Racemization in Suzuki couplings: A quantitative study using 4-hydroxyphenylglycine and tyrosine derivatives as probe molecules

Article abstract of DOI:10.1021/jo0621266

(Chemical Equation Presented) Reaction conditions considered to be typical in Suzuki couplings can cause significant (up to 34% of the unwanted enantiomer) loss of optical purity in sensitive substrates such as hydroxyphenylglycine 1. This may be remedied using sodium succinate instead of sodium carbonate as base, but chemical yields are somewhat lower. Optically pure biaryl amino acids related to those found in the chloropeptins and vancomycin were synthesized by Suzuki coupling of 1 with indolylboronic acids 6-8 and with cyclic boronic acid 9.

Full text of DOI:10.1021/jo0621266

Racemization in Suzuki Couplings:
A Quantitative Study Using
-Hydroxyphenylglycine and Tyrosine Derivatives
systematic investigation into the racemization produced in
certain aryl-aryl Suzuki couplings.
Initially, we focused on the model reactions shown in Scheme
4
1, where the aryl halide coupling partner is a racemizeable probe
as Probe Molecules
molecule and its counterpart a simple arylboronic acid. Any
racemization in the biaryl products should be quantifiable by
†
†
†
5
M o` nica Prieto, Silvia Mayor, Katy Rodr ´ı guez,
HPLC analysis on a chiral stationary phase.
,
†
,†,‡
Paul Lloyd-Williams,* and Ernest Giralt*
We chose the brominated and protected derivatives of (R)-
phenylglycine 1 and (S)-tyrosine 2 as probe molecules. While
the former provided a sensitive test molecule on account of its
susceptibility to base-induced racemization, we reasoned that
it might also give rise to atypically high quantities of the
enantiomeric Suzuki coupling product. We opted to use a
tyrosine derivative as a second probe molecule in order to gauge
more accurately what we felt would be more typical levels of
Department of Organic Chemistry, UniVersitat de Barcelona,
Mart ´ı i Franqu e` s, 1-11, Barcelona E-08028, Spain; and
IRBB-Parc Cient ´ı fic de Barcelona, Josep Samitier 1,
E-08028 Barcelona, Spain
lloydwilliams@ub.edu; egiralt@pcb.ub.es
6
,7
substrate racemization. We elected to use inexpensive and
easily prepared arylbromides as substrates in all Suzuki cou-
plings. Although they are somewhat less reactive than the more
ReceiVed October 13, 2006
8
,9
expensive aryl iodides, they are considerably more reactive
than the corresponding aryl chlorides.
We proposed attempting to suppress any racemization
produced in these model couplings by judicious choice of
inorganic base or other reaction parameters. We planned to
continue our study by submitting the phenylglycine derivatives
1
to Suzuki couplings with more complex aryl partners in order
to produce optically pure biaryl amino acids related to those
found in certain natural products. In particular, we were
1
0,11
interested in employing the indolylboronic acids
6-8 and
1
2
the cyclic boronic acid derivative 9. The indolyl amino acids
1
0-12 produced are models of the biaryl bisamino acids found
Reaction conditions considered to be typical in Suzuki
couplings can cause significant (up to 34% of the unwanted
enantiomer) loss of optical purity in sensitive substrates such
as hydroxyphenylglycine 1. This may be remedied using
sodium succinate instead of sodium carbonate as base, but
chemical yields are somewhat lower. Optically pure biaryl
amino acids related to those found in the chloropeptins and
vancomycin were synthesized by Suzuki coupling of 1 with
indolylboronic acids 6-8 and with cyclic boronic acid 9.
13,14
in the chloropeptins.
The biaryl amino acid 16, derived from
Suzuki coupling product 13 by subsequent synthetic manipula-
tions, is an analog of actinoidinic acid found in vancomycin.15
Pd(Ph3P)4 was used as catalyst, and all couplings were carried
out in a mixture of toluene-MeOH-H2O. The inorganic salts
KOAc, (CH2COONa)2, NaHCO3, Na2CO3, and K3PO4 were
screened as bases (see below). Each series of couplings was
carried out using similar stoichiometries, catalyst batches,
solvent compositions, concentrations, reaction times, and tem-
peratures. The conditions employed reflect typical current
The Suzuki coupling has established itself1-3 as a powerful
and versatile synthetic procedure for the formation of carbon-
carbon bonds. Advances and improvements in the chemistry of
the reaction have allowed increasingly complex coupling
(5) Steinauer, R.; Chen, F. M. F.; Benoiton, N. L. J. Chromatogr. 1985,
3
25, 111-126.
(6) Benoiton, N. L. Chemistry of Peptide Synthesis; CRC Press: Boca
Raton, 2006.
partners to be exploited, including those incorporating multiple
(7) Lloyd-Williams, P.; Albericio, F.; Giralt, E. Chemical Approaches
to the Synthesis of Peptides and Proteins; CRC Press: Boca Raton, 1997.
functional groups and stereogenic elements.1
(8) Espu n˜ a, G.; Arsequell, G.; Valencia, G.; Barluenga, J.; P e´ rez, M.;
Since addition of base is normally necessary to ensure good
yields in reasonable reaction times, there is a risk that the more
Gonz a´ lez, J. M. Chem. Commun. 2000, 1307-1308.
(9) Espu n˜ a, G.; Arsequell, G.; Valencia, G.; Barluenga, J.; Alvarez-
Gutierrez, J. M.; Ballesteros, A.; Gonzalez, J. M. Angew. Chem., Int. Ed.
Engl. 2004, 43, 325-329.
4
sensitive stereogenic centers in a molecule will racemize. To
our knowledge, this issue has not been addressed in any
published study to date. Here we disclose the results of a
(
10) Moyer, M. P.; Shiurba, J. F.; Rapoport, H. J. Org. Chem. 1986, 51,
106-5110.
11) Prieto, M.; Zurita, E.; Rosa, E.; Mu n˜ oz, L.; Lloyd-Williams, P.;
5
(
Giralt, E. J. Org. Chem. 2004, 69, 6812-6820.
(12) Brown, A. G.; Crimmin, M. J.; Edwards, P. D. J. Chem. Soc., Perkin
Trans. 1 1992, 123-130.
*
To whom correspondence should be addressed. Phone: 34-93-402-1260.
Fax: 34-93-339-7878.
†
Department of Organic Chemistry, Universitat de Barcelona
IRBB-Parc Cientific de Barcelona.
‡
(13) Deng, H. B.; Jung, J. K.; Liu, T.; Kuntz, K. W.; Snapper, M. L.;
Hoveyda, A. H. J. Am. Chem. Soc. 2003, 125, 9032-9034.
(14) Shinohara, T.; Deng, H. B.; Snapper, M. L.; Hoveyda, A. H. J.
Am. Chem. Soc. 2005, 127, 7334-7336.
(1) Nicolaou, K. C.; Bulger, P. G.; Sarlah, D. Angew. Chem., Int. Ed.
Engl. 2005, 44, 4442-4489.
(
(
(
2) Suzuki, A. J. Organomet. Chem. 2002, 653, 83-90.
3) Lloyd-Williams, P.; Giralt, E. Chem. Soc. ReV. 2001, 30, 145-157.
4) Benoiton, N. L. Int. J. Peptide Protein Res. 1994, 44, 399-400.
(15) Nicolaou, K. C.; Boddy, C. N. C. J. Am. Chem. Soc. 2002, 124,
10451-10455.
1
0.1021/jo0621266 CCC: $37.00 © 2007 American Chemical Society
Published on Web 01/10/2007
J. Org. Chem. 2007, 72, 1047-1050
1047

SCHEME 1. Model Reactions
TABLE 1. Na2CO3 Stoichiometry vs Racemization in Suzuki
Couplings between Probe Molecules 1 or 2 and Arylboronic Acid 3
practice for such couplings. Each reaction was performed at
least twice and repeated as necessary to establish yield
reproducibility. No attempt was made to maximize these yields
since the primary objective of this investigation was to ascertain
the effect of variation of inorganic base strength and stoichi-
ometry on the degree of racemization engendered.
probe
Na2CO3 (equiv)
yield
R/S (%)
1
1
1
2
2
2
1.50
0.75
0.25
3.10
1.50
0.75
57%
75%
55%
65%
61%
47%
66:34
80:20
87:13
100:0
100:0
100:0
TABLE 2. Base Type and Stoichiometry vs Racemization in
Suzuki Couplings between Probe Molecule 1 and Arylboronic Acid
3
base
equiv
1.50
yield 4
R/S (%)
Na2CO3 (pKa 10.3)
57%
75%
55%
70%
75%
52%
55%
52%
44%
54%
40%
36%
66:34
80:20
87:13
84:16
90:10
94:6
100:0
100:0
100:0
100:0
100:0
100:0
0
0
.75
.25
NaHCO3 (pKa 6.35)
1.50
0
0
.75
.50
(
CH2COONa)2 (pKa 5.61)
3.1
2
1
.1
.5
KOAc (pKa 4.74)
3.1
2
1
.1
.5
Optically pure probe molecules 1 and 2 were synthesized
using standard chemistry. (See Supporting Information for
reaction schemes and experimental data.) In preliminary experi-
ments we noted that Suzuki couplings between 1 and 4-meth-
oxyphenylboronic acid 3 gave biaryl 4 in yields and levels of
optical purity that depended on the amount of Na2CO3 used.
With 1.5 equiv of this base with respect to 1, a yield of 57% of
carried out with different quantities of these bases are sum-
marized in Table 2.
Although NaHCO3 is a weaker base than Na2CO3, it promoted
similar yields of biaryl 4 in Suzuki couplings between 1 and 3.
However, the optical purity of the coupling product was
significantly improved in comparison to that obtained using Na2-
CO3. We were unable to suppress racemization completely using
these bases. On the other hand, use of either (CH2COONa)2 or
KOAc, both of which are weaker than NaHCO3, did allow
optically pure biaryl 4 to be obtained, albeit at the cost of
diminished chemical yields in comparison with Na2CO3 and
NaHCO3. Marginally better yields of optically pure biaryl 4 were
obtained using (CH2COONa)2.
Having achieved the synthesis of the optically pure biaryls
in our model Suzuki couplings, we proceeded to subject 1 to
Suzuki coupling with indolylboronic acids 6-8 and cyclic
boronic acid 9. Yields and optical purities for these couplings,
mediated by both sodium carbonate and sodium succinate, are
listed in Table 3.
4
was obtained. HPLC analysis of this product on a chiral
stationary phase indicated 34% of the unwanted enantiomer.
On reducing the amount of base to 0.75 equiv, yields of 4 rose
to 75% and optical purity was also improved (20% of the
undesired enantiomer). When the amount of base was lowered
further to 0.25 equiv, yields of 4 were 55% with 13% of the
unwanted enantiomer still being detected.
Similar Suzuki couplings involving 2, on the other hand, gave
moderate to good yields of optically pure biaryl 5. Even when
3
.1 equiv of Na2CO3 were used, no unwanted enantiomer was
detected (see Table 1).
In attempting to suppress racemization in the Suzuki cou-
plings between 1 and 4-methoxyphenylboronic acid 3, we
screened the inorganic bases, KOAc, (CH2COONa)2, and
NaHCO3, as alternatives to Na2CO3. The results of couplings
When 1.5 equiv of Na2CO3 was used as base in couplings
between 1 and 6-8, biaryls 10-12 were produced in yields of
85%, 68%, and 77%, respectively. HPLC analysis of biaryl 10
1
048 J. Org. Chem., Vol. 72, No. 3, 2007

TABLE 3. Comparison of Sodium Carbonate and Sodium
Succinate in Suzuki Couplings between Probe Molecule 1 and
Indolylboronic Acids 6-8 and Cyclic Boronic Acid 9
Methyl (R)-N-Boc-3-(indol-5-yl)-4-methoxyphenylglycinate
10). KH (32 mg, 0.80 mmol) was suspended in anhydrous THF
0.2 mL) under an argon atmosphere at 0 °C. 5-Bromoindole (152
(
(
yield (%)
R/S (%)
mg, 0.78 mmol) in anhydrous THF (1.3 mL) was added, and the
mixture was stirred for 15 min. After cooling to -78 °C, a solution
ArB(OH)2 biaryl Na2CO3 (CH2COONa)2 Na2CO3 (CH2COONa)2
t
of BuLi in pentane (1.53 mmol), previously cooled to -78 °C,
6
7
8
9
10
11
12
13
85
68
77
54
52
46
42
20
71:29
a
a
100:0
100:0
100:0
100:0
was added dropwise. The mixture was brought to rt, stirred for 10
min, and re-cooled to -78 °C. B(OMe)
was added, and stirring was continued for a further 3 h at rt. H
5 mL) was added, and the mixture was extracted with AcOEt (2
3
(0.18 mL, 1.53 mmol)
2
O
62:38
(
a
Enantiomer ratios were not determined for these couplings
× 10 mL). The aqueous phase was acidified to pH 1 with 10%
HCl and re-extracted with AcOEt (3 × 10 mL). The combined
4
organic extracts were dried over anhydrous MgSO and filtered.
on a chiral stationary phase indicated that 29% of the enantiomer
had been produced. When these same couplings were carried
out using 2.1 equiv of (CH2COONa)2 as base, yields of 10-12
dropped to 52%, 46%, and 42%, respectively, but all three
compounds were optically pure. Indolylphenylglycines 10-12
are related to the biaryl amino acids found in the chloropeptins.
The Suzuki coupling between probe molecule 1 and the cyclic
boronic acid derivative 9 gave analogous results. When Na2-
CO3 was used as base biaryl 13 was formed in 54% yield. HPLC
analysis on a chiral stationary phase indicated the presence of
The solvents were evaporated, leaving the crude indolylboronic acid
6
as a pale brown oil.
Solutions of Pd(PPh ) (10 mg, 0.009 mmol) in toluene (0.5 mL)
3 4
and of the crude indolylboronic acid 6 (2/5 of the product prepared
above) in MeOH (0.4 mL) were added to a stirred solution of
sodium succinate (65 mg, 0.4 mmol) and methyl (R)-N-Boc-3-
bromo-4-methoxyphenylglycinate (72 mg, 0.19 mmol) in toluene
2
(0.6 mL) and H O (0.4 mL) under an argon atmosphere. The
mixture was brought to 90 °C and stirred for 60 min. The additions
of indolylboronic acid 6 (1/5 of the crude prepared above) in MeOH
3
8% of the epimer. Using 2.1 equiv of (CH2COONa)2 as base
3 4
(0.2 mL) and Pd(PPh ) (5 mg, 0.005 mmol) in toluene (0.3 mL)
were repeated a further three times at 60 min intervals. The mixture
was then stirred for a further 5 h. The solvents were evaporated,
and the crude product was purified by column chromatography
the coupling yield dropped to 20% but 13 was produced
optically pure. This biaryl alcohol was transformed via tosylate
1
1
4, azide 15, and amine 16 into the protected amino derivative
7, which may be considered to be a model of actinoidinic acid,
[
silica gel, AcOEt in hexanes (0-15%)], giving the product as a
yellow oil (41 mg, 52%): IR (film NaCl) 3405, 1701, 1607, 1499,
1
(
a component of the vancomycin family of antibiotics. No
racemization was observed in these subsequent synthetic opera-
tions.
-1 1
368, 1250, 1163 cm ; H NMR (200 MHz, d
6
-acetone) δ 1.42
9H, s), 3.68 (3H, s), 3.80 (3H, s), 5.29 (1H, d, J 8.0), 6.51 (1H,
m), 6.75 (1H, d, J 6.8), 7.06 (1H, d, J 8.4), 7.28-7.47 (5H, m),
This study shows that reaction conditions that are considered
to be typical for Suzuki couplings can give rise to significant
loss of optical purity with sensitive substrates such as 1. This
can be completely suppressed using sodium succinate as base,
although yields are somewhat lower. The sensitivity of probe
molecule 1 to racemization makes it an excellent test case for
screening reaction conditions in Suzuki couplings of other
racemization-sensitive substrates.
7.72 (1H, s), 10.27 (1H, s); 13C NMR (50 MHz, d -acetone) δ 28.5
6
(CH ), 52.5 (CH ), 55.9 (CH ), 58.3 (CH), 79.4 (C), 102.5 (CH),
3
3
3
111.3 (CH), 112.3 (CH), 121.9 (CH), 124.1 (CH), 125.7 (C), 125.8
CH), 127.8 (CH), 128.8 (C), 129.8 (C), 130.0 (C), 130.8 (CH),
33.1 (C), 136.3 (C), 157.4 (C), 172.5 (C); MS m/z 449 [(M +
(
1
+
+
K) , 68%], 433 [(M + Na) , 100%]; HRMS calcd for C23
26 2 5
H N O
3
Cl, c 1.1).
+
•
(M ), 410.1842: found, 410.1845; [R]
D
-92.3 (CH
Methyl (R)-N-Boc-3-(2′,4′-dimethoxy-6′-hydroxymethylphen-
n
yl)-4-methoxyphenylglycinate (13). A solution of BuLi in hexanes
4.2 mmol) was added to a stirred solution of 3,5-dimethoxy-2-
(
Experimental Section
bromobenzyl alcohol (494 mg, 20 mmol) in anhydrous THF (4 mL)
under an argon atmosphere at -78 °C. The mixture was allowed
Methyl (R)-N-Boc-3-(4′-methoxyphenyl)-4-methoxyphenyl-
to warm to rt over 15 min. After re-cooling to -78 °C, B(OMe)
3
glycinate (4). Solutions of Pd(PPh
0.8 mL) and of 4-methoxyphenylboronic acid (32 mg, 0.21 mmol)
in MeOH (0.35 mL) were added to a solution of sodium succinate
91 mg, 0.57 mmol) and methyl (R)-N-Boc-3-bromo-4-methoxy-
phenylglycinate (100 mg, 0.27 mmol) in toluene (0.8 mL) and H
0.7 mL) under an argon atmosphere. The resulting mixture was
stirred at 90 °C for 1 h. The additions of 4-methoxyphenylboronic
acid (17 mg, 0.11 mmol) in MeOH (0.35 mL) and Pd(PPh (4
3 4
) (7 mg, 0.006 mmol) in toluene
(0.45 mL, 4.0 mmol) was added and the mixture was allowed to
(
2
warm to rt and stirred for 2 h. H O (4 mL) was added, and the
phases were separated. The aqueous phase was extracted with
AcOEt (10 mL), acidified to pH 1 with 10% HCl, and re-extracted
with AcOEt (3 × 10 mL). The combined organic extracts were
(
2
O
(
4
dried over anhydrous MgSO . Filtration and solvent removal gave
the crude boronic acid 9 as a pale yellow semisolid.
3 4
)
mg, 0.003 mmol) in toluene (0.4 mL) were repeated a further three
times at 60 min intervals. The mixture was then stirred for a further
Solutions of Pd(PPh (40 mg, 0.035 mmol) in toluene (1 mL)
3 4
)
and of the crude arylboronic acid 9 (1/3 of the product prepared
above) in MeOH (0.9 mL) were added to a stirred mixture of
sodium succinate (254 mg, 1.57 mmol) and methyl (R)-N-Boc-3-
bromo-4-methoxyphenylglycinate (184 mg, 0.5 mmol) in toluene
(2 mL) and H O (1 mL) under an argon atmosphere. The mixture
2
was heated to 90 °C and stirred for 4 h. The additions of arylboronic
acid 9 (1/3 of the crude prepared above) in MeOH (0.8 mL) and
5
h. After cooling, the solvents were evaporated and the crude
product was purified by column chromatography [silica gel, AcOEt
in hexanes (0-13%)], furnishing 4 as a white semisolid (56 mg,
5
1
2%): mp 129-132 °C; IR (film NaCl) 3377, 2956, 2933, 2838,
-
1 1
746, 1715, 1519, 1495, 1248, 1167, 1049, 1030 cm ; H NMR
) δ 1.44 (9H, s), 3.72 (3H, s), 3.80 (3H, s), 3.84
3H, s), 5.29 (1H, d, J 7.2), 5.51 (1H, d, J 6.0), 6.92-6.96 (3H,
(400 MHz, CDCl
3
(
3 4
Pd(PPh ) (40 mg, 0.035 mmol) in toluene (1 mL) were repeated
13
m), 7.27-7.39 (2H, m), 7.42-7.46 (2H, m); C NMR (100 MHz,
CDCl ), 52.6 (CH ), 55.3 (CH ), 55.6 (CH ), 57.0
CH), 80.1 (C), 111.4 (CH), 113.5 (CH), 127.0 (CH), 129.0 (C),
twice more at 4-h intervals. The mixture was then stirred for a
further 16 h at 90 °C. The solvents were evaporated, and the crude
product was purified by column chromatography [silica gel, AcOEt
in hexanes (20-55%)], furnishing the product as a yellow oil (45
mg, 20%): IR (film NaCl) 3384, 2927, 2856, 1717, 1605, 1507,
3
) δ 28.3 (CH
3
3
3
3
(
1
29.3 (CH), 130.2 (C), 130.6 (CH), 130.8 (C), 154.8 (C), 156.5
+
(
C), 158.8 (C), 171.9 (C); MS m/z 440 [(M+K) , 97%], 424
+
+•
-1 1
[(M+Na) , 100%]; HRMS calcd for C22
H27NO
6
(M ), 401.1838:
1488, 1466, 1264, 1200, 1160, 1059, 1034 cm ; H NMR (400
found, 401.1856; [R]
D
-95.9 (CHCl , c 1.30).
3
3
MHz, CDCl ) δ 1.43 (18H, s), 3.67 (3H, s), 3.68 (3H, s), 3.71
J. Org. Chem, Vol. 72, No. 3, 2007 1049

(
3H, s), 3.72 (3H, s), 3.73 (6H, s), 3.86 (6H, s), 4.19-4.31 (4H,
m), 5.27 (2H, d J 6.8), 5.50 (2H, sa), 6.49-6.50 (2H, m), 6.71
1H, d, J 2.0), 6.72 (1H, d, J 2.0), 6.95 (1H, d, J 8.8), 6.96 (1H, d,
J 8.4) 7.10 (1H, d, J 2.4), 7.13 (1H, d, J 1.6), 7.30-7.35 (2H, m);
Acknowledgment. This work was supported by funds from
MEC-FEDER (Bio 2005-00295), MEC (CTQ2006-12460/BQU)
and Generalitat de Catalunya (SGR and CeRBa).
(
Supporting Information Available: Reaction schemes for the
synthesis of probe molecules 1 and 2; experimental procedures for
the synthesis of compounds 1, 2, 5, 11, 12, and 17-21 together
13
C NMR (75 MHz, CDCl
5.7 (CH ), 55.8 (CH ), 63.5 (CH
CH), 111.4 (CH), 125.5 (C), 127.5 (CH), 128.7 (C), 131.2 (CH),
41.5 (C), 154.8 (C), 157.2 (C), 158.0 (C), 160.3 (C), 171.8 (C);
3
) δ 28.2 (CH
3 3 3
), 52.5 (CH ), 55.3 (CH ),
5
(
1
3
3
2
), 80.0 (C), 98.3 (CH), 104.0
1
13
with their characterization data; copies of the H and C NMR
spectra for compounds 1, 2, 4, 5, 10-13, 17, and 18-21. This
material is available free of charge via the Internet at http://
pubs.acs.org.
+
+
MS m/z 500 [(M + K) , 38%], 484 [(M + Na) , 100%]; HRMS
calcd for C24
+
•
H31NO
8
(M ), 461.2050: found: 461.2065; [R]
D
-64.5 (CHCl
3
, c 1.00).
JO0621266
1
050 J. Org. Chem., Vol. 72, No. 3, 2007