Enantioselective transfer hydrogenation of ketone catalysed by artificial metalloenzymes derived from bovine β-lactoglobulin

Article abstract of DOI:10.1039/c2cc36980j

Artificial metalloproteins resulting from the embedding of half-sandwich Ru(II)/Rh(III) fatty acid derivatives within β-lactoglobulin catalysed the asymmetric transfer hydrogenation of trifluoroacetophenone with modest to good conversions and fair ee's. T

Full text of DOI:10.1039/c2cc36980j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 11984–11986
www.rsc.org/chemcomm
COMMUNICATION
Enantioselective transfer hydrogenation of ketone catalysed by artificial
metalloenzymes derived from bovine b-lactoglobulinw
Alice Chevalleyab and Michele Salmain*ab
`
Received 26th September 2012, Accepted 24th October 2012
DOI: 10.1039/c2cc36980j
Artificial metalloproteins resulting from the embedding of
half-sandwich Ru(II)/Rh(III) fatty acid derivatives within
b-lactoglobulin catalysed the asymmetric transfer hydrogenation of
trifluoroacetophenone with modest to good conversions and fair ee’s.
papain from Carica papaya. Depending on the metal cofactor,
10
papain acquired the ability to accelerate a Diels–Alder reaction,
+ 11
to catalyse either the reduction of NAD(P) , or the TH of
trifluoroacetophenone (TFAP), albeit with modest enantio-
1
2
selectivity. More generally, the presence of a nucleophilic group
within a broad cavity is an adequate feature to construct artificial
Design of metalloproteins has received much attention during
the last years mostly to generate (enantioselective) biocatalysts
13
metalloenzymes. In search for an alternative protein host and
inspired by the work performed with the biotin–avidin system, we
have chosen bovine b-lactoglobulin (b-LG) to design an artificial
transfer hydrogenase for the stereoselective synthesis of secondary
alcohols.
1
operating under mild conditions. One significant advantage
of such hybrid constructs is to clearly separate the reactivity,
mostly brought about by the metal cofactor, from the selectivity
2
brought about by the protein host. Three main strategies
b-LG is the most abundant whey protein in cow’s milk. It is
produced as two genetic variants called A and B differing by 2
amino acids. Its sequence contains 162 amino acids that
mainly fold into an 8-stranded antiparallel b-barrel typical
of the lipocalin superfamily, thus forming a calyx lined with
concur to the design of artificial metalloenzymes, depending
on whether the nonnative metal cofactor is incorporated into
the protein host by either dative, covalent or supramolecular
3
anchoring methods. In practice, protein scaffolds suitable for
artificial metalloprotein design should meet criteria of chemical/
physical stability, abundance and knowledge of 3D-structure.
When dealing with supramolecular constructs, ligands/inhibitors
with high affinity for the protein host should be known and their
binding mode well defined. As far as enantioselective catalysis
is concerned, supramolecular constructs based on the biotin–
1
4
hydrophobic residues. X-ray structural characterizations at
pH 6.2, 7.1 and 8.2 showed that loop EF undergoes a
pH-dependent conformational change at residues 85–90. This
mobile loop acts as a lid controlling ligand access to the
1
5
calyx. It is now well established that a range of hydrophobic
ligands such as fatty acids bind to b-LG with submicromolar
(
strept)avidin system have proved the most efficient ones to date,
4
16
dissociation constants and close to 1 : 1 stoichiometry. Structural
studies have shown that the fatty acid aliphatic chain occupies the
b-LG’s hydrophobic cavity with the carboxyl function directed
towards the calyx entrance and the rest more or less buried as a
in terms of versatility, reactivity and selectivity.
Asymmetric transfer hydrogenation (ATH) of ketones is a
powerful route to access chiral secondary alcohols. Several
6
Ru(II), Rh(III) and Ir(III) d piano-stool complexes, including
17
function of the alkyl chain length (Fig. S1, ESIw).
chiral mono N-tosylated vicinal diamine or bipyridyl-type ligands,
All these features prompted us to synthesize compounds 1
and 2 derived from palmitic and lauric acid, respectively, and
also carrying a dipyridylamide moiety. These compounds were
obtained by acylation of 2,2-dipyridylamine with lauroyl
chloride or palmitoyl chloride (Scheme 1). Reaction of 1 or
have been shown to be efficient catalysts for the (A)TH of ketones
5–7
in neat water using formate as hydrogen source.
Metal-
catalysed (A)TH involves transient formation of a metal-hydride
species followed by hydride and proton transfer from the metal
8
and the diamine ligand (or the solvent) to the substrate. Since this
2
with [Cp*Rh(m-Cl)Cl] , [(p-cym)Ru(m-Cl)Cl] or [(benz)-
2 2
reaction can be carried out in water under mild conditions, it has
9
been amenable to artificial metalloenzyme design.
Ru(m-Cl)Cl)]2 afforded the monomeric, monocationic com-
plexes 3–7 (Scheme 1) with high yield. Their H NMR
1
We have recently entered the field of artificial metalloenzymes
by covalently anchoring several half-sandwich complexes to
spectra were consistent with metal coordination to the two
pyridine rings of the dipyridylamide moiety. These complexes
are water-soluble to some extent and can be conveniently kept
as DMSO solutions where they display long-term stability.
Since TFAP was efficiently reduced in the presence of
a
Chimie ParisTech, Laboratoire Charles Friedel, 11 rue Pierre et Marie
Curie, 75005 Paris, France. E-mail: Michele-salmain@chimie-paristech.fr;
Fax: +33 1 43260061; Tel: +33 1 44276732
CNRS, UMR 7223, Paris, France.
b
6
2+
6
[(Z -C
of TH of TFAP were carried out at 40 1C in water containing
M formate (initial pH = 7.5) and 1 or 2 mol% of each
6 6 2
Me )Ru(bpy)(OH )] and formate in water, catalytic tests
E-mail: Michele-salmain@chimie-paristech.fr;
Fax: +33 1 43260061; Tel: +33 1 44276732
w Electronic supplementary information (ESI) available: Detailed
experimental procedures, CD spectra. See DOI: 10.1039/c2cc36980j
1
complex. Gradual appearance of a-(trifluoromethyl)benzyl alcohol
1
1984 Chem. Commun., 2012, 48, 11984–11986
This journal is c The Royal Society of Chemistry 2012

View Article Online
Scheme 1 Synthesis of ligands and complexes.
Fig. 2 Circular dichroism spectra of b-LG (A), 6Cb-LG (B) and
2Cb-LG (C).
addition of 0.8 molar eq. of the Rh complexes 3 or 6, two induced
CD bands appeared in the spectra, one positive band at 271 nm
and one negative broad band centred at 360 nm (Fig. 2 and
Fig. S7, ESIw). These bands were assigned to ligand-centred and
2
0
metal-centred electronic transitions, respectively. In these cases,
the chromophore portion of 3 and 6 also became located in a
chiral environment in the presence of b-LG. Upon addition of
0.8 molar eq. of complex 4, 5 or 7 to b-LG, a change of the near-
UV CD spectrum was observed only for 7 in the form of an
ellipticity decrease in the 250–290 nm spectral range (Fig. S8–S10,
ESIw). Binding of complexes 4 and 5 to b-LG was nevertheless
indirectly shown by the ability of both complexes to displace the
binding of 1 (Fig. S11 and S12, ESIw). In the far-UV region
(200–250 nm), no change was observed in the spectrum of b-LG
after addition of 1 or 4 (Fig. S13 and S14, ESIw). Thus, the new
ligands and complexes derived from palmitic and lauric acid were
all able to bind b-LG without disturbing the protein secondary
structure. Displacement assays with palmitic acid or ligand 1 gave
a good indication that binding occurred at the calyx. However, the
slight differences in the CD spectra of the protein–ligand/complex
mixtures may reflect subtle differences of interaction among the
ligands/complexes.
Fig. 1 Transfer hydrogenation of TFAP catalysed by complexes or
artificial metalloenzymes. Conditions: [TFAP] = 5 or10 mM, 1 or
2
mol% catalyst.
(TFMBA) was monitored by reverse phase HPLC analysis of the
reaction mixtures (Fig. S2 and S3, ESIw). Conversion to the alcohol
product was quantitative after 24 h in the presence of the Rh(III)
complexes 3 and 6, whereas conversion reached 48% at best after
91 h with the Ru(II) complex 4 (Fig. 1). The benzene derivative 5
gave a lower conversion than the p-cymene complex 4, as previously
7,18
observed for analogous compounds and the lauric derivative 7
gave a slightly slower rate than the palmitic derivative 4. A
second set of kinetic experiments with the sole Rh(III) catalysts
indicated that conversion was quantitative in 2.5 h for complex
Having established that ligands and complexes were able to bind
b-LG, catalytic tests of transfer hydrogenation of TFAP were
performed in the presence of 1 or 2 mol% b-LG and 0.8 eq. of
complex 3, 4 or 6. Control experiments were also performed with
b-LG replaced by hen egg white lysozyme (HEWL) a protein of
similar size (14.3 instead of 18.6 kDa) but different folding or
avidin (Av), a protein displaying a b-barrel folding. Gradual
appearance of TFMBA was again monitored by HPLC analysis
of the reaction mixtures. TH of TFAP was observed in the presence
of b-LG with conversion reaching 68% for 6 C b-LG, 64% for
3 C b-LG and 14% for 4 C b-LG after 72 h (Fig. 1). Conversely,
conversion was quantitative in 24 h with Av and 6. Nevertheless,
TFBMA was formed at a considerably slower rate in the presence
of HEWL or Av and even more in the presence of b-LG with
respect to catalyst 6 alone (Fig. S15, ESIw). Indeed, the turnover
6
and 4 h for 3 (Fig. S4, ESIw). The two rhodium complexes
thus appeared much more reactive than the Ru complexes in the
catalysis of TH of TFAP.
Binding of ligands 1 or 2 and complexes 3–7 to b-LG was
next studied in phosphate buffer pH 7.5 by circular dichroism
(CD). In the near-UV region, the CD spectrum of b-LG
displayed a pattern of negative bands between 250 and
3
0
10 nm owing to its aromatic residues. Upon addition of
.8 molar eq. of ligand 1 or 2, a positive Cotton effect was
observed at 271 nm where the chromophoric moiety absorbs
Fig. 2 and Fig. S5, ESIw). Such induced CD of an achiral
(
guest in the presence of a protein host may result from (i) the
chiral conformation of the guest chromophore preferred by
the binding site or (ii) intermolecular chiral exciton
1
9
À1 À1
coupling between ligand and protein chromophores. In any
case, acquisition of optical activity provided evidence for
interaction between b-LG and 1 or 2. Furthermore, addition
of excess palmitic acid to the b-LG–1 mixture resulted in the
complete disappearance of the extrinsic CD band (Fig. S6, ESIw).
Thus, ligand 1 and palmitic acid share the same binding site. Upon
frequency was equal to 12.5 h for 6 alone, 8 h in the presence
À1
À1
of HEWL, 4 h in the presence of Av and only 1 h for 6Cb-LG
(Table S1, ESIw). The protein may act as an inhibitor by
2
1
competitive coordination to the metal centre. Alternatively,
reaction may be slowed down by limited access of the substrate
to the active site (or a combination of both).
This journal is c The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 11984–11986 11985

View Article Online
Table 1 Transfer hydrogenation of TFAP in the presence of
metalloenzymes
3 J. Steinreiber and T. R. Ward, Coord. Chem. Rev., 2008, 252,
751–766.
a
4
T. R. Ward, Acc. Chem. Res., 2011, 44, 47–57; M. E. Wilson and
G. M. Whitesides, J. Am. Chem. Soc., 1978, 100, 306–307; C.-C. Lin,
C.-W. Lin and A. S. C. Chan, Tetrahedron: Asymmetry, 1999, 10,
b
ee (%) [Conv.] (%)
Entry
1
Complex
Protein
Time (h)
3
4
6
6
6
6
6
b-LG
b-LG
b-LG
b-LG (A)
b-LG (B)
Av
48
72
48
72
72
24
9.5
16 [50]
20 [14]
17 [50]
26 [68]
24 [72]
0 [100]
0 [84]
1
887–1893.
c
2
5
X. Wu, C. Wang and J. Xiao, Platinum Met. Rev., 2010, 54, 3–19;
J. Canivet, G. Suss-Fink and P. Stepnicka, Eur. J. Inorg. Chem.,
3
4
5
6
7
2
007, 4736–4742; C. Leiva, C. Lo and R. H. Fish, J. Organomet.
Chem., 2010, 695, 145–150; I. Nieto, M. S. Livings, J. B. I. Sacci,
L. E. Reuther, M. Zeller and E. T. Papish, Organometallics, 2011,
HEWL
3
0, 6339–6342; P. Govindaswamy, J. Canivet, B. Therrien,
G. Suss-Fink, P. Stepnicka and J. Ludvik, J. Organomet. Chem.,
007, 692, 3664–3675; Y. Himeda, N. Onozawa-Kamatsuzaki,
a
b
Reaction conditions: [TFAP] = 5 mM, 2% catalyst. Determined
by chiral GC after extraction of the organics with diisopropyl ether or
by chiral HPLC. The (R)-configuration is assigned according to
2
H. Sugihara, H. Arakawa and K. Kasuga, J. Mol. Catal. A:
Chem., 2003, 195, 95–100; Y. Himeda, N. Onozawa-Komatsuzaki,
S. Miyazawa, H. Sugihara, T. Hirose and K. Kasuga, Chem.–Eur. J.,
c
ref. 23. [TFAP] = 10 mM, 1% catalyst.
2
008, 14, 11076–11081.
S. Ogo, T. Abura and Y. Watanabe, Organometallics, 2002, 21,
964–2969.
C. Romain, S. Gaillard, M. K. Elmkaddem, L. Toupet,
C. Fischmeister, C. M. Thomas and J.-L. Renaud, Organometallics,
6
7
Finally, the amount of (R)- and (S)-enantiomers was measured
by chiral GC analysis of extracts or chiral HPLC analysis of the
reaction medium (Table 1 and Fig. S16–S18, ESIw). Induction of
enantioselectivity for the (R)-isomer was observed with all the b-LG
constructs with subtle differences depending on the metal cofactor
and the protein host. Indeed, ee was independent of the fatty acid
chain (entries 1 and 3) but slightly higher for the Ru(II) complex
compared to the Rh(III) analogue derived from palmitic acid
2
2
010, 29, 1992–1995.
S. E. Clapham, A. Hadzovic and R. H. Morris, Coord. Chem. Rev.,
004, 248, 2201–2237.
8
2
9 M. Creus, A. Pordea, T. Rossel, A. Sardo, C. Letondor,
A. Ivanova, I. Le Trong, R. E. Stenkamp and T. R. Ward, Angew.
Chem., Int. Ed., 2008, 47, 1400–1404; C. Letondor, N. Humbert
and T. R. Ward, Proc. Natl. Acad. Sci. U. S. A., 2005, 102,
(
entries 1 and 2). Surprisingly, both genetic variants of b-LG yielded
higher ee’s than the protein sample comprising both variants
entries 3–5). We and others have observed by ESI-MS analysis
that the commercial b-LG sample (mixtures of genetic variants)
4
683–4887; C. Letondor, A. Pordea, N. Humbert, A. Ivanovna,
S. Mazurek, M. Novic and T. R. Ward, J. Am. Chem. Soc., 2006,
28, 8320–8328.
1
(
1
0 B. Talbi, P. Haquette, A. Martel, F. de Montigny, C. Fosse,
S. Cordier, T. Roisnel, G. Jaouen and M. Salmain, Dalton Trans.,
2010, 39, 5605–5607.
22
contained several lactosylated forms due to sample preparation.
Conversely, no lactosylation was detected for the pure variants.
Since no difference between the ee’s was observed between the two
variants taken separately (entries 4 and 5), heterogeneity of the
b-LG sample was solely responsible for the lower selectivity. When
b-LG was replaced by HEWL or Av, no enantioselectivity was
detected (entries 6 and 7), providing evidence that enantiopreference
arose from incorporation of the metal cofactors within b-LG.
In conclusion, a combination of easily accessible Ru(II)/
Rh(III) complexes derived from palmitic or lauric acid and
inexpensive, readily available and crystallisable b-lactoglobulin
yielded supramolecular constructs displaying catalytic activity
in the transfer hydrogenation of trifluoroacetophenone. Promising
enantioselectivities were observed as a result of insertion of the fatty
acid derivatives within the b-LG’s binding site. Current efforts are
aimed at solving the X-ray structure of metal cofactorCb-LG
constructs so as to gain insight into the interaction between both
species at the atomic level and eventually improve their activity and
selectivity. Extension of this work to other ketones and imines is
also under investigation.
11 P. Haquette, B. Talbi, L. Barilleau, N. Madern, C. Fosse and
M. Salmain, Org. Biomol. Chem., 2011, 9, 5720–5727.
1
2 N. Madern, B. Talbi and M. Salmain, Appl. Organomet. Chem.,
DOI: 10.1002/aoc.2929.
13 L. Panella, J. Broos, J. Jin, M. W. Fraaije, D. B. Janssen,
M. Jeromius-Stratingh, B. L. Feringa, A. J. Minnaard and
J. G. de Vries, Chem. Commun., 2005, 5656–5658; M. T. Reetz,
M. Rentzsch, A. Pletsch and M. Maywald, Chimia, 2002, 56,
7
21–723; M. T. Reetz, M. Rentzsch, A. Pletsch, M. Maywald,
P. Maiwald, J. J.-P. Peyralans, A. Maichele, Y. Fu, N. Jiao,
F. Hollmann, R. Mondiere and A. Taglieber, Tetrahedron, 2007,
3, 6404–6414; C. A. Kruithof, M. A. Casado, G. Guillena,
`
6
M. R. Egmond, A. van de Kerk-van Hoof, A. J. R. Heck, R. J.
M. Klein Gebbink and G. van Koten, Chem.–Eur. J., 2005, 11,
6
869–6877; L. Rutten, B. Wieczorek, J. Mannie, C. A. Kruithof,
H. P. Dijkstra, M. R. Egmond, M. Lutz, R. Gebbink, P. Gros and
G. van Koten, Chem.–Eur. J., 2009, 15, 4270–4280; B. Wieczorek,
A. Traff, P. Krumlinde, H. P. Dijkstra, M. R. Egmond, G. van
Koten, J. E. Backvall and R. Gebbink, Tetrahedron Lett., 2011, 52,
1
601–1604.
1
1
4 G. Kontopidis, C. Holt and L. Sawyer, J. Dairy Sci., 2004, 87, 785–796.
5 B. Y. Qin, M. C. Bewley, L. K. Creamer, H. M. Baker, E. N. Baker
and G. B. Jameson, Biochemistry, 1998, 37, 14014–14023.
16 D. Frapin, E. Dufour and T. Haertle, J. Protein Chem., 1993, 12,
43–449.
4
The Agence Nationale de la Recherche (ANR) is gratefully
acknowledged for financial support (project ‘‘Artzymes’’ grant
number ANR-11-BS07-027-01).
1
7 J. I. Loch, A. Polit, P. Bonarek, D. Olszewska, K. Kurpiewska,
M. Dziedzicka-Wasylewska and K. Lewinski, Int. J. Biol. Macromol.,
2012, 50, 1095–1102.
1
1
8 J. Canivet, L. Karmazin-Brelot and G. Suss-Fink, J. Organomet.
Chem., 2005, 690, 3202–3211.
9 F. Zsila, Z. Bikadi, I. Fitos and M. Simonyi, Curr. Drug Discovery
Technol., 2004, 1, 133–153.
Notes and references
1
T. Heinisch and T. R. Ward, Curr. Opin. Chem. Biol., 2010, 14,
20 G. Gupta, K. T. Prasad, B. Das and K. M. Rao, Polyhedron, 2010,
29, 904–910.
21 M. Poizat, I. Arends and F. Hollmann, J. Mol. Catal. B: Enzym.,
2010, 63, 149–156.
184–199; M. R. Ringenberg and T. R. Ward, Chem. Commun.,
2011, 47, 8470–8476; Y. Lu, Angew. Chem., Int. Ed., 2006, 45,
5588–5601; Y. Lu, N. Yeung, N. Sieracki and N. M. Marshall,
Nature, 2009, 460, 855–862; Y. Lu, Curr. Opin. Chem. Biol., 2005,
, 118–126.
P. J. Deuss, R. den Heeten, W. Laan and P. C. J. Kamer,
22 J. Leonil, D. Molle, J. Fauquant, J. L. Maubois, R. J. Pearce and
S. Bouhallab, J. Dairy Sci., 1997, 80, 2270–2281.
23 Z. Cakl, S. Reimann, E. Schmidt, A. Moreno, T. Mallat and
A. Baiker, J. Catal., 2011, 280, 104–115.
9
2
Chem.–Eur. J., 2011, 17, 4680–4698.
1
1986 Chem. Commun., 2012, 48, 11984–11986
This journal is c The Royal Society of Chemistry 2012