A Baylis-Hillman/ozonolysis route towards (±) 4,5-dihydroxy-2,3- pentanedione (DPD) and analogues

Article abstract of DOI:10.1016/j.tetlet.2005.07.102

The Baylis-Hillman reaction between 2-(tert-butyldimethylsilyloxy)ethanal and 3-buten-2-one followed by desilylation gave rise to the corresponding α-methylene-β,γ-dihydroxy ketone further converted by reductive ozonolysis of the carbon-carbon double bond into racemic 4,5-dihydroxy-2,3- pentanedione (DPD), a significant molecule in bacterial cell-cell communication systems. The same sequence applied to other substrates allowed the preparation of chain elongated analogues and 5-O-acylated derivatives of DPD.

Full text of DOI:10.1016/j.tetlet.2005.07.102

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 6495–6498
A Baylis–Hillman/ozonolysis route towards
± ± 4,5-dihydroxy-2,3-pentanedione (DPD± and analogues
(
Marine Frezza, Laurent Soul e` re, Yves Queneau and Alain Doutheau*
Laboratoire de Chimie Organique, UMR CNRS-UCBL 5181, Institut National des Sciences Appliqu e´ es,
0 avenue A. Einstein, 69621 Villeurbanne, France
2
Received 7 June 2005; revised 19 July 2005; accepted 20 July 2005
Available online 9 August 2005
Abstract—The Baylis–Hillman reaction between 2-(tert-butyldimethylsilyloxy)ethanal and 3-buten-2-one followed by desilylation
gave rise to the corresponding a-methylene-b,c-dihydroxy ketone further converted by reductive ozonolysis of the carbon–carbon
double bond into racemic 4,5-dihydroxy-2,3-pentanedione (DPD), a significant molecule in bacterial cell–cell communication sys-
tems. The same sequence applied to other substrates allowed the preparation of chain elongated analogues and 5-O-acylated deriv-
atives of DPD.
Ó 2005 Elsevier Ltd. All rights reserved.
Many bacterial species are able to sense their own pop-
ulation size through a cell–cell communication system,
named quorum sensing (QS), based on the production
and exchange of diffusible extracellular signaling mole-
new access to DPD and analogues. Despite its very sim-
ple structure, the synthesis of DPD is not an easy task
owing, in particular, to its instability at high concentra-
tions. At the time we initiated this work, only two syn-
theses of the (S) enantiomer had been reported in the
8
1
cules called autoinducers. Since a link between this pro-
2
cess and virulence factor production or biofilms
literature, both involving the oxidation of a triple bond
3
9,10
formation has been established for a number of patho-
to generate the a-diketone function.
We imagined an
genic bacteria, the design of QS inhibitors has recently
attracted considerable interest for the development of
anti-infection therapies. Among the various autoinduc-
ers which have been identified so far, a borate, known as
autoinducer-2 (AI-2), formed from the metabolic prod-
uct (S) 4,5-dihydroxy-2,3-pentanedione (DPD) (Scheme
alternative approach in which the a-diketo-b-hydroxy
moiety in DPD would be generated by the reductive
ozonolysis of the carbon–carbon double bond in a-
methylene-b-hydroxy ketones of type 4 resulting from
a Baylis–Hillman reaction.
4
1
), is produced by a large number of both Gram-nega-
Thus, using dimethyl sulfide as the reducing agent, the
byproducts formed during the last step will be formalde-
hyde and dimethylsulfoxide. The latter compounds
being reputed non-toxic, at least at low concentrations,
the purification of DPD or analogues would become
unnecessary before biological assays. In addition,
asymmetric Baylis–Hillman reaction being well docu-
5
tive and Gram-positive bacteria.
As part of a programme aimed at obtaining QS inhibi-
tors,
6
,7
we recently became interested in designing a
HOHO
O
OH O
O
OH
OH
OH O
1
1
B
mented, the preparation of non-racemic compounds
could be considered. The very recent report of Vander-
HO
O
O
1
2
leyden and co-workers describing a synthesis of (S)-
DPD by reductive ozonolysis of (S)-1,2-dihydroxy-4-
methyl-4-penten-3-one (Scheme 2) prompts us to report
our preliminary results.
AI-2
Scheme 1.
DPD
4
1
3
When carried out in THF at 0 °C, the Baylis–Hillman
reaction between 2-(tert-butyldimethylsilyloxy) ethanal
1a and 3-buten-2-one 2a (4 equiv) in the presence of
Keywords: Quorum sensing; AI-2; Baylis–Hillman reaction;
Ozonolysis.
1
4
*
Corresponding author. Tel.: +33 (0)4 72 43 82 21; fax: +33 (0)4 72 43
8 96; e-mail: alain.doutheau@insa-lyon.fr
8
DABCO (0.25 equiv) was completed in 21 h and
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.07.102

6496
M. Frezza et al. / Tetrahedron Letters 46 (2005) 6495–6498
1
9
OH
(ꢀ 65:35) of anti/syn diastereoisomers, which were
incompletely separated by column chromatography.
The deprotection step was carried out on each pure ste-
HO
ref 12
(
S)-DPD
O
2
0
reoisomer to give the corresponding anti or syn diol 4c,
which were submitted to ozonolysis and led to expected
Scheme 2.
2
1
dicarbonyl compounds 5c. As seen above in the case of
a, when 5b and 5c were reacted with excess 1,2-phenyl-
ene-diamine, they were quantitatively transformed into
5
afforded the expected enone 3a in 74% yield (Scheme 3
and Table 1). Removal of the silyl protecting group
led to the corresponding diol 4a which was subjected
to ozonolysis. When conducted on a diluted (ꢀ10 mM)
2
2
the corresponding quinoxaline 8b and 8c. Conversely,
when carried out from a-dimethyl substituted aldehydes
1c and 2a, the Baylis–Hillman reaction did not take
2
3
1
5
24
CD OD solution, the reaction gave rise to 5a (DPD).
place.
3
As already reported,9,10,12 the H NMR spectrum of
DPD is rather complicated due to the presence of three
isomers in equilibrium: the open diketone, and its two
1
The method was also successfully applied for the prepa-
ration of derivatives bearing 5-OH enzymolabile mask-
ing groups, which could be used as in vivo precursors
of DPD. Thus, the Baylis–Hillman reaction between
1
6
cyclic anomers 6 and 7, in about a 0.5:1:1 ratio. In
order to get structure verification, this mixture was
submitted to the action of 1,2-phenylene-diamine
and we observed that it was totally con-
verted into the quinoxaline 8a (Scheme 4).
1
3
25
acetoxyacetaldehyde 9a or pivaloxyacetaldehyde 9b
and 2a led to the corresponding adducts 10a and 10b
9
,10,12
26
(
3 equiv)
in very good yields. Further, ozonolysis gave the dike-
1
7,18
27
tonic compounds 11a and 11b (Scheme 5).
We then extended this approach to the preparation of
some chain elongated analogues of DPD by varying
the nature of both the vinyl ketone and the silyloxyalde-
hyde. A similar sequence was carried out from aldehyde
In conclusion, we developed a short reaction sequence
leading to racemic DPD. The sequence was successfully
applied for the preparation of elongated chain ana-
logues or 5-O-acylated derivatives. We are presently
exploring the possibilities of our method to obtain more
elaborated analogues of DPD as well as non-racemic
compounds.
1
a and 4-penten-3-one 2b (Scheme 3 and Table 1) to give
diketone 5b. The Baylis–Hillman reaction of a-methyl
substituted aldehyde 1b with 2a gave 3c as a mixture
O
O
OH
O
OH
O
R1
R1
R2
RO
a
c
R1
R3
R3
R2
TBDMSO
R3
5a-c
+
R2
HO
O
1
1
1
a R1 = R2 = H
b R1 = H, R2 = CH3
c R1 = R2 = CH3
2a R3 = CH3
2b R3 = C H5
3a-c R = TBDMS
4a-c R = H
b
2
a R = R2 = H, R3 = CH3
1
b R1 = R2 = H, R3 = C H5
2
TBDMS : tert-butyldimethylsilyl
c R1 = H, R2 = R3 = CH3
Scheme 3. Synthetic sequence to compounds 5. Reagents and conditions: (a) THF, DABCO (0.25 equiv), 0 °C. (b) TBAF (1 equiv), THF, rt. (c) O
MeOH, À78 °C, then DMS, À78 °C to rt.
3
,
Table 1.
Aldehyde
Vinyl-ketone
Baylis–Hillman adduct
Reaction time (h)
Yield (%)
Deprotected product
Yield (%)
1a
1a
1b
2a
2b
2a
3a
3b
3c
21
30
24
74
80
70
4a
4b
78
90
71
78
4c anti
4c syn
(
anti/syn 65:35)
O
O
R2
R1
N
N
HO
H4
H5
OH
H1
O
CH3
OH
HO
H7
H8
OH
H2
H3
CH3
O
CH3
HO
O
OH
O
H6
OH
H9
5
a (DPD)
6 (rac)
7 (rac)
8a R1 = H, R2 = CH3
8
8
b R1 = H, R2 = C H5
c R1 = R2 = CH3
2
Scheme 4. Structures of DPD, its two cyclic anomers 6 and 7, and quinoxalines 8.

M. Frezza et al. / Tetrahedron Letters 46 (2005) 6495–6498
OH
OH
6497
O
O
O
a
b
+
2a
RCOO
RCOO
RCOO
O
9
9
a
b
10a (87%)
10b (89%)
11a
11b
a R = CH3
b R = tert-Bu
Scheme 5. Synthetic sequence to compounds 11. Reagents and conditions: (a) THF, DABCO (0.25 equiv), 0 °C, 20 h. (b) O
DMS, À78 °C to rt.
3
, MeOH, À78 °C, then
Acknowledgements
(1 mL) was added and the resulting solution was concen-
trated under vacuum to a residual volume of about 1 mL.
1
1
6. H NMR (500 MHz, D
2
O). 5a (20%) + 6 (40%) + 7 (40%).
Financial support from MENESR and CNRS is grate-
fully acknowledged. M.F. thanks the MENESR for a
scholarship.
d 4.35 (dd, J = 6.9, J = 5.7 Hz, H-4), 4.16 (m, 2H, H-5,
H-8), 4.03 (dd, J = 6.0, J = 3.5 Hz, H-7), 3.95 (dd, J = 7.4,
J = 3.5 Hz, H-1), 3.80 (m, 2H, H-3, H-9), 3.63 (dd,
J = 11.9, J = 7.4 Hz, H-2) 3.56 (dd, J = 9.4, J = 5.7 Hz,
H-6), 2.35 (s, CH ), 1.42 (s, CH ), 1.39 (s, CH ).
3
3
3
References and notes
Assignments are based on published data and 2D-COSY
experiments.
1
1
2
. For a recent review, see: Lyon, G. J.; Muir, T. W. Chem.
Biol. 2003, 10, 1007–1021.
. Finch, R. G.; Pritchard, D. I.; Bycroft, B. W.; Williams,
P.; Stewart, G. S. J. Antimicrob. Chemother. 1998, 42, 569–
17. H NMR (300 MHz, D
2
O). 8a d 8.04 (m, 1H), 7.92 (m,
1H), 7.80 (m, 2H), 5.30 (dd, J = 6.7, J = 4.5 Hz, 1H), 3.99
(dd, J = 12, J = 4.5 Hz 1H), 3.90 (dd, J = 12, J = 6.7 Hz,
1H), 2.77 (s, 3H) in agreement with published data.
1
5
71.
18. Minor signals were observed in the H NMR spectra of
3
. Rice, S. A.; McDougald, D.; Kumar, N.; Kjelleberg, S.
Curr. Opin. Investig. Drugs 2005, 6, 178–184.
. Raffa, R. B.; Iannnuzzo, J. R.; Levine, D. R.; Saeid, K. K.;
Schwartz, R. S. C.; Sucic, N. T.; Terleckyj, O. D.;
Young, J. M. J. Pharmacol. Exp. Ther. 2005, 312, 417–
DPD (three singulets in the 2.0–2.2 ppm area and a
multiplet at 3.95 ppm) and of 8a (multiplet at 7.49 and
7.70 ppm).
4
19. The anti stereochemistry was temporarily assigned to the
major isomer by analogy with the results reported in the
literature for Baylis–Hillman reactions between a-alkoxy-
aldehydes and vinyl ketones or esters: Drewes, S. E.;
Manickum, T.; Roos, G. H. P. Synth. Commun. 1988, 18,
1065–1070.
4
23.
5
6
7
. Chen, X.; Schauder, S.; Potier, N.; Van Dorsselaer, A.;
Pelczer, I.; Bassler, B. L.; Hughson, F. M. Nature 2002,
4
15, 545–549.
. Reverchon, S.; Chantegrel, B.; Deshayes, C.; Doutheau,
20. In these cases, compounds 4c are in equilibrium with small
amounts (25% for 4c anti, 5% for 4c syn) of isomeric cyclic
A.; Cotte-Pattat, N. Bioorg. Med. Chem. Lett. 2002, 12,
1
153–1157.
anomers.
1
. Castang, S.; Chantegrel, B.; Deshayes, C.; Dolmazon, R.;
Gouet, P.; Haser, R.; Reverchon, S.; Nasser, W.; Hug-
ouvieux-Cotte-Pattat, N.; Doutheau, A. Bioorg. Med.
Chem. Lett. 2004, 14, 5145–5149.
21. H NMR (500 MHz, D
2
O). 5b d 4.38 (dd, J = 7.3,
J = 6.0 Hz, 1H), 4.16 (dd, J = 10.1, J = 6.0 Hz, 1H),
4.15 (dd, J = 9.5, J = 7.3 Hz, 1H), 4.02 (dd, J = 6.0,
J = 3.2 Hz, 1H), 3.94 (dd, J = 7.3, J = 4.0 Hz, 1H), 3.79
(dd, J = 10.1, J = 3.2 Hz, 1H), 3.77 (dd, J = 12.0,
J = 4.0 Hz, 1H), 3.62 (dd, J = 12.0, J = 7.3 Hz, 1H),
3.53 (dd, J = 9.5, J = 6.0 Hz, 1H) 2.77 (m, 2H), 1.75 (m,
4H), 1.03 (t, J = 7.1 Hz, 3H), 0.98 (t, J = 7.5 Hz, 3H), 0.97
8
. Hoffman, T.; Schieberle, P. J. Agric. Food Chem. 1998, 46,
2
35–241.
9
. Meijler, M. M.; Hom, L. G.; Kaufmann, G. F.; McKen-
zye, K. M.; Sun, C.; Moss, J. A.; Matsushita, M.; Janda,
K. D. Angew. Chem., Int. Ed. 2004, 43, 2106–2108.
2
(t, J = 7.5 Hz, 3H). (300 MHz, D O) 5c anti d 3.97 (m,
1
1
1
0. Semmelhack, M. F.; Campagna, S. R.; Federle, M. J.;
Bassler, B. L. Org. Lett. 2005, 7, 569–572.
1. Matsui, K.; Takizawa, S.; Sasai, H. Tetrahedron Lett.
1H), 3.93 (d, J = 7.9 Hz, 1H), 3.76 (m, 1H), 3.56 (d,
J = 6.0 Hz, 1H), 1.40 (s, 3H), 1.39 (s, 3H), 1.31 (d,
J = 6.4 Hz, 3H), 1.29 (d, J = 6.4 Hz, 3H). 5c syn d 4.37 (m,
1H), 4.30 (m, 1H), 4.05 (d, J = 5.3 Hz, 1H), 3.86 (d,
J = 4.5 Hz, 1H), 1.41 (s, 3H), 1.36 (s, 3H), 1.24 (d,
2
005, 46, 1943–1946, and references cited therein.
2. De Keersmaecker, S. C. J.; Varszegi, C.; van Boxel, N.;
Habel, L. W.; Metzger, K.; Daniels, R.; Marchal, K.; De
Vos, D.; Vanderleyden, J. J. Biol. Chem. 2005, 280, 19563–
J = 6.8 Hz, 3H), 1.15 (d, J = 6.8 Hz, 3H).
1
22. H NMR (300 MHz, D
2
O). 8b d 8.06 (m, 1H), 7.98 (m,
1
9568.
3. Brzezinski, L. J.; Rafel, S.; Leahy, J. W. Tetrahedron 1997,
3, 16423–16434.
1H), 7.83 (m, 2H), 5.33 (dd, J = 6.8, J = 4.9 Hz, 1H), 3.99
(dd, J = 11.7, J = 4.9 Hz, 1H), 3.93 (dd, J = 11.7,
J = 6.8 Hz, 1H), 3.11 (m, 2H), 1.36 (t, J = 7.5 Hz, 3H).
8c anti d 8.10 (m, 1H), 7.98 (m, 1H), 7.83 (m, 2H), 5.06 (d,
J = 6.8 Hz, 1H), 4.26 (m, 1H), 2.81 (s, 3H), 1.33 (d,
J = 6.4 Hz, 3H). 8c syn d 8.05 (m, 1H), 7.95 (m, 1H), 7.83
(m, 2H), 5.04 (d, J = 5.7 Hz, 1H), 4.32 (m, 1H), 2.78 (s,
3H), 1.19 (d, J = 6.4 Hz, 3H).
1
1
5
4. Nicolaou, K. C.; Liu, J.-J.; Yang, Z.; Ueno, H.; Sorensen,
E. J.; Claiborne, C. F.; Guy, R. K.; Hwang, C.-K.;
Nakada, M.; Nantermet, P. G. J. Am. Chem. Soc. 1995,
1
17, 634–644.
1
5. A solution of 4a (8 mg, 62 lmol) in CD OD (6 mL) was
3
cooled to À78 °C and ozone was bubbled through the
solution. The reaction mixture first turned light yellow and
after about 30 min, became light green. At this time, the
solution was purged with oxygen until the green colour
23. The aldehydes 1b and 1c were prepared by the reductive
ozonolysis of known tert-butyldimethylsilyl protected
3-buten-2-ol and 2-methyl-3-buten-2-ol, respectively.
24. The starting aldehyde 1c was recovered unchanged after
30 h of stirring at 0 °C. This failure was probably due to
the sterically hindered carbonyl function of the neopen-
tylic type aldehyde 1c.
disappeared and Me S (45 lL, 10 equiv) was added. The
2
mixture was then allowed to warm at rt and stirred for
1
2
6 h. To 1 mL of the colourless reaction mixture, D O

6498
M. Frezza et al. / Tetrahedron Letters 46 (2005) 6495–6498
2
5. Audouard, C.; Fawcett, J.; Griffiths, G. A.; Percy, J. M.;
Pintat, S.; Smith, C. A. Org. Biomol. Chem. 2004, 528–
41.
3
NMR (300 MHz, CDCl ) d 6.24 (s, 1H), 6.18 (d,
J = 1.1 Hz, 1H), 4.75 (m, 1H), 4.24 (dd, J = 3.4,
5
11.3 Hz, 1H), 4.13 (dd, J = 6.8, 11.3 Hz, 1H), 3.00 (d,
1
13
13
2
6. Compounds 4a–c and 10a,b gave H, C NMR and
J = 5.7 Hz, 1H), 2.35 (s, 3H), 2.08 (s, 3H). C NMR
HRMS data in agreement with the proposed structures.
Data for 4b and 10a are given as examples. 4b H NMR
(75 MHz, CDCl ) d 199.64, 171.42, 146.60, 127.61, 69.33,
67.61, 26.21, 20.86. HRMS-CI MH calcd for C H O :
8 13 4
3
1
+
(
1
300 MHz, CD OD) d 6.29 (s, 1H), 6.14 (d, J = 1.1 Hz,
H), 4.63 (m, 1H), 3.60 (dd, J = 3.8, 11.3 Hz, 1H), 3.36
173.0736; found: 173.0817.
27. 11a H NMR (300 MHz, D
3
1
2
O) d 4.32 (m, 1H), 4.13 (m,
2H), 2.38 (s, 3H), 2.09 (s, 3H); HRMS-CI MH1 calcd for
C H O : 175.0606; found: 175.0607. 11b H NMR
+
(
dd, J = 6.8, 11.3 Hz, 1H), 2.78 (m, 2H), 1.06 (t,
1
3
J = 7.2 Hz, 3H); C NMR (75 MHz, CD OD) d 203.95,
1
3
7
11
5
+
49.60, 125.86, 71.31, 66.93, 32.16, 8.56; HRMS-CI MH
(300 MHz, D
2
O) d 4.33 (m, 1H), 4.12 (m, 2H), 2.37 (s,
calcd for C : 145.0865; found: 145.0866. 10a H
7
13 3
H O
3H), 1.19 (s, 9H).