Highly efficient and site-selective phosphane modification of proteins through hydrazone linkage: Development of artificial metalloenzymes

Article abstract of DOI:10.1002/anie.201002174

A joint effort: A novel, highly efficient, and selective procedure for phosphane modification of proteins is reported (see scheme). This method involves cysteine modification with a maleimide containing a hydrazide functional group and subsequent hydrazone formation with phosphane aldehydes. Mono- and bidentate phosphane ligands were successfully coupled to several proteins, one of which was coordinated to rhodium to give an artificial metalloenzyme.

Full text of DOI:10.1002/anie.201002174

Angewandte
Chemie
DOI: 10.1002/anie.201002174
Metalloenzymes
Highly Efficient and Site-Selective Phosphane Modification of Proteins
through Hydrazone Linkage: Development of Artificial
Metalloenzymes**
Peter J. Deuss, Gina Popa, Catherine H. Botting, Wouter Laan,* and Paul C. J. Kamer*
Transition metal/phosphane complexes are well known to be
one of the most successful classes of homogeneous catalysts
for several industrial reactions such as hydroformylation and
hydrogenation.[1] However, for a large number of trans-
formations existing phosphane-based catalysts lack the
desired selectivity, which is a driving force for the continuous
development of novel phosphane ligand systems.[2] One
bioinspired approach is the creation of hybrid catalysts by
the introduction of synthetic catalysts into biopolymers.
Covalent or noncovalent merging of transition-metal catalysts
with proteins generates the opportunity of combining chem-
ical and genetic methods for performance optimization.[3]
Protein-based artificial metalloenzymes employing phos-
phane ligands have mainly been developed using noncovalent
anchoring approaches, using either antibodies raised against a
diphosphane/rhodium complex[4] or the very successful biotin-
avidin system,[5] mainly developed in the group of Ward[6] and
later subjected to directed evolution by the group of Reetz.[7]
Reetz et al. also covalently introduced a diphosphane ligand
in lipases through a phosphonate linkage, but hydrolytic
lability of the linker hampered application of these elegant
systems.[8] The drawback of the approaches outlined above is
that the protein structure space that can be combined with
phosphane ligands is very limited. Herein we report the
development of a site-specific covalent anchoring method
which will allow the introduction of phosphanes in a wide
variety of protein structures, demonstrated by modifying
three structurally different proteins with a small library of
phosphane ligands.
2L) domain of multifunctional enzyme type 2 containing an
18 ꢀ long and 9 ꢀ wide hydrophobic tunnel,[9] 2) photoactive
yellow protein (PYP),[10] which exhibits a small hydrophobic
pocket,[11] and 3) the sensing blue-light using FAD (BLUF)
domain of activator of photopigment and puc expression A
(AppA)[12] containing a cleft that binds flavins.[13] Several
single-cysteine variants for covalent modification with phos-
phanes were created by site-directed mutagenesis. Modifica-
tion of the nucleophilic thiol of a unique cysteine is a widely
employed strategy for site-selective bioconjugation.[14]
A
cysteine can be introduced at virtually any position within a
protein structure by site-directed mutagenesis and then
selectively modified, using for example, alkyl halides or
maleimides. This approach has already been successfully
applied for the development of artificial metalloenzymes
bearing ligands having donor atoms other than phosphorus.[15]
We recently reported the synthesis of artificial metal-
loenzymes by direct modification of the unique cysteine
moiety of PYP using phosphanes which contain an 1,1’-
carbonyldiimidazole (CDI) activated carboxylic acid.[16] How-
ever, for SCP-2L and AppA this method lacked the desired
chemoselectivity.[17a] Nonprotected phosphane-containing
maleimides cannot be synthesized because of the nucleophilic
character of the phosphane, which leads to the formation of
phosphonium salts and phosphorus ylides.[18] By using borane-
and sulfur-protected maleimido phosphanes, highly selective
cysteine modification was easily achieved, however, depro-
tection proved incompatible with or inefficient for these
protein constructs.[17b]
In seeking to exploit proteins to induce shape selectivity in
catalytic reactions, we selected several proteins having differ-
ent cavity architectures: 1) sterol carrier protein-2-like (SCP-
Hydrazone formation between a hydrazide and aldehyde
or ketone is a common bioconjugation method that proceeds
under mild reaction conditions in water.[19] The reaction has
also been reported to be compatible with phosphanes,[20]
prompting us to explore the use of this method for the
bioconjugation of phosphanes. The commercially available
cross-linker 1 was used for the cysteine-selective introduction
of a hydrazide, affording the protein–1 product (Scheme 1a,
Figure 1).
[*] P. J. Deuss, G. Popa, Dr. C. H. Botting, Dr. W. Laan,
Prof. Dr. P. C. J. Kamer
School of Chemistry, University of St. Andrews
North Haugh, St. Andrews (UK)
E-mail: wwl1@st-andrews.ac.uk
pcjk@st-andrews.ac.uk
Upon mixing of the hydrazide-modified proteins with
excess diphenylphosphane-containing benzaldehydes P1–P3
under inert atmosphere and stirring overnight, a quantitative
conversion into the corresponding phosphane-modified pro-
teins was observed by mass spectroscopy (ES+; Scheme 1b,
Figure 1, and Table 1 entries 1–12). The excess of insoluble
phosphane aldehyde was easily removed by centrifugation
and subsequent washing with buffer in a centrifugal concen-
trator. The procedure causes minimal loss of protein, as
determined by Bradford assays, resulting in typical modified-
[**] We thank the European Union (Marie Curie excellence grant MEXT-
2004-014320); NEST Adventure STREP Project artizymes (contract
no. FP6-2003-NEST-B3 15471); Network of Excellence Idecat
(Idecat-CT-2005-011730); COST action (CM0802 PhoSciNet),
EASTCHEM, and Sasol for funding. We also thank Dr. T. Glumoff
(University of Oulu, Finland) and Prof. Dr. K. J. Hellingwerf
(University of Amsterdam) for providing plasmids and the Well-
come Trust for funding the purchase of the instruments.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201002174.
Angew. Chem. Int. Ed. 2010, 49, 5315 –5317
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5315

Communications
Table 1: Results of bioconjugation of hydrazide-modified proteins with
phosphane aldehydes P1–P6[a]
Entry
Protein–1
P
Calculated
mass [Da]
Observed
mass [Da][b]
1
2
3
4
5
6
7
8
SCP-2L V83C–1
SCP-2L V83C–1
SCP-2L V83C–1
PYP–1
PYP–1
PYP–1
AppA Y21C–1
AppA Y21C–1
AppA Y21C–1
AppA Q63C–1
AppA Q63C–1
AppA Q63C–1
SCP-2L V83C–1
AppA Y21C–1
AppA Q63C–1
SCP-2L V83C–1
SCP-2LV 83C–1
AppA Q63C–1
P1
P2
P3
P1
P2
P3
P1
P2
P3
P1
P2
P3
P4
P5
P5
P5
P6
P6
13830.2
13830.2
13830.2
16317.8
16317.8
16317.8
15850.5
15850.5
15850.5
15885.5
15885.5
15885.5
14114.9
16085.8
16120.8
14065.3
14161.8
16216.9
13830.4Æ0.3
13830.8Æ0.6
13829.9Æ0.5
16317.0Æ1.1
16318.2Æ0.8
16317.4Æ0.6
15850.6Æ0.6
15850.9Æ0.6
15850.0Æ0.5
15900.2Æ0.8[c]
15884.9Æ0.6
15884.6Æ0.8
14130.1Æ1.0[c]
16083.4Æ2.8
16120.2Æ1.1
14065.9Æ0.7
14163.5Æ3.0
16216.2Æ0.9
9
10
11
12
13
14[d]
15[d]
16[e]
17[e]
18[d]
Scheme 1. a) Modification of unique cysteine-containing proteins with
hydrazide–maleimide 1. b) Bioconjugation of hydrazide-modified
proteins with phosphane aldehydes P1–P6.
[a] Reaction conditions: 2–10 equivalents phosphane (P1–P6) in aque-
ous buffer pH 6–7; full conversion of protein–1 was observed unless
stated otherwise. [b] Main peak for modified protein. [c] Main peak
corresponds to the peak for the protein containing the oxidized
phosphane. [d] Some protein–1 observed. [e] Protein–1 observed as
main peak and only weak signal for phosphane-modified protein–1 was
found.
Figure 1. Processed mass spectra (ES+) of SCP-2L V83C (light gray),
SCP-2L V83C–1 (dark gray), and SCP-2L V83C–1–P3 (black)
protein yields of over 95%. The 31P NMR spectra of PYP–1
and SCP-2LV83C–1 modified with P3 confirmed that the
modified proteins contained a free phosphane, showing broad
signals at d values ranging from À2 to À8 ppm (Figure 2 and
Figure 3a). In both spectra, the signal appears to consist of at
least two overlapping broad peaks. This signal might be the
result of the presence of different conformations of the
conjugates, caused by either the imine bond, the chiral center
formed by the maleimide–sulfide bond, or conformation of
the protein. The possibility that the different signals originate
from modification of different amino acid side chains of the
protein was ruled out by detailed analysis of tryptic digests.
Moreover, no protein modification was observed after treat-
ment of unmodified proteins with the phosphane aldehydes.
Also by employing Ellmanꢁs reagent, the absence of free thiol
groups in the conjugates was confirmed upon analysis by mass
spectroscopy (ES+).[21]
Figure 2. 31P{1H} NMR spectrum of PYP–1 modified with P3.
although the signal for the hydrazone-modified proteins was
often still be observed (Table 1 entries 13–18). However,
conversions were still found to be greater than 90% after
correction of the ES+ data for differences in ionization
efficiencies (entries 13–15 and 18). Unfortunately, several of
the conjugates containing diphosphanes P5 and P6 were
prone to protein precipitation, which hampered the determi-
nation of the conversion and severely compromised the yield
(entries 16 and 17). It appeared that the more reactive
phosphane-benzaldehyde-type substrates were more suitable
than the aliphatic phosphane aldehydes, as the products were
less prone to precipitation. Consequently, typical yields of
The method was extended to diphosphanes P4–P6, again
affording the desired conjugates with high selectivity,
5316
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5315 –5317

Angewandte
Chemie
opment holds great promise for the use of a wide range of
protein structures as templates for the preparation of
phosphane bearing artificial metalloenzymes.
Received: April 13, 2010
Published online: June 22, 2010
Keywords: bioconjugation · hydrazones · metalloenzymes ·
.
phosphanes · rhodium
[1] P. W. N. M. van Leeuwen, Homogeneous Catalysis: Understand-
ing the Art, Kluwer Academic Publishers, Dordrecht, 2004.
[2] a) J. A. Gillespie, D. L. Dodds, P. C. J. Kamer, Dalton Trans.
2010, 39, 2751; b) Phosphorus Ligands in Asymmetric Catalysis;
Synthesis and Applications (Ed.: A. Bꢂrner), Wiley-VCH,
Weinheim, 2008.
Figure 3. 31P{1H} NMR spectra of a) SCP-2L V83C–1–P3. b) SCP-
2L V83C–1–P3 treated with [Rh(acac)(CO)2]. Peaks between d=2 and
0 ppm are probably the result of protein phosphorylation as those are
also observed in the native protein (see the Supporting Information).
[3] a) J. Steinreiber, T. R. Ward, Coord. Chem. Rev. 2008, 252, 751;
b) T. Heinisch, T. R. Ward, Curr. Opin. Chem. Biol. 2010, 14, 184.
[4] H. Yamaguchi, T. Hirano, H. Kiminami, D. Taura, A. Harada,
Org. Biomol. Chem. 2006, 4, 3571.
[5] a) M. E. Wilson, G. M. Whitesides, J. Am. Chem. Soc. 1978, 100,
306; b) C.-C. Lin, C.-W. Lin, A. S. C. Chan, Tetrahedron:
Asymmetry 1999, 10, 1887.
[6] a) J. Pierron, C. Malan, M. Creus, J. Gradinaru, I. Hafner, A.
Ivanova, A. Sardo, T. R. Ward, Angew. Chem. 2008, 120, 713;
Angew. Chem. Int. Ed. 2008, 47, 701; b) G. Klein, N. Humbert, J.
Gradinaru, A. Ivanova, F. Gilardoni, U. E. Rusbandi, T. R.
Ward, Angew. Chem. 2005, 117, 7942; Angew. Chem. Int. Ed.
2005, 44, 7764.
[7] M. T. Reetz, J. J. P. Peyralans, A. Maichele, Y. Fu, M. Maywald,
Chem. Commun. 2006, 4318.
[8] M. T. Reetz, M. Rentzsch, A. Pletsch, M. Maywald, Chimia 2002,
56, 721.
protein–1–P4 (entry 13) were similar to those obtained with
P1–P3 (> 95%).
To demonstrate the formation of protein/phosphane/
metal complex, SCP-2LV83C–1–P3 was treated with one
equivalent of [Rh(acac)(CO)2] (acac = acetylacetonate). The
mass spectrum of the obtained mixture showed a signal
corresponding to the SCP-2LV83C–1–P3 modified with a
rhodium carbonyl fragment as main peak (Figure 4). The
31P NMR spectrum recorded after the reaction revealed a
[9] A. M. Haapalainen, D. M. F. van Aalten, G. Merilaeinen, J. E.
Jalonen, P. Pirilae, R. K. Wierenga, J. K. Hiltunen, T. Glumoff, J.
Mol. Biol. 2001, 313, 1127.
[10] T. E. Meyer, E. Yakali, M. A. Cusanovich, G. Tollin, Biochem-
istry 1987, 26, 418.
[11] G. E. O. Borgstahl, D. R. Williams, E. D. Getzoff, Biochemistry
1995, 34, 6278.
[12] S. Anderson, V. Dragnea, S. Masuda, J. Ybe, K. Moffat, C. Bauer,
Biochemistry 2005, 44, 7998.
[13] J. S. Grinstead, S.-T. D. Hsu, W. Laan, A. M. J. J. Bonvin, K. J.
Hellingwerf, R. Boelens, R. Kaptein, ChemBioChem 2006, 7,
187.
[14] Bioconjugate Techniques, 2nd ed. (Ed.: G. T. Hermanson),
Acadamic Press, San Diego, 2008.
[15] a) R. R. Davies, M. D. Distefano, J. Am. Chem. Soc. 1997, 119,
11643; b) M. T. Reetz, M. Rentzsch, A. Pletsch, A. Taglieber, F.
Hollmann, R. J. G. Mondiere, N. Dickmann, B. Hoecker, S.
Cerrone, M. C. Haeger, R. Sterner, ChemBioChem 2008, 9, 552.
[16] W. Laan, B. K. Muꢃoz, R. den Heeten, P. C. J. Kamer, Chem-
BioChem 2010, DOI: 10.1002/cbic.201000159.
Figure 4. Processed mass spectra (ES+) of SCP-2L V83C–1–P3 (black)
and SCP-2L V83C–1–P3 after treatment with [Rh(acac)(CO)2] (gray).
[17] Reference to this phenomenon can be found in references [4]
and [14] as well as R. den Heeten, B. K. Muꢃoz, G. Popa, W.
Laan and P. C. J. Kamer, Dalton Trans. 2010, DOI: 10.1039/
C0DT00239A.
[18] E. Hedaya, S. Theodoropulos, Tetrahedron 1968, 24, 2241.
[19] a) T. P. King, S. W. Zhao, T. Lam, Biochemistry 1986, 25, 5774;
b) A. Dirksen, P. E. Dawson, Bioconjugate Chem. 2008, 19, 2543.
[20] K. K. Hii, S. D. Perera, B. L. Shaw, M. Thornton-Pett, J. Chem.
Soc. Dalton Trans. 1992, 2361.
phosphorus shift of almost 50 ppm to give broad signals
between d = 45 and 43 ppm (Figure 3b). These results indi-
cate successful formation of a rhodium phosphane complex
attached to the protein.
In conclusion, we have developed the first highly efficient
and widely applicable method for cysteine-selective biocon-
jugation of phosphane ligands. Also a protein/phosphane/
rhodium complex was successfully synthesized. Whereas
Ward and others demonstrated the power of combining
phosphane/transition-metal catalysts with proteins, this devel-
[21] J. M. Chalker, C. S. C. Wood, B. G. Davis, J. Am. Chem. Soc.
2009, 131, 16346.
Angew. Chem. Int. Ed. 2010, 49, 5315 –5317
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5317