Synthesis of a heterogeneous artificial metallolipase with chimeric catalytic activity

Article abstract of DOI:10.1039/c5cc02450a

A solid-phase strategy using lipase as a biomolecular scaffold to produce a large amount of Cu²⁺-metalloenzyme is proposed here. The application of this protocol on different 3D cavities of the enzyme allows creating a heterogeneous artificial

Full text of DOI:10.1039/c5cc02450a

ChemComm
View Article Online
View Journal
COMMUNICATION
Synthesis of a heterogeneous artificial
metallolipase with chimeric catalytic activity†
Cite this:DOI: 10.1039/c5cc02450a
a
a
b
c
b
M. Filice,* O. Romero, J. Gutierrez-Fernandez, B. de las Rivas, J. A. Hermoso
´
´
a
Received 24th March 2015,
Accepted 27th April 2015
and J. M. Palomo
DOI: 10.1039/c5cc02450a
www.rsc.org/chemcomm
A solid-phase strategy using lipase as a biomolecular scaffold to However, the consequence of this catalytic site modification of
2+
produce a large amount of Cu -metalloenzyme is proposed here. enzymes relies on the inexorably suppression of native catalytic
2
The application of this protocol on different 3D cavities of the enzyme activity. Despite its enormous promising potential, many issues
allows creating a heterogeneous artificial metallolipase showing about artificial metalloenzymes should be addressed and solved
chimeric catalytic activity. The artificial catalyst was assessed in to be synthetically useful. Among all, good availability of the
Diels–Alder cycloaddition reactions and cascade reactions showing biomolecular scaffold, high activity and selectivity, recyclability
excellent catalytic properties.
and stability are some of the bottlenecks that still hamper their
definitive explosion. Moreover, although the vast majority of
3a,f
1
Since the first reports, the design of artificial metalloenzymes the reported artificial enzymes exhibited high catalytic activity
has rapidly been converted into an important topic in biological when used in their free homogeneous form and small scale, they
and inorganic chemistry due to their potential applications in were unstable to harsher reaction conditions (high temperatures,
2
synthetic chemistry, nanoscience and biotechnology. Several high solvent and substrate concentrations) being only useful
elegant and exciting strategies combining chemical and bio- under limited conditions without recyclability.
logical processes together with computational methods for the
Herein, we present a solid-phase strategy achieving a large amount
incorporation of abiotic metal cofactors within biological macro- of heterogeneous chimeric metalloenzyme with excellent catalytic
3
molecules have been developed. Generally, all the reported strate- properties (in terms of activity, selectivity, stability and reusability)
gies fall down inside one of these general classes: (i) supramolecular starting from a lipase as a biomolecular scaffold. Appling the
4
anchoring (by a strong affinity ligand, such as biotin), (ii) dative proposed protocol (based on the combination of molecular biology,
anchoring (by direct metal coordination with an aminoacid orienting immobilization strategies, solid-phase bioorganic modifi-
3f
residue) or (iii) covalent attachment (by selective anchoring of cations and bioinformatics tools) on different regions of hydrolase’s
5
a metal-binding ligand on an a-amino acid). Independently of structure, a highly active and selective artificial Diels–Alderase was
the used technique, the final goal of each strategy is to selectively finally created. The generation of the artificial machinery in an
modify precise zones of the biomolecular scaffold in order enzyme site different than the catalytic pocket permitted to retain
6
to convert them into efficient artificial catalytic sites. To this the native hydrolytic activity. Hence, a chimeric artificial enzyme
scope, the artificial activity is generally created inside natural showing dual catalytic activity (native plus artificial one) and
existing cavities of proteins or inside the catalytic pocket of useful in cascade reactions was satisfactorily created.
3
a
selected enzymes. Considering the well-known enantioselective
The Geobacillus thermocatenulatus lipase (GTL), for which the
7
properties of enzymes, especially the latter case is widely con- three-dimensional structure is known, was chosen as a protein
sidered one of the most used strategies in order to confer scaffold. As primary choice, the native catalytic cavity was selected
selectivity to a catalytically active but unselective metal complex. to develop and optimize the designed solid-phase strategy creating
the artificial catalytic site. To this scope, an engineered triple
a
b
c
mutant was created by replacement of two naturally occurring
cysteines (Cys 65 and Cys 296) by serine (GTL*) and by genetically
encoding one artificial Cys residue as replacement of catalytic
Departamento de Biocat a´ lisis, Instituto de Cat a´ lisis, CSIC, Marie Curie 2,
Campus UAM, 28049, Madrid, Spain. E-mail: marco.filice1@gmail.com
Departamento de Cristalograf ´ı a y Biolog ´ı a Estructural, Instituto de Qu ´ı mica-F ´ı sica
Rocasolano (CSIC), Serrano 119, 28006, Madrid, Spain
8
Ser114 (GTL*/114) (Fig. S1, ESI†). Hence, to efficiently create
Departamento de Biotecnolog ´ı a Microbiana, Instituto de Ciencia y Tecnolog ´ı a de
alimentos y Nutrici o´ n, (CSIC), Madrid, Spain
the artificial catalytic machinery inside the GTL structure, an easy-
handling and scalable solid-phase strategy was then designed
(Fig. 1). After E. coli expression and ultrasound-mediated cell
†
Electronic supplementary information (ESI) available: Synthetic experimental
details and additional figures and tables. See DOI: 10.1039/c5cc02450a
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
Table 1 DA cycloaddition reaction catalysed by different heterogeneous
a
artificial metallolipase derivatives
b
c
d
Entry Catalyst
C (%) endo/exo (%) ee (%)
2+
1
2
3
4
5
C-Cu
SB-eNH
SB-Lys-GTL*/114-C-Cu
SB-eNH -GTL*/196-C-Cu
SB-Lys-GTL*/196-C-Cu
28
51
99
92
94
98.7
nd
93.5
o5
o5
18
nd
92
2
+
+
2
-GTL*/114-C-Cu
2+
2
2
o5
2+
Fig. 1 Solid-phase synthesis of artificial metallolipase. (I) Purification by
reversible adsorption of the GTL* mutant on the solid support; (II) site-
directed chemical modification with 1 and insertion of catalytic metal; (III)
metallolipase desorption. Overall yield: 93%. Surface colour codes: green:
GTL’s lid residues; purple: terminal amino group.
98
a
All reactions were performed with 1 mM of 2, 10 mM of freshly
distilled 3, 100 mg of heterogeneous artificial biocatalyst (0.238 mol%)
at 4 1C for 72 h in 1 mL of 5% ACN in 10 mM MES buffered solution
pH 6. Conversion. The yield was calculated by HPLC. Determined by
b
c
d
HPLC. Determined by HPLC.
disruption, the GTL*/114 biocatalyst was selectively adsorbed
on commercial butyl-Sepharose (SP) (Table S1, ESI†) in quanti- strategies described above were also promoted on the Sepabeads
tative yield and high purity as demonstrated by SDS-PAGE support (SB, hydrophobic matrix) achieving the SB-eNH -GTL*/
2
2+
2+
(
SP-C4-GTL*/114) (Fig. S2, ESI†). After DTT-mediated thiol 114-C-Cu and SB-Lys-GTL*/114-C-Cu heterogeneous catalysts
reduction, in order to promote the selective conjugation to target in high overall yields (Table S1, ESI†). To test the effectiveness of
Cys, the SP-C4-GTL*/114 derivative was incubated in a solution with the applied strategy and characterize the catalytic potential of the
ten-fold excess of the phenanthroline-based ligand 1 that it is generated artificial site (especially in terms of stereo and enantio-
characterized by the presence of the iodoacetamide reactive group selectivity), the heterogeneous artificial biocatalysts were assessed
2,9
for selective conjugation to Cys (Fig. 1). The modification reaction in the Diels–Alder cycloaddition reaction (DA) of azachalcone 2
of the enzyme showed an almost quantitative coupling efficiency with cyclopentadiene 3 as the benchmark reaction (Table 1 and
2,3b,e
2+
(96% by Ellman’s assay) (Table S1, ESI†). The selective conjugation Table S2, ESI†).
Using a free C-Cu complex, 28% conversion
of 1 to the GTL*/114 mutant (GTL*/114-C) was further checked by and no enantioselectivity were observed (Table 1, entry 1) whereas
2+
MALDI and tryptic digestion confirming that only the target residue the soluble GTL*/114-C-Cu artificial metallolipase catalysed the
Cys114 was functionalized (Fig. S3 and S4, ESI†). To insert the reaction in 53% conversion, indicating that the DA reaction was
catalytic metal into the 3D pocket of lipase, the modified SP-C4- enhanced by the enzyme scaffold probably due to the protein-
GTL*/114-C derivative was then incubated with ten-fold excess of mediated stabilization of metal catalysts (Table S2, ESI†). The
2
+
2+
aqueous solution of Cu salt. After detergent-mediated desorption unmodified GTL*/114 mutant incubated with Cu salt did not
from the SP-C4 hydrophobic derivative, the artificial heterogeneous promote any conversion, excluding any possible unspecific
2
+
catalyst GTL*/114-C-Cu was finally created (Fig. 1; Table S1 and adsorption of catalytic metal on the protein scaffold (Table S2,
Fig. S1, ESI†). Thanks to this scalable solid phase strategy, about ESI†). The immobilization of the artificial enzyme on the hydro-
10 mg of pure artificial metallolipase per g of support were easily philic Sepharose support resulted in completely inactive catalysts
obtained in three practical steps and 93% overall yield (Fig. 1). (Table S2, ESI†). When immobilized on SB through the terminal
Compared to the majority of protocols reported in the literature, amino, the artificial catalyst showed results similar to those
in terms of operative steps, reproducibility and achieved amounts, achieved by the soluble artificial metallolipase whereas almost
this strategy is clearly found to be more reliable and efficient. quantitative conversion was achieved by using the GTL*/114-C-
2+
Once the key scalable solid-phase synthesis of artificial metallo- Cu artificial catalyst immobilized on Sepabeads by means of
enzymes is optimized, different orienting immobilization multipoint covalent linkage. In this case, the diastereoselectivity
protocols were considered in order to evaluate the immobiliza- (from 94 up to 98.7) and enantioselectivity (from 5 up to 18% ee)
tion effect on catalytic properties (activity, selectivity, stability, values were enhanced (Table 1, entry 3). Therefore, these results
handling and consequently its recyclability features). Specifi- confirmed the key role of the immobilization strategy design,
cally, two different orienting covalent immobilizations of GTL*/ the hydrophobic character of the supporting matrix as well as
2+
1
14-C-Cu were promoted on the Sepharose support (SP, hydro- the selected protein orientation in the immobilization process
philic matrix): the single covalent linkage through the terminal being two crucial parameters. In fact, it has been previously
amino group (placed on the opposite side of the enzyme structure described that the multipoint covalent immobilization through
compared to the catalytic site entrance position) (SP-eNH -GTL*/114- the Lys-richest region of the enzyme surface directly encloses also
2
2+
C-Cu ) and the multiple covalent linkage through the Lys-richest the GTL lid (a freely moving oligopeptide arm placed nearby the
2
+ 10
region of the protein surface (SP-Lys-GTL*/114-C-Cu ) (Fig. S5 catalytic pocket entrance) finally fixing its open and catalytically
1
0
and Table S1, ESI†). To evaluate the possible effect related to the active conformation on the support surface. Nevertheless, the
composition of supporting material, the orienting immobilization best retrieved enantiomeric value is still low.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015

View Article Online
ChemComm
Communication
Hence, in order to allow a better understanding of the low By applying the previously optimized solid-phase strategy, the
2
+
impact of the tridimensional enzyme cavity on the metal selectivity, artificial GTL*/196-C-Cu catalyst was prepared in high overall
a molecular docking study of ligand 1 covalently bound to Cys114 yields (91%) and purity (Fig. 1 and Fig. S1–S3 and Table S1, ESI†).
was attempted using the Genetic Optimization for Ligand Docking The immobilization on Sepabeads of artificial metallolipase by
11
(
GOLD) software (Fig. S6, ESI†). In order to discriminate the the single and multipoint covalent orienting protocols previously
potential ability of this artificial enzyme to selectively produce described permitted us to achieve, respectively, the SB-eNH
2
-
2
+
2+
diastereoisomers and/or enantiomers, the four potential products GTL*/196-C-Cu and SB-Lys-GTL*/196-C-Cu heterogeneous
a,b and 5a,b (Table 1) were modelled and docked. The best artificial catalysts in high yields (Fig. S5 and Table S1, ESI†).
retrieved solutions corresponded to both endo isomers with no The catalysts were then tested in the Diels–Alder reaction
4
2+
clear enantiopreference between 5a and 5b products (Fig. S7, (Table 1 and Table S2, ESI†). The soluble GTL*/196-C-Cu as well
2
+
ESI†). Hence, in silico simulations resulted in agreement with as SB-eNH -GTL*/196-C-Cu catalysts were not active in the cyclo-
2
2+
experimental evidence supporting the low enantioselectivity for addition reaction whereas the SB-Lys-GTL*/196-C-Cu catalyst
artificial DA reaction carried out in the native catalytic pocket showed a quantitative conversion with a diastereoselectivity value
(for more details, see the ESI†). In order to identify a more of 93.5% (endo) and, more interesting, an excellent enantio-
promising enzyme cavity able to confer higher enantioselectivity selectivity (ee) value of 92% (R,R) (Table 1, entry 5; Fig. S10,
to the Cu catalyst, we decided to evaluate other regions of the ESI†). A possible explanation for the extremely different results
GTL surface as possible targets to create an artificial catalytic site. related to the orienting immobilization strategy could rely
Recently, we have synthesized novel semisynthetic GTL enzymes by on the evidence that the 196 site is placed in a zone of the lid
site-specific incorporation of tailor-made peptides on a single-Cys (b-winged helix turn-helix domain) that is accessible to reaction
mutant (196 position, GTL*/196) placed in a well-defined pocket of medium only when GTL exists in its open and active conforma-
the enzyme lid. As a result, the selectivity of this modified enzyme tion. Hence, considering the effect related to the immobilization
8
in different chemical processes was highly improved. Thus, we orientation promoted by the one point covalent attachment
2
selected this position as a possible candidate to create a new through eNH together with the aqueous surrounding medium,
unnatural catalytic cavity into the protein. First, we checked the the lipase will mainly exist in its closed and catalytically inactive
viability of this zone by using the same docking analysis strategy conformation. As a consequence, the artificial catalytic site 196
1
2
previously described. Docking results indicated that, despite Cys will be secluded from the incoming reactants. On the other
96 being located in a wider cavity than 114, many protein residues hand, considering the previously detailed fixation and rigidification
1
2+
are involved in C-Cu ligand stabilization (Fig. 2 and Fig. S8 and of the open and catalytically active conformation of the GTL*/196
S9, ESI†). Among them, especially Arg93 plays a critical role in the structure on the SB surface, the orientation promoted by the
orientation of the ligand by establishing a cation–p interaction with multipoint covalent attachment seems to be critical in order to
the phenanthroline ring. This network of interactions provides a generate and stabilize the new tridimensional unnatural cavity
2+
precise orientation for ligand 1 that orients the reactive Cu to a responsible for the observed high selectivity of the metal catalyst
wide channel able to accept the incoming reactant 2 (Fig. 2 and (Table 1, entry 5; Fig. S10, ESI†). Also in this case, the observed
Fig. S8, ESI†). According to GOLD scores, the R,R-endo enantiomer 5a experimental results overlapped the in silico simulation.
is retrieved as the best solution whereas, due to the multiple clashes,
the S,S-endo and both exo isomers represent a poorly feasible 196-C-Cu heterogeneous artificial metalloenzyme has been
conformation (Fig. 2 and Fig. S9, ESI;† the docking simulation demonstrated by its reuse in three DA reaction cycles maintaining
The robustness and large-scale applicability of SB-Lys-GTL*/
2
+
was carried out using crystallographic data previously published almost intact its catalytic properties (Fig. S11, ESI†). Moreover, this
8
by ourselves, for more details, see the ESI†).
catalyst was applied in a 5-fold scaled-up DA reaction in the
Hence, encouraged by this favourable in silico simulation, presence of 50% (v/v) of acetonitrile and 5 times higher substrate
the single Cys GTL*/196 mutant was produced (Fig. S1, ESI†). concentration maintaining intact the activity and selectivity.
2
+
2+
Fig. 2 Docking result showing the anchoring site and stabilization of C-Cu and product 5a at the Cys196 site. (A) Overall view of the GTL*/196-C-Cu
2+
open structure. The lid zone is coloured in blue. (B) Details of C-Cu stabilization at site 196. Ligand (sticks coloured grey for C atoms and Cu orange
spheres) and relevant residues (coloured yellow for C atoms) involved in their stabilization are depicted and labelled. (C) Surface details of R,R-endo product
a and C-Cu stabilization at site 196. (D) Surface clashes (coloured in magenta) generated by the 5b S,S-endo product at site 196.
2+
5
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
in order to generate and stabilize a novel unnatural catalytic cavity
into the lipase structure. Consequently to the creation of the
artificial catalytic site in an enzyme cavity different than the native
catalytic pocket, a hybrid artificial enzyme with excellent chimeric
catalytic activity and useful in Diels–Alder cycloaddition reaction as
well as cascade reaction has been successfully generated and
characterized. Extending this strategy to other enzymes, ligands
and catalytic metals, we envisage the creation of a combinatorial
library of programmable artificial enzymes. Thus, newer and more
exciting horizons in metalloenzyme research, development and
Fig. 3 Domino one-pot synthesis of aminoarene 8 starting from nitroarene
2+
ester 6 catalyzed by SB-Lys-GTL*/196-C-Cu
.
Consequently to the creation of an unnatural and active practical application will be opened.
catalytic site in an enzyme domain not involved in the ‘wild-
This work has been sponsored by CSIC, by grants from the
type’ catalytic activity, a chimeric artificial metalloenzyme with Spanish Ministry of Economy and Competitiveness (BFU2011-
multicatalytic activities (native plus artificial one) and hence 25326; to J.A.H.) and the biomedicine program of government
3d,13
potentially useful in cascade reactions has been then achieved.
of Autonomous Community of Madrid (S2010/BMD-2457; to
To assess its potential, the multicatalytic activity of the new J.A.H.). The authors thank Sabino Veintemillas-Verdaguer
2+
immobilized metallolipase SB-Lys-GTL*/196-C-Cu was evaluated (ICMM-CSIC) for the assistance.
in the cascade synthesis of aminoarene 8 starting from nitroarene
ester 6 (Fig. 3). In the first step, the quantitative ‘native’ enzymatic
hydrolysis of 6 to obtain the chromogenic 4-nitrophenol 7 was
Notes and references
carried out (Fig. 3 and Fig. S12, ESI†). After that, in order to assess
1 M. E. Wilson and G. M. Whitesides, J. Am. Chem. Soc., 1978,
2+
100, 306.
also the artificial activity of the metal cofactor, the Cu -mediated
quantitative reduction of 7 to generate target product 8 was
subsequently promoted (Fig. 3). In fact, very recently, the catalytic
2
3
J. C. Lewis, ACS Catal., 2013, 3, 2954.
(a) D. Hamels and T. R. Ward, Comprehensive Inorganic Chemistry II,
2nd edn, Elsevier Ltd, New York, 2013, 737; (b) J. Bos and G. Roelfes,
Curr. Opin. Chem. Biol., 2014, 19, 135; (c) V. M. Robles, P. Vidossich,
A. Lled ´o s, T. R. Ward and J. D. Mar ´e chal, ACS Catal., 2014, 4, 833;
2+
activity of complexed Cu atoms in the synthesis of such com-
14
pounds has been demonstrated. As negative control, after the
first enzymatic hydrolysis, the reduction reaction was promoted
after chimeric catalyst filtration and using a SB-CHO-GTL*/196-C
(
d) M. Filice, M. Marciello, M. D. P. Morales and J. M. Palomo, Chem.
Commun., 2013, 49, 6876; (e) M. T. Reetz and N. Jiao, Angew. Chem.,
Int. Ed., 2006, 45, 2416; ( f ) I. Drienovsk ´a , A. Rioz-Mart ´ı nez,
A. Draksharapua and G. Roelfes, Chem. Sci., 2015, 6, 770; (g) J. Bos,
F. Fusetti, A. J. M. Driessen and G. Roelfes, Angew. Chem., Int. Ed.,
2012, 51, 7472.
C. M. Thomas and T. R. Ward, Chem. Soc. Rev., 2005, 34, 337.
D. K. Garner, L. Liang, D. A. Barrios, J.-L. Zhang and Y. Lu, ACS
Catal., 2011, 1, 1083.
2+
catalyst not incubated with Cu salt solution.
In both cases, no conversion was observed, confirming the
crucial role of the chimeric artificial metallolipase.
To establish a kinetic comparison of each catalytic process,
the reaction rate constant (k) was calculated for both reactions
4
5
6 T. Matsuo and S. Hirota, Bioorg. Med. Chem., 2014, 22, 5638.
7
C. Carrasco-L ´o pez, C. Godoy, B. de Las Rivas, G. Fern ´a ndez-Lorente,
J. M. Palomo, J. M. Guis ´a n, R. Fern ´a ndez-Lafuente, M. Mart ´ı nez-
Ripoll and J. A. Hermoso, J. Biol. Chem., 2009, 13, 4365.
(
Fig. S12, ESI†). The achieved values for the enzymatic ester
2+
hydrolysis and the Cu mediated nitro-to-amine reduction were
0
À1
.107 and 0.036 min respectively. Their comparison shows that the
8 O. Romero, M. Filice, B. de las Rivas, C. Carrasco-Lopez, J. Klett,
A. Morreale, J. A. Hermoso, J. M. Guisan, O. Abian and J. M. Palomo,
Chem. Commun., 2012, 48, 9053.
multicatalytic artificial metallolipase possesses comparable native
and artificial activities the last being a crucial requirement in order to
9
J. M. Chalker, G. J. L. Bernardes, Y. A. Lin and B. G. Davis,
Chem. – Asian J., 2009, 4, 630.
13,15,16
promote cascade reactions with these hybrid systems.
1
0 G. Fernandez-Lorente, C. A. Godoy, A. A. Mendes, F. Lopez-Gallego,
V. Grazu, B. de Las Rivas, J. M. Palomo, J. Hermoso, R. Fernandez-
Lafuente and J. M. Guisan, Biomacromolecules, 2008, 9, 2553.
In conclusion, we here presented a reliable solid-phase strategy
transforming a lipase into a multiactive artificial metalloenzyme.
Combining different protein engineering protocols (molecular 11 M. J. Hartshorn, M. L. Verdonk, G. Chessari, S. C. Brewerton,
W. T. M. Mooij, P. N. Mortenson and C. W. Murray, J. Med. Chem.,
biology, orienting immobilization, solid-phase bioorganic modifi-
2007, 50, 726.
cation and bioinformatics tools), a large amount of heterogeneous
artificial metallolipase with advanced catalytic properties (in
terms of activity, selectivity, stability and recyclability) and useful
in different synthetic reactions was satisfactorily obtained in
1
2 M. Filice, J. M. Guisan, M. Terreni and J. M. Palomo, Nat. Protoc.,
2012, 7, 1783.
3 M. Filice and J. M. Palomo, ACS Catal., 2014, 4, 1588.
4 U. Sharma, P. Kumar, N. Kumar, V. Kumar and B. Singh, Adv. Synth.
Catal., 2010, 352, 1834.
1
1
high overall yields. The careful selection of the modification site 15 V. K o¨ hler, Y. M. Wilson, M. D u¨ rrenberger, D. Ghislieri, E. Churakova,
T. Quinto, L. Kn ¨o rr, D. H ¨a ussinger, F. Hollmann, N. J. Turner and
T. R. Ward, Nat. Chem., 2013, 5, 93.
6 J. Muschiol, C. Peters, N. Oberleitner, M. D. Mihovilovic,
inside the 3D enzyme structure and the careful immobilization
protocol design (as fine combination of enzyme orientation and
1
supporting matrix selection) were shown to be critical parameters
U. T. Bornscheuer and F. Rudroff, Chem. Commun., 2015, 51, 5798.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015