Palladium-catalyzed α-arylation of tetramic acids

Article abstract of DOI:10.1021/jo900799y

(Chemical Equation Presented) A mild, racemization-free, palladium-catalyzed α-arylation of tetramic acids (2,4-pyrrolidinediones) has been developed. Various amino acid-derived tetramic acids were cleanly arylated by treatment with 2 mol % of Pd(OAc)

Full text of DOI:10.1021/jo900799y

Palladium-Catalyzed r-Arylation of Tetramic Acids
Morten Storgaard,†,‡ Florencio Zaragoza Do¨rwald,‡,§ Bernd Peschke,‡ and David Tanner*,†
Technical UniVersity of Denmark, Department of Chemistry, 201 KemitorVet, DK-2800 Kgs. Lyngby,
Denmark, and NoVo Nordisk A/S, Biopharm. Chemistry, DK-2760 MåløV, Denmark
dt@kemi.dtu.dk
ReceiVed April 17, 2009
A mild, racemization-free, palladium-catalyzed R-arylation of tetramic acids (2,4-pyrrolidinediones) has
been developed. Various amino acid-derived tetramic acids were cleanly arylated by treatment with 2
mol % of Pd(OAc)2, 4 mol % of a sterically demanding biaryl phosphine, 2.3 equiv of K2CO3 or K3PO4,
and aryl chlorides, bromides, or triflates in THF. With conventional heating, conversions >95% could
be attained after 1 h at 80 °C, whereas microwave-induced heating led to much shorter reaction times (5
min at 110 °C). The electron density of the aryl electrophile had no effect on their reactivity: both electron-
rich and electron-poor aryl chlorides and bromides or triflates led to good yields. Ortho-substituted aryl
halides and heteroaryl halides, however, did not undergo the title reaction.
Introduction
Tetramic acids are ꢀ-keto-γ-lactams which are slightly acidic
(pKa ≈ 6.4).1,2 Depending on solvent, concentration, and
temperature, tetramic acids can exist as both an enol (4-hydroxy-
3-pyrrolin-2-one) and a keto tautomer (2,4-pyrrolidindione) (see
Figure 1).1,3,4
The structural unit of tetramic acids has been known for more
than 100 years,5 and it is found in many biologically active
natural products,3 typically either as 3-acyl or 4-O-alkyl
derivatives, examples being althiomycin,6a,b dolastatin 15,6c,d
and epicoccamide.6e Tetramic acids are important intermediates
FIGURE 1. Tetramic acids can exist as both an enol and a keto
tautomer.1,3,4
in the synthesis of statins,7a,b ꢀ-hydroxy γ-amino acids,7c and
lactams7d which are inhibitors of renin. Renin is involved in
the renin-angiotensin system (blood pressure and fluid regulat-
ing system in the body), hypertension, congestive heart failure,
and development of HIV. Furthermore, tetramic acid derivatives
have been reported as key intermediates for the synthesis of
analogues of penicillins and cephalosporins7e and 4-substituted
3-hydroxy-1H-pyrrole-2,5-dione derivatives7f which are inhibi-
tors of glycolic acid oxidase and thus potentially useful drugs
for the treatment of calcium oxalate renal lithiasis (kidney
stones) and primary hyperoxalurias, which is an inborn error
of metabolism resulting in increased urinary excretion of oxalate.
2-Ethyl-4,6-dimethylphenyl-substituted tetramic acid derivatives
have been described in the patent literature as novel pesticides
† Technical University of Denmark.
‡ Novo Nordisk A/S.
§ Current address: Lonza AG, Rottenstrasse 6, CH-3930 Visp, Switzerland.
(1) Mulholland, T. P. C.; Foster, R.; Haydock, D. B. J. Chem. Soc., Perkin
Trans. 1 1972, 2121–2128.
(2) Mulholland, T. P. C.; Foster, R.; Haydock, D. B. J. Chem. Soc., Perkin
Trans. 1 1972, 1225–1231.
(3) (a) Royles, B. J. L. Chem. ReV. 1995, 95, 1981–2001. (b) Schobert, R.;
Schlenk, A. Bioorg. Med. Chem. 2008, 16, 4203–4221.
(4) Steyn, P. S.; Wessels, P. L. Tetrahedron Lett. 1978, 47, 4707–4710.
(5) (a) Benary, E. Chem. Ber. 1907, 40, 1079–1083. (b) Benary, E. Chem.
Ber. 1911, 44, 1759–1765.
5032 J. Org. Chem. 2009, 74, 5032–5040
10.1021/jo900799y CCC: $40.75  2009 American Chemical Society
Published on Web 05/27/2009

Palladium Catalyzed R-Arylation of Tetramic Acids
FIGURE 2. Retrosynthetic analysis of amino benzyl ketones.
SCHEME 1. General Synthesis of N-Boc-Protected Amino
SCHEME 2. Proposed r-Arylation of Tetramic Acids with
an Aryl Halide (X ) Cl, Br, I)
Acid-Derived Tetramic Acids8,9
and herbicides.7g Recently, methods for incorporation of amino
acid-derived tetramic acids into peptides have been developed,8
giving rise to more stable tripeptides. Tetramic acids derived
from amino acids are easily synthesized in good yield from
commercially available N-Boc amino acids and Meldrum’s acid
(2,2-dimethyl-1,3-dioxane-4,6-dione) via DCC (N,N′-dicyclo-
hexylcarbodiimide) or EDC (1-ethyl-3-(3-dimethylaminopropyl)
carbodiimide) activation (see Scheme 1).8,9
Only a few examples in the literature have described the
3-aryl tetramic acids,10a,b the most important being the use of
3-phenyl 5-olefinic tetramic acids as novel glycine site N-
methyl-D-aspartate receptor antagonists, for the treatment of
neurological diseases.10c The development of a solid-phase
synthesis of substituted 3-aryl tetramic acids has also been
described.10d However, none of these methods utilize the easy
synthesis of N-Boc protected tetramic acids described above,
and the methods are not general, requiring the use of strong
base (e.g., KHMDS or NaOEt) and many synthetic steps. None
of these methods make it easy to efficiently synthesize a broad
range of 3-aryl-substituted amino acid-derived tetramic acids
as potentially interesting biologically active compounds. We
therefore wished to develop a useful method for the synthesis
of 3-aryl tetramic acids from the readily available N-Boc amino
acid-derived tetramic acids.
Specifically, we want to use these compounds as building
blocks for C-terminal modified peptides toward the preparation
of peptidyl enzyme inhibitors. For example, ring-opening of the
3-aryl tetramic acids followed by decarboxylation would lead
to a new type of amino benzyl ketones (see Figure 2), which
subsequently can be coupled to the C-terminal of a peptide.
Traditionally, R-arylated ketones or carboxylic acid deriva-
tives have been synthesized by nucleophilic aromatic substitution
reactions (SNAr) of aryls substituted with electron-withdrawing
groups by reaction with stabilized enolates11 or via copper-
catalyzed enolate reaction with 2-bromobenzoic acid.12 These
methods all have drawbacks and are not very general. Usually,
they require harsh reaction conditions, which are not suitable
for protected, enantiomerically pure amino acid derivatives.
Using a palladium-catalyzed R-arylation would be much more
efficient since these reactions are typically more general, mild,
and broad in substrate scope.
The literature reports a number of palladium-catalyzed
R-arylation conditions for different substrates containing elec-
tron-withdrawing groups such as ketones,13a aldehydes,13b
malonates,13c cyanoesters,13c sulfones,13d esters,13e amides,13f
protected amino acids,13g piperidinones,13h and nitriles.13i Only
a few examples of R-arylation of 1,3-dicarbonyls have been
described, and most of them are nonchiral and synthetically
undemanding compounds. Most of the examples have been
reported by Buchwald and co-workers,13a using substrates such
as diethyl malonate, 1,3-cyclohexanedione, and 1,3-cyclopen-
tanedione. Very recently more functionalized substrates have
been subjected to palladium-catalyzed arylation, e.g. the sp2
arylation of azine N-oxides,14 R-arylation of highly function-
(6) (a) Yamaguchi, H.; Nakayama, Y.; Takeda, K.; Tawara, K. J. Antibiot.,
Ser. A 1957, 10, 195–200. (b) Fujimoto, H.; Kinoshita, T.; Suzuki, H.; Umezawa,
H. J. Antibiot. 1970, 23, 271–275. (c) Pettit, G. R.; Kamano, Y.; Dufresne, C.;
Cerny, R. L.; Herald, C. L.; Schmidt, J. M. J. Org. Chem. 1989, 54, 6005–
6006. (d) Pettit, G. R.; Thornton, T. J.; Mullaney, J. T.; Boyd, M. R.; Herald,
D. L.; Singh, S.-B.; Flahive, E. J. Tetrahedron 1994, 50, 12097–12108. (e)
Wright, A. D.; Osterhage, C.; Ko¨nig, G. M. Org. Biomol. Chem. 2003, 1, 507–
510.
(7) (a) Jouin, P.; Catro, B.; Nisato, D. J. Chem. Soc., Perkin Trans. 1 1987,
1177–1182. (b) Schmidt, U.; Riedl, B.; Haas, G.; Griesser, H.; Vetter, A.;
Weinbrenner, S. Synthesis 1993, 216–220. (c) Fehrentz, J.-A.; Bourdel, E.;
Califano, J.-C.; Chaloin, O.; Devin, C.; Garrouste, P.; Lima-Leite, A.-C.; Llinares,
M.; Rieunier, F.; Vizavonna, J.; Winternitz, F.; Loffet, A.; Martinez, J.
Tetrahedron Lett. 1994, 35, 1557–1560. (d) Wittenberger, S. J.; Baker, W. R.;
Donner, B. G.; Hutchins, C. W. Tetrahedron Lett. 1991, 32, 7655–7658. (e)
Hlubucek, J. R.; Lowe, G. J. Chem. Soc., Chem. Commun. 1974, 419–420. (f)
Rooney, C. S.; Randall, W. C.; Streeter, K. B.; Ziegler, C.; Cragoe, E. J., Jr.;
Schwam, H.; Michelson, S. R.; Williams, H. W. R.; Eichler, E.; Duggan, D. E.;
Ulm, E. H.; Noll, R. M. J. Med. Chem. 1983, 26, 700–714. (g) Fischer, R.;
Lehr, S.; Feucht, D.; Losel, P.; Malsam, O.; Bojack, G.; Auler, T.; Hills, M. J.;
Kehne, H.; Rosinger, C. H. United States Patent Application Publication, US2007/
0225167, September 27, 2007.
(10) (a) Larsen, S.; Bernstein, J. J. Am. Chem. Soc. 1950, 72, 4447–4452.
(b) Andrews, M. D.; Brewster, A.; Moloney, M. G. Tetrahedron: Asymmetry
1994, 5, 1477–1478. (c) Mawer, I. M.; Kulagowski, J. J.; Leeson, P. D.;
Grimwood, S.; Marshall, G. R. Bioorg. Med. Chem. Lett. 1995, 5, 2643–2648.
(d) Matthews, J.; Rivero, R. A. J. Org. Chem. 1998, 63, 4808–4810.
(11) Heckmann, J. Ann. Chem. 1883, 220, 128–146.
(12) Bruggink, A.; McKillop, A. Tetrahedron 1975, 31, 2607–2619.
(13) (a) Fox, J. M.; Huang, X.; Chieffi, A.; Buchwald, S. L. J. Am. Chem.
Soc. 2000, 122, 1360–1370. (b) Terao, Y.; Fukuoka, Y.; Satoh, T.; Miura, M.;
Nomura, M. Tetrahedron Lett. 2002, 43, 101–104. (c) Beare, N. A.; Hartwig,
J. F. J. Org. Chem. 2002, 67, 541–555. (d) Mitin, A. V.; Kashin, A. N.;
Beletskaya, Russ. J. Org. Chem. 2004, 40, 802–812. (e) Lloyd-Jones, G. C.
Angew. Chem., Int. Ed. 2002, 41, 953–956. (f) Hama, T.; Liu, X.; Culkin, D. A.;
Hartwig, J. F. J. Am. Chem. Soc. 2003, 125, 11176–11177. (g) Lee, S.; Beare,
N. A.; Hartwig, J. F. J. Am. Chem. Soc. 2001, 123, 8410–8411. (h) Filippis, A.;
de Pardo, D. G.; Cossy, J. Tetrahedron 2004, 60, 9757–9767. (i) Culkin, D. A.;
Hartwig, J. F. Acc. Chem. Res. 2003, 36, 234–245.
(8) (a) Hosseini, M.; Kringelum, H.; Murray, A.; Tønder, J. E. Org. Lett.
2006, 8, 2103–2106. (b) Hosseini, M.; Grau, J. S.; Sørensen, K. K.; Søtofte, I.;
Tanner, D.; Murray, A.; Tønder, J. E. Org. Biomol. Chem. 2007, 5, 2207–2210.
(c) Hosseini, M.; Tanner, D.; Murray, A.; Tønder, J. E. Org. Biomol. Chem.
2007, 5, 3486–3494.
(9) (a) Ma, D.; Ma, J.; Ding, W.; Dai, L. Tetrahedron: Asymmetry 1996, 7,
2365–2370. (b) Courcambeck, J.; Bihel, F.; Michelis, C. D.; Que´le´ver, G.; Kraus,
J. L. J. Chem. Soc., Perkin Trans. 1 2001, 1421–1430.
(14) Schipper, D. J.; Campeau, L.-C.; Fagnou, K. Tetrahedron 2009, 65,
3155–3164.
J. Org. Chem. Vol. 74, No. 14, 2009 5033

Storgaard et al.
FIGURE 3. Ligands used in the initial screening experiments.
TABLE 1. Results from Initial Screening of Bases and Ligands
TABLE 2. Increased Temperature and Prolonged Reaction Time
entry
ligand
base
conv (%)a
1
2
3
4
5
4-8b
all fourb
Cs2CO3
K3PO4
Na2CO3
K2CO3
<5
15
36
-
36
entry
base
temp (°C)
time (h)
conv (%)a
9
9
9
9
1
2
3
4
5
6
7
8
Cs2CO3
K3PO4
Na2CO3
K2CO3
Cs2CO3
K3PO4
80
80
80
72
72
72
72
16
16
16
16
21
33
<5
32
25
71
<5
52
c
80
Determined by H NMR. b Ligands 4-8 tested with Cs2CO3, K3PO4,
a
1
100
100
100
100
Na2CO3, and K2CO3, respectively, in 20 experiments. c No significant
1
Na2CO3
K2CO3
product formation determined by H NMR or LC-MS.
a
1
Determined by H NMR.
alized cyclohexanones,15 and asymmetric intramolecular R-ary-
lation of aldehydes.16 However, to the best of our knowledge,
tetramic acids have never before been subjected to this kind of
transformation (see Scheme 2), and we set out to determine
suitable reaction conditions.
step procedure with addition of base at reduced temperature
and then the actual arylation at elevated temperature.13h Buch-
wald and co-workers13a reported the first use of a weak inorganic
base, K3PO4, in palladium-catalyzed R-arylations of ketones.
Finally, a very important parameter is the choice of ligand. This
can be difficult, since many different ligands have been reported
to work in R-arylation of carbonyl substrates, with a broad
variety of stereo- and electronic properties. In general the ligands
are either mono- or bis-phosphines. With a rational selection
of parameters it should be easier to find the optimal reaction
conditions for the R-arylation of tetramic acids (see Scheme
2).
Even though no general reaction conditions exist in the
literature, it was possible to discern a general trend for the
reaction of substrates similar to tetramic acids, e.g., 1,3-
dicarbonyl compounds, cyclic substrates, and amino acids, from
the literature. It was found that an R-arylation is usually
conducted with either Pd(OAc)2 or Pd2(dba)3 with use of an
aryl bromide or iodide.13 Aryl chlorides are often too unreactive
for this type of chemistry. Many different solvents can be used,
but THF, toluene, dioxane, MeCN, or DMF are the most
common. The bases used can be divided into two groups: weak
inorganic bases such as Cs2CO3, K2CO3, K3PO4, Na3PO4, or
Na2CO3 and strong organic bases such as NaOtBu, KHMDS,
NaHMDS, LDA, or LiN(SiMe2Ph)2. The choice of base is
strongly dependent on the pKa value of the substrate but, in
general, strong bases can be used for most simple substrates.
However, if base-sensitive functionalities are present in the
molecule or deprotonation can cause racemization, strong bases
may give problems. Sometimes strong bases even require a two-
Results and Discussion
Initially, we tried to arylate Boc-pyPhe-OH (1) (the prefix
“py” is used to indicate that the amino acid is converted to a
tetramic acid8) with 4-bromoanisole (2) in the presence of 2
mol % of Pd(OAc)2 in THF at 80 °C overnight for the synthesis
of the 3-aryl tetramic acid 3 (see Table 1). Four different weak
inorganic bases were chosen: Cs2CO3, K3PO4, Na2CO3, and
K2CO3, respectively, in 2.3 equiv inspired by results published
by Buchwald and co-workers.13a We prioritized the screening
of a variety of weak bases because they are much more
compatible with functional groups, and because tetramic acids
(15) Zhao, Y.; Zhou, Y.; Liang, L.; Yang, X.; Du, F.; Li, L.; Zhang, H. Org.
Lett. 2009, 11, 555–558.
(16) Garc´ıa-Fortanet, J.; Buchwald, S. L. Angew. Chem., Int. Ed. 2008, 47,
8108–8111.
5034 J. Org. Chem. Vol. 74, No. 14, 2009

Palladium Catalyzed R-Arylation of Tetramic Acids
TABLE 3. Variation of Equivalents of Aryl Bromide 2, Pd(OAc)2,
Ligand 9, and K3PO4
FIGURE 4. Di-tert-butyl biaryl phosphine ligands.
Ar-Br 2
(equiv)
Pd(OAc)2
(mol %)
ligand 9
(mol %)
K3PO4
(equiv)
conv
(%)a
entry
1
2
3
4
5
2.0
1.0
1.0
1.0
1.0
2
2
4
2
4
4
4
8
8
4
2.3
5.0
2.3
2.3
2.3
43
15
>95
>95
79
a
1
Determined by H NMR.
are slightly acidic. The phosphine ligands P(tBu)3 (4), P(o-tolyl)3
(5), rac-BINAP (6), Xantphos (7), and DPPF (8) (see Figure
3) were screened with the four bases in 20 initial experiments.
Unfortunately, none of these conditions gave rise to any
significant product formation (conv <5%) and only starting
materials were isolated upon acidic workup (see Table 1, entry
1). However, to our gratification screening experiments with 4
mol % of biaryl phosphine ligand 9 gave promising results. Not
surprisingly, a major difference among the bases was observed,
K3PO4 and K2CO3 giving comparable results, 36% conversion
(entries 3 and 5), whereas Na2CO3 did not give any product
formation at all (entry 4). Cs2CO3 gave an intermediate result
with a conversion of 15% (entry 2).
To further increase the conversion with 4 mol % of biaryl
phosphine ligand 9 and 2.3 equiv of K3PO4, a series of
experiments at elevated temperature (100 °C) and a series with
prolonged reaction time (3 days) was conducted (see Table 2).
No significant change in conversions was observed after 3 days
(entries 1-4). On the other hand, the conversion was increased
at 100 °C overnight, especially with K3PO4, which almost gave
a 2-fold increase in conversion to 71% (entry 6). In both
experimental series, the use of Na2CO3 still did not give any
significant product formation (entries 3 and 7).
We reasoned that inefficient activation of the catalyst may
cause the low to moderate conversions obtained so far. Therefore
we tested the screening reaction at 80 °C with all four bases,
respectively, with 2 mol % of Pd2(dba)3 as a direct source of
Pd(0). However, we found that there was no improvement in
conversion, and we therefore assumed that the problem with
low conversion was not due to the nature of the palladium
catalyst.
Having a set of reaction conditions giving a moderate
conversion and a catalyst that presumably is sufficiently
activated, we screened a set of different equivalents with regard
to 4-bromoanisole (2), Pd(OAc)2, biaryl phosphine ligand 9, and
K3PO4 (see Table 3). Increasing the equivalents of 2 from 1.0
to 2.0 only increased the conversion slightly (entry 1), whereas
increasing the equivalents of K3PO4 to 5.0 gave a significant
reduction in conversion (entry 2). Furthermore, we examined
the effect of catalyst and ligand loading. Increasing the loading
of both catalyst and ligand to 4 and 8 mol %, respectively, gave
full conversion of the starting material to the desired product 3
FIGURE 5. Dicyclohexyl (13-19) and diphenyl (20) biaryl phosphine
ligands.
(entry 3). The same was true with increased ligand loading only
(entry 4). It was found that an excess of ligand was essential,
since 4 mol % of Pd(OAc)2 and 4 mol % of biaryl phosphine
ligand 9 gave a conversion of 79% (entry 5).
On the basis of these results, it seemed likely that screening
other biaryl phosphine ligands might give full conversion
without increased ligand loading. We therefore screened three
classes of commercially available biaryl phosphine ligands: a
series of di-tert-butyl biaryl phosphines 10-12 (see Figure 4),
a series of dicyclohexyl biaryl phosphines 13-19, and a single
diphenyl biaryl phosphine 20 (see Figure 5). Only 2-di-tert-
butylphosphino-2′,4′,6′-triisopropylbiphenyl (12) gave full con-
version with 4 mol % of ligand loading (see Table 4, entry 3),
the two other di-tert-butyl biaryl phosphines (10 and 11) gave
only poor conversion (entries 1 and 2). All dicyclohexyl biaryl
phosphine ligands 13-19 and the diphenyl biaryl phosphine
ligand 20 gave only traces of product. Apparently, this reaction
requires a sterically hindered and electron-rich ligand and the
di-tert-butyl substituents are essential for reactivity, which is
demonstrated by the absence of reactivity with the analogous
dicyclohexyl biaryl phosphine ligand 16.
Before moving on with substrate scope and limitations, we
analyzed the enantiomeric purity of the product 3, by means of
chiral HPLC. Fortunately, little racemization had occurred (ee
97%), which was expected due to the use of mild base.
Having arrived at these optimized reaction conditions we
wished to examine the scope and limitations of the reaction by
testing other different aryl coupling partners, namely aryl
chlorides, iodides, tosylates, and triflates. For comparison
reasons we chose to screen 4-methoxy derivates only (see Table
5). To our delight, 4-chloroanisole (21) reacted identically (entry
1) compared to 4-bromoanisole (2), as did aryl triflate 24 (entry
4). Use of aryl triflates expands the scope of the reaction further
because it allows conversion of phenols into functional coupling
partners very easily. Aryl iodide 22 and aryl tosylate 23 gave
J. Org. Chem. Vol. 74, No. 14, 2009 5035

Storgaard et al.
TABLE 4. Screening of Different Biaryl Phosphine Ligands
perature no reaction occurred and at 60 °C only 11% conversion
was observed after 1 h. Therefore a temperature at 80 °C was
chosen for further experiments.
With these suitable reaction conditions in hand, we once again
examined the catalyst loading. Reducing the loading of
Pd(OAc)2 to 1 mol % and ligand 12 to 2 mol %, only 35%
conversion was achieved after 1 h. However, full conversion
was achieved with overnight reaction times. As previously
shown, K3PO4 and K2CO3 gave similar conversions. We chose
to use K2CO3 exclusively because it is the most inexpensive.
Control experiments with no palladium catalyst or ligand were
performed, but no product was formed, as expected.
entry
ligand
conv (%)a
1
2
3
4
10
11
12
13-20
5
Using the optimized reaction conditions we tested a broad
range of aryl chlorides with different substituents, electron-
donating (EDG) as well as electron-withdrawing groups (EWG),
and with different disubstitution patterns (see Table 6). Both
meta- and para-disubstituted aryl chlorides reacted efficiently
giving full conversion after 1 h for most of the substrates.
Electron-donating groups such as ethers, alcohols, and amines
(entries 3, 6, and 7) worked well and the same was true for a
variety of electron-withdrawing groups like nitriles, nitro groups,
ketones, and esters (entries 10, 12, 13, and 15). Chlorobenzene
(25) itself also reacted smoothly giving full conversion after
1 h (entry 1). Coupling of 4-chlorophenol (29), 4-chloroaniline
(30), and 4-chlorobenzoic acid (37) did not proceed to comple-
tion after 1 h (entries 6, 7, and 14), but full conversion was
achieved overnight (16 h). Apparently, the unprotected func-
tional groups slowed down the reaction. Protection of the aniline
nitrogen as in (N-Boc)-4-chloroaniline (31) gave full conversion
after 1 h (entry 8). In the case of a free aliphatic amine (entry
9), the unprotected nitrogen completely quenched the reaction.
To synthesize a halogen-substituted product, we tested the
chemistry with 1-bromo-4-chlorobenzene (39) (entry 16) and
the 4-chloro product 54 was formed exclusively. This is
reasonable because bromides react faster than chlorides. Finally,
we examined a couple of similar ortho-substituted aryl chlorides,
but none of them gave any significant product formation after
16 h (entries 4 and 11). For the case of 2-chloroanisole (27) we
also tested 2-bromoanisole (28) to examine if the more reactive
bromide would react, but that was not the case (entry 5). It is
not surprising that ortho-substituted aryl halides did not react
at all, since there is much more steric hindrance around the
halogen. Use of less sterically demanding ligands did not solve
this problem.
21
>95
<5
a
1
Determined by H NMR.
TABLE 5. Testing of Different Coupling Partners: Aryl Halides,
Tosylate, and Triflate
In all cases, the crude product was isolated as the enol
tautomer upon acidic workup. However, during flash chroma-
tography we discovered some degree of shifting in equilibrium
toward the keto tautomer, which gives more complex NMR
spectra. To shift back the equilibrium we found that suspension
in EtOAc and treatment with 10% KHSO4 was suitable, which
eventually dissolved the compound completely as the solubilities
of the keto and enol tautomers are quite different. An example
of this equilibrium shifting is shown in Figure 6 with the 3-aryl
tetramic acid 3. To the left is shown the enol tautomer and the
a
1
Determined by H NMR.
only traces of the product 3 (entries 2 and 3). This observation
is important because aryl chlorides are generally much cheaper
than the corresponding iodides and a wider range of com-
mercially available compounds exists.
To fine-tune the chemistry, reaction time and temperature
were taken into consideration again. It was found that the
reaction is in fact complete in less than an hour at 80 °C with
4-chloroanisole (21) as the coupling partner. The product is
stable under the reaction conditions and the reaction time can
be extended to 16-20 h with no product decomposition. In
addition, lower temperatures were also tested. At room tem-
1
appurtenant H NMR spectrum in DMSO-d6 and to the right
1
the two possible keto tautomers. The H NMR spectrum of
3-keto is more complex because of broad peaks which might
be a result of the coexistence of both a cis and trans tautomer.
For both the enol and the keto tautomers, LC-MS (5f95%
MeCN in H2O added 0.05% TFA) showed one peak with the
same retention time. The acidic conditions apparently shift the
equilibrium to one of the tautomers independent of the initial
5036 J. Org. Chem. Vol. 74, No. 14, 2009

Palladium Catalyzed R-Arylation of Tetramic Acids
equilibrium position. In the 1H NMR spectra (see Figure 6) the
chemical shift of H5 is 4.71 ppm (dd) for the enol tautomer but
4.28 ppm (m) for the keto tautomers (H5′). The coupling pattern
of the H6 diastereotopic protons also changes; for the enol-
tautomer 3 the two protons are two well-resolved double
TABLE 6. Substrate Scope with Different Substituted Aryl
Chlorides
doublets (H6a + H6 ), whereas the keto tautomers (3-keto) show
b
a multiplet in the region slightly upfield (H6′ + H6′ ). On the
a
b
1
basis of H NMR it was not possible to determine the ratio
between the cis/trans tautomers.
Besides the aryl chlorides we also examined the scope of
the reaction with a series of chloro-substituted heterocycles (see
Figure 7). Three different pyridines 55-57, 2-chloropyrimidine
(58), 5-chloro-1-methyl-1H-imidazole (59), two chloro thiophenes
(60 and 61), and finally a bromo-substituted heterocycle,
3-bromothiophene (62), were tested. Unfortunately, none of
them gave any significant formation of product after 16 h. It is
plausible that the heteroatoms simply coordinate to palladium
resulting in an unreactive complex.
The reaction between Boc-pyPhe-OH (1) and 4-chloroanisole
(27) was tested with microwave heating and it was found that
full conversion (>95%) was achieved within only 5 min at 110
°C.
Finally, we wanted to expand the scope with other tetramic
acids than Boc-pyPhe-OH (1). A series of functionalized tetramic
acids were chosen: Boc-pyTyr(tBu)-OH (63), Boc-pyLys(Cbz)-
OH (64), Boc-pyArg(Pbf)-OH (65), Boc-pyThr(OtBu)-OH (66),
Boc-pyAsp(OtBu)-OH (67), and the glycine-derived tetramic
acid Boc-pyGly-OH (68). These were all subjected to the
optimized reaction conditions with 4-chloroanisole (21) as the
coupling partner (see Table 7). The chosen tetramic acids
represent a broad variety of functional side chains and different
protecting groups. Most of them gave similar yields compared
to the previous results, but Boc-pyAsp(tBu)-OH (67) and
especially Boc-pyGly-OH (68) gave much lower yield of the
corresponding 3-aryl tetramic acids 73 and 74, respectively
(entries 5 and 6).
So far, all reactions were conducted in THF. We then revisited
the initial reaction and tested some other solvents (see Table
8). Dioxane, MeCN, and DMF gave very low conversions (<5%)
no matter which base was used (Cs2CO3, K3PO4, Na2CO3, and
K3CO3, respectively) (entries 1-3). However, running the
reaction in toluene gave a significantly different result (entries
4-7). First of all, conversions were all much higher than the
equivalent experiments in THF (cf. Table 1, entries 2-5), even
with Na2CO3 a conversion of 29% was achieved (see Table 8,
entry 6). Full conversion was achieved with K3PO4, which gave
only 36% conversion in THF with the very same ligand (9).
Steric properties of the ligand are therefore not the only factor
dependent on the efficiency of the catalytic system.
Surprisingly, the crude product isolated from the toluene/
K3PO4 reaction was exclusively the keto tautomer (3-keto cis/
trans, see Figure 6), whereas THF gave the enol tautomer 3
when the same acidic workup procedure was used. To inves-
tigate this point further, we tried to shift the equilibrium of the
enol tautomer 3 by dissolving it in toluene, but nothing happened
based on TLC analysis. When adding aqueous 10% KHSO4 the
enol tautomer was slowly shifted toward the keto tautomers (3-
keto cis/trans), which have a significantly different Rf value.
Isolation as the keto tautomers is unfortunately not always
straightforward, and we discovered that it is easily shifted back
to the enol tautomer. Formation of the less stable keto tautomers
is not easy and it is rather unpredictable, probably because the
a Purified by flash chromatography. b No significant product formation
determined by 1H NMR or LC-MS. c Not purified by flash
chromatography, but crude 1H NMR is provided in the Supporting
Information. d Only the 4-chlorobenzene product was observed
determined by the isotope pattern of the molecular ion (LC-MS) of the
product (see the Supporting Information).
J. Org. Chem. Vol. 74, No. 14, 2009 5037

Storgaard et al.
1
FIGURE 6. Observed tautomeric equilibrium and appurtenant H NMR spectra (upfield region); the spectrum to the left belongs to the enol
tautomer of 3 and the spectrum to the right to the keto tautomer (3-keto cis/trans).
solvent and the workup conditions, but shifted back to the enol
tautomer upon EtOAc/10% KHSO4 treatment. A variety of
different substrates was tested and a range of functionalities are
tolerated, e.g., Boc- or Cbz-protected amines, Pbf-protected
guadinine groups, ethers, esters, ketones, alcohols, nitriles, and
nitro groups. Heterocycles and unprotected amines are not
FIGURE 7. Heteroaryl halides which did not undergo the title
compatible with this chemistry. Aryl chlorides, bromides, and
triflates all coupled nicely, whereas aryl iodides and tosylates
did not work. With respect to the substitution pattern of the
aryl chloride, electron-withdrawing as well as electron-donating
groups showed similar reactivity and meta- and para-substituted
aryl chlorides reacted identically. Due to steric hindrance ortho-
substituted aryl chlorides did not react. The title reaction can
be facilitated by microwave heating with reaction time down
to 5 min at 110 °C.
Synthesis of the previously mentioned amino benzyl ketones
is currently under development in our laboratory. These building
blocks will ultimately be used for the preparation of C-terminal
modified peptidyl enzyme inhibitors.
R-arylation.
equilibrium shifting is dependent on concentration, temperature,
pH, and solvent. This is comparable with the literature regarding
the keto-enol equilibrium of tetramic acids,1,3,4 as previously
described.
On the basis of the literature,13a we propose the following
mechanism for the coupling reaction (see Figure 8). The catalytic
cycle is assumed to be initiated by reduction of Pd(II) to the
active Pd(0), which might happen by a homocoupling of the
tetramic acids. Oxidation of phosphine ligands is another well-
known pathway for generation of Pd(0). However, Barder and
Buchwald reported recently that dialkylbiaryl phosphines are
highly resistant toward oxidation by molecular oxygen.17
Following reduction of Pd(II) to Pd(0), oxidative addition of
the aryl halide 25 takes place, then transmetalation by the
potassium enolate of the tetramic acid 1. Upon reductive
elimination the product is released and Pd(0) re-enters the
catalytic cycle. The desired product 40 can be isolated by acidic
workup.
Experimental Section
General Arylation Procedure. A vial was charged with dry
THF (3.0 mL), tetramic acid (1.00 mmol, 1.00 equiv), 2-di-tert-
butylphosphino-2′,4′,6′-triisopropylbiphenyl (12) (17 mg, 0.04
mmol, 0.04 equiv), K2CO3 (318 mg, 2.30 mmol, 2.30 equiv), and
an aryl chloride (1.00 mmol, 1.00 equiv). N2 was bubbled through
the reaction mixture and Pd(OAc)2 (4 mg, 0.02 mmol, 0.02 equiv)
was added, then the vial was filled with N2, sealed with a screw
cap, and placed in an aluminum heating block. The mixture was
stirred vigorously at 80 °C for 1 h (or 16 h, cf. Tables 6 and 7).
After being cooled to ambient temperature, the crude mixture was
transferred to a separatory funnel with 10% aqueous KHSO4 (10
mL) and extracted with EtOAc (30 mL + 20 mL). The combined
organic phases were dried over anhydrous Na2SO4, filtered, and
evaporated in vacuo. The yellow crude product was purified by
flash chromatography (5f10% MeOH in EtOAc, in some cases
up to 20% MeOH) affording the pure product typically as a keto/
enol tautomer mixture. The product was subsequently suspended
In conclusion, we have developed a new, mild, and racem-
ization-free palladium-catalyzed R-arylation of tetramic acids
giving rise to 3-aryl amino acid-derived tetramic acids. Through
optimization it was found that 2 mol % of Pd(OAc)2 and 4 mol
% of 2-di-tert-butylphosphino-2′,4′,6′-triisopropyl-1,1′-biphenyl
(12) gave full conversion in THF at 80 °C after 1 h for most
substrates. The two weak inorganic potassium bases, K2CO3
and K3PO4, worked equally well. The product could be isolated
as either the enol or the keto tautomer depending on reaction
(17) Barder, T. E.; Buchwald, S. L. J. Am. Chem. Soc. 2007, 129, 5096–
5101.
5038 J. Org. Chem. Vol. 74, No. 14, 2009

Palladium Catalyzed R-Arylation of Tetramic Acids
TABLE 7. Substrate Scope with Different Tetramic Acids
TABLE 8. Initial Reaction Conducted in Toluene and Screened
with Four Different Bases
entry
solvent
base
conv (%)a
1
2
3
4
5
6
7
dioxane
MeCN
DMF
toluene
toluene
toluene
toluene
all fourb
all fourb
all fourb
Cs2CO3
K3PO4
<5
<5
<5
86
>95
29
88
Na2CO3
K2CO3
a Determined by 1H NMR. b Tested with the four bases Cs2CO3,
K3PO3, Na2CO3, and K2CO3, respectively.
a Purified by flash chromatography. b Flash chromatography did not
successfully purify the product.
in EtOAc (50 mL). Ten percent aqueous KHSO4 (50 mL) was added
and the biphasic system was stirred vigorously at room temperature
until complete dissolution of the compound. The mixture was
transferred to a separatory funnel and the organic layer was
separated. The aqueous layer was extracted with EtOAc (20 mL)
and the combined organic layers were dried over anhydrous Na2SO4,
filtered, evaporated in vacuo, and dried overnight in high vacuum,
which afforded the pure product mostly as the enol tautomer.
(5S)-5-Benzyl-1-(tert-butyloxycarbonyl)-4-hydroxy-3-(4-meth-
oxyphenyl)-1,5-dihydropyrrol-2-one (3). Following the general
method for the arylation afforded 79% (313 mg) of the desired
product as a pale brown solid. 1H NMR (DMSO-d6) δ 12.11 (br s,
FIGURE 8. Proposed reaction mechanism for the R-arylation of
tetramic acids.
1H), 7.48 (d, J ) 8.6 Hz, 2H), 7.21-7.13 (m, 3H), 6.98 (d, J )
6.8 Hz, 2H), 6.86 (d, J ) 8.8 Hz, 2H), 4.71 (dd, J ) 4.3, 2.3 Hz,
1H), 3.72 (s, 3H), 3.44 (dd, J ) 14.0, 4.9 Hz, 1H), 3.26 (dd, J )
13.8, 1.9 Hz, 1H), 1.53 (s, 9H). 13C NMR (DMSO-d6) δ 168.9,
167.6, 157.7, 149.0, 134.2, 129.5, 128.7, 127.9, 126.8, 122.9, 113.2,
J. Org. Chem. Vol. 74, No. 14, 2009 5039

Storgaard et al.
105.1, 81.2, 57.9, 55.0, 34.6, 27.9. HRMS (m/z) calcd for
C46H50N2O10Na [2M + Na]+ 813.3358, found 813.3367. Anal.
Calcd for C23H25NO5: C, 69.86; H, 6.37; N, 3.54. Found: C, 69.84;
H, 6.52; N, 3.49. Mp 142-145 °C. IR (neat) ν 2975, 2930, 1750
(strong), 1363, 1284, 1251, 1147, 1095, 833, 699 cm-1. Chiral
HPLC: 4.57 min (minor) and 5.78 min (major) gave an enantiomeric
excess of 97%.
27.9. HRMS (m/z) calcd for C44H46N2O8Na [2M + Na]+ 753.3146,
found 753.3153. Anal. Calcd for C22H23NO4: C, 72.31; H, 6.34;
N, 3.83. Found: C, 71.97; H, 6.60; N, 3.96. Mp 86-88 °C. IR
(neat) ν 3082, 3061, 2977, 2928, 1753, 1702, 1661, 1645 (strong),
1397, 1359, 1298, 1149, 694 cm-1
.
(5S)-5-(4-tert-Butoxybenzyl)-1-(tert-butyloxycarbonyl)-4-hy-
droxy-3-(4-methoxyphenyl)-1,5-dihydropyrrol-2-one (69). Fol-
lowing the general method for the arylation afforded 69% (324
mg) of the desired product as a pale brown solid. 1H NMR (DMSO-
d6) δ 12.02 (s, 1H), 7.42 (d, J ) 9.1 Hz, 2H), 6.88 (d, J ) 8.6 Hz,
2H), 6.83 (d, J ) 9.1 Hz, 2H), 6.77 (d, J ) 8.6 Hz), 4.66 (dd, J )
4.6, 2.8 Hz, 1H), 3.71 (s, 3H), 3.39 (dd, J ) 13.9, 4.6 Hz, 1H),
3.19 (dd, J ) 13.9, 2.5 Hz, 1H), 1.53 (s, 9H), 1.16 (s, 9H). 13C
NMR (DMSO-d6) δ 168.8, 167.6, 157.7, 153.7, 149.0, 130.0, 128.9,
128.7, 123.2, 122.8, 113.1, 105.4, 81.1, 77.7, 57.9, 55.0, 34.1, 28.4,
27.9. HRMS (m/z) calcd for C54H66N2O12Na [2M + Na]+ 957.4508,
found 957.4516. Mp 147-153 °C. IR (neat) ν 2975, 2932, 1748,
1643, 1607, 1514, 1392, 1363, 1290, 1247, 1150 (strong), 1095,
Microwave-Assisted Synthesis of (5S)-5-Benzyl-1-(tert-buty-
loxycarbonyl)-4-hydroxy-3-(4-methoxyphenyl)-1,5-dihydropy-
rrol-2-one (3). A microwave vial was charged with Boc-pyPhe-
OH (1) (289 mg, 1.00 mmol, 1.00 equiv), 2-di-tert-butylphosphino-
2′,4′,6′-triisopropylbiphenyl (12) (17 mg, 0.04 mmol, 0.04 equiv),
Pd(OAc)2 (4 mg, 0.02 mmol, 0.02 equiv), and K2CO3 (318 mg,
2.30 mmol, 2.30 equiv). Dry THF (3.0 mL) was added and used to
carefully rinse the inside of the vial (for safety reasons no solid
may be stuck on the glass!). 4-Chloroanisole (21) (143 mg, 0.08
mL, 1.00 equiv) was added and N2 was bubbled into the vial to
secure an inert reaction atmosphere. The vial was capped, sealed,
and heated to 110 °C in a microwave synthesizer for 5 min. After
being cooled to ambient temperature, the crude product was
neutralized with 10% KHSO4 (10 mL) and EtOAc (30 + 20 mL)
was added. The combined organic layers were dried over anhydrous
Na2SO4 and evaporated in vacuo. The crude product was analyzed
by 1H NMR in DMSO-d6, which showed full conversion (>95%),
and the spectrum was identical with that of the product obtained
by conventional heating (80 °C, 1 h).
894, 830 cm-1
.
Acknowledgment. Morten Storgaard thanks Novo Nordisk
A/S, Corporate Research Affairs, and the Danish Ministry of
Science, Technology and Innovation for financial support.
Supporting Information Available: Experimental proce-
dures and characterization of all new compounds except those
mentioned in the Experimental Section (compounds 3, 40, and
69) and copies of 1H and 13C NMR spectra (including 1H NMR
spectra of compounds 45 and 52), analytical HPLC chromato-
grams of compounds not provided with elemental analysis, chiral
HPLC chromatogram of compound 3, and MS (TOF ES+) of
compound 54. This material is available free of charge via the
Internet at http://pubs.acs.org.
(5S)-5-Benzyl-1-(tert-butyloxycarbonyl)-4-hydroxy-3-phenyl-
1,5-dihydropyrrol-2-one (40). Following the general method for
the arylation afforded 75% (273 mg) of the desired product as a
1
pale brown solid. H NMR (DMSO-d6) δ 12.30 (br s, 1H), 7.52
(dd, J ) 8.3, 1.3 Hz, 2H), 7.28 (t, J ) 7.6 Hz, 2H), 7.22-7.15 (m,
4H), 7.01-6.98 (m, 2H), 4.73 (dd, J ) 4.8, 2.5 Hz, 1H), 3.45 (dd,
J ) 13.9, 4.8 Hz, 1H), 3.27 (dd, J ) 14.0, 2.4 Hz, 1H), 1.54 (s,
9H). 13C NMR (DMSO-d6) δ 170.2, 167.4, 149.0, 134.2, 130.5,
129.5, 127.9, 127.7, 127.5, 126.8, 126.3, 105.4, 81.2, 57.9, 34.6,
JO900799Y
5040 J. Org. Chem. Vol. 74, No. 14, 2009