Carbohydr. Chem. 1983, 1, 331. (b)
Laguzza, B. C.; Ganem, B. Tetrahedron Lett. 1981, 22, 1483. (b)
Moreau, B.; Lavielle, S.; Marquet, A. Tetrahedron Lett. 1977, 18, 2591.
(18) Elamin, B.; Anantharamaiah, G. M., Royer, G. P., Means, G.
E. J . Org. Chem. 1979, 44, 3442.
(19) After this work was completed, a publication appeared20 which
used a similar strategy to synthesize 3-azido-3-deoxyshikimic acid.
(20) Kim, C. U.; Lew, W.; Williams, M. A.; Liu, H.; Zhang, L.;
Swaminathan, S.; Bischofberger, N.; Chen, M. S.; Mendel, D. B.; Tai,
C. Y.; Laver, W. G.; Stevens, R. C. J . Am. Chem. Soc. 1997, 119, 681.
(21) Mu¨ller, R.; Mu¨ller, M.; Yu, T.-W.; Floss, H. G. Unpublished
results.
(22) Chaudhuri, S.; Anton, I. A.; Coggins, J . R. Methods Enzymol.
1987, 142, 315.
(16) Caron, M.; Sharpless, K. B. J . Org. Chem. 1985, 50, 1560.

Synthesis of (-)-3-Amino-3-deoxyquinic Acid
J . Org. Chem., Vol. 63, No. 26, 1998 9755
294 (16.6), 293 (100), 171 (24.8), 153 (22.2), 143 (16.6), 139
(10), 126 (15.5), 105 (18.8). Rf ) 0.45 (EtOAc-hexane 1:1).
C15H16O6 (292.3).
that under different incubation conditions or with ex-
tracts of cells grown under different conditions these two
substrates can be used more efficiently for the biosyn-
thesis of AHBA and thus rifamycin B. The presence of
an enzyme in A. mediterranei, encoded by the rifI gene
in the rifamycin biosynthetic gene cluster, which specif-
ically oxidizes aminoquinate and aminoshikimate, but
not quinate, at C-3 suggests a physiological role for these
compounds in rifamycin formation, for example in a
possible salvage pathway of AHBA formation.
(-)-Meth yl (1R,3R,4S,5R)-1-(Ben zoyloxy)-5-(N,N-d iben -
zyla m in o)-3,4-d ih yd r oxycycloh exa n e Ca r boxyla te (5). A
mixture of 880 mg of epoxide 4 (3 mmol), 2 mL of dibenzy-
lamine (10 mmol), and 0.2 mL of Ti(OiPr)4 (0.68 mmol) was
stirred at room temperature under argon for 48 h. After the
mixture had been passed through a short column of silica gel
(3 × 3 cm, EtOAc-hexane 1:1 400 mL, followed by 500 mL of
EtOAc), the combined eluates were evaporated. The residue
was subjected to a Kugelrohr distillation (120 °C/0.1 Torr, 15
min). The product in the remaining residue was purified by
flash chromatography (SiO2, EtOAc/hexane 1:1). Recrystalli-
zation from EtOAc-petroleum ether gave 587 mg 5 (1.2 mmol,
Exp er im en ta l Section
(-)-Meth yl (1S,3R,4R,5R)-1-(Ben zoyloxy)-3-O-(m eth yl-
su lfon yl)-4,5-d ih yd r oxycycloh exa n e Ca r boxyla te (3). To
a solution of 486 mg (1.5 mmol) of 1-O-benzoyl-4,5-O-isopro-
pylidenequinide (2) in 5 mL of MeOH was added a solution of
95 mg (1.75 mmol) NaOMe in 1 mL MeOH. After 30 min at
room temperature, the solution was diluted with 50 mL of
ether and washed with brine, NH4Cl solution, NaHCO3 solu-
tion, and brine. The organic layer was dried over Na2SO4, and
the solvent was evaporated.
The residue was dissolved in 5 mL of dry CH2Cl2. Pyridine
(300 µL, 4 mmol) and a solution of 350 mg of methanesulfonic
anhydride (2 mmol) in 0.5 mL of dry CH2Cl2 were added. After
30 min at room temperature, the solution was diluted with
50 mL of ether and washed with brine, NH4Cl, NaHCO3, and
brine. The organic layer was dried over Na2SO4, and the
solvent was evaporated.
To a stirred solution of the residue in 5 mL of THF/H2O 4:1
at 0 °C was added 0.45 mL of trifluoroacetic acid. The resulting
solution was allowed to warm to room temperature and left
overnight. The reaction mixture was diluted with 50 mL of
EtOAc and washed with brine, NH4Cl, NaHCO3, and brine.
The organic layer was dried over Na2SO4, and the solvent was
removed under reduced pressure. The crude product was
purified by flash chromatography (SiO2, gradient of EtOAc-
hexane 1:1 to 3:1) to give 495 mg (1.3 mmol, 85%) of 3 as a
white foam.
1
40%) of colorless needles, mp 184 °C. H NMR (CDCl3): 1.85
(t, J ) 13.02 Hz, 1H), 1.98 (dd, J ) 14.4, 1.5 Hz, 1H), 2.73 (d,
br, 2H), 3.35 (d, J ) 13.1 Hz, 2 H), 3.37 (m, 1H), 3.67 (dd, J )
10.5, 3.2 Hz, 1H), 3.76 (s, 3H), 3.87 (d, J ) 13.1 Hz, 2H), 4.24
(m, br, 1H), 7.20-7.24 (m, 10H), 7.37 (t, J ) 7.6 Hz, 2H), 7.55
(t, J ) 7.4 Hz, 1H), 7.84 (d, J ) 7.3 Hz, 2H). 13C NMR
(CDCl3): 28.99, 33.50, 51.83, 52,62, 53.26, 67.41, 69.80, 79.96,
127.27, 128.21, 128.49, 128.85, 129.92, 130.01, 132.95, 138.50,
20
165.30, 172.08. [R]D -51° (c ) 0.0043 acetone). Rf ) 0.7
(EtOAc-hexane 1:1). MS (FAB): 490 (100), 400 (7), 368 (12).
HR-MS (FAB):
29H31NO6 (489.3).
(-)-Meth yl (1R,3R,4S,5R)-5-Am in o-1-(ben zoyloxy)-3,4-
C29H32NO6 calcd 490.2229, obsd 490.2245.
C
d ih yd r oxycycloh exa n e Ca r boxyla te. To the protected ami-
noquinic acid 5 (800 mg, 1.4 mmol) in 2.5 mL of a 4.4% solution
of formic acid in methanol was added 600 mg of palladium
black (Aldrich). After complete reaction (30 min) the mixture
was passed through a short silica gel column with 10 mL of
MeOH. Purification was performed by preparative TLC (SiO2,
EtOAc-MeOH 3:1) to give 400 mg (90%, about 90% pure) of
1
the debenzylated product, mp 175 °C. H NMR (CDCl3): 1.98
(t, br, 2H), 2.66 (t, br, 2H), 3.60-3.65 (m, br, 2H), 3.62 (s, 3H),
4.06 (s, br, 1H), 7.32 (t, 2H), 7.44 (t, 1H), 7.97 (d, 2H). Rf )
0.15 (EtOAc-MeOH 3:1). MS (FAB): 310 (100), 188 (20), 170
(15). HR-MS:
15H19NO6 (309.32).
(-)-(1R,3R,4S,5R)-5-Am in o-1,3,4,-tr ih yd r oxy-cycloh ex-
C15H20NO6 calcd 310.1290, obsd 310.1282.
1H NMR (CDCl3): 2.14 (m, 2H), 2.81 (d, J ) 14.0 Hz, 2H),
3.08 (s, 3H), 3.69 (m, 1H), 3.71 (s, 3H), 4.25 (q, J ) 3.0 Hz,
1H), 5.12 (ddd, J ) 12.9, 4.0 Hz, 1H), 7.41 (t, J ) 7.3 Hz, 2H),
7.55 (t, J ) 7.3 Hz, 1H), 7.99 (d, J ) 8.1 Hz, 2H). 13C NMR
(CDCl3): 33.8, 37.85, 38.56, 52.96, 68.84, 72.65, 78.39, 79.81,
128.51, 129.44, 129.92, 133.5, 165.36, 170.97. C16H20O9S
(388.4).
(-)-Meth yl (1S,3R,4R,5R)-1-(Ben zoyloxy)-3,4-ep oxy-5-
h yd r oxycycloh exa n e Ca r boxyla te (4). To a stirred solution
of 4.2 g of mesylate 3 (10.8 mmol) and 3.8 mL of DBU (25
mmol) in 100 mL of dry THF was added 50 mg of LiBr at -0
°C. The reaction mixture was allowed to warm to room
temperature. After 10 min the reaction mixture was passed
through a short silica gel column (4 × 4 cm, eluent 400 mL
EtOAc-hexane 1:1, then 2 l EtOAc). Rechromatography (SiO2,
30 × 3 cm, EtOAc-hexane 1:1, 800 mL) gave pure epoxide 4
(2.5 g, 8.5 mmol, 80%) as a colorless oil.
C
a n eca r boxylic Acid (Am in oqu in ic Acid ) (6). To 400 mg
(1.2 mmol) of the product from the previous step in 10 mL of
methanol at 0 °C was added 100 mg of LiOH (2.4 mmol) in 5
mL of H2O. After 12 h at room temperature more LiOH was
added (50 mg in 2 mL of H2O, total 3.6 mmol). The hydrolysis
was complete after 5 h. The reaction mixture was loaded on a
short ion exchange column (10 × 1 cm, Dowex AG 1 × 8,
gradient H2O to 3 M formic acid). The product eluted with 0.5
M formic acid, and evaporation of solvent gave 180 mg (80%)
of a white powder, mp 201-203 °C. [R]D20 -44° (c ) 0.004 H2O).
1H NMR (D2O): 1.93-2.10 (m, 4H), 3.42 (dt, J ) 10.5 Hz, 4.1
Hz, 1H), 3.56 (dd, J ) 10.5 Hz, 3.0 Hz, 1H), 4.07 (d, J ) 3.0
Hz, 1H). 13C NMR (D2O): 37.89, 37.96, 48.85, 70.22, 72.29,
76.14, 181.23. MS (ES-):190.1 (100), 121.0 (22). HR-MS
(FAB): C7H14NO5 calcd 192.0872, obsd 192.0871. C7H13NO5
(191.19).
1H NMR (CDCl3): 2.01 (d, J ) 15.5 Hz, 1H), 2.09 (dd, J )
12.6, 10.6 Hz, 1H), 2.28 (ddd, J ) 12.6, 5.2, 1.5 Hz, 1H), 3.13
(ddd, J ) 15.5, 4.9, 1.9 Hz, 1H), 3.39 (m, 1H), 3.47 (t, J ) 4.3
Hz, 1H), 3.73 (s, 3H), 4.04 (ddd, J ) 10.6, 5.5, 2.2 Hz, 1H),
7.43 (t, J ) 7.5 Hz, 2H), 7.57 (t, J ) 7.7 Hz, 1H), 7.98 (d, J )
8.1 Hz, 2H). 13C NMR (CDCl3): 31.37, 35.33, 52.82, 53.35,
55.13, 65.53, 78.81, 128.52, 129.29, 129.88, 133.55, 165.35,
171.72. [R]D20: -68° (c ) 0.01 acetone). MS (ES+): 295 (2.7),
Ack n ow led gm en t. This work was supported by a
grant from the National Institutes of Health (AI20264
to H.G.F.). M.M. and R.M. thank the Deutsche For-
schungsgemeinschaft for postdoctoral support.
J O981291L