Enhanced catalytic activity in organic solvents using molecularly dispersed haemoglobin-polymer surfactant constructs

Article abstract of DOI:10.1039/c3cc46101g

The surface of haemoglobin (Hb) is chemically modified to produce molecular dispersions of discrete core-shell Hb-polymer surfactant bionanoconjugates in water and organic solvents. The hybrid nanoconstructs exhibit peroxidase-like catalytic activity with enhanced turnover rates compared with native Hb in water. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3cc46101g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Enhanced catalytic activity in organic solvents using
molecularly dispersed haemoglobin–polymer
surfactant constructs†
Cite this: Chem. Commun., 2013,
49, 9561
Received 9th August 2013,
Accepted 3rd September 2013
Yixiong Zhang, Avinash J. Patil,* Adam W. Perriman* and Stephen Mann
DOI: 10.1039/c3cc46101g
www.rsc.org/chemcomm
The surface of haemoglobin (Hb) is chemically modified to produce use of dendritic polymers6 or copolymers,7 and encapsulation in
molecular dispersions of discrete core–shell Hb–polymer surfactant silica gels.8 In particular, the covalent coupling of poly(ethylene)
bionanoconjugates in water and organic solvents. The hybrid nano- glycol (PEG) to the surface of enzymes (PEGylation) has been
constructs exhibit peroxidase-like catalytic activity with enhanced used extensively to increase the efficiency of biocatalysis in
turnover rates compared with native Hb in water.
organic solvents.9 Although the amphiphilic nature of PEG
increases the range of available dielectric media, the resulting
Enzymes have evolved to optimise their efficiency in aqueous dispersions often contain large aggregates of the PEG-modified
environments, and as a consequence exhibit a strong dynamical enzyme molecules that curtail intrinsic homogenous catalytic
and conformational dependence on the nature of molecular performance.
interactions with their solvent envelope. Accordingly, the catalytic
Herein, we describe a novel method using a solvent-free liquid
efficiency of most enzymes dispersed in organic solvents is protein10 to produce molecular dispersions of an enzymatically
significantly retarded, as solvent exposure commonly leads to active haemoglobin (Hb)–polymer surfactant conjugate in a range
protein aggregation and denaturation.1 These inhibitory pro- of organic solvents. The dispersed constructs comprise a pre-
cesses can be partly rationalized by the substantially different served globular protein core surrounded by a compact polymer
dielectric constants of organic solvents when compared with surfactant corona, and exhibit enhanced catalytic turnover rates
water. The latter can be considered a poor solvent for proteins in depending on the dielectric constant of the organic solvent.
the sense that it favours the compact structure of the globular The methodology involves the carbodiimide (EDC)-mediated
fold. Conversely, polar organic solvents such as ethanol have a cationization of surface accessible amino acid carboxylate side
propensity to strip away tightly bound water molecules and chains to produce a super-charged polycationic haemoglobin
solvate the hydrophobic core of a protein, which in turn stabi- (C-Hb; zeta potential, +52 mV; native Hb, +7 mV), followed by
lises the denatured state.2 Proteins dispersed in highly hydro- electrostatic tethering of the PEG-based anionic surfactant
phobic solvents, however, commonly remain folded and can be (C9H19-C6H4-(OCH2CH2)20O(CH2)3SO3ꢀK+) (S) to yield a charge-
extremely stable, although in many cases there is a loss of activity neutral aqueous conjugate ([C-Hb][S]) (Fig. 1), which when
due to the reduction in the population of functional conformers lyophilised and then annealed at 30 1C produced a viscous
due to molecular aggregation.3
solvent-free liquid (ESI† Methods). Matrix-assisted laser deposition/
Enzyme reactions in organic solvents are commonly under- ionisation time-of-flight (MALDI-TOF) mass spectroscopy (Fig. S1,
taken using dispersions of freeze-dried powders, where the ESI†) gave a cationization efficiency of ca. 96%, which corresponded
activity arises predominately from aggregates rather than discrete
molecular entities. Accordingly, a number of approaches have
been developed to increase the dispersibility and hence enzyme
activity in organic reaction media. These include covalent
modifications of enzyme structures using lipids4 or surfactants,4
entrapment of enzymes in reverse micelles or colloidosomes,5
Centre for Organized Matter Chemistry, School of Chemistry, University of Bristol,
Bristol BS8 1TS, UK. E-mail: chawp@bristol.ac.uk, avinash.patil@bristol.ac.uk;
Fax: +44-(0)117-9290509; Tel: +44-(0)117-3317215
Fig. 1 Schematic illustration showing protein cationization using N,N0-dimethyl-
1,3-propanediamine (DMPA) followed by electrostatic coupling of poly(ethylene
glycol) 4-nonylphenyl 3-sulfopropyl ether, potassium salt (S) to yield a charge
neutral stoichiometric conjugate [C-Hb][S]. Blue and red regions on the surface of
the haemoglobin structures show the anionic and cationic residues respectively,
and S molecules are represented using purple tubes.
† Electronic supplementary information (ESI): Experimental procedures, MALDI-
TOF, DLS, UV-Vis and Michaelis–Menten plots are provided. See DOI: 10.1039/
c3cc46101g
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 9561--9563 9561

View Article Online
Communication
ChemComm
Fig. 2 DLS particle size distributions at 25 1C for (a) aqueous solutions of native Hb
(blue), cationized Hb (green), polymer surfactant S (black) and [C-Hb][S] (purple),
and (b) [C-Hb][S] in acetonitrile (orange), ethanol (grey), and isopropanol (red).
to the modification of approximately 44 carboxylates per protein,
and equated to a total of 100 surfactant binding sites.
The solvent-free [C-Hb][S] liquid could be readily dispersed in
water, acetonitrile, ethanol and isopropanol, and the resulting clear
solutions remained stable for six months without any observable
signs of aggregation (Fig. S2, ESI†). Conversely, aqueous solutions
of native Hb aggregated immediately when added to any of the
organic solvents (Fig. S2, ESI†). Dynamic light scattering (DLS)
profiles gave average hydrodynamic diameters of 6.3 or 7.7 nm for
aqueous dispersions of C-Hb or [C-Hb][S] (Fig. 2a), consistent with
the formation of a molecularly dispersed protein–polymer surfac-
tant construct with a densely packed polymer surfactant corona.
In contrast, DLS profiles of the [C-Hb][S] conjugate dispersed in
Fig. 3 (a) CD spectra, (b) protein secondary structure content, and (c) UV-Vis
spectra for native Hb in water (blue), C-Hb in water (green), and [C-Hb][S] in
water (purple), acetonitrile (orange), ethanol (grey) and isopropanol (red).
acetonitrile, ethanol or isopropanol gave particle size distributions alpha helical content (ca. 80%) after protein cationization or
with larger mean diameters of 8.7, 8.9 or 8.9 nm respectively surfactant conjugation (Fig. 3b), indicating that the polymer
(Fig. 2b). The data confirmed that the [C-Hb][S] conjugate was coronal layer effectively screened the globular protein core from
dispersed without significant aggregation as a discrete molecular adverse interactions with the organic solvents. UV-Vis spectra of
entity, and that the polymer surfactant shell was extended in the aqueous solutions of native Hb or C-Mb showed Soret bands at
organic solvents, presumably via effective solvation and extension 406 or 412 nm, and Q bands at 503, 541, 578, 631 or 532 and
of the hydrophobic nonylphenyl surfactant tails.
562 nm, respectively (Fig. 3c). The data were consistent with a
Circular dichroism (CD) and UV-Vis spectroscopies were used mixture of a five-coordinate high spin and six-coordinate hydroxyl-
to investigate the protein structure associated with the [C-Hb][S] bound low spin states for native Hb,11 and a six-coordinate low
conjugate after dissolution in aqueous or organic media. CD spin configuration with a strongly associated water molecule or
spectra from all the [C-Hb][S] dispersions showed characteristic bis-ligated distal histidine for the cationized protein.11 In contrast,
a-helical features at approximately 195, 208 and 222 nm, origi- UV-Vis spectra of the [C-Hb][S] conjugate were consistent with a
nating from the amide p–p* and n–p* transitions (Fig. 3a). There six-coordinate low spin state when dispersed in water, or in
was only minimal variation between the CD traces, and decon- organic solvents such as isopropanol or acetonitrile. However,
volution of the spectra showed no significant reduction in the when dispersed in ethanol, the protein–polymer surfactant
c
9562 Chem. Commun., 2013, 49, 9561--9563
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
conjugate displayed a UV-Vis spectrum similar to that of substrate affinity in acetonitrile to the absence of hydrogen
aqueous Hb, albeit with a slightly blue-shifted Soret band bonding in this reaction medium. An increase in the propensity
(403 nm) (Fig. 3c). UV-Vis spectra from free hemin dispersed for hydrogen bonding would also explain the smaller Km value
in isopropanol, ethanol or acetonitrile (Fig. S3, ESI†) gave Soret for ethanol compared with isopropanol.
band peak positions at 401, 400, and 379 nm respectively,
indicating that exposure of [C-Hb][S] to the organic solvents of the surface of Hb can be used to prepare a protein–polymer
did not result in haem removal. surfactant conjugate that can be molecularly dispersed in water or
In conclusion, we have demonstrated that nanoscale engineering
Native Hb can be used to spontaneously cleave H2O2 to a range of organic solvents without loss of structure or enzymatic
catalyse the oxidation of organic pollutants such as phenols function. Significantly, the presence of a condensed amphiphilic
and polycyclic aromatic hydrocarbons,12 and this incipient polymer-surfactant corona not only inhibited protein aggregation
peroxidase-like activity was probed by monitoring the oxidation in organic solvents, but also increased the peroxidase-like activity
of o-phenylenediamine (OPD) to phenazine by Hb and [C-Hb][S] of the [C-Hb][S] construct in water, acetonitrile, ethanol or iso-
under aqueous and non-aqueous conditions at a H2O2 concen- propanol. Although we have used haemoglobin as an archetype of
tration of 15 mM. The catalytic turnover rates (kcat) and Michaelis– a globular protein with enzymatic activity, we anticipate that
Menten constants (Km) were determined by fitting the initial rates our methodology could be successfully applied to industrially
using the Michaelis–Menten equation (Fig. S4 and S5, ESI†), and relevant enzymes, such as lipases, amylases or proteases to yield
the resulting kinetic parameters of native Hb and [C-Hb][S] are molecularly dispersed functional bioconjugates for application
listed in Table S1 (ESI†). Significantly, an increase in kcat was in a wide range of reaction media.
evident for the [C-Hb][S] conjugate when dispersed in water or a
We thank the EPSRC (Cross-disciplinary Interfaces Program),
range of organic solvents compared with the peroxidase-like ERC (Advanced Grant) and University of Bristol, UK for Financial
activity of unmodified Hb. For the organic solvents, the kcat support for AWP, SM and AJP respectively.
values increased in the sequence isopropanol o ethanol o
Notes and references
acetonitrile, with a kcat value approximately six-fold higher in
acetonitrile when compared with the native protein. The posi-
tive correlation with increasing solvent polarity was consistent
with associated changes in the dielectric constant of the active
site resulting in a reduction of the energy of the activated
substrate–enzyme complex. Moreover, given the similarity in
the secondary structure and absorption band structure of the
[C-Hb][S] conjugates in the organic solvents, it is possible that
the dependence on increased solvent polarity was driven by
an increase in the distribution of functional motions in the
protein structure.13 In contrast, an approximately 30% decrease
in the kcat values compared with native Hb in water was observed
when unmodified Hb was dispersed in the above organic
solvents. We attribute this to deactivation of a proportion of
the protein molecules due to the inaccessibility of the active sites
accompanying aggregation, as well as distortions in the surface
hydration layer and denaturation of the enzyme in the presence
of apolar solvents.14
1 (a) A. Zaks and A. M. Klibanov, Science, 1984, 224, 1249; (b) E. P. Hudson,
R. K. Eppler and D. S. Clark, Curr. Opin. Biotechnol., 2005, 16, 637.
2 (a) R. Affleck, Z. F. Xu, V. Suzawa, K. Focht, D. S. Clark and J. S.
Dordick, Proc. Natl. Acad. Sci. U. S. A., 1992, 89, 1100; (b) M. E. Diaz-
Garcia and M. J. Valencia-Gonzalez, Talanta, 1995, 42, 1763.
3 (a) A. Zaks and A. M. Klibanov, J. Biol. Chem., 1988, 263, 3194;
(b) D. S. Clark, Philos. Trans. R. Soc. London, Ser. B, 2004, 359,
1299.
4 (a) Y. Okahata, Y. Fujimoto and K. Ijiro, Tetrahedron Lett., 1988,
29, 5133; (b) Y. Okahata and K. Ijiro, Bull. Chem. Soc. Jpn., 1992,
65, 2411; (c) M. Goto, N. Kamiya, M. Miyata and F. Nakashio,
Biotechnol. Prog., 1994, 10, 263; (d) V. M. Paradkar and J. S.
Dordick, J. Am. Chem. Soc., 1994, 116, 5009.
5 (a) R. Biswas and S. K. Pal, Chem. Phys. Lett., 2004, 387, 221;
(b) C. Z. Wu, S. Bai, M. B. Ansorge-Schumacher and D. Y. Wang,
Adv. Mater., 2011, 23, 5694; (c) R. Krishna Kumar, M. Li, S. N. Olof,
A. J. Patil and S. Mann, Small, 2013, 9, 357.
6 (a) M. E. Cosulich, S. Russo, S. Pasquale and A. Mariani, Polymer,
2000, 41, 4951; (b) O. Yemul and T. Imae, Biomacromolecules, 2005,
6, 2809.
7 V. Depp, J. L. Kaar, A. J. Russell and B. S. Lele, Biomacromolecules,
2008, 9, 1348.
8 J. Kim and J. W. Grate, Nano Lett., 2003, 3, 1219.
9 (a) H. Gaertner, T. Watanabe, J. V. Sinisterra and A. Puigserver,
J. Org. Chem., 1991, 56, 3149; (b) Z. Yang, M. Domach, R. Auger,
F. X. Yang and A. J. Russell, Enzyme Microb. Technol., 1996, 18, 82;
(c) M. J. Hernaiz, J. M. Sanchez-Montero and J. V. Sinisterra, Enzyme
Microb. Technol., 1999, 24, 181.
The peroxidase-like activity of native Hb and [C-Hb][S] in
water gave Km values of 0.34 ꢁ 0.03 and 0.056 ꢁ 0.002 mM
respectively (Table S1, ESI†). The increase in substrate affinity
for [C-Hb][S] was attributed to an increase in steric restriction by
the densely packed polymer surfactant corona. Under aqueous 10 (a) A. W. Perriman and S. Mann, ACS Nano, 2011, 5, 6085;
(b) A. W. Perriman, H. Colfen, R. W. Hughes, C. L. Barrie and
S. Mann, Angew. Chem., Int. Ed., 2009, 48, 6242; (c) A. W. Perriman,
A. P. S. Brogan, H. Colfen, N. Tsoureas, G. R. Owen and S. Mann,
conditions, the PEG domain of the polymer surfactant is expected
to form rigid hydrogen bonded networks with water molecules,
and the hydrophobic nonylphenol moiety to be buried within the
surrounding coronal layer. Considerably smaller Km values
were also obtained for assays performed on [C-Hb][S] in ethanol
(0.049 ꢁ 0.004 mM) and isopropanol (0.19 ꢁ 0.03 mM),
indicating high affinity for the substrate molecules under these
conditions. Conversely, a significant increase in Km (0.30 ꢁ 0.02 mM)
was observed for the peroxidase-like activity of [C-Hb][S] in
acetonitrile. As the [C-Hb][S] construct showed a similar expan-
sion in the polymer surfactant corona when dispersed in the
Nat. Chem., 2010, 2, 622; (d) A. J. Patil, N. McGrath, J. E. Barclay,
D. J. Evans, H. Colfen, I. Manners, A. W. Perriman and S. Mann,
Adv. Mater., 2012, 24, 4557.
11 (a) M. Ikedasaito, H. Hori, L. A. Andersson, R. C. Prince, I. J.
Pickering, G. N. George, C. R. Sanders, R. S. Lutz, E. J. Mckelvey
and R. Mattera, J. Biol. Chem., 1992, 267, 22843; (b) Q. C. Li and
P. A. Mabrouk, JBIC, J. Biol. Inorg. Chem., 2003, 8, 83; (c) M. P. Roach,
A. E. Pond, M. R. Thomas, S. G. Boxer and J. H. Dawson, J. Am.
Chem. Soc., 1999, 121, 12088.
12 Q. Wang, Q. Gao and J. Shi, J. Am. Chem. Soc., 2004, 126, 14346.
13 M. J. Liszka, M. E. Clark, E. Schneider and D. S. Clark, Annu. Rev.
Chem. Biomol. Eng., 2012, 3, 77.
above organic solvents (Fig. 2b), we attributed the reduced 14 H. Ogino and H. Ishikawa, J. Biosci. Bioeng., 2001, 91, 109.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 9561--9563 9563