KcsA: mutational analysis of ion selectivity with molecular dynamics

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1

KcsA: Mutational Analysis
of Ion Selectivity with Molecular Dynamics

By:
Melia Tabbakhian

Master of Science in Physiology and Biophysics

University of Southern California

August 2013

2
Table of Contents
Abstract 4
Chapter 1: Introduction 6
Chapter 2: Methods 15
Chapter 3: Results 21
Chapter 4: Discussion 29
References 33

3
Acknowledgments

Foremost, I would like to express my sincere gratitude to my advisor Professor
Robert Farley for the continuous support of my M.S. study and research, for his
patience, motivation, enthusiasm and immense knowledge. His guidance helped me
during the entirety of my research and the writing of this thesis. I could not have
imagined a better advisor and mentor for my M.S. study.

A most special thanks goes to Dr. Van Ngo and Dr. Darko Stefanovski for their
unbelievable assistance in my research at every step of the way, for their stimulating
discussions, for the countless days spent generating data and for their constant
support of new ideas.

I would like to thank the rest of my thesis committee: Prof. Robert Chow, Prof.
Harvey Kaslow, and Prof. Vijay Kaldra for their encouragement, insightful
comments, and support in my future work.

4

Master’s Thesis
KcsA: A Mutational Analysis of Ion Selectivity with Molecular Dynamics
Abstract
The mechanism of ion selectivity in K
+
-selective ion channels has not yet
been determined. In previous studies it was found that K
+
ions were allowed
entrance into the channels’ selectivity filter while Na
+
ions were excluded from the
selectivity filter at threonine 75. To determine ion selectivity in the bacterial K
+
-
selective channel KcsA, mutational analysis was conducted on threonine 75 using
molecular dynamics simulations. A step-wise pulling protocol was used to pull both
K
+
and Na
+
ions through the wild type and mutant channels in order to estimate any
changes of conductance or selectivity by the use of force profiles. Mutations were
chosen specifically to investigate the importance of each of the side groups of
threonine to selectivity. The three mutations studied changed threonine 75 to
serine, valine or cysteine. In comparison to the wild type, both the valine and serine
mutants excluded K
+
and Na
+
from entering into the selectivity filter while the
cysteine mutant allowed both the K
+
and Na
+
ion to enter the selectivity filter.
Interpretation of this data indicated that valine and serine mutants both decrease
the conductance as well as the selectivity ratio of the KcsA channel. However, the
cysteine mutant, which lost both methyl and hydroxyl groups, allows entry of both
K
+
and Na
+
ions into the selectivity filter and may cause an increase in conductance
as well as a decrease in the selectivity ratio. Results from molecular dynamics
simulations help investigators to look at particle movement on a short temporal
5
scale as well as helps make future experimental design more efficient. In future
work, mutations will be inserted into KcsA DNA using PCR and conductance
measurements will be taken using the patch clamp technique. These experiments
will help determine mutational effects on selectivity and conductance of the actual
ion channel.

6
Chapter 1: Introduction: Ion Channel
Ion channels are integral membrane proteins that form pores within cell
membranes. The subsequent movement of ions down their electrochemical
potential gradients at different rates generates resting potentials. By gating the
stream of specific ions across the cell membrane and regulating the flow of ions
across secretory and epithelial cells, ion channels establish resting membrane
potentials, create ion gradients across cell membranes, and shape action potentials
and other electrical signals
[9]
.
Action potentials depend completely on the ability of K
+
channels and Na
+

channels to open and close at the appropriate times. Maintaining ion selectivity
within these ion channels is key to the viability of cells and mutations within sodium
or potassium ion channels can cause hyperexcitability or hypoexcitability within
functioning neuronal cells and are strongly associated with neurological disorders

[8]
. The K
+
selective KcsA ion channel is a highly selective channel in which the
selectivity mechanism is not yet understood. The KcsA channel, found in bacterial
cells, is a pH gated ion channel that allows the passage of potassium ions at low pH.
This membrane ion channel is a tetrameric channel consisting of a cytoplasmic gate,
a water-filled central vestibule, and a selectivity filter.

7

Figure 1- Side view of two subunits of the KcsA channel. Closed gate conformation. Red and purple
balls represent ions in each of the stabilizing carbonyl sites (S1-S4). Normally, ions occupy every
other carbonyl site, while water molecules occupying the sites in between. This is a composite view
of two crystal structures, one with ions occupying S1 and S3, the other with ions in S2 and S4
[21]

KcsA has been crystallized under different conditions, including one with
high K+ (400 mM) and one with low K+ (3 mM). The low-K structure represents a
non-conducting state of the channel and the high-K structure is compatible with ion
conduction
[21]
. A third structure became available during the performance of this
work. This structure has the largest distance separating threonine 112 residues in
the KcsA subunits (32Å) and is called the open structure. Preliminary work on this
project was done using the high-K structure, and the mutagenesis simulations were
done on the open structure in order to approximate the conducting channel as
closely as possible.
Channel Structure
A: Selectivity Filter
B: Water Vestibule
C: Gate (closed conformation)

A
B
C
Figure 2- Top view of four subunits of the KcsA
channel. Ions (yellow) flow through the center
of the four subunits making their way across
the cell membrane
[22]

8
In the high-K structure, the four subunits come together to create a gate-like
barrier in the shape of a “V” (Figure 1) blocking the entrance of ions. At low pH, the
channel structure opens up, spreading out the lower legs of each subunit creating an
opening for ions to pass into the water vestibule. Once the channel moves into the
conducting state, ions make their way into the water vestibule by simple diffusion.
Ions randomly approach the selectivity filter where they may or may not be allowed
passage through this narrow opening. The K
+
selectivity ratio over Na
+
for the KcsA
channel is at least 150 and may be as high as 1,000
[13]
. Crystal structures indicate
that K
+
selective channels bind dehydrated potassium ions in the selectivity filter
[12]
. Selectivity filters of K
+
-selective bacterial and animal channels contain the
conserved amino acid sequence, TVGYG; threonine, valine, glycine, tyrosine, glycine.
Highly conserved sequences in any given protein indicate a highly selective and
precise function for that protein. The characteristic TVGYG sequence motif of the
potassium channel suggests that ion selectivity is closely linked to the selectivity
filter. Early mutational work done on the bacterial KcsA selectivity filter established
that mutations made to any of the five amino acids resulted in inaccurate ion
selectivity. A tyrosine to valine (Y75V) mutation in amino acid 75 caused a ten-fold
decrease in potassium selectivity over sodium
[14]
. The amino acid chain backbone
faces the inside of the selectivity filter while the side chains face out towards the
rest of the protein (Figure 3). Within the selectivity filter, electro-negative carbonyl
oxygen atoms aligned along the center of the filter (Figure 4), creating four stable
potassium-binding sites
[15]
.
9
Threonine 75 is the first amino acid the ion encounters as it enters the
selectivity filter from the vestibule and as such, is a good candidate for participation
in the mechanism that discriminates between Na
+
and K
+
.

Figure 3: Illustrates the four stable potassium positions and the amino acids that comprise each
binding site
[20]

Figure 4: (Left) K+ ion (magenta sphere) is shown bound in the stabilizing site of the selectivity filter
(green ribbon). The backbone carbonyl groups of the site are shown as spheres. (Right) View looking
down the pore (z-axis)
[16]
.

Another computational method, steered molecular dynamics, evaluates
energy interaction, potentials of mean force, and interactive properties of
biomolecules and their response to external force
[17]
. SMD simulations yield
important structural information about the structure-function relationship of the
ligand (ion)-channel system as well as the mechanism underlying ion selectivity.
Atoms can be manipulated to move in a certain direction by applying a force and
observing the resultant movement of the atom. The system is set up with an ion of
choice bound to a “dummy” atom by a spring that is assigned a spring constant (k).
The ion and dummy atom begin on the same coordinate point and as the simulation
10
pulls the dummy atom along the specified axis, the attached ion moves in
accordance to the tightness or looseness (k constant) of the spring.
Although the ion ultimately moves along the same axis as the pulling force,
the ion also has the freedom to interact with its surrounding protein structures and
encounter energy barriers. The pulled particle must be moving at a constant
velocity, set by the investigator, so as the ion encounters these energy barriers along
its path the system must increase the amount of pulling force applied to the particle
in order to maintain this velocity. Once the dummy particle makes it through the
protein the output will contain a list of coordinates and the forces applied at those
points indicating where the energy barriers are located.
A variant method to SMD is the step-wise pulling method. In step-wise
pulling the dummy particle is moved in a step-wise fashion while the ion follows on
the spring. The dummy particle is progressed by a distance of lambda ( at each
time step (Figure 5). Between each of the time steps the simulation is allowed a
relaxation period for the ion to equilibrate in each of the lambda positions. For the
time period the ion equilibrates at each lambda, ten positions were collected,
recorded, and plotted (Figure 6).
11

Figure 5: Ion pulled through an ion channel in a stepwise manner. X-axis represents time and y-axis
represents the lambda distance the dummy particle is moved.

Figure 6: Equilibration at each lambda for the ion. The ion is allowed to equilibrate at each lambda
for a certain amount of time (t). Instep-wise pulling, ten positions are collected during the
equilibration and plotted for each lambda step-wise pull. Sampling ten positions allows predictions
for where each ion spends the most time.

The integration of all the data can indicate the stable positions for
different ions. Determining where each ion is most stable can give a substantial
amount of information on the site of ion selectivity and a possible mechanism that
hasn’t been fully uncovered. With the use of Steered Molecular Dynamics (SMD),
acquiring theoretical data has offered probable theories for ion selectivity and
potential Na
+
ion barrier sites. In the steered simulation, single sodium and
potassium ions were, in two separate simulations, pulled through the selectivity
filter at constant velocity. In order for the ions to maintain constant velocity while
12
traveling through the channel, they must overcome energy barriers within the
channel. Each time the ion encounters an energy barrier the simulation must
intensify the force applied in order to overcome these barriers to maintain constant
velocity. Energy barriers in the simulation output are identified where the forces
increase along the pulling axis (z-axis). Using data from multiple simulations,
measurements from the center of mass of the channel to the location of the energy
barriers identified amino acids that may play a role in ion selectivity.

In step-wise pulling simulation of Na
+
and K
+
ions through the KcsA
Potassium channel we see some clear differences in stable locations and energy
minima. Energy minima represent locations of low external energy on the ion and
therefore high stability for an ion. As the ions lave the water vestibule and prior to
entering the selectivity filter, each ion experiences the first, and relatively large, free
energy difference at the threonine 75 position. Looking at structural data we see
that amino acids 75 to 79 constitute the selectivity filter of the KcsA channel made
up of the amino acid backbones. Prior to entering the S4 stable carbonyl site the
ions must pass by threonine 75 (Figure 7). In a comparing two step-wise pulling
systems, it is found that the Na
+
experiences much more stability and an energy
minima as it reaches threonine versus the K
+
ion which does not experience this
Figure 7: Ions are stabilized within each of the S1-
S4 cavities. Each stabilizing space is created by the
carbonyl backbone and by a hydroxyl side chain
(S4 only, provided by threonine 75).
13
same energy minima. A probable hypothesis, which will be tested, is that the
hydroxyl group on the threonine side chain, pointing down towards the water
vestibule, greatly stabilizes the positively monovalent sodium and rejecting its entry
into the selectivity filter.
NAMD enables us to generate multiple threonine 75 mutational
structures that can be implemented into step-wise pulling simulations. The three
mutant structures include: valine (r-side chain replaces the partially negative
hydroxyl group of threonine with a neutral methyl group), cysteine (r-side chain
contains no hydroxyl group but a thiol side group) and serine (r-group contains a
single hydroxyl side group)(Figure 8).

Threonine: Methyl-Hydroxyl Side Group

Valine: Dimethyl Side Group

Cysteine: Thiol Side Group
14

Serine: Hydroxyl side group
Figure 8: Amino acid structures for mutational analysis. Mutations are based on the different side
chain electronegativity.

The cysteine mutation (T75C) replaces the hydroxyl with an SH group and will
determine whether the hydroxyl side group is truly responsible for creating the
stabilizing effects of Na
+
over K
+
. The serine mutation (T75S) has a hydroxyl group
on its side chain as well as an accompanying methyl group. If Na
+
stability is due to
the threonine hydroxyl group, the valine mutation should show little change to the
energy minima while the cysteine mutation should causes a decreased stability at
the S4 site. Lastly, the valine mutation maintains the methyl group while replacing
the hydroxyl group with a second methyl group. In theory, this should create a less
stable Na
+
site due to the lack of an electronegative side group, causing Na
+
to either
bounce back out of the selectivity filter or continue through with less resistance.
Each of these mutations will provide evidence to either strengthen or weaken the
hypothesis that the threonine 75, hydroxyl group provides stability to Na
+
ion,
omitting its passage through the channel, and identifying a possible selectivity
mechanism.

15
Chapter 2: Methods
Computational methods, such as molecular dynamics simulations, are often
used to analyze the movement and interaction of atoms and molecules within
biological systems on a spatial or temporal scale that is not readily accessible to
direct measurement. Molecular dynamics simulations can be used to analyze
interatomic and intermolecular interactions and also to derive protein-ion energy
profiles and to predict stable protein conformations
[1][10]
. Simulation of large
molecules, however, requires much more computing power and one-way to achieve
such simulations is to utilize parallel computers. Parallel computing is a form of
computation in which many calculations are carried out simultaneously, operating
on the principle that large problems can often be divided into smaller ones, which
are then solved concurrently ("in parallel")
[2]
. NAMD is a molecular dynamics
simulation package that is designed to run efficiently on such parallel machines for
simulating large molecular systems
[3]
. NAMD is a simulation package written using
the CHARM++ parallel programming model that runs on the popular molecular
graphics program VMD (Visual Molecular Dynamics) for simulation setup and
trajectory analysis
[4] [5]
. CHARM++ is a package of force fields, written in the C++
computing language, which contains parameter values for atomic mass, van der
Waals radius, partial charge for individual atoms, equilibrium values of bond
lengths, bond angles, dihedral angles for pairs, triplets, and quadruplets of bonded
atoms, and values corresponding to the effective spring constant for each potential.
Force field refers to the form and parameters of mathematical functions used to
describe the potential energy of a system of interacting particles. "All-atom" force
16
fields provide parameters for every type of atom in a system, including hydrogens
(allowing for solvation of the protein). In an all atom simulation each atom is
simulated as a single particle with its own specified radius and net charge. Bonded
interactions are treated as "springs" (each with an assigned spring constant) with an
equilibrium distance equal to the calculated bond length. Force field calculates the
molecular system's potential energy (E) in a given conformation as a sum of
individual energy terms. The lowest energy region of the potential energy function
is the state the protein most occupies at thermal equilibrium. NAMD is comprised of
a common potential energy function that includes bonded and nonbonded bonds.
E = E
bonded
+ E
nonbonded

E
bonded
= E
stretch
+E
bend
+ E
torsional

E
nonbonded
= E
electrostatic
+ E
van der Waals

The most stable protein structures are in the lowest energy states and using
molecular dynamics simulations, protein conformations and protein-ion
interactions can be predicted based on the parameters set by the CHARM++ force
fields. The simulation ultimately identifies a protein structure with the lowest
overall potential energy, hence the term, energy minimization. Energy minimization
is the process in which the simulation program, NAMD, computes the equilibrium
configuration of the molecule under specific conditions. During energy
minimization, the molecule fluctuates under specific environmental conditions until
it reaches its lowest, most stable, energy state. This steady state molecular structure
17
can then be used as the protein structure in further molecular dynamics simulations
such as Steered Molecular Dynamics (SMD)
[11]

In order to run a molecular dynamics simulation, NAMD requires four file
types; PDB, PSF, parameter and a configuration file. The PDB file is the Protein Data
Bank file, which stores atomic coordinates and/or velocities for the system. This file
can either be created or found on the Protein Data Bank site online. A PSF, Protein
Structure File, stores structural information of the protein, such as various types of
bonding interactions. PSF files contain all of the molecule-specific information
needed to apply a particular force field to a molecular system. This structure file is
derived from the PDB crystal structure file using a topology file as the conversion
tool. Topology files contains all of the information needed to convert a list of
residue names into a complete PSF structure file and contains internal coordinates
that allow the automatic assignment of coordinates to hydrogens and other atoms
missing from a crystal structure PDB file. The parameter file encodes the specific
force fields needed for the system. One such force field that NAMD is able to use is
the CHARMM force field previously mentioned. A CHARMM force field parameter
file contains all of the numerical constants needed to evaluate forces and energies,
given a PSF structure file and the PDB atomic coordinates. The parameter file is
closely tied to the topology file that was used to generate the PSF file, and the two
are typically distributed together and given matching names. A NAMD configuration
file specifies what dynamics options and values that NAMD should use, such as the
number of time-steps to run, initial temperature, and how long the simulation
should continue in addition to many other parameters of the computer job
[1][7]
.
18
The investigator has control of the options and values in this file and thus
controls how the system will be simulated. The values within the configuration file
are often adjusted as an investigator advances through a simulation as a result of
the progression of the simulation and the output of likely “crash” reports (reports
displayed if the simulation is unable to continue and the system crashes). The fate
and progression of a simulation depends on the stability of a system and often based
on the values specified within the configuration file. The options and values
specified determine the exact behavior of NAMD, what features are active or
inactive. The NAMD configuration file is given to NAMD on the command line of the
simulation and specifies aspects of the simulation including addition of a lipid
bilayer, water molecules, ions in solution, temperature and pH etc. During a typical
NAMD simulation, atoms and bonds are allowed to interact and fluctuate for a set
amount of time under the given force fields. For some systems, like folded proteins,
a few nanoseconds would be enough to assess stability, volume fluctuations, etc.
For calculations needed for large scale MD simulations, hundreds of nanoseconds
are needed for sampling the conformational space.
The resultant trajectories of each particle after having experienced the
various intermolecular and intermolecular interactions can be visualized using VMD
(visual molecular dynamics). VMD is the visualization component of MDScope, a
set of tools for interactive problem solving in structural biology, which also includes
the parallel MD program, NAMD, and the MDCOMM software used to connect the
visualization and simulation programs
[6]
. MDComm defines a protocol for transfer
of static and dynamic molecular data between applications, provides a library for
19
use by applications to send and receive this data. Both VMD and NAMD use the
MDComm software, which links these two applications into a single environment for
interactive simulation of biomolecular assemblies. VMD is the platform on which
the NAMD simulation package runs and additionally the program that solvates and
incorporates ions into the loaded structure. VMD was primarily developed as a tool
for viewing and analyzing the results of molecular dynamics simulations, but it also
includes tools for working with volumetric data, sequence data, and arbitrary
graphics objects.
With these computational methods, the forces between particles and their
potential energies are defined by the molecular mechanics force fields that each
atom experiences within the system as a result of the presence and movement of all
other atoms in the system. Ultimately, multipart numerical calculations and system
parameters can uncover stability of complex systems and provide the tools to
calculate changes in the system that occur as a result of perturbations. This is a very
useful, inexpensive and accurate way, for example, to predict the possible outcomes
of introducing mutations into proteins, observing interactions between amino acids
and monitoring the pathways of ions through protein channels. In addition to
analyzing outcomes that result from changes in protein structure, molecular
dynamics simulations can specify the environment of a protein. The environment
can be specified to either emulate a protein’s natural environment or to simulate a
hypothetical environment in order to predict results of a changed environment. In
order to successfully complete a molecular dynamics simulation, an investigator
needs a protein structure file, the protein data bank file, a configuration file and the
20
force field parameter file. Protein crystal structures found within the Protein Data
Bank are loaded into the desired molecular dynamics program; parameters are set
by the investigator to accurately portray the desired environment then the
simulation begins. By applying an external force to either the protein or any atom
within the constructed system, there will be a resultant “force field” due to various
molecular interactions. Given these force fields, investigators are able to calculate
energy profiles at different time steps and different locations within the system as
the simulation progresses for a fixed amount of time. Using precise measurements
from a given center of mass versus the force profile, energy barriers may be
identified at specific angstrom distances from the center of mass and further
hypothesis may be formed concerning the movement of ions, proteins and atoms.
The technology to accurately emulate protein systems and simulate
molecular interactions is a huge step in the direction of theoretical science. With the
use of molecular dynamics, plausible theories are formed based on the potential
energy profiles as the system experiences different forces. Computer simulation is a
powerful tool for either solving or clarifying scientific problems as numerical
experiments can be performed for new systems without synthesizing or purchasing
materials. In addition to being cost effective and time efficient, molecular dynamics
allows users to analyze data on a larger temporal scale. MD allows for the
visualization of atomic movement and mechanisms on a much larger viewing scale
and a much slower time scale.

21
Chapter 3: Results
The preliminary work done on the high-K
+
KcsA structure was
provided by collaborators; Darko Stefanovski and Van Ngo. The closed structure
was the first structure that was available for structural analysis. With the use of the
Step-Wise pulling protocol, both a Na
+
and K
+
ion were pulled through the channel,
one lambda step at a time. At each lambda value the corresponding z-value of the
ion was recorded and as a result the stable position of the ions were determined.
(Å) 0 1 2 3 4 5 6 7 8 9 10 11 12
Stable
Positions
of K
+
(Å)
2.1
4.4
2
4.5
4.6 4.7
4.8
7
5.2
7.1
7.1 7.2 7.2 7.3 7.4 7.5 7.6
Stable
Positions
of Na
+
(Å)
0.4
2.3
4.9
2.4
5
5.2 5.2 5.3 5.5 5.5 5.6
5.6
5.8
5.7
7
6.7
5.8
6.8
8.8
6.8
9
(Å) 13 14 15 16 17 18 19 20 21 22 23 24
Stable
Positions
of K
+
(Å)
7.7 7.8
7.8
10.3
7.9
7.9
10.3
13.9
8
10.3
10.7
15.7
10.6
14.5
16
17
10.6
15.8
17.7
17.5
22.7
18
22.7
18.1
23
24.2

Stable
Positions
of Na
+
(Å)
7
8.8
9 9 9.3 9.1 9.2 9.3 9.4
9.6
11.8
21.9
9.5
10.5
22.4
9.5
11.3
15.7
22.5
9.7
15.8
22.1
24

Figure 9: Step-wise pulling protocol data

The table above has the lambda values highlighted in yellow with the
corresponding z-values for each ion highlighted in green. Within a step-wise pulling
protocol, moving the dummy particle by a distance of 1  (each pull moves a
the ion is in its z position. Looking at the K
+
ion data, when the dummy particle is at
22
a lambda of 6 to 15, the K
+
ion maintains its position at approximately z = 7.5. The z-
position of z = 7-8 in the closed structure of the KcsA channel corresponds to the
threonine 75. The Na
+
ion seems to find a stable position when lambda reaches a
value of 2 to 11, which corresponds to a z position of approximately 5.5. Any z-
position less than 7 is in the water vestibule and before the selectivity filter.

Figure 10: Ionic Frequency within each z position

The frequency histogram above illustrates the frequency of the z-
coordinate for each ion recorded as a step wise pulling protocol was conducted for
both the Na
+
and K
+
ions. The frequency is the number of times each ion visits the
corresponding z positions. A high frequency for an ion correlates with a stable
binding site for the ion. The Na
+
ions remain outside of the S4 binding site while the
K
+
ions are found with their highest frequency within the S4 binding site. In
between the junctions of S4 and S3, and S3 and S2 there is a high frequency of Na
+

ions. Within the S3 and S2 binding sites there is a high frequency of K
+
ions.
23

Figure 11: F and F for K
+
and Na
+

The graph on the left displays the change in free energy for each ion
relative to the reference position (z = 0) as  in incremented by  = 1Å. The zero
position of lambda represents the center of mass for the system where we find the
hydrated ion in the water vestibule. As the step wise pulling progresses, the dummy
particle is pulled by 1  step each 0.5ns and the ion follows with a certain amount of
associated free energy needed to move the ion. The K
+
ion has a spontaneous
movement from the zero position to its lowest energy position at a lambda of 7. The
Na
+
ion also spontaneously moves from the zero position to its lowest energy
position at a lambda of 5. The graph on the right is the difference in F for each ion
at each lambda value (F). The highest F is found at a lambda of 11; at which
point the differences in F between the Na
+
and K
+
ions begin to decrease. In graph
A we see the amount of free energy necessary to move each ion from one lambda to
Graph A: F necessary to move ion to
each lambda position.

Graph B: F based on graph A.

24
the next. Graph B then plots the differences between the free energies between the
Na
+
and K
+
ions. There is a positive slope from a = 5-11 because while the
potassium ion moves into the S4 binding site, the sodium ion is trapped outside of
the selectivity filter unable to enter S4 until the pulling force can overcome the
stable binding of the ion to the position at z = 5-6. As the  value increases with
each pull, the change in free energy also increases since the z-value of the ion isn’t
changing. Only when we are at a  of 11, the amount of force experienced by the Na+
ion reaches a point at which it overpowers the forces keeping the ion within the
stable position. F reaches a peak of 3.7kcal/mol because at this  the K
+
is still
firmly bound in the S4 site while the Na
+
ion is finally moving out of its stable
position outside of the selectivity filter to within the filter. The Na
+
ion remains
outside of the selectivity filter because of a possible ionic barrier found around 6 to
7 angstroms from the center of mass of the system. This position corresponds to the
threonine 75 amino acid. With these preliminary results, a new project arose
focusing on mutational analysis of the threonine 75 amino acid.
It was discovered with the preliminary data that the KcsA ion channel
selectively excludes Na
+
ions from entering into the selectivity filter past threonine
75. By implementing mutations to the threonine 75 residue, the side groups that
occupy the opening of the selectivity filter as well as provide stabilizing oxygen
atoms to the S4 binding site can be changed. Implementing the three mutations and
doing a step-wise pulling with K
+
ions through the wild type and mutant channels,
generated data showing either the allowance or restriction of K
+
into the different
mutant structure. Figure 12 illustrates snapshots of the K
+
ion within the high-K
+
25
KcsA channel at a lambda of 7 for each mutant channel. In panel 1, the wild type
KcsA selectivity filter allows for the entrance of the K
+
ion while in panel 2, the
Serine mutant channel denies the K
+
entrance into the filter. In panel 3, the valine
mutant channel also excludes the K
+
ion but in contrast, panel 4, the cysteine mutant
allows the entry of the K
+
ion.

Figure 12: K
+
ion within KcsA channel at  of 7. (Panel 1- Wild type. Panel 2 - T75S. Panel 3 - T75V.
Panel 4- T75C)

In the wild type channel the hydroxide side group points up towards
the K
+
ion while the methyl group is pointing towards the water vestibule. In the
serine mutant channel, the hydroxyl group is pointing downwards towards the
water vestibule while in the valine mutant there are no hydroxyl side groups but
rather two methyl groups. However in the cysteine mutant channel, where there
are no hydroxyl or methyl side groups, the thiol side chain didn’t restrict K
+
access
into the S4 site.

26

Figure 13: Na
+
ion within KcsA channel at  of 7(Panel 1- Wild type. Panel 2 - T75S. Panel 3 - T75V.
Panel 4- T75C)

In figure 13, the wild type channel restricts Na
+
entrance into the
selectivity filter but in the cysteine mutant channel the Na
+
ion is able to enter into
the selectivity filter. Free energy profiles for the K
+
ion through each of the mutant
and wild type channels (Figure 14) shows a clear correlation between the wild type
and cysteine mutant channel while the valine and serine channels experience a
higher F around a  of 7 which is the location of the entrance to the selectivity
filter. Free energy profiles have been generated for each of the mutant channels and
are illustrated in figures 14, 15, and 16. For figures 14 to 17 x-axis represents  and
y-axis represents F.

27

Figure 14: Free Energy Profile- K
+
through the WT and all mutant channels.

Figure 15: Free energy profile -Na
+
and K
+
through the WT and CYS mutant channel.

Figure 16: Free Energy Profile- Na
+
and K
+
through SER mutant channel.
-10
0
10
20
30
40
50
60
0 5 10 15 20 25 30
SER
VAL
THR
CYS
-10
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30
K-THR
Na-THR
K-CYS
Na-CYS
-10
0
10
20
30
40
50
60
70
80
90
0 5 10 15 20 25 30
Na-SER
K-SER
28

Figure 17: Free Energy Profile- Na
+
and K
+
through VAL mutant channel.

-10
0
10
20
30
40
50
60
70
0 10 20 30
Na-Val
K-Val
29
Chapter 4: Discussion
The results of the molecular dynamics simulation on the wild type
KcsA channel confirmed that K
+
ions are allowed into the selectivity filter past
threonine 75 while Na
+
ions are restricted from passing past the threonine 75
residue. It is this residue then that must have bearing on the selection process of the
KcsA potassium ion channel. We did mutational analysis using molecular dynamics
by creating mutant sequences in the threonine 75 position. The mutations were
strategically chosen in order to determine importance of each of the side groups.
The results from the simulation showed that the valine and serine mutations didn’t
allow either the K
+
or Na
+
ions to enter into the selectivity filter. However, the
cysteine mutation allowed access to both the K
+
and Na
+
ions into the selectivity
filter. These results suggest that there is a possible loss of selectivity and increased
conductance in the cysteine mutant while there is a loss of selectivity and a decrease
in conductance in both the serine and valine mutant channels. The cysteine mutant
allows both ions to enter the selectivity filter and therefore decreases the selectivity
ratio of K
+
: Na
+
ions. Free energy profiles (Figure 14) of the wild type and cysteine
mutant channel show that for both potassium and sodium ions, the amount of force
necessary to pull the ion from one position to the next is very similar for the two
ions. In the serine mutant (Figure 16) there is little difference between the Na
+
and
K
+
free energy profiles and free energy barriers are both higher than the wild type
ion channel for entry into the selectivity filter. There is similar resistance against
both ions when the threonine 75 is replaced with a serine residue. Since the free
energy profiles for both the K
+
and Na
+
ions in the serine mutant have no significant
30
difference, it means that there is a loss of selectivity for the ions. In this mutant, the
methyl group is lost but the hydroxyl group is maintained, and this suggests that the
methyl group is highly important to the process of ion selectivity. The methyl group
can either be solely responsible for the selection process or must be present in
conjunction with the hydroxyl group in order to establish the high selectivity of the
KcsA channel. The valine mutant (Figure 17) is very similar to the serine mutant in
that it blocks the entrance of both Na
+
and K
+
. The valine mutant has lost the
hydroxyl group and has two methyl groups. With the loss of the hydroxyl group, a
larger force is needed to pull both ions into the selectivity filter, which might
indicate that the mutant would have a reduced conductance relative to the wild
type. In addition, there seems to be no difference in the amount of force necessary
to pull each ion into the selectivity filter, which means that not only did the channel
have a decrease in conductance, but it displays no selectivity for one ion over
another. In the cysteine mutant (Figure15), in which there is a loss of both the
hydroxyl and methyl groups, we find that both sodium and potassium ions enter
into the selectivity filter with almost the same free energy profile as the wild type
with a K
+
ion. This indicates that without either the methyl or hydroxyl side groups
normally present in the threonine, there is a complete loss of selectivity between the
two ions. Since both ions are now capable of passing beyond the threonine 75
position and entering the selectivity filter, it can be predicted that the conductance
of the T75C mutant channel will slightly increase. Without the exclusion of the Na
+

ion, the channel is free to move either ion without risk of the Na
+
blocking the
channel.
31
The data indicate that both the methyl and hydroxyl side groups of the
threonine amino acid are necessary to create a highly selective and conductive
potassium ion channel. Without either of the side groups there are large shifts in
the amount of force necessary to move an ion through the channel. The serine and
valine mutant channels which have each lost either the methyl or the hydroxyl
group, but not both, lose the ability to allow either Na
+
or K
+
ions into the selectivity
filter unless pulled with a higher than normal force. Whereas the cysteine mutant,
which has lost both the methyl and hydroxyl side groups, allows admittance to both
the Na
+
and K
+
ions into the selectivity filter with the same force as a non-mutated
channel with K
+
. Further experiments and mutational analysis of KcsA DNA will
help test these hypotheses as well as provide more information about the
conductance and selectivity of ions. KcsA plasmid DNA will be used in a
Quickchange procedure to make each of the 3 mutations. The DNA will then be
inserted into E. coli cells in order to allow the mutant membrane channel to be
synthesized. Once the cells grow to a high enough concentration, the membrane
channels will be extracted and placed into a lipid bilayer for patch clamp
experiments and we will measure channel conductance and reversal potentials. The
relative amount of current carried by each ion (i.e., ion selectivity) can be estimated
from the intercept of the current-voltage line with the voltage x-axis while the
conductance can be determined by the slope of the current-voltage line. In regards
to selectivity of the mutants compared to the wild type channel, the value of the
intersection of the current-voltage plots of the mutants with the x-axis compared to
32
that of a wild type channel can tell whether or not selectivity has maintained or
shifted towards the Na
+
ion.

33
References

1. Phillips, James C., et al. "Scalable molecular dynamics with NAMD." Journal of
computational chemistry 26.16 (2005): 1781-1802.

2. Phillips, James C., et al. "NAMD: Biomolecular simulation on thousands of
processors." Supercomputing, ACM/IEEE 2002 Conference. IEEE, 2002.

3. Kalé, Laxmikant, et al. "NAMD2: Greater scalability for parallel molecular
dynamics." Journal of Computational Physics 151.1 (1999): 283-312.

4. Phillips, James C., et al. "Chemistry, 26: 1781-1802, 2005. Humphrey, W., Dalke, A.
and Schulten, K.," VMD-Visual Molecular Dynamics", J. Molec. Graphics, 1996, vol. 14,
pp. 33-38. &Stephan Frickenhaus, AWI, 2007, som_pak-3.1 modified for dihedral
periodicity, vmd-scripts trajsort/mkpic in tcl." Chemistry 26 (2005): 1781-1802.

5. Revealed, Voltage Clamp, et al. "Voltage-Gated Ion Channels Review and Electrical
Excitability." Neuron 20 (1998): 371-380.

6. Isralewitz, Barry, et al. "Steered molecular dynamics investigations of protein
function." Journal of Molecular Graphics and Modelling 19.1 (2001): 13-25.

7. Phillips, J., et al. "NAMD TUTORIAL." Windows Version. University of Illinois. NIH
Resource for Macromolecular Modelling and Bioinformatics Beckman Institute
(2007).

8. Nelson, Mark T., et al. "NAMD: a parallel, object-oriented molecular dynamics
program." International Journal of High Performance Computing Applications 10.4
(1996): 251-268.

9. Humphrey, William, Andrew Dalke, and Klaus Schulten. "VMD: visual molecular
dynamics." Journal of molecular graphics 14.1 (1996): 33-38.

10. Stone, John E., Justin Gullingsrud, and Klaus Schulten. "A system for interactive
molecular dynamics simulation." Proceedings of the 2001 symposium on Interactive
3D graphics. ACM, 2001.

11. Rush, Anthony M., et al. "A single sodium channel mutation produces hyper-or
hypoexcitability in different types of neurons." Proceedings of the National Academy
of Sciences 103.21 (2006): 8245-8250.

12. Latorre, Ramon, and Christopher Miller. "Conduction and selectivity in
potassium channels." Journal of Membrane Biology 71.1 (1983): 11-30.

13. Williamson, I. M., et al. "The potassium channel KcsA and its interaction with the
lipid bilayer." Cellular and molecular life sciences 60.8 (2003): 1581-1590.
34

14. Shrivastava, Indira H., et al. "K
+
versus Na
+
Ions in a K Channel Selectivity Filter:
A Simulation Study." Biophysical journal 83.2 (2002): 633-645.

15. Nimigean, Crina M., and Toby W. Allen. "Origins of ion selectivity in potassium
channels from the perspective of channel block." The Journal of general physiology
137.5 (2011): 405-413.

16. Dixit, Purushottam D., and Dilip Asthagiri. "Thermodynamics of ion selectivity in
the KcsA K+ channel." The Journal of general physiology 137.5 (2011): 427-433.

17. Izrailev, Sergei, et al. "Steered molecular dynamics." Computational molecular
dynamics: challenges, methods, ideas 4 (1998): 39-65.

18: Essen, Lars-Oliver, and Ulrich Koert. "Ion-channel engineering." Annual Reports
Section" C"(Physical Chemistry) 104 (2008): 165-188.

19. Uysal, Serdar, et al. "Crystal structure of full-length KcsA in its closed
conformation." Proceedings of the National Academy of Sciences 106.16 (2009):
6644-6649.

20. Gullingsrud, Justin. "KcsA Potassium Channel." KcsA Potassium Channel.
Theoretical and Computational Biophysics Group, 27 Jan. 2006. Web. 17 June 2013.

21. Zhou, Y., et al., Chemistry of ion coordination and hydration revealed by a K
channel-Fab complex at 2.0 A resolution. Nature, 2001. 414: p. 43-48.

Abstract (if available)

Abstract The mechanism of ion selectivity in K⁺-selective ion channels has not yet been determined. In previous studies it was found that K⁺ ions were allowed entrance into the channels' selectivity filter while Na⁺ ions were excluded from the selectivity filter at threonine 75. To determine ion selectivity in the bacterial K⁺-selective channel KcsA, mutational analysis was conducted on threonine 75 using molecular dynamics simulations. A step-wise pulling protocol was used to pull both K⁺ and Na⁺ ions through the wild type and mutant channels in order to estimate any changes of conductance or selectivity by the use of force profiles. Mutations were chosen specifically to investigate the importance of each of the side groups of threonine to selectivity. The three mutations studied changed threonine 75 to serine, valine or cysteine. In comparison to the wild type, both the valine and serine mutants excluded K⁺ and Na⁺ from entering into the selectivity filter while the cysteine mutant allowed both the K⁺ and Na⁺ ion to enter the selectivity filter. Interpretation of this data indicated that valine and serine mutants both decrease the conductance as well as the selectivity ratio of the KcsA channel. However, the cysteine mutant, which lost both methyl and hydroxyl groups, allows entry of both K⁺ and Na⁺ ions into the selectivity filter and may cause an increase in conductance as well as a decrease in the selectivity ratio. Results from molecular dynamics simulations help investigators to look at particle movement on a short temporal scale as well as helps make future experimental design more efficient. In future work, mutations will be inserted into KcsA DNA using PCR and conductance measurements will be taken using the patch clamp technique. These experiments will help determine mutational effects on selectivity and conductance of the actual ion channel.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Elucidating the mechanism of ion permeation through a potassium-selective ion channel

PDF

Structural-functional study of the paracellular ion pore

Asset Metadata

Creator Tabbakhian, Melia A. (author)

Core Title KcsA: mutational analysis of ion selectivity with molecular dynamics

School Keck School of Medicine

Degree Master of Science

Degree Program Physiology and Biophysics

Publication Date 08/05/2013

Defense Date 06/19/2013

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag ion channel,ions,KcsA,molecular dynamics,OAI-PMH Harvest,potassium,sodium,step-wise pulling

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Farley, Robert A. (committee chair), Chow, Robert H. (committee member), Kalra, Vijay K. (committee member), Kaslow, Harvey R. (committee member)

Creator Email mtstanford1@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-316791

Unique identifier UC11293859

Identifier etd-Tabbakhian-1965.pdf (filename),usctheses-c3-316791 (legacy record id)

Legacy Identifier etd-Tabbakhian-1965.pdf

Dmrecord 316791

Document Type Thesis

Format application/pdf (imt)

Rights Tabbakhian, Melia A.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA