pubs.acs.org/crystal
Article
Electrostatic Potentials around the Proteins Preferably Crystallized
by Ammonium Sulfate
Yan Guo, Liang Qu, Noritaka Nishida, and Tyuji Hoshino*
Cite This: https://dx.doi.org/10.1021/acs.cgd.0c01136
Downloaded via CHIBA UNIV on December 12, 2020 at 02:35:22 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
ACCESS
Metrics & More
Read Online
Article Recommendations
sı Supporting Information
*
ABSTRACT: Ammonium sulfate (AS) is one of the most popular
precipitants in protein crystallization. However, AS is not always effective
for all kinds of proteins. Some proteins are easily crystallized by utilizing AS,
while others are not. Polyethylene glycols (PEGs) is also frequently used in
protein crystallization. To clarify the reason a protein has a preference for a
precipitant in crystal growth, we investigated electrostatic potentials for 100
kinds of proteins that had been crystallized by high concentrations of AS.
Most of the proteins have a common shape for the isosurface of their
electrostatic potentials. The positive and negative areas of electrostatic
potential are almost equally separated. The contact between the positive and
negative areas is limited to a narrow region on the protein surface. The
separation of the electrostatic potential is neat even at the contact region. In
contrast, the separation between positive and negative areas is not clear for
the proteins preferably crystallized by PEGs. The positive and negative areas of electrostatic potential are fragmentary, and the
isosurface at the contact region is complicated. These findings suggest that not only the local interaction between AS ions and a
protein molecule but also the surroundings of the protein are responsible for crystal growth.
■
INTRODUCTION
X-ray crystallography is the primary experimental method for
determining a protein structure. However, the crystal growth of
protein molecules is a critical step for this method. Various
parameters influence the molecular interaction of proteins and
are related to protein crystal growth.1 Hence, the optimization
of crystallization conditions is a time-consuming process. In
particular, the selection of the precipitant agent is a key
parameter for obtaining good-quality protein crystals.2
Ammonium sulfate (AS) and polyethylene glycol (PEG) are
the most popular precipitants for inducing protein crystallization.3 Some proteins are easily crystallized by AS, while
others are not. Instead, some proteins are preferably crystallized by PEGs. The reason for the different preferences of
precipitants among proteins is still unclear, despite the fact that
many studies on the functions of precipitants have been carried
out.4−6
The function of AS in protein crystallization has been
extensively studied.7,8 The protein solubility is affected by the
ionic strength of the solution and the net charge of the
protein.9 The surface tension at the water−protein interface
and the activity of water in the hydration layer are related to
the solubility. According to a well-accepted mechanism, the
Hofmeister series of ions for protein hydration, strongly
hydrated anions such as sulfate and phosphate compete for
water molecules in the second hydration layer, leading to a
decrease in water activity in the first layer for solvating the
protein. The strongly hydrated anions further increase the
© XXXX American Chemical Society
surface tension of the solvent, causing the protein to minimize
its solvent-accessible surface area. The decrease in water
activity and the increase in surface tension eventually reduce
the protein solubility by the salting-out effect. Another effect of
salts on proteins is the protein−ion interaction. The electric
charge of the protein is an essential factor related to solubility.
Strongly hydrated ions induce a steady protein−ion
interaction, which has an influence on the apparent charge of
proteins.10,11 AS is one of the strong salting-out agents and is
widely used in crystallization due to its high degree of
solubility, low heat of solution, and weak denaturation action
on the protein tertiary structure. Consequently, protein crystal
growth by AS is attributed to the effects of hydration and
dehydration involving competition for water molecules.
Although the salting-out mechanism is broadly accepted,
understanding the reason for the preference of precipitants is
still not clear.
In our previous study,12 one kind of protein was crystallized
with three different precipitants, including AS and PEGs. A
computational analysis showed that AS ions were anisotropiReceived: August 13, 2020
Revised: December 1, 2020
A
https://dx.doi.org/10.1021/acs.cgd.0c01136
Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
pubs.acs.org/crystal
Article
precipitant agents. The difference in the shapes of electrostatic
potential was quantitatively characterized by an index
representing polarity. The aim of this study is to clarify the
reason for the preference for precipitants in protein crystal
growth.
cally distributed around the protein, which was reflected in the
molecular packing of the crystals. Since the molecular packing
of a protein crystal is linked to the space group, the
relationships between space groups and precipitants were
investigated for several kinds of proteins.13,14 We also
performed X-ray structure analysis by growing single crystals
for four types of proteins.15 Two of them are known to be
easily crystallized by AS, while the other two are not easily
crystallized by AS. The two proteins not easily crystallized by
AS are human serum albumin (HSA) and hen egg white
lysozyme (HEWL). The crystal growth of these two proteins
by AS was difficult, and optimization of the cryoprotectant also
took a tremendous amount of work. X-ray diffractions from
most of the HSA crystals were inadequate for structure
analysis. Hence, no crystal structure had been deposited in the
PDB for HSA crystallized with AS until our study. Molecular
dynamics simulations of the four proteins with AS ions
indicated that the distribution of AS ions for the proteins easily
crystallized by AS was not random but highly anisotropic
around the protein, with the ions localized in two areas. The
localized distribution was caused by the electrostatic potential
around the protein, and AS ions were accumulated in the
regions where the absolute value of the electrostatic potential
was high. These findings suggested that the shape of the
isosurface of electrostatic potential was responsible for the ease
of crystal growth with AS.
The electrostatic potential is highly correlated with the
shape and electric properties of proteins. The influence of an
electrostatic interaction on the protein−protein association
was studied in terms of stabilization, desolvation, and
kinetics.16 Two proteins favorably interact with each other at
a specific position, and then protein association is considerably
directional. A free energy calculation suggested that an
electrostatic contribution could not fully compensate for the
desolvation of the residues involved in the salt bridge or
hydrogen bond formation.17 The driving force for protein
folding and binding was primarily a hydrophobic interaction,
and an electrostatic interaction due to salt bridges and
hydrogen bonds was a complementary effect. Another
calculation suggested that the attractive electrostatic force
between a pair of charges, each of them being on different
protein molecules, prominently stabilized the protein crystal
with two proteins facing each other to compensate for their
opposite charges.18 A theoretical study suggested that the
major cost in free energy for protein association was the
entropic loss of counterions that made proteins electrically
neutral.19 An electric field has sometimes been applied for
protein crystallization.20,21 A review pointed out that the
crystallization process could be significantly enhanced by the
application of electric fields in terms of reduction of the
nucleation time, control of the location of nucleation, increase
in the yield, control of the crystal size, enhancement of the
crystal quality, control of the crystal orientation, and control of
the polymorphism.22 Accordingly, it is clear from these
previous findings that the electrostatic force or electrostatic
potential has a significant influence on protein crystallization.
In this work, we selected 100 crystal structures crystallized
by high concentrations of AS. Electrostatic potentials were
drawn for the surrounding spaces of the proteins. The shapes
of electrostatic potentials were contrasted to those of 20
proteins that had been crystallized by PEGs. In addition, the
electrostatic potentials were examined for the proteins that had
been crystallized by moderate concentrations of AS or other
■
METHODS
Calculation of Electrostatic Potential. The crystal structures
analyzed in this work were downloaded from the protein data bank
(PDB). The downloaded PDB files were filtered under the following
criteria. (1) For 100 structures, proteins were crystallized by AS at
high concentrations over 2.0 M. (2) For 10 structures, proteins were
crystallized by AS at low concentrations of less than 2.0 M. (3) A
crystal was grown in the apo form without ligands or inhibitors. (4)
Duplication of the same kinds of protein should be avoided. (5)
Crystal structures should not contain a large number of missing
residues. Then, we selected the structures for analysis from the filtered
PDB files in terms of the presence of precise descriptions of
resolution, pH condition, heteroatoms, etc. The procedure for the
selection is illustrated in Figure S1. For comparison, 20 crystal
structures obtained by using PEGs were selected when the average
molecular weight was over 3000 and the concentration of PEGs in
solution was over 30%. The structures were filtered by criteria similar
to those of AS. Additionally, 10 structures of the proteins crystallized
by other kinds of salts or agents were included. For a detailed analysis,
20 structures of the proteins crystallized by PEGs with the inclusion of
low concentrations of AS were added. In total, 160 crystal structures
were surveyed in this work. The PDB codes, protein names,
crystallization conditions, and molecular properties are given in
Table S1.
The electrostatic potential around every protein was depicted for all
of the downloaded crystal structures as follows. In preparation, the
missing residues in the crystal structure were added by homology
modeling with Modeler 9.2.23 However, the missing residues at the Nand C-terminal sides were not generated because the terminal
residues were flexible and reliable structure prediction was usually
impossible. When the biological unit was not a monomer, the
calculation model was built in an oligomeric form as shown in Table
S1. The pKa calculation was performed by Propka 3.1,24 in which the
pKa values of the titratable residues were predicted by the electrostatic
continuum theory. The protonation state of each titratable residue
was selected so that the protein net charge became zero at a certain
pH: i.e., the isoelectric point. For example, when the calculated pKa of
an Asp residue is higher than the pH, the name of the residue is set to
ASH instead of ASP. Similarly, the residue name of a Lys is set to
LYN when the calculated pKa is lower than the pH. The hydrogen
atoms were generated by the leap module of AmberTools16.25 The
electrostatic potential was obtained by solving the Poisson−
Boltzmann equation by Delphi 5.1,26 and the potential map was
visualized using Chimera 1.1227 or PyMOL 1.7.28
Quantitative Estimation of Electrostatic Potential. The
amplitude of the polarity due to electrostatic potential was estimated
by two electric point charges and their distance. The isosurface of the
electrostatic potential was drawn at the contour values of +0.5 and
−0.5 in units of kT/e. Provided that the isosurface has a shape
consisting of two spheres, one positive point charge and one negative
point charge were placed at the centers of the spheres. The distance
between the two point charges was calculated as an index representing
the electric polarity. Since the distance depended on the size of the
protein, each calculated value was scaled by dividing by the cube root
of the mass weight of the respective protein.
Crystallization of an Example Protein. A single crystal of
chitosanase was grown by utilizing AS. An Escherichia coli strain,
Rosetta (DE3) pLysS, was transformed with the pET50b vector
(Novagen) containing the gene coding N-terminal side-truncated
chitosanase. Chitosanase was expressed as a 6×His-fused Nus-tagconjugated form due to the use of pET50b vector. After preculture in
LB medium at 37 °C, the protein was expressed at 28 °C overnight by
induction with 0.2 mM isopropyl-β-thiogalactopyranoside (IPTG) at
B
https://dx.doi.org/10.1021/acs.cgd.0c01136
Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
pubs.acs.org/crystal
Article
Figure 1. Positive and negative isosurfaces of electrostatic potential around the 30 selected proteins. Out of 30 proteins, 20 proteins were
crystallized by high concentrations of AS and 10 were crystallized by PEGs. The 20 AS-crystallized proteins are illustrated in a blue frame, and the
10 PEGs-crystallized proteins are illustrated in a red frame. The positive and negative isosurfaces are shown in blue and red meshes representation,
respectively. The contour values of the isosurfaces are +0.5 and −0.5 in units of kT/e. Protein molecules are represented by green cartoons.
building were carried out using Refmac32 and Phenix.33 The
molecular structure and electron density map were visualized by
COOT.34
an OD600 value of 0.6. The conjugated protein was purified by Co
metal-affinity chromatography, followed by the cleavage of the 6×Hisfused Nus-tag by human rhinovirus 3C (HRV-3C) protease. The
cleaved Nus-tag, HRV-3C protease, and uncleaved conjugated protein
were removed by Ni-nitrilotriacetic acid (NTA) agarose resin. The
protein was further purified by gel filtration with a running buffer of
10 mM Tris-HCl at pH 8.0 and 150 mM NaCl. Finally, the protein
was concentrated to 5.4 mg/mL.
A single crystal of chitosanase was grown by the sitting drop vapor
diffusion method at 18 °C. To set up a droplet for crystallization, 1.0
μL of protein solution was mixed with 1.0 μL of a precipitant solution
consisting of 100 mM sodium citrate at pH 5.0 and 3.4 M AS. The
mixture was placed on the crystallization stage of a well plate, with the
reservoir filled with 300 μL of the precipitant solution. The crystals
were grown within 2 weeks without any seeding technique. The
crystals were cryoprotected by brief immersion in a solution
containing 25% (v/v) glycerol, followed by freezing in liquid nitrogen.
X-ray diffraction data were collected at 100 K on the BL-17A
beamline of the Photon Factory (PF, Tsukuba, Japan). The diffraction
data were indexed, scaled, and merged with XDS.29 Intensities were
converted into structure factors, and 5% of the reflections were
flagged for Rfree calculations. The protein structure was determined by
the molecular replacement with MolRep in the CCP4 software
package30 using a structure obtained by the PEGs-grown crystal (PDB
code: 1V5C)31 as a search model. Structure refinement and model
■
RESULTS AND DISCUSSION
The electrostatic potentials were depicted for all of the 160
proteins with determination of the protonation state of every
titratable residue at a pH value at which the net charge of the
protein was zero. Figure 1 shows the electrostatic potentials at
the surrounding spaces of 20 typical proteins that are
preferably crystallized by high concentrations of AS. All of
the 20 proteins display a common shape in electrostatic
potential. The positive and negative areas are almost equally
separated, and each area has a spherical lobe form. The protein
molecule is positioned at the boundary region of the two areas.
Hence, each spherical lobe arises from the protein and the
spherical lobes expand in opposite directions. It should be
noted that the positive and negative areas distinctly split even
at their boundary, and the contact region is limited in the
vicinity of the protein. Electrostatic potentials around the 10
proteins that are preferably crystallized by PEGs are shown for
comparison. The separation between the positive and negative
areas is unclear, and their boundary is ambiguous. The positive
C
https://dx.doi.org/10.1021/acs.cgd.0c01136
Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
pubs.acs.org/crystal
Article
Figure 2. Electrostatic potentials for the 140 proteins. The contour values of the positive and negative isosurfaces are +0.5 and −0.5 in units of kT/
e, respectively. The positive and negative isosurfaces are shown in blue and red mesh representationx, and the protein molecule is represented by a
green cartoon. The 110 AS-crystallized proteins are displayed in a blue frame, and the 20 PEGs-crystallized proteins are displayed in a red frame.
Six proteins in the green frame were crystallized by sodium formate, and two proteins in the yellow frame were crystallized by glycerol. The other
two proteins in grey were crystallized by ethylene glycol and sodium chloride.
by limitation to the apo form. As described for the 20 typical
proteins shown in Figure 1, the positive and negative areas are
distinctly separated, and each of the areas has a spherical lobe
form. In other words, the positive and negative areas of the
electrostatic potential clearly split, with both areas being round
in shape and having almost equal volumes. The next 10
proteins have been crystallized by AS at concentrations less
than 2.0 M. The maximum AS concentration for crystal growth
was 1.7 M among these 10 proteins. The positive and negative
and negative areas are interlaced with each other and deviate
from a spherical form.
The electrostatic potentials for all of the 140 proteins are
shown in Figure 2. The initial 100 proteins have been
crystallized by AS at concentrations above 2.0 M. About 1200
structures have so far been deposited in the PDB as crystals
grown by AS at high concentrations above 2.0 M. However,
crystals suitable for the present study were greatly reduced by
elimination due to duplication of the same kind of protein and
D
https://dx.doi.org/10.1021/acs.cgd.0c01136
Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
pubs.acs.org/crystal
areas are clearly separated for some proteins such as 1SKZ and
1TOH. However, the separation is not clear for a few other
proteins such as 5GO6 and 5WKO. For the 20 proteins that
have been crystallized by PEGs, the boundary between the
positive and negative areas is not clear, with some exceptions
such as 1V5C. In 1AO6 and 1Y62, the positive and negative
areas are separated, but the boundary is not distinguishable. An
additional 6, 2, and 2 proteins have been crystallized by sodium
formate, glycerol, and other precipitants. The proteins
crystallized by sodium formate show a distinct separation of
the positive and negative areas, as seen in the proteins
crystallized by high concentrations of AS. Therefore, it can be
concluded that the positive and negative areas neatly split in
the electrostatic potentials for the proteins preferably crystallized by precipitants of strongly hydrated ions such as AS and
sodium formate.
All of the electrostatic potentials in Figures 1 and 2 are
depicted at the pH condition that made the protein net charge
zero. Namely, the pH is equal to the isoelectric point of the
protein. If the protonation states at pH 7.0 are employed for
the respective titratable residues, the protein molecule is not
neutral. For reference, the electrostatic potentials are drawn at
pH 7.0 in Figures S2 and S3. In many proteins, either a positive
or negative area becomes dominant, and one large sphere
appears with the protein molecule centered. Therefore, a large
volume is occupied by either of the positive and negative
electrostatic potentials.
Among the 20 crystal structures crystallized by PEGs, the
electrostatic potential of 1V5C exceptionally displayed a clear
separation of negative and positive areas. Since the electrostatic
potential has a shape common to AS-crystallized proteins,
crystal growth by AS will also be possible for the protein of
1V5C. The PDB entry, 1V5C, is the crystal structure of the
inactive form of chitosanase, an enzyme to hydrolyze partly
acetylated chitosan. The inactive form of chitosanase was
crystallized in a previous study by a precipitant of 20% (w/v)
PEG4000, including 0.4 M AS at pH 3.7.31 The space group of
1V5C was I222, and the maximum resolution was 2.00 Å. In
the present work, we attempted crystallization of chitosanase
by AS. Initially, we searched a solution matrix composed of 12
different AS concentrations of 1.4, 1.6, 1.8, 2.0, 2.2, 2.4, 2.6,
2.8, 3.0, 3.2, 3.4, and 3.6 M and 4 different pH conditions of
5.5, 6.5, 7.5, and 8.5. Since crystal growth was observed at AS
concentrations of 3.2 and 3.4 M at pH 5.5, a focused matrix
was prepared at AS concentrations of 3.0, 3.2, 3.4, and 3.6 M
and pH conditions of 4.5, 5.0, 5.5, and 6.0. Protein crystals
were obtained in several wells of the focused solutions. A single
crystal of chitosanase for structure analysis was grown under
the condition of 3.4 M AS at pH 5.0. Neither PEGs nor the
organic ingredients were included in the precipitant solution.
The space group of the AS-grown crystal was P212121, and the
maximum resolution was 1.74 Å (Table S2). Two chitosanase
molecules were included in the asymmetric unit, unlike 1V5C.
The structures of the two chitosanase molecules were almost
the same. The root-mean-square deviation (RMSD) between
two molecules was 0.131 Å. The structure of the AS-grown
crystal is almost the same as that of the PEGs-grown crystal.
The crystal structure of one of the chitosanase molecules in the
asymmetric unit was superimposed on the 1V5C structure
(Figure 3a). The RMSD between them was 0.136 Å. A slight
structural difference was observed only at the loop region of
residues 107−111.
Article
Figure 3. (a) Structures of the superimposition of AS-crystallized
chitosanase onto PEGs-crystallized chitosanase. Protein molecules are
represented by cartoons with colors changing from blue to red as the
residue goes from the N- to C-terminal sides for the AS-crystallized
molecule. The PEGs-crystallized molecule is shown in gray. Water
molecules are not shown for clarity. (b) Electrostatic potentials for the
AS-crystallized chitosanase with surrounding molecules. A molecule of
interest is represented by the ribbon in magenta, and the surrounding
molecules are shown by the surfaces in green. The positive and
negative isosurfaces are drawn in blue and red, with their contour
values at +0.5 and −0.5 in units of kT/e, respectively. (c) Electrostatic
potentials for the PEGs-crystallized chitosanase with surrounding
molecules. The coloring and drawings are the same as in (b).
One chitosanase molecule in the AS-grown crystal is in
contact with six surrounding molecules in the crystal packing.
The electrostatic potential of the AS-grown chitosanase was
drawn with the surrounding molecules (Figure 3b). The
positive and negative spherical lobes of the electrostatic
potential are located at the spaces unoccupied by other
E
https://dx.doi.org/10.1021/acs.cgd.0c01136
Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
pubs.acs.org/crystal
Article
Figure 4. Isosurfaces of electrostatic potential around the 20 proteins crystallized by PEGs with inclusion of low concentrations of AS. The positive
and negative isosurfaces are shown in blue and red mesh representations, respectively. The contour values of the isosurfaces are +0.5 and −0.5 in
units of kT/e. Protein molecules are represented by green cartoons.
protein molecules. Positive and negative spherical lobes of
electrostatic potentials are observed in most of the proteins.
The lobes are positioned at the spaces not occupied by other
molecules. The molecules in close contact are likely to be
positioned at the node of the two lobes. The proteins in the
AS-grown crystal tend to make contact with other molecules in
the areas where the absolute value of the electrostatic potential
is low. Therefore, the electrostatic potential is responsible for
the molecular packing of protein crystals, which will stabilize
the mixture of AS ions and the proteins in crystal growth. The
protein crystals contain a large amount of solvent,35 and the
protein atoms occupy only about 47% of the unit cell in
volume on average.36 Because the molecular size of AS is much
larger than that of water, AS molecules occupy about 11% of
the total volume in a 3.0 M AS solution.15 The distribution of
AS ions around the protein is highly anisotropic.15 Hence, it is
reasonable to assume that the accumulation of precipitant ions
and the packing of protein molecules are complementary to
each other in crystallization.
In a previous study,31 chitosanase was crystallized by
PEG4000 with inclusion of 0.4 M AS in the precipitant
solution. As mentioned above, the distribution of AS ions is
complementary to the molecular packing in protein crystals.
Hence, AS may assist in the crystallization even for PEGsbased conditions in the case in which the shape of the
electrostatic potential is similar to those of the proteins
preferably crystallized by AS. Then, we selected 20 proteins
that were crystallized by PEGs with the presence of a low
concentration of AS. The electrostatic potentials for the 20
proteins are depicted in Figure 4. Most of the proteins have a
common shape; i.e., the positive and negative areas are almost
equally separated with spherical lobe forms. The protein
molecule is positioned at the boundary region of the two areas.
The electrostatic potentials shown in Figure 4 suggest that
many of the 20 proteins can be crystallized by AS, as
molecules in the crystal. Molecules in close contact are likely to
be positioned at the boundary region between two spherical
lobes. In other words, the molecule of interest makes close
contact with the surrounding molecules in areas where the
absolute value of the electrostatic potential is low. According to
a simulation study of AS ions in a protein crystal droplet,15 the
distributions of AS ions around a protein were anisotropic in
the case of proteins easily crystallized by AS. AS ions tended to
accumulate in two local areas of the surrounding space of the
protein. The distribution of AS ions around the protein
coincided with the shape of the electrostatic potential.
Therefore, AS ions are considered to be attracted to the
areas where the absolute value of the electrostatic potential is
high in the crystallization of chitosanase by AS. The ASattracted areas are identical with the spaces unoccupied by
protein molecules in the crystal packing. The electrostatic
potential in the PEGs-grown crystal was also drawn with the
surrounding molecules for comparison (Figure 3c). The
locations of the surrounding molecules seem to be related to
the electrostatic potential. With one exception, the surrounding
molecules tend to make contact with the molecule of interest
at the boundary regions between two spherical lobes where the
absolute value of the electrostatic potential is low.
Furthermore, no molecule is observed in the positive area of
the electrostatic potential. Since the PEGs-grown crystal was
reported to be generated with a low concentration of AS,31 the
mixed AS ions might assist protein growth by local
accumulation in accordance with the electrostatic potential
even for a PEGs-grown crystal.
Since the molecular packing in a crystal is relevant to the
electrostatic potential in the case of chitosanase as described
above, the relationship between the molecular packing and
electrostatic potential was examined for the 20 selected
proteins (Figure S4). The isosurface of the electrostatic
potential around the protein is depicted with the surrounding
F
https://dx.doi.org/10.1021/acs.cgd.0c01136
Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
pubs.acs.org/crystal
biological unit is sometimes effective for the crystallization by
AS. For 5E7T, the crystal structure was obtained in the
octadecamer. One unit of 5E7T is composed of three parts,
and each part is comprised of six chains. The positive and
negative areas of the electrostatic potential are well separated
due to the association of the three parts, as seen in Figure 2.
The crystal structure of 2YQ0 was obtained in the homooctamer. The association of eight chains gives a fine separation
of the positive and negative areas in Figure 2. For 3IXF and
2QFK, the calculation was performed in the monomer form for
calculation in Figure 4 in spite of the fact that a homodimer
was denoted in the PDB. In both 3IXF and 2QFK, two chains
are connected only through the Asp-Arg salt bridge and no
close contact is observed between the two chains. Further, the
side chains of Asp and Arg at the salt bridge are flexible
because a few alternative conformations are assigned to both
Asp and Arg in the crystal structure. This means that the
relative positions of the two chains are highly variable. For
comparison, we also calculated the electrostatic potential in the
homodimer form. The positive and negative areas of the
electrostatic potential in the monomer are separated more
remarkably in comparison to those in the homodimer. The
crystals for 3IXF and 2QFK were both grown by PEGs with
the inclusion of a low concentration of AS. Since the
distribution of AS ions is complementary to the molecular
packing in protein crystals, AS may assist in the crystallization
by suppressing the flexibility of the connected chains.
AS and PEGs are two major precipitants in protein
crystallization.3 Therefore, these two agents will be first
examined in the search for optimal crystallization conditions.
Many commercial screening kits indeed contain a variety of
concentrations of AS solutions and different kinds of PEGs.39
If some structural information on a protein is available in
advance, we can judge which should be chosen for a detailed
search of crystallization conditions between AS or PEGs. For
example, when a protein crystal is obtained with a precipitant,
but the resolution of X-ray diffraction is not satisfactorily high,
a better quality of protein crystals is occasionally required.
When no good crystal condition is found in screening kits, a
clue for the probable crystallization condition is helpful. Due to
the recent progress of calculation techniques,40,41 a protein
structure can be predicted by homology modeling or de novo
computation. The electrostatic potential from the predicted
structure is beneficial to prioritize the crystallization conditions
that should be examined in detail in experiments. The optimal
zone of precipitant concentrations is sometimes narrow for
crystal growth with AS. The electrostatic potential will be
helpful to judge whether screening with many different
concentrations of AS should be applied to a limited amount
of a protein sample instead of the sparse-condition screening
for other organic precipitants. The electrostatic potential will
also be utilized as a guide for how to modify recombinant
proteins suitable for crystal growth with AS. Modeling of the
protein structure from low-resolution X-ray diffraction or
computer prediction and modification of the protein by
introducing amino acid mutations based on calculation of the
electrostatic potential will provide a suggestion for promoting
crystallization with good quality.
demonstrated for 1V5C. A mixture of AS and PEGs seems
effective for several proteins such as 2GTY and 2P8V. The
electrostatic potentials of these proteins show clear positive
and negative areas, while their shapes are largely deformed
from the spherical lobe form.
To quantitatively estimate the separation of positive and
negative areas of the electrostatic potential, we applied the
approximation with point charges to the spherical shape of the
potential. Provided that the electrostatic potential is caused by
positive and negative point charges placed at the centers of the
respective lobes, the separation is computed from the distance
between the two point charges. The separation distances are
given in Table S3 for all of the 160 proteins, and the average
values are shown in Table 1. The average distance is large for
Table 1. Scaled Separation Distance Estimated by Point
Charge Approximation
precipitant
high concn of AS
moderate concn of AS
PEGs
sodium formate
othersa
AS-included PEGs
distance (Å/3√Mw)
1.94
1.98
1.59
1.89
1.70
2.04
±
±
±
±
±
±
Article
0.60
0.64
0.55
0.52
0.95
0.42
a
Glycerol, ethylene glycol, and sodium chloride.
the 100 proteins that were crystallized by high concentrations
of AS. The distance is also large for the 10 proteins crystallized
by low concentrations of AS. In contrast, the distance is small
for the 20 PEGs-crystallized proteins. Hence, the amplitude of
the separation in electrostatic potential can be evaluated by
approximation with point charges. The distance is also large for
the proteins crystallized by sodium formate. Accordingly, the
degree of positive and negative separation is high for the
proteins preferably crystallized by strongly hydrated ions.
In a previous study, the relationship between AS
concentration and success in protein crystallization was
surveyed.3 The results of the study suggested that protein
crystals were grown in a wide range of AS concentrations from
0 to 4.0 M and that the range had a peak at concentrations of
1.6−2.0 M. The results of another study suggested that the
optimal range of AS concentrations for successful crystal
growth was 1.5−2.5 M.37 Those studies also indicated that the
pH condition for crystal growth did not correlate with the
isoelectric point.3,37 Two isoelectric points are shown in Table
S1. One is the theoretical value obtained by the ProtParam
tool of ExPASy,38 and the other is the pH value that makes the
net charge zero on the basis of pKa calculation for titratable
residues by Propka.24 These two isoelectric points are
consistent (R2 = 0.86). These isoelectric points, however,
hardly correlate with the pH values at which protein crystals
were grown in experiments (R2 = 0.02 and 0.03). This low
correlation is compatible with results of previous studies on the
relationship between isoelectric point and experimental pH in
crystal growth.3,37 Table S1 shows other properties such as the
Matthews coefficient and molecular weight. No correlation was
found between those properties. There was no correlation
between those properties even for subsets of the proteins such
as the 100 structures crystallized by AS concentrations over 2.0
M and the 20 structures crystallized by PEGs.
The electrostatic potentials were fundamentally calculated in
the form of the biological unit shown in the PDB. The
■
CONCLUSION
Electrostatic potentials were examined for 160 kinds of
proteins to clarify the reason some proteins were preferably
crystallized by AS. Among the 160 proteins, 100 had been
G
https://dx.doi.org/10.1021/acs.cgd.0c01136
Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
pubs.acs.org/crystal
Noritaka Nishida − Graduate School of Pharmaceutical
Sciences, Chiba University, Chiba 260-8675, Japan
crystallized by high concentrations of AS and 10 had been
crystallized by moderate concentrations of AS. The electrostatic potentials were compared with those of 20 other proteins
crystallized by PEGs. For comparison, 10 proteins crystallized
by sodium formate, glycerol, ethylene glycol, and sodium
chloride were included. Further, 20 proteins crystallized by ASincluded PEGs were also included. The electrostatic potentials
for the proteins preferably crystallized by AS showed a
common shape in their isosurfaces. The positive and negative
areas were almost equally separated in volume, each area
having a spherical lobe form, and the protein molecule was
positioned at the node of the two areas. In contrast, the
separation of the positive and negative areas was ambiguous for
the proteins preferably crystallized by PEGs. The proteins
crystallized by sodium formate showed a shape in their
electrostatic potential similar to that of AS-crystallized
proteins. The separation was not clear for the proteins
crystallized by glycerol. Therefore, proteins were preferably
crystallized by salts when they had a clear separation of positive
and negative areas in the electrostatic potential around them.
The amplitude of the separation of electrostatic potential was
estimated by approximating the two spherical shapes with
positive and negative point charges. On the basis of the
findings for the difference in electrostatic potential, we
demonstrated that crystallization by AS was possible for a
protein that had been reported to be crystallized by PEGs if the
electrostatic potential showed a clear separation of positive and
negative areas. The electrostatic potential was illustrated with
surrounding molecules in the crystal packing. The spaces
unoccupied by surrounding molecules coincided with regions
where the absolute values of the electrostatic potential were
high. The protein molecules in a crystal made contact with
each other in regions where the absolute value of the
electrostatic potential was low. Therefore, the shape of the
isosurface of the electrostatic potential was responsible for the
molecular arrangement in a crystal and the preference for a
precipitant in crystal growth.
■
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.cgd.0c01136
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
Calculations were performed at the Research Center for
Computational Science, Okazaki, Japan, and at the Information
Technology Center of the University of Tokyo. X-ray
diffraction data were acquired at the Photon Factory, Tsukuba,
Japan (proposal no. 2018G613). A part of this work was
supported by a grant for Scientific Research C from the Japan
Society for the Promotion of Science (18K07138). The crystal
structure of the inactive form of chitosanase was deposited
with the PDB. The accession code is 7CJU.
■
REFERENCES
(1) Gavira, J. A. Current trends in protein crystallization. Arch.
Biochem. Biophys. 2016, 602, 3−11.
(2) Dumetz, A. C.; Chockla, A. M.; Kaler, E. W.; Lenhoff, A. M.
Comparative Effects of Salt, Organic, and Polymer Precipitants on
Protein Phase Behavior and Implications for Vapor Diffusion. Cryst.
Growth Des. 2009, 9, 682−691.
(3) McPherson, A.; Cudney, B. Optimization of crystallization
conditions for biological macromolecules. Acta Crystallogr., Sect. F:
Struct. Biol. Commun. 2014, 70, 1445−1467.
(4) Okur, H. I.; Hladílková, J.; Rembert, K. B.; Cho, Y.; Heyda, J.;
Dzubiella, J.; Cremer, P. S.; Jungwirth, P. Beyond the Hofmeister
Series: Ion-Specific Effects on Proteins and Their Biological
Functions. J. Phys. Chem. B 2017, 121, 1997−2014.
(5) Dumetz, A. C.; Chockla, A. M.; Kaler, E. W.; Lenhoff, A. M.
Protein Phase Behavior in Aqueous Solutions: Crystallization, LiquidLiquid Phase Separation, Gels, and Aggregates. Biophys. J. 2008, 94,
570−583.
(6) Yamanaka, M.; Inaka, K.; Furubayashi, N.; Matsushima, M.;
Takahashi, S.; Tanaka, H.; Masaru Sato, S.; Kobayashi, T.; Tanaka, T.
Optimization of salt concentration in PEG-based crystallization
solutions. J. Synchrotron Radiat. 2011, 18, 84−87.
(7) Collins, K. D. Ions from the Hofmeister series and osmolytes:
effects on proteins in solution and in the crystallization process.
Methods 2004, 34, 300−311.
(8) Dempsey, C. E.; Mason, P. E.; Jungwirth, P. Complex Ion Effects
on Polypeptide Conformational Stability: Chloride and Sulfate Salts
of Guanidiniumand Tetrapropylammonium. J. Am. Chem. Soc. 2011,
133, 7300−7303.
(9) Dumetz, A. C.; Chockla, A. M.; Kaler, E. W.; Lenhoff, A. M.
Protein Phase Behavior in Aqueous Solutions: Crystallization, LiquidLiquid Phase Separation, Gels, and Aggregates. Biophys. J. 2008, 94,
570−583.
(10) Shukla, D.; Schneider, C. P.; Trout, B. L. Complex Interactions
between Molecular Ions in Solution and Their Effect on Protein
Stability. J. Am. Chem. Soc. 2011, 133, 18713−18718.
(11) Dumetz, A. C.; Snellinger-O’Brien, A. M.; Kaler, E. W.;
Lenhoff, A. M. Patterns of protein-protein interactions in salt
solutions and implications for protein crystallization. Protein Sci.
2007, 16, 1867−1877.
(12) Fudo, S.; Qi, F.; Nukaga, M.; Hoshino, T. Influence of
precipitants on molecular arrangement and space group of protein
crystals. Cryst. Growth Des. 2017, 17, 534−542.
(13) Qi, F.; Fudo, S.; Neya, S.; Hoshino, T. A cluster analysis on the
structural diversity of protein crystals, exemplified by human
immunodeficiency virus type 1 protease. Chem. Pharm. Bull. 2014,
62, 568−577.
ASSOCIATED CONTENT
sı Supporting Information
*
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.cgd.0c01136.
PDB codes of proteins and their molecular properties,
statistics in X-ray crystallography, flowchart for selection
of the crystal structures for analysis, electrostatic
potentials depicted in the protonation state at pH 7.0,
and the 20 selected AS-crystallized proteins with
surrounding molecules (PDF)
■
Article
AUTHOR INFORMATION
Corresponding Author
Tyuji Hoshino − Graduate School of Pharmaceutical Sciences,
Chiba University, Chiba 260-8675, Japan; orcid.org/
0000-0003-4705-4412; Phone: +81-43-226-2936;
Email: hoshino@chiba-u.jp
Authors
Yan Guo − Graduate School of Pharmaceutical Sciences,
Chiba University, Chiba 260-8675, Japan
Liang Qu − Graduate School of Pharmaceutical Sciences,
Chiba University, Chiba 260-8675, Japan
H
https://dx.doi.org/10.1021/acs.cgd.0c01136
Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
pubs.acs.org/crystal
(14) Qi, F.; Fudo, S.; Neya, S.; Hoshino, T. A dominant factor for
structural classification of protein crystals. J. Chem. Inf. Model. 2015,
55, 1673−1685.
(15) Kitahara, M.; Fudo, S.; Yoneda, T.; Nukaga, M.; Hoshino, T.
Anisotropic distribution of ammonium sulfate ions in protein
crystallization. Cryst. Growth Des. 2019, 19 (11), 6004−6010.
(16) Sheinerman, F. B.; Norel, R.; Honig, B. Electrostatic aspects of
protein−protein interactions. Curr. Opin. Struct. Biol. 2000, 10, 153−
159.
(17) Hendsch, Z. S.; Tidor, B. Do salt bridges stabilize proteins? A
continuum electrostatic analysis. Protein Sci. 1994, 3, 211−226.
(18) Takahashi, T.; Endo, S.; Nagayama, K. Stabilization of protein
crystals by electrostatic interactions as revealed by a numerical
approach. J. Mol. Biol. 1993, 234, 421−432.
(19) Schmit, J. D.; Whitelam, S.; Dill, K. Electrostatics and
aggregation: How charge can turn a crystal into a gel. J. Chem.
Phys. 2011, 135, 085103.
(20) Hammadi, Z.; Astier, J. P.; Morin, R.; Veesler, S. Protein
crystallization induced by a localized voltage. Cryst. Growth Des. 2007,
7, 1472−1475.
(21) Koizumi, H.; Fujiwara, K.; Uda, S. Control of nucleation rate
for tetragonal hen-egg white lysozyme crystals by application of an
electric field with variable frequencies. Cryst. Growth Des. 2009, 9,
2420−2424.
(22) Alexander, L. F.; Radacsi, N. Application of electric fields for
controlling crystallization. CrystEngComm 2019, 21, 5014.
(23) Fiser, A.; Š ali, A. Modeller: Generation and Refinement of
Homology-Based Protein Structure Models. Methods Enzymol. 2003,
374, 461−491.
(24) Søndergaard, C. R.; Olsson, M. H.; Rostkowski, M.; Jensen, J.
H. Improved treatment of ligands and coupling effects in empirical
calculation and rationalization of pKa values. J. J. Chem. Theory
Comput. 2011, 7, 2284−2295.
(25) Case, D. A.; Betz, R. M.; Botello-Smith, W.; Cerutti, D. S. T.E.;
Cheatham, I.; Darden, T. A.; Duke, R. E.; Giese, T. J.; Gohlke, H.;
Goetz, A. W.; Homeyer, N.; Izadi, S.; Janowski, P.; Kaus, J.;
Kovalenko, A.; Lee, T. S.; LeGrand, S.; Li, P.; Lin, C.; Luchko, T.;
Luo, R.; Madej, B.; Mermelstein, D.; Merz, K. M.; Monard, G.;
Nguyen, H.; Nguyen, H. T.; Omelyan, I.; Onufriev, A.; Roe, D. R.;
Roitberg, A.; Sagui, C.; Simmerling, C. L.; Swails, J.; Walker, R. C.;
Wang, J.; Wolf, R. M.; Wu, X.; Xiao, L.; York, D. M.; Kollman, P. A.
Amber16; University of California: San Francisco, 2016.
(26) Li, L.; Li, C.; Sarkar, S.; Zhang, J.; Witham, S.; Zhang, Z.;
Wang, L.; Smith, N.; Petukh, M.; Alexov, E. DelPhi: a comprehensive
suite for DelPhi software and associated resources. BMC Biophys.
2012, 5, 9.
(27) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.;
Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. UCSF Chimera - a
visualization system for exploratory research and analysis. J. Comput.
Chem. 2004, 25, 1605−1612.
(28) The PyMOL Molecular Graphics System, Ver. 1.7; Schrödinger,
LLC.
(29) Kabsch, W. XDS. Acta Crystallogr., Sect. D: Biol. Crystallogr.
2010, D66, 125−132.
(30) Winn, M. D.; Ballard, C. C.; Cowtan, K. D.; Dodson, E. J.;
Emsley, P.; Evans, P. R.; Keegan, R. M.; Krissinel, E. B.; Leslie, A. G.
W.; McCoy, A.; McNicholas, S. J.; Murshudov, G. N.; Pannu, N. S.;
Potterton, E. A.; Powell, H. R.; Read, R. J.; Vagin, A.; Wilson, K. S.
Overview of the CCP4 suite and current developments. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2011, D67, 235−242.
(31) Adachi, W.; Sakihama, Y.; Shimizu, S.; Sunami, T.; Fukazawa,
T.; Suzuki, M.; Yatsunami, R.; Nakamura, S.; Takenaka, A. Crystal
structure of family GH-8 chitosanase with subclass II specificity from
Bacillus sp. K17. J. Mol. Biol. 2004, 343, 785−795.
(32) Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Refinement of
Macromolecular Structures by the Maximum-Likelihood method.
Acta Crystallogr., Sect. D: Biol. Crystallogr. 1997, D53, 240−255.
(33) Liebschner, D.; Afonine, P. V.; Baker, M. L.; Bunkóczi, G.;
Chen, V. B.; Croll, T. I.; Hintze, B.; Hung, L. W.; Jain, S.; McCoy, A.
Article
J.; Moriarty, N. W.; Oeffner, R. D.; Poon, B. K.; Prisant, M. G.; Read,
R. J.; Richardson, J. S.; Richardson, D. C.; Sammito, M. D.; Sobolev,
O. V.; Stockwell, D. H.; Terwilliger, T. C.; Urzhumtsev, A. G.; Videau,
L. L.; Williams, C. J.; Adams, P. D. Macromolecular structure
determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr. 2019, D75, 861−877.
(34) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K. Features
and development of Coot. Acta Crystallogr., Sect. D: Biol. Crystallogr.
2010, D66, 486−501.
(35) Oswald, R.; Pietzsch, M.; Ulrich, J. A view inside the nature of
protein crystals. J. Cryst. Growth 2017, 469, 176−179.
(36) Kantardjieff, K. A.; Rupp, B. Matthews coefficient probabilities:
Improved estimates for unit cell contents of proteins, DNA, and
protein−nucleic acid complex crystals. Protein Sci. 2003, 12, 1865−
1871.
(37) Gaur, R. K. Ammonium sulfate concentration optimization and
its relation with protein parameters for crystallization. IIOAB J. 2018,
9, 28−36.
(38) http://ca.expasy.org/tools/protparam.html.
(39) For example: https://hamptonresearch.com/product-CrystalScreen-Crystal-Screen-2-Crystal-Screen-HT-1.html; http://www.
rigakureagents.com/c-1-crystallography-screens.aspx.
(40) Xu, D.; Zhang, Y. Ab initio protein structure assembly using
continuous structure fragments and optimized knowledge-based force
field. Proteins: Struct., Funct., Genet. 2012, 80, 1715−1735.
(41) Raman, S.; Vernon, R.; Thompson, J.; Tyka, M.; Sadreyev, R.;
Pei, J.; Kim, D.; Kellogg, E.; DiMaio, F.; Lange, O.; Kinch, L.; Sheffler,
W.; Kim, B.-H.; Das, R.; Grishin, N. V.; Baker, D. Structure prediction
for CASP8 with all-atom refinement using Rosetta. Proteins: Struct.,
Funct., Genet. 2009, 77, 89−99.
I
https://dx.doi.org/10.1021/acs.cgd.0c01136
Cryst. Growth Des. XXXX, XXX, XXX−XXX