The synthesis of 6,6-difluoroshikimic acid

Article abstract of DOI:10.1016/j.tetlet.2004.03.033

The synthesis of 6,6-difluoroshikimic acid (4) has been achieved in nine steps from the enantiopure diol 9, which is derived from microbial dihydroxylation of iodobenzene. The synthetic strategy has also been demonstrated to be applicable to the preparati

Full text of DOI:10.1016/j.tetlet.2004.03.033

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 3429–3432
The synthesis of 6,6-difluoroshikimic acid
Jane L. Humphreys, David J. Lowes, Karen A. Wesson and Roger C. Whitehead*
Department of Chemistry, The University of Manchester, Oxford Road, Manchester M13 9PL, UK
Received 23February 2004; revised 25 February 2004; accepted 3March 2004
Abstract—The synthesis of 6,6-difluoroshikimic acid (4) has been achieved in nine steps from the enantiopure diol 9, which is derived
from microbial dihydroxylation of iodobenzene. The synthetic strategy has also been demonstrated to be applicable to the prep-
aration of other 6-substituted analogues of shikimic acid.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
The exciting discovery that 2 and 3 possess antiparasitic
properties has stimulated renewed interest in the design
and synthesis of novel inhibitors of enzymes on the
shikimate pathway. In this Letter, we report the syn-
thesis of compound 4, the final member of the series of
6-fluoroshikimic acids, using an approach, which we
have also demonstrated to be amenable to the prepa-
ration of other 6-substituted analogues of shikimic acid.
The fluorinated analogues 2 and 3 of ())-shikimic acid
(1) (Fig. 1) have attracted a great deal of scientific
interest.1–3 Both compounds display in vitro antibacte-
rial activity against a range of E. coli strains with (6S)-6-
fluoroshikimic acid (2) being the more potent agent
(MIC against E. coli K-12 of 0.1 lg/mL compared with
64 lg/mL for (6R)-6-fluoroshikimic acid (3)).4 Both
fluorinated analogues are substrates for the shikimate
transport system of E. coli5 and importantly, the (6S)-
isomer 2 has been shown to be protective against a range
of bacterial intraperitoneal challenges in mice.4 Fol-
lowing the discovery that the shikimic acid pathway is
vital for the survival of parasites of the phylum Api-
complexa,6;7 it has recently been disclosed that both 2
and 3 inhibit the growth of the malaria parasite,
P. falciparum.8 Interestingly, in contrast to the situation
in E. coli, the (6R)-compound 3 was reported to be
significantly more potent than the (6S)-isomer 2 in this
assay.
2. Results and discussion
Recently, we described the synthesis of vinyl bromide 6
in four steps from diol 5 (Scheme 1).9
Oxidation of the allylic hydroxyl in 6 gave the expected
enone which, on treatment with the nucleophilic fluori-
nating agent [bis-(2-methoxyethyl)]-aminosulfurtrifluor-
ide (DeoxoFluorꢁ)10 was converted to the gem-
difluoride 7. Unfortunately, all attempts to introduce a
carboxyl substituent at C1 of 7 using Pd(0) chemistry, as
well as using other trans-metallation protocols, met with
failure.9 It is well documented that aryl and vinyl bro-
mides are less reactive in Pd(0) catalysed C–C bond
forming reactions than the corresponding iodides and
we decided, therefore, to turn our attention to the
preparation of the analogue of 7 bearing an iodine atom
at C1. We thus embarked on the synthesis of 6,6-
difluoroshikimic acid using the enantiopure iodo-diol 9
as starting material (Scheme 2).11 Following the general
procedure of Hudlicky,12 protection of the vicinal diol of
9 as its acetonide followed by face-selective cis-dihydr-
oxylation of the 3,4-double bond gave diol 10. Selective
protection of the allylic hydroxyl of 10 as its benzyl ether
was accomplished in excellent yield via an intermediate
CO H
CO H
2
CO H
2
2
1
2
1
R
F
2
6
5
R
F
OH
3
4
HO
OH
HO
OH
HO
1
1
2
2
OH
OH
R
R
= H
= F
R
R
= F
= H
OH
4
2
3
(-)-shikimic acid
1
Figure 1.
* Corresponding author. Tel.: +44-161-275-7856; fax: +44-161-275-
4939; e-mail: roger.whitehead@man.ac.uk
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.03.033

3430
J. L. Humphreys et al. / Tetrahedron Letters 45 (2004) 3429–3432
Br
Br
Br
CO CH
2
3
F
F
OH
O
O
O
F
F
O
BnO
BnO
O
O
BnO
HO
OCH
OCH
3
3
OCH
3
O
O
OH
H CO
H CO
3
3
H CO
3
5
6
7
8
Scheme 1.
equimolar amounts of the desired compound 17 and an
isomer 16 (Scheme 3). Pleasingly, in the model system, it
was possible to develop mild conditions, which allowed
efficient isomerisation of 16, via the presumed interme-
diacy of a transient allylic carbenium ion 18, to give 17.
I
I
I
O
O
O
1
i), ii)
iii)
OH
OH
2
3
6
5
11
R = H
O
BnO
HO
I
iv)
4
12
R = Ac
OH
10
OR
9
v), vi)
I
Fluorodeoxygenation of a-iodo enone 14 proceeded
smoothly to give an approximately 1:1 mixture of
difluorides 19 and 20 from which the desired gem-di-
fluoride 20 could be isolated in adequate yield (Scheme
4). Unfortunately, we were unable to discover condi-
tions under which vinyl fluoride 19 could be isomerised
to gem-difluoride 20, presumably reflecting the insta-
bility of the highly oxygenated allylic carbenium ion,
which is a necessary intermediate in this interconversion.
We were pleased, however, to find that Pd(0) mediated
carbonylation of 20 proceeded smoothly and in rea-
sonable yield to give the conjugated ester 21.18
O
OH
O
vii)
OCH3
O
BnO
BnO
OCH3
O
O
H3CO
H3CO
14
13
Scheme 2. Reagents and conditions: (i) (CH3O)2C(CH3)2, pTSA,
CH2Cl2; (ii) OsO4 (cat.), NMO, tBuOH, H2O, 94% over two steps; (iii)
Bu2SnO, C6H5CH3, CH3OH, D then BnBr, Bu4NI, C6H5CH3, 13 0ꢂC,
91%; (iv) Ac2O, DMAP, py, CH2Cl2; (v) TFA/H2O (6:1), rt; (vi)
butan-2,3-dione, (CH3O)3CH, CSA, CH3OH, D; (vii) DMSO,
(COCl)2, Et3N, CH2Cl2, )78 ꢂC to rt, 36% over three steps.
With compound 21 in hand, all that remained to be
accomplished was complete deprotection to give the
target material. Several plausible routes were considered
with the main requirement being for conditions that
would allow efficient removal of the benzyl protecting
group without recourse to hydrogenolysis procedures
and that were mild enough to avoid competitive,
destructive aromatisation processes. Ultimately, the
preparation of 4 was achieved in two steps (Scheme 5).
Firstly, the butane diacetal (BDA) group was removed
in quantitative fashion by stirring 21 in a mixture of
TFA and water (6:1) at room temperature. Secondly,
and somewhat surprisingly, removal of the benzyl pro-
tecting group and concomitant ester hydrolysis was
accomplished by heating a solution of the diol 22 in
stannylene acetal.13;14 The regioselectivity of this reac-
tion was confirmed by acetylation of the remaining free
1
hydroxyl of 11 to give 12: comparison of the H NMR
spectra of the two compounds confirmed a significant
downfield shift of the resonance assigned to C(4)H in
compound 12 (dH: ꢀ4.41 for 11, 5.53for 12). trans-
Ketalisation of 11 using the conditions of Ley et al.15
gave an inseparable mixture of butane diacetals, which
was oxidised using Swern conditions16 to give 14 in a
disappointing yield (ꢀ15%) over two steps from 11. A
lengthier three-step procedure was thus developed,
involving initial hydrolysis of 11 to give the corre-
sponding triol, followed by ketalisation to give an
inseparable mixture of diacetals 13 and then oxidation
to provide the desired enone 14 in an acceptable overall
yield of 36%.
I
I
I
I
ii)
+
F
O
F
F
i)
F
+
With enone 14 in hand, we were able to investigate the
key fluorodeoxygenation step necessary for introduction
of geminal fluorines at C6. Previous work in our
research group17 had indicated that fluorodeoxygen-
ation of model compound 15 under standard conditions,
was not only low-yielding but also gave approximately
F
15
16
17
18
Scheme 3. Reagents and conditions: (i) Morph-DAST, C6H6, 80 ꢂC,
24 h, 26% combined yield of 16 and 17; (ii) 4 A mol sieves, CH2Cl2, rt,
8 h, quant.
ꢀ
I
I
I
CO CH
2
3
F
F
O
O
F
F
i)
OCH
ii)
F
F
+
O
O
O
O
BnO
BnO
BnO
BnO
OCH
OCH
OCH
3
3
3
3
O
O
O
H CO
3
H CO
3
H CO
3
H CO
3
14
19
20
21
Scheme 4. Reagents and conditions: (i) DeoxoFluorꢁ, rt, 72 h, 30% of 19, 36% of 20; (ii) Pd(OAc)2, diisopropylethylamine, tri-2-furylphosphine,
CH3OH, CO, DMF, rt, 24 h, 56%.

J. L. Humphreys et al. / Tetrahedron Letters 45 (2004) 3429–3432
3431
been demonstrated by the preparation of a differentially
protected synthetic precursor to the known antimicro-
bial agent (6S)-6-fluoroshikimic acid (2).
CO CH
2
3
CO CH
2
CO H
3
2
F
F
F
i)
3
ii)
F
F
F
O
O
BnO
OH
BnO
OH
HO
OCH
OH
22
OH
H CO
3
21
4
Acknowledgements
Scheme 5. Reagents and conditions: (i) TFA/H2O (6:1), rt, 3h; (ii)
concd HCl/H2O (1:1), 60–70 ꢂC, 12 h then HPLC, 68% over two steps.
We acknowledge, with thanks, the EPSRC for funding
(J.L.H, D.J.L and K.A.W.) and we wish to acknowledge
the EPSRC National Mass Spectrometry Service Centre
in Swansea for the provision of high resolution mass
measurements. We are particularly grateful to Professor
Derek Boyd of the Queens University in Belfast for the
generous donation of diol 9 and also to Rehana Sung
for assistance with HPLC purification of compound 4.
ꢀ6 M HCl at 60–70 ꢂC for 12 h.19a;b;c Analysis by 1H
NMR of the crude product from this sequence indicated
that no aromatisation had occurred and the target
compound 4 was generated in essentially pure form.20
A long-term goal of our research endeavours has been
the development of a ‘divergent’ synthetic strategy,
which will allow the preparation of a variety of
6-substituted analogues of shikimic acid. In this regard
we envisaged the allylic alcohol 13 to be a pivotal
intermediate, however, preparation of a pure sample of
this compound had not been possible from the acetonide
11 (vide supra). We therefore turned our attention to the
‘aromatic Finkelstein reaction’ recently developed by
Buchwald and co-workers, which allows the conversion
of aryl bromides to the corresponding aryl iodides by
the action of a catalytic quantity of CuI, KI and a
diamine additive.21 In the original publication, one
example of a high-yielding halogen exchange of a vinyl
bromide was reported. Encouraged by this, we initiated
an investigation into the halogen exchange of vinyl
bromide 6 with a view to preparing a sample of the vinyl
iodide 13. Pleasingly, reaction of 6 under the conditions
described by Buchwald, resulted in quite clean conver-
sion to the vinyl iodide 13, which was isolated in 72%
yield after purification by flash chromatography
(Scheme 6). Fluorodeoxygenation of 13 using the
nucleophilic fluorinating agent DAST (Et2NSF3)22a;b
proceeded with inversion of configuration to give the
allylic fluoride 23 and subsequent Pd(0) mediated car-
bonylation occurred reasonably smoothly to give the
conjugated ester 24.23 The successful outcome of this
short synthetic sequence demonstrates the potential
versatility of the vinyl iodide 13 as a useful precursor for
the synthesis of other shikimic acid analogues.
References and notes
1. Sutherland, J. K.; Watkins, W. J.; Bailey, J. P.; Chapman,
A. K.; Davies, G. M. J. Chem. Soc., Chem. Commun. 1989,
1386–1387.
2. Duggan, P. J.; Parker, E.; Coggins, J.; Abell, C. Bioorg.
Med. Chem. Lett. 1995, 5, 2347–2352.
3. Song, C.; Jiang, S.; Singh, G. Tetrahedron Lett. 2001, 42,
9069–9071.
4. Davies, G. M.; Barrett-Bee, K. J.; Jude, D. A.; Lehan, M.;
Nichols, W. W.; Pinder, P. E.; Thain, J. L.; Watkins, W.
J.; Wilson, R. G. Antimicrob. Agents Chemother. 1994, 38,
403–406.
5. Jude, D. A.; Ewart, C. D. C.; Thain, J. L.; Davies, G. M.;
Nichols, W. W. Biochim. Biophys. Acta 1996, 1279, 125–
129.
6. Roberts, F.; Roberts, C. W.; Johnson, J. J.; Kyle, D. E.;
Krell, T.; Coggins, J. R.; Coombs, G. H.; Millous, W. K.;
Tzipori, S.; Fergusson, D. J. P.; Chakrabarti, D.; McLeod,
R. Nature 1998, 801–805.
7. Roberts, C. W.; Roberts, F.; Lyons, R. E.; Kirisits, M. J.;
Mui, E. J.; Finnerty, J.; Johnson, J. J.; Ferguson, D. J. P.;
Coggins, J. R.; Krell, T.; Coombs, G. H.; Milhous, W. K.;
Kyle, D. E.; Tzipori, S.; Barnwell, J.; Dame, J. B.; Carlton,
J.; McLeod, R. J. Infectious Diseases 2002, 185, S25–S36.
8. McConkey, G. A. Antimicrob. Agents Chemother. 1999,
43, 175–177.
9. Begum, L.; Box, J. M.; Drew, M. G. B.; Harwood, L. M.;
Humphreys, J. L.; Lowes, D. J.; Morris, G. A.; Redon, P.
M.; Walker, F. M.; Whitehead, R. C. Tetrahedron 2003,
59, 4827–4841.
In summary, we have prepared the novel analogue 4 of
())-shikimic acid in nine steps from the enantiopure diol
9. The allylic alcohol 13, although not directly accessible
in pure form from 9, has been synthesised via applica-
tion of Buchwald’s ‘aromatic Finkelstein reaction’.21
The potential utility of 13 as a diversification point for
the synthesis of other analogues of ())-shikimic acid has
10. Lal, G. S.; Pez, G. P.; Pesaresi, R. J.; Prozonic, F. M.;
Cheng, H. J. Org. Chem. 1999, 64, 7048–7054.
11. The diol 9 was generously donated by Professor Derek
Boyd at The Queens University of Belfast. It is commer-
cially available from the QUESTOR Centre at the Queens
University of Belfast. For further details, contact Profes-
Br
I
I
CO2CH3
OH
OH
F
i)
ii)
iii)
F
O
BnO
O
BnO
O
BnO
OCH3
O
24
OCH3
BnO
OCH3
O
OCH3
O
O
O
H3CO
H3CO
H3CO
6
13
23
H3CO
Scheme 6. Reagents and conditions: (i) CuI, KI, N,N0-dimethylethylenediamine, nBuOH, D, 72%; (ii) Et2NSF3, CH2Cl2, 0 ꢂC to rt, 53% (iii)
Pd(OAc)2, diisopropylethylamine, tri-2-furylphosphine, CH3OH, CO, DMF, rt, 24 h, 58%.

3432
J. L. Humphreys et al. / Tetrahedron Letters 45 (2004) 3429–3432
sor Derek Boyd, The School of Chemistry, The Queens
University of Belfast, David Keir Building, Stranmills
Road, Belfast BT9 5AG, Northern Ireland.
19. (a) For some examples of acid mediated removal of benzyl
protecting groups, see: Baker, W.; Brown, N. C. J. Chem.
Soc. 1948, 2303–2307; (b) Kametani, T.; Yagi, H.; Satoh,
F.; Fukumoto, K. J. Chem. Soc. 1968, 271–275; (c) Marsh,
12. Hudlicky, T.; Price, J. D.; Rulin, F.; Tsunoda, T. J. Am.
Chem. Soc. 1990, 112, 9439–9440.
13. Simas, A. B. C.; Pais, K. C.; da Silva, A. A. T. J. Org.
Chem. 2003, 68, 5426–5428.
14. For a review see: David, S.; Hanessian, S. Tetrahedron
1985, 41, 643–663.
15. For a review see: Ley, S. V.; Baeschlin, D. K.; Dixon, D.
J.; Foster, A. C.; Owen, D. R.; Ince, S. J.; Priepke, H. W.
M.; Reynolds, D. J. Chem. Rev. 2001, 101, 53–80.
16. Mancuso, A. J.; Swern, D. Synthesis 1981, 165–185.
17. Box, J. M.; Harwood, L. M.; Whitehead, R. C. Synlett
J. P.; Goodman, L. J. Org. Chem. 1965, 30, 2491–2492.
25
D
20. Spectroscopic data for compound 4. ½aŠ )128.0 (c 0.10,
H2O); dH (400 MHz; D2O) 3.78 (1H, ddd, J 9.6, 4.0, 1.6,
C(4)H), 3.99 (1H, ꢀdt, J 12.2, 9.6, C(5)H), 4.38 (1H, br t,
J 4.0, C(3)H), 6.91 (1H, dd, J 4.4, 2.0, C(2)H); dF
(282.1 MHz; D2O) )104.48 (1F, br d, J 276.9, one of
C(6)F2), )109.24 (1F, dd, J 276.9, 12.2, one of C(6)F2);
m=z (negative ion electrospray) 209 (60%, [M)H]À), 189
(100); (found 209.0270, C7H7F2O5 ([M)H]À) requires
209.0267).
1997, 571–573.
21. Klapars, A.; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124,
14844–14845.
22. (a) Middleton, W. J. J. Org. Chem. 1975, 40, 574–578;
19
18. Spectroscopic data for compound 21. Mp 131–133 ꢂC; ½aŠ
+2.9 (c 1.36, CH2Cl2); mmax (film)/cmÀ1 1733s (C@O); dDH
(400 MHz; CDCl3) 1.39 and 1.43 (2 · 3H, 2 · s, 2 · butyl
CH3), 3.30 and 3.36 (2 · 3H, 2 · s, 2 · acetal OCH3), 3.82
(3H, s, CO2CH3), 3.96 (1H, dd, J 10.8, 3.8, C(4)H), 4.25
(1H, dd, J 5.8, 3.8, C(3)H), 4.50 (1H, dt, J 13.6, 10.8,
C(5)H), 4.68 (1H, d, J 11.2, one of benzyl CH2), 5.05 (1H,
d, J 11.2, one of benzyl CH2), 7.08 (1H, dd, J 5.8, 2.4,
C(2)H), 7.29–7.40 (3H, m, aromatic m- and p-CH), 7.45
(2H, d, J 7.6, aromatic o-CH); dC (75 MHz; CDCl3) 17.90
and 17.96 (2 · butyl CH3), 48.45 and 48.47 (2 · acetal
OCH3), 52.76 (CO2CH3), 66.87 (dd, J 21.3, 18.4, C(5)H),
67.10 (dd, J 6.0, 1.2, C(4)H), 70.06 (C(3)H), 74.81 (benzyl
CH2), 99.57 and 99.81 (2 · acetal C), 115.95 (t, J 245.1,
C(6)F2), 128.28, 128.59 and 128.72 (aromatic CH), 129.1
(dd, J 26.5, 22.9, C(1)), 138.38 (aromatic ipso-C), 141.3(t,
J 6.6, C(2)H), 163.23 (t, J 1.4, C@O); dF (376 MHz;
CDCl3) )106.45 (1F, ddd, J 278.3, 10.8, 2.4, one of
C(6)F2), )107.40 (1F, dd, J 278.3, 13.6, one of C(6)F2);
m=z (CI/NH3) 446 (MNHþ4 , 100%), 414 (25), 340 (52), 188
(25), 102 (67); (found 446.1987, C21H30F2NO7 (MNHþ4 )
requires 446.1990).
(b) Hudlicky, M. Org. React. (NY) 1988, 35, 513–637.
23. Spectroscopic data for compound 24. Mp 113–115 ꢂC; ½aŠ
19
+34.5 (c 1.08, CHCl3); mmax (film)/cmÀ1 1730s (C@O); dDH
(300 MHz; CDCl3) 1.47 and 1.49 (2 · 3H, 2 · s, 2 · butyl
CH3), 3.38 and 3.47 (2 · 3H, 2 · s, 2 · acetal OCH3), 3.69
(1H, dd, J 11.4, 3.4, C(4)H), 3.89 (3H, s, CO2CH3), 4.20
(1H, dd, J 6.0, 3.4, C(3)H), 4.57 (1H, ddd, J 21.3, 11.4, 7.1,
C(5)H), 4.73(1H, d, J 11.4, one of benzyl CH2), 5.05 (1H,
d, J 11.4, one of benzyl CH2), 5.38 (1H, dd, J 49.2, 7.1,
C(6)H), 6.91 (1H, d, J 6.0, C(2)H), 7.34–7.51 (5H, m,
aromatic CH); dC (75 MHz; CDCl3) 17.95 and 17.99
(2 · butyl CH3), 48.30 and 48.36 (2 · acetal OCH3), 52.52
(CO2CH3), 68.12 (d, J 4.6, C(4)H), 68.32 (d, J 8.1, C(5)H),
70.61 (C(3)H), 74.10 (benzyl CH2), 88.65 (d, J 175.7,
C(6)FH), 99.21 and 99.66 (2 · acetal C), 128.06, 128.47
and 128.63(aromatic CH), 131.95 (d, J 19.2, C(1)),
137.26 (d, J 5.4, C(2)H), 138.71 (aromatic ipso-C), 165.30
(C@O); dF (376 MHz; CDCl3) )183.86 (dd, J 49.2, 21.3,
C(6)HF); m=z (positive ion electrospray) 433 (100%,
[M + Na]þ).