Download Homo-Dimeric Spherulin 3a: A Single

Biol. Chem., Vol. 380, pp. 89 – 94, January 1999 · Copyright © by Walter de Gruyter · Berlin · New York
Short Communication
Homo-Dimeric Spherulin 3a: A Single-Domain Member
of the ␤␥-Crystallin Superfamily
Michael Kretschmar, Eva-Maria Mayr
and Rainer Jaenicke*
Institut für Biophysik und Physikalische Biochemie,
Universität Regensburg, D-93040 Regensburg, Germany
* Corresponding author
The ␤␥-crystallin superfamily of eye lens proteins
comprises a class of structurally related members
with a wide variety of different functions. Common
features of these proteins are 1. the Greek-key motif
of antiparallel ␤-sheets, called the crystallin fold,
and 2. the high intrinsic long-term stability. Spherulin
3a (S3a), a dormant protein from the spherules of
Physarum polycephalum, is the only known single-domain protein within the ␤␥-crystallin family. Based on
sequence homology and ‘domain swapping’, it has
been proposed to represent an evolutionary ancestor
of present-day eye lens crystallins. Since S3a is highly
expressed in spherulating plasmodia of P. polycephalum under a variety of stress conditions, it can
be assumed that the protein may serve as a compatible solute in the cytosol of the slime mold. In order to
investigate the stability and other physicochemical
properties of a single-domain all-␤ protein, we isolated natural S3a. For the large-scale purification, the
recombinant protein was cloned and expressed in
Escherichia coli. The detailed spectral and biochemical analysis proved the recombinant protein to be
authentic. In its native form, S3a is dimeric. Due to its
exposed cysteine residues (Cys4), in the absence of
reducing agents intermolecular disulfide cross-linking leads to the formation of higher oligomers. In order
to preserve the native quaternary structure without
aggregation artifacts in denaturation/renaturation
experiments, the Cys4 씮Ser mutant (S3a C4S) was
produced. Both the wild-type protein and its mutant
are indistinguishable in their physicochemical properties. At pH 3 – 4, both proteins form a stable compact
intermediate (A-state). Concentration-dependent thermal and chemical denaturation showed that the equilibrium unfolding of S3a obeys the simple two-state
model with no significant occurrence of folding intermediates.
Key words: ␤␥-Crystallin superfamily / Greek-key motif /
Long-term stability / Physarum polycephalum /
Two-state model.
Proteins of the ␤␥-crystallin superfamily are closely related in their sequence as well as in their gene and protein
structure, all consisting of domains with two Greek-key
motifs of antiparallel ␤-sheets. They show high long-term
stability, in the case of the eye lens crystallins without
turnover or degradation over the whole life time of the organism (Jaenicke, 1994). Based on sequence homology,
the encystment-specific spherulin 3a (S3a) from the slime
mold Physarum polycephalum was predicted to be structurally related to the N-terminal domain of ␥B-crystallin
(Bernier et al., 1987; Wistow, 1990). Although the sequence identity of the two proteins is no more than 20%,
all amino acid residues essential for the ␤␥-crystallin fold
are found to be conserved. The predicted structural similarity was confirmed by 3D NMR (Rosinke et al., 1997).
Being the only known member of the ␤␥-crystallin family with only one domain, S3a is a particularly suitable molecule for biophysical studies on the anomalous stability of
the all-␤ crystallin fold observed for a number of proteins.
Since thermodynamic measurements require significant
amounts of protein, S3a was cloned and expressed in
Escherichia coli, and the recombinant protein purified to
homogeneity. In order to exclude artifacts caused by intermolecular cross-linking, Cys4, the only cysteine residue
present in S3a, was substituted by serine. Here we present
the isolation and characterization of the natural, recombinant and mutant proteins.
S3a was isolated from P. polycephalum spherules by
sequential anion- and cation-exchange and subsequent
hydrophobic chromatography. Spherules were obtained
by starvation of plasmodia (Jump, 1954) and harvested by
centrifugation. The cells were disrupted using a Braun homogenizator with glass beads (ø = 0.5 mm) in 50 mM Naphosphate pH 7.0, 5% Triton X-100 (w/v), 50 mM EDTA,
300 mM NaCl. The soluble fraction was dialyzed against
50 mM Mops/NaOH pH 7.5, 1 mM EDTA (buffer A) and applied onto a DE52 column (Whatman) preequilibrated with
buffer A. The flow-through containing the S3a was dialyzed against 10 mM Na-phosphate pH 6.0, 1 mM EDTA
(buffer B). After loading onto a SP-Sepharose Fast-Flow
column preequilibrated with buffer B, S3a was in the
flow-through which was dialyzed against 50 mM Tris/HCl
pH 8.5, 1 mM EDTA (buffer C). The solution was applied to
a Resource Q column (Pharmacia). Ammonium sulfate
was added to the S3a containing flow-through to a final
concentration of 2 M. The solution was applied to a Resource ISO column. Again, S3a was found in the flowthrough. The efficiency of the single purification steps and
the homogeneity of the final product is illustrated in Fig-
Unauthenticated
Download Date | 6/15/17 5:22 AM
90
M. Kretschmar et al.
Fig. 1 SDS-PAGE Illustrating the Purification of Spherulin S3a.
Lane 1: molecular mass standard, lane 2: crude extract of
spherules, lane 3: flow-through DE-52, lane 4: flow-through SPSepharose FF, lane 5: flow-through Resource Q, lane 6: flowthrough Resource ISO, lane 7: purified recombinant S3a.
ure 1. The yield was 1 mg of natural S3a from 10 g of
wet-weight spherules.
Since this yield is very low, we produced S3a recombinantly in E. coli. All molecular cloning procedures were
performed according to Sambrook et al. (1989). The plasmid for cytoplasmic expression of S3a in E. coli was constructed by cloning the gene of S3a in the expression plasmid pET11a (Studier et al., 1990) via the NdeI and BamHI
restriction sites. The gene was PCR-amplified from the
plasmid pBR322/LAV2-3 (Bernier et al., 1987) containing
the S3a cDNA using synthetic oligonucleotide primers
(primer EM11: 5⬘-TAT ACA AAA ACA CAT ATG TCG GTC
TGC AAA GGA GTT TCT G-3⬘; primer EM12: 5⬘-CAT TAT
TAT CAC TAG GAT CCT TAT CAT GCG CTC TCA AAC-3⬘).
For the construction and expression of the Cys4 씮Ser
mutant of S3a, the plasmid pETSph3a was used. Site-directed mutagenesis was performed according to Kunkel
et al. (1987) using the helper phage M13K07 (Geisselsoder
et al., 1987) and the kit supplied by Bio-Rad (Hercules, CA,
USA). The following oligonucleotide MK1 was used: 5⬘TCC AGA TCC TTT GGA GAC CGA CAT ATG TAT-3⬘. The
mutation-carrying plasmids were identified by restriction
analysis and verified by sequencing the whole gene using
an ABI 373 A Sequencer (Perkin Elmer). For recombinant
expression, E. coli BL21(DE3) cells (Studier and Moffatt,
1986) harboring the expression plasmids pETSph3a or
pETSph3a C4S, were grown in 2 l of LB medium containing ampicillin (100 ␮g/ml) at 30 °C. At OD550nm = 1.0, isopropyl-␤-D-thiogalactopyranoside (IPTG) was added to a
final concentration of 1 mM. The cells were grown for further 20 h, and subsequently harvested by centrifugation
and resuspended in buffer B. Cell disruption was performed by passing the cells twice through a French press
cell at 18 000 psi. The soluble fraction of the lysate was dialyzed against buffer B to remove traces of salt, and applied to a SP-Sepharose Fast-Flow column (Pharmacia)
preequilibrated with buffer B. Protein was eluted using a
linear gradient from 0 to 200 mM sodium chloride in buffer
B. Fractions containing S3a were pooled and ammonium
sulfate was added to a final concentration of 2 M. The solution was then applied to a Phenyl-Sepharose FF column
(Pharmacia) preequilibrated with 10 mM Na-Phosphate
pH 7.0, 1 mM EDTA, 2 M ammonium sulfate (buffer D). After
washing the column with buffer C, spherulin was eluted
with a linear gradient from 2 to 0 M ammonium sulfate in
buffer B. Fractions containing the purified protein were
collected and dialyzed against 25 mM Mops/NaOH pH 7.0,
1 mM EDTA. Typically 5 mg of purified S3a or S3a C4S per
liter of cell culture were obtained by this purification procedure. As judged by silver-stained SDS-polyacrylamide
gels, recombinant S3a and S3a C4S were purified to
⬎ 99.5% homogeneity. All purification steps were performed at room temperature. N-terminal sequencing of
recombinant S3a demonstrated that the N-terminal methionine residue was removed. Natural S3a is chemically
modified at the N-terminus, so that protein sequencing
was not possible.
The authenticity of the recombinant protein was confirmed by a detailed spectroscopic analysis (Figure 2). Absorbance, fluorescence and CD spectra of recombinant
and natural S3a are identical within experimental error.
S3a contains one tryptophan, two tyrosine and eight
phenylalanine residues. The absorption spectrum of the
native protein has only a maximum at 280 nm, with the
second maximum of tryptophan (Schmid, 1997) reduced
to a shoulder at 290 nm by the influence of the tyrosines. In
the wavelength range between 260 and 270 nm the contributions of the eight phenylalanine residues are visible. Denaturation in 6 M GdmCl leads to a slight blue shift and a
decrease of the maximum (Figure 2A). The specific absorption coefficient of the native protein, A280nm,0.1%,1cm =
0.82, was determined according to Gill and von Hippel
(1989). Native S3a shows a fluorescence spectrum with an
emission maximum at 323 nm, indicating that the tryptophan residue is buried in the hydrophobic core of the protein (Schmid, 1997). The low quantum yield suggests that
the tryptophan residue is quenched by a neighbouring
amino acid. After denaturation in 6 M guanidinium chloride
(GdmCl), the solvent exposed tryptophan of S3a has an
emission maximum at 352 nm which is 1.4-fold higher
than the emission maximum of the native protein (Figure
2B). The large difference in the spectra of native and denatured S3a was used to study equilibrium unfolding transitions. In contrast to many other proteins, native spherulin
shows a near-UV CD spectrum with no negative ellipticity.
The shape is very similar to the absorption spectrum.
Again, a maximum at 280 nm and a shoulder at 290 nm are
visible. Furthermore, the spectrum has a small second
maximum at 259 nm. As expected, the spectrum of the
denatured protein shows no ellipticity (Figure 2C). The farUV CD spectrum, which normally is determined by contributions of the amide bonds, gives information about the
secondary structure. As observed for other all-␤ proteins,
S3a shows a maximum at 201 nm; the typical negative ellipticity between 250 and 210 nm is not detectable. Instead, S3a is characterized by a minimum with negative
ellipticity at 188 nm (Figure 2D). Similar characteristics in
Unauthenticated
Download Date | 6/15/17 5:22 AM
Spherulin 3a from P. polycephalum
the far-UV CD were also observed for the cellulose-binding domain of cellobiohydrolase I from Trichoderma reesei
(Mattinen et al., 1997) and the cold-shock protein homolog from Thermotoga maritima (Welker et al., 1998).
91
Probably the influence of the numerous aromatic residues
is responsible for the anomalous dichroic absorption in the
far-UV range. The spectrum of the denatured protein
shows no ellipticity between 250 and 205 nm.
Fig. 2 Spectroscopic Characterization of Native Natural S3a (nat wt), Native Recombinant S3a (rec wt) and Denatured Recombinant S3a.
(A) Absorption spectra at 1 mg/ml protein. (B) Fluorescence spectra at 40 ␮g/ml concentration. (C) Near-UV CD spectra and (D)
far-UV CD spectra at 1 mg/ml protein concentration of natural wt (– – –) and recombinant wt (—) in 25 mM Mops/NaOH pH 7.0, 1 mM EDTA
and recombinant wt in 6 M GdmCl, 25 mM Mops/NaOH pH 7.0, 1 mM EDTA (• • • •). UV-absorption spectra were measured in a Cary 1/3
Varian spectrophotometer, fluorescence emission spectra in a Jobin Yvon Spex FluoroMax-2 spectrometer, and CD spectra in an AVIV
62D spectropolarimeter.
Table 1 Determination of the Molecular Mass of Spherulin S3a and Its S3a C4S Mutant in
25 mM Mops/NaOH pH 7.0, 1 mM EDTA.
Protein
S3a
S3a C4S
a
b
Additive
Analytical gelfiltration
Analytical ultracentrifugation
Vel (ml)
Mb (kDa)
na
s20,w (S)
M (kDa)
n
83.2
63.1
36.3
21.9
–
8
6
4
2
–
3.49
39.4
4
5 mM DTT
8.6
9.3
10.8
12.2
–
2.19
21.2
2
150 nM NaCl
12.2
21.9
2
2.24
21.6
2
150 nM NaCl
n, state of association: 2, 4, 6, 8 stand for dimer, tetramer, hexamer and octamer, respectively. Ranges of error: ⌬MGF ⬇ 8%, ⌬s20,w ⬇ 3%, ⌬MAUC ⬍ 5%
molecular mass
Unauthenticated
Download Date | 6/15/17 5:22 AM
92
M. Kretschmar et al.
The molecular mass of S3a was determined by mass
spectrometry, analytical gel filtration and analytical ultracentrifugation. Mass spectrometry measurements resulted in a molecular mass of S3a monomer of 11 152 Da,
corresponding to monomeric S3a (Mr calc = 11 151). Gel filtration and ultracentrifugation experiments were performed in order to determine the state of association
(Table 1). Surprisingly, the elution profile of S3a applied onto a Superdex 75 HR gel filtration column exhibits several
Fig. 3 Gel Filtration Chromatography of S3a and S3a C4S.
Analytical gel filtration of S3a wildtype (– – –) and S3a C4S (—) at
1 mg/ml were performed at room temperature using a Superdex
75 HR size exclusion column (Pharmacia) in 25 mM Mops/NaOH
pH 7.0, 1 mM EDTA, 150 mM NaCl and monitored by absorbance
at 280 nm. Molecular mass standards were: 1. bovine serum albumin (68 kDa), 2. chymotrypsinogen (25 kDa), 3. RNase A (14 kDa),
and 4. bovine pancreatic trypsin inhibitor (6.5 kDa).
peaks (Figure 3). Each peak fraction contained S3a as
identified by SDS-PAGE. Compared to the calibration
curve, the peaks correspond to dimers, tetramers, hexamers and octamers. This apparent heterogeneity is indicative for a slow dissociation/association equilibrium. Figure
3 illustrates the situation at relatively high protein concentration at which the equilibrium favours the tetrameric
quaternary structure. Ultracentrifugation yields dimers in
the presence of DTT and tetramers and higher states of
oligomerization in the absence of DTT. As S3a contains
one cysteine residue per subunit, higher states of association may be easily explained by the formation of interchain disulfide bridges. The sedimentation analysis in the
presence of DTT as a reducing agent clearly proves this interpretation to be valid. To ensure a well-defined state of
association of the native protein and to avoid side reactions of the cysteine residues (e.g., aggregation of the
thermally denaturated protein), the cysteine in position 4
was replaced by serine by site-directed mutagenesis. The
mutant protein S3a C4S shows identical physicochemical
properties compared to the natural and the recombinant
wild-type protein (data not shown). Thus, for all further
measurements S3a C4S was used. The gel-filtration experiments with S3a C4S show only one peak in the elution
profile, indicating a molecular mass of 21.9 kDa, in accordance with the dimeric native state (Figure 3). Analytical
ultracentrifugation confirmed this result, yielding a molecular mass of 21.6 ⫾ 1.6 kDa (Table 1).
The pH dependence of the secondary and tertiary
structure of S3a C4S was investigated by far-UV and nearUV CD spectroscopy (Figure 4). The unaltered far-UV CD
spectra in Figure 4A showed that at pH 4 – 7 the secondary
structure of the protein remains unaffected. At pH 3, only a
slight decrease of the maximum at 201 nm was observed;
Fig. 4 pH-Dependent Near-UV and Far-UV CD Spectra of S3a C4S.
(A) Near-UV CD spectra and (B) far-UV CD spectra of S3a C4S at 1 mg/ml protein concentration and varying pH: pH 7, 25 mM Mops/NaOH,
1 mM EDTA; pH 4, 25 mM Na-formiate, 1 mM EDTA; pH 3, 25 mM glycine/HCl, 1 mM EDTA; pH 2, 25 mM glycine/HCl,
1 mM EDTA.
Unauthenticated
Download Date | 6/15/17 5:22 AM
Spherulin 3a from P. polycephalum
93
Fig. 5 Equilibrium Denaturation Transitions of S3a C4S.
(A) GdmCl-induced equilibrium transitions of S3a C4S at 0.3 mg/ml protein concentration, monitored by the change in fluorescence intensity at 360 nm (쎲), dichroic absorption at 213 nm (❍), sedimentation coefficient (왖) and molecular mass (왕) from sedimentation velocity and sedimentation equilibrium, respectively. (B) Equilibrium transitions at 10 µg/ml (쎲), 100 µg/ml (왖) and 300 µg/ml (쐽) protein concentration, monitored by the relative change in fluorescence intensity. Denaturant-induced equilibrium transitions were determined by
incubating the protein at various GdmCl concentrations until equilibrium was reached at 25 °C (Pace, 1986). The denaturation buffer
contained 25 mM Mops/NaOH pH 7.0, 1 mM EDTA. For refolding experiments, the protein was denatured in 6 M GdmCl, 25 mM Mops/NaOH
pH 7.0, 1 mM EDTA for 16 hours prior to dilution into buffers containing different concentrations of GdmCl. Sedimentation velocity and sedimentation equilibrium runs were performed in a Beckman Spinco model E ultracentrifuge equipped with a high-sensitivity UV scanning
system at 20 °C. Double-sector cells with sapphire windows were used in an AnG rotor. The scanning wavelength was 280 nm. (C) Temperature induced denaturation at protein concentrations of 0.1 mg/ml (쎲) and 1 mg/ml (❍), monitored by dichroic absorption at 213 nm in
25 mM Mops/NaOH pH 7.0, 1 mM EDTA, 0.8 M GdmCl, heating rate 0.4 K/min.
at pH 2 the spectrum differed significantly from the spectrum of the native protein. In the near-UV CD spectrum
(Figure 4B), the decrease in signal intensities at pH 4 and
pH 3 points to a loss of tertiary structure, whereas the farUV CD spectra still showed native-like characteristics.
Obviously, at low, non-denaturing pH, S3a is in an A-state
like conformation. At pH 2, native-like and denatured molecules seem to be in equilibrium.
The conformational stability of S3a was determined by
GdmCl- and temperature-dependent equilibrium transitions (Figure 5). GdmCl-induced changes of the quaternary, tertiary and secondary structure were measured at
different protein concentrations, making use of analytical
ultracentrifugation, fluorescence intensity and ellipticity
(Figure 5A). With all three methods, fully reversible and
highly cooperative monophasic transitions with midpoints
at 1.2 ⫾ 0.1 M GdmCl were obtained. Equilibrium transitions with protein concentrations between 10 to 300 ␮g/ml
(monitored by the change of the fluorescence intensity at
360 nm) show a slight concentration dependence, with a
shift of the transition midpoint of ⬃ 0.3 M GdmCl (Figure
5B). The temperature-induced transitions were measured
in the presence of 0.8 M GdmCl to decrease the transition
temperature and to achieve full reversibility (Figure 5B). In
the absence of GdmCl, S3a ‘melts’ at ⬃ 65 °C and aggregates simultaneously (data not shown). Again, between
0.1 and 1 mg/ml a monophasic, highly cooperative transition with a concentration dependent melting point was detected. For S3a C4S, upon increasing the protein concentration from 0.1 to 1.0 mg/ml, the midpoint of transition is
shifted from 50 to 52 °C; evidently, unfolding is coupled to
the dissociation of the dimer. This finding, together with
the observation that different methods (detecting changes
in the quaternary, tertiary and secondary structure) show
coinciding monophasic transitions at the same protein
concentration, allows the conclusion that the folding of
S3a follows the simple two-state model. The complete
thermal reversibility in the presence of GdmCl, even at
high protein concentrations, makes S3a a good candidate
for a more detailed thermodynamic and kinetic analysis
which is in process.
Acknowledgements
We thank Dr. D. Pallotta for the S3a cDNA and Dr. E. Holler and
Hermine Reisner for the spherules of P. polycephalum. Fruitful
discussions with Dr. S. Miller are gratefully acknowledged. Work
was supported by the Deutsche Forschungsgemeinschaft
(Ja78/33-3) and the Fonds der Chemischen Industrie.
References
Bernier, F., Lemieux, G., and Pallotta, D. (1987). Gene families
encode the major encystment-specific proteins of P. polycephalum plasmodia. Gene 59, 265 – 277.
Geisselsoder, J., Whitney, F., and Yuckenberg, P. (1987). Efficient
site-directed in vitro mutagenesis. Biotechniques 5, 786-791.
Gill, S.C., and von Hippel, P.H. (1989). Calculation of protein extinction coefficients from amino acid sequence data. Anal.
Biochem. 182, 319 – 326.
Jaenicke, R. (1994). Eye-lens proteins: Structure, superstructure,
stability, genetics. Naturwissenschaften 81, 423 – 429.
Jump, J.A. (1954). Studies on sclerotization in P. polycephalum.
Amer. J. Bot. 41, 561 – 567.
Unauthenticated
Download Date | 6/15/17 5:22 AM
94
M. Kretschmar et al.
Kunkel, T.A., Roberts, J.D., and Zakour, R.A. (1987). Rapid and
efficient site-specific mutagenesis without phenotypic selection. Meth. Enzymol. 154, 367 – 382.
Mattinen, M.-L., Kontteli, M., Kerovuo, J., Linder, M., Annila, A.,
Lindenberg, G., Reinikainen, T., and Drakenberg, T. (1997).
Three-dimensional structures of three engineered cellulosebinding domains of cellobiohydrolase I from Trichoderma reesei. Prot. Sci. 6, 294 – 303.
Pace, C.N. (1986). Determination and analysis of urea and GdmCl denaturation curves. Meth. Enzymol. 131, 266 – 280.
Rosinke, B., Renner, Ch., Mayr, E.-M., Jaenicke, R., and Holak,
T.A. (1997). Ca2+-loaded Spherulin 3a from P. polycephalum
adopts the prototype ␥-crystallin fold in aequeous solution. J.
Mol. Biol. 271, 645 – 655.
Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular
Cloning: A Laboratory Manual, 2nd ed. (Cold Spring Harbor,
N.Y., USA: Cold Spring Harbor Laboratory Press).
Schmid, F.X. (1997). Spectral methods of characterizing protein
conformation and conformational changes. In: T. E. Creighton
(ed.), Protein Structure: A Practical Approach. (Oxford, New
York, Tokio: IRL Press), pp. 261 – 298.
Studier, F.W., and Moffatt, B.A. (1986). Use of bacteriophage T7
RNA polymerase to direct selective high-level expression of
cloned genes. J. Mol. Biol. 189, 113 – 130.
Studier, F.W., Rosenberg, A. H., Dunn, J.J., and Dubendorff, J.W.
(1990). Use of T7 RNA polymerase to direct expression of
cloned genes. Meth. Enzymol. 185, 60 – 89.
Welker, C., Böhm, G., Schurig, H., and Jaenicke, R. (1998). Cloning, overexpression, purification and physicochemical characterization of a cold shock protein homolog from Thermotoga
maritima. Prot. Sci., in press.
Wistow, G. (1990). Evolution of a protein superfamily: Relationship between vertebrate lens crystallins and microorganism
dormancy proteins. J. Mol. Evol. 30, 140 – 145.
Received September 18,1998; accepted October 23,1998
Unauthenticated
Download Date | 6/15/17 5:22 AM