Simplifying pyridoxal: Practical methods for amino acid dynamic kinetic resolution

Article abstract of DOI:10.1021/ol100319b

Figure presented Metal complexes of picolinaldehyde are identified as low-cost and environmentally benign catalysts, providing high reaction rates and turnovers for the racemization of amino acids. These pyridoxal surrogates demonstrate activity toward a variety of amino acid esters. Applications to chemoenzymatic dynamic kinetic resolutions provide access to amino acids in high yields and with excellent enantioselectivities, demonstrating their compatibility with protease-mediated transformations.

Full text of DOI:10.1021/ol100319b

ORGANIC
LETTERS
2010
Vol. 12, No. 9
1916-1919
Simplifying Pyridoxal: Practical Methods
for Amino Acid Dynamic Kinetic
Resolution
Albert E. Felten, Gangguo Zhu, and Zachary D. Aron*
Department of Chemistry, Indiana UniVersity, 800 East Kirkwood AVenue,
Bloomington, Indiana 47405-7102
zaron@indiana.edu
Received February 6, 2010
ABSTRACT
Metal complexes of picolinaldehyde are identified as low-cost and environmentally benign catalysts, providing high reaction rates and turnovers
for the racemization of amino acids. These pyridoxal surrogates demonstrate activity toward a variety of amino acid esters. Applications to
chemoenzymatic dynamic kinetic resolutions provide access to amino acids in high yields and with excellent enantioselectivities, demonstrating
their compatibility with protease-mediated transformations.
Chiral amines are critical components of pharmaceuticals,
catalysts, and functional biomaterials.1-3 Despite advances
in enantioselective synthesis, the majority of amines continue
to be manufactured in racemic form, requiring a subsequent
resolution to access pure material.1 An attractive strategy to
improve material throughput would use dynamic kinetic
resolutions that combine amine racemization catalysts with
resolving agents to convert racemic mixtures to single
antipodes.4 Unfortunately, amine racemization remains a
significant challenge, with harsh conditions, high loadings,
toxic metals, restricted substrate scopes, and prohibitive costs
limiting practical applications.5-7
Biomimetic catalysts would provide an attractive alterna-
tive for amine racemization. The primary inspiration for
amine racemization is the enzyme cofactor pyridoxal-5′-
(4) Griengl, H.; Faber, K.; Steinreiber, J. Chem.sEur. J. 2008, 14, 8060–
8072.
(5) (a) Reetz, M. T.; Schimossek, K. Chimia 1996, 50, 668–669. (b)
Parvulescu, A.; De Vos, D.; Jacobs, P. Chem. Commun. 2005, 5307–5309.
(c) Paetzold, J.; Backvall, J. E. J. Am. Chem. Soc. 2005, 127, 17620–17621.
(d) Blacker, A. J.; Stirling, M. J.; Page, M. I. Org. Process Res. DeV. 2007,
11, 642–648. (e) Gastaldi, S.; Escoubet, S.; Vanthuyne, N.; Gil, G.; Bertrand,
(1) For a review on the industrial synthesis of chiral amines, see: Breuer,
M.; Ditrich, K.; Habicher, T.; Hauer, B.; Kesseler, M.; Sturmer, R.; Zelinski,
T. Angew. Chem., Int. Ed. 2004, 43, 788
.
M. P. Org. Lett. 2007, 9, 837–839.
(2) For select examples of chiral amines in catalysis, see: (a) Mukhejee,
S.; Yang, J. W.; Hoffmann, S.; List, B. Chem. ReV. 2007, 107, 5471–5569.
(b) Desimoni, G.; Faita, G.; Jorgensen, K. A. Chem. ReV. 2006, 106, 3561–
(6) (a) Olivard, J.; Metzler, D. E.; Snell, E. E. J. Biol. Chem. 1952,
199, 669–674. (b) Clark, J. C.; Phillipps, G. H.; Steer, M. R. J. Chem.
Soc., Perkin Trans. 1 1976, 475–481. (c) Tabushi, I.; Kuroda, Y.; Yamada,
M. Tetrahedron Lett. 1987, 28, 5695–5698. (d) Battacharya, A.; Araullo-
Mcadams, C.; Meier, M. B. Synth. Commun. 1994, 24, 2449–2459. (e) Liu,
L.; Breslow, R. Tetrahedron Lett. 2001, 42, 2775–2777. (f) Zimmermann,
3651
.
(3) For select examples of chiral amines in functional materials, see:
(a) Puiggali, J.; Subirana, J. A. J. Pept. Sci. 2005, 11, 247–249. (b) Xie,
D.; Park, J.-G.; Faddah, M.; Zhao, J.; Khanijoun, H. K. J. Biomater. Appl.
2006, 21, 147–165. (c) Broyer, R. M.; Quaker, G. M.; Maynard, H. D.
V.; Beller, M.; Kragl, U. Org. Process Res. DeV. 2006, 10, 622–627
(7) Servi, S.; Tessaro, D.; Pedrocchi-Fantoni, G. Coord. Chem. ReV.
2008, 252, 715–726
.
J. Am. Chem. Soc. 2007, 130, 1041–1047
.
.
10.1021/ol100319b  2010 American Chemical Society
Published on Web 04/05/2010

phosphate (PLP), which uses Schiff base and quinonoid
formation to racemize amino acids under mild conditions
(Figure 1).6a,d,8 Analogues such as pyridoxal, salicylaldehyde,
Table 1. Effect of Metal on the Racemization of Phe(OMe)a
entry
metal/aldehyde
Xb
TONc
1
2
3
4
5
6
7
8
AgNO3/picolinaldehyde
Ti(OiPr)4/picolinaldehyde
Co(AcAc)3/picolinaldehyde
Cu(OTf)2/picolinaldehyde
Y(OTf)3/picolinaldehyde
Ni(AcAc)2/picolinaldehyde
FeSO4/picolinaldehyde
Ca(OTf)2/picolinaldehyde
Zn(OTf)2/picolinaldehyde
Zn(OTf)2/picolinaldehyded
ZnCl2/picolinaldehyde
99.2
99.1
95.6
88.9
67.3
46.8
33.4
29.9
27.0
30.2d
25.7
24.2
22.2
98.3
58.3
0.4
0.45
2.3
5.9
Figure 1. Pyridoxal-5′-phosphate (PLP)-mediated amino acid
racemization.
19.8
38.0
54.8
60.6
65.4
57.9d
68.0
71.0
75.2
0.6
9
and 3,5-dinitrosalicylaldehyde have been used in dynamic
kinetic resolutions, but low activity and/or prohibitive costs
limit their utility.6 We report that simple metal complexes
of picolinaldehyde efficiently catalyze amino acid ester
racemization under conditions compatible with enzymatic
resolution. Our reactions facilitate dynamic kinetic resolu-
tions of amino acids and provide an inexpensive and green
alternative to existing methods. This observation is particu-
larly noteworthy since monosubstituted pyridines have not
previously been reported to catalytically activate amines.9-11
Our rationale for using picolinaldehyde metal complexes
was based on the hypothesis that metal binding would
facilitate Schiff base formation in a manner analogous to
the 3-hydroxyl group of pyridoxal-5′-phosphate (PLP; see
Figure 1).12,13 These complexes are particularly attractive
as cation binding is anticipated to enhance reactivity through
coordination to the pyridine nitrogen, stabilizing the electron-
rich quinonoid tautomer B (Scheme 1).9
10d
11
12
13
14
15
ZnClO4/picolinaldehyde
Zn(OAc)2/picolinlaldehyde
Zn(OTf)2/4-pyridinecarboxaldehyde
Zn(OTf)2/pyridoxal
27.2
a Reactions were run for 30 min at 0.5 M at room temperature, using
0.02 equiv of aldehyde and 0.02 equiv of Zn(OTf)2. b As determined by
HPLC. c TON was determined from reactions halted at 30 min.14,15
d Reaction was run using free Phe(OMe) in the presence of 1 equiv of
trifluoroacetic acid.
efforts revealed that amine racemization occurs fastest in
protic solvents and in the presence of ammonium salts, which
are necessary for promoting Schiff base exchange. Under
optimal conditions, complexes of Zn(OTf)2 and picolinal-
dehyde proved highly effective, racemizing Phe(OMe) at
rates 4-fold faster than Zn(OTf)2/pyridoxal (Table 2, entries
1 and 6). Consistent with our hypothesis, complexes of
4-pyridinecarboxaldehyde that lack the ability to effectively
chelate metals are largely inactive (Table 1, entry 14).
The racemization of Phe(OMe) with various multivalent
cations (select examples shown, Table 1) demonstrated
distinctive trends. Among metal salts, dicationic species with
small atomic radii are most effective,16 whereas the coun-
terion has little effect when incorporated on the metal or as
Scheme 1. Picolinaldehyde-Catalyzed Amine Racemization
(11) (a) Casella, L.; Gullotti, M. Inorg. Chem. 1983, 22, 2259–2266.
(b) Casella, L.; Guillotte, M. Inorg. Chem. 1986, 25, 1293–1303. (c) Chavez-
Gil, T. E.; Yasaka, M.; Senokuchi, T.; Sumimoto, M.; Kurosaki, H.; Goto,
M. Chem. Commun. 2001, 2388–2389.
(12) (a) Nitschke, J. R. Angew. Chem., Int. Ed. 2004, 43, 3073. (b)
Schultz, D.; Nitschke, J. R. Angew. Chem., Int. Ed. 2006, 45, 2453–2456.
(c) Schultz, D.; Nitschke, J. R. J. Am. Chem. Soc. 2006, 128, 9887–9892.
(13) (a) Crugeiras, J.; Rios, A.; Riverios, E.; Richard, J. P. J. Am. Chem.
Soc. 2009, 131, 15815–15824. (b) Crugeiras, J.; Rios, A.; Riveiros, E.;
Amyes, T. L.; Richard, J. P. J. Am. Chem. Soc. 2008, 130, 2041–2050. (c)
Toth, K.; Richard, J. P. J. Am. Chem. Soc. 2007, 129, 3013–3021. (d) Auld,
D. S.; Bruice, T. C. J. Am. Chem. Soc. 1967, 89, 2083–2089.
(14) TON was defined as the number of successful catalyst/substrate
interactions per mol of catalysts that occurred within the time span of the
given reaction. These numbers were calculated using the following equation
derived from methods described by Carpenter:15 TON(T) ) [ln(100/ee)]/
(cat) where ee ) ee of substrate at given time T, and (cat) ) loading of the
catalyst relative to substrate.
Initial studies established picolinaldehyde metal chelates
as amino acid racemization catalysts (Table 1). Optimization
(8) Eliot, A. C.; Kirsch, J. F. Annu. ReV. Biochem. 2004, 73, 383–415
.
(9) Several reports exist that describe monosubstituted pyridines acting
as stoichiometric PLP analogues; however, no catalytic examples have been
described (see ref 10).
(15) Carpenter, B. K. Determination of Organic Reaction Mechanisms;
Wiley & Sons: New York, 1984; pp 35-37.
(16) Shannon, R. D.; Prewitt, C. T. Acta Crystallogr. 1969, B25, 925–
946.
(10) (a) Maley, J. R.; Bruice, T. C. J. Am. Chem. Soc. 1968, 90, 2843.
(b) Bachmann, S.; Knudsen, K. R.; Jorgensen, K. A. Org. Biomol. Chem.
2004, 2, 2044–2049
.
Org. Lett., Vol. 12, No. 9, 2010
1917

To improve our understanding of catalyst structure and
explore reaction scope, we have examined amine activation
with a variety of substrates. These experiments measured
catalytic activity in the context of deuterium incorporation
to allow simple substrates such as Gly(OMe) to be directly
assessed (Table 3). Among amino acids, less sterically
Table 2. Effect of Aldehyde Electronics on the Racemization of
Phe(OMe)a
Table 3. Deuterium Incorporation of Selected Aminesa
a Reactions were run for 30 min at 0.5 M at room temperature, using
0.02 equiv of aldehyde and 0.02 equiv of Zn(OTf)2. b As determined by
HPLC. c TON was determined from reactions halted at 30 min.14,15 d TOF
was determined from TON determined at five time points, run in triplicate
for each reaction (5-65% conversion). e Reactions were run using Ca(OTf)2
in lieu of Zn(OTf)2.
a Reactions were run for 30 min at 0.5 M at room temperature, using
0.02 equiv of aldehyde and 0.02 equiv of Zn(OTf)2, with % conversions
determined by 1H NMR. b The value of kobs was determined from time
courses run in triplicate and analyzed as pseudo-first-order.
an exogenous salt (Table 1, entries 9-13). This result is
surprising considering that the Schiff base changes from an L2-
to an LX-type ligand upon deprotonation (Scheme 1). The lack
of salt effects suggests that the active species does not
incorporate these counterions into the metal ligand sphere.
Our studies found that incorporation of either σ- or
π-electron-withdrawing substituents on picolinaldehyde slowed
the reaction rate, whereas electron-rich substituents acceler-
ated racemization (Table 2). This observation was unexpected
as electron-poor substituents on the aldehyde were anticipated
to accelerate Schiff base exchange by increasing imine
electrophilicity and to facilitate proton transfer by stabilizing
the electron-rich quinonoid intermediate. This phenomenon
was examined further by performing Phe(OMe) racemization
in the presence of either Zn(OTf)2 or Ca(OTf)2 with both
5-cyanopicolinaldehyde and picolinaldehyde (Table 2, entries
1 and 2). In the case of picolinaldehyde, little change in
reaction rate was observed, whereas with 5-cyanopicolinal-
dehyde, the use of Ca(OTf)2 resulted in a 20-fold rate
increase over reactions with Zn(OTf)2. The unusual trends
in pyridine substitution and catalysis, coupled with the
observation that different metals affect substituted pyridines
in different ways, may reflect variations in the metal-binding
properties of the substituted pyridines. Accordingly, electron-
poor pyridines would act as weaker σ-donors to disfavor
metal binding and thus inhibit the formation of catalytically
active complexes.
hindered substrates have proven to be more reactive, with
Gly(OMe) incorporating deuterium fastest and Val(OMe)
incorporating it slowest (Table 3, entries 3 and 7). These
results suggest that either Schiff base exchange is rate-
limiting or that steric congestion around the catalyst impairs
deprotonation. In contrast, reactions are largely insensitive
to the ester substituent, with methyl and t-butyl esters
incorporating deuterium at similar rates (Table 3, entries 1
and 2). The presence of the carboxylate appears to play a
significant role in mediating catalyst activity, with minimal
deuterium incorporation seen with p-nitrobenzyl amine
(Table 3, entry 9) and relatively slow incorporation in the
case of the sterically unencumbered cyanomethyl amine
(Table 3, entry 8). The high reactivity of carboxylate-
containing substrates suggests that the carboxylate is directly
bound to the cation, with the insensitivity toward ester
substitution suggesting that the carbonyl oxygen provides
the coordination (Figure 2).
Figure 2
.
Proposed three-coordinate metal/substrate complex.
1918
Org. Lett., Vol. 12, No. 9, 2010

To examine our system in the context of chemoenzymatic
dynamic kinetic resolutions, we examined the reactions of a
variety of amino acid esters in the presence of both Alcalase
and picolinaldehyde metal complexes (Table 4). Wang and
Since protic acid is generated, lithium carbonate was
introduced to buffer the reaction and maintain enzyme
activity. Product racemization is avoided by running reactions
in a mixture of t-butanol and water, which results in the
precipitation of amino acid products. Under optimal condi-
tions, highly enantioenriched products can be directly ac-
cessed by a simple filtration. Aromatic as well as straight-
and γ-branched-chain amino acids are resolved in good yields
with high enantiopurity, whereas ꢀ-branched amino acids
are poorly resolved by Alcalase (data not shown).
Table 4. Dynamic Kinetic Resolutions of Amino Acid Estersa
In summary, a simple biomimetic catalyst has been
introduced that mediates low-cost and environmentally
friendly racemizations of amino acids. Applications to
dynamic kinetic resolutions are described and demonstrate
the compatibility of our catalyst with chemoenzymatic
strategies. The key advantage of this study is that it provides
a readily available amino acid racemization catalyst which
is capable of interacting with a diverse array of substrates.
Future efforts will explore the utility of these simple
pyridoxal analogues in a variety of transformations.
Acknowledgment. We acknowledge the financial support
of Indiana University.
Supporting Information Available: Experimental pro-
cedures describing the synthesis of 4-dimethylaminopicoli-
naldehyde, general procedures for transformations depicted
in Tables 1-4, and tables providing raw data from which
Tables 1-4 were generated. This material is available free
of charge via the Internet at http://pubs.acs.org.
a Reactions were run for 4 h at 0.2 M at room temperature, using 0.1
equiv of aldehyde, 0.05 equiv of Zn(OTf)2, 0.6 equiv of LiCO3, and 1 mL/
mmol alcalase solution. b Isolated amino acid. c As determined by HPLC.
co-workers have previously established that Alcalase, an
inexpensive and readily available proteolytic enzyme mix-
ture,17 was an effective resolving agent for pyridoxal-
catalyzed dynamic kinetic resolutions.17c Our development
of picolinaldehyde-catalyzed dynamic kinetic resolutions
required several changes to our racemization conditions.
OL100319B
(17) (a) Chen, S.-T.; Wang, K.-T.; Wong, C.-H. J. Chem. Soc., Chem.
Commun. 1986, 1514–1516. (b) Kijima, T.; Ohshima, K.; Kise, H. J. Chem.
Technol. Biotechnol. 1994, 59, 61–65. (c) Chen, S.-T.; Huang, W.-H.; Wang,
K.-T. J. Org. Chem. 1994, 59, 7580–7581. (d) Chen, S.-T.; Wang, K.-T.
J. Chin. Chem. Soc. 1999, 46, 301–311.
Org. Lett., Vol. 12, No. 9, 2010
1919