A nearly isosteric photosensitive amide-backbone substitution allows enzyme activity switching in ribonuclease S

Article abstract of DOI:10.1021/ja069048o

ψ[CS-NH]⁴-RNase S, a site specific modified version of RNase S obtained by thioxylation (O/S exchange) at the Ala⁴-Ala ⁵- peptide bond, was used to evaluate the impact of protein backbone photoswitching on bioactivity. ψ[CS-NH]⁴-RNase S was yielded by recombination of the S-protein and the respective chemically synthesized thioxylated S-peptide derivative. Comparison with RNase S revealed similar thermodynamic stability of the complex and an unperturbed enzymatic activity toward cytidine 2′,3′-cyclic monophosphate (cCMP). Reversible photoisomerization with a highly increased cis/trans isomer ratio of the thioxopeptide bond of ψ[CS-NH]⁴-RNase S in the photostationary state occurred under UV irradiation conditions (254 nm). The slow thermal reisomerization (t^1/2 = 180 s) permitted us to determine the enzymatic activity of cis ψ[CS-NH]⁴-RNase S by measurement of inital rates of cCMP hydrolysis. Despite thermodynamic stability of cis ψ[CS-NH]⁴-RNase S, its enzymatic activity is completely abolished but recovers after reisomerization. We conclude that the thioxopeptide bond modified polypeptide backbone represents a versatile probe for site-directed photoswitching of proteins.

Full text of DOI:10.1021/ja069048o

Published on Web 03/31/2007
A Nearly Isosteric Photosensitive Amide-Backbone
Substitution Allows Enzyme Activity Switching in
Ribonuclease S
†
†
†
‡
Dirk Wildemann, Cordelia Schiene-Fischer, Tobias Aum u¨ ller, Annett Bachmann,
‡
†
,†
Thomas Kiefhaber, Christian L u¨ cke, and Gunter Fischer*
Contribution from the Max-Planck Research Unit for Enzymology of Protein Folding,
Weinbergweg 22, D-06120 Halle/Saale, Germany, and Biozentrum, UniVersity of Basel,
Klingelbergstrasse 50/70, CH-4056 Basel, Switzerland
Received December 18, 2006; E-mail: fischer@enzyme-halle.mpg.de
4
Abstract: ψ[CS-NH] -RNase S, a site specific modified version of RNase S obtained by thioxylation (O/S
4
5
exchange) at the Ala -Ala - peptide bond, was used to evaluate the impact of protein backbone
4
photoswitching on bioactivity. ψ[CS-NH] -RNase S was yielded by recombination of the S-protein and the
respective chemically synthesized thioxylated S-peptide derivative. Comparison with RNase S revealed
similar thermodynamic stability of the complex and an unperturbed enzymatic activity toward cytidine 2′,3′-
cyclic monophosphate (cCMP). Reversible photoisomerization with a highly increased cis/trans isomer ratio
4
of the thioxopeptide bond of ψ[CS-NH] -RNase S in the photostationary state occurred under UV irradiation
1/2
conditions (254 nm). The slow thermal reisomerization (t ) 180 s) permitted us to determine the enzymatic
4
activity of cis ψ[CS-NH] -RNase S by measurement of inital rates of cCMP hydrolysis. Despite
4
thermodynamic stability of cis ψ[CS-NH] -RNase S, its enzymatic activity is completely abolished but
recovers after reisomerization. We conclude that the thioxopeptide bond modified polypeptide backbone
represents a versatile probe for site-directed photoswitching of proteins.
Introduction
practical approach involves using substituted azobenzene moi-
eties as site-specific side chain or backbone modifications that
Numerous attempts have been made to reversibly control the
bioactivity of a protein by introducing a photoresponsive element
into the protein chain.1-3 The success of the approach was
demonstrated by generation of photoswitchable variants of
allow polypeptide chains to sample two regions of the confor-
mational space determined by the azobenzene isomerization.3
However, conformational flexibility of the linkages coupling
the photoresponsive element to the polypeptide chain is likely
to dampen effects of the azobenzene group isomerization and
,14-16
enzymes,4 neuronal ion channels,
-7
8-10
and an E. coli tran-
scriptional activator.11 Typically, azobenzene and spiropyrane
groups can be easily introduced in polypeptides and convey them
to respond to irradiation with a conformational change. The cis/
trans isomerization of azobenzene moieties involves a fairly
large conformational change that is rapid on the time scale of
17
makes effects difficult to interpret. Consequently, the photot-
riggered cis/trans isomerization of the phenylazophenylalanine
residue at position 11 of RNase S, which is near its active site,
did not lead to substantial effects on enzyme activity.14 The
hypothesis was made that the task of faithfully converting a
light signal into a predetermined structural change of a protein
is best accomplished by photoswitching the conformation of
the backbone directly.
1
2,13
conformational interconversions of a polypeptide chain.
A
†
‡
Max-Planck Research Unit for Enzymology of Protein Folding.
University of Basel.
(
1) Pieroni, O.; Fissi, A.; Angelini, N.; Lenci, F. Acc. Chem. Res. 2001, 34,
-17.
2) Borisenko, V.; Woolley, G. A. J. Photochem. Photobiol. A 2005, 173, 21-
It is worth noting that the natural protein backbone itself
provides a conformational change that has some features in
common with the cis/trans isomerization of the azobenzene
moiety. The trans to cis interconversion of the peptide bond
9
(
2
8.
(
3) Moroder, L. J. Pept. Sci. 2005, 11, 258-261.
(
4) Aizawa, M.; Namba, K.; Suzuki, S. Arch. Biochem. Biophys. 1977, 182,
3
05-310.
(CRsC(dO)sNRsC′R; R) H, alkyl) shortens the distance
(5) Weston, D. G.; Kirkham, J.; Cullen, D. C. Biochim. Biophys. Acta 1999,
1
428, 463-467.
(
6) Muranaka, N.; Hohsaka, T.; Sisido, M. FEBS Lett. 2002, 510, 10-12.
(12) Wachtveitl, J.; Sp o¨ rlein, S.; Satzger, H.; Fonrobert, B.; Renner, C.; Behrendt,
R.; Oesterhelt, D.; Moroder, L.; Zinth, W. Biophys. J. 2004, 86, 2350-
2362.
(13) Lednev, I. K.; Ye, T. Q.; Hester, R. E.; Moore, J. N. J. Phys. Chem. 1996,
100, 13338-13341.
(14) James, D. A.; Burns, D. C.; Woolley, G. A. Protein Eng. 2001, 14, 983-
991.
(15) Woolley, G. A. Acc. Chem. Res. 2005, 38, 486-493.
(16) Renner, C.; Moroder, L. ChemBioChem 2006, 7, 868-878.
(17) Kumita, J. R.; Flint, D. G.; Smart, O. S.; Woolley, G. A. Protein Eng.
2002, 15, 561-569.
(
7) Nakayama, K.; Endo, M.; Majima, T. Bioconjugate Chem. 2005, 16, 1360-
1
366.
(
8) Banghart, M.; Borges, K.; Isacoff, E.; Trauner, D.; Kramer, R. H. Nat.
Neurosci. 2004, 7, 1381-1386.
(
9) Volgraf, M.; Gorostiza, P.; Numano, R.; Kramer, R. H.; Isacoff, E. Y.;
Trauner, D. Nat. Chem. Biol. 2006, 2, 47-52.
(
(
10) Chambers, J. J.; Banghart, M. R.; Trauner, D.; Kramer, R. H. J.
Neurophysiol. 2006, 96, 2792-2796.
11) Bose, M.; Groff, D.; Xie, J.; Brustad, E.; Schulz, P. G. J. Am. Chem. Soc.
2
006, 128, 388-389.
4910
9
J. AM. CHEM. SOC. 2007, 129, 4910-4918
10.1021/ja069048o CCC: $37.00 © 2007 American Chemical Society

Ribonuclease S Backbone Photoswitch
A R T I C L E S
between the CR and the CR′ atom by approximately 0.8 Å, and
versions (in the case of the recombination of modified S-peptides
18,19
14,31-34
the cis/trans ratio increases in the photostationary state.
This
with S-protein, the term RNase S’ is also in use).
We
isomer-specific chain displacement, in some proteins, propagates
into the distal part of the molecule.20 Unfortunately, the
wavelength range (<220 nm) of peptide absorbance, which must
be used for photoswitching, is not compatible with the photo-
chemical stability of proteins. In addition, fast re-equilibration
rates prevent the photoinduced cis population from applying
isomer-specific effects on a coupled reaction. We have been
developing a new approach to the peptide bond as a photore-
sponsive element of a protein that utilizes features of the near-
present results herein that suggest that backbone peptide bond
thioxylation located in the N-terminal region of the S-peptide
provides a reversible on/off photoswitch for the enzymatic
activity of RNase S.
4
We provide evidence that the incorporation of the trans Ala -
5
ψ[CS-NH]-Ala - segment into the S-peptide chain does not
4
5
impair the formation of the noncovalent (Ala -ψ[CS-NH]-Ala )
4
S-peptide/S-protein complex (ψ[CS-NH] -RNase S) and its
hydrolytic activity against cytidine 2′,3′-cyclic monophosphate
4
isosteric thioxopeptide bond (sC(dS)sNRs; R ) H, alkyl)
(cCMP). Irradiation of a ψ[CS-NH] -RNase S sample resulted
substitution.2
1-23
The process of photoisomerization of short
in a markedly increased nonequilibrium population of cis ψ-
4
thioxopeptides is completed within the picosecond time scale
[CS-NH] -RNase S that retained less than 1% cCMP cleaving
at a high quantum yield and does not exhibit a more prolonged
activity. The data of this study are indicative of a major chemical
change in the active site due to a predefined one-bond
phototrigger at a remote site.
reaction time for a longer peptide chain.2
4,25
Biologically, the
O/S exchange causes minor effects as long as a bioreaction does
not directly involve the respective amide carbonyl group. In
particular, it does not greatly affect secondary structure forma-
tion, and the bioactivity of the thioxopeptide derivative of
Experimental Section
General Considerations. All chemicals and reagents were purchased
with the highest purity commercially available (Fluka, Merck, Aldrich,
Sigma, Bachem, Novabiochem, Advanced ChemTech). The S-protein
(Grade XII) was purchased from Sigma. Peptide synthesis was
performed using a Syro II multiple peptide synthesizer (MultiSynTech,
Germany). Analytical HPLC was performed using solvent gradients
of B (ACN, 0.1% TFA) in A (H O, 0.1%TFA) on a Pharmacia LKB
2
HPLC system in combination with the peak integration software “PC
Integration Pack” (Kontron Instruments) and a LiChroCART 125-4 RP
ligands was found to be similar to that of the oxopeptide
congener.2
6-29
However, the lack of methods is evident for the
site-specific thioxylation of a mature polypeptide chain and the
chemical synthesis of an enzyme thioxylated at a predefined
position. In addition, classical strategies of chemical ligation,
which are normally used for synthesis of site-specific modified
polypeptides, are not suitable because the presence of thioxo-
peptide bonds is not compatible with the subsequent synthesis
of the thioester moiety which is necessary for the ligation
procedure.
One way to reconcile the latter drawback would be for a
protein which constitutes a bioactive molecule by noncovalently
recruiting an oligopeptide. In this case, the oligopeptide can be
prepared in the form of a mono- or polythioxylated molecule.
We were tempted to utilize RNase S to prepare an enzymati-
cally active protein containing a single thioxopeptide bond.
RNase S is composed of two RNase A fragments obtained by
subtilisin cleveage, the S-peptide (consisting of residues 1 to
8
(5 µm) column (Merck). Detection was performed at 220 nm, and
the solvent flow rate was 1 mL/min. Preparative HPLC was performed
on a Hibar RT 250-25 LiChrosorb RP-select B (7 µM) column using
an HPLC system from SYKAM (Germany). Concentrations of the
4
S-protein, S-peptide, and ψ[CS-NH] -S-peptide were determined using
-
1
-1
-1
the extinction coefficients ꢀ ) 9560 M cm , ꢀ257.5 ) 195 M
2
80
-1
-
1
-1
cm , and ꢀ266 ) 12 000 M cm , respectively.
Peptide Synthesis. Whereas the S-peptide was synthesized using a
standard protocol of solid-phase peptide synthesis, a special protocol
4
was used for the ψ[CS-NH] -S-peptide and Ac-Ala-ψ[CS-NH]-Phe-
NH
as described elsewhere.35 Peptides were purified by preparative
2
HPLC.
20), and the S-protein (consisting of residues 21 to 124) that
+
Ac-Ala-ψ[CS-NH]-Phe-NH
H ] found m/z ) 294.0; t
2
: [M + H ] calcd m/z ) 294.1; [M
constitute the active enzyme after mixing both components in
solution.30 It is well-known that RNase S dissociates and
associates in a pH-dependent manner allowing the substitution
of the natural S-peptide of RNase S by chemically modified
+
+
R
, 19.8 min (10-25% B in 30 min).
+
+
S-peptide: [M + H ] calcd m/z ) 2148.1; [M + H ] found m/z )
4
2
148.4; t
R
+
, 18.85 min (5-35% B in 30 min). ψ[CS-NH] -S-peptide:
+
[M + H ] calcd m/z ) 2164.1; [M+H ] found m/z ) 2164.1; t
R
, 18.86
min (5-35% B in 30 min).
(
18) Tsukahara, T.; Ishiura, S.; Sugita, H. Int. J. Biochem. 1991, 23, 79-83.
19) Wang, Y.; Purrello, R.; Spiro, T. G. J. Am. Chem. Soc. 1989, 111, 8274-
Isothermal Titration Calorimetry. Calorimetric experiments were
performed using a VP-ITC titration calorimeter (MicroCal, Inc,
Northampton, MA) at 25 °C. All solutions were degassed under vacuum
prior use. Solutions of the respective S-peptide derivative and S-protein
were dialyzed against 50 mM sodium acetate buffer (pH 6.0, 100 mM
NaCl). The volume of the protein solution (50 µM) in the sample cell
was 1.4 mL. The injection syringe was filled with 300 µL solution of
the respective S-peptide derivative (200 µM). Each titration experiment
consisted of a single 1 µL injection followed by 16 identical 5 µL
injections of the peptide solution.
(
8
276.
(
20) Reimer, U.; Fischer, G. Biophys. Chem. 2002, 96, 203-212.
21) Zhao, J.; Wildemann, D.; Jakob, M.; Vargas, C.; Schiene-Fischer, C. Chem.
Commun. 2003, 22, 2810-2811.
(
(
22) Frank, R.; Jakob, M.; Thunecke, F.; Fischer, G.; Schutkowski, M. Angew.
Chem. 2000, 112, 1163-1165; Angew. Chem., Int. Ed. 2000, 39, 1120-
1
122.
(
23) Zhao, J.; Micheau, J. C; Vargas, C.; Schiene-Fischer, C. Chem.sEur. J.
2
004, 10, 6093-6101.
(
24) Helbing, J.; Bregy, H.; Bredenbeck, J.; Pfister, R.; Hamm, P.; Huber, R.;
Wachtveitl, J.; De Vico, L.; Olivucci, M. J. Am. Chem. Soc. 2004, 126,
8
823-8834.
(
(
(
(
(
(
25) Satzger, H.; Root, C.; Gilch, P.; Zinth, W.; Wildemann, D.; Fischer, G. J.
Phys. Chem. B 2005, 109, 4770-4775.
26) Schutkowski, M.; Neubert, K.; Fischer, G. Eur. J. Biochem. 1994, 221,
(31) Richards, F. M.; Vithayathil, P. J. J. Biol. Chem. 1959, 234, 1459-1465.
(32) Wyckoff, H. W.; Tsernoglou, D.; Hanson, A. W.; Knox, J. R.; Lee, B.;
Richards, F. M. J. Biol. Chem. 1970, 245, 305-328.
(33) Liu, D.; Karanicolas, J.; Yu, C.; Zhang, Z.; Woolley, G. A. Biorg. Med.
Chem. Lett. 1997, 7, 2677-2680.
(34) Hamachi, I.; Hiraoka, T.; Yamada, Y.; Shinkai, S. Chem. Lett. 1998, 27,
537-538.
(35) Wildemann, D.; Drewello, M.; Fischer, G.; Schutkowski, M. Chem.
Commun. 1999, 18, 1809-1810.
4
55-461.
27) Kruszynski, M.; Kupryszewski, G.; Ragnarrson, U.; Alexandrova, M.;
Strbak, V.; Tonon, M. C.; Vaudry, H. Experentia 1985, 41, 1576-1577.
28) Angyal, R.; Strbak, V.; Alexandrova, M.; Kruszynski, M. Endocrinol. Exp.
1
985, 19, 213-219.
29) Alexandrova, M.; Strbak, V.; Herman, Z. S.; Stachura, Z.; Kruszynski, M.
Endocrinol. Exp. 1987, 21, 43-49.
30) Raines, R. T. Chem. ReV. 1998, 98, 1045-1066.
J. AM. CHEM. SOC.
9
VOL. 129, NO. 16, 2007 4911

A R T I C L E S
Wildemann et al.
CD Spectroscopic Measurements. Measurements were performed
at 25 °C in 50 mM sodium acetate buffer (pH 6.0, 0.1 M NaCl) using
4
a Jasco-710 spectropolarimeter. The CD spectrum of ψ[CS-NH] -
RNase S was yielded by mixing of the S-protein (45 µM) and ψ[CS-
4
NH] -S-peptide (29.25 µM) and succeeding incubation at 25 °C for 15
min. According to the association constant, which was determined under
the same conditions by ITC measurements, 96% of the thioxylated
S-peptide is complexed by the S-protein. The CD spectrum of this
mixture was corrected regarding the part of the S-protein and thioxylated
S-peptide that are still free in solution, and the resulting spectrum was
Figure 1. Trans and cis isomeric forms of an amidic thioxo peptide bond
(R1, R2 ) alkyl).
Results and Discussions
4
normalized to a ψ[CS-NH] -RNase S concentration of 50 µM. The
Preparation and Biophysical Characterization of ψ[CS-
CD spectrum of RNase S was obtained according to the same procedure
with concentrations of 45 µM and 225 µM for the S-protein and the
S-peptide, respectively (99.5% saturation of the S-protein), and was
also normalized to an RNase S concentration of 50 µM.
4
NH] -RNase S. Figure 1 shows the isomerization reaction
underlying the phototriggered backbone conformational change.
The sequence of the designed thioxylated S-peptide is H2N-
Lys-Glu-Thr-Ala-ψ[CS-NH]-Ala-Ala-Lys-Phe-Glu-Arg-Gln-
Stopped-Flow Measurements. Determination of koff: Measurements
were performed using a Pistar-180 stopped-flow spectrophotometer
4
His-Nle-Asp-Ser-Ser-Thr-Ser-Ala-Ala-OH (ψ[CS-NH] -S-
35
(Applied Photophysics) at 25 °C in 50 mM sodium acetate buffer (pH
peptide) and was synthesized as described elsewhere. Because
1
3
6
.0, 0.1 M NaCl). The 2.5 mL injection syringe of the stopped-flow
of common problems with oxidation, the Met of the natural
S-peptide was replaced by norleucine (Nle), a substitution that
spectrometer was filled with a 6.9 mM solution of the S-peptide, and
4
37,38
4
the 1 mL injection syringe, with a 126 µM solution of ψ[CS-NH] -
is tolerated by RNase S.
To generate ψ[CS-NH] -RNase
RNase S (mixture containing 135 µM S-protein and 135 µM ψ[CS-
4
S the S-protein and ψ[CS-NH] -S-peptide were allowed to
4
NH] -S-peptide preincubated for 15 min at 25 °C). Concentrations in
recombine by mixing at pH 6.0 (50 mM acetate buffer) at 25
the sample cell (path length: 10 mm) directly after mixing of both
°
C for 15 min. As a reference RNase S was prepared by
solutions, meaning at the time point t
0
of the kinetic measurement, were
recombination of S-protein with the corresponding non-thioxy-
lated S-peptide that also contained an Nle at position 13.
Comparative analysis of the S-peptide/S-protein complex stabili-
ties by isothermal titration calorimetry (ITC) revealed that ψ-
4
33.7 µM ψ[CS-NH] -RNase S and 4.93 mM S-peptide. Detection was
performed using the CD signal of the thioxo peptide bond n-π*
transition at 336 nm.
Determination of kon: Measurements were performed at 25 °C in
4
6
-1
[
CS-NH] -RNase S (Ka ) (1.4 ( 0.04) × 10 M ) was
5
0 mM sodium acetate buffer (pH 6.0, 0.1 M NaCl) using an SX-18
thermodynamically slightly more stable than RNase S (Ka )
MV stopped-flow spectrophotometer (Applied Photophysics). The final
overall concentration of the S-protein within the sample cell was 5
µM. All data were obtained under pseudo-first-order conditions ([S-
peptide derivative] g 5 × [S-protein]). Emission was detected at 315
nm (excitation: 280 nm). The rate constant kobsd was calculated from
the respective time course using a single-exponential equation.
6
-1
(
1.0 ( 0.04) × 10 M ). These values are in the range reported
39,40
for the S-peptide/S-protein interaction under our conditions.
When free in solution, the S-peptide adopts a random coil con-
formation at 25 °C. In contrast, this part of the molecule forms
an R-helix that spans the amino acids at position 3 to 13 of
4
1-43
4
NMR Measurements. All NMR experiments were performed using
a DRX500 spectrometer (Bruker, Rheinstetten, Germany). All spectra
were processed with the XWINNMR 3.5 software (Bruker) and
referenced to external 2,2-dimethyl-2-silapentane-5-sulfonate.
RNase S.
The peptide carbonyl group of Ala is engaged
8
in a hydrogen bond donated by the NH group of Phe, whereas
the Ala -NH is solvent exposed (PDB file 2RLN). Based on
5
near-UV CD spectra the chromogenic properties of the thioxo
Fluorescence Measurements. Spectra were recorded using a
Fluorolog 3 fluorescence spectrometer (Jobin Yvon) and a 10 × 10
mm quartz cuvette. All spectra are corrected for buffer fluorescence
and device specific effects.
peptide bond allowed the determination of microscopic rate con-
stants of ψ[CS-NH] -RNase S formation according to eq 1.
4
Determination of the Enzyme Kinetic Constants. Measurements
were performed in 0.1 M MES buffer (pH 6.0, 0.1 M NaCl) at 25 °C,
3
6
according to Wang et al. The respective S-peptide derivative was
incubated for 15 min with the S-protein, and the measurement was
then started by substrate addition (cCMP). Final concentrations in the
4
In its unbound state, ψ[CS-NH] -S-peptide showed two
samples were 2 µM for the S-protein, 200 µM for the S-peptide, and
4
thioxo group derived bands in the near-UV CD spectrum,
representing the π-π* and n-π* transition at 265 and 336 nm,
1
00 µM for the ψ[CS-NH] -S-peptide. Accordingly, it was guaranteed
that at least 99% of the S-protein is in the complex with the respective
respectively (Figure 2). Examination at 336 nm revealed ψ-
S-peptide derivative.
4
[
CS-NH] -RNase S formation by a change ∆[Θ]MRW ) 7900
Irradiation Experiments. Irradiation was performed using a Xenon/
Mercury lamp in combination with a monochromator (Photon Technol-
ogy International). The light was conducted to the sample by a fiber
-
2
-1
deg cm dmol from positive to negative ellipticity in the
4
(37) Thomson, J.; Ratnaparkhi, G. S.; Varadarajan, R.; Sturtevant, J. M.;
Richards, F. M. Biochemistry 1994, 33, 8587-8593.
(38) Rocchi, R.; Scatturin, A.; Moroder, L.; Marchiori, F.; Tamburro, A. M.;
Scoffone, E. J. Am. Chem. Soc. 1969, 91, 492-496.
optic cable. The ψ[CS-NH] -RNase S was yielded by preincubation
4
(15 min) of a mixture containing the S-protein and ψ[CS-NH] -S-
peptide at the indicated concentrations.
(
39) Schreier, A. A.; Baldwin, R. L. J. Mol. Biol. 1976, 105, 409-426.
In all UV and CD spectroscopic measurements it was guaranteed
that the UV light of the spectrometer has no significant influence on
the thioxo peptide bond cis/trans ratio.
(40) Rocchi, R.; Borin, G.; Marchiori, F.; Moroder, L.; Peggion, E.; Scoffone,
E.; Crescenzi, V.; Quadrifoglio, F. Biochemistry 1972, 11, 50-57.
41) Zegers, I.; Maes, D.; Dao-Thi, M. H.; Poortmans, F.; Palmer, R.; Wyns,
L. Protein Sci. 1994, 3, 2322-2339.
(
(
42) Wyckoff, H. W.; Hardman, K. D.; Allewell, N. M.; Inagami, T.; Tsernoglou,
D.; Johnson, L. N.; Richards, F. M. J. Biol. Chem. 1967, 242, 3749-3753.
(36) Wang, M. H.; Wang, Z. X.; Zhao, K. Y. Biochem. J. 1996, 320, 187-192.
(43) Kartha, G.; Bello, J.; Harker, D. Nature 1967, 213, 862-865.
4
912 J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007
9

Ribonuclease S Backbone Photoswitch
A R T I C L E S
Figure 2. CD measurements. The figure shows the near UV-CD spectra
4
of the ψ[CS-NH] -S-peptide (solid line), S-protein (dotted line), RNase S
4
(
dashed dotted line), and ψ[CS-NH] -RNase S (dashed line) at 50 µM
concentration (path length 10 mm). Insert: zoom of the range between 300
4
and 380 nm in the case of the ψ[CS-NH] -S-peptide (solid line) and ψ-
4
[
°
CS-NH] -RNase S (dashed line). Measurements were performed at 25
C in 50 mM sodium acetate buffer (pH 6.0, 0.1 M NaCl).
Figure 4. Determination of the kon value of the S-peptide/S-protein-
complex (RNase S) formation. (A) Time course of the fluorescence decrease
during formation of RNase S. (B) Concentration dependence of the kinetics
reaction. Measurements were performed at 25 °C in 50 mM sodium acetate
buffer (pH 6.0; 0.1 M NaCl). Overall concentration of the S-protein within
the sample cell was 5 µM. All data were obtained under pseudo-first-order
conditions ([S-peptide] g 5 × [S-protein]). Emission was detected at 315
nm (excitation: 280 nm). The rate constant kobsd was calculated from the
respective time course using a single-exponential equation (solid line). As
Figure 3. Determination of koff. The figure shows the time course of
4
displacement of the ψ[CS-NH] -S-peptide from the S-protein by addition
shown in (A), for an S-peptide concentration of 25 µM a value of k
obsd
)
4
-1
a large excess of the S-peptide to the ψ[CS-NH] -RNase S. Measurements
(11.2 ( 0.7) s was obtained. The kon value of the S-peptide/S-protein
were performed using the CD signal at 336 nm. Concentrations in the sample
complex formation is equivalent to the initial slope of the curve displayed
4
5
cell (path length: 10 mm) were 33.7 µM ψ[CS-NH] -RNase S and 4.93
by the dashed line in (B) and was determined to be (4.37 ( 0.24) × 10
mM S-peptide. The excess was shown to be sufficient by increasing the
concentration of the S-peptide until the observed rate constant was
concentration independent. Using a single-exponential equation, a koff value
M-1 s-1.
Table 1. Parameters Describing the Interaction between the
S-Protein and the Respective S-Peptide Derivative
-
1
of (0.22 ( 0.01) s was determined. The solid line represents the first-
order fit. All measurements were performed at 25 °C in 50 mM sodium
acetate buffer (pH 6.0, 0.1 M NaCl).
S-peptide
derivative
k
M
on
,
k
s
off
,
K
a
,a
M
K
,b
a
1
105
-
1
s-1
-
1
106
-1
106
-
M
4
bound state. Notably, the spectral region above 320 nm lacked
CD signals for S-peptide, S-protein, and RNase S. Thus, the
Ψ[CS-NH] -S-peptide 4.40 ( 0.55 0.22 ( 0.01c 1.4 ( 0.04 2.0 ( 0.34
S-peptide
4.37 ( 0.24 0.44 ( 0.04 1.0 ( 0.04
4
supply of an excess of S-peptide to preformed ψ[CS-NH] -
a
Determined by isothermal titration calorimetry. b Calculated with the
4
RNase S displaced ψ[CS-NH] -S-peptide from the complex
equation K ) k /k . c Calculated with the equation k ) k /K ; K from
a
on off
off
on
a
a
ITC measurements. Measurements were performed in 50 mM sodium acetate
and gave rise to a time-dependent increase in the CD signal at
buffer (pH 6.0; 100 mM NaCl) at 25 °C.
3
36 nm that followed pseudo-first-order kinetics (Figure 3). In
the RNase S dissociation reaction the rate of displacement of
fluorescently labeled S-peptides was already shown to be similar
that thioxylation at position 4 of the S-peptide has no effect on
4
kon. Furthermore, Ka for ψ[CS-NH] -RNase S formation, as
4
4
4
to that of koff. The half time of dissociation of ψ[CS-NH] -
-1
calculated from the equation Ka ) kon/koff ) (2.0 ( 0.34) M ,
-
1
RNase S was found to be 3.1 s (koff ) (0.22 ( 0.1) s ). We
is rather similar to the value measured by isothermal titration
calorimetry (Table 1). The rate constant koff for RNase S, which
is not directly available under our conditions, can then be
4
next determined kon for ψ[CS-NH] -RNase S formation by
using stopped-flow fluorescence at Tyr emission of S-protein.
Using this method, it was also possible to determine kon for the
S-peptide/S-protein recombination under the same conditions
-1
calculated from koff ) kon/Ka ) (0.44 ( 0.04) s . This indicates
that the O/S exchange contributed little to the noncovalent
interactions determining RNase S stability, and this small
contribution is exclusively found in the koff rate. Taken together
these results demonstrate that, as a precondition for the
suitability of this thioxo peptide bond in photoswitching
5
-1 -1
(
Figure 4). A kon value of about 4.4 × 10 M
s
was obtained
for both, the modified and unmodified RNase S, thus showing
(
44) Goldberg, J. M.; Baldwin, R. L. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,
2
019-2024.
J. AM. CHEM. SOC.
9
VOL. 129, NO. 16, 2007 4913

A R T I C L E S
Wildemann et al.
4
Figure 5. Photoisomerization of ψ[CS-NH] -S-peptide: UV spectroscopic
4
characterization. The figure shows the UV spectrum of the ψ[CS-NH] -
S-peptide measured before irradiation (solid line), in the photostationary
state of irradiation at 254 nm (dashed line) and after 5 cycles of irradiation/
re-equilibration (circles). Insert: The time dependence of re-equilibration
(
1
circles) has revealed a first-order rate constant kc/t of (5.7 ( 0.005) ×
-
3 -1
0
s
and was measured following the absorption at 287 nm. The solid
line represents the first-order fit. All measurements were performed at 10
4
°
C with 18.3 µM of the ψ[CS-NH] -S-peptide in 50 mM sodium acetate
buffer (pH 6.0, 100 mM NaCl, path length: 10 mm). Irradiation was
performed for 5 min at 254 nm.
experiments, the O/S substitution at position 4 has only a minor
influence on the thermodynamic constants which characterize
the S-peptide/S-protein complex.
4
Photoisomerization of Thioxo Compounds. The Ala -ψ-
5
4
[
CS-NH]-Ala moiety of the ψ[CS-NH] -S-peptide provides
Figure 6. Ac-Ala-ψ[CS-NH]-Phe-NH : Determination of the cis content
2
1
an example of a isosteric replacement of a secondary amide
peptide bond that probably all have >99% trans conformation
in the ground state of unstructured polypeptides.
of the ψ[CS-NH] -S-peptide with UV light at 254 nm has
resulted in a bathochromic shift of the 265 nm UV absorption
band and a slightly reduced maximum intensity (Figure 5) that
is typical of an increased cis/trans ratio.2
in the photostationary state of 254 nm irradiation by H NMR measurements.
A sample containing 3.07 mM Ac-Ala-ψ[CS-NH]-Phe-NH2 in 33 mM
sodium phosphate buffer (pH 7.5, 5% D2O) was irradiated with UV light
23,45,46
Irradiation
1
at 254 nm until the photostationary state was reached and then the H NMR
4
measurement started. (A) From the time course of the cis/trans ratio (circles)
-
3 -1
a value of (1.15 ( 0.03) × 10
s
was calculated for the first-order rate
constant kc/t characterizing the re-equilibration process at 5 °C. The solid
line represents the first-order fit. By extrapolation of the regression curve
to the time point t ) 0, a maximum cis content of 19.9% in the
photostationary state was determined. The cis isomer percentage was
calculated by integration of the respective thioxo alanine â-methyl proton
signal (trans, 1.33 ppm; cis, 0.63 ppm) as shown in (B) (I, before irradiation;
II, after 7.5 min of re-equilibration; III, after 1 h of re-equilibration). The
following parameters were used for the NMR data collection: relaxation
time, 1.2 s; number of scans per experiment, 16; spectral width, 14 ppm
1,23
The thermal re-
equilibration after irradiation is characterized by a first-order
-
3
-1
rate constant kc/t ) 5.7 × 10
s
at 10 °C about 1000-fold
4
5,47
slower as compared to an oxopeptide bond (Figure 5).
This
slow rate of isomerization allowed us to explore the individual
catalytic parameters for each of the two isomeric forms of the
enzyme. These forms exist as transient populations in the
photostationary state contingent to a cis/trans isomerization at
a single predefined peptide bond.
The lack of an NMR-sensitive isomeric reporter group has
limited direct determination of the cis/trans ratio in ψ[CS-
NH] -S-peptides. Fortunately, an indirect method, which is
based on the nearly sequence-independent π-π* transition in
the UV/vis spectra of thioxopeptides could serve as a basis for
determining the cis/trans ratio of the ψ[CS-NH] -S-peptide and
ψ[CS-NH] -RNase S in their photostationary states. If the
isomerizing thioxopeptide bond is the linker group between
phenylalanine and alanine such as those occurring in Ac-Ala-
ψ[CS-NH]-Phe-NH2, the alanine â-methyl group experienced
a large isomer specific anisotropy shift of 0.7 ppm that is upfield
(
7002.8 Hz); 90° puls length, 7.4 µs (2 dB); time domain size, 4 K;
spectrometer frequency, 500.13 MHz.
doublets, permitted us to calculate the cis/trans ratio in the
photostationary state after irradiation at 254 nm (Figure 6). The
cis content increased from ,0.5% to 19.9% in the photosta-
tionary state at 3.07 mM peptide under the conditions of the
NMR experiment. If examined under similar conditions the
NMR-based knowledge about the cis/trans ratio gave rise to
the UV/vis spectrum of the pure cis and trans isomers (Figure
4
45
4
4
7
). The Ac-Ala-ψ[CS-NH]-Phe-NH2 exhibited absorption
maxima in its cis and trans state at 281 and 265 nm with molar
-1 -1
absorption coefficients of ꢀ281 ) 15 215 M
1
s
and ꢀ265 )
-1 -1
2 000 M s , respectively. Notably, the cis/trans ratio of the
photostationary state rises to 40% if the peptide concentration
was lowered from 3 mM to 20 µM (see the Supporting
Information). The reason is that at 3 mM the transparency of
the sample is too low for an efficient photoisomerization.
Corresponding isomer-specific band positions for the ψ[CS-
1
for the H NMR signal of the cis isomer. The intensity pattern
of the two sets of methyl signals, the cis isomer and trans isomer
(
(
(
45) Scherer, G.; Kramer, M. L.; Schutkowski, M.; Reimer, U.; Fischer, G. J.
Am. Chem. Soc. 1998, 120, 5568-5574.
46) Pappenberger, G.; Aygun, H.; Engels, J. W.; Reimer, U.; Fischer, G.;
Kiefhaber, T. Nat. Struct. Biol. 2001, 8, 452-458.
4
NH] -S-peptide were calculated to be 281 (cis) and 265 nm
47) Schiene-Fischer, C.; Fischer, G. J. Am. Chem. Soc. 2001, 123, 6227-6231.
(trans).
4914 J. AM. CHEM. SOC.
9
VOL. 129, NO. 16, 2007

Ribonuclease S Backbone Photoswitch
A R T I C L E S
Figure 7. UV spectroscopic characterization of the thioxo peptide bond
cis/trans photoisomerization of Ac-Ala-ψ[CS-NH]-Phe-NH2. The following
UV spectra are shown: Ac-Ala-ψ[CS-NH]-Phe-NH2 before irradiation
(solid line), in the photostationary state of irradiation at 254 nm (dashed
line), after five cycles of irradiation/re-equilibration (circles) and the
calculated spectrum of the pure cis thioxo peptide bond isomer form (dotted
line). The measurements were performed at 5 °C in 50 mM sodium acetate
buffer (pH 6.0, 100 mM NaCl, path length: 10 mm). For irradiation
(duration: 5 min, wavelength: 254 nm), the same solution as in the case
of the NMR measurements was used. After irradiation this concentrated
solution was transferred into a cuvette containing the appropriate amount
of buffer, resulting in a final thioxo peptide concentration of 18.35 µM.
The measurement was then started immediately. Using a value of 19.9%
for the cis content of the irradiated sample, as determined by the NMR
measurements, the UV spectrum of the pure cis isomeric form was
calculated. Thereby, it was assumed that the UV spectrum of the nonirra-
diated sample represents the spectrum of the pure trans isomeric form,
because the NMR measurements have revealed that without irradiation the
cis content is , 0.5%.
4
Figure 8. Irradiation of ψ[CS-NH] -RNase S with UV-light at 254 nm.
A) UV spectroscopic characterization of the re-equilibration process after
(
4
thioxo peptide bond cis/trans photoisomerization of ψ[CS-NH] -RNase
S. The measurements were performed at 10 °C in 50 mM sodium acetate
4
Short-time irradiation of ψ[CS-NH] -RNase S leads also to
reversible changes in the UV spectrum. But during long-time
buffer (pH 6.0, 100 mM NaCl, path length: 10 mm). The sample contained
4
7
9.6 µM S-protein and 36 µM ψ[CS-NH] -S-peptide ensuring that >98%
irradiations an irreversible change in the UV spectrum of ψ-
of the S-peptide derivative is bound to the S-protein. Before irradiation
duration, 5 min; wavelength, 254 nm), the sample was stirred for 15 min.
4
[
CS-NH] -RNase S was observed (Figure 8). On the contrary
(
to the reversible effect, the irreversible effect does not depend
on the thioxo peptide bond isomerization because it was also
observed in the case of the S-protein and RNase S, respectively.
Given the similarity of isomer-specific changes of the Cd
After irradiation the first measurement was started immediately followed
by further measurements after increasing times of re-equilibration. (B) Long-
time irradiation at 254 nm leads to irreversible changes in the UV spectrum
4
of ψ[CS-NH] -RNase S. Measurements were performed using the same
conditions as described in (A). Before the respective measurement was
started, the sample was irradiated the indicated time and then stirred for a
further 60 min without irradiation (re-equilibration of the thioxo peptide
bond isomeric state).
23
S-derived π-π* transition for Ac-Ala-ψ[CS-NH]-Phe-NH2
4
and ψ[CS-NH] -S-peptide, spectra monitored in the photosta-
4
4
tionary states of ψ[CS-NH] -S-peptide and ψ[CS-NH] -RNase
S can be used to calculate the cis/trans ratio of the irradiated
samples (see the Supporting Information). At a concentration
level of 18.3 µM we obtained 30.3% and 27.5% cis for ψ[CS-
constant k is similar to that found for the unbound ψ[CS-
c/t
4
NH] -S-peptide. This indicates that only the specific binding
4
of the ψ[CS-NH] -S-peptide by the S-protein is responsible
4
4
NH] -S-peptide and ψ[CS-NH] -RNase S, respectively. Ac-
for the effect of the S-protein on k . These results are consistent
c/t
4
4
Ala-ψ[CS-NH]-Phe-NH2 and ψ[CS-NH] -S-peptide subjected
with a stable cis ψ[CS-NH] -RNase molecule that does not
4
to five cycles of irradiation/re-equlibration did not show any
sign of an irreversible event (Figures 5, 7).
dissociate into the components after switching the Ala -ψ[CS-
5
NH]-Ala moiety in cis conformation.
4
Re-equilibration after irradiation of thioxopeptides gave rise
to rate constants of the reversible thioxo peptide bond cis/trans
isomerization by monitoring the time-dependent decrease in
absorbance at 287 nm (Figure 9). At 10 °C the first-order rate
constants for thioxopeptide bond isomerization in the case of
Definitive proof of the stability of cis ψ[CS-NH] -RNase S
4
8
required investigation of (ψ[CS-NH], Trp )-RNase S that was
4
8
obtained from chemically synthesized (ψ[CS-NH], Trp )-S-
peptide and S-protein by recombination. The association
4
8
constant characterizing the stability of the (ψ[CS-NH], Trp )-
RNase S at 10 °C, the same temperature used in the irradiation
4
4
the ψ[CS-NH] -S-peptide and ψ[CS-NH] -RNase were kc/t
-
3
-1
-3 -1
5
-1
)
(5.7 ( 0.005) × 10
s
and kc/t) (3.9 ( 0.02) × 10 s ,
experiments, has a value of (5.6 ( 0.4) × 10 M and is
respectively. Importantly, the order of irradiation and recom-
significantly lower than the value measured at 25 °C for ψ-
4
4
bination steps involving S-protein and ψ[CS-NH] -S-peptide
[CS-NH] -RNase. In this context it is also important to note
does not affect the observation of a lower cis/trans isomerization
that at 10 °C compared to 25 °C the S-peptide/S-protein complex
stability is about 10-fold higher.
The fluorescence of the singular Trp residue found in the
whole complex is considerably quenched due to (ψ[CS-
4
37
rate for ψ[CS-NH] -RNase. In the presence of an excess of
4
unmodified S-peptide, which displaced ψ[CS-NH] -S-peptide
4
from the irradiated ψ[CS-NH] -RNase, the isomerization rate
J. AM. CHEM. SOC.
9
VOL. 129, NO. 16, 2007 4915

A R T I C L E S
Wildemann et al.
Figure 11. The thioxopeptide bond photoisomerization does not influence
Figure 9. Re-equilibration process after the thioxo peptide bond trans to
cis photoisomerization in the ψ[CS-NH] -S-peptide: influence of the
4
8
the fluorescence emission spectrum of the (ψ[CS-NH] ,Trp )-RNase S.
4
4
8
The spectra represent the (ψ[CS-NH] ,Trp )-RNase S before irradiation
dotted line), in the photostationary state of the thioxopeptide bond
S-protein on kc/t. All measurements were performed at 10 °C in 50 mM
sodium acetate buffer (pH 6.0, 100 mM NaCl). Before starting the
measurements, each sample was irradiated for 5 min with UV light at 254
nm. The re-equilibration process was followed at 287 nm. In every case,
(
photoisomerization (solid line) and after 2 h of re-equilibration (circles).
The irreversible decrease in fluorescence results from the small amount of
the tryptophan photobleaching during irradiation at 254 nm (see the
Supporting Information). Measurements were performed at 10 °C in 50
mM sodium acetate buffer (pH 6.0) containing 100 mM NaCl. The
4
the concentration of the ψ[CS-NH] -S-peptide was 2 µM. (Circles) free
4
-3 -1
ψ[CS-NH] -S-peptide: kc/t ) (5.7 ( 0.03) × 10
s ; (squares) in
-
3
-1
presence of 60 µM S-protein: kc/t ) (3.9 ( 0.02) × 10
s ; (triangles)
4
8
concentration of the (ψ[CS-NH] ,Trp )-RNase S was 10 µM. The sample
in presence of 60 µM S-protein and 200 µM S-peptide (the S-peptide was
4
contained 60 µM of the S-protein ensuring that >96% of the (ψ[CS-NH] ,-
-
3
added directly after termination of irradiation): kc/t ) (5.7 ( 0.03) × 10
8
Trp )-S-peptide (final concentration in the sample: 10 µM) is bound to the
-
1
s
. For the first and second case the respective linear regression is shown
S-protein. Fluorescence resulting from the nonbounded, excess S-protein
was subtracted from the respective spectrum. The fluorescence excitation
was performed at 297 nm. Irradiation was performed for 5 min at 254 nm.
by the dashed and solid line, respectively.
Table 2. Enzyme Kinetic Constants for Hydrolysis of the
Substrate Cytidine 2′,3′-Cyclic Monophosphate (cCMP) by RNase
4
S and ψ[CS-NH] -RNase S
a
b
K
m
K
p
k
2
k
2
/K
m
(s-1)
(M- s-1)
1
enzyme
(mM)
(
µ
M)
4
Ψ[CS-NH] -RNase S 0.6 ( 0.02 75 ( 1.1 3.6 ( 0.2 6000 ( 533
RNase S
0.6 ( 0.02 75 ( 0.6 3.9 ( 0.13 6500 ( 433
a
The Kp value represents the dissociation constant of the enzyme/product
complex, since the reaction product, cytidine 3′-monophosphate, is an
inhibitor of RNase S. b k2 represents the reaction rate constant of the slowest
step of the catalyzed substrate hydrolysis. Measurements were performed
in 0.1 M MES buffer (pH 6.0, 0.1 M NaCl) at 25 °C according to Wang et
36
al.
4
8
Figure 10. Fluorescence emission spectra of the (ψ[CS-NH] ,Trp )-S-
ents are readily attained by using appropriate concentrations of
starting materials. In essence, only minor alterations were found
for k2 (rate constant of the slowest step of the enzyme catalyzed
4
8
peptide and (ψ[CS-NH] ,Trp )-RNase S. Measurements were performed
at 10 °C in 50 mM sodium acetate buffer (pH 6.0) containing 100 mM
4
8
NaCl. The concentration of the (ψ[CS-NH] ,Trp )-S-peptide (solid line)
4
8
4
and (ψ[CS-NH] ,Trp )-RNase S (dashed line) was 10 µM. The sample of
substrate hydrolysis) and Km values of both ψ[CS-NH] -RNase
4
8
the (ψ[CS-NH ,Trp )-RNase S contained 60 µM of the S-protein ensuring
S and RNase S (Table 2), and the constants were similar to
4
8
that >96% of the (ψ[CS-NH ,Trp )-S-peptide (final concentration in the
sample: 10 µM) is bound to the S-protein. Fluorescence resulting from the
nonbounded, excess S-protein was subtracted from the respective spectrum.
Samples were excitated at 297 nm. The quenching of the tryptophan
3
6
those reported for RNase S. Because the thioxopeptide site
does not make contact with the S-protein and the thioxocarbonyl
is located in a space a long distance from the catalytic residues
of RNase S, insensitivity of the catalytic parameter to this
4
8
fluorescence results from specific binding of (ψ[CS-NH] ,Trp )-S-peptide
by the S-protein. This was demonstrated by addition of a large excess of
the S-peptide to the preformed (ψ[CS-NH] ,Trp )-RNase S that replaces
the (ψ[CS-NH] ,Trp )-S-peptide from the S-protein (see the Supporting
4
4
8
substitution is comprehensible. To assess how ψ[CS-NH] -
4
8
RNase S responds to the light-induced trans to cis intercon-
Information).
4
5
version of the Ala -ψ[CS-NH]-Ala moiety, initial rates of
cCMP cleavage have been investigated before and after irradia-
tion at 254 nm. Briefly, preformed ψ[CS-NH] -RNase S was
4
8
4
NH], Trp )-RNase S formation (Figure 10) allowing sensitive
detection of complex dissociation during irradiation. After
irradiation at 254 nm we found fluorescence quenching com-
pletely retained with the consequent preservation of the original
irradiated and the hydrolytic reactions were started by addition
of cCMP into the sample cell. Time courses of cCMP hydrolysis
were measured by monitoring the absorbance increase at 290
nm at 10 °C and were found to be linear over the initial phase
of the reaction (10 s). This indicates that cCMP depletion was
negligible and alterations of the activity of irradiated ψ[CS-
4
8
population of cis (ψ[CS-NH], Trp )-RNase S (Figure 11).
Photocontrol of ψ[CS-NH] -RNase S Activity. The cleav-
age of a PO bond in cytidine 2′,3′-cyclic monophosphate
cCMP) has been investigated to compare the enzymatic
properties of RNase S and ψ[CS-NH] -RNase S. Conditions
of quantitatively formed RNase S complexes from the constitu-
4
4
(
NH] -RNase S were absent in this time range (Figure 12A).
4
Furthermore, the results of the irradiation experiments with (ψ-
4
8
[CS-NH], Trp )-RNase S (Figures 10 and 11) discussed above
4916 J. AM. CHEM. SOC.
9
VOL. 129, NO. 16, 2007

Ribonuclease S Backbone Photoswitch
A R T I C L E S
4
Figure 13. Regain of the ψ[CS-NH] -RNase S activity after the
thioxopeptide bond photoisomerization. Kinetic measurements were per-
formed at 10 °C in 50 mM sodium acetate buffer (pH 6.0, 100 mM NaCl).
4
A mixture of 60 µM S-protein and 2 µM ψ[CS-NH] -S-peptide was
4
preincubated for 15 min. At these concentrations, >98% of ψ[CS-NH] -
S-peptide is complexed by the S-protein. The mixture was then irradiated
for 5 min at 254 nm. After the respective time of re-equilibration the
measurement was started by addition of 100 µM cCMP and the corre-
sponding initial velocity V0 was determined. The relative initial rate rel. V0
was calculated by setting the initial rate of the nonirradiated sample to 100%.
For the regain of the enzymatic activity, data analysis using a single-
-
3 -1
exponential equation resulted in a rate constant of (3.6 ( 0.5) × 10
s .
activity regain was found to follow first-order kinetics with k
-
3 -1
)
(3.6 ( 0.5) × 10
s
that agrees well with the value kc/t )
directly measured for the thioxo peptide
-
3 -1
(3.9 ( 0.02) × 10
s
4
4
bond cis/trans isomerization of ψ[CS-NH] -RNase S (Figures
9, 13). Second, different cis levels of ψ[CS-NH] -RNase S in
Figure 12. Initial rates of the ψ[CS-NH] -RNase S catalyzed cCMP
hydrolysis: Influence of the thioxopeptide bond photoisomerization. (A)
Enzyme kinetic measurements were performed before irradiation [circles;
V0 ) (1.88 ( 0.04) µM/s], at the photostartionary state of irradiation
4
the photostationary state, which could be achieved by variations
of the irradiation wavelength, were linearly correlated with initial
rates of cCMP hydrolysis (Figure 12B). To ensure reversibility
[diamonds; V0 ) (1.31 ( 0.02) µM/s] and after re-equilibration (triangles;
V0 ) (1.93 ( 0.04) µM/s]. In the first and second case the corresponding
regression line is shown by the dotted and solid line, respectively. Kinetic
measurements were performed at 10 °C in 50 mM sodium acetate buffer
4
of the photoswitching, photoisomerized ψ[CS-NH] -RNase S
was allowed to re-equilibrate for 60 min followed by enzyme
activity monitoring (Figure 12A). The original level of enzyme
activity (1.93 ( 0.04) µM/s was achieved after this procedure.
What is most striking is that the magnitude of the cis ψ[CS-
(
pH 6.0, 100 mM NaCl). Final concentrations in the sample were 60 µM
4
4
S-protein, 2 µM ψ[CS-NH] -S-peptide where >98% of the ψ[CS-NH] -
S-peptide is complexed by the S-protein. Irradiation at 254 nm was
performed for 5 min. Data collection was started immediately after substrate
addition (100 µM cCMP). Thereby every 1 s one data point was collected.
Because of the re-equilibration process of the thioxo peptide bond cis/trans
equilibrium, a small part of ∆A290 in the case of the irradiated sample results
from the small amount (∼1%) of the thioxo peptide bond cis to trans
isomerization that happens in parallel during the time of the enzymkinetic
measurement. This small amount of signal change was subtracted guarantee-
ing that the graph of the irradiated sample in the figure represents only the
substrate hydrolysis. (B) Correlation between the effect of irradiation on
4
4
NH] -RNase S fraction of the irradiated ψ[CS-NH] -RNase S
4
and the relative loss of ψ[CS-NH] -RNase S activity can be
quantitatively matched (Figure 12B). Given a stable complex
4
structure the cis ψ[CS-NH] -RNase S complex must be
considered as an RNase inactive enzyme variant. This last result
4
sets ψ[CS-NH] -RNase S apart from [phenylazophenylala-
4
4
the ψ[CS-NH] -RNase S activity and the content of the cis thioxo peptide
nine] -RNase S since the latter side-chain photomodulatable
bond after irradiation. Cis contents of 27.5% and 20.5% were yielded by
irradiation with UV light at 254 and 260 nm, respectively. The kinetic
measurements were performed as those described above. The slope of the
enzyme variant hardly affected catalytic properties after adopting
14,33
the cis conformation.
It is somewhat a mystery why a remote
4
regression line is -1.07 ( 0.13, indicating that the ψ[CS-NH] -RNase S
peptide bond trans to cis isomerization caused reduction of the
species with the cis thioxo peptide bond seems to be inactive.
level of cCMP hydolysis to <1 % of those for the trans ψ-
4
[
CS-NH] -RNase S variant. From our data it is unlikely that
4
along with the linear signal response in the activity measure-
photoisomerization displaces cis ψ[CS-NH] -S-peptide from
trans ψ[CS-NH] -RNase S with a concomitant total loss of
4
4
ments (Figure 12A) indicate that cis ψ[CS-NH] -RNase S does
not dissociate in the constituents after its formation. Surprisingly
initial rates of cCMP hydrolysis were found to be (1.88 ( 0.04)
µM/s and (1.31 ( 0.02) µM/s for the nonirradiated state and
the photostationary state of ψ[CS-NH] -RNase S, respectively.
Furthermore, using the same conditions, irradiation of the non-
enzyme activity.
Notably, it is the first turn of the short R-helix encompassing
4
5
residues 3-13 that includes the Ala -ψ[CS-NH]-Ala moiety.
The cis conformation can be postulated to produce a local
interruption of the helical conformation leaving other helical
turns intact. However, from the invariance of Ka to the O/S
replacement it is suggested that the CdS ‚ ‚ ‚ ‚ HsN Phe8
hydrogen bond plays only a minor role for the helix stability,
and thus removal has a low impact on helix positioning. In
contrast, the insertion of a cis peptide linkage at the N-terminus
4
thioxylated enzyme has no effect on the acitivity. Thus, the
4
activity loss of 30% of ψ[CS-NH] -RNase S after irradiation
appears to result from a partial interconversion of trans ψ[CS-
4
4
NH] -RNase S into cis ψ[CS-NH] -RNase S. Two different
approaches provided proof of this. First, the rate constant of
J. AM. CHEM. SOC.
9
VOL. 129, NO. 16, 2007 4917

A R T I C L E S
Wildemann et al.
4
might apply compensatory components to the helix dipole that
has been implicated in acid/base properties of pendent side
chains.48 At the helix termini the pKa of a histidine side chain
motions differ between trans ψ[CS-NH] -RNase S and the cis
ψ[CS-NH] -RNase S variant variants.
4
Conclusion
can be shifted up to 2 units when compared with the normal
value of histidinium residues.4
9,50
Major residues involved in
UV/vis irradiation of the site-specifically thioxylated 20-mer
S-peptide of RNase S or its ψ[CS-NH] -RNase S complex, in
RNase S catalysis include His12 and His that display pKa
values (RNaseA) of 5.8 and 6.2, respectively. The unperturbed
119
4
51
which the ground state all-trans conformation of the S-peptide
portion is nearly complete, resulted in a photostationary state
with about 30% cis isomer for the secondary amide thioxopep-
tide bond. A slow first-order reaction of thermal re-equilibration
allowed sufficient time for assaying the influence of the
increased cis population on enzymatic reactions. Photoswitching
intrinsic pKa of a proteinaceous His residue was given to be
5
2
6
.5 for both surface and buried histidine. The C-terminus of
12
the S-peptide derived R-helix presents His prone to such a
helix dipole induced pKa shift. Thus, an upward shift of the
pKa value by reducing the helical dipole might severely interfere
with the role of His as an acid/base catalyst. Alternatively, an
isomerization-mediated chain displacement of the N-terminus
12
4
trans ψ[CS-NH] -RNase S thus provides the unique feature
of giving an enzyme the opportunity to respond to a well-defined
one-bond conformational change situated at a predefined peptide
bond of the enzyme.
4
of ψ[CS-NH] -RNase S might indirectly affect its active site
structure. Although the N-terminal corrupted, the core part of
4
the ψ[CS-NH] -S-peptide helix would remain connected to the
A surprising result of the first application of this method is
that the conformation of the remote Ala -ψ[CS-NH]-Ala
4
S-protein. New ψ[CS-NH] -S-peptide/S-protein contacts could
4
5
be formed in the cis isomer that could not be inferred from
available RNase S structures with an all-trans S-peptide
segment. However, because of results which were obtained with
moiety, which was never considered to be important for
catalysis, has been implicated in the control of the catalytic
machinery of RNase S. Whatever the reason, the one-bond
photoswitching of the backbone is likely to find utility in the
evaluation of folding-function relationships. A combination of
a thioxopeptide scan through polypeptide segments with pho-
toswitching may be especially informative.
1
2
3
6
des-Lys, Glu, Thr -RNase S, and [Pro ]-RNase S, respectively,
the total loss of enzymatic activity by isomerization-mediated
chain displacement of the N-terminus ore interruption of the
4
first turns of the S-peptide R-helix of ψ[CS-NH] -RNase S
seems to be less likely.5
3-55
There is evidence by using theoretical investigations of
concerted conformational fluctuations to suggest a correlation
between the network of protein vibrations and rate enhancement
Acknowledgment. We thank Dr. A. Schierhorn and Dr. M.
Kipping for mass spectroscopic measurements and Dr. J.
Fangh a¨ nel for the ITC measurements. We gratefully acknowl-
edge B. H o¨ kelmann and M. Seidel for excellent technical
assistance. This work was supported by the “Fond der Chemis-
chen Industrie”.
5
6
by enzymes. In light of these data, uncoupling of backbone
motions from substrate turnover steps might also account for
4
the loss of enzymatic activity in the cis ψ[CS-NH] -RNase S
variant. Experimental hints have emerged that such a mechanism
5
7,58
could be effective in dihydrofolate reductase catalysis.
However, it is currently not known whether internal protein
Supporting Information Available: Determination of content
4
4
of the cis ψ[CS-NH] -S-peptide and cis ψ[CS-NH] -RNase
S and additional control experiments about photoswitching of
(
(
(
(
(
48) Joshi, H. V.; Meyer, M. S. J. Am. Chem. Soc. 1996, 118, 12038-12044.
49) Lodi, P.; Knowles, J. R. Biochemistry 1993, 32, 4338-4343.
50) Sancho, J.; Serrano, L.; Fersht, A. R. Biochemistry 1992, 31, 2253-2258.
51) Markley, J. L. Biochemistry 1975, 14, 3546-3553.
4
8
(ψ[CS-NH], Trp )-RNase S are described in detail. This
material is available free of charge via the Internet at
http://pubs.acs.org.
52) Jensen, J. H.; Li, H.; Robertson, A. D.; Molina, P. A. J. Phys. Chem. A
2
005, 109, 6634-6643.
(
(
(
(
53) Moroder, L.; Marchiori, F.; Rocchi, R.; Fontana, A.; Scoffone, E. J. Am.
JA069048O
Chem. Soc. 1969, 91, 3921-3926.
54) Marchiori, F.; Rocchi, R.; Moroder, L.; Fontana, A.; Scoffone, E. J. Am.
Chem. Soc. 1968, 90, 5889-5894.
(57) Rod, T. H.; Radkiewicz, J. L.; Brooks, C. L. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 6988-6985.
(58) Watney, J. B.; Agrawal, P. K.; Hammes-Schiffer, S. J. Am. Chem. Soc.
2003, 125, 3747-3750.
55) Marchiori, F.; Borin, G.; Moroder, L.; Rocchi, R.; Scoffone, E. Biochim.
Biophys. Acta 1972, 257, 210-221.
56) Agarwal, P. K. J. Am. Chem. Soc. 2005, 127, 15248-15256.
4918 J. AM. CHEM. SOC.
9
VOL. 129, NO. 16, 2007