Surface-enhanced 2D IR spectroscopy to probe

voltage- and pH – sensitive peptides and ion channels

in a single bilayer

Ion channels play vital roles in biological processes. These channels have been long

recognized for their role in controlling electrically excitable tissues such as muscles and nerves.

However, there is still debate as to the mechanism with which the ions permeate the channel,

specifically if water passes through the channel with the ions or if collisions between the ions are

necessary for transduction. Previous work in the Zanni group has utilized two-dimensional infrared

(2D IR) spectroscopy to determine the method of ion permeation in a model ion channel, KcsA.

2D IR spectroscopy is an excellent technique for this study because vibrational frequencies and

lineshapes are excellent probes of molecular structure and electrostatic environment. Since the

publication of the original report, others have questioned our conclusions and have found different

ways to average the modeled spectra to match our experimental spectra. Therefore, we need to do

more in depth studies of KcsA. We also are working towards a voltage-dependent experiment that

will allow us to study the impact of an applied potential on the protein of interest. In order to do

these potential experiments, I have been working to develop a surface-sensitive 2D IR method

using localized surface plasmon enhancement in order to work with proteins embedded in a single

bilayer. This work has yielded promising results in a model peptide and we are working to

implement this same technique into larger systems.

Research Plan:

Research Objectives

The objective of my research is to determine the mechanism of ion permeation through ion

channels by using two-dimensional infrared (2D IR) spectroscopy, a technique known for its

inherent time and structural resolution. We begin by following up on previous work in the Zanni

group, studying the potassium (K+) channel from Streptomyces lividans (KcsA) in micelles using 13C/18O isotope labels.1,2 However, since we want to probe the effect of an external stimulus on

ion occupancy and general structure, we are moving this work into a single bilayer. This is

necessary in order to facilitate a potential drop across the membrane-bound protein of interest.

With stacked bilayers, the potential required to trigger protein movement increases additively with

each successive layer, leading to unnecessarily dangerous applied potential. With a single bilayer,

the applied potential can be as low as 50 mV.3 However, studying a protein in a single bilayer

yields a very low signal due to the inherently small amount of material present. This can be

remedied by the recent push towards surface-enhanced spectroscopies. Using thin, rough gold as

a source of plasmon enhancement, combined with the inherent non-linear scaling of 2D IR, we

have been able to take spectra of proteins in a bilayer.4,5

Determine the ion occupancy of KcsA – Work by the Zanni group that determined the ion

occupancy of KcsA to predict the mechanism of permeation has recently come under scrutiny.6,7

In short, it was put forth that our data was inconclusive as to the presence or lack of water in the

selectivity filter (a short, highly conserved sequence making up the channel). We need to collect

more spectra of KcsA with different isotope labeling schemes in the filter to better determine ion

occupancy. Work done by the Skinner Group at the University of Chicago has indicated that in

labelling Val76 and Gly77 individually, spectra that are more narrow and red-shifted would be

indicative of the water-free (hard-knock) mechanism and broader, blue-shifted spectra will

indicate water in the channel during transduction (soft-knock mechanism).

Surface-sensitive techniques – In order to be able to observe a spectrum of a single bilayer, while

being able to place a voltage across it, we use a thin, conductive layer of rough gold to function

both as an electrode and a source of plasmon enhancement. Furthermore, by taking advantage of

the ease of sulfur-gold chemistry, we can easily tether a monolayer of the sample to the gold. The

signal enhancement from the gold has been shown to have enhancements of up to 6000x. A further

3x enhancement can be gained by using a reflection geometry for a 3-nm thick thermally

evaporated film of gold.8 However, we have found that the 3-nm film is not macroscopically

conductive. To overcome this, I am working to find a thickness of gold that is conductive while

yielding suitable enhancements in the mid-IR, without lineshape distortions. This could also be

accomplished by adding a non-plasmonic conductive layer under the gold (e.g. indium tin oxide

(ITO) with a small amount of rough, plasmonic gold coated on top.9 We will investigate this

technique by studying the model peptaibol, alamethicin, comparing its static pH and voltage

dependences, as it forms pores in the tethered membrane. We will translate these static experiments

into transient, voltage-jump experiments to gain insight into the dynamics of voltage-gating.

Background and Significance

Ion Channels and Pores – Nerves and muscles have been known to be electrically excitable for

over 200 years. However, the movement of ions across the membrane to trigger this excitation was

not discovered until the early twentieth century by Julius Bernstein and built upon by many, such

as Nobel Prize winners Hodgkin, Huxley, and MacKinnon.10–13 Ion channels are so central to cell

physiology and communication, that study of these channels has expanded from monitoring the

ionic flux with voltage- and patch-

clamp experiments to determining

the crystallographic structure of

the membrane-bound proteins

themselves.

A notable feature of ion

channels, which generally control

the flow of Na+, K+, Ca2+, and Cl-

ions is that they have extremely

high ion specificity while

maintaining an ion flux of millions

of ions per second, nearly

matching the diffusion limit.10 The

model potassium channel, KcsA,

has been extensively used in order

to study the mechanism of ion

permeation.14–17 Selection for one

ion over another occurs in the

selectivity filter, a highly

conserved sequence (TVGYG) in

K+ channels composed of four

binding sites bound by five

backbone carbonyls on each of the

monomers forming the tetrameric

channel.11,17 Currently, there are

two competing models for ion

transduction: the hard-knock and

soft-knock (or knock-on) models

as shown in Figure 1. In the hard-

knock model, the potassium ions

are completely desolvated as they

enter the selectivity filter. As a

new K+ ion enters, coulombic forces repel the neighboring K+ ion, causing them to move through

the channel.6,18 In the soft-knock model, the selectivity filter is filled with alternating K+ ions and

water. The new K+ ion enters with an accompanying water, pushing another K+ ion and water out

into the extracellular side of the channel.1,19–23 Experiments, such as x-ray crystallography, single-

channel patch-clamp experiments, and computational modeling, have generally supported the soft-

knock mechanism, though these techniques were not able to give definitive evidence.22–24

However, in recent years, several theoretical groups have proposed the hard-knock model, which

they support by its ability to explain the high ion throughput and selectivity of the channel as the

energy penalty of desolvation greatly favors K+ over Na+.6

Previous work by the Zanni group resulted in a significant advance in the field, yielding

data that could be matched to results from molecular dynamics (MD) simulations. In 2016,

Figure 1. KcsA (two of four monomers shown) with the two proposed

methods of ion permeation A) the soft-knock and B) the hard-knock

with K+ ions in yellow and water molecules in blue. Backbone

carbonyls are shown in the selectivity filter.

Kratochvil and colleagues took 2D IR spectra of the KcsA channel in micelles.1 In order to observe

the environment of the selectivity filter, 13C/18O isotope labels at specific residues were used. The

additional mass of the labels red-shifts the amide I peak corresponding to the labelled amino acid

by approximately 60 cm-1, far enough to distinguish it from the rest of the protein. This method

provides a vibrational probe of the secondary structure and electrostatic environment of a single

residue.25–28 The isotope labels are sensitive to the electrostatics of their environment because

interactions with the partial negative charge on the 18O will effectively increase its mass, causing

a red-shifted peak in the spectrum. In this case, the inward-facing carbonyls that make up the ion

binding sites are sensitive to the electrostatics of the ions and molecules in the selectivity filter.

Placing isotope labels in three locations (Val76, Gly 77, and Gly79) allowed Kratochvil et al. to

observe the ion positions in the channel. In order to see the isotope labels, which were otherwise

masked due to the size of the protein (41 kDa), a non-isotope labeled spectrum was subtracted

from the isotope labelled spectrum. Data from the isotope-labelled experiments of KcsA in

micelles showed good agreement with spectra derived from MD simulation trajectories of the soft-

knock model. A state which included the carbonyl of Val76 flipped out of the selectivity filter was

included to create better agreement with experiment. This state has also been observed in NMR

structural studies of the channel as well as other MD simulations, though not through x-ray

crystallography.29–31

Papers published in the last year have called these results into question.6,7 One showed that

it was possible to generate a spectrum similar to experiment by using a weighted average of states

from the hard-knock model. They argued that the hard-knock model also has a better explanation

to the energetics of ion selectivity. They calculated the difference in energy required to desolvate

the K+ ion is much less than the Na+ ion. This difference explains the rapid K+ ion flow while Na+

ions are blocked. They also argue against the inclusion of the Val76 flipped state as previous work

has indicated that it corresponds to a non-conductive filter. However, in order to fit the hard-knock

model to our data, they have a configuration with four successive K+ ions in the channel account

for 25 percent of the contributing states, which would involve large repulsive forces in the channel.

We take this result to indicate that more experimental data is needed in order to constrain the

weights of the individual occupancies and come to a general conclusion. This may include

individual carbonyl isotope labels as detailed in future work below. Finally, some of these models

take into account the presence of a 50-mV potential drop across the bilayer, which was not used

in the previous experiments.18

Surface Enhanced 2D IR spectroscopy – Previous work to study ion channels have relied on

structural information from x-ray crystallography and NMR. X-ray crystallography has atomic

resolution, but is inherently static and the crystallization process can impact structure and ion

occupancy.11,15,29,32 Furthermore, it is difficult to distinguish between internal waters and ions.33

NMR gives both structurally- and time-resolved information but is limited as to the size of

protein.34 The dynamics of ion channels systems have been studied with time resolved current data

from patch-clamp experiments. Patch clamp experiments have good time resolution, but lack

overall structural information. In order to get residue-specific structural information from patch-

clamp experiments, time-intensive point-mutation must be done.32,35 2D IR has the ability to give

residue-specific structural information with inherent femtosecond time resolution and is therefore

an excellent method to study these systems without compromising structural or dynamical

information.4,25,36–38

Briefly, 2D IR measures the third-order response function of the vibrational transitions of

the sample after three light-matter interactions (see the SI for a more details). The vibrations of the

molecule are sensitive to which atoms are vibrating and the chemical environment.4 In proteins, a

commonly probed vibration is the amide I mode, largely composed of the backbone carbonyl

stretch, which absorbs between 1600 and 1680 cm-1.39,40 In our measurements, we observe the

frequencies in this region to determine the secondary structures present in the protein and how they

change. The main secondary structures, α-helix, β-sheet, and random coil, have distinct center

frequencies (1650 cm-1, 1620 cm-1, and 1645 cm-1 respectively) and lineshapes (narrower defined

structures and broad for random coil).4,39 Lineshapes also give insights into inhomogeneous and

homogenous broadening which are indicative of chemical environment. 2D IR is helpful in

distinguishing between these structures because the signal scales with μ4, rather than μ2 in 1D IR.

This increases signal and narrows the lineshape of some peaks while suppressing noise, making it

easier to resolve different secondary structures.4,25

In order to continue our work with membrane-embedded KcsA, we want to work in a single

bilayer. Doing so will allow us to apply external stimuli due to the voltage-divider-like nature of

the membrane as well as simulate a more biomimetic set up.3 2D IR spectroscopy is an odd-ordered

experiment, and is therefore not surface-specific. However, it is possible to generate surface

sensitivity, which is necessary to study the small number of molecule in a single bilayer.41 This is

commonly done by utilizing plasmon enhancement, similar to surface-enhanced Raman

spectroscopy (SERS) and surface-enhanced infrared spectroscopy (SEIRAS). There are several

methods of plasmon enhancement, including nanoarrays of various geometries to create specific

wavelengths of plasmonic resonance in a region of interest.42–46 Our method uses a thin, rough

gold film with small islands of gold on the surface. Between the islands are ‘hot spots’ which are

created when light is highly confined in the gaps, resulting in a higher concentration of the electric

field.41 The peak plasmon frequency is in the visible range, but extends into the mid-IR (vide infra).

For our experiments, this non-resonant enhancement leads to smaller enhancements than a

resonant enhancement method like SERS, but is still on the order of 1000x.8 Thicker gold red-

shifts the plasmon frequency, thus modulating the enhancement. However, thicker gold leads to

lineshape distortions, so a

compromise must be struck

between enhancement and spectral

quality.9

Recent work in the Zanni

and Fayer groups have further

manipulated the spectrometer

geometry to yield further surface

enhancements.8,47 In a BOXCARs

geometry, which is fully non-

collinear, the intensity of the local

oscillator (LO), which is overlaid

with the signal electric field to

produce the heterodyned signal, can

be independently modulated in

order to reduce the background.4

Due to the technical challenges of

collecting purely absorptive spectra

in this geometry, a pump-probe

geometry, which is partially

collinear and self-heterodyned (i.e.

the probe pulse acts as the LO), is popularly employed. However, due to the self-heterodyned

nature of the pump-probe geometry, the LO intensity is tied to the probe intensity and directly

influences the signal strength. To combat this, we took advantage of the fundamental qualities of

different polarizations of light as described in the Fresnel equations. For p-polarized light, there is

Figure 2. Reflectivity of s- and p- polarized light at a CaF2-air

interface, where Brewster’s angle is where the p-polarized light has

zero reflectivity.

an angle (Brewster’s angle) at which when it hits an interface, no light is reflected as shown in

Figure 2. At near-Brewster’s angle reflection geometries, therefore, it is possible to excite the

sample with the probe beam, but attenuate the LO strength such that the background is suppressed,

effectively enhancing the signal in the reflected direction. This was first demonstrated by the Fayer

group, and later applied to biological samples in strongly absorbing solvents in my first year. This

method itself gives a 4x enhancement that can be combined with localized surface plasmons for

enhancements on the scale of 104.8

Research Progress

I have developed a method for surface-enhanced 2D IR spectroscopy using a nanometers-thin,

thermally-evaporated layer of rough gold on a calcium fluoride window to enhance the signal from

a lipid bilayer. I have also demonstrated the ability to not only see signals from a model peptide,

but also changes in the peptide caused by changes in its environment.

A single monolayer of 1,2-dipalmitoyl-sn-glycero-3-phosphothioethanol (DPTE) was

created by incubating a solution of the DPTE in ethanol with a 2 nm thick gold-coated window in

order to form a gold-sulfur bond, tethering the head group of the lipid to the window.48 Vesicles

of 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (POPC) were allowed to fuse to the monolayer

of DPTE. The window was washed thoroughly to leave a single bilayer behind. Spectra of both

the DPTE and the mixed-lipid bilayer show the success of this technique seen in Figure 3. The

peaks arise from the ester carbonyls which link the head group to the tails.49 The presence of

multiple peaks is likely due to the difference in the two tails of POPC and interactions with the

gold. Furthermore, it is important to note that there are no substantial signals arising in the amide

I region, which is vital in being able to take peptide spectra without background correction.

To determine if we could observe a peptide embedded in the membrane, alamethicin (alm),

a 20-residue model peptaibol, was chosen to test our experimental methods. Alm associates with

the membrane, and has both α- and 310-helical character.50,51al A change of voltage or pH causes

alm to form bundles and insert itself into a membrane, initiating its antimicrobial activity.48,50,52–55

The alm was deposited onto the membrane at a 10:1 lipid:peptide ratio. The sample was dried then

rehydrated with D2O at a neutral or basic pH. 2D IR spectra of this sample were then taken

demonstrating the sensitivity of our methods as seen in figure 4. These spectra are consistent with

previous linear IR and SFG spectra taken of alm. We attribute the 1670 cm-1 peak to the α-helical

half of the peptide. The 1635 cm-1 peak arises from the 310-helix, consistent with both SFG spectra

of alm and 2D IR spectra of 310-helices.52,56

Figure 3. Representative 2D IR spectra of the ester carbonyl region for A) A monolayer of DPTE and B) a

single bilayer of DPTE and fused POPC vesicles both tethered to 2-nm of thermally evaporated gold on a

CaF2 window in D2O.

A B

Pu

mp

Fre

qu

ency

(cm

-1)

Pu

mp

Fre

qu

ency

(cm

-1)

Probe Frequency (cm-1

) Probe Frequency (cm

-1)

The differences between the pH 6.8

and the pH 12 spectra demonstrate our

sensitivity to the change of environment that

the amide backbone is experiencing. There

are evident shifts between the relative

intensities of the 1670 cm-1 and the 1630 cm-

1 peaks. This can be explained by a change in

angle as the peptide inserts into the membrane

or due to peptide-peptide interations.48,52,57,58

In this system, membrane-peptide

interactions that are created can also

influence the amide I frequency as the

bundles are formed. We can also consider

possible changes in secondary structure. For

example, the alm is destabilized when it

forms a 310-helix, causing it to insert into the

membrane in order to stabilize the structure.

Most importantly, these results gave us the

confidence to move beyond small peptides.

Being able to detect these structural changes

with the small amount of sample is

encouraging as we move to larger systems

where we will need to detect changes in a

single isotope label.

In order to test our abilities to run alm

as a voltage-gated experiment, several

modifications were needed. First, the thin

gold-coated substrates were not

macroscopically conductive, despite being

thicker than the percolation threshold

(thickness needed for conductivity). This

required us to determine what thickness of

gold yielded a good signal without lineshape distortions previously observed when adding a large

amount of plasmonic material to a CaF2 window.9 Thicknesses of gold between 2 and 10 nm of

gold were tested for both conductivity and signal ‘quality’ with alamethicin. Conductance was

determined by measuring the resistance across approximately one centimeter of the coated window

with a multimeter. The standard of reasonable conductance of the electrode was taken to be less

than 50-kΩ of resistance, as previously noted.9 It was determined that 8 nm of thermally evaporated

gold was the minimum thickness that resulted in an appropriate resistance, and spectra were taken

using 10 nm and 8 nm windows.

Figure 4. Representative 2D IR spectra of the amide I

region for A) alamethicin at pH 6.8 B) alamethicin at pH

12 both deposited on a single DPTE/POPC bialyer

tethered to 2-nm of thermally evaporated gold on a CaF2

window in D2O.

A

B

A new sample cell was also needed to perform these experiments to allow for the

connection of electronics to the sample. To this end, I designed and 3D printed several iterations

of a spectro-electrochemical cell (Figure 5a). The cell is made of plastic and is therefore safer to

use than our conductive metal sample cells. We use conductive aluminum tape to both connect the

gold (working electrode) to the potentiometer and act as a counter electrode (Figure 5b). Since we

are focused on placing a constant potential over a membrane, we are not currently concerned about

having a reference electrode. Due to evaporation issues in losing solution over the course of

averaging, a flow cell design was also implemented so that liquid could be replenished over the

course of the experiment.

Some promising voltage-dependent data were taken with a 10 nm gold-coated window, but

were not replicable. This was due to generally a large negative offset in signal that washed out the

peptide signal, perhaps due to the presence of hot carriers. This signal was not seen in a control

experiment at a negative waiting time (t2) where the probe pulse arrives before the pump pulse,

indicating that it is not scatter. Other experimental set ups attempted included 8 nm of gold in both

a transmission and reflection geometry, 10 microns of gold in a reflection geometry, and 10 nm of

palladium topped with 2 nm of gold. In the successful preliminary experiment, spectra of the alm

on the lipid bilayer in a KCl/D2O buffer were taken at 0 and ±900 mV. The 0-mV spectrum was

subtracted from each the -900- and +900-mV spectra in order to both eliminate the non-resonant

response of the thick gold and act as a background subtraction shown in figure 6. This yielded

difference spectra qualitatively similar to the basic pH spectrum.

Figure 5. A) Designed spectro-electrochemical flow cell for the voltage-dependent 2D IR studies. The slot

visible at the top of the window is for inserting the conductive aluminum tape. B) Proposed electrochemical

experiment set up with alamethicin on a DPTE/POPC fused bilayer tethered to a thin gold film on a CaF2

window. Space between the two CaF2 windows is created by a 56-μm spacer.

A B

Due to

aforementioned

difficulties, we

are attempting to

create a more

reproducible

window as an

electrode.

Towards this

end, our

collaborator in

the Arnold group

is beginning to

make substrates

with a 5 nm layer

of spin-coated

then annealed

indium tin oxide

(ITO) under 3

nm of thermally

evaporated gold.

This may work

better due to the lack of plasmon from the ITO so we are able to use the thickness of the gold we

found to be best for the plasmonic enhancement of monolayers. We are still observing a negative

response from the material, but have been able to subtract it out to see an alm signal by a peptide-

free spectrum from a peptide spectrum.

While working

on this new

experimental set up, we

have also been

collecting surface-

enhanced spectra of

KcsA to clarify the

mechanism of ion

permeation. As before,

we prepared these

samples by creating a

monolayer of DPTE on

a calcium fluoride

window coated with 3

nm of gold. Buffered

KcsA in vesicles were

deposited, allowed to

form into a bilayer, then

gently washed to

eliminate any stacked

bilayers. They were then

exchanged with D2O 50

times to eliminate any

Figure 7. Representative 2D IR spectra of

the amide I region for A) unlabeled KcsA

B) Val76, Gly 77, and Gly79 labelled

KcsA C) a zoom of the unlabeled spectrum

subtracted from the labeled spectrum to

leave just the isotope labels. The two

samples were deposited on the same DPTE

coated window with a 3-nm gold thermally

evaporated film, hydrated in D2O with a 12

μm spacer.

A

B

B

C

Figure 6. Spectra of alamethicin in a tethered

lipid bilayer on 10 nm thermally evaporated

Au at A) 0V B) -900mV and C) the 0V

spectrum subtracted from the -900 mV

spectrum, yielding qualitative similarities to

the basic pH spectrum.

Pu

mp

Fre

qu

ency

(cm

-1)

Pu

mp

Fre

qu

ency

(cm

-1)

Pu

mp

Fre

qu

ency

(cm

-1)

Pu

mp

Fre

qu

ency

(cm

-1)

Pu

mp

Fre

qu

ency

(cm

-1)

Pu

mp

Fre

qu

ency

(cm

-1)

Probe Frequency (cm-1

) Probe Frequency (cm-1

)

Probe Frequency (cm-1

)

Probe Frequency (cm-1

) Probe Frequency (cm-1

)

Probe Frequency (cm-1

)

A B

C

H2O from appearing in the amide I region as described in previously. The spots are then rehydrated

in D2O. Preliminary spectra have shown consistency with the work of previous work in the Zanni

group as shown in figure 7. Furthermore, we are able to see isotope labels by using a subtraction

scheme where an unlabeled spectrum is subtracted from an isotope labelled spectrum by

normalizing to the α-helical peak.1 This allows the bulk of the amide I region to be eliminated,

making the isotope labels visible. We are now prepared to take spectra of the newly labelled KcsA

as detailed in my future plans.

Future Plans

Finish work with alamethicin for publication – In the next few months we will finish work on the

alm project. Once we have replicable voltage data, we will write a paper on the static voltage

technique. I simply need to finalize the flow cell set up and to find a suitable combination of

conductive materials to form the electrode on the CaF2 window. I will then take the voltage data,

compare the results both to our own pH-modulated data, and SFG data from other groups. This

will pave the way for more complex systems by serving as a strong proof of concept experiment.

KcsA mechanism validation – As we finish the work with alamethicin, we will transition to KcsA.

We want to add additional constraints to the existing models by adding data from new isotope

labelling schemes in the selectivity filter. With our surface-enhanced method, we believe that we

will be able to observe single isotope labels, rather than the multi-label scheme employed in

previous work. We can then observe Val76 and Gly77 individually in samples prepared with native

protein ligation by our collaborators as detailed in the SI. These sites are of significant interest

after modeling work done by the group of Jim Skinner at the University of Chicago (private

correspondence). They modeled both the hard- and soft-knock models including both the flipped

and unflipped states of Val76. They then took the trajectories and used them to generate weighted

2D IR spectra of each individual label as explained in the SI. The models yielded readily

distinguishable spectra. In both cases, the soft-knock mechanism yielded an inhomogenously

broadened features between 1570 and 1590 cm-1, while the hard knock mechanism had narrower

peaks with maxima between 1550 and 1570 cm-1. The differences in peak width is especially

apparent in the case of Gly77, and examples of the different types of broadening are showing in

figure S4. Experiments with the suggested labels will definitively clarify the mechanism for KcsA,

which serves as the model for K+ channels. Furthermore, our data benefits computational models

that can use this data to further constrain averaged spectra.

Many models of the mechanism of K+ permeation through KcsA has included a 50 mV

drop across the bilayer.18 Using the static voltage methods I am working to develop, we plan to

repeat our KcsA experiments with an applied voltage. This will allow us to rectify any differences

between our experimental results and competing models by providing data using the parameters

of the simulated data. Again, this is a case of more data contributing both to a better understanding

of biological processes and also providing data with which to test simulations and models.

Transient Voltage-Jump experiment – Finally, I will work to develop a 2D IR voltage-jump (V-

jump) experiment. The end goal is to be able to bridge the transient ion current data currently

collected from patch clamp experiments and the static structural data gleaned from

crystallography, taking

advantage of the

structural and temporal

resolution of 2D IR

spectroscopy. This will

entail creating a new

pulsed experiment

where we will create a

voltage using an

arbitrary wave

generator (AWG)

triggered off of the

laser. Using a function

generator, we are able

to control both the amplitude and duration of the applied potential. We are writing a code to collect

2D IR spectra at different voltage waiting times (Tv) after the laser pulse (Figure 8), similar to

temperature-jump experiments. This will allow us to observe the structural dynamics of voltage-

gating in real time.

There are two processes that we plan to observe using this method. First, we will be able

to observe the ion permeation through the channel by using isotope labels in the selectivity filter

of an ion channel and again comparing to models. This will allow us to compare the mechanism

we inferred through our static experiments with what is happening in real time. Second, we will

be able to observe allosteric processes that occur in the more complex voltage-gated K+ channel,

KvAP. KvAP uses an arginine-rich S4 helix to sense a potential drop.59–62 The change in voltage

elicits a yet-to-be-determined physical result in the paddle. There are a few existing theories: the

sliding helix (or canonical) model where the S4 helix moves out of the membrane and turns, the

paddle model where the angle of the voltage sensing domain changes and the twisted S4 model

where a kink in the center of the helix causes the two helical segments to move separately.63 Using

the V-jump technique, we will be able to observe the movement of the voltage paddle. By

comparing these results with modeled spectra, we will then be able to determine the correct

mechanism of gating. We could, also, determine if the movement of the voltage paddle is the rate

determining step, or if there is a currently unknown structural rearrangement between the paddle

moving and the pore opening that determines the response time of the opening of the ion channel.

This can easily be extracted by determining if the movement of the paddle is synchronous with the

pore opening (making it the rate determining step) or before (indicating an unknown middle step).

This is only the start of what could be asked by the V-jump experiment.

Figure 8. Proposed pulse sequence for the transient voltage experiments. After zero

potential, a potential will quickly be applied. Then at a delay time, tv a 2D IR

spectrum will be taken. After averaging the 2D IR spectrum at the first value of tv,

the delay time will change, allowing for spectra taken at several points after the

applied voltage in order to measure the structural response to the potential.

Supporting Information:

2D IR spectroscopy: Two-dimensional infrared spectroscopy measures the third-order response

function of a sample as a result of three light-matter interactions exciting the vibrational transitions

of a molecule. The methods of collecting the spectra vary, but all use an identical pulse sequence

(figure S1) where three pulses of

light interact with a sample to create

a signal. The signal can then be

heterodyned by either with the final

probe pulse or with a separate local

oscillator. The first pulse puts the

sample in a coherence state and it

stays there during the coherence

time, t1. The second pulse then

brings it into a population state that

evolves during the population time, t2. Finally, the third pulse will cause a coherence between the

ground and first excited state or the first and second excited states, stimulating a combined ground

state bleach and stimulated emission or excited state absorption signals respectively.4

The direction of this signal depends on the phase matching geometry utilized. The Zanni

group uses a pump-probe geometry in which the first two pump pulses are collinear and spatially

offset from the third, pump, pulse resulting in a signal that is emitted in the same direction as the

probe. This set up is convenient for a few reasons. First, the signal is self-heterodyned in that we

use the probe pulse as our local oscillator to extract the phase information from the sample.

Secondly, the rephasing and non-rephasing signals are emitted in the same direction, allowing us

to cut down on data collection time and yielding purely absorptive data directly.64,65

The two pump pulses are created by using a germanium acousto-optical amplifier (Ge-

AOM). Briefly, this is used as a pulse shaper to turn our pump pulse into two, phase-controlled

pump pulses separated by different time delays. The incoming pulse is first Fourier transformed

from the time to frequency domain with a grating. An acoustic wave, or mask, is applied to the Ge

crystal, acting as a diffraction grating in the frequency domain for the incoming light. After the

Ge-AOM, a second grating is used to return the pulse to the time-domain, yielding the double

pump pulse.25,64 The Ge-AOM can be modulated on a shot-to-shot basis.66 We generate our spectra

in the time-frequency domain such that the pump axis is generated by scanning the delay t1 and

performing a Fourier transform, while the probe axis is collected by the dispersed frequencies on

an MCT array.

The interference between the electric fields of the signal and the local oscillator contains

the information in this experimental set up. The signal in the frequency domain is

𝑆 ∝ ∫ |(𝐸𝐿𝑂(𝜔) + 𝐸𝑠𝑖𝑔(3)(𝜔))|

2

𝑑𝜔 ≈ 𝐸𝐿𝑂2 (𝜔) + 𝐸𝑠𝑖𝑔

(3)2(𝜔) + 2𝐸𝐿𝑂(𝜔)𝐸𝑠𝑖𝑔

(3)(𝜔)

where 𝐸𝐿𝑂2 (𝜔) is the intensity of the pump pulse (local oscillator), 𝐸𝑠𝑖𝑔

(3)2(𝜔) is the intensity of the

signal, and the 2𝐸𝐿𝑂(𝜔)𝐸𝑠𝑖𝑔(3)(𝜔) cross term is the signal of interest. We are able to isolate this

cross term again by using the pulse shaper. Every other pulse takes a spectrum as described, while

the intermittent pulse turns ‘off’ the pump by using the Ge-AOM as a chopper. As a result, we

have spectra that are just the local oscillator by itself. We can then subtract off the local oscillator

as is done is absorption spectroscopy where the signal is approximately

𝑆 = log (𝑝𝑢𝑚𝑝 𝑜𝑛

𝑝𝑢𝑚𝑝 𝑜𝑓𝑓) ≈ log (

𝐸𝐿𝑂2 (𝜔)𝐸𝑠𝑖𝑔

(3)2(𝜔) + 2𝐸𝐿𝑂(𝜔)𝐸𝑠𝑖𝑔

(3)(𝜔)

𝐸𝐿𝑂2 (𝜔)

) ≈ 2𝐸𝐿𝑂(𝜔)𝐸𝑠𝑖𝑔

(3)(𝜔)

𝐸𝐿𝑂2 (𝜔)

Figure S1. Pulse sequence for a 2D IR experiment.

Eq. S1

Eq. S2

The ability to modulate the pump on and off is clearly very useful in data collection. In total, the

use of the pulse shaper aids in spectra collection, allows us to reduce pump scatter through phase

cycling, and allows for faster data collection.4

Sample Preparation: Alamethicin derived from Trichoderma viride purchased from Sigma-Aldrich

was diluted to a concentration of 2 mg/mL in methanol for storage purposes. Lipids, 1-palmitoyl-

2-oleoyl-glycero-3-phosphocholine (POPC) and 1,2-Dipalmitoyl-sn-Glycero-3-

Phosphothioethanol (DPTE) (Figure S2) both in chloroform at a concentration of 10 mg/mL were

purchased from Avanti Polar Lipids Inc. Calcium fluoride (CaF2) windows (25 mm x 2 mm) were

purchased from Crystran.

Gold films (Au% > 99.99%) were deposited via thermal evaporation to varying thicknesses

at the rate of 0.1 Å/s at a pressure of 8 x 10-7 torr onto the CaF2 window. Films of indium tin oxide

(ITO) and gold were created first by sputter-coating the window with 5 nm of ITO and annealing

at 350 °C for 1.5 hours. The gold was then deposited to the desired thickness as described. They

layered metal (Pd/Au) samples were created by first thermally evaporating the palladium, then

adding the gold on top separately.

A 100-μL aliquot of DPTE in

chloroform (10 mg/mL) and 250-μL

aliquot of POPC in chloroform (10

mg/mL) were dried down under N2 and

lyophilized overnight to remove

remaining solvent. DPTE was rehydrated

in 10 mL of spectroscopic-grade ethanol.

The coated window was then submerged

in the DPTE/ethanol solution overnight.

The window was then rinsed with ethanol

and allowed to dry. For alamethicin

samples, POPC vesicles were prepared by

rehydrating the POPC in 100 mM NaCl buffer and applying five freeze-thaw cycles in isopropanol

and dry ice and lukewarm water respectively. The solution was then placed in a bath sonicator for

10 minutes to create approximately 50-nm diameter vesicles. Window was then submerged in the

POPC vesicle solution overnight to allow for vesicle rupture. The window was then washed with

deionized water. Alamethicin was prepared by diluting 5.2 μL of the stock solution in 50 μL of

deionized water. This was then distributed over half of the lipid-coated CaF2 window. Assuming

total surface coverage of lipid, this should create a 20:1 lipid:peptide ratio. The window was dried

under N2 and then rehydrated with D2O or phosphate buffer (pH 6, 12) in D2O and a second

window was added separated with a 56 μm spacer.

For the KcsA samples, the DPTE monolayer was created in the same fashion. KcsA (1.5

mg/mL) in 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) (0.25% w/v) and 50 mM Tris,

150 mM KCl, and 1 mM DTT buffer (pH 7.5), both those that were unlabeled and labelled, were

deposited in separated 0.2-μL spots on the same window, and gently rinsed to remove excess

sample. The spots were then exchanged 50 times with D2O by adding 0.2 μL at a time and allowing

it to dry under N2. The spots were then rehydrated with 0.2 μL of D2O and placed between two

CaF2 windows with a 12-μm spacer.

2D IR data acquisition: A Coherent Libra regenerative amplifier centered at 800 nm (3.4 W, 50 fs)

pump an optical parametric amplifier (OPA) (Light Conversion TOPAS). The output signal and

idler pulses are then mixed in a AgGaS2 difference frequency generation (DFG) crystal to generate

the mid-IR light centered at 6 μm with pulses of <100 fs with energies around 18 μJ. The mid-IR

Figure S2. The structures of the lipids

A

B

is then split with a 90/10 CaF2 beam splitter into the pump and probe lines. The pump line is

directed through an acousto-optical modulator (AOM) which shapes the pulses as previously

described. The polarization (s- or p-) of both the pump and probe lines are modulated by successive

half-wave plates and polarizers. A 7.5-cm focal length parabolic mirror is used to focus the

spatially and temporally overlapped pulses at the sample. Due to the pump-probe geometry used,

the probe/signal beam is then directed into a monochromator to be dispersed onto a 32-pixel MCT

array.

MD Simulations:

The route from MD simulation to spectra is outlined briefly here.4,67 Spectra are derived from MD

simulations by taking the frequency trajectory of the vibration of interest, in our case the isotope-

labelled backbone carbonyl, and from the trajectory determine the distribution of frequencies. This

distribution will yield a frequency fluctuation correlation function, which can be used to determine

the lineshape function. The lineshape function is key in creating the response function. Then, the

response functions for the rephasing and non-rephasing signals can be treated like collected time-

domain data, and Fourier transformed and summed to create the purely absorptive spectra. Because

2D IR spectra take an ensemble average of the sample of interest, it is necessary to take into

account the different ion occupancies of selectivity filter. This was done by taking the trajectories

of each of the possible ion occupancies for both models individually and determine the resulting

spectra. These spectra were then combined in a weighted average to create the final modeled

spectra.

Figure S4. Modeled spectra with A) homogenous broadening B) partially inhomogeneous broadening C)

inhomogeneous broadening.

A B C

References: (1) Kratochvil, H. T.; Carr, J. K.; Matulef, K.; Annen, A. W.; Li, H.; Maj, M.; Ostmeyer, J.; Serrano, A. L.;

Raghuraman, H.; Moran, S. D.; et al. Instantaneous Ion Configurations in the K+ Ion Channel Selectivity

Filter Revealed by 2D IR Spectroscopy. Science 2016, 353 (6303), 1040–1044.

(2) Kratochvil, H. T.; Maj, M.; Matulef, K.; Annen, A. W.; Ostmeyer, J.; Perozo, E.; Roux, B.; Valiyaveetil, F.

I.; Zanni, M. T. Probing the Effects of Gating on the Ion Occupancy of the K + Channel Selectivity Filter

Using Two-Dimensional Infrared Spectroscopy. J. Am. Chem. Soc. 2017, 139 (26), 8837–8845.

(3) Su, Z.; Shodiev, M.; Leitch, J. J.; Abbasi, F.; Lipkowski, J. Role of Transmembrane Potential and Defects

on the Permeabilization of Lipid Bilayers by Alamethicin, an Ion-Channel-Forming Peptide. Langmuir

2018, 34 (21), 6249–6260.

(4) Hamm, P.; Zanni, M. Concepts and Methods of 2D Infrared Spectroscopy; 2011; Vol. 9781107000.

(5) Rezus, Y. L. A.; Selig, O. Impact of Local-Field Effects on the Plasmonic Enhancement of Vibrational

Signals by Infrared Nanoantennas. Opt. Express 2016, 24 (11), 12202.

(6) Kopec, W.; Köpfer, D. A.; Vickery, O. N.; Bondarenko, A. S.; Jansen, T. L. C.; de Groot, B. L.; Zachariae,

U. Direct Knock-on of Desolvated Ions Governs Strict Ion Selectivity in K+ Channels. Nat. Chem. 2018, 10

(8), 813–820.

(7) Langan, P. S.; Vandavasi, V. G.; Weiss, K. L.; Afonine, P. V.; el Omari, K.; Duman, R.; Wagner, A.;

Coates, L. Anomalous X-Ray Diffraction Studies of Ion Transport in K+ Channels. Nat. Commun. 2018, 9

(1), 4540.

(8) Petti, M. K.; Ostrander, J. S.; Saraswat, V.; Birdsall, E. R.; Rich, K. L.; Lomont, J. P.; Arnold, M. S.; Zanni,

M. T. Enhancing the Signal Strength of Surface Sensitive 2D IR Spectroscopy. J. Chem. Phys. 2019, 150

(2), 024707.

Figure S5. UV-vis spectra of the various coated CaF2

windows

Figure S6. Signal from 5 nm ITO/3nm Au coated window,

representative of the background response of the conductive

substrates.

(9) Lotti, D.; Hamm, P.; Kraack, J. P. Surface-Sensitive Spectro-Electrochemistry Using Ultrafast 2D ATR IR

Spectroscopy. J. Phys. Chem. C 2016, 120 (5), 2883–2892.

(10) Petkov, G. V. Chapter 16 – Ion Channels. In Pharmacology; 2009; pp 387–427.

(11) Declan A. Doyle, João Morais Cabral, Richard A. Pfuetzner, Anling Kuo, Jacqueline M. Gulbis, Steven L.

Cohen, Brian T. Chait, R. M. The Structure of the Potassium Channel: Molecular Basis of K+ Conduction

and Selectivity. Science (80-. ). 1998, 280 (5360), 69–77.

(12) Hodgkin, A. L.; Huxley, A. F.; Katz, B. Measurement of Current‐voltage Relations in the Membrane of the

Giant Axon of Loligo. J. Physiol. 1952, 116 (4), 424–448.

(13) Hille, B. Ionic Channels of Excitable Membranes 2nd Edition. Journal of The Electrochemical Society.

1987, p 343.

(14) Valiyaveetil, F. I.; Zhou, Y.; MacKinnon, R. Lipids in the Structure, Folding, and Function of the KcsA K+

Channel. Biochemistry 2002, 41 (35), 10771–10777.

(15) Zimmer, J.; Doyle, D. A.; Günter Grossmann, J. Structural Characterization and PH-Induced

Conformational Transition of Full-Length KcsA. Biophys. J. 2006, 90 (5), 1752–1766.

(16) Zhou, Y.; Morais-Cabral, J. H.; Kaufman, A.; Mackinnon, R. Chemistry of Ion Coordination and Hydration

Revealed by a K+channel-Fab Complex at 2.0 Å Resolution. Nature 2001, 414 (6859), 43–48.

(17) LeMasurier, M.; Heginbotham, L.; Miller, C. KcsA: It’s a Potassium Channel. J. Gen. Physiol. 2001, 118

(3), 303–314.

(18) Kopfer, D. A.; Song, C.; Gruene, T.; Sheldrick, G. M.; Zachariae, U.; de Groot, B. L.; Quigley, A.; Grieben,

M.; Goubin, S.; Mukhopadhyay, S.; et al. Ion Permeation in K+ Channels Occurs by Direct Coulomb

Knock-On. Science (80-. ). 2014, 346 (6207), 352–355.

(19) Iwamoto, M.; Oiki, S. Counting Ion and Water Molecules in a Streaming File through the Open-Filter

Structure of the K Channel. J. Neurosci. 2011, 31 (34), 12180–12188.

(20) Zhou, Y.; MacKinnon, R. The Occupancy of Ions in the K+selectivity Filter: Charge Balance and Coupling

of Ion Binding to a Protein Conformational Change Underlie High Conduction Rates. J. Mol. Biol. 2003,

333 (5), 965–975.

(21) Morais-Cabral, J. H.; Zhou, Y.; MacKinnon, R. Energetic Optimization of Ion Conduction Rate by the K+

Selectivity Filter. Nature 2001, 414 (6859), 37–40.

(22) Bernèche, S.; Roux, B. Energetics of Ion Conduction through the K+channel. Nature 2001, 414 (6859), 73–

77.

(23) Berneche, S.; Roux, B. A Microscopic View of Ion Conduction through the K+ Channel. Proc. Natl. Acad.

Sci. 2003, 100 (15), 8644–8648.

(24) Noskov, S. Y.; Bernéche, S.; Roux, B. Control of Ion Selectivity in Potassium Channels by Electrostatic and

Dynamic Properties of Carbonyl Ligands. Nature 2004, 431 (7010), 830–834.

(25) Middleton, C. T.; Woys, A. M.; Mukherjee, S. S.; Zanni, M. T. Residue-Specific Structural Kinetics of

Proteins through the Union of Isotope Labeling, Mid-IR Pulse Shaping, and Coherent 2D IR Spectroscopy.

Methods 2010, 52 (1), 12–22.

(26) Ding, B.; Laaser, J. E.; Liu, Y.; Wang, P.; Zanni, M. T.; Chen, Z. Site-Specific Orientation of an a-Helical

Peptide Ovispirin-1 from Isotope-Labeled SFG Spectroscopy. J. Phys. Chem. B 2013, 117 (47), 14625–

14634.

(27) Moran, S. D.; Zanni, M. T. How to Get Insight into Amyloid Structure and Formation from Infrared

Spectroscopy. J. Phys. Chem. Lett. 2014, 5 (11), 1984–1993.

(28) Torres, J.; Kukol, A.; Goodman, J. M.; Arkin, I. T. Site-Specific Examination of Secondary Structure and

Orientation Determination in Membrane Proteins: The Peptidic 13C=18O Group as a Novel Infrared Probe.

Biopolymers 2001, 59 (6), 396–401.

(29) Jekhmane, S.; Medeiros-Silva, J.; Li, J.; Kümmerer, F.; Müller-Hermes, C.; Baldus, M.; Roux, B.;

Weingarth, M. Shifts in the Selectivity Filter Dynamics Cause Modal Gating in K + Channels.

(30) Fowler, P. W.; Abad, E.; Beckstein, O.; Sansom, M. S. P. Energetics of Multi-Ion Conduction Pathways in

Potassium Ion Channels. J. Chem. Theory Comput. 2013, 9 (11), 5176–5189.

(31) Bernèche, S.; Roux, B. A Gate in the Selectivity Filter of Potassium Channels. Structure 2005, 13 (4), 591–

600.

(32) Recordings, P.; Matulef, K.; Valiyaveetil, F. I. Potassium Channels. 2018, 1684, 181–191.

(33) Arkhipova, V.; Guskov, A.; Slotboom, D.-J. Analysis of the Quality of Crystallographic Data and the

Limitations of Structural Models. J. Gen. Physiol. 2017, 149 (12), 1091–1103.

(34) Yu, H. Extending the Size Limit of Protein Nuclear Magnetic Resonance. Proc. Natl. Acad. Sci. 2002, 96

(2), 332–334.

(35) Kornreich, B. G. The Patch Clamp Technique: Principles and Technical Considerations. J. Vet. Cardiol.

2007, 9 (1), 25–37.

(36) Rosenfeld, D. E.; Nishida, J.; Yan, C.; Kumar, S. K. K.; Tamimi, A.; Fayer, M. D. Structural Dynamics at

Monolayer–Liquid Interfaces Probed by 2D IR Spectroscopy. J. Phys. Chem. C 2013, 117 (3), 1409–1420.

(37) Mukherjee, P.; Kass, I.; Arkin, I.; Zanni, M. T. Picosecond Dynamics of a Membrane Protein Revealed by

2D IR. PNAS March 2006, 7 (10), 3528–3533.

(38) Woys, A. M.; Almeida, A. M.; Wang, L.; Chiu, C. C.; McGovern, M.; De Pablo, J. J.; Skinner, J. L.;

Gellman, S. H.; Zanni, M. T. Parallel β-Sheet Vibrational Couplings Revealed by 2D IR Spectroscopy of an

Isotopically Labeled Macrocycle: Quantitative Benchmark for the Interpretation of Amyloid and Protein

Infrared Spectra. J. Am. Chem. Soc. 2012, 134 (46), 19118–19128.

(39) Barth, A. The Infrared Absorption of Amino Acid Side Chains. Progress in Biophysics and Molecular

Biology. 2000, pp 141–173.

(40) Barth, A.; Zscherp, C. What Vibrations Tell Us about Proteins. Quarterly Reviews of Biophysics. 2002, pp

369–430.

(41) Kraack, J. P.; Hamm, P. Surface-Sensitive and Surface-Specific Ultrafast Two-Dimensional Vibrational

Spectroscopy. Chem. Rev. 2017, 117 (16), 10623–10664.

(42) Aouani, H.; Rahmani, M.; Šípová, H.; Torres, V.; Hegnerová, K.; Beruete, M.; Homola, J.; Hong, M.;

Navarro-Cía, M.; Maier, S. A. Plasmonic Nanoantennas for Multispectral Surface-Enhanced Spectroscopies.

J. Phys. Chem. C 2013, 117 (36), 18620–18626.

(43) Kraack, J. P.; Sévery, L.; Tilley, S. D.; Hamm, P. Plasmonic Substrates Do Not Promote Vibrational Energy

Transfer at Solid–Liquid Interfaces. J. Phys. Chem. Lett. 2018, 9 (1), 49–56.

(44) Ataka, K.; Stripp, S. T.; Heberle, J. Surface-Enhanced Infrared Absorption Spectroscopy (SEIRAS) to Probe

Monolayers of Membrane Proteins. Biochim. Biophys. Acta - Biomembr. 2013, 1828 (10), 2283–2293.

(45) Uchida, T.; Osawa, M.; Lipkowski, J. SEIRAS Studies of Water Structure at the Gold Electrode Surface in

the Presence of Supported Lipid Bilayer. J. Electroanal. Chem. 2014, 716, 112–119.

(46) Sarid, D.; Challener, W. Localized Surface Plasmons. In Modern Introduction to Surface Plasmons; 2013;

pp 201–255.

(47) Nishida, J.; Yan, C.; Fayer, M. D. Enhanced Nonlinear Spectroscopy for Monolayers and Thin Films in

Near-Brewster’s Angle Reflection Pump-Probe Geometry. J. Chem. Phys. 2017, 146 (9).

(48) Su, Z.; Jay Leitch, J.; Abbasi, F.; Faragher, R. J.; Schwan, A. L.; Lipkowski, J. EIS and PM-IRRAS Studies

of Alamethicin Ion Channels in a Tethered Lipid Bilayer. J. Electroanal. Chem. 2018, 812, 213–220.

(49) Lewis, R. N. A. H.; Mcelhaney, R. N.; Pohle, W.; Mantsch, H. H. Components of the Carbonyl Stretching

Band in the Infrared Spectra of Hydrated 1,2-Diacylglycerolipid Bilayers: A Reevaluation; 1994; Vol. 67.

(50) Leitgeb, B.; Szekeres, A.; Manczinger, L.; Vágvölgyi, C.; Kredics, L. The History of Alamethicin: A

Review of the Most Extensively Studied Peptaibol. Chem. Biodivers. 2007, 4 (6), 1027–1051.

(51) Bechinger, B. Structure and Functions of Channel-Forming Peptides: Magainins, Cecropins, Melittin and

Alamethicin. Journal of Membrane Biology. 1997, pp 197–211.

(52) Ye, S.; Nguyen, K. T.; Chen, Z. Interactions of Alamethicin with Model Cell Membranes Investigated Using

Sum Frequency Generation Vibrational Spectroscopy in Real Time in Situ. J. Phys. Chem. B 2010, 114 (9),

3334–3340.

(53) Yang, P.; Wu, F.-G.; Chen, Z. Lipid Fluid–Gel Phase Transition Induced Alamethicin Orientational Change

Probed by Sum Frequency Generation Vibrational Spectroscopy. J. Phys. Chem. C 2013, 117 (33), 17039–

17049.

(54) Marsh, D. Orientation and Peptide−Lipid Interactions of Alamethicin Incorporated in Phospholipid

Membranes: Polarized Infrared and Spin-Label EPR Spectroscopy. Biochemistry 2009, 48 (4), 729–737.

(55) Yin, P.; Burns, C. J.; Osman, P. D. J.; Cornell, B. A. A Tethered Bilayer Sensor Containing Alamethicin

Channels and Its Detection of Amiloride Based Inhibitors.

(56) Maekawa, H.; Formaggio, F.; Toniolo, C.; Ge, N.-H. Chain Length Dependence of Two-Dimensional

Infrared Spectral Pattern Characteristic to 310-Helix Peptides; 2009; pp 415–417.

(57) Roman Volinsky, †,‡; Sofiya Kolusheva, †,‡; Amir Berman, *,†,§ and; Raz Jelinek*, †,‡. Microscopic

Visualization of Alamethicin Incorporation into Model Membrane Monolayers. 2004.

(58) Nagao, T.; Mishima, D.; Javkhlantugs, N.; Wang, J.; Ishioka, D.; Yokota, K.; Norisada, K.; Kawamura, I.;

Ueda, K.; Naito, A. Structure and Orientation of Antibiotic Peptide Alamethicin in Phospholipid Bilayers as

Revealed by Chemical Shift Oscillation Analysis of Solid State Nuclear Magnetic Resonance and Molecular

Dynamics Simulation. Biochim. Biophys. Acta - Biomembr. 2015, 1848 (11), 2789–2798.

(59) Schmidt, D.; Cross, S. R.; MacKinnon, R. A Gating Model for the Archeal Voltage-Dependent K+Channel

KvAP in DPhPC and POPE:POPG Decane Lipid Bilayers. J. Mol. Biol. 2009, 390 (5), 902–912.

(60) Ruta, V.; Chen, J.; MacKinnon, R. Calibrated Measurement of Gating-Charge Arginine Displacement in the

KvAP Voltage-Dependent K+ Channel. Cell 2005, 123 (3), 463–475.

(61) Bruhova, I.; Zhorov, B. S. KvAP-Based Model of the Pore Region of Shaker Potassium Channel Is

Consistent with Cadmium- and Ligand-Binding Experiments. Biophys. J. 2005, 89 (2), 1020–1029.

(62) Cohen, B. E.; Grabe, M.; Jan, L. Y. Answers and Questions from the KvAP Structures. Neuron. 2003, pp

395–400.

(63) Sands, Z. A.; Grottesi, A.; Sansom, M. S. P. The Intrinsic Flexibility of the Kv Voltage Sensor and Its

Implications for Channel Gating. Biophys. J. 2006, 90 (5), 1598–1606.

(64) Shim, S.-H.; Zanni, M. T. How to Turn Your Pump–probe Instrument into a Multidimensional

Spectrometer: 2D IR and Vis Spectroscopiesvia Pulse Shaping. Phys. Chem. Chem. Phys. 2009, 11 (5), 748–

761.

(65) DeFlores, L. P.; Nicodemus, R. A.; Tokmakoff, A. Two-Dimensional Fourier Transform Spectroscopy in

the Pump-Probe Geometry. Opt. Lett. 2007, 32 (20), 2966.

(66) Kearns, N. M.; Mehlenbacher, R. D.; Jones, A. C.; Zanni, M. T. Broadband 2D Electronic Spectrometer

Using White Light and Pulse Shaping: Noise and Signal Evaluation at 1 and 100 KHz. Opt. Express 2017,

25 (7), 7869.

(67) Wang, L.; Middleton, C. T.; Zanni, M. T.; Skinner, J. L. Development and Validation of Transferable

Amide i Vibrational Frequency Maps for Peptides. J. Phys. Chem. B 2011, 115 (13), 3713–3724.