Originally published In Press as doi:10.1074/jbc.M107889200 on November 1, 2001
J. Biol. Chem., Vol. 277, Issue 2, 1419-1425, January 11, 2002
Crystal Structures of the Semireduced and
Inhibitor-bound Forms of Cyclic Nucleotide Phosphodiesterase
from Arabidopsis thaliana*,
Andreas
Hofmann
§,
Melissa
Grella¶,
Istvan
Botos,
Witold
Filipowicz
, and
Alexander
Wlodawer
From the Macromolecular Crystallography
Laboratory, NCI, National Institutes of Health, Frederick,
Maryland 21702 and the Friedrich Miescher-Institut,
4002 Basel, Switzerland
Received for publication, August 16, 2001, and in revised form, October 28, 2001
 |
ABSTRACT |
The crystal structure of the semireduced
form of cyclic nucleotide phosphodiesterase (CPDase) from
Arabidopsis thaliana has been solved by molecular
replacement and refined at the resolution of 1.8 Å. We have previously
reported the crystal structure of the native form of this enzyme, whose
main target is ADP-ribose 1",2"-cyclic phosphate, a product of the tRNA
splicing reaction. CPDase possesses six cysteine residues, four of
which are involved in forming two intra-molecular disulfide bridges.
One of these bridges, between Cys-104 and Cys-110, is opened in the
semireduced CPDase, whereas the other remains intact. This change of
the redox state leads to a conformational rearrangement in the loop
covering the active site of the protein. While the native structure
shows this partially disordered loop in a coil conformation, in the semireduced enzyme the N-terminal lobe of this loop winds up and elongates the preceding 
-helix. The semireduced state of CPDase also
enabled co-crystallization with a putative inhibitor of its enzymatic
activity, 2',3'-cyclic uridine vanadate. The ligand is bound within the
active site, and the mode of binding is in agreement with the
previously proposed enzymatic mechanism. Selected biophysical
properties of the oxidized and the semireduced CPDase are also discussed.
| |
INTRODUCTION |
Appr>p1 is a product
generated during tRNA splicing in yeast, plants, and also vertebrates
(1-3). It is further processed into Appr-1"p in a reaction catalyzed
by a cyclic nucleotide phosphodiesterase (CPDase), which cleaves the
2"-phosphoester (4, 5). CPDases, characterized in wheat (6),
Arabidopsis thaliana (1, 5), and yeast (1, 7, 8), constitute
one group within a large family of proteins that includes at least four
different classes of enzymes having cyclic phosphodiesterase or related
activities. Members of this family, or enzymatically competent domains
thereof, are of comparable lengths (about 200 residues) and share two
similarly spaced tetrapeptide signature motifs (H-
-(T/S)-,
being a hydrophobic residue) (8).
The first crystal structure for any member of this family was reported
recently by us for the CPDase from A. thaliana (9). The
crystal structure showed two almost symmetrical lobes formed by the
non-contiguous parts of the peptide chain. Each lobe consists of a
three- or four-stranded antiparallel 
-sheet, respectively, constituting the inner core of the protein (see Fig. 1A).
The arrangement of the -sheets resembles an open barrel flanked on the outside of each lobe by two antiparallel -helices. Despite close
similarity between both lobes, their multiple connectivity results in
only a single globular domain rather than in a two-domain fold. The
structural elements within each lobe show similarities to many proteins
involved in RNA binding and to kinases; however, their arrangement
results in a new protein fold with unique features. A surface loop
covering the putative active site showed a high degree of disorder and
was thus assumed to be flexible.
The active site was found to be located in a water-filled cavity and is
composed of the tandem signature motif residues H- -(T/S)- .
These residues, as well as Tyr-124, participated in the coordination of
a sulfate ion. Based on these findings, an enzymatic mechanism was
proposed that employs the nucleophilic attack of a water molecule,
which is activated by His-119. Additional residues of the active site
motif (Thr-44, Ser-121, Tyr-124) were assigned a stabilizing function
by keeping the substrate locally fixed via appropriate coordination. In
the last step, the proton from the iminium group of His-42 is
transferred to the free 2'-oxygen of the phospho-ribosyl moiety.
Because this transfer leaves His-42 in the basic state and His-119 in
the acidic state, the system must be restored, most likely via fast
proton transfer using two ordered water molecules present in the cavity.
CPDase possesses six cysteine residues, four of which form two
dithioether linkages as determined by mass spectrometry and by the
crystal structure (9). Cys-64 and Cys-177 are located in helix 2 and
the C-terminal -strand 7, respectively (see Fig. 1C).
Both structural elements are directly adjacent and are covalently
linked through the formation of a disulfide bridge between these two
cysteine residues. The loop region (100) exists in a coil
conformation and is exposed to the solvent. The flap-like shape of this
loop is maintained by the presence of a second disulfide bridge between
Cys-104 and Cys-110. This cystine ties together the upstream and
downstream moieties of the loop, thereby stabilizing the overall
conformation. The remaining two cysteines, Cys-86 and Cys-159, are
found in isolated and shielded positions and thus cannot form disulfide
bonds. Therefore, the oxidized state of CPDase contains two disulfides
and two free sulfhydryl groups. The assumption that the flexible loop
is indeed an important feature of CPDase and the presence of a
disulfide bridge within this stretch lead us to investigate the
possible effects of reducing environments on the structure and activity
of the protein.
Although the catalytic activity of the wild-type A. thaliana
CPDase was investigated in an earlier study (5), the effect of
mutations or redox potentials was not characterized. It is known from a
mutation study on Saccharomyces cerevisiae CPDase that the
four residues of the tandem signature motif are crucial for the
catalytic activity of this protein (8). Because the Arabidopsis and the yeast protein are thought to be
homologous, these results have stimulated interpretation of the
behavior of the plant enzyme as well. However, the yeast enzyme does
not show the same cysteine pattern in its primary structure.
In the present study, we provide characterization of the (semi-)reduced
state of the A. thaliana CPDase as well as the crystal structure of CPDase in complex with 2',3'-cyclic uridine vanadate, a
putative inhibitor of its enzymatic activity. If used below without
specifying its source, the name CPDase refers to the
Arabidopsis enzyme, whereas enzymes from other sources are
explicitly identified. Biophysical characterization of the CPDase from
Arabidopsis, including phosphodiesterase activity
experiments with the wild-type protein and several mutants, is in
progress and will be published elsewhere.
| |
EXPERIMENTAL PROCEDURES |
Purification of Recombinant Protein--
CPDase was expressed in
Escherichia coli strain BL21(DE3) using a construct in
pET11d (5). The recombinant protein features a C-terminal
His6 tag and was purified by affinity chromatography using a Ni2+-nitrilotriacetic acid resin. Typical
expression volumes of 8 liters of bacterial culture yielded about 50 mg
of protein. The purity after affinity chromatography was 92 to
95%.
Preparation of 2',3'-Cyclic Uridine Vanadate--
The protocol
for preparation of 2',3'-cyclic uridine vanadate followed the report of
Borah et al. (10). 6.11 g (25 mmol) of uridine was
dissolved in 1.5 ml of water and mixed with 2.0 g (17 mmol) of
NH4VO3 in 1.5 ml of hot water. Formation of the product is accompanied by yellow coloring of the solution. The substrate was used without further purification.
Crystallization, Data Collection, and Structure
Solution--
Cubic-shaped crystals of semireduced CPDase were grown
in acidic ammonium sulfate conditions (1.2-2.0 M
(NH4)2SO4, 0.1 M NaOAc, pH 5.0) using the vapor diffusion hanging drop method. The drops consisted of 3 µl of reservoir solution, 1 µl of
1,4-dithio-DL-threitol (DTT; final concentration, 12.5 mM), and 3 µl of protein solution (10 mg/ml in 100 mM NaCl, 20 mM Tris-HCl, pH 8.0). Preparation of ligand-bound crystals was attempted by both soaking and
co-crystallization; in the latter case, 1 µl of ligand solution was
added to the crystal drop. These ligands were Appr>p (5 mg/ml),
2',3'-cAMP (10 mg/ml), or U-V (saturated). Growth time of the
semireduced crystals was about 8 weeks. Crystals of the inhibitor-bound
semireduced protein were very small, appeared after 7 months of growth,
and exhibited only limited diffraction power. The crystals were
prepared for cryogenic data collection by flash soaking (<1 min) in
the cryogenic buffer (15% glycerol, 2.0 M
(NH4)2SO4, 0.1 M NaOAc,
pH 5.0). Diffraction data were collected on beamline X9B of the
National Synchrotron Light Source, Brookhaven, NY, equipped with an
ADSC Quantum-4 CCD detector. Data analysis and reduction were performed
with HKL2000 (11); the statistics are summarized in Table I.
Indexing of the diffraction patterns was successful when rhombohedral
crystal symmetry was assumed. The self-rotation function for 
= 180°, calculated with GLRF (12), revealed the presence of
crystallographic 2-fold axes at 
= 30°, 90°, and 150°.
The final space group was thus determined to be R32, with two molecules per asymmetric unit for CNP25 (semireduced CPDase) and with one molecule per asymmetric unit for CNP29 (inhibitor-bound semireduced CPDase). The structures were solved by molecular replacement with the
standalone version of the program AMoRe (13) using the truncated (
90-118) monomer of the oxidized CPDase (9) as a search model. The
solution for CNP25 yielded a correlation coefficient of 0.567 for the
first molecule and 0.789 for the second molecule (R-factor, 0.343). With CNP29, the solution had a correlation coefficient of 0.580 (R-factor, 0.350); the next highest peak was 0.253 (R-factor, 0.486).
Model Building and Refinement--
In CNP25, the two monomers
found in the asymmetric unit are related by a non-crystallographic
symmetry axis almost perpendicular to (001) at the height of
of the z axis. The axis deviates by ~4° from
the (110) direction; this is the most likely reason for the absence of
any significant peaks in the self-rotation function. The initial model
was rebuilt and refined in a number of cycles of visual inspection and
manual adjustments with the program O (14), interspersed with
computational refinement. The latter was carried out using the
conjugate gradient method with CNS 1.0 (15), employing the standard
crystallographic residual target function. Typical protocols consisted
of a positional refinement followed by simulated annealing, grouped and
individual B-factor refinement, and the final positional
refinement. A bulk solvent model and overall anisotropic
B-factor correction were applied throughout the procedure.
After including a manually built water model, the structure of
semireduced CPDase was refined to an R-factor of 0.192 (Rfree = 0.258). Geometrical properties of the
model were analyzed with the program PROCHECK (16).
The refinement of CNP29 followed a similar procedure to the one
described above. However, due to only moderate data quality (rather
incomplete and with extensive ice rings), this structure was difficult
to refine and required repeated cycling of manual adjustments and
computational refinement steps. It has to be noted that the
completeness of reflections used for refinement in the working set is
only 84%, and the number of reflections in the test set was 650. The
final R-factor is 0.213, and we consider the high
Rfree of 0.398 to be reflective of the missing
reflections due to incomplete data and to a model that was not able to
describe all features present in the electron density maps.
Additionally, there seems to be a high degree of flexibility within
this structure, leading to more frequent disordered residues. This
observation is in agreement with the rather high average
B-factor for the data and the model. The model of the U-V
ligand was built with Insight II (17), and its geometry was optimized
with the PM3 Hamiltonian function (18) of MOPAC (17). Parametrization
for refinement was carried out with XPLO2D (19). Table I
summarizes the refinement statistics for
both structures (more details are available as supplementary
material).
Graphical Representation--
Ribbon drawings and graphical
representations of electron densities and protein models presented in
this work were generated by MOLSCRIPT (20) or BOBSCRIPT (21) and were
rendered with POVRay (22) and Raster3d (23).
Urea-induced Unfolding--
The folding stability of
the protein in the presence and absence of reducing conditions was
investigated by urea-induced equilibrium denaturation. The unfolding
process was monitored by intrinsic fluorescence. Samples consisted of
about 2 µM protein in 100 mM NaCl, 20 mM Tris (pH 8.0). The reducing conditions were achieved through the presence of 1 mM -mercaptoethanol. Urea was
present in 17 separate samples with concentrations ranging from 0 to 8 M. The samples were prepared 30 min prior to the
measurements to allow for equilibration. Fluorescence emission spectra
were recorded on a Perkin Elmer LS 50B luminescence spectrometer
using two excitation wavelengths, 
exc = 280 nm and
exc = 295 nm, respectively. All fluorescence spectra
were corrected against buffer-only samples and analyzed offline with
the program AFDP (24). Three independent denaturing series were carried
out for each condition. Each excitation set was analyzed by calculating
an i-curea relation in which i = I(unfolded)/I(folded)
(emission intensity analysis) and a -curea relation
(wavelength analysis). Because two-state unfolding reactions were
observed, it was possible to calculate stability energies as described
by Pace (25); for this purpose, data from the emission intensity
analysis were used.
Circular Dichroism--
Circular dichroic spectra were recorded
with an AVIV 202 spectrometer. The final protein concentrations in the
samples were about 2 µM. For each sample, three separate
CD spectra were collected and averaged and corrected offline with the
program ACDP (24). Correction for each spectrum was against the
respective buffer-only spectrum.
| |
RESULTS |
Following our first report on the structure of the cyclic
nucleotide phosphodiesterase from A. thaliana, we have now
solved the structure of the semireduced and inhibitor-bound forms of this enzyme. CPDase possesses six cysteine residues, four of which are
engaged in disulfide bridges in the oxidized species. Attempts to
reduce the dithioether bonds lead to crystallization conditions very
similar to those for the oxidized species but with DTT present at
concentrations higher than 10 mM. The crystals obtained
from these conditions have a much denser packing arrangement than the ones obtained with the oxidized CPDase species. At the same time, the
space group changes from P6122 (oxidized form) to R32
(semireduced form) in a polymorphic transition. This is a very
conservative transition because the rhombohedral system is a subgroup
of the hexagonal crystal system. Accordingly, the non-crystallographic symmetry trimer observed with the hexagonal structure has now become a
crystallographic trimer centered on the 3-fold axis.
Structure of the Semireduced CPDase (CNP25)--
In the oxidized
CPDase structure, one of the disulfide bridges (Cys-104-Cys-110) is
located in an exposed surface loop with the bridge separating the two
lobes of this loop (100). The dithioether bond in the loop region
is absent in the new structure, and the former loop conformation is
reorganized to elongate helix 3 by two turns (see Fig.
1B), which moves residues
102-115 much closer toward the protein core. This reorganization also
forces Cys-104 and Cys-110 into completely different environments,
preventing any contact with each other (Figs. 1D and
2). The second disulfide bridge
(Cys-64-Cys-177) connects helix 2 with the terminal strand 7.
Despite the high concentration of DTT in the crystallization drop, this
cystine is found to be unchanged compared with the oxidized CPDase
structure. Because one of two disulfide bridges is still present in the
current crystal structure, we name this structure type
semireduced.
View larger version (41K):
[in this window]
[in a new window]
|
Fig. 1.
Refolding of the loop region of CPDase upon
reduction. The top panel shows front views of
oxidized (A) and semireduced (B) CPDase; residues
of the tandem motif (His-42, Thr-44, His-119, Ser-121) are depicted
explicitly. Shown on the bottom panel are side views of the
oxidized (C) and semireduced (D) CPDase with
locations of all six cysteine residues. The color coding is
green for the terminal lobe, blue for the transit
lobe, and red for the loop region.
|
|
View larger version (41K):
[in this window]
[in a new window]
|
Fig. 2.
Disulfide bridges in the oxidized and
semireduced states. Stereo pictures of the 2Fo  Fc electron density contoured at 1  around the
cysteine residues Cys-64/Cys-177 (upper left part) and
Cys-104/Cys-110 (lower part). The upper panel
shows the oxidized, and the lower panel shows the
semireduced CPDase structure, respectively.
|
|
Superposition of the oxidized CPDase (molecule 1) and the
semireduced species (molecule 1) shows only minor deviations in the
protein core, mainly around the stretch of residues 161-171. The
refolding of the loop segment into a helical structure does not affect
the overall conformation of the core, as reflected by a root mean
square deviation of 0.63 Å between the oxidized and the semireduced
structure. For comparison, the two non-crystallographic symmetry-related molecules of the semireduced structure show almost the
same root mean square deviation, namely 0.67 Å (Table
II).
Structure of the Inhibitor-bound Semireduced CPDase--
The two
data sets obtained from the semireduced CPDase co-crystallized with U-V
are of moderate quality and not entirely complete. However, the length
of time necessary for crystal growth and the paucity of available
crystals prevented us from improving data quality. Flexibility and
disorder in this structure added further to the difficulties in
refinement and model building. Nevertheless, the density clearly shows
conserved conformation for the core residues and of the active site
cleft. The structure in helix 3 and the loop region is similar to
that observed with the semireduced CPDase without a ligand, although
the local conformations seem to be different. The density in this
region indicates considerable disorder, and residues Asn-106 and
Phe-108 were modeled as alanines because of insufficient information
about their side chain positions. Extra density for U-V was clearly
visible within the active site, even at the early stages of refinement.
The vanadium atom assumes the position of the sulfate sulfur atom
observed in the oxidized species, whereas residues His-42, Thr-44,
His-119, Ser-121, and Tyr-124 coordinate all vanadate oxygen atoms (see
Fig. 3). The 5'-hydroxyl group of the
ligand interacts with Trp-12 and Ser-10, and the imide nitrogen of the
uracilyl group is hydrogen-bonded to the backbone carbonyl of Thr-163.
Further stabilization of the nucleotide base is provided by hydrophobic
contacts with Phe-84 and Trp-171. The density observed for the
nucleotide base allows for two other conformations, indicating a
possible rotation around the C1-N bond, which connects the base to the
ribosyl body. However, these two alternate conformations were not
modeled in this structure.
View larger version (42K):
[in this window]
[in a new window]
|
Fig. 3.
Inhibitor-bound form of CPDase.
Top panel, stereo view of the active site cavity with
superposition of important residues in the three CPDase structures:
oxidized (blue), semireduced (green), and semireduced,
inhibitor-bound (orange). The U-V ligand is depicted in
red. Bottom panel, stereo picture of the
2Fo Fc electron
density contoured at 1 around the U-V ligand within the active
site.
|
|
The structure reported here also features a secondary binding site for
U-V, which is found on the surface of the protein. The ligand is
coordinated there by Arg-31, the backbone carbonyl of residue 36, and
the backbone amide of residue 38, and it occupies a position on the
crystallographic 3-fold axis. The density of this site is not very well
defined for the nucleotide base but is apparent for the rest of the ligand.
With respect to data quality, we expressly note that the information
obtained from this structure is limited to the binding mode of the
nucleotide. The conclusions drawn from the observed ligand-protein
interactions are valid because the density within the active site
region and the surrounding protein core is unambiguous.
Folding Stability--
Because the structures of both the fully
oxidized and a semireduced species of CNPase are now available, it is
necessary to analyze the respective stabilities of these two distinct
forms of the protein. Although the existence of a semireduced state of
CPDase in the presence of 1 mM -mercaptoethanol is not
proven experimentally, the assumption of its presence is reasonable. For both species, we find similar stability parameters (Table III) and also comparable intrinsic
fluorescence behavior (data not shown). With the semireduced species,
the wavelength analysis of the unfolding series shows some anomaly in
the low concentration range. At 280 nm excitation, it appears as if a
first transition takes place at c1/2 = 0.9 M,
whereas the main transition is still observed at c1/2 = 3.9 M. Data collected at 295 nm excitation are unstable in the low concentration range but do not allow for a conclusion of an extra
transition. Intensity analysis, by contrast, shows a clear two-state
unfolding reaction without any anomalies. We therefore conclude that
some side reactions might occur with fluorophore groups affecting the
emission wavelength. However, the unfolding behavior of CPDase displays
clear two-state characteristics in both the oxidized and semireduced
redox state.
Substrate and DTT Effects on the Protein Structure as Measured by
CD--
Using the molar ellipticity at 222 and 208 nm, CPDase does not
show significant differences in the presence or absence of either
Appr>p or DTT. Calculation of secondary structure contents with the
program CDNN (26) shows no differences when compared with the native CD
spectrum (data not shown). The overall increase of -helical content
by elongation of helix 3 is ~4%, which is presumably too small to
be detected by CD.
| |
DISCUSSION |
Following our initial work on the structure of the native oxidized
form of CPDase from A. thaliana, we undertook extensive efforts to obtain ligand-bound CPDase structures. However, soaking or
co-crystallization with Appr>p, 2',3'-cAMP, and U-V proved unsuccessful (data not shown). The hypothesis evolved that this failure
might be due to our inability to grow crystals under conditions other
than acidic or of changing pH of the mother liquor without dissolving
crystals. Cross-linking of crystals obtained under acidic conditions
and rebuffering at higher pH values led to severe crystal damage and/or
loss of diffraction (data not shown).
At the same time, we set up experiments targeting reduction of the two
disulfide bridges found in the oxidized CPDase structure. After
obtaining a new crystal form in the presence of DTT, which also
promised higher crystal quality than the oxidized species, we again
attempted to obtain a ligand-bound structure. However, crystal growth
was extremely slow, and only two crystals became available for
diffraction experiments.
The crystal structure obtained from this new crystal form exhibits a
conformational change within the loop region in which one of the two
cystines (Cys-104-Cys-110) is located (see Fig. 1). Although this
bridge is reduced and the N-terminal lobe of the loop winds up into
helical turns to elongate helix 3, the other cystine
(Cys-64-Cys-177) remains oxidized; the protein is in its semireduced
redox state. This surprising finding can be interpreted as a balance
between two opposing forces: a conformational force that is aimed at
the energy-minimized structure and a redox force that is driven by a
combination of redox potentials. The loop was identified even in the
first (oxidized) structure as a very flexible region of the protein.
DTT exhibits a liberating effect on the disulfide bridge
Cys-104-Cys-110 because the covalent connection ties the loop into a
certain (coil-)conformation that switches into a helical structure upon
release of the strain. This increases the energetic stabilization
because the helix conformation is very favorable and even more because
an existing helix (3) becomes elongated. The second disulfide bridge
remains intact, although it should have been accessible to the reducing
agent. The local environment shows that -strand 7 anneals very
smoothly to helix 2, bringing Cys-177 into very close contact with
Cys-64. Upon the opening of the covalent bond, the two sulfur atoms
would still be in close vicinity if one assumes that the protein core does not change its overall conformation. Additionally, Phe-60 provides
aromatic shielding to the cystine (distance: 3.8 Å), thereby
stabilizing the covalent bond. Holding the conformation tightly
together, the protein almost forces this disulfide bridge to be formed.
This scenario is strongly supported by the results from the unfolding
experiments. Although we cannot prove that the experimental setup of 1 mM -mercaptoethanol does indeed generate the semireduced and not the fully reduced species, the outcome is not impaired at all.
CPDase displays the same unfolding characteristics in the presence or
absence of a reducing agent as seen from the stability analysis
(cf. Table III). Furthermore, the intrinsic fluorescence behavior in both experiments is indistinguishable (not shown). This
allows for the conclusion that the fluorophores, which are all located
in the protein core, assume the same conformation in both experimental
environments, thus suggesting that the overall structure of the central
part of the protein does not depend on the oxidation state of the
disulfide bridges.
An analogous situation has been reported previously for human
epidermal-type fatty acid-binding protein (FABP). Hohoff et al. (27) describe the crystal structure of FABP, which contains six cysteine residues, similarly to CPDase. Although two of these cysteines are in isolated positions, the other four are paired in close
vicinity. Despite the absence of any reducing agent in the
crystallization media, these authors (27) report one oxidized (Cys-120-Cys-127) and one reduced (Cys-67-Cys-87) cystine. The remarkable feature in the case of FABP is that Cys-67 and Cys-87 are
still located close enough to be able to form a covalent bond without
any structural rearrangements. This is, of course, different than
CPDase, in which the semireduced species displays a complete conformational rearrangement of the loop segment leading to the movement of both cysteine residues a long distance apart. Very recently, a similar phenomenon has been reported with the OxyR transcription factor in which the redox switch from the oxidized to the
reduced form of Cys-199-Cys-208 results in structural changes within
the regulatory domain (28); an -helical turn is formed while a
-strand downstream transforms into a coiled conformation. Unlike
CPDase, however, in the case of OxyR, these changes lead to different
oligomeric associations. Because of the importance of this process for
protein regulation, this phenomenon has been termed "fold editing."
The main effect of disulfide cross-links in proteins is a decrease in
conformational entropy. However, this can happen by limiting the
conformational freedom of the unfolded peptide chain or by local
interactions in the folded state or a combination of both. The impact
of disulfide bonds on the stability of proteins has been investigated
in great detail in the past. The effect of mutations, substituting one
or both of the cysteine residues, is frequently studied in different
systems. The inability of the protein to form the dithioether bond
because of a mutational substitution is always accompanied by a loss in
stability energy (e.g. see Ref. 29). Looking at different
redox states, it has been found with RNase A that the two cystines
connecting the N and C termini have a much more enhancing effect
on protein stability than the two embedded cystines in the core (30).
Zhang et al. (31) presented a detailed study on phage T4
lysozyme to assess the relation of the conformational entropy of
cystines with the ring size of the dithioether-linked loop. Three
different disulfide-containing constructs were characterized in their
oxidized and reduced states, and a decrease in stability for all
reduced states was found compared with the respective oxidated species.
Bovine -lactalbumine, a close homologue of chicken lysozyme, has
four disulfide cross-links. A kinetic study on the reduction of these
cystines revealed that there is one disulfide bond being reduced much
faster than the other and that this superreactivity is most likely due
to the geometric strain imposed by the local conformation in the
oxidized state (32). In this context, CPDase behaves exceptionally in that the reduction of the cystine formed by Cys-104-Cys-110 does not
lead to a considerable decrease in folding stability. It seems extremely likely, though, that the loss of stability is compensated by
the release of conformational strain within the loop region of the
oxidized species and the gain in stability when rearranging from a
coiled into a helical structure. This argument can also be extended to
explain the remaining cystine, Cys-64-Cys-177, which finds itself in a
very stabilized environment in which the secondary structure elements
and the neighboring residues provide an unrestrained conformation,
making the formation of the dithioether bond very favorable.
CPDase has been shown to hydrolyze Appr>p to Appr-1"p but also
nucleoside 2',3'-cyclic phosphates (N>p) to nucleoside 2'-phosphates (N-2'p). The usage of U-V in this context resembles a 2',3'-cyclic phosphate substrate and enables valid conclusions because CPDase can
also process N>p substrates (5). The binding mode of U-V supports and
is in full agreement with the proposed catalytic mechanism (9).
Although correct positioning of the ligand is ensured by Ser-10,
Trp-12, and Thr-163 through interactions with peripheral groups of the
substrate molecule, the cyclic vanadate substituting for the cyclic
phosphate is held in place and coordinated by the residues of the
tandem signature motif. Thr-44 and Ser-121 act as stabilizing groups by
coordinating the vanadate oxygen atoms. Tyr-124, which is not part of
the signature motif but was postulated previously to take part in the
catalytic reaction (9), is also coordinating the vanadate. The
supporting role for these residues agrees with the proposed catalytic
reaction mechanism because His-119 is believed to act as the initial
base. Activated by the backbone carbonyl of Met-117, the base abstracts
a proton from a conserved water molecule, thereby generating the
attacking nucleophilic hydroxide ion. Both histidine residues, His-42
and His-119, are part of a coordination network with water and ligand atoms, which supports their roles as catalytic acid and base, respectively. No tight specific interactions are seen between the
ribose ring and protein residues, allowing for a flexible positioning
of this part, which is necessary to accommodate either 2',3'-cyclic
nucleotides or 1",2"-cyclic phosphates. The nucleobase of U-V sitting
in the 1'-position (or "meta" with respect to the cyclic
vanadate) extends toward the protein residues lining the entrance path
of the active site cleft. Taking the vanadate position as fixed,
Appr>p, which is a 1",2"-cyclic phosphate, would fit sterically into
the active site by aligning the five-membered ribose rings of U-V and
Appr>p, but the coordination pattern has to be different because the
constitution of the ribose ring is different in Appr>p. Also, the AMP
extension, which connects to the 5"-position of the cyclic phosphate
ribose ("para" position), would run along the entrance
path of the active site, which is the only way to fit this type of
substrate sterically into the space provided by the protein cleft.
| |
CONCLUSION |
The present work provides further insights into the structural
properties of CPDase from A. thaliana, which was the first enzyme out of a new family of phosphodiesterases for which a crystal structure has been reported. Although the first structure of CPDase (9)
was for the native (oxidized) species, we were now able to obtain the
structure of CPDase in the semireduced state. The protein undergoes a
redox-induced conformational rearrangement in the loop region because
of the release of sterical strain from a cystine. The phenomenon that a
second cystine of CPDase was not influenced by the change in the redox
potential of its environment is most likely due to the conformational
force of the protein, which tightly holds both participating cysteine
residues in place. Admittedly, one cannot exclude that this
conformational force is not an intrinsic property of the protein but an
artifact generated by crystal packing forces. However, unfolding
studies of both the oxidized and the (semi-)reduced species show no
differences of stability parameters or intrinsic protein fluorescence
behavior. We therefore conclude that this conformational force is most
likely a fold-associated feature of CPDase.
The structure of CPDase in a complex with its putative inhibitor
uridine vanadate is in excellent agreement with the proposed catalytic
mechanism and adds further weight to its validity. In an earlier
report, we speculated that Tyr-124, which is not part of the tandem
signature motif, might play a role in the enzymatic reaction (9). This
hypothesis is supported by the current study; moreover, Ser-10 was
identified as an additional binding residue for 2',3'-cyclic nucleotide
substrates by CPDase. The preference of CPDase for Appr>p rather than
2',3'-cyclic nucleotides is most likely due to steric considerations,
which are more in favor of a para- than a
meta-arrangement of cyclic phosphate and a ribosyl extension
group. Experiments aimed at investigation of the enzymatic functions
and the effect of mutations are currently in progress.
| |
ACKNOWLEDGEMENTS |
We thank George Pavlakis for access to the
luminescence spectrometer and Zbigniew Dauter for advice on data
collection and analysis. The Friedrich Miescher-Institut is part
of the Novartis Research Foundation.
| |
FOOTNOTES |
*
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
This work is dedicated to Robert Huber on the occasion of his
65th birthday.
The on-line version of this article (available at
http://www.jbc.org) contains Table IV, which provides additional
details about the refinement statistics for CNP25 and CNP29.
The atomic coordinates and the structure factors (code 1jh6, 1jh7) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).
§
To whom correspondence should be sent. Tel.: 301-846-5033; Fax:
301-846-7101; E-mail: hofmanna@ncifcrf.gov.
¶
Present address: Commonwealth Biotechnologies, Inc.,
601 Biotech Drive, Richmond, VA 23235.
Published, JBC Papers in Press, November 1, 2001, DOI 10.1074/jbc.M107889200
| |
ABBREVIATIONS |
The abbreviations used are:
Appr>p, adenosine-diphosphate ribose-1",2"-cyclic phosphate;
CPDase, cyclic
nucleotide phosphodiesterase;
U-V, 2',3'-cyclic uridine vanadate;
2', 3'-cAMP, adenosine 2',3'-cyclic phosphate;
(N>p), nucleoside
2',3'-cyclic phosphates;
(N-2'p), nucleoside 2'-phosphates;
DTT, 1,4-dithio-DL-threitol.
| |
REFERENCES |
| 1.
|
Culver, G.,
McCraith, S.,
Zillmann, M.,
Kierzek, R.,
Michaud, N.,
LaReau, R.,
Turner, D.,
and Phizicky, E.
(1993)
Science
261,
206-208
|
| 2.
|
Zillmann, M.,
Gorovsky, M. A.,
and Phizicky, E. M.
(1991)
Mol. Cell. Biol.
11,
5410-5416
|
| 3.
|
Zillmann, M.,
Gorovsky, M.,
and Phizicky, E.
(1992)
J. Biol. Chem.
267,
10289-10294
|
| 4.
|
Culver, G.,
Consaul, S.,
Tycowski, K.,
Filipowicz, W.,
and Phizicky, E.
(1994)
J. Biol. Chem.
269,
24928-24934
|
| 5.
|
Genschik, P.,
Hall, J.,
and Filipowicz, W.
(1997)
J. Biol. Chem.
272,
13211-13219
|
| 6.
|
Tyc, K.,
Kellenberger, C.,
and Filipowicz, W.
(1987)
J. Biol. Chem.
262,
12994-13000
|
| 7.
|
Martzen, M. R.,
McCraith, S. M.,
Spinelli, S. L.,
Torres, F. M.,
Fields, S.,
Grayhack, E. J.,
and Phizicki, E. M.
(1999)
Science
286,
1153-1155
|
| 8.
|
Nasr, F.,
and Filipowicz, W.
(2000)
Nucleic Acids Res.
28,
1676-1683
|
| 9.
|
Hofmann, A.,
Zdanov, A.,
Genschik, P.,
Ruvinov, S.,
Filipowicz, W.,
and Wlodawer, A.
(2000)
EMBO J.
19,
6207-6217
|
| 10.
|
Borah, B.,
Chen, C. W.,
Egan, W.,
Miller, M.,
Wlodawer, A.,
and Cohen, J. S.
(1985)
Biochemistry
24,
2058-2067
|
| 11.
|
Otwinowski, Z.,
and Minor, W.
(1997)
Methods Enzymol.
276,
307-326
|
| 12.
|
Tong, L.,
and Rossmann, M.
(1990)
Acta Crystallogr. Sect. A
46,
783-792
|
| 13.
|
Navaza, J.
(1994)
Acta Crystallogr. Sect. A
50,
157-163
|
| 14.
|
Jones, T. A.,
Zou, J. Y.,
Cowan, S.,
and Kjeldgaard, M.
(1991)
Acta Crystallogr. Sect. A
47,
110-119
|
| 15.
|
Brünger, A. T.,
Adams, P. D.,
Core, G. M.,
DeLano, W. L.,
Gros, P.,
Grosse-Kunstleve, R. W.,
Jiang, J. S.,
Kuszewski, J.,
Nilges, M.,
Pannu, N. S.,
Read, R. J.,
Rice, L. M.,
Simonson, T.,
and Warren, G. L.
(1998)
Acta Crystallogr. Sect. D Biol. Crystallogr.
54,
905-921
|
| 16.
|
Laskowski, R. A.,
MacArthur, M. W.,
Moss, D. S.,
and Thornton, J. M.
(1993)
J. Appl. Crystallogr.
26,
283-291
|
| 17.
| Molecular Simulations, Inc., Insight II, version 2000,
San Diego, CA
|
| 18.
|
Stewart, J. J. P.
(1990)
J. Comput Aided Mol. Des.
4,
1-105
|
| 19.
|
Kleywegt, G. J.,
and Jones, T. A.
(1997)
Methods Enzymol.
277,
208-230
|
| 20.
|
Kraulis, P. J.
(1991)
J. Appl. Crystallogr.
24,
946-950
|
| 21.
|
Esnouf, R. M.
(1999)
Acta Crystallogr. Sect. D Biol. Crystallogr.
55,
938-940
|
| 22.
| Persistence of Vision Development Team (1999) POVRay
(www.povray.org), 3.1 9
|
| 23.
|
Merrit, E. A.,
and Bacon, D. J.
(1997)
Methods Enzymol.
277,
505-524
|
| 24.
| Hofmann, A. & Wlodawer, A. (2002) Bioinformatics,
in press
|
| 25.
|
Pace, C. N.
(1995)
Methods Enzymol.
259,
538-554
|
| 26.
|
Böhm, G.
(1997)
CDNN: CD Spectra Deconvolution, Version 2.1.
, Universität Halle, Halle, Germany
|
| 27.
|
Hohoff, C.,
Börchers, T.,
Rüstow, B.,
Spener, F.,
and van Tilbeurgh, H.
(1999)
Biochemistry
38,
12229-12239
|
| 28.
|
Choi, H. J.,
Kim, S. J.,
Mukhopadhyay, P.,
Cho, S.,
Woo, J. R.,
Storz, G.,
and Ryu, S. E.
(2001)
Cell
105,
103-113
|
| 29.
|
Liu, Y.,
Breslauer, K.,
and Anderson, S.
(1997)
Biochemistry
36,
5323-5335
|
| 30.
|
Klink, T. A.,
Woycechowsky, K. J.,
Taylor, K. M.,
and Raines, R. T.
(2000)
Eur. J. Biochem.
267,
566-572
|
| 31.
|
Zhang, T.,
Bertelsen, E.,
and Alber, T.
(1994)
Nat. Struct. Biol.
1,
434-438
|
| 32.
|
Kuwajima, K.,
Ikeguchi, M.,
Sugawara, T.,
Hiraoka, Y.,
and Sugai, S.
(1990)
Biochemistry
29,
8240-8249
|
| 33.
|
Brünger, A. T.
(1992)
Nature
355,
472-474
|
Copyright © 2002 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
B. Schwer, A. Aronova, A. Ramirez, P. Braun, and S. Shuman
Mammalian 2',3' cyclic nucleotide phosphodiesterase (CNP) can function as a tRNA splicing enzyme in vivo
RNA,
February 1, 2008;
14(2):
204 - 210.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
L. K. Wang, B. Schwer, M. Englert, H. Beier, and S. Shuman
Structure-function analysis of the kinase-CPD domain of yeast tRNA ligase (Trl1) and requirements for complementation of tRNA splicing by a plant Trl1 homolog
Nucleic Acids Res.,
January 20, 2006;
34(2):
517 - 527.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Kozlov, J. Lee, D. Elias, M. Gravel, P. Gutierrez, I. Ekiel, P. E. Braun, and K. Gehring
Structural Evidence That Brain Cyclic Nucleotide Phosphodiesterase Is a Member of the 2H Phosphodiesterase Superfamily
J. Biol. Chem.,
November 14, 2003;
278(46):
46021 - 46028.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. Mazumder, L. M. Iyer, S. Vasudevan, and L. Aravind
Detection of novel members, structure-function analysis and evolutionary classification of the 2H phosphoesterase superfamily
Nucleic Acids Res.,
December 1, 2002;
30(23):
5229 - 5243.
[Abstract]
[Full Text]
[PDF]
|
|
|
Copyright © 2002 by the American Society for Biochemistry and Molecular Biology.
TOP
ABSTRACT
INTRODUCTION
