Acta Cryst. (2009). F65, 111–115doi:10.1107/S1744309108040578
Acta Crystallographica Section F
Crystallization and preliminary crystallographic
characterization of the extrinsic PsbP protein of
photosystem II from Spinacia oleracea
J. Kohoutova ´,aI. Kuta ´
Smatanova ´,a,bJ. Brynda,b,c
M. Lapkouski,a,bJ. L. Revuelta,d
J. B. Arellanoeand R. Ettricha,b*
aInstitute of Systems Biology and Ecology,
Academy of Sciences of the Czech Republic,
Za ´mek 136, 37333 Nove ´ Hrady, Czech
Republic,bInstitute of Physical Biology,
University of South Bohemia, Za ´mek 136,
37333 Nove ´ Hrady, Czech Republic,cInstitute
of Molecular Genetics and Institute of Organic
Chemistry and Biochemistry, Academy of
Sciences of the Czech Republic, Flemingovo
nam. 2, 166 37 Praha 6, Czech Republic,
dDepartamento de Microbiologı ´a y Gene ´tica,
Instituto de Microbiologı ´a Bioquı ´mica,
Universidad de Salamanca/CSIC, Campus
Miguel de Unamuno, 37007 Salamanca, Spain,
andeDepartamento de Estre ´s Abio ´tico, Instituto
de Recursos Naturales y Agrobiologı ´a de
Salamanca (IRNASA–CSIC), Apartado 257,
37071 Salamanca, Spain
Correspondence e-mail: email@example.com
Received 9 October 2008
Accepted 2 December 2008
Preliminary X-ray diffraction analysis of the extrinsic PsbP protein of photo-
system II from spinach (Spinacia oleracea) was performed using N-terminally
His-tagged recombinant PsbP protein overexpressed in Escherichia coli.
Recombinant PsbP protein (thrombin-digested recombinant His-tagged PsbP)
stored in bis-Tris buffer pH 6.00 was crystallized using the sitting-drop vapour-
diffusion technique with PEG 550 MME as a precipitant and zinc sulfate as an
additive. SDS–PAGE analysis of a dissolved crystal showed that the crystals did
not contain the degradation products of recombinant PsbP protein. PsbP
crystals diffracted to 2.06 A˚resolution in space group P212121, with unit-cell
parameters a = 38.68, b = 46.73, c = 88.9 A˚.
Oxygenic photosynthesis is the process by which light energy is
converted into chemical energy. This process takes place in the
thylakoid membrane of higher plants, algae and cyanobacteria, in
which the membrane-embedded pigment–protein complex photo-
system II (PSII) performs light-driven oxidation of water with con-
comitant reduction of the plastoquinone pool. As a result of the light-
driven redox reaction, molecular oxygen is released as a byproduct
from the oxygen-evolving complex (OEC) located on the lumenal
side of PSII (Barber, 2003). The extrinsic proteins stabilizing the
OEC catalytic centre differ in prokaryotic and eukaryotic oxyphoto-
trophs (Seidler, 1996; De Las Rivas et al., 2004). The OEC of higher
plants consists of an inorganic Mn4Ca cluster and extrinsic proteins
named PsbO, PsbP, PsbQ and PsbR, which create the correct ionic
environment during water oxidation. The PSII of nongreen algae and
cyanobacteria also includes PsbO, but two different extrinsic proteins,
PsbU and PsbV, are associated with the OEC catalytic centre
(Seidler, 1996; Suorsa et al., 2006). For a better understanding of the
water splitting, knowledge of the structure of PSII and its extrinsic
proteins is an essential prerequisite. The recently determined three-
dimensional X-ray structure of a cyanobacterial PSII has notably
improved upon the previous partial structures of the bacterial
complex (Zouni et al., 2001; Kamiya & Shen, 2003; Loll et al., 2005).
However,these cyanobacterial PSII structures provide no clues to the
possible arrangement of the extrinsic PsbP and PsbQ proteins on the
lumenal side of higher plant PSII. While PsbO is ubiquitous to all
known oxyphototrophs, PsbP and PsbQ are not. Moreover, PsbU and
PsbVare not homologous in sequence or structure to PsbP or PsbQ,
although they play the same functions as the PsbP and PsbQ proteins
of the higher plant and green algal PSII (Shen & Inoue, 1993). In vitro
studies have demonstrated that PsbP and PsbQ are involved in Ca2+
and Cl?retention in PSII and also form a barrier that is open to
substrates and products but closed to exogenous reductants (Seidler,
1996). More recently, the physiological roles of PsbP and PsbQ have
been tested in transgenic plants in which the levels of these two
proteins were severely downregulated. Whereas PsbP was demon-
strated to be essential for the regulation and stability of PSII (Ifuku,
Yamamoto et al., 2005; Yi et al., 2007), PsbQ was found to be
dispensable (Ifuku, Yamamoto et al., 2005). In recent years, efforts to
# 2009 International Union of Crystallography
All rights reserved
elucidate the X-ray structures of PsbP and PsbQ have been made in
order to shed further light on their role in the OEC of green plant
PSII. The crystal structure of PsbQ from Spinacia oleracea has been
solved to 1.49 A˚resolution (Balsera et al., 2005). To date, recombi-
nant PsbP proteins from several plant species (spinach, cucumber and
tobacco) have been tested for crystallization, but only the crystal
structure of PsbP protein from Nicotiana tabacum has been solved (to
1.60 A˚ resolution; Ifuku et al., 2004). Partial degradation at the
N-terminus of the protein was observed in these PsbP crystals (Ifuku
et al., 2003). However, the region prone to degradation is reported to
be functionally relevant (Ifuku, Nakatsu et al., 2005). Here, we report
the successful purification and crystallization of recombinant PsbP
protein from S. oleracea without partial degradation.
2. Materials and methods
2.1. Isolation of spinach PsbP protein overexpressed in Escherichia
coli with a His anchor
Coding regions for the mature PsbP protein were amplified by
high-fidelity PCR using the clone pSoc23.81E1 as a template (Jansen
et al., 1987). This clone was a kind gift from Professor R. G. Herr-
mann. The PCR primers used were 50-CCATATGGCCTATGGA-
GAAGCTGCTAAT-30(forward) and 50-GGGATCCTTAAGCAA-
CACTGAAAGAACT-30(reverse) containing restriction sites for
NdeI and BamHI (bold nucleotides) and a start codon and a termi-
nation anticodon (italicized nucleotides), respectively. The resulting
PCR product was purified using the GFX PCR DNA and gel band
purification kit (GE Healthcare Biosciences AB, Uppsala, Sweden)
and cloned into EcoRV-digested Bluescript II SK+vector (BSK+).
The PCR-amplified psbP insert was sequenced in both directions in
an automated sequencer using the reverse and universal primers to
verify the nucleotide sequence. The plasmid DNA containing the
psbP insert was consecutively cut with NdeI and BamHI and cloned
into pET-12a (Unigene) pre-digested with NdeI and BamHI. The
resulting construction was named JR2591. The chosen forward
primer including an NdeI restriction site introduced a methionine
residue in the first N-terminal position of the recombinant PsbP
protein. JR2591 was transferred into E. coli BL21(DE3)pLysS. The
region of the upstream leader ompT of pET12a, encoding the outer
membrane protein specified by ompT, was removed when using the
restriction enzymes NdeI and BamHI. Transformed cells named B95
were stored at 193 K in a 20% glycerol solution. The PsbP expression
resulting from JR2591 was minimal regardless of temperature or the
optical density of the cell culture after IPTG induction. As a conse-
quence of the poor expression, the pET-28b+ vector (Novagen) was
chosen as an alternative expression system. BSK+containing the
psbP insert was subsequently cut with NdeI and BamHI as described
above and cloned into a pET-28b+ vector pre-digested with NdeI and
BamHI. The resulting construction was named JR3133. In this case,
the recombinant PsbP protein contained a His-tagged sequence and a
thrombin cleavage site in the N-terminal region. JR3133 was trans-
ferred into E. coli BL21(DE3)pLysS. Transformed cells named B152
were stored at 193 K in a 20% glycerol solution.
2.2. Purification of recombinant PsbP and His-tagged PsbP proteins
B152 cells containing the JR3133 construct were grown at 310 K in
LB medium supplemented with 50 mg ml?1kanamycin. When the
optical density of the culture at 580 nm reached a value of 0.6,
overexpression of the His-tagged recombinant PsbP protein was
initiated by adding 1 mM IPTG. The cells were incubated for 18 h at
303 K and then harvested by centrifugation, suspended in 20 mM
potassium phosphate, 1 mM EDTA, 0.5 M NaCl pH 7.4 (buffer A)
and finally passed through a French press. After removing the un-
broken material by centrifugation at 15 000g for 45 min at 277 K, the
protein suspension was loaded onto an Ni Sepharose High Perfor-
mance column (GE Healthcare Biosciences AB, Uppsala, Sweden)
pre-equilibrated with buffer A. The His-tagged recombinant PsbP
protein was eluted with a linear gradient of increasing imidazole
concentration from 0 to 0.5 M in 20 mM potassium phosphate, 1 mM
EDTA, 0.5 M NaCl, 0.5 M imidazole pH 7.4 (buffer B). Fractions
enriched in the His-tagged recombinant PsbP protein were pooled,
concentrated and washed with 20 mM bis-Tris, 1 mM EDTA pH 6.0
(buffer C) using centrifugal filter devices (Amicon Ultra 10 000
molecular-weight cutoff, 15 ml capacity) from Millipore (Billerica,
Massachusetts, USA). The His-tagged recombinant PsbP protein was
loaded onto an SP Sepharose Fast Flow cation-exchange column (GE
Healthcare Biosciences AB, Uppsala, Sweden) pre-equilibrated with
buffer C and eluted with a linear gradient of increasing salt
concentration from 0 to 0.7 M in 20 mM bis-Tris, 1 mM EDTA, 1 M
NaCl pH 6.0 (buffer D). Fractions enriched in the His-tagged
recombinant PsbP protein were pooled, washed with 20 mM bis-Tris,
0.1 M NaCl pH 6.0 (buffer E) and finally concentrated to 15 mg ml?1
using centrifugal filter devices (Amicon Ultra 10 000 molecular-
weight cutoff, 15 ml capacity) from Millipore (Billerica, Massachu-
setts, USA). The first 17 amino acids of the N-terminal region of the
His-tagged recombinant PsbP protein were enzymatically removed
with thrombin (Calbiochem, EMD Biosciences Inc., San Diego,
California, USA) by incubating 5 mg of protein per unit of protease
in buffer E for 3 h at room temperature. As the result of protease
cleavage, the His-tagged sequence was removed and the resulting
recombinant PsbP protein contained four additional amino acids (i.e.
GSHM) at the N-terminus compared with the native PsbP protein
from spinach. The recombinant PsbP protein was passed through a
Superdex 75 column (GE Healthcare Biosciences AB, Uppsala,
Sweden) pre-equilibrated with 20 mM bis-Tris, 1 mM EDTA, 0.2 M
NaCl pH 6.0 (buffer F) at a flow rate of 0.2 ml min?1. Fractions
containing the recombinant PsbP protein were collected and
concentrated to 15 mg ml?1using centrifugal filter devices (Amicon
Ultra 10 000 molecular-weight cutoff, 4 ml capacity) from Millipore
(Billerica, Massachusetts, USA). SDS–PAGE according to Laemmli
(1970) with a total acrylamide content of 12% in the separating gel
was performed to test the purity of the recombinant PsbP protein.
The gel was stained with Coomassie R-250.
2.3. Protein crystallization and data collection
The recombinant PsbP protein was washed with 20 mM bis-Tris pH
6.0 and concentrated to a final concentration of about 15 mg ml?1
using centrifugal filter devices (Amicon Ultra 10 000 molecular-
weight cutoff, 4 ml capacity) from Millipore (Billerica, Massachusetts,
USA). The protein was crystallized at room temperature using the
sitting-drop vapour-diffusion technique. The Crystallization Basic Kit
for Proteins and the Crystallization Extension Kit for Proteins
(Sigma) were used to find initial crystallization conditions. A 2 ml
volume of protein solution (15 mg ml?1) was mixed with 2 ml of
various reservoir solutions and equilibrated against 700 ml reservoir
solution. Screening showed that PEG 550 MME, PEG 6000 and PEG
8000 were effective precipitants. Univalent and bivalent ions were
tested as additives in subsequent experiments. The conditions were
optimized by varying the protein concentration, additives and buffer
systems (type, concentration and pH).
Crystals were tested using the in-house X-ray diffractometer of the
Institute of Organic Chemistry and Biochemistry and the Institute of
Kohoutova ´ et al.
Acta Cryst. (2009). F65, 111–115
Molecular Genetics of the Academy of the Sciences of the Czech
Republic in Prague. Protein crystals were measured and complete
data sets were collected using an in-house X-ray diffractometer
constituted of an FR591 rotating-anode X-ray generator (Bruker–
Nonius), a double-mirror X-ray optical system (XOS), a MAR 345
image-plate detector and a Cryostream cooling system (Oxford
Cryosystem) using Cu K? radiation. Crystals were cryocooled in a
nitrogen-gas stream at 110 K. A total of 231?of diffraction images
were collected, each covering 0.5?crystal rotation. The measured
data were integrated using MOSFLM (Leslie, 1999) and scaled using
SCALA (Evans, 1997), both from the CCP4 package (Collaborative
Computational Project, Number 4, 1994).
3. Results and discussion
Successful purification and crystallization of the recombinant PsbP
protein from S. oleracea is reported in this paper. Protein purification
began with overexpression of the recombinant PsbP protein with a
six-His tag at the N-terminus (His-tagged PsbP protein), which was
followed by a series of chromatographic steps including the
application of metal-affinity and ion-exchange chromatography. Size-
exclusion chromatography was added as the final step of the protein-
purification procedure to separate the recombinant PsbP protein
from the protease and the cleaved His tag (Fig. 1). A test of the
stability of the recombinant His-tagged PsbP and recombinant PsbP
proteins in different buffers (0.1 M MES pH 6.0, 20 mM bis-Tris pH
6.0, 0.1 M sodium cacodylate buffer pH 6.5 and 20 mM potassium
phosphate buffer pH 7.4) showed that they displayed highest stability
in 20 mM bis-Tris buffer pH 6.0 (data not shown). Two different
buffers (20 mM bis-Tris buffer, 0.1 M NaCl pH 6.00 and 20 mM
potassium phosphate buffer, 0.1 M NaCl pH 7.4) were tested and
compared for thrombin activity. No differences in the cleavage of the
His anchor by thrombin were observed (data not shown). Bis-Tris
buffer was found to be an optimal buffer for all the purification steps
that followed purification of the His-tagged PsbP protein and also for
Both recombinant proteins (His-tagged PsbP and PsbP) were used
for crystallization trials. However, only recombinant PsbP protein
was found to form three-dimensional crystals, which were observed
Acta Cryst. (2009). F65, 111–115 Kohoutova ´ et al.
SDS–PAGE analysis of the purification steps of the spinach recombinant PsbP
protein. Lane 1, molecular-weight markers (kDa); lane 2, supernatant after IPTG
induction of B152 cells; lane 3, the His-tagged PsbP protein purified by metal-
affinity chromatography; lane 4, His-tagged PsbP protein purified by ion-exchange
chromatography; lane 5, recombinant PsbP protein after Superdex 75 column
PsbP crystals from three different conditions: (a) 13% PEG 8000, 0.1 M Tris–HCl
pH 7.5, 10 mM ZnCl2, (b) 18% PEG 6000, 20 mM zinc acetate, 0.1 M Tris–HCl pH
7.0 and (c) 16% PEG 550 MME, 0.1 M Tris–HCl pH 7.5, 10 mM ZnSO4.
under three different conditions (Fig. 2). Optimization of the crys-
tallization conditions showed that the following condition yielded the
best quality crystals: 15 mg ml?1PsbP protein in 20 mM bis-Tris
buffer pH 6.0 mixed in a 1:1 ratio with reservoir solution containing
20% PEG 550 MME, 0.1 M Tris–HCl pH 7.5 and 5 mM ZnSO4and
equilibrated at 288 K for 3 d (Fig. 3). SDS–PAGE analysis of the
dissolved crystals showed that crystals were only formed from mature
protein; no partial degradation of PsbP was observed (Fig. 4).
Formation of PsbP crystals under the described conditions was found
to be advantageous in comparison with previously published crys-
tallization conditions (Ifuku et al., 2003), which yielded PsbP crystals
that contained partially degraded protein. Apart from the concrete
crystallization conditions, the shorter time needed for crystal growth
(3 d for PsbP protein from S. oleracea compared with two weeks for
PsbP protein from N. tabacum) might play a role in preventing
degradation. Diffraction data were collected to 2.06 A˚resolution
from a protein crystal directly frozen in mother liquor at 110 K
(Fig. 3). Crystal parameters and data-collection statistics are
summarized in Table 1. Our attempts to solve the phase problem by
molecular replacement were successful when the structure of the
PsbP protein from N. tabacum (PDB code 1v2b; Ifuku et al., 2004)
was used as a search model. The R factor and Rfreefactor were 0.453
and 0.463, respectively, and fell to 0.446 and 0.458, respectively,
during the initial three cycles of rigid-body refinement for data in the
resolution range 44.5–3.8 A˚. The orthorhombic crystal form (space
group P212121) contained one molecule in the asymmetric unit, with a
solvent content of 36.2% (VM= 1.92 A˚3Da?1).
The expression, purification and crystallization of stable recombinant
forms of the extrinsic PsbP protein of PSII are essential steps in the
elucidation of its three-dimensional structure and its interaction with
other PSII subunits. Knowledge of the crystallographic structures of
PsbP from tobacco (Ifuku et al., 2004) and spinach will allow us to
relate their structures to the data available from biophysical experi-
ments (Raman spectroscopy, FTIR spectroscopy and molecular
dynamics). Likewise, the structural information obtained on this PSII
extrinsic protein will lead to the design of further experiments
(vibrational spectroscopy, NMR, protein–protein cross-linking,
surface plasmon resonance and atomic force microscopy) in order to
Kohoutova ´ et al.
Acta Cryst. (2009). F65, 111–115
Difraction image of the spinach recombinant PsbP protein crystallized using 20% PEG 550 MME, 0.1 M Tris–HCl pH 7.5 and 5 mM ZnSO4.
Data-collection statistics for the PsbP crystal.
Values in parentheses correspond to the highest resolution shell.
Unit-cell parameters (A˚)
Resolution limits (A˚)
No. of unique reflections
Wilson B (A˚2)
a = 38.72, b = 46.79, c = 89.01
of reflection hkl.
ijIiðhklÞ ? hIðhklÞij=P
iIiðhklÞ, where Ii(hkl) is the ith observa-
tion of reflection hkl and hI(hkl)i is the weighted average intensity for all observations i
better understand the role of extrinsic proteins in PSII of higher
Support from the Institutional Research Concept of the Academy
of Science of the Czech Republic (AVOZ60870520, AVOZ50520514
and AVOZ0550506) and from the Ministry of Education of the Czech
Republic (LC 06010 and MSM6007665808) and the Grant Agency of
the Czech Republic (203/08/0114) are gratefully acknowledged. This
work was also funded by grants BFU2007-68107-C02-02/BMC (to
JBA) and AGL2005-07245-C03-03 (to JLR) from the Ministerio de
Education y Ciencia, Spain. We would also like to thank Ms Esther
Ferna ´ndez for technical assistance.
Balsera, M., Arellano, J. B., Revuelta, J. L., De Las Rivas, J. & Hermoso, J. A.
(2005). J. Mol. Biol. 350, 1051–1060.
Barber, J. (2003). Q. Rev. Biophys. 36, 71–89.
Collaborative Computational Project, Number 4 (1994). Acta Cryst. D50,
De Las Rivas, J., Balsera, M. & Barber, J. (2004). Trends Plant Sci. 9, 18–25.
Evans, P. R. (1997). Jnt CCP4/ESF–EACBM Newsl. Protein Crystallogr. 33,
Ifuku, K., Nakatsu, T., Kato, H. & Sato, F. (2003). Acta Cryst. D59, 1462–1463.
Ifuku, K., Nakatsu, T., Kato, H. & Sato, F. (2004). EMBO Rep. 5, 362–367.
Ifuku, K., Nakatsu, T., Shimamoto, R., Yamamoto, Y., Ishihara, S., Kato, H. &
Sato, F. (2005). Photosyn. Res. 84, 251–255.
Ifuku, K., Yamamoto, Y., Ono, T., Ishihara, S. & Sato, F. (2005). Plant Physiol.
Jansen, T., Rother, C., Steppuhn, J., Reinke, H., Beyreuther, K., Jansson, C.,
Andersson, B. & Herrmann, R. G. (1987). FEBS Lett. 216, 234–240.
Kamiya, N. & Shen, J. R. (2003). Proc. Natl Acad. Sci. USA, 100, 98–102.
Laemmli, U. K. (1970). Nature (London), 227, 680–685.
Leslie, A. G. W. (1999). Acta Cryst. D55, 1696–1702.
Loll, B., Cern, J., Saenger, W., Zouni, A. & Biesiadka, J. (2005). Nature
(London), 438, 1040–1044.
Seidler, A. (1996). Biochim. Biophys. Acta, 1277, 35–60.
Shen, J. R. & Inoue, Y. (1993). Biochemistry, 32, 1825–1832.
Suorsa, M., Sirpio, S., Allahverdiyeva, Y., Paakkarinen, V., Mamedov, F.,
Styring, S. & Aro, E. (2006). J. Biol. Chem. 281, 145–150.
Yi, X. P., Hargett, S. R., Liu, H., Frankel, L. K. & Bricker, T. M. (2007). J. Biol.
Chem. 282, 24833–24841.
Zouni, A., Witt, H. T., Kern, J., Fromme, P., Krauss, N., Saenger, W. & Orth, P.
(2001). Nature (London), 409, 739–743.
Acta Cryst. (2009). F65, 111–115Kohoutova ´ et al.
SDS–PAGE analysis of dissolved PsbP crystals obtained from four different
conditions: lane 1, 20% PEG 550 MME, 0.1 M Tris–HCl pH 7.0, 20 mM ZnCl2; lane
2, 15% PEG 550 MME, 0.1 M Tris–HCl pH 7.0, 20 mM ZnSO4; lane 3, 15% PEG
550 MME, 0.1 M Tris–HCl pH 7.5, 20 mM ZnSO4; lane 4, 15% PEG 8000, 0.1 M
Tris–HCl pH 7.5, 20 mM ZnCl2; lane 5, molecular-weight markers (kDa).