An improved procedure for the synthesis of enaminones - Dimer building blocks in β-strand mimetics

Article abstract of DOI:10.1055/s-2005-872094

@-Tides have been shown to have the same characteristics as a peptide in the β-strand conformation and to have the ability to self-associate into dimeric β-sheets. Aza-cyclohexaenaminones, obtained by condensation of a protected azacyclohexa-3,5-dione and amino acid esters, are the key building-blocks in the synthesis of @-tides. An improved three-step synthetic sequence to these enaminones has been developed that takes advantage of microwave-assisted chemistry in two of the steps to enhance the reaction rates. It was also found that the enaminone building blocks can be obtained by direct condensation of the aza-cyclohexa-3,5-dione with amino acid esters, without prior activation of the diketone. Multivariate design was used to optimize this microwave-assisted condensation, resulting in a short reaction time (300 s) and high yields (67-94%). Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2005-872094

2590
PAPER
An Improved Procedure for the Synthesis of Enaminones – Dimer Building
Blocks in b-Strand Mimetics
An
Improved
P
roc
n
e
dure for the
d
Synthesis of
r
E
namin
e
ones as Larsson,a Sara Spjut,a Jan Kihlberg,*a,b Fredrik Almqvist*a
a
Department of Chemistry, Organic Chemistry, Umeå University, 901 87 Umeå, Sweden
Medicinal Chemistry, AstraZeneca R&D Mölndal, 431 83 Mölndal, Sweden
b
Fax +46(90)138885; E-mail: jan.kihlberg@chem.umu.se; E-mail: fredrik.almqvist@chem.umu.se
Received 11 January 2005; revised 11 April 2005
chaperones and pilus subunits can inhibit key protein–
protein interactions required for pilus assembly. Further-
more, a library of designed peptides was constructed and
used to elucidate the structure–activity relationships for
Abstract: @-Tides have been shown to have the same characteris-
tics as a peptide in the b-strand conformation and to have the ability
to self-associate into dimeric b-sheets. Aza-cyclohexaenaminones,
obtained by condensation of a protected azacyclohexa-3,5-dione
and amino acid esters, are the key building-blocks in the synthesis this interaction.3 Peptides have poor bioavailability as a
of @-tides. An improved three-step synthetic sequence to these
consequence of low absorption and a high propensity to
enaminones has been developed that takes advantage of micro-
undergo degradation by peptidases. Therefore, we wanted
wave-assisted chemistry in two of the steps to enhance the reaction
to construct b-strand mimetics, which would ideally both
rates. It was also found that the enaminone building blocks can be
improve bioavailability and biological affect, as com-
obtained by direct condensation of the aza-cyclohexa-3,5-dione
pared to the pilus-derived peptides. In our design we
thought about using alternating aza-cyclohexenones in
between ordinary amino acid residues, in an attempt to ri-
gidify the mimetic in an elongated b-strand conformation
and to eradicate all peptide bonds (Figure1). To our de-
light, Bartlett et al. have prepared such pentamer peptido-
mimetics, so called @-tides, consisting of amino acids
with 1,2-dihydro-3(6H)-pyridones at alternate positions.4a
These b-strand mimetics were shown to have the same
characteristics as a peptide in a b-strand conformation and
to have the ability to self-associate into dimeric b-sheets.
Here we present an improved microwave-assisted syn-
thetic sequence to the desired @-tide building blocks,
which also enable diversification desirable for library syn-
thesis.
with amino acid esters, without prior activation of the diketone.
Multivariate design was used to optimize this microwave-assisted
condensation, resulting in a short reaction time (300 s) and high
yields (67–94%).
Key words: enaminones, b-strand mimetics, microwave-assisted
synthesis, multivariate design, optimization
b-Strands and b-sheets are secondary structural elements
involved in many protein–protein and protein–DNA inter-
actions of importance in biological systems, and in the
processes associated with disease as well as normal func-
tion.1 One such process is the molecular machinery of pi-
lus assembly in uropathogenic Escherichia coli (UPEC),
where b-sheet formation is crucial both during chaperone-
mediated folding and in the transport of pilus subunits in
the periplasm, as well as in the formation of mature pili on
the bacterial cell-surface.2 In an effort to develop inihibi-
tors of pilus assembly we have shown that peptides corre-
sponding to the b-strand motifs from periplasmic
The previous synthesis of @-tide building blocks 1 started
from commercially available 3,5-dichloropyridine 2,
which consisted of some time-consuming steps that pro-
ceeded in modest yields (Scheme 1).4 The method also
employed expensive and toxic reagents [e.g. Yb(OTf)3,
O
O
R
R
R
O
N
N
H
N
H
N
H
R'
O
O
O
O
N
OR''
N
H
O
R
O
R
O
O
O
H
N
H
N
H
N
N
H
N
H
R
O
R
O
R
Figure 1 A scaffold with alternating aza-cyclohexenones is anticipated to adopt an extended b-strand conformation and preserve some of the
hydrogen bonding interactions required for b-sheet formation. Enaminones 1 are key building blocks in the synthesis of such b-strand mimetics.
SYNTHESIS 2005, No. 15, pp 2590–2596
x
x.
x
x
.
2
0
0
5
Advanced online publication: 22.07.2005
DOI: 10.1055/s-2005-872094; Art ID: Z01105SS
© Georg Thieme Verlag Stuttgart · New York

PAPER
An Improved Procedure for the Synthesis of Enaminones
2591
OMe
Cl
OMe
1 N HCl
a
b
O
N
N
N
OMe
Cl
OMe
2
3
O
O
O
R1
R1
c
d
OR2
+
O
N
OR2
H2N
O
N
N
H
O
O
O
O
O
4
6–14
H3C
O
O
CH3
OH
e
f
O
N
O
N
OSO2Mes
N
H
O
O
O
5
15
Scheme 1 General scheme for synthesis of @-tide building blocks 5–13.
Reaction conditions used by Phillips et al.: a) NaOMe (8 equiv), DMF, 80 °C, 38 h; (68%) b) NaBH4, allyl chloroformate, MeCN, –45 °C, 1.5
h; (80%) e) MesSO2Cl, K2CO3, CH2Cl2, r.t., 4 h; (69%) f) H-L-Leu-O-t-Bu, H-L-Phe-O-t-Bu, H-L-Ile-O-t-Bu or H-L-Thr(t-Bu)O-t-Bu,
Yb(OTf)3 or Sn(OTf)2, i-Pr2NEt, CH2Cl2, r.t., 24 h; (73–81%) d) TFA, r.t., 2 h; (100%)
Reaction conditions used in this paper: a) NaOMe (6 equiv), DMF, microwave, 230 °C, 20 min; (60%) b) NaBH4, allyl chloroformate, MeCN,
–45 °C, 1.5 h; (80%) c) Me- or t-Bu-esters of amino acids (Table 2), Et3N, NMP–C6H6 (90:10), microwave, 140 °C, 5 min; (48–94%) d) 0.1 M
LiOH (aq), THF, r.t., 1.5 h; (100%).16
Sn(OTf)2] for the conjugate-addition–elimination step made us consider combining benzene with a solvent hav-
(Scheme 1, step f), which is inappropriate for pharmaceu- ing a higher polarizability in order to assist microwave
tical applications. The published method does therefore heating. Moreover, such a co-solvent would also increase
not appear to be completely satisfactory when making the solubility of the reactants and hence enable synthesis
larger libraries of compounds, or in industrial applica- of more diverse enaminones. Consequently, a small inves-
tions. In an effort to develop an efficient and fast route to tigation of suitable co-solvents was performed and enami-
building blocks 6–14, which have diverse structures, our none 6 was synthesized using a microwave temperature of
attention was first directed to the use of microwave-assist- 110 ºC, a reaction time of six minutes in benzene contain-
ed synthesis to enhance the rate of the methoxylation of ing either 4% DMSO, DMF, or N-methylpyrrolidine
3,5-dichloropyridine (2). In addition, it appeared that the (NMP). The amount of enaminone 6 formed in the pres-
synthetic sequence could be shortened by one step ence of the different co-solvents was compared using LC-
through direct condensation of alloc-azacyclohexadione 4 MS and showed that up to twice as much product was
with an amino acid to give the corresponding enaminone formed when NMP was employed as the co-solvent com-
without prior activation of diketone 4 (cf. 5, Scheme 1).
pared to DMSO or DMF.
It was found that the methoxylation of 3,5-dichloropyri- Initial attempts towards microwave-assisted synthesis of
dine (2) progressed smoothly in DMF and that all starting enaminone 6 directly from diketone 4 resulted in low
material was consumed at a microwave temperature of yields, which indicated the decomposition of enaminone
230 °C within a reaction time of 20 minutes. The isolated 6 and/or diketone 4 at these elevated temperatures and
yield was comparable with that from the published proce- prolonged reaction times. Therefore, different reaction
dure (60% vs. 68%), which was conducted at 80 °C for 38 variables were optimized using multivariate experimental
hours with 8 equivalents of sodium methoxide.4b Thereaf- design in an effort to increase the yield.6 Multivariate ex-
ter, alloc-protected diketone 4 was prepared from 3,5- perimental design is a tool that can maximize the knowl-
dimethoxypyridine 3 as described earlier starting by re- edge of the reaction conditions with as few experiments as
duction of the pyridine ring and in situ alloc-protection of possible. In addition, a properly performed design also
the resulting secondary amine. Subsequent hydrolysis of gives information about interaction effects between the
the enol ethers resulted in alloc-protected diketone 4.4a different variables, which can not be obtained by chang-
Diketone 4 is known to be somewhat labile and decom- ing the variables one by one. Thus, a full factorial design6a
poses over time,4a but we found it to be stable for exten- with three different variables, temperature, reaction time,
sive periods of time when it was frozen in a benzene and the amount of diketone 4 was set up. Three central
matrix (36.5 mg/mL).
points were also included so that the design would allow
detection of curvature in the resulting model, giving a to-
tal of eleven experiments (Figure 2 and Table 1).
Benzene is often used as a solvent in the synthesis of
enamines,5 however, the low polarizability of benzene
Synthesis 2005, No. 15, 2590–2596 © Thieme Stuttgart · New York

2592
A. Larsson et al.
PAPER
Table 1 Experimental Design for the Synthesis of 6
Entry
Temperature Time (s)
(°C)
Diketone
(equiv)
Yield (%)
1
2
100
140
100
140
100
140
100
140
120
120
120
300
300
900
900
300
300
900
900
600
600
600
1.05
1.05
1.05
1.05
1.75
1.75
1.75
1.75
1.4
21
100
44
69
9
3
4
5
6
66
48
65
81
92
76
7
8
Figure 2 Experimental design for optimization of enaminone syn-
thesis. A full factorial design using three variables – temperature, re-
action time, and amount of diketone 4 – resulting in a total of eleven
experiments including three central points.
9
10
11
1.4
1.4
The yield for each experiment was calculated using LC-
MS and a calibration curve was constructed using differ-
ent concentrations of pure enaminone 6 corresponding to
yields ranging from 10–90%. Calculated yields from the
eleven experiments were then correlated with the different
variables using multiple linear regression (MLR). This
gave a model with five coefficients which explained 93%
of the variation in the response (R2Y = 0.93) and had good
predictive properties according to cross-validation with a
Q2 of 0.60. The regression coefficients (Figure 3, a)
showed that the temperature was positively correlated
with the yield while the reaction time and the amount of
diketone did not significantly affect the yield.
servation that the diketone and/or the product decom-
posed when the reaction times were too long or elevated
temperatures were used. Furthermore, a negatively corre-
lated square term of the temperature indicated that the op-
timal temperature actually is within the investigated
interval. A plot of temperature against time, holding the
diketone variable constant at the lowest setting and having
the yield as response, predicted that optimal conditions for
the reaction could be achieved with a temperature of 140
°C and a reaction time of five minutes (Figure 3, b). Thus,
a diverse set of eight amino acids were selected and the
corresponding enaminones were synthesized (Table 2, en-
tries 1–8) in order to test the optimal conditions predicted
by the model.
Moreover, a negatively correlated cross-term between the
temperature and time was in good agreement with the ob-
Figure 3 Multiple linear regression (MLR) model of the reaction conditions used for synthesis of enaminone 6. a) Regression coefficients
showing the investigated variables and their correlation to the yield of 6. b) Contour plot of the predicted yield (cf. boxes) as a function of tem-
perature and reaction time using 1.05 equivalents of diketone 4 and the methyl ester of leucine.
Synthesis 2005, No. 15, 2590–2596 © Thieme Stuttgart · New York

PAPER
An Improved Procedure for the Synthesis of Enaminones
2593
Table 2 Enaminone Synthesis Using Optimized Conditions
O
O
R1
Cl–
140 C, 5 min
°
R1
OR2
+
AllocN
OR2
H3N
1 equiv TEA
AllocN
N
H
10% cosolvent in C6H6
O
O
O
4
6–14
Entry
1
Compound
R1
R2
Cosolvent
Diketone (equiv)
1.05
Yield (%)
88
CH3
CH3
6
Me
NMP
2
3
7
8
Me
Me
NMP
NMP
1.05
1.05
69
90
NHBoc
Ot-Bu
O
4
9
t-Bu
NMP
1.05
94
N
H
5
6
10
11
Proline
t-Bu
NMP
NMP
1.05
1.05
68
67
Me
H3C
CH3
7
8
12
13
Me
DMF
DMF
1.05
1.05
48
52
OH
t-Bu
H3C
H3C
H3C
OH
9
14
11
Me
Me
NMP
NMP
1.05
73
Ot-Bu
CH3
10
1.05 + 0.5a
100
a An additional portion of diketone 4 was added and the microwave irradiation procedure was repeated.
Use of the optimized reaction conditions gave isolated properties to NMP, which prevents the use of simple flash
yields in the range 67–94% of @-tide building blocks for chromatography for purification. In order to circumvent
six of the amino acids used (Table 2, entries 1–6). Further- this problem 12 and 13 were re-synthesized, using water-
more, the reaction could be forced to completion by the extractable DMF as a cosolvent instead of NMP, but only
addition of more diketone 4 (0.5 equiv) and repeating the moderate yields (48% and 52%, respectively) were ob-
reaction once more (cf. enaminone 11, Table 2, entries 6 tained (Table 2, entries 7,8). Fortunately, this setback can
and 10). The synthesis of enaminone 10 from proline be avoided by using protecting groups such as t-Bu or Boc
shows that the method can be extended to secondary on amino acids with polar side-chains (Table 2, entries
amines. However, proline has limited use as a building 2,3). Use of t-Bu-protected threonine for the synthesis of
block in b-strand mimetics due to its tendency to initiate enaminone 14 also raised the isolated yield to 73% com-
turn formation in peptides. In addition, enaminone 10 was pared to 52% when the side chain of threonine was left un-
formed in a 1:1 cis/trans ratio (the amide bond), according protected (Table 2, entries 8, 9).
to NMR spectroscopy, which may be a complicating fac-
tor in some peptidomimetics.
Most commercially available polar amino acids carry acid
labile protecting groups, such as Boc and t-Bu, on their
If the side chains of the amino acids are too polar, e.g. side-chains. To allow orthogonal deprotection, the base
serine and threonine, the amount of NMP has to be raised labile methyl group was used for protection of the a-car-
to 10% in order for the reaction to proceed. Although the boxylic acid in the amino acids used to prepare most of the
reaction appeared to proceed well according to LC-MS, compounds 6–14. However, for hydrolyzing methyl es-
the isolated yields were low after purification. This could ters the method has to be mild enough so as not to cause
be explained by the unfortunate fact that the resulting epimerization of the a-carbon of the amino acid. Use of a
enaminones 12 and 13 have similar chromatographic 0.1 M aqueous solution of LiOH in THF is known to sup-
Synthesis 2005, No. 15, 2590–2596 © Thieme Stuttgart · New York

2594
A. Larsson et al.
PAPER
press epimerization.7 Enaminone 6 was hydrolyzed to
enaminone 15 (Scheme 1) in quantitative yield using
these conditions. To investigate if any racemization had
occurred, the reaction mixture was analyzed by chiral
HPLC [(S,S) Whelk-O1]. A single peak was obtained, in-
dicating that hydrolysis proceeded without racemization.
The synthesis was then repeated starting from racemic
leucine. In this case the two enantiomers were detected
with baseline separation confirming that synthesis of 15
had been achieved without racemization.
Alloc-azacyclohexenone (4)
Prepared according to published procedures.4a
Enaminones 6–14; General Procedure
Amino acid esters (0.2 mmol) and Et3N (28 mL, 0.2 mmol) were
added to the alloc-protected diketone 4 (41.4 mg, 0.21 mmol) in
NMP–benzene (10:90, 1.6 mL) and the solution was heated by mi-
crowave irradiation in a monomode instrument at 140 °C for 300 s.
A sat. soln of NH4Cl (1.5 mL) was added and the mixture was ex-
tracted with EtOAc (3 × 5 mL). The combined organic layers were
washed with brine (5 mL), dried over MgSO4, and concentrated. Pu-
rification by flash column chromatography gave enaminones 6–14
as yellow oils.
A new three-step synthetic sequence to @-tide building
blocks has been developed. Microwave-assisted chemis-
try was used in two of the steps to enhance the reaction
rates. Multivariate design was used to optimize the key
step, i.e. synthesis of enaminones without prior activation
of the intermediate aza-cyclohexa-3,5-dione, which is
condensed with amino acid esters. Overall, this gave a fast
and efficient synthetic method that allows the use of struc-
turally diverse amino acids, for instance in the synthesis of
@-tide libraries. Work is now in progress to synthesize
@-tides that mimic the G1 b-strand of the type 1 pilus
chaperone FimC. The ability of such @-tides to interfere
with protein–protein interactions required for pilus as-
sembly in uropathogenic E. coli will be investigated.
Alloc-Ach-Leu-OMe (6)
Purification by column flash chromatography (EtOAc–heptane,
70:30) gave 6 as a yellow oil. Yield: 56.8 mg (88%); [a]D –36 (c 1.0,
CHCl3).
IR: 3253, 3072, 2956, 1743, 1704, 1563, 1440, 1226 cm–1.
1H NMR (400 MHz, CDCl3): d = 6.63 (d, J = 7.96 Hz, 1 H), 5.73–
5.86 (m, 1 H), 5.02–5.23 (m, 3 H), 4.49 (d, J = 5.5 Hz, 2 H), 4.25 (s,
2 H), 3.90–4.00 (m, 3 H), 3.62 (s, 3 H), 1.58 (d, J = 6.0 Hz, 3 H),
0.81 (dd, J = 13.45, 5.49 Hz, 7 H).
13C NMR (100 MHz, CDCl3): d = 191.2, 172.6, 161.3, 154.9, 132.2,
118.0, 95.1, 66.7, 53.8, 52.5, 50.6, 44.0, 40.8, 24.8, 22.6, 21.9.
HRMS (FAB): m/z calcd for C16H25N2O5 [M + H+], 325.1763;
found, 325.1754.
All reactions were carried out under an inert atmosphere with anhy-
drous solvents under anhydrous conditions, unless otherwise stated.
TLC was performed on Silica Gel 60 F254 (Merck) with detection
by UV light and staining with anisaldehyde or KMnO4. Flash col-
umn chromatography was performed on silica gel (Matrex, 60 Å,
35–70 mm, Grace Amicon). Microwave-assisted synthesis was car-
ried out on a SmithCreator microwave device using intended sep-
tum-sealed reaction vials. The 1H and 13C NMR spectra were
recorded on a Bruker DRX-400 in CDCl3 [residual CHCl3 (dH 7.26
ppm) or CDCl3 (dC 77.0 ppm) as internal standard] at 298 K. First
order chemical shifts and coupling constants were obtained from
one-dimensional spectra. Optical rotations were measured with a
Perkin-Elmer 343 polarimeter at 20 °C. IR spectra were recorded on
ATI Mattson Infinity Series FTIRTM spectrometer. Mass spectrom-
etry was performed on a Waters micromass ZQ. High-resolution
mass spectra [FAB+] were recorded on a JEOL JMS-SX 102 spec-
trometer.
Alloc-Ach-Lys(Boc)-OMe (7)
Purification by flash column chromatography (EtOAc–heptane,
90:10) gave 7 as a yellow oil. Yield: 60.6 mg (69%); [a]D 3.7 (c 1.0,
CHCl3).
IR: 3278, 3070, 2933, 1702, 1554, 1464, 1234, 928, 730 cm–1.
1H NMR (400 MHz, CDCl3): d = 5.65–5.98 (m, 2 H), 5.18–5.33 (m,
2 H), 5.11 (s, 1 H), 4.66 (br s, 1 H), 4.60 (d, J = 4.6 Hz, 2 H), 4.30
(s, 2 H), 3.98–4.11 (m, 3 H), 3.75 (s, 3 H), 3.09 (d, J = 6.3 Hz, 2 H),
1.76–1.92 (m, 2 H), 1.20–1.52 (m, 13 H).
13C NMR (100 MHz, CDCl3): d = 191.1, 171.9, 156.2, 154.8, 132.2,
118.0, 95.8, 79.3, 66.6, 55.0, 53.4, 52.7, 50.6, 44.1, 39.6, 31.0, 29.8,
28.4, 22.2.
HRMS (FAB): m/z calcd for C21H34N3O7 [M + H+], 440.2397;
found, 440.2395.
Alloc-Ach-Asp(O-t-Bu)-OMe (8)
Purification by flash column chromatography (EtOAc–heptane,
90:10) gave 8 as a yellow oil.
The experiments were planned using full-factorial design6a and
were done in randomized order. Sample analysis was carried out by
diluting the reaction mixture (20 mL) with H2O–MeCN (1.98 mL,
95:5) before injection onto a LC-MS column. Integrated peak areas
were quantified and the yield was determined using a calibration
curve. The results were analyzed by MLR using MODDE 6.0.8
Yield: 68.6 mg (90%); [a]D 42 (c 1.0, CHCl3).
IR: 3266, 3062, 2979, 1702, 1550, 1438, 1367, 1226, 1153 cm–1.
1H NMR (400 MHz, CDCl3): d = 5.85–5.98 (m, 1 H), 5.74 (br s, 1
H), 5.19–5.35 (m, 2 H), 5.18 (s, 1 H), 4.62 (d, J = 5.3 Hz, 2 H),
4.23–4.33 (m, 3 H), 4.01–4.14 (m, 2 H), 3.78 (s, 3 H), 2.87 (dd,
J = 16.7, 4.9 Hz, 1 H), 2.76 (dd, J = 16.8, 5.2 Hz, 1 H), 1.43 (s, 9 H).
13C NMR (100 MHz, CDCl3): d = 191.1, 170.4, 169.0, 159.5, 154.8,
132.3, 118.0, 95.9, 82.2, 66.6, 52.9, 51.5, 50.6, 44.1, 36.7, 27.9.
HRMS (FAB): m/z calcd for C18H27N2O7 [M + H+], 383.1818;
found, 383.1819.
3,5-Dimethoxypyridine (3)
3,5-Dichloropyridine (2; 0.74 g, 5 mmol) and NaOMe (0.81 g, 15
mmol) were added to freshly distilled DMF (18 mL) and the mix-
ture was heated by microwave irradiation in a monomode instru-
ment at 230 °C for 10 min. To the cooled reaction mixture NaOMe
(0.81 g, 15 mmol) was added and the reaction was heated by micro-
wave irradiation for an additional 10 min at 230 °C. The reaction
was quenched with Et2O (40 mL) and H2O (40 mL). The aqueous
layer was extracted with Et2O (3 × 20 mL) and the combined organ-
ic layers were washed with brine (3 × 20 mL), dried over MgSO4,
and concentrated. Purification by flash column chromatography
(EtOAc–heptane, 50:50) gave 3 as a clear oil (0.42 g, 60%). Struc-
tural data are in agreement with previously published data.4b
Alloc-Ach-Trp-O-t-Bu (9)
Purification by flash column chromatography (EtOAc–heptane,
80:20) gave 9 as a yellow oil.
Yield: 82.8 mg (94%); [a]D –74 (c 1.0, CHCl3).
IR: 3301, 2979, 1691, 1556, 1438, 1232, 1160, 750 cm–1.
Synthesis 2005, No. 15, 2590–2596 © Thieme Stuttgart · New York

PAPER
An Improved Procedure for the Synthesis of Enaminones
2595
1H NMR (400 MHz, CDCl3): d = 8.17 (br s, 1 H), 7.50 (d, J = 7.6
Hz, 1 H), 7.36 (d, J = 8.1 Hz, 1 H), 7.20 (t, J = 14.1, 7.0 Hz, 1 H),
7.12 (t, J = 14.7, 7.1 Hz, 1 H), 7.00 (d, J = 3.0 Hz, 1 H), 5.85–5.96
(m, 1 H), 5.48 (br s, 1 H), 5.16–5.33 (m, 3 H), 4.60 (d, J = 5.7 Hz,
2 H), 4.26–4.33 (m, 1 H), 4.08–4.16 (m, 2 H), 4.05 (s, 2 H), 3.38
(dd, J = 14.8, 5.0 Hz, 1 H), 3.25 (dd, J = 14.6, 5.4 Hz, 1 H), 1.40 (s,
9 H).
13C NMR (100 MHz, CDCl3): d = 191.1, 170.1, 160.1, 154.7, 136.0,
132.2, 129.3, 127.5, 123.0, 122.1, 119.6, 118.3, 117.9, 111.4, 109.2,
95.6, 83.0, 66.6, 56.2, 50.5, 44.1, 27.8, 27.0.
Alloc-Ach-Thr-O-t-Bu (13)
The solvent used for the reaction was DMF–benzene (10:90). Puri-
fication by flash column chromatography (EtOAc–MeOH, 95:5)
gave 13 as a yellow oil.
Yield: 36.9 mg (52%); [a]D –51 (c 1.0, CHCl3).
IR: 3299, 3073, 2977, 1695, 1549, 1444, 1239, 1108 cm–1.
1H NMR (400 MHz, CDCl3): d = 6.15 (d, J = 7.4 Hz, 1 H), 5.83–
5.96 (m, 1 H), 5.18–5.33 (m, 3 H), 4.60 (d, J = 5.1 Hz, 2 H), 4.24–
4.43 (m, 3 H), 3.97–4.13 (m, 2 H), 3.87 (dd, J = 8.2, 3.0 Hz, 1 H),
1.46 (s, 9 H), 1.24 (d, J = 6.5 Hz, 3 H).
HRMS (FAB): m/z calcd for C24H30N3O5 [M + H+], 440.2185;
found, 440.2184.
13C NMR (100 MHz, CDCl3): d = 191.6, 169.2, 161.8, 154.9, 132.2,
118.1, 95.6, 83.4, 67.7, 66.8, 61.0, 50.5, 44.2, 27.9, 20.8, 20.0.
HRMS (FAB): m/z calcd for C17H27N2O6 [M + H+], 355.1869;
found, 355.1860.
Alloc-Ach-Pro-O-t-Bu (10)
Purification by column flash chromatography (EtOAc) gave 10 as a
yellow oil.
Alloc-Ach-Thr(O-t-Bu)-OMe (14)
Purification by flash column chromatography (EtOAc–MeOH,
80:20) gave pure 14 as a yellow oil.
Yield: 47.5 mg (68%); [a]D –96 (c 1.0, CHCl3).
IR: 2977, 2879, 1702, 1635, 1569, 1446, 1226, 1147 cm–1.
1H NMR (400 MHz, CDCl3): d = 5.83–5.97 (m, 1 H), 5.16–5.34 (m,
3 H), 5.01 (s, 1 H), 4.60 (s, 2 H), 3.87–4.50 (m, 6 H), 3.47–3.72 (m,
1 H), 3.25–3.47 (m, 2 H), 2.14–2.29 (m, 2 H), 1.92–2.14 (m, 3 H),
1.45 (d, J = 7.1 Hz, 10 H).
13C NMR (100 MHz, CDCl3): d = 190.4, 189.9, 170.4, 170.3, 159.7,
155.0, 132.3, 117.9, 97.8, 97.4, 83.0, 82.6, 66.6, 61.5, 61.1, 50.3,
50.0, 48.8, 48.3, 43.3, 43.1, 30.7, 29.9, 27.9, 23.5, 22.4.
Yield: 53.8 mg (73%); [a]D –12 (c 1.0, CHCl3).
IR: 3257, 3020, 2929, 1700, 1597, 1519, 1439, 1216 cm–1.
1H NMR (400 MHz, CDCl3): d = 5.83–6.01 (m, 1 H), 5.43 (d, J =
7.4 Hz, 1 H), 5.05–5.36 (m, 3 H), 4.62 (d, J = 5.1 Hz, 2 H), 4.33 (s,
2 H), 4.23 (q, J = 5.7 Hz, 1 H), 3.98–4.13 (m, 2 H), 3.83 (dd, J = 8.2,
3.0 Hz, 1 H), 3.71 (s, 3 H), 1.22 (d, J = 6.5 Hz, 3 H), 1.11 (s, 9 H).
13C NMR (100 MHz, CDCl3): d = 191.2, 170.2, 161.0, 154.8, 132.3,
118.0, 95.9, 74.7, 67.2, 66.7, 60.6, 52.5, 50.5, 44.4, 28.2, 21.6.
HRMS (FAB): m/z calcd for C18H29N2O6 [M + H+], 369.2026;
found, 369.2029.
HRMS (FAB): m/z calcd for C18H27N2O5 [M + H+], 351.1920;
found, 351.1914.
Alloc-Ach-Ile-OMe (11)
Purification by flash column chromatography (EtOAc–heptane,
70:30) gave 11 as a yellow oil.
Alloc-Ach-Leu (15)
The ester protected enaminone 6 (39.8 mg, 0.12 mmol) was dis-
solved in THF (12 mL) and LiOH (0.1 M aq; 1.4 mL) was added.
After stirring at r.t. for 1.5 h the reaction mixture was concentrated.
The remaining solution was co-concentrated from toluene. The res-
idue was suspended in MeOH (4 mL) and Amberlite IR-120 before
the solid phase was filtered off and the solution was concentrated to
yield pure 15 as a yellow oil; yield: 38.1 mg (100%). Structural data
are in agreement with previously published data.4a
Yield: 43.5 mg (67%); [a]D –2.2 (c 1.0, CHCl3).
IR: 3251, 3062, 2962, 1702, 1549, 1438, 1230 cm–1.
1H NMR (400 MHz, CDCl3): d = 5.85–5.99 (m, 1 H), 5.41 (d,
J = 6.6 Hz, 1 H), 5.19–5.35 (m, 2 H), 5.16 (s, 1 H), 4.62 (d, J = 5.0
Hz, 2 H), 4.18–4.38 (m, 2 H), 3.94–4.15 (m, 3 H), 3.77 (s, 3 H),
1.84–1.96 (m, 1 H), 1.46–1.59 (m, 1 H), 1.21–1.35 (m, 1 H), 0.95 (t,
J = 14.8, 7.3 Hz, 3 H), 0.90 (d, J = 6.7 Hz, 3 H).
13C NMR (100 MHz, CDCl3): d = 191.2, 171.4, 154.5, 132.3, 118.1,
96.1, 66.7, 59.4, 53.5, 52.6, 50.8, 44.4, 37.6, 25.9, 15.0, 11.7.
HRMS (FAB): m/z calcd for C16H25N2O5 [M + H+], 325.1763;
Acknowledgment
We thank Einar Nilsson at Lund University for HRMS analysis.
This work was funded by grants from the Knut and Alice Wallen-
berg Foundation and the Swedish Research Council and the Göran
Gustafsson Foundation for Research in Natural Science and Medi-
cine.
found, 325.1765.
Alloc-Ach-Ser-OMe (12)
The solvent used for the reaction was DMF–benzene (10:90). Puri-
fication by flash column chromatography (EtOAc–MeOH, 90:10)
gave 12 as a yellow oil.
Yield: 28.7 mg (48%); [a]D –3.7 (c 1.0, CHCl3).
References
IR: 2940, 2901, 1688, 1503, 1299, 922 cm–1.
(1) (a) Prusiner, S. B.; Scott, M. R.; DeArmond, S. J.; Cohen, F.
E. Cell 1998, 93, 337. (b) Fitzgerald, P. M. D.; McKeever,
B. M.; Vanmiddlesworth, J. F.; Springer, J. P.; Heimbach, J.
C.; Leu, C. T.; Herber, W. K.; Dixon, R. A. F.; Darke, P. L.
J. Biol. Chem. 1990, 265, 14209. (c) Clark, K. L.; Halay, E.
D.; Lai, E. S.; Burley, S. K. Nature (London) 1993, 364,
412. (d) Brown, J. H.; Jardetzky, T. S.; Gorga, J. C.; Stern,
L. J.; Urban, R. G.; Strominger, J. L.; Wiley, D. C. Nature
(London) 1993, 364, 33. (e) Acharya, R.; Fry, E.; Stuart, D.;
Fox, G.; Rowlands, D.; Brown, F. Nature (London) 1989,
337, 709.
1H NMR (400 MHz, CDCl3): d = 6.07 (br s, 1 H), 5.84–5.99 (m, 1
H), 5.19–5.35 (m, 2 H), 5.15 (s, 1 H), 4.16 (d, J = 4.7 Hz, 2 H),
4.26–4.38 (m, 2 H), 4.11–4.18 (m, 1 H), 4.03–4.10 (m, 2 H), 3.94–
4.03 (m, 2 H), 3.81 (s, 3 H), 3.38 (br s, 1 H).
13C NMR (100 MHz, CDCl3): d = 191.5, 170.2, 160.8, 155.0, 132.2,
118.2, 95.7, 66.8, 61.6, 57.2, 53.0, 50.6, 44.2, 38.0.
HRMS (FAB): m/z calcd for C13H19N2O6 [M + H+], 299.1243;
found, 299.1251.
Synthesis 2005, No. 15, 2590–2596 © Thieme Stuttgart · New York

2596
A. Larsson et al.
PAPER
(5) Alt, G. H.; Cervinka, O.; Cook, A. G.; Haynes, L. W.;
(2) (a) Sauer, F. G.; Futterer, K.; Pinkner, J. S.; Dodson, K. W.;
Hultgren, S. J.; Waksman, G. Science (Washington, D.C.)
1999, 285, 1058. (b) Choudhury, D.; Thompson, A.;
Stojanoff, V.; Langermann, S.; Pinkner, J.; Hultgren, S. J.;
Knight, S. D. Science (Washington, D.C.) 1999, 285, 1061.
(3) Larsson, A.; Johansson, S.; Pinkner, J. S.; Hultgren, S. J.;
Almqvist, F.; Kihlberg, J.; Linusson, A. J. Med. Chem. 2005,
48, 935.
Kuehne, M. E.; Malhotra, S. K.; Paukstelis, J. V.; Stamhuis,
E. J. Enamines: Synthesis structure and reactions; Marcel
Dekker Inc.: New York, 1969.
(6) (a) Box, G. E. P.; Hunter, W. G.; Hunter, J. S. Statistics for
experimenters. An introduction to design data analysis and
model building; John Wiley & Sons Inc.: New York, 1978.
(b) Andersson, P. M.; Lundstedt, T.; Abramo, L. J.
Chemometr. 1996, 10, 379.
(7) (a) Mogemark, M.; Elofsson, M.; Kihlberg, J. Org. Lett.
2001, 3, 1463. (b) Emtenas, H.; Ahlin, K.; Pinkner, J. S.;
Hultgren, S. J.; Almqvist, F. J. Comb. Chem. 2002, 4, 630.
(8) Modde 6.0, Umetrics AB Box 7960, 907 19 Umeå, Sweden.
(4) (a) Phillips, S. T.; Rezac, M.; Abel, U.; Kossenjans, M.;
Bartlett, P. A. J. Am. Chem. Soc. 2002, 124, 58.
(b) Testaferri, L.; Tiecco, M.; Tingoli, M.; Bartoli, D.;
Massoli, A. Tetrahedron 1985, 41, 1373. (c) Phillips, S. T.;
Piersanti, G.; Rüth, M.; Gubernator, N.; Lengerich, B.;
Bartlett, P. A. Org. Lett. 2004, 6, 4483.
Synthesis 2005, No. 15, 2590–2596 © Thieme Stuttgart · New York