Azurin-poly(N-isopropylacrylamide) conjugates by site-directed mutagenesis and their thermosensitive behavior in electron-transfer processes

Article abstract of DOI:10.1002/anie.200804440

(Chemical Equation Presented) Searching for "intelligence": Azurin-PNIPAM conjugates were prepared by site-directed mutagenesis followed by protein reconstitution by using imidazole-conjugated poly(N- isopropylacrylamides). The polymer-bound imidazole act

Full text of DOI:10.1002/anie.200804440

Communications
DOI: 10.1002/anie.200804440
Protein–Polymer Hybrids
Azurin–Poly(N-isopropylacrylamide) Conjugates by Site-Directed
Mutagenesis and their Thermosensitive Behavior in Electron-Transfer
Processes**
Nadine Rosenberger, Armido Studer,* Nobuyuki Takatani, Hiroshi Nakajima, and
Yoshihito Watanabe*
Peptide and protein polymer conjugates have recently
received increased attention.[1] The activity and structure of
the biohybrid materials can be modulated as a function of the
polymer attached. If “smart” stimuli-responsive polymers are
bound, the corresponding bioconjugates respond to external
stimuli such as light, pH, or temperature.[2] As “smart”
polymer, the thermoresponsive poly(N-isopropylacrylamide)
(PNIPAM)[3] has been used to prepare “intelligent” biohybrid
materials.[4] Below 328C the polymer is soluble in water in the
chain-extended hydrated state. Above the so-called lower
critical solution temperature (LCST) polymer chains collapse
and turn into their hydrophobic state. This phase switch can
be used to induce biological responses of PNIPAM–biomol-
ecule conjugates.[2,4,5]
Various approaches for conjugation of proteins with
polymers have been followed:[1] 1) reaction of polymers
bearing reactive end groups with proteins, 2) reaction of
polymerization initiators with proteins and subsequent poly-
merization, and 3) cofactor reconstitution by using cofactor-
terminated polymers.[6] If more than one reactive site in the
protein is accessible to the activated polymer or initiator,
nonspecific multiple modification is possible by applying
approaches (1) or (2). This problem can be circumvented by
using bioengineering to prepare modified proteins bearing
defined reactive sites for selective attachment of the polymer
or initiator.[1f,2,7]
is a well-characterizied blue copper protein involved in
bacterial electron-transfer chains.[8] Recent progress in elec-
trochemistry has enabled the efficient electrical communica-
tion between azurin and an electrode. Consequently, azurin is
considered to be a potential component in biodevices.[9]
Canters and co-workers have shown that the imidazole of
His117, which acts as a Cu ligand in the active site of the blue
copper protein azurin of Pseudomonas aeruginosa, can be
replaced by imidazole derivatives in His117Gly mutants.[10]
The activity of imidazole-reconstituted mutants could be
restored. Guided by these studies we decided to synthesize
imidazole-terminated PNIPAM by nitroxide-mediated radi-
cal polymerization[11,12] for preparation of “smart” mutated
azurin–PNIPAM conjugates by reconstitution of the
His117Gly azurin mutant with imidazole-terminated
PNIPAM. In this way PNIPAM will be introduced to the
active site of azurin (Figure 1) and is therefore expected to be
located between azurin and its redox partner protein,
cytochrome c. The presence of PNIPAM should selectively
alter the protein–protein interaction as a function of temper-
ature.
Herein we present a conceptually different approach for
preparation of protein–polymer conjugates with azurin, which
[*] N. Rosenberger, Prof. Dr. A. Studer
Organisch-Chemisches Institut
Westfꢀlische Wilhelms-Universitꢀt
Corrensstrasse 40, 48149 Mꢁnster (Germany)
Fax: (+49)251-83-36523
Figure 1. The active site of a reconstituted azurin–PNIPAM conjugate.
E-mail: studer@uni-muenster.de
We show herein by kinetic measurements that PNIPAM
bound to the active site of azurin can serve as a modulator of
the electron transfer (ET) from cytochrome c to the azurin–
PNIPAM conjugate as a function of temperature. The fact
that short-chain PNIPAM cannot affect the ET process
provides a clue of how PNIPAM disturbs the interaction
between cytochrome c and azurin.
The synthesis of imidazole-terminated PNIPAM is de-
scribed in Scheme 1. Radical bromination (Br2, hn, CCl4) of 4-
ethylbenzonitrile gave 1, which was converted to alkoxyamine
2 by using a known procedure.[5] Reduction of 2 with lithium
aluminum hydride (LAH) and subsequent coupling with
Dr. N. Takatani, Dr. H. Nakajima, Prof. Dr. Y. Watanabe
Department of Chemistry, Graduate School of Science
Nagoya University, Chikusa-ku, 464-8602 (Japan)
E-mail: yoshi@nucc.cc.nagoya-u.ac.jp
[**] We thank the DFG and the JSPS for funding (International Research
Training Group Mꢁnster/Nagoya). Dr. Kunishige Kataoka (Kana-
zawa University) is acknowleged for providing an expression system
of azurin-1 and advice on cultivation. We thank Prof. Dr. S. Yamago
and Y. Ukai (Kyoto University) for conducting GPC measurements of
PNIPAM and S. Ozawa (Nagoya University) for the reconstitution of
azurin-PNIPAM(34200).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200804440.
1946
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1946 –1949

Angewandte
Chemie
lyzed. The reconstitution protocol had to be slightly modified
for the longer-chain PNIPAM derivative (34200 gmolꢀ1); a
20-fold excess of PNIPAM was used in that case. The
reconstituted holoprotein precipitated when heated to 408C
and could be separated readily from non-reconstituted
apoprotein (about 43% was reconstituted, see the Supporting
Information). Since azurin–PNIPAM(34200) precipitated
above the LCST, it was not suitable for ET measurements
and was not further used in these studies.
UV/Vis spectra of azurin–PNIPAM (3800 gmolꢀ1) and
azurin–PNIPAM (9800 gmolꢀ1) were recorded at 258C and
358C and compared with the spectra of azurin WT and the
azurin mutant reconstituted with 4-methylimidazole
(Figure 2). The ratio of absorbance at 280 nm to the
absorbance at 620 nm is a measure for successful reconstitu-
tion. This ratio was nearly identical for the mutants and azurin
WT which indicated complete reconstitution of the type-1
copper center. The azurin–PNIPAM hybrids showed no
spectral change when heated above the LCST of the polymer.
Therefore, expulsion of the imidazole–PNIPAM ligand
(ligand exchange with water) of the mutant at higher temper-
ature was ruled out.7
Scheme 1. Synthesis of imidazole–PNIPAM conjugates 6 with
Mn =3800–34200 gmolꢀ1. IIDQ=isobutyl-1,2-dihydro-2-isobutoxy-1-
quinoline carboxylate.
trityl-protected imidazole propionic acid 4[13] gave radical
initiator 5 after deprotection. Nitroxide-mediated polymeri-
zation was conducted in C6D6 with 0.5—2 mol% 5 at 1258C in
1
sealed tubes. Conversion was determined by H NMR spec-
troscopy; the mean molecular weight (3800–34200 gmolꢀ1)
and polydispersity index (PDI = 1.14–1.39) were determined
by mass spectrometry and gel permeation chromatography
(GPC) (see the Supporting Information).[14,15] LCSTs of the
imidazole–PNIPAM conjugates 6 were found to lie at around
328C as measured by 1H NMR spectroscopy (imid–PNIPAM
3800 gmolꢀ1: 30.58C; imid–PNIPAM 9800 gmolꢀ1: 31.68C;
imid–PNIPAM 34200 gmolꢀ1: 32.68C; see the Supporting
Information).[16]
The mutagenesis of His117 (base code CAC) into Gly
(base code GGC) was carried out by polymerase chain
reaction of azurin plasmid DNA by using primers with the
necessary mutation. Azurin His117Gly apoprotein was
expressed and purified in analogy to the protocol of Canters
et al. for expression of azurin wild type (WT).[17] The purity of
the apoprotein was checked by SDS-PAGE and UV/Vis
spectroscopy (see the Supporting Information).
The reconstitution of azurin with
6
(3800 and
9800 gmolꢀ1) was carried out by a modified literature
protocol:[18] Cu(NO3)2·3H2O (1 equiv) was added to a solu-
tion of mutant apoprotein in 20 mm MES-NaOH (pH 6.0)
leading to an increase of absorbance at 420 nm due to
formation of type-2 copper azurin. A threefold excess of 6 and
NaCl (0.2m) were then added, and the solution was stirred at
48C until formation of the characteristic blue type-1 copper
complex was complete. After desalting and precipitation of
free 6 the resulting azurin–PNIPAM conjugates were ana-
Figure 2. a) UV/Vis spectra of WT azurin (black line) and an azurin
mutant reconstituted with 4-methylimidazole (red line). b) UV/Vis
spectra of azurin reconstituted with PNIPAM(3800) recorded at 258C
(black line) and at 358C (red line); UV/Vis spectra of azurin recon-
stituted with PNIPAM(9800) recorded at 258C (green line) and at
358C (blue line).
Angew. Chem. Int. Ed. 2009, 48, 1946 –1949
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1947

Communications
Reconstituted mutants were also analyzed by EPR
spectroscopy. Azurin WT and the azurin–PNIPAM(9800)
mutant showed very similar spectra (Figure 3). The coordi-
nation geometry at the Cu center of the mutant reconstituted
with imidazole–PNIPAM was virtually identical to that of
wild-type azurin as judged based on Ak(58ꢀ10ꢀ4 cmꢀ1), gk-
(2.248), and g?(2.055) values.[10a,18] Similar results were also
obtained for azurin–PNIPAM(3800) (see the Supporting
Information).
Figure 3. EPR spectra of the reconstituted azurin–PNIPAM(9800)
mutant (a) and WT azurin (b) at 80 K.
Recorded CD spectra of WTazurin and azurin–PNIPAM-
(9800) at 258C and 358C indicated that the mutant virtually
retained the original protein conformation of WT azurin at
258C as well as at 358C (see the Supporting Information).
Finally, we studied ET processes between reduced cyto-
chrome c and WT azurin and the reconstituted mutants at 25
and 358C. We used cytochrome c552 from Thermus thermo-
philus HB8 as the electron donor instead of the natural redox
partner, cytochrome c from Pseudomonas aeruginosa,
because cytochrome c552 shows high conformational stability
under the experimental conditions. Moreover, the hydro-
phobic patch located at the surface around the heme in
cytochrome c552 was expected to force hydrophobic interac-
tions with azurin to allow electron transfer. Kinetic experi-
ments were conducted in a UV cell under strictly anaerobic
conditions to avoid atmospheric oxidation of reduced cyto-
chrome c552 (see the Supporting Information). To this end, the
oxidized form of a reconstituted azurin mutant was added to a
solution of reduced cytochrome c552, and the disappearance of
the Soret band at 416 nm was recorded as a function of time at
the given temperature (Figure 4).
Figure 4. Electron-transfer processes from reduced cytochrome c552 to
oxidized WT azurin (a), to the azurin–PNIPAM(3800) mutant (b), and
to the azurin–PNIPAM(9800) mutant (c) at 258C (blue dots) and at
358C (red dots).
solvated PNIPAM tail has a small but measurable effect on
the activity of the protein. At 358C a slightly larger rate
constant was obtained (3.2 ꢀ 105 mꢀ1 sꢀ1). In this case the
PNIPAM tail was most likely too small to induce any
thermosensitive behavior.[20] Indeed, the ET rate constant
could be altered as a function of temperature for azurin–
PNIPAM(9800). At 258C we determined an apparent rate
constant of 2.5 ꢀ 105 mꢀ1 sꢀ1 (similar to the value obtained for
the mutant bearing the smaller tail). However, at 358C, the
For WT azurin the apparent rate constant, k = 9.8 ꢀ
105 mꢀ1 sꢀ1, obtained at 258C is in agreement with literature
values.[19] A similar k value was obtained at 358C (1.0 ꢀ
106 mꢀ1 sꢀ1). For azurin–PNIPAM(3800) ET transfer was
about an order of magnitude slower at 258C (1.9 ꢀ
105 mꢀ1 sꢀ1) than with the wild-type azurin, showing that the
rate constant dropped to a value of 6.2 ꢀ 104 mꢀ1 sꢀ1.[21]
A
fourfold lower activity was obtained for the 9800 mutant at
higher temperature.
1948
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1946 –1949

Angewandte
Chemie
11288; S. Kulkarni, C. Schilli, B. Grin, A. H. E. Mꢃller, A. S.
We believe that structural changes of the PNIPAM tail
(for polymers with Mn around 10000 gmolꢀ1) on a temper-
ature increase from 25 to 358C influences protein recognition
of cytochrome c552 with azurin. Hydrophobic interactions at
the active site of azurin, where the polymer is located in our
mutant, are known to be important for the interaction of
cytochrome c and azurin in the wild-type enzymes. The
collapsed PNIPAM tail would partially block the interaction
site on azurin, which significantly reduces the accessibility of
cytochromec552 to the site, while hydrated PNIPAM provides
space for cytochrome c552 to access the site for ET although
full contact between the proteins is restricted. The different
results obtained with PNIPAM(3800) and PNIPAM(9800)
presumably reflect to what extent the polymer tail covers the
hydrophobic protein/protein recognition site.
In conclusion, we have shown that polymer–azurin
hybrids can be prepared by His117Gly mutations and
subsequent reconstitution with imidazole-terminated poly-
mers. The polymer acts as a ligand for the copper ion in the
active site. This is the first report on the preparation of a
polymer–protein biohybrid by polymer–ligand reconstitution
in a metalloprotein. Importantly, this approach allows the
site-specific attachment of the polymer. Moreover, the
polymer is located at the active site of the metalloprotein
which is important for tuning enzyme activity. Structure and
activity were restored upon protein reconstitution. For
biohybrid materials bearing larger PNIPAM tails enzyme
activity could be altered as a function of temperature. The ET
rate constant decreased by a factor of 4 at higher temperature.
For protein–polymer hybrids bearing smaller PNIPAM tails
thermosensitivity was not observed.
Hoffman, P. S. Stayton, Biomacromolecules 2006, 7, 2736; C.
Boyer, J. Liu, V. Bulmus, T. P. davis, C. Barner-Kowollik, M. H.
Stenzel, Macromolecules 2008, 41, 5641.
[5] K. Molawi, A. Studer, Chem. Commun. 2007, 5173.
[6] M. J. Boerakker, N. E. Botterhuis, P. H. H. Bomans, P. M.
Frederik, E. M. Meijer, R. J. M. Nolte, N. A. J. M. Sommerdijk,
Chem. Eur. J. 2006, 12, 6071, and references therein.
[7] a) P. S. Stayton, T. Shimoboji, C. Long, A. Chilkoti, G. Chen,
J. M. Harris, A. S. Hoffman, Nature 1995, 378, 472; b) T.
Shimoboji, E. Larenas, T. Fowler, A. S. Hoffman, P. S. Stayton,
Bioconjugate Chem. 2003, 14, 517; c) A. Deiters, T. A. Cropp, D.
Summerer, M. Mukherji, P. G. Schultz, Bioorg. Med. Chem. Lett.
2004, 14, 5743; d) B. R. Griffith, B. L. Allen, A. C. Rapraeger,
L. L. Kiessling, J. Am. Chem. Soc. 2004, 126, 1608; e) K. L.
Heredia, D. Bontempo, T. Ly, J. T. Byers, S. Halstenberg, H. D.
Maynard, J. Am. Chem. Soc. 2005, 127, 16955.
[8] E. Vijgenboom, J. E. Busch, G. W. Canters, Microbiology 1997,
143, 2853.
[9] a) R. Rinaldi, A. Biasco, G. Maruccio, V. Arima, P. Visconti, R.
Cingolani, P. Facci, F. De Rienzo, R. Di Felice, E. Molinari, M. P.
Verbeet, G. W. Canters, Appl. Phys. Lett. 2003, 82, 472; b) G.
Maruccio, A. Biasco, P. Visconti, A. Bramanti, P. P. Pompa, F.
Calabi, R. Cingolani, R. Rinaldi, S. Corni, R. Di Felice, E.
Molinari, M. R. Verbeet, G. W. Canters, Adv. Mater. 2005, 17,
816.
[10] a) T. den Blaauwen, M. ven de Kamp, G. W. Canters, J. Am.
Chem. Soc. 1991, 113, 5050, and references therein; b) T. E.
de Jongh, A.-M. M. van Roon, M. Prꢄdencio, M. Ubbink, G. W.
Canters, Eur. J. Inorg. Chem. 2006, 3861, and references therein.
[11] a) C. J. Hawker, A. W. Bosman, E. Harth, Chem. Rev. 2001, 101,
3661; b) A. Studer, Chem. Eur. J. 2001, 7, 1159; c) A. Studer,
Chem. Soc. Rev. 2004, 33, 267; d) A. Studer, T. Schulte, Chem.
Rec. 2005, 5, 27.
[12] Controlled radical polymerization for the synthesis of biocon-
jugates: J. Nicolas, G. Mantovani, D. M. Haddleton, Macromol.
Rapid Commun. 2007, 28, 1083.
[13] G. Lelais, P. Micuch, D. Josien-Lefebvre, F. Rossi, D. Seebach,
Helv. Chim. Acta 2000, 65, 2229.
[14] a) T. Schulte, K. O. Siegenthaler, H. Luftmann, M. Letzel, A.
Studer, Macromolecules 2005, 38, 6833; b) S. Kehr, H. Luftmann,
e-Polym. 2007, 010.
[15] Moleluar weights and PDIs determined by GPC agreed well
with the data obtained by mass spectrometry (see the Supporting
Information).
[16] A. Larsson, D. Kuckling, M. Schꢁnhoff, Colloids Surf. A 2001,
190, 185.
[17] M. van de Kamp, F. C. Hali, N. Rosato, A. Agro, G. W. Canters,
Biochim. Biophys. Acta Bioenerg. 1990, 1019, 283.
[18] T. den Blaauwen, G. W. Canters, J. Am. Chem. Soc. 1993, 115,
1121.
[19] P. Rosen, I. Pecht, Biochemistry 1976, 15, 775.
[20] The LCST of the protein–polymer conjugates (3800 and
9800 gmolꢀ1) could not be determined. We assume the LCSTs
to be comparable with those for the imidazole-terminated
PNIPAMs (around 328C) according to a publication by Stayton
et al. In this article the LCST of a streptavidin–PNIPAM
conjugated was comparable with the LCST of free PNIPAM. P.
Stayton, T. Shimboji, C. Long, A. Chilkoti, G. Chen, J. M. Harris,
A. S. Hoffman, Nature 1995, 378, 472.
Received: September 9, 2008
Revised: November 4, 2008
Published online: February 6, 2009
Keywords: bioconjugates · cytochrome · enzymes ·
.
point mutations · polymers
[1] a) M. A. Gauthier, H.-A. Klok, Chem. Commun. 2008, 2591;
b) I. S. Carrico, Chem. Soc. Rev. 2008, 37, 1423; c) H. G. Bꢁrner,
Macromol. Chem. Phys. 2007, 208, 124; d) H. G. Bꢁrner, H.
Schlaad, Soft Matter 2007, 3, 394; e) K. L. Heredia, H. D.
Maynard, Org. Biomol. Chem. 2007, 5, 45; f) P. Thordarson,
B. L. Droumaguet, K. Velonia, Appl. Microbiol. Biotechnol.
2006, 73, 243; g) H.-A. Klok, J. Polym. Sci. Part A 2005, 43, 1;
h) J. Nicolas, G. Mantovani, D. M. Haddleton, Macromol. Rapid
Commun. 2007, 28, 1083.
[2] A. S. Hoffman, P. S. Stayton, Macromol. Symp. 2004, 207, 139.
[3] M. Heskins, J. E. Guillet, J. Macromol. Sci. Part A 1968, 2, 1441.
Review: H. G. Schild, Prog. Polym. Sci. 1992, 17, 163.
[4] PNIPAM–protein conjugates: D. Bontempo, H. D. Maynard, J.
Am. Chem. Soc. 2005, 127, 6508; H. D. Maynard, K. L. Heredia,
R. C. Li, D. P. Parra, V. Vꢂzquez-Dorbatt, J. Mater. Chem. 2007,
17, 4015; C. Boyer, V. Bulmus, J. Liu, T. P. Davis, M. H. Stenzel,
C. Barner-Kowollik, J. Am. Chem. Soc. 2007, 129, 7145; P. De, M.
Li, S. R. Gondi, B. S. Sumerlin, J. Am. Chem. Soc. 2008, 130,
[21] Control of enzyme activity by responsive redox polymers: B.
Nagel, A. Warsinke, M. Katterle, Langmuir 2007, 23, 6807.
Angew. Chem. Int. Ed. 2009, 48, 1946 –1949
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 1949