This article was downloaded by: [Kungliga Tekniska Hogskola]

On: 05 October 2014, At: 00:27

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Journal of Biomolecular Structure and Dynamics

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tbsd20

Structural Determinants of V. cholerae CheYs that

Discriminate them in FliM binding: Comparative
Modeling and MD Simulation Studies
a a
Jhimli Dasgupta & Jiban K. Dattagupta
a
Crystallography and Molecular Biology Division , Saha Institute of Nuclear Physics , 1/AF
Bidhannagar, Kolkata , 700064 , India
Published online: 15 May 2012.

To cite this article: Jhimli Dasgupta & Jiban K. Dattagupta (2008) Structural Determinants of V. cholerae CheYs that
Discriminate them in FliM binding: Comparative Modeling and MD Simulation Studies, Journal of Biomolecular Structure and
Dynamics, 25:5, 495-503, DOI: 10.1080/07391102.2008.10507196

To link to this article: http://dx.doi.org/10.1080/07391102.2008.10507196

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Journal of Biomolecular Structure &
Dynamics, ISSN 0739-1102
Volume 25, Issue Number 5, (2008)
©Adenine Press (2008)

Structural Determinants of V. cholerae Jhimli Dasgupta

CheYs that Discriminate them in FliM binding: Jiban K. Dattagupta*
Comparative Modeling and MD Simulation Studies
Crystallography and
http://www.jbsdonline.com Molecular Biology Division
Saha Institute of Nuclear Physics
Downloaded by [Kungliga Tekniska Hogskola] at 00:27 05 October 2014

Abstract
1/AF Bidhannagar
Chemotaxis of Vibrio cholerae is a complex process where multiple paralogues of various
chemotaxis genes participate. V. cholerae contains five copies of the response regulator Kolkata, 700064, India
protein CheY (CheYV) and the role played by these CheY homologs in chemotaxis and viru-
lence are investigated only through a few in vivo studies. As identification of the molecular
features that discriminate CheYVs in terms of FliM binding is necessary for the detailed
understanding of chemotaxis and pathogenesis, we built the models of CheYVs through com-
parative modeling and MD simulation was performed on each model in their phosphorylated
and Mg+2 bound state. Our analysis identified the key structural elements, unique to CheY3V,
which complement the N-terminal part of FliMV and we explained how the structure, shape,
and surface properties of the FliM binding pocket of other CheYVs abrogate this function.
Furthermore, we have provided the structural basis of a putative cross species interaction
between CheYE and FliMV, identified in a recent in vivo study.

Key words: Vibrio cholerae; Chemotaxis; FliM binding; Comparative modeling; and MD
simulation.

Introduction

The ability of the motile bacteria to swim toward or away from specific environ-
mental stimuli to provide cells with a survival advantage is called chemotaxis,
which is mediated by briefly reversing the direction of rotation of the flagellar
motors (1). Chemotaxis has been extensively studied in Escherichia coli and Sal-
monella typhimurium, where the process is governed by the activity of single copy
of each chemotaxis protein in a linear pathway (2). But chemotaxis in Vibrio
cholerae is far more complex having multiple paralogues of various chemotaxis
genes. Although the requirement of chemotaxis are established for full virulence
of V. cholerae (3, 4), functional elucidation of the multiple chemotaxis genes are
restricted only to a few in vivo studies.

In E. coli, methyl-accepting chemotaxis proteins activate CheA through a linker

protein CheW (5). Activated CheA phosphorylates the response regulator CheY.
Phosphorylated CheY interacts with FliM to influence the direction of flagellar
rotation from counterclockwise to clockwise, reorienting the cell from smooth
swimming to tumbling motion (6).
*
Phone: +91-33-2321-4986
V. cholerae contains five copies of the response regulator protein CheY (7). At- Fax: +91-33-2337-4637
tempts to identify the cheY gene responsible for the motion of flagellum showed Email: jibank.dattagupta@saha.ac.in

Abbreviations: CheYV, Vibrio cholerae CheY; CheYE, E. coli CheY; FliMV, Vibrio choerae FliM;
CheYE, E. coli CheY; MD, Molecular Dynamics. 495
 496 that the insertional disruption of the cheY4 gene results in decreased motility while
insertional duplication of this gene increases motility (8). In contrast, a deletion
mutant of cheY3 was found to impair chemotaxis (9) and an in vivo study reported
Dasgupta and Dattagupta that only CheY3V directly switches the flagellar rotation (10).

However, to date, no report is available that investigate the molecular basis of

FliM binding for V. cholerae CheY(s), though it is required to understand the che-
motaxis of V. cholerae in more detail. To identify the molecular features of V. chol-
erae CheY proteins that discriminate them in terms of FliM binding, we built the
models of CheY -1V, -2V, -3V, and -4V through comparative modeling (CheY0V is
excluded, reason is given later) and Molecular Dynamics (MD) simulation of 1 ns
was performed on each model in their Mg+2 bound and phosphorylated state. Our
analysis identified the key structural elements, unique to CheY3V, that make it the
most suitable for FliM binding and the structural features of other three CheYVs,
responsible for the abrogation of FliM binding. A rational molecular explanation
for the putative cross species interaction between CheYE and FliMV, identified
experimentally in a recent in vivo study (10), is also provided here.
Downloaded by [Kungliga Tekniska Hogskola] at 00:27 05 October 2014

Materials and Methods

Comparative Modeling of CheYs of V. cholerae

Sequences of CheYVs were blasted (using WU-BLAST) against PDB and the
templates for modeling were chosen, based on the highest sequence identity.
Models were built using 3D-JIGSAW homology modeling server (11), version
2.0. For CheY1V and CheY4V models, CheY of S. typhimurium (PDB code:
2CHE; 12) was chosen as template as each of CheY1V and CheY4V possess 41%
sequence identity with S. typhimurium CheY, while their identity with their next
homologue, E. coli CheY (PDB code: 1F4V; 13), are of 40% and 38%, respec-
tively. PhoP receiver domain of B. subtilis (PDB code: 1MVO; 14) was chosen
as a template for CheY2V showing an identity of 35% while S. typhimurium CheY
(PDB code: 2CHE) was the second best showing 33% identity with CheY2V.
CheY3V, showed 66% identity with S. typhimurium CheY (PDB code: 2CHE; 12)
and E. coli CheY (PDB code: 1F4V; 13); however, we have chosen E. coli CheY
as a template for CheY3V.

As FliM binds with phosphorylated CheY, we considered BeF3 and Mg+2 bound
CheY structure (PDB code: 1F4V; 13) to model the active site and the FliM bind-
ing face of each CheYV. Energy minimizations of the models were performed
using DISCOVER-3 module of InsightII using the steepest descent algorithm
(down to a gradient of <100 kcal mol-1 Å-1) followed by conjugate gradient mini-
mization (down to <10 kcal mol-1 Å-1) with an initial backbone restrain and then
on the solvated unrestrained models.

Modeling of the N-terminal Part of V. cholerae FliM

N-terminal 16 residues of FliME, known to be responsible for CheYE binding, differ

from FliMV at five positions (Fig. 1B). The model of FliMV was made by mutat-
ing the N-terminal peptide (1-16) of FliME (PDB code: 1F4V; 13) in silico, using
the BIOPOLYMER module of InsightII, followed by energy minimization with
a main-chain restrain. To make CheYV-FliMV model, CheY3V was superposed
on CheYE part of CheYE-FliME complex (PDB code: 1F4V) and the transformed
coordinate of CheY3V was appended with the coordinate of FliMV. The resulting
CheYV-FliMV model was subjected to energy minimization and subsequent simu-
lation. We have used main-chain restrain at very initial stage of energy minimiza-
tion of FliMV model. As at this stage, the model was in vacuo, we kept the restrain
to avoid artifacts. All subsequent energy minimizations and MD simulations were
then performed on the unrestrained complex model of CheYV and FliMV.
 A 497

FliM Binding of CheY

Figure 1: (A) Sequence alignment of CheYVs with

CheYE. ‘*’ represent residues that constitute the ac-
tive site, and ‘•’ represent the residues involved in FliM
binding; (B) Alignment of N-terminal 18 residues of
FliME with FliMV.
Downloaded by [Kungliga Tekniska Hogskola] at 00:27 05 October 2014

Incorporations of Mg+2 and Phosphate Ion at the Active Site of CheYV Models

Each CheYV model was locally superposed on BeF3 and Mg+2 bound CheYE con-
sidering the active site residues D12, D13, and D57 (CheYE numbering). The co-
ordinates of the Mg+2 ion and BeF3 were then retrieved and appended with the
transformed coordinate of CheYV. Be and F atoms were replaced by P and O- and
the bond distances and angles were adjusted according to the geometry specifica-
tions of the BUILDER module of InsightII. OD2 atom of D57 (CheYE numbering)
served as the 4th coordination of P.

Energy Minimization and MD Simulation

To start with, hydrogen atoms were generated at pH 7.0 and energy minimized keep-
ing the heavy atoms fixed. Potentials were assigned using CVFF forcefield with a
distance independent dielectric constant of 4.0. Atom based method was used to
treat non-bonded interactions with a cutoff distance of 9.5Å. Models were solvated
with a 10Å water shell using the SOAK utility of DISCOVER-3 and the water mol-
ecules were minimized and simulated for 30ps, keeping the protein models fixed
(15). The protein-solvent assembly was then used as starting point for NVT (con-
stant volume and temperature) MD simulations at 298K using the Verlet velocity
algorithm (16) to generate possible stable conformations. In each case, a total simu-
lation run of 1ns was performed with 1fs time step. Structures were validated using
PROCHECK and CONTACT of CCP4 suite (17). Last 100ps was considered as
production run. Twenty snapshots, with an interval of 5 ps, were collected from the
last 100 ps production run and used for the analysis of the hydrogen-bonding and
hydrophobic interactions, using ‘O’ (18) and InsightII. The interactions that appear
in all the snapshots were considered to have 100% frequency of occurrence.

To test the reliability of the simulation protocol, used here, we performed a simula-
tion run of 1 ns on the crystal structure of CheYE-FliME at 298K and the r.m.s. (root
mean square) fluctuations of the simulated structure were compared with the rmsf
values derived from the crystallographic B-factors (Fig. 2) and they are in good
agreement with each other. No major structural deviation was observed around the
active site and the FliM binding site. All reported interactions between CheYE and
FliME are retained in the simulated structure.
 498 Results

Sequence Comparison Among the CheYV Proteins

Dasgupta and Dattagupta
Multiple amino acid sequence alignment of all five CheYVs using
ClustalW shows significantly weak identity (of about 17%). Iden-
tical sequences mainly cover the residues critical for phosphory-
lation and metal binding. However, pair-wise sequence compari-
sons among the CheYVs show about 30-39% identity (Fig. 1A
and Supplementary Table I). Interestingly, for pair-wise sequence
comparison, the regions having identical residues are distributed
differently for different pair. These facts probably indicate that
the four CheYVs, CheY -1V, -2V, -3V, and -4V have evolved to per-
form different cellular functions, mediated by phosphorylation.

CheY0V, on the other hand, is substantially different from the oth-

er four CheYVs, as revealed by sequence comparisons (Supple-
mentary Table I) and several amino acid insertions (four residue
Downloaded by [Kungliga Tekniska Hogskola] at 00:27 05 October 2014

after D12/D13 and >20 residue at the C-terminus). These obser-

vations, combined with the available literature on CheY0V, con-
vinced us to exclude it from our modeling and simulation run.

CheYV Models, Their Overall Structural Organization and

Site of Phosphorylation

As expected, the CheYV models exhibit the structural organiza-

tion of (β/α)5 like other response regulators. The Cα superpo-
sition of the simulated average structure of CheY3V on CheYE
produced the lowest rmsd (root mean square deviation) value of
0.7Å, which was 1.2Å, 1.5Å, and 1.0Å for CheY -1V, -2V, and
-4V, when compared with CheYE.

In CheYE, D57 is the site of phosphorylation and D12, D13,

T87, and K109 surrounding it comprise the active site. The car-
boxylate groups of D13 and D57 coordinate with the Mg2+ ion
(13); whereas K109, T87, and Mg2+ interact directly with the
phosphate group and are believed to assist the phosphorylation
and dephosphorylation events. All the active site residues are
conserved in the CheYVs, considering the conservative replace-
ment of T87→S in CheY2V (Fig. 1A).

In the simulated average structures of CheYVs, the active site

residues are seen in proper orientation and geometry to coordi-
nate Mg2+ and the phosphate group (Fig. 3). No amino acid vari-
ation is found around the active site of any CheYV, which would
hamper phosphorylation. However, two conserved residues of
CheYE, A88 of β4α4 loop, and M17 of α1 helix, implicated in
CheA binding, are substituted in CheY1V, CheY2V, and CheY4V
(Fig. 1A). The residue corresponding to A88 of CheYE is Thr in
CheY1V, CheY2V, and CheY4V, whereas M17 is Val in CheY1V,
CheY2V, and Ile in CheY4V. Interestingly, both of these residues
are conserved in CheY3V (Fig. 1A).

Difference in the Surface Properties of CheYVs

Figure 2: The root mean square fluctuation (rmsf) values are compared for
at the FliM Binding Pocket
CheYE-FliME complex structures. The dotted line represents the rmsf values
calculated from crystallographic B-factor and solid line represents the rmsf I95 and A99 of α4 helix of CheYE are known to be significant in
values calculated from MD simulations. (A) Residues 91-109 of CheYE, (B) forming the hydrophobic wall for FliM binding. In CheY -1V, -2V,
Residues 119-125 of CheYE, and (C) Residues 4-16 of FliME.
 and -4V, the residue corresponding to I95 is either Arg or Lys (Fig. 1A). Although the 499
residue corresponding to A99 is a hydrophobic residue Ile in CheY2V, the correspond-
ing position of CheY1V and CheY4V is occupied by Arg and Lys, respectively. These
substitutions suffice to destroy the hydrophobicity of the CheY surface that faces the
FliM Binding of CheY
hydrophobic part of FliM (Fig. 4). In addition, the polar residues, known to form
hydrogen bonds with FliM, are also found to be replaced in CheY -1V, -2V, and -4V,
compared to CheYE, by the residues of different polarity and sizes (Fig. 1A). Posi-
tively charged K91 of CheYE is replaced either by Pro or Glu in these three CheYVs
while K119 is replaced by Val or Thr (Fig. 1A). A Tyr106→Trp mutant of CheYE
showed that Trp at this position prefers to stay in ‘inside’ position and favors FliM
binding (19, 20). Snapshots of our CheY4V model show W107 stays predominantly
in ‘inside’ conformation, which is favorable for FliM binding. However, analysis
of CheY4V model revealed that the conformation and surface potential of its β4-α4
region is substantially different from CheYE, which is possibly caused by the replace-
ments K91(CheYE)→S and K92→P (Fig. 1). This structural change not only steri-
cally hinders the approach of FliMV with an undistorted conformation toward α5-β4-
α4 face of CheY4V, but also destroys an important salt bridge interaction with FliMV,
as P93 of CheY4V is incapable of forming that as seen in case of K92 of CheYE.
Downloaded by [Kungliga Tekniska Hogskola] at 00:27 05 October 2014

The surface potentials for the simulated average structures of CheY -1V, -2V, and
-4V and their comparisons with CheYE revealed that, in these three CheYVs, the
residue types and their spatial arrangement not only change the surface charge
distributions but also destroy the depth of the hydrophobic cavity (Fig. 4), which
may in turn hinder FliM binding. However, these important residues are found to
be conserved in CheY3V and the surface charge distribution of CheY3V shows a
high degree of resemblance with CheYE at FliM binding site. Similarity of overall
surface potential and the depth of the hydrophobic cavity between CheY3V and
CheYE make CheY3V a suitable candidate for FliM binding (Fig. 4). The detailed
interactions between CheY3V and FliMV and their comparison with CheYE-FliME
complex are discussed in the next section.

Interactions of CheY3V with FliMV

In the CheYE-FliME complex, residues D3-S7 of FliME adopt an extended con-

formation and interact with CheYE both at the backbone and the side chain level
whereas residues Q8-N16 of FliME form two helical turns and pack with the α5-β4-
α4 face of CheYE (13). Hydrophobic interactions, which account for ~70% of the
interface, are formed by L6, I11, and L15 of FliME interacting with I95, A99, and
A103 of the α4 helix of CheYE. Y106 of CheYE, the well-known conserved aro-
matic residue at β5 (19) adopts an ‘inside’ position and in combination with another
β5 residue V108, forms the hydrophobic core for FliM binding. K119 and K122,
belonging to the α5 helix of CheYE, form salt bridges with D12 and N16 of FliME.

Structural resemblance of CheY3V with CheYE and their similar surface properties,
that seem to be suitable for FliM binding, drove us to run MD simulation on the com-
plex model CheY3V-FliMV. Analysis of the production run trajectories revealed that
CheY3V-FliMV complex resemble its E. coli counterpart in terms of the hydrogen
bonds, hydrophobic, and salt bridge interactions (Fig. 5) despite a couple of substitu-
tions at the N-terminal of FliMV, compared to FliME (Fig. 1B). The hydrogen bonds
and hydrophobic interactions that show more than 80% frequency of occurrence (see
Materials and Methods) in the production run of CheY3V-FliMV complex are con-
sidered and compared with the interactions of CheYE-FliME complex (Table I). The
hydrogen bonding interactions, observed between D3-S4 of FliME and K91-K92
of CheYE in CheYE-FliME are expected to be maintained in CheY3V-FliMV as the
contacts between T2-D3 of FliMV and K94-R95 of CheYV (Fig. 5B). A network of
hydrogen bonds is additionally expected in CheY3V-FliMV, connecting FliMV with
R76 of CheY3V through N107 and R129, which is not seen in CheYE-FliME com-
plex possibly due to the substitutions with smaller amino acids at the corresponding
 500 A B C

Dasgupta and Dattagupta

D E
Downloaded by [Kungliga Tekniska Hogskola] at 00:27 05 October 2014

Figure 4: (A) Surface representation of CheYE with bound FliME (in stick); (B, C, and E) Surface
representation of CheY -1V, -2V, -4V, respectively; (D) Surface respresentation of CheY3V with bound
FliMV (in stick).

A B
α4

T2 K94
α5 L14 α4 R95
D3
H15
T2

Figure 3: The disposition of the active side residues. C D

(A) D12, D13, D57, N59 (carbonyl O), and K109 along
with BeF3 and Mg+2 in CheYE–FliME complex (PDB
code: 1F4V); (B) D15, D16, D60, N62 (carbonyl O), K125 V106

and K112 along with bound PO3 and Mg+2 in the simu- α5 A102

lated average model of CheY3V.

L14
Y109
K122 H15
H15 α4
D11
L13
I10 I98

Q7

Figure 5: Interactions of FliMV (yellow) with CheY3V (magenta); (A) Overall representation showing the
binding of FliMV with the α5-β4-α4 face of CheY3V; (B & C) Hydrogen bonding and salt bridge interac-
tions between CheY3V and FliMV; (D) Hydrophobic interactions between FliMV and the α4 of CheY3V.

A B

V106
Figure 6: (A) CheYE (cyan) is superposed on CheY3V L14
A103
(magenta) part of CheY3V-FliMV (yellow) complex.
(B) Sphere model of CheY3V-FliMV complex, showing
hydrophobic residues of FliMV in blue, and hydrophilic
Y109
and others in violet. The patch of negative charge, ex-
pected after K120→Glu mutation is shown in brown. α4
 positions of CheYE. The salt bridge made by K119 and hydrogen bond made by 501
K122 with D12 and N16, respectively, in CheYE-FliME complex are maintained
by K122 and K125 (of CheY3V) with D11 and H15 (of FliMV) in CheY3V-FliMV
(Fig. 5C and Table I). The hydrophobic interactions observed between I95, A99,
FliM Binding of CheY
and A103 of CheYE with L6, I11, and L15 of FliME that account for the ~70% of
Table I
the interface, are maintained in CheY3V-FliMV complex by I98, A103, and V106 Hydrogen bonds and hydrophobic interactions in CheYE-
of CheY3V with L5, I10, and L14 of FliMV (Fig. 5D and Table I). FliME and CheYV-FliMV.

Intermolecular hydrogen bonds

Role of V106 of CheY3V in relation to Asn16→His15 substitution of FliMV
FliME CheYE FliMV CheY3V

H15 of FliMV is found to be in place to interact with the carbonyl O of Y109 T2 OG1 K94 NZ
of CheY3V (Fig. 5C), similar to the interaction of N16 of FliME with Y106 D3 OD1 K91 NZ D3 O R95 N
D3 OD2 K91 NZ D3 OD1 R95 NH1
of CheYE. However, in order to accommodate the bigger aromatic side chain
D3 OD2 R95 NH1
of H15 (FliMV) compared to N16 (FliME), an outward movement of the L14- S4 O K92 N
H15 region of FliMV is observed in the model of CheY3V-FliMV. Due to this S4 OG K92 NZ
movement, hydrophobic interaction observed between L15 (FliME) and A103 L6 N A90 O L5 N A93 O
of CheYE in the CheYE-FliME complex would have been reduced in CheY3V- Q8 OE1 V108 N Q7 OE1 V111 N
FliMV. Although A103 of CheYE is conserved in CheY -1V, -2V, and -4V, it is Q8 NE2 Y106 O Q7 NE2 Y109 O
Downloaded by [Kungliga Tekniska Hogskola] at 00:27 05 October 2014

a Val in CheY3V. This A103 (CheYE)→V106 (CheY3V) substitution plays a D12 OD1 K119 NZ D11OD1 K122 NZ
vital role in regaining the hydrophobic interactions with L14 of FliMV, since D16 O K122 NZ H15 O K125 NZ
V106 of CheY3V accounts for the extra space created by the outward move- D16 OD1 Y106 N
D16 OD2 Y106 O H15 NE2 Y109 O
ment of C-terminal part of the FliM (Figs. 5D and 6A). This kind of space
compensation for FliM binding can not be expected from other three CheYVs, Intermolecular hydrophobic interactions
since Ala is present at this crucial position for them. L6 I95 L5 I98
I11 V108 I10 V111
I11 I95 I10 I98
Probable Structural Basis of the Cross-species Interaction
L15 I95
Between CheYE and FliMV L15 Y106 L14 Y109
L15 A103 L14 V106
A recent in vivo study (10) reported that though CheY3V could act efficiently on
FliME, CheYE only partially interact with FliMV. They predicted that FliMV would
not favor CheYE binding due to an electric repulsion between D8 of FliMV and E117
of CheYE. We have performed a careful scrutiny of CheYE-FliME complex structure
and observed that E117 of CheYE is more than 15Å away from A9 of FliME. There-
fore, we can expect that D8 of FliMV, which corresponds to A9 of FliME (Fig. 1B),
would be about 15Å away from E117 of CheYE (corresponds to K120 of CheY3V in
Figs. 1A and 6B) in the putative cross species complex. As it is hard to believe that
electric repulsion has any influence to disfavor the interaction at this distance, we
prepared a model of CheYE and FliMV and identified two plausible reasons, which
can be responsible for the reduction in interaction between CheYE and FliMV. While
V106 of CheY3V (A103 of CheYE) is able to interact hydrophobically with L14 of
FliMV, (as it fills up the extra space, caused by the movement of FliM helix to ac-
commodate H15 side chain), A103 at this position of CheYE is unable to maintain
that hydrophobic contact with L14 of FliMV (Fig. 6A). Moreover, the network of
hydrogen bonds, that has been observed to connect FliMV with CheY3V in CheY3V-
FliMV complex involving residues N107 and R129, is not possible in case of CheYE-
FliMV, as S104 and K126 of CheYE are incapable of forming the network due to their
size differences from their CheY3V counterparts. The knockout cells, expressing the
mutant K120 (CheY3V)→Glu, reorient less often than the cells expressing the wild
type protein (10). Our analysis reveals that a number of negatively charged residues
(D30, D124, E128) line up around K120 of CheY3V and a K120→E mutation at this
position, coupled with those charged residues, create a patch of negative charge on
the surface of CheY3V (Fig. 6B). The relationship between this negative patch on
CheY3V surface and the behavioral change observed in knockout cells expressing
K120→E CheY3V mutant, however, cannot be ruled out.

Discussion

In this study, we have compared the amino acid sequences of CheYVs and analyzed
 502 their modeled structures to understand the key structural elements that discriminate
CheYVs in terms of FliM binding. Considerable diversity is observed at the primary
structural level, as evident by a weak sequence identity among these four (Fig. 1
Dasgupta and Dattagupta and Supplementary Table I). The residues, strictly conserved in all of them, mainly
constitute the active site and are properly oriented to coordinate the Mg+2 ion and in
the correct geometry to be phosphorylated, which is the same for all four CheYVs.
However, despite a considerable identity observed in the pairwise sequence com-
parisons, the identical regions are distributed differently for different pairs. These
two facts presumably imply that the four CheYV proteins are evolved to perform
different cellular function, mediated by phosphorylation.

A comparison of the structural data, coupled with a mapping of amino acid distri-
bution, reveals that while some residues at the FliM binding pocket of CheY -1V,
-2V, -4V destroy the depth and hydrophobicity of the pocket, other residues reduce
the feasibility of salt bridge and hydrogen bonding interactions, required for FliM
binding. On the contrary, CheY3V possesses a suitable depth and hydrophobicity
at the FliM binding pocket and the residues that form the hydrogen bonds with
FliMV are found in correct orientation. Interestingly, a conserved Ala of other
Downloaded by [Kungliga Tekniska Hogskola] at 00:27 05 October 2014

CheYVs and CheYE is replaced by V106 in CheY3V, which plays a crucial role to
interact hydrophobically with FliMV. But this kind of hydrophobic interaction is
not possible for other three CheY3Vs with a shorter Ala at this position. There-
fore, mere conservation of the residues, capable of maintaining the salt bridge and
hydrophobic interactions at the FliM binding surface are not enough to make a
CheYV suitable for FliMV binding. Rather, a correct combination of the residues
at CheYV-FliMV interface, capable of complementing each other, is essential and
CheY3V is the only candidate, among these four, to possess all required molecular
features to complement the N-terminal part of FliMV.

Acknowledgements

We thank Dr. U. Sen of C&MB Division, SINP for his comments and suggestions
and Sophia Tsai of MCB, USC, USA for her assistance in checking the language of
the manuscript. The work was partially supported by the Council of Scientific and
Industrial Research (No. 21/0653/06/EMR-II) and Department of Biotechnology
(No. BT/PRO0139/R&D/15/11/96), Government of India.

Supplemental Materials

Supplemental materials can be acquired, free of charge, from the contact author or
from Adenine Press for $25.

References and Footnotes

1. S. M. Block, J. E. Segall, and H. C. Berg. J Bacteriol 154, 312-323(1983).

2. J. J. Falke, R. B. Bass, S. L. Butler, S. A. Chervitz, and M. A. Danielson. Annu Rev Cell Dev
Biol 13, 457-512 (1997).
3. G. W. Jones and R. Freter. Infect Immun 14, 240-245 (1976).
4. R. Freter and P. C. O’Brien. Infect Immun 34, 222-233 (1981).
5. J. A. Gegner, D. R. Graham, A. F. Roth, and F. W. Dahlquist. Cell 70, 975-982 (1992).
6. A. Bren and M. Eisenbach. J Mol Biol 312, 699-709 (2001).
7. J. F. Heidelberg, J. A. Eisen, W. C. Nelson, R. A. Clayton, M. L. Gwinn, R. J. Dodson, D.
H. Haft, E. K. Hickey, J. D. Peterson, L. Umayam, S. R. Gill, K. E. Nelson, T. D. Read, H.
Tettelin, D. Richardson, M. D. Ermolaeva, J. Vamathevan, S. Bass, H. Qin, I. Dragoi, P. Sell-
ers, L. McDonald, T. Utterback, R. D. Fleishmann, W. C. Nierman, and O. White. Nature
406, 477-483 (2000).
8. R. Banerjee, S. Das, K. Mukhopadhyay, S. Nag, A. Chakrabortty, and K. Chaudhuri. FEBS
Letters 532, 221-226 (2002).
9. S. M. Butler and A. Camilli. Proc Natl Acad Sci USA 101, 5018-5023 (2004).
10. A. Hyakutake, M. Homma, M. J. Austin, M. A. Boin, C. C. Hase, and I. Kawagishi. J Bac-
teriol 187, 8403-8410 (2005).
11. P. A. Bates, L. A. Kelley, R. M. MacCallum, and M. J. Sternberg. Proteins Suppl 5, 39-
46 (2001).
 12. A. M. Stock, E. Martinez-Hackert, B. F. Rasmussen, A. H. West, J. B. Stock, D. Ringe, and 503
G. A. Petsko. Biochemistry 32, 13375-13380 (1993).
13. S. Y. Lee, H. S. Cho, J. G. Pelton, D. Yan, R. K. Henderson, D. S. King, L. Huang, S. Kustu,
E. A. Berry, and D. E. Wemmer. Nat Struct Biol 8, 52-56 (2001). FliM Binding of CheY
14. C. Birck, Y. Chen, F. M. Hulett, and J. P. Samama. J Bacteriol 185, 254-261 (2003).
15. J. Dasgupta, U. Sen, and J. K. Dattagupta. Protein Engineering 16, 489-496 (2003).
16. W. C. Swope and H. C. Anderson. J Chem Phys 76, 637–649 (1982).
17. Collaborative Computational Project, Number 4, Acta Crystallogr D50, 760-763 (1994).
18. T. A. Jones, J. Y. Zou, S. W. Cowan, and M. Kjeldgaard. Acta Crystallogr A47, 110-
119 (1991).
19. X. Zhu, C. D. Amsler, K. Volz, and P. Matsumura. J Bacteriol 178, 4208-4215 (1996).
20. X. Zhu, J. Rebello, P. Matsumura, and K. Volz. J Biol Chem 272, 5000-5006 (1997).
Downloaded by [Kungliga Tekniska Hogskola] at 00:27 05 October 2014

Date Received: June 20, 2007

Communicated by the Editor Manju Bansal