Article

The Pathogenic A116V Mutation Enhances Ion-Selective Channel Formation by Prion Protein inMembranes

Ambadi Thody Sabareesan,1 Jogender Singh,1 Samrat Roy,1,2,3 Jayant B. Udgaonkar,1,* and M. K. Mathew1,*1National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bengaluru, India; 2Biocon Bristol Myers Squibb ResearchCenter, Bengaluru, India; and 3School of Biotechnology, Kalinga Institute of Industrial Technology (KIIT) University, Bhubaneswar, India

ABSTRACT Prion diseases are a group of fatal neurodegenerative disorders that afflict mammals. Misfolded and aggregatedforms of the prion protein (PrPSc) have been associated with many prion diseases. A transmembrane form of PrP favored bythe pathogenic mutation A116V is associated with Gerstmann-Straussler-Scheinker syndrome, but no accumulation of PrPSc isdetected. However, the role of the transmembrane form of PrP in pathological processes leading to neuronal death remains un-clear. This study reports that the full-length mouse PrP (moPrP) significantly increases the permeability of living cells to Kþ, andforms Kþ- and Ca2þ-selective channels in lipid membranes. Importantly, the pathogenic mutation A116V greatly increases thechannel-forming capability of moPrP. The channels thus formed are impermeable to sodium and chloride ions, and are blockedby blockers of voltage-gated ion channels. Hydrogen-deuterium exchange studies coupled with mass spectrometry (HDX-MS)show thatupon interactionwith lipid, thecentral hydrophobic region (109–132) of theprotein is protectedagainst exchange,makingit a good candidate for inserting into themembrane and lining the channel. HDX-MS also shows a dramatic increase in the protein-lipid stoichiometry for A116VmoPrP, providing a rationale for its increased channel-forming capability. The results suggest that ionchannel formation may be a possible mechanism of PrP-mediated neurodegeneration by the transmembrane forms of PrP.

INTRODUCTION

Transmissible spongiform encephalopathies, generallyknown as prion diseases, are a group of fatal neurodegener-ative disorders that can affect several mammalian species,including humans. The prion protein (PrP), which normallyis glycophosphatidylinositol (GPI) anchored to the cellmembrane, is critical for the transmission and pathogenesisof these disorders. The normal function of PrP is un-known (1). Disease pathology is usually associated with aconformational conversion of native monomeric cellularPrP (PrPC) to an aggregated oligomeric form (PrPSc). How-ever, there is increasing evidence that this feature is animportant, but not sufficient, factor in disease etiology.Although PrPSc is well established as the infectious form,it may not be the direct cause of neurodegeneration, at leastin some prion diseases. It appears that alternative forms ofPrP (both soluble and membrane-bound), which differfrom PrPSc in both structural and biochemical properties,may have important roles in prion-mediated neurodegener-

Submitted December 7, 2015, and accepted for publication March 7, 2016.

*Correspondence: jayant@ncbs.res.in or mathew@ncbs.res.in

Editor: Hagan Bayley.

1766 Biophysical Journal 110, 1766–1776, April 26, 2016

http://dx.doi.org/10.1016/j.bpj.2016.03.017

� 2016 Biophysical Society

ation (2). The mechanisms underlying neurodegeneration,however, remain unclear (3–5).

Soluble forms of PrP have been found to occur naturally inthe cytoplasm of neurons inmany parts of the brain (6). Cyto-solic PrP, which does not have the GPI anchor, has beenshown to cause neurodegenerative features, in the absenceof any significant accumulation of PrPSc, in culturedneuronal cell lines as well as in animal models of prion dis-ease (7–9). Soluble forms of PrP without the GPI anchor mayalso be secreted, and secreted PrP can also be toxic (10).Anchorless PrP expression has been shown to cause braindamage in transgenic mice (7). Moreover, PrP can be shedfrom the cell surface upon cleavage, either within the proteinor at the linkage to the GPI anchor, under physiologicalconditions (11,12). Unanchored soluble protein has beenfound in both the cerebrospinal fluid and nasal secretionsof patients (13,14). Clearly, it has become important to studythe mechanisms by which soluble PrP, such as bacteriallyproduced recombinant PrP, may induce toxicity in cells.

The extent of neurotoxicity of non-PrPSc disease-linkedforms of PrP correlates well with their ability to interact withlipid membranes (7,15–19). Interestingly, transmembraneforms of PrP have been shown to induce neurodegeneration

Ion Channel Formation by Prion Protein

even when no evidence for PrPSc can be detected in the brain(15). In particular, it appears that brain material from rodentsthat have succumbed to Gerstmann-Straussler-Scheinker(GSS) syndrome-associated prion variants is not infectiouswhen injected into the brains of normal rodents (20). Severalmutations in PrP enhance the formation of transmembranePrP, including a single Ala to Val mutation at residue position117 (A117V) in human PrP (15). The A117V mutation is apathogenic mutation associated with GSS syndrome, butthe mechanism by which it acts is not known.

A possible mechanism by which PrP could be toxic issuggested by studies of the interactions between lipid mem-branes and peptides derived from PrP: peptides derived fromsequence stretches 82–146, 105–126, and 185–206 havethe capability to form ion channels (21–25). Further, pep-tides covering the sequence stretches 105–126, 105–135,118–135, and 185–206 induce features of neurotoxicity incultured cell lines and mouse brain slices (22,24,26–29).However, these peptides are not present in physiologicalconditions, and it has not yet been shown whether intact,full-length, native PrP has the capability to form ion chan-nels. Although the N-terminal unstructured domain of PrPhas been shown to be lipophilic (30), PrP-membrane inter-actions remain to be characterized in detail.

In this study, a detailed biophysical characterization of theinteraction of recombinant wild-type (WT) as well as A116V(corresponding to A117V in human PrP) full-length mousePrP (moPrP) with lipid membranes was carried out usingmultiple spectroscopic techniques and black lipid membrane(BLM) electrophysiology. It is shown that both WT andA116V moPrP bind to lipid membranes and form well-defined ion channels. Importantly, the extents of interactionand ion channel formation are much higher for A116VmoPrP than for WT moPrP. The channels formed by thesetwo proteins allow the transit of potassium and calciumions, but, surprisingly, are unable to conduct sodium or chlo-ride ions. The channel blockers lanthanum chloride, 4-ami-nopyridine (4-AP), and tetraethylammonium (TEA) blockthe channels. Hydrogen-deuterium exchange studies coupledwith mass spectrometry (HDX-MS) suggest that the centralhydrophobic region (109–132) of the protein becomes buriedin the lipid membrane, and the amount of buried protein is~10-fold higher in the case of A116V moPrP. Patch-clampstudies on living cells show that whole-cell currents elicitedby A116V moPrP are larger than those elicited by WTmoPrP. The results of this study suggest that ion channel for-mation could be a possible mechanism of PrP-mediated neu-rodegeneration by the transmembrane forms of PrP.

MATERIALS AND METHODS

Protein expression and purification

The full-length moPrP (23–231; WT moPrP) and the C-terminal domain of

moPrP (121–231; CTD moPrP) were expressed and purified as described

previously (31,32). The purified protein was then used either directly or

after treatment to remove any very small amounts of oligomer that escaped

detection by dynamic light scattering (DLS). For treatment, the moPrP var-

iants were first incubated in 8 M urea at pH 4, 25�C, for 1 h to denature anypossible aggregates present. The proteins were then refolded in 10 mM so-

dium acetate (pH 4) using a Sephadex G-25 HiTrap desalting column in

conjunction with an AKTA Basic high-performance liquid chromatography

system (GE HealthCare Life Sciences, Pittsburgh, PA). The moPrP variants

at a concentration of 1 mM, pH 4, were then mixed with 8-anilinonaphtha-

lene-1-sulfonic acid (ANS) at a final concentration of 10 mM. The samples

were incubated at room temperature for 30 min before fluorescence spectra

were acquired. The fluorescence spectra were acquired from 400 to 600 nm,

with excitation at 365 nm. The interactions of treated and untreated proteins

(both the WT and the mutant variant) with lipids and cell membranes were

found to be identical.

Preparation of unilamellar liposomes

Unilamellar liposomes were made from 1,2-diphytanoyl-sn-glycero-3-

phosphocholine (DPhPC; Avanti Polar Lipids, Alabaster, AL) and choles-

terol (1:1 v/v) in 5 mM HEPES buffer (pH 7.4) containing 5 mM CaCl2,

100 mM NaCl, and 3 mM dextran (10 kDa) as described earlier (33).

Calcium release was measured as described in Supporting Materials and

Methods in the Supporting Material.

Planar BLM experiments

Planar lipid bilayers were made frommonolayers by using a modification of

the technique previously described by Montal and Mueller (34). Ion selec-

tivity was assayed using asymmetric buffer conditions. In general, the

cis chamber contained 1 M KCl and the composition of the trans chamber

was varied as described in Supporting Materials and Methods.

Whole-cell patch-clamp experiments

Whole-cell patch-clamp experiments were conducted using standard

methods. These methods and all other experimental procedures are

described in detail in Supporting Materials and Methods.

RESULTS

WT and A116V moPrP bind to lipid membranesand induce calcium release

To determine whether WT moPrP and A116V moPrP bindto and permeabilize membranes, to show that addition ofmonomeric moPrP can result in channel formation in mem-branes, and to characterize the ion selectivities of thesechannels, studies were carried out with unilamellar lipo-somes and BLM.

Unilamellar liposomes were made of DPhPC and choles-terol. Incubation of 1 mM WT or A116V moPrP in 5 mMHEPES buffer (pH 7.4) at 25�C resulted in an increase intryptophan fluorescence emission intensity together with ablue shift in the emission maximum (Figs. 1 a and S1, aand b). Both the spectral shift and the enhancement of inten-sity increased with an increase in the concentration of lipo-somes, indicating a change in the tryptophan environmenttoward lower polarity. The limiting amplitude of tryptophanfluorescence at high lipid concentrations was z40% higher

Biophysical Journal 110, 1766–1776, April 26, 2016 1767

FIGURE 1 moPrP binds to lipid membranes and

induces calcium release. (a) Dependence of the

change in total emitted tryptophan fluorescence of

1 mMA116V (>) and WT (B) moPrP upon addi-

tion of increasing concentrations of liposomes.

(b) Increase in the intensity of Fura-2 fluorescence

upon the release of calcium from liposomes;

100 nM of A116V (dashed line) or WT (solid

line) moPrP was added to the liposomes at time

zero. The inset in (b) shows log-log plots of the

initial rate of calcium releaseversus protein concen-

tration for A116V (>) and WT (B) moPrP. The

slope estimated from the plot is z1.0 for both

proteins. (c and d) Blockage of calcium release

from the liposomes upon incubation with (c)

100 nMWTmoPrP and (d) 100 nMA116V moPrP.

In both panels, liposomes were incubated with

100 nM WT or A116V (solid line) moPrP in the

presence of the sodium channel blocker tetrodo-

toxin (500 nM, dashed line), potassium channel

blocker 4-AP (50 nM, dashed double dotted line),

potassium channel blocker TEA (1 mM, dashed

single dotted line), or calcium channel blocker

LaCl3 (50nM,dotted line), eachofwhichwas added

to the sample before the start of recording. In all

panels, the thick black dotted line represents the

fluorescence signal upon complete lysis of the lipo-

somes using 10% Tween 20, and the small-dotted

line represents the baseline fluorescence signal

from the untreated liposomes. In panel (a), the error

bars represent standard deviations calculated from

three independent experiments. In panels (b) and

(c), the kinetic traces are representative traces.

Sabareesan et al.

for A116V moPrP than for WT moPrP. The dissociationconstant for binding observed for A116V moPrP was alsosignificantly lower (30 ng/mL vs. 60 ng/mL for WT moPrP),suggesting that A116V moPrP has a greater propensity tointeract with lipid membranes than does WT moPrP. BothWT and A116V moPrP appeared to retain their native struc-tures in the presence of liposomes, as seen by both circulardichroism (CD) and infrared absorption spectroscopy(Fig. S2), with no clear evidence for the lipid-induced struc-tural conversion in the C-terminal region reported earlier(35). It should be noted, however, that under the conditionsof the CD experiment, less than 10% of the protein would belipid associated (see Supporting Materials and Methods).Consequently, a conformational switch would not be ex-pected to be detectable by CD. The Fourier transforminfrared (FTIR) experiment, on the other hand, was carriedout with a protein/lipid ratio at which most of the proteinwould be associated with lipid, and any large conforma-tional change would have been detected. The absence of asignificant spectral change in the FTIR experiment wouldsuggest that either no significant conformational changeoccurred in the protein upon lipid association, a significantconformational change occurred in only a small fractionof the protein molecules, or a significant conformationalchange occurred in only a small portion of each lipid-asso-ciated protein molecule.

1768 Biophysical Journal 110, 1766–1776, April 26, 2016

To checkwhether themoPrP-lipid interaction affectsmem-brane integrity, a calcium release experiment was performedwith liposomes (see Supporting Materials and Methods).Over a 100-fold concentration range (10–1000 nM), bothWT and A116V moPrP induced calcium release from lipo-somes (Figs. 1 b and S1, c and d). The observed gradualincrease of the fluorescence signal corresponding to Ca2þ

release was consistent with transporter/channel activity.The rate was higher for A116V moPrP than for WT moPrP.A log-log plot of the variation of the initial rate of Ca2þ

release with protein concentration was linear with a slope of1 for both moPrP variants (Fig. 1 b, inset). Had the rate-limiting step been oligomerization of the protein at the mem-brane before membrane insertion and channel formation, ahigher-order dependence on protein concentration wouldhave been expected. This suggested that the rate-limitingstep in channel formation involves a single protein molecule,possibly inserting into the membrane or undergoing a confor-mational change. After the rate-limiting step, the protein islikely to oligomerize rapidly to form channels in the mem-brane (see Discussion).

It should be noted that DLS measurements indicated thatboth WT and A116V moPrP remained monomeric in thebuffer used for the measurements shown in Fig. 1 in theabsence of vesicles, even when the protein concentrationwas 25mM(Fig. S3, b and d). Size-exclusion chromatography

Ion Channel Formation by Prion Protein

experiments (Fig. S3 d) confirmed that the protein existed as amonomer in solution in the absence of vesicles.

WT and A116V moPrP induce calcium release byforming ion channels in lipid membranes

Ca2þ release from liposomes can occur due to eithertransporter/channel activity or membrane disruption. Thedefinitive experiment to distinguish between these two pos-sibilities is to detect currents passing through individualchannels in planar membranes (BLM). Current steps corre-sponding to the transient formation of single channels of10–20 pS conductance were detected across membranesconsisting of equal parts of DPhPC and cholesterol in5 mM HEPES buffer containing 0.25 M CaCl2, with bothWT and A116V proteins (Fig. 2). A116V moPrP formedchannels within 30 min of addition into the BLM chambereven at 50 nM, at which concentration WT moPrP did notshow any channel formation activity over a period of 12 h(Fig. 2 a). WT moPrP did form channels at higher proteinconcentrations, and did so within 30 min at concentrationsofR250 nM (Fig. 2 b). Currents were also observed in sym-metric 1 M KCl and 5 mM CaCl2, with 50 nM A116V or250 nM WT moPrP (Figs. 2, 3, and S4).

Single-channel currents were observed over a wide rangeof voltages, indicating that they are not voltage gated. Cur-rents through single channels were found to vary linearlywith voltage regardless of whether the conducting ion wasKþ or Ca2þ (Figs. 3, 4, and S4). This observation of Ohmicconductance suggests that the channels were stable, becauseotherwise, nonlinear current-voltage (I-V) plots would havebeen observed. The conductance characteristics were alsofound to be highly reproducible. The slope conductancesfor potassium (1 M) and calcium (250 mM) ions were16 5 1.8 pS and 20 5 2.1 pS for A116V moPrP, and10 5 1.2 pS and 11 5 1.6 pS for WT moPrP, respectively

(Figs. 3, S4, and S5), indicating a fourfold preference forCa2þ over Kþ in both sets of channels. Interestingly, themean open time of the channels was very long, around300 ms for WT and 500 ms for A116V channels in symmet-ric 1 M KCl (Fig. S6). The extraordinarily long open timesalso suggest that conducting channels, once formed, havevery high kinetic stability.

WT and A116V moPrP ion channels conduct KD

and Ca2D ions, but not NaD and Cl� ions

Surprisingly, no opening and closing events were detectedeven 12 h after addition of the proteins into a BLM chamberwith symmetric 1 M NaCl (Figs. 3, e and f, and S4 e). Thelack of channel events could be due either to a failure toform channels in the lipid membrane in the presence ofNaCl or to an inability of moPrP channels to conduct so-dium and chloride ions (Fig. 3, e and f). WT and A116VmoPrP ion channel formation was carried out in symmetric1 M KCl, after which the buffer in the trans chamber wasreplaced with a buffer containing a mixture of 0.5 M KCland 0.5 M NaCl. The slope conductance at positive andnegative potentials was 15 pS and 8 pS, respectively—a ra-tio of close to 2. In addition, the reversal potential for thecurrents was 27 mV, which is close to the Nernst potentialfor Kþ in this configuration (Figs. 4, a and b, and S7 a).

To further confirm this ion-selective feature of moPrPchannels, current recordings were conducted in an asym-metric buffer configuration with 1 M KCl on the cis sideand 1 M NaCl on the trans side. Proteins were added intothe cis side of the chamber, where 1 M KCl was present(see Supporting Materials and Methods). Current recordingswere conducted 1 h after addition of proteins into the BLMchamber. Channel events were present upon application ofpositive potentials, but not negative potentials (Figs. 4, cand d, and S7 c). These experiments demonstrated that

FIGURE 2 moPrP forms ion channels in lipid

membranes. Current was recorded at þ70 mV

and �70 mV 1 h after addition of 50 nM WT

moPrP (a), 250 nM WT moPrP (b), 50 nM

A116V moPrP (c), or 250 nM A116V moPrP (d)

to a BLM chamber with 5 mM HEPES buffer

(pH 7.4) containing 0.25 M CaCl2. BLM composi-

tion: DPhPC/cholesterol (1:1 w/v). Except for

panel (a), all panels show 30 s current traces. Panel

(a) shows a full trace (60 s) of current recording

with 50 nMWT-moPrP. The spikes near the begin-

ning and end of the recordings are due to capacitive

transients. The insets in panel (a) show expanded

sections of the recordings. For traces acquired

at þ70 mV, upward deflections indicate channel

opening, whereas at �70 mV downward deflec-

tions indicate channel opening.

Biophysical Journal 110, 1766–1776, April 26, 2016 1769

FIGURE 3 A116V moPrP forms ion channels

that conduct Kþ and Ca2þ ions. Current traces

from A116V channels in DPhPC/cholesterol (1:1)

membranes held at þ 70 mV and �70 mV. (a, c,

and e) Symmetrical buffer compositions of 1 M

KCl (a), 0.25 M CaCl2 (c) and 1 M NaCl (e)

were used. (b, d, and f) I-V plots for the lowest-

conductance state observed in the corresponding

recordings to the left (see Supporting Materials

and Methods). The single-channel conductances

calculated from the data in panels (b) and (d) are

16 5 1.8 pS and 20 5 2.1 pS, respectively. For

traces acquired at þ70 mV, upward deflections

indicate channel opening, whereas at �70 mV,

downward deflections indicate channel opening.

Sabareesan et al.

preinserted channels are refractory to the passage of Naþ

and Cl� ions.Reagents that are known to block voltage-gated ion

channels were tested for their ability to affect moPrPchannels. La3þ, which is a known blocker of Ca2þ channels(36), blocked Ca2þ release from liposomes (Fig. 1, c andd), as well as both Ca2þ and Kþ currents in BLM mea-surements (Figs. S8 and S9). 4-AP, a known blocker ofKþ channels (37), blocked both Kþ and Ca2þ currents inBLM measurements (Figs. S8 and S9), as well as Ca2þ

1770 Biophysical Journal 110, 1766–1776, April 26, 2016

release from liposomes (Fig. 1, c and d). TEA, a blockerof Kþ channels, significantly reduced the rate of Ca2þ

release from liposomes, whereas tetrodotoxin, a blockerof voltage-gated Naþ channels, had little effect. The obser-vation that both La3þ and 4-AP blocked movement ofKþ and Ca2þ suggested that both ions flow through thesame channel created by the insertion of moPrP into themembrane.

It should be noted that DLS measurements showedthat both WT and A116V moPrP remained native and

FIGURE 4 A116V moPrP forms an ion-selec-

tive channel. (a and c) Current traces from

A116V channels (50 nM) in DPhPC/cholesterol

(1:1) membranes held at þ 70 mV and �70 mV

in 5 mM HEPES (pH 7.4) buffer containing (a)

1.0 M KCl and 5 mM CaCl2 (cis side) and 0.5 M

KCl, 0.5 M NaCl, and 5 mM CaCl2 (trans side),

and (c) 1.0 M KCl and 5 mM CaCl2 (cis side)

and 1 M NaCl and 5 mM CaCl2 (trans side).

(b and d) I-V plots for the lowest-conducting state

in the recordings shown in panels (a) and (c),

respectively. The intercept in panel (b) is at

27 mV, close to the calculated Nernst potential

for Kþ in this configuration. In panel (c), the initial

spike is due to a capacitive transient. The inset

shows expanded sections of the recordings. For

traces acquired at þ70 mV, upward deflections

indicate channel opening, whereas at �70 mV

downward deflections indicate channel opening.

Ion Channel Formation by Prion Protein

monomeric in the buffers used for BLM measurements ob-tained in the absence of any lipid (Fig. S10).

The central hydrophobic region of moPrP isimportant for channel formation

HDX-MS studies were carried out in the presence andabsence of liposomes to obtain sequence-specific informa-tion about the structural changes brought about in WT andA116V moPrP upon interaction with lipid membranes. Seg-ments of the protein that interact with the lipid membranemay be expected to be shielded from solvent and exchangeof amide protons. The location of protected hydrogens canbe determined by proteolytic digestion followed by MSusing a previously established peptide map (38). In theabsence of liposomes, A116V moPrP showed a pattern ofdeuterium incorporation very similar to that of WT moPrP(Fig. 5), indicating that the A116V mutation does not induceany major structural changes in moPrP.

FIGURE 5 Deuterium incorporation at 30 s into different sequence seg-

ments of WT and A116V moPrP in the presence and absence of liposomes

at 25�C, pH 7.0. The percent deuterium incorporation into each peptic frag-

ment was determined from its mass calculated from the overall centroid of

the isotopic envelope in the mass spectrum, relative to the mass of the cor-

responding peptide in 90% D2O (see Supporting Materials and Methods).

The amino acid residue 22 at the N-terminus is Met. Error bars represent

the standard error in the data from two independent experiments.

Both WTand A116V moPrP showed increased deuteriumincorporation by ~25% in helix 1 and decreased deuteriumincorporation by ~25% in helix 2 in the presence of lipo-somes compared with buffer alone (Fig. 5). The abovechanges are signatures of oligomer formation: oligomersof moPrP have been shown to exhibit increased deuteriumincorporation in the helix 1 region along with reduceddeuterium incorporation in helix 2 regions (38). Hence, itappears that in the presence of liposomes, both WT andA116V moPrP are in a conformation similar to that ofoligomers. It should be noted that in the absence of lipo-somes, both WT and A116V moPrP are monomeric at theconcentrations used here (Fig. S3).

Interestingly, the sequence segment 109–132 showed abimodal mass distribution for WT and A116V moPrP in thepresence of liposomes, whereas it showed a unimodal massdistribution in solution (Fig. 6). A bimodal mass distributionin a sequence segment is indicative of conformational hetero-geneity in the population of proteinmolecules with respect tothat sequence segment (38,39). In one subpopulation ofmoPrPmolecules, the rate ofHDX into the sequence segment

FIGURE 6 The sequence segment 109–132 in WT and A116V moPrP

shows protection against HDX in the presence of liposomes. Mass spectra

for the sequence segment 109–132 for WT and A116V moPrP at 30 s of

deuterium labeling show bimodal mass distributions in the presence of lipo-

somes, compared with the unimodal distribution seen in the absence of li-

posomes. The controls of protonated (0% D) and deuterated (90% D)

peptide fragments are also shown. The m/z of the þ33 charge state of the

intact protein in the case of A116V moPrP is seen to partially overlap

with the m/z of the sequence segment 109–132. In the presence of lipo-

somes, the area under the lower mass (protected) peak for the sequence

segment 109–132 is 5% and 35% in WT and A116V moPrP, respectively.

Biophysical Journal 110, 1766–1776, April 26, 2016 1771

Sabareesan et al.

109–132 is significantly lower than in the remaining popu-lation, suggesting that it is structured or physically seques-tered away from the solvent due to its interaction withlipid membranes. Consequently, the two subpopulations ofsequence segment 109–132 have different masses, leadingto a bimodal distribution. Although the sequence segment109–132 showed a bimodal mass distribution for bothWT and A116V moPrP, the fraction of molecules thatwere protected against HDX was significantly higher forA116V moPrP (~35%) than for WT moPrP (~5%) (Fig. 6).These observations indicated that the partitioning of proteinmolecules into lipid membranes via the sequence segment109–132 was greater in the case of A116V moPrP than inthe case of WT moPrP, and provided a rationale for theincreased channel-forming activity of A116V moPrP.

It should be noted that the fluorescence binding data(Fig. 1 a) suggest that all protein molecules would be boundto liposomes at the concentrations used for the HDX experi-ments (see SupportingMaterials andMethods). It is thereforesurprising that the HDX measurements indicate that only~5% of WT moPrP molecules and ~35% of A116VmoPrPmolecules are bound to the liposomes in a manner thatprotects segment 109–132 from HDX. An alternate interpre-tation is that in both cases, all protein molecules are indeedbound to the liposomes in a manner that protects segment109–132 against HDX, but this segment has a higher pro-tection factor (39) upon interaction with a lipid membranein the case of A116V moPrP than in the case of WT moPrP.

C-terminal domain of moPrP does not form anyion channel

The HDX-MS experiments indicated that it is the central hy-drophobic region (sequence segment 109–132) that interactswith the lipid membrane, and that this segment may beburied inside the lipid membrane. It is likely that it is thisregion that is involved in ion channel formation. In earlierstudies, it was shown that a peptide (residues 105–126)derived from the central hydrophobic region forms ion chan-

1772 Biophysical Journal 110, 1766–1776, April 26, 2016

nels in lipid membranes (25). To confirm the importanceof the central hydrophobic region, BLM experiments werecarried out using the C-terminal domain of moPrP (CTD-moPrP) (residues 121–231), which lacks most of the centralhydrophobic region. As expected, CTD-moPrP did not showany channel activity in BLM experiments (Fig. S11), indi-cating that ion channel formation requires the central hydro-phobic region of the protein or the preceding unstructuredregion.

WT and A116VmoPrP elicit ionic currents in livingcells

Cells were exposed to full-length proteins while patch-clamped in the whole-cell voltage-clamp configuration tomeasure plasma membrane conductance. Channel openingand closing events would be undetectable using otherconfigurations of the patch clamp, since the single-channelconductance, as estimated based on the BLM experiment,would be too low to measure in physiological buffer.Fig. 7 a shows whole-cell current recordings from cells sub-jected to sweeps of transmembrane potential from�100 mVto þ100 mV. At þ60 mV, compared with the control cells,the current increased fourfold in cells exposed to 0.5 mMWT moPrP, and by sixfold in cells exposed to A116Vprotein at the same concentration, a 50% difference in effi-cacy. The corresponding increases in cells exposed to 1 mMmoPrP were 10-fold and 12-fold, respectively.

A detailed analysis of the I-V traces in Fig. 7 a indicatedthat the reversal potential of the currents seen in the pres-ence of added protein was more negative than that in controlcells. This brings the resting membrane potential closer tothe Nernst potential for Kþ, suggesting that the enhancedconductance was Kþ selective. Treatment with the Kþ chan-nel blocker 4-AP resulted in a substantial reduction of therecorded currents (Fig. 7 b).

It should be noted that CD and DLS measurements indi-cated that both WT and A116V moPrP remained native andmonomeric in the buffer used for the measurements shown

FIGURE 7 Whole-cell patch-clamp recordings

were made from HEK 293T cells treated with

moPrP variants. (a) I-V plots for cells that were

treated with recording buffer (control) or different

concentrations of WT moPrP or A116V moPrP for

a period of 1 h. (b) I-V plots for cells treated with

moPrP variants in the presence and absence of the

potassium channel blocker 4-AP (see Supporting

Materials and Methods). Both panels present the

mean 5 standard deviation calculated from inde-

pendent measurements performed across six to

eight cells.

Ion Channel Formation by Prion Protein

in Fig. 7, in the absence of cells, even when the protein con-centration was 20 mM (Fig. S12).

PrP appears to insert into the membrane in itsmonomeric native form

A specific concern was whether the native monomeric pro-tein had become contaminated with very small amounts ofoligomer during its preparation, and whether it was the olig-omer that interacted with and formed ion channels in thelipid membrane. In this study, native moPrP was shown byboth DLS and size-exclusion chromatography measure-ments to be monodisperse and monomeric in the buffersused for the liposome experiments (data not shown), theBLM experiments (Fig. S10), and the cell electrophysiologyexperiments (Fig. S12). To further confirm the absence ofany oligomer, native moPrP was first unfolded in 8 Murea (which would also disaggregate any oligomer present)and refolded (see Supporting Materials and Methods), andthen shown by fluorescence measurements to lack the abilityto bind to the hydrophobic dye ANS (Fig. S3 a). ANS isknown to bind to prion oligomers even when the oligomersare present at very low relative amounts along with mono-mer (40,41). Fig. S3 a indicates, for example, that even50 nM oligomer would be detectable by the ANS bindingassay. Monomeric protein treated in this way possessedthe same ability to interact with lipid membranes as did pro-tein that had not been so treated (data not shown). Thus, itappears that the ability of moPrP to associate with lipidmembranes is a property of the monomeric native protein,although it cannot be ruled out that oligomerization andmembrane insertion are induced at the membrane-solutioninterface. It should also be noted that several aspects ofthe data suggest that the initial interaction between the pro-tein and the lipid membrane involves monomeric protein. Inthe calcium efflux experiments (Fig. 1 b), the plot of the log-arithm of the initial rate versus the logarithm of monomerconcentration had a slope of 1, indicating that monomerinsertion into the liposome was the rate-limiting step inchannel formation. In the BLM experiments, discrete uni-tary channel opening events were observed (Figs. 2, 3, and4), whereas oligomers formed by WT moPrP, even at a con-centration of 100 nM, have been shown to rupture the BLM(30) and to not elicit discrete unitary currents. It appears thatthe protein oligomerizes to form channels only after it asso-ciates with the membrane as a monomer.

A second concern was whether the monomeric native pro-tein was heterogeneous and contained multiple subpopula-tions of monomers, of which only one subpopulation wasable to interact with lipid membrane. In fact, it has been re-ported that monomeric PrP can be made to exist in multipleconformations, and one such conformation was shown tobe highly toxic to cultured cell lines (42). The HDX-MS ex-periments (Fig. 5), however, did not provide any evidence ofconformational heterogeneity in the monomer of either WT

or A116V moPrP in the absence of liposomes, as they did inthe case of oligomers and fibrils in previous studies (39,43).NMR studies of native moPrP also provide no evidence ofmore than one monomeric conformation (44).

DISCUSSION

The primary motivation behind this work was to obtain abetter understanding of the interaction of PrP with lipidmembranes, in light of the strong correlation that has beenobserved between such interactions and pathogenesis in thecase of prion diseases that do not involve the formation ofaggregates of PrPSc (45). In particular, it is known that thepathogenic mutation A117V in the central hydrophobicregion of human PrP converts normal PrP into a transmem-brane form, and induces neurodegenerative symptoms in theabsence of any significant accumulation of protease-resistantaggregated PrP (15,16). Hence, one goal was to understandhow this mutation might lead to neurodegeneration inducedby either the transmembrane or soluble form of the protein.

moPrP forms discrete channels in artificial andcell membranes, selective for KD and Ca2D

This study establishes that soluble full-length moPrPinteracts with membranes, including cell membranes, andpermeabilizes them. Had the enhanced permeability beendue to a detergent-like rupture of the membrane, or to theformation of a nonselective pore, the effect on the restingmembrane potential would have been to shift it toward0 mV. The observed shift in the resting membrane potentialof affected cells toward the Nernst potential of Kþ suggeststhe formation of a Kþ-selective leak across the plasmamembrane. Indeed, 4-AP, a Kþ channel blocker, partiallyblocked prion-induced currents observed in the whole-cellpatch-clamp experiment. A more detailed characterizationin artificial membrane systems demonstrates that the pro-teins form well-defined channels that are permeable to Kþ

and Ca2þ and can be blocked by 4-AP and TEA, whichblock Kþ channels, and by La3þ, which blocks Ca2þ chan-nels. Tetrodotoxin, a blocker of voltage-gated Naþ channels,has a limited effect. In the isolated planar bilayer system, thechannels are Ohmic in character, whereas the currentsinduced in cells have nonlinear I-V characteristics, suggest-ing either different behaviors in the two systems or theinduction of endogenous channels in cells. A116V moPrP,which has a mutation in the central hydrophobic region ofthe protein that favors formation of the transmembraneform of the protein, was much more efficacious in bindingto membranes and forming channels in all three systemsstudied. Interestingly, the same region of the pathogenictransmembrane form of PrP has been shown to interactwith cell membranes (15). Hence, it appears that the patho-genic transmembrane form of PrP might act by formingchannels in cell membranes.

Biophysical Journal 110, 1766–1776, April 26, 2016 1773

Sabareesan et al.

moPrP can release Ca2þ from liposomes. Inhibition ofthis release by specific blockers is diagnostic of a trans-porter-mediated mechanism of release. The observation ofdiscrete current steps of uniform conductance in the BLMconfirms that the transporter is a channel (Figs. 1, 2, 3, 4,S1, S4, and S7). Elimination of conductance fluctuationsby the same reagents confirms a channel origin as opposedto transient membrane defects causing permeability.Although both WT and A116V moPrP form channels, thechannel-forming propensity is higher for the mutant variantcompared with WT moPrP (Figs. 1 and 2). As shown byHDX-MS studies (Fig. 6), the increased channel activityof A116V moPrP is correlated with its enhanced propensityto insert into lipid membranes. The lack of CTD-moPrPactivity suggests that the unstructured N-terminal region(residues 23–120) is important for lipid membrane bindingas well as ion channel formation (Fig. S11).

Recently, it was reported that the interaction of lipid-anchored forms of PrP with membranes is very differentfrom that observed for anchor-free forms (46). These differ-ences could well affect the channel-formation activity,limiting extension of our results to GPI-anchored forms.Indeed, when GPI-anchored WT PrPC is overexpressed intransgenic mice, it does not appear to induce neurotoxicity.Only when PrPC was expressed as an anchorless protein intransgenic mice did it appear to be capable of inducingtoxicity (9). Nevertheless, when pathogenic mutant variantsof PrP are expressed in cells with the GPI anchors intact,they have been reported to induce spontaneous ionic cur-rents as detected by whole-cell patch-clamp measurements(47). Hence, channel-forming activity could be an inherentproperty of PrP irrespective of anchorage.

Nature of the PrP-membrane interaction

BLM experiments show homogeneous single-channelconductance values for both WT and A116V moPrP, afeature that was not observed in earlier studies using PrPpeptides and recombinant PrP aggregates (22). The obser-vation of homogeneous single-channel conductances dem-onstrates that the responsible channels are formed by asingle kind of PrP assembly in the lipid membrane.

HDX-MS studies demonstrate that, at least in the case ofA116V moPrP, the lipid membrane protects the sequencesegment 109–132 or a part of it. A monomeric oligopeptideof 20–22 residues is very unlikely to form a transmembranechannel. Hence, oligomerization would be necessary toform the channels. Although soluble oligomers formed byPrP can interact with and permeabilize lipid membranes(38,48), the absence of any preformed oligomeric form insolution under the measurement conditions used here(Figs. S10 and S12) suggests that PrP can form channelswith discrete conductances in lipid membranes even whenpresent in the monomeric form in solution. The otherwiseunstructured sequence segment 109–132 must form second-

1774 Biophysical Journal 110, 1766–1776, April 26, 2016

ary structure and oligomerize when inserted into the lipidmembrane.

The HDX-MS data are consistent with a restructuringand/or sequestration of the central hydrophobic regionin the presence of lipid. Indeed, this sequence is highlyconserved among all mammalian PrPs and has been shownto serve as a transmembrane anchor in certain scenarios(1,15). Peptides comprising residue stretches 105–126,118–135, and 105–135, derived from the central hydrophobicregion, have all been shown to interact with lipid membranes(22,24,26–28). At least in vitro, peptides comprising residuestretches 105–126 and 82–146 of PrP have shown channel-forming abilities in different lipid membranes (21,25).

Previously, it was reported that a PrP lacking the centralhydrophobic region formed pores upon expression in celllines, and that the formation of these pores was due to thetransient interaction of the N-terminal positively chargedmotif (23KKRPK27) with the lipid membrane (47,49). How-ever, the HDX-MS data reported here do not show anyevidence of a significant change in the solvent accessibilityof this segment in the intact protein in the presence of lipid(Fig. 7). It is conceivable that transient interactions may notbe picked up by this technique.

The structure of PrP ion channels may be similarto those formed by pore-forming toxins

A clue as to how PrP might form ion channels is provided bythe pore-forming toxins. These toxins can exist either in astable water-soluble state or as integral membrane proteins(50), a property that appears to be shared by PrP. The mem-brane-integrated structures of pore-forming proteins may beeither a-helical or b-barrel (50,51). That of PrP appears tobe principally a-helical (Fig. S2), although the conversionof short (10%) sequence stretches into b-structure wouldnot be detectable (see Results). The membrane-inserting re-gions of some toxins form a-helical hairpins that are 40–60residues long, whereas those of other toxins form b-hairpinsthat are 15–25 residues long (52,53). The sequence stretch109–132 in PrP, which shows protection against HDX inthe presence of lipid membranes, is of the right length toform a membrane-spanning b-hairpin, but not an a-helicalhairpin. Importantly, it contains seven glycine residues,including two XGGX motifs that are thought to be impor-tant in lipid-membrane interactions (54), such as in the for-mation of a b-hairpin in the membrane.

In the case of the pore-forming toxin a-hemolysin, acentral sequence stretch of ~27 residues, which is rich inglycine residues, forms a b-hairpin, and seven protein mol-ecules oligomerize to form a 14-stranded b-barrel pore inthe membrane (55). It is tempting to postulate that PrP toointeracts with lipid membrane and forms a pore in the mem-brane in a similar manner. It should be noted that althoughthe pores may form through association of b-hairpins as aconsequence of protein oligomerization in both cases, the

Ion Channel Formation by Prion Protein

rest of the a-hemolysin structure is also b-sheet (55),whereas PrP appears to maintain a largely helical structure(Fig. S2). The conversion of 10% of the sequence intob-hairpin would not be detectable in CD or FTIR spectra.The mean open times of the channels formed by PrP areunusually long (Fig. S4), although not as long as thoseformed by a-hemolysin (56). The substantially shortermean open times and smaller conductances of the PrP chan-nels, as compared with the a-hemolysin channels, makes itunlikely that PrP has toxin-like behavior.

It is known that hydrophobic interactions between thehydrophobic surfaces of sequence stretches from the proteinand the zwitterionic phospholipids on the membrane surfaceplay a major role in protein-lipid binding (57). The mutationAla116 to Val116 in the central hydrophobic region increasesthe hydrophobic nature of the sequence and introduces abranched amino acid residue with a high b-sheet propensityin place of a residue with a high a-helix propensity. Itis likely that the binding of both WT and A116V moPrPincreases the local protein concentration at the membrane,which favors oligomerization within the lipid membrane.The slope of 1 of the log-log plot of the calcium releaserate versus protein concentration suggests that the rate-limiting step in channel formation could be either a confor-mational transition in the monomer or its insertion into themembrane (Fig. 3). Oligomerization of the inserted stretchof the protein would then lead to channel formation.

In the case of both A116V and WT moPrP, both the cal-cium and potassium conductances are intermediate betweenthe conductances of voltage-gated Naþ and Kþ channels(20–50 pS) and store-operated calcium channels (~20 fS)(58). On the other hand, the mean open time for the prionchannels is a few hundred milliseconds (Fig. S4), in contrastto, say, voltage-gated Naþ channels, which have mean opentimes of a few milliseconds. The mean channel conductanceis independent of voltage (Fig. S6), as is the probabilityof channel opening, implying that significant amounts ofCa2þ can enter during a single opening at resting membranepotentials.

In conclusion, this study conducted with recombinantmoPrP suggests that the ability to form an ion channel is aninherent property of PrP, and that the propensity to formchannels is increased by a pathogenic mutation in the centralhydrophobic region. It is thus conceivable that mutant PrPswith a high propensity to form channels could mediate suffi-cient influx of Ca2þ to be pathogenic. Studies are currentlyunder way to establish whether monomeric PrP can be toxicto cells, at least under some conditions, and whether suchtoxicity is linked to the ability of the protein to elicit calciumcurrents in cell membranes.

SUPPORTING MATERIAL

SupportingMaterials and Methods and twelve figures are available at http://

www.biophysj.org/biophysj/supplemental/S0006-3495(16)30068-6.

AUTHOR CONTRIBUTIONS

A.T.S., J.S., S.R., J.B.U., and M.K.M. designed the experiments and wrote

the manuscript. A.T.S., J.S., and S.R. carried out the experiments and

analyzed the data.

ACKNOWLEDGMENTS

We thank members of our laboratories for discussions.

J.B.U. is a recipient of a JC Bose National Research Fellowship from the

Government of India. This work was funded by the Tata Institute of Funda-

mental Research and the Department of Biotechnology, Government of

India.

REFERENCES

1. Aguzzi, A., F. Baumann, and J. Bremer. 2008. The prion’s elusivereason for being. Annu. Rev. Neurosci. 31:439–477.

2. Biasini, E., J. A. Turnbaugh,., D. A. Harris. 2012. Prion protein at thecrossroads of physiology and disease. Trends Neurosci. 35:92–103.

3. Caughey, B., and P. T. Lansbury. 2003. Protofibrils, pores, fibrils, andneurodegeneration: separating the responsible protein aggregatesfrom the innocent bystanders. Annu. Rev. Neurosci. 26:267–298.

4. Harrison, C. F., K. J. Barnham, and A. F. Hill. 2007. Neurotoxic speciesin prion disease: a role for PrP isoforms? J. Neurochem. 103:1709–1720.

5. Solomon, I. H., J. A. Schepker, and D. A. Harris. 2010b. Prion neuro-toxicity: insights from prion protein mutants. Curr. Issues Mol. Biol.12:51–61.

6. Mironov, A., Jr., D. Latawiec, ., P. J. Peters. 2003. Cytosolic prionprotein in neurons. J. Neurosci. 23:7183–7193.

7. Ma, J., R. Wollmann, and S. Lindquist. 2002. Neurotoxicity and neuro-degeneration when PrP accumulates in the cytosol. Science. 298:1781–1785.

8. Park, K. W., and L. Li. 2008. Cytoplasmic expression of mouse prionprotein causes severe toxicity in Caenorhabditis elegans. Biochem. Bio-phys. Res. Commun. 372:697–702.

9. Thackray, A. M., C. Zhang, ., R. Bujdoso. 2014. Cytosolic PrP canparticipate in prion-mediated toxicity. J. Virol. 88:8129–8138.

10. Hay, B., S. B. Prusiner, and V. R. Lingappa. 1987. Evidence for asecretory form of the cellular prion protein. Biochemistry. 26:8110–8115.

11. Parkin, E. T., N. T. Watt,., N. M. Hooper. 2004. Dual mechanisms forshedding of the cellular prion protein. J. Biol. Chem. 279:11170–11178.

12. Taylor, D. R., E. T. Parkin,., N. M. Hooper. 2009. Role of ADAMs inthe ectodomain shedding and conformational conversion of the prionprotein. J. Biol. Chem. 284:22590–22600.

13. Bessen, R. A., J. M. Wilham, ., J. A. Wiley. 2012. Accelerated shed-ding of prions following damage to the olfactory epithelium. J. Virol.86:1777–1788.

14. Altmeppen, H. C., J. Prox, ., M. Glatzel. 2013. Roles of endoproteo-lytic a-cleavage and shedding of the prion protein in neurodegenera-tion. FEBS J. 280:4338–4347.

15. Hegde, R. S., J. A. Mastrianni, ., V. R. Lingappa. 1998. A transmem-brane form of the prion protein in neurodegenerative disease. Science.279:827–834.

16. Hegde, R. S., P. Tremblay,., V. R. Lingappa. 1999. Transmissible andgenetic prion diseases share a common pathway of neurodegeneration.Nature. 402:822–826.

17. Aguzzi, A., C. Sigurdson, and M. Heikenwaelder. 2008b. Molecularmechanisms of prion pathogenesis. Annu. Rev. Pathol. 3:11–40.

Biophysical Journal 110, 1766–1776, April 26, 2016 1775

Sabareesan et al.

18. Caughey, B., G. S. Baron, ., M. Jeffrey. 2009. Getting a grip onprions: oligomers, amyloids, and pathological membrane interactions.Annu. Rev. Biochem. 78:177–204.

19. Wang, X., F. Wang, ., J. Ma. 2006. The interaction between cyto-plasmic prion protein and the hydrophobic lipid core of membranecorrelates with neurotoxicity. J. Biol. Chem. 281:13559–13565.

20. Tateishi, J., T. Kitamoto,., N. Heldt. 1990. Immunochemical, molec-ular genetic, and transmission studies on a case of Gerstmann-Strauss-ler-Scheinker syndrome. Neurology. 40:1578–1581.

21. Bahadi, R., P. V. Farrelly,., M. Salmona. 2003. Channels formed witha mutant prion protein PrP(82-146) homologous to a 7-kDa fragment indiseased brain of GSS patients. Am. J. Physiol. Cell Physiol.285:C862–C872.

22. Kagan, B. L., R. Azimov, and R. Azimova. 2004. Amyloid peptidechannels. J. Membr. Biol. 202:1–10.

23. Quist, A., I. Doudevski, ., R. Lal. 2005. Amyloid ion channels: acommon structural link for protein-misfolding disease. Proc. Natl.Acad. Sci. USA. 102:10427–10432.

24. Alier, K., Z. Li,., J. H. Jhamandas. 2010. Ionic mechanisms of actionof prion protein fragment PrP(106-126) in rat basal forebrain neurons.J. Neurosci. Res. 88:2217–2227.

25. Lin, M. C., T. Mirzabekov, and B. L. Kagan. 1997. Channel formationby a neurotoxic prion protein fragment. J. Biol. Chem. 272:44–47.

26. Pillot, T., L. Lins, ., R. Brasseur. 1997. The 118-135 peptide of thehuman prion protein forms amyloid fibrils and induces liposomefusion. J. Mol. Biol. 274:381–393.

27. O’Donovan, C. N., D. Tobin, and T. G. Cotter. 2001. Prion protein frag-ment PrP-(106-126) induces apoptosis via mitochondrial disruption inhuman neuronal SH-SY5Y cells. J. Biol. Chem. 276:43516–43523.

28. Chabry, J., C. Ratsimanohatra, ., T. Pillot. 2003. In vivo and in vitroneurotoxicity of the human prion protein (PrP) fragment P118-135independently of PrP expression. J. Neurosci. 23:462–469.

29. Sonkina, S., I. I. Tukhfatullina, ., J. Cladera. 2010. Interaction of theprion protein fragment PrP 185-206 with biological membranes: effecton membrane permeability. J. Pept. Sci. 16:342–348.

30. Taylor, D. R., and N. M. Hooper. 2006. The prion protein and lipidrafts. Mol. Membr. Biol. 23:89–99.

31. Jain, S., and J. B. Udgaonkar. 2008. Evidence for stepwise formation ofamyloid fibrils by the mouse prion protein. J. Mol. Biol. 382:1228–1241.

32. Moulick, R., and J. B. Udgaonkar. 2014. Thermodynamic characteriza-tion of the unfolding of the prion protein. Biophys. J. 106:410–420.

33. Godbole, A., A. K. Dubey, ., M. K. Mathew. 2013. MitochondrialVDAC and hexokinase together modulate plant programmed celldeath. Protoplasma. 250:875–884.

34. Montal, M., and P. Mueller. 1972. Formation of bimolecular mem-branes from lipid monolayers and a study of their electrical properties.Proc. Natl. Acad. Sci. USA. 69:3561–3566.

35. Wang, F., F. Yang, ., J. Ma. 2007. Lipid interaction converts prionprotein to a PrPSc-like proteinase K-resistant conformation underphysiological conditions. Biochemistry. 46:7045–7053.

36. Nathan, R. D., K. Kanai,., W. Giles. 1988. Selective block of calciumcurrent by lanthanum in single bullfrog atrial cells. J. Gen. Physiol.91:549–572.

37. Thompson, S. 1982. Aminopyridine block of transient potassium cur-rent. J. Gen. Physiol. 80:1–18.

38. Singh, J., A. T. Sabareesan,., J. B. Udgaonkar. 2012. Development ofthe structural core and of conformational heterogeneity during the con-version of oligomers of the mouse prion protein to worm-like amyloidfibrils. J. Mol. Biol. 423:217–231.

1776 Biophysical Journal 110, 1766–1776, April 26, 2016

39. Singh, J., and J. B. Udgaonkar. 2013. Dissection of conforma-tional conversion events during prion amyloid fibril formation usinghydrogen exchange and mass spectrometry. J. Mol. Biol. 425:3510–3521.

40. Gerber, R., A. Tahiri-Alaoui, ., W. James. 2007. Oligomerization ofthe human prion protein proceeds via a molten globule intermediate.J. Biol. Chem. 282:6300–6307.

41. Bjorndahl, T. C., G. P. Zhou,., D. S. Wishart. 2011. Detailed biophys-ical characterization of the acid-induced PrP(c) to PrP(b) conversionprocess. Biochemistry. 50:1162–1173.

42. Zhou, M., G. Ottenberg, ., C. I. Lasmezas. 2012. Highly neurotoxicmonomeric a-helical prion protein. Proc. Natl. Acad. Sci. USA.109:3113–3118.

43. Singh, J., and J. B. Udgaonkar. 2015. Structural effects of multiplepathogenic mutations suggest a model for the initiation of misfoldingof the prion protein. Angew. Chem. Int. Ed. Engl. 54:7529–7533.

44. Moulick, R., R. Das, and J. B. Udgaonkar. 2015. Partially unfoldedforms of the prion protein populated under misfolding-promoting con-ditions: characterization by hydrogen exchange mass spectrometry andNMR. J. Biol. Chem. 290:25227–25240.

45. Chesebro, B., M. Trifilo, ., M. Oldstone. 2005. Anchorless prionprotein results in infectious amyloid disease without clinical scrapie.Science. 308:1435–1439.

46. Chu, N. K., W. Shabbir, ., C. F. Becker. 2014. A C-terminal mem-brane anchor affects the interactions of prion proteins with lipid mem-branes. J. Biol. Chem. 289:30144–30160.

47. Solomon, I. H., J. E. Huettner, and D. A. Harris. 2010. Neurotoxic mu-tants of the prion protein induce spontaneous ionic currents in culturedcells. J. Biol. Chem. 285:26719–26726.

48. Singh, J., H. Kumar, ., J. B. Udgaonkar. 2014. Rational stabilizationof helix 2 of the prion protein prevents its misfolding and oligomeriza-tion. J. Am. Chem. Soc. 136:16704–16707.

49. Westergard, L., J. A. Turnbaugh, and D. A. Harris. 2011. A nine aminoacid domain is essential for mutant prion protein toxicity. J. Neurosci.31:14005–14017.

50. Gouaux, E. 1997. Channel-forming toxins: tales of transformation.Curr. Opin. Struct. Biol. 7:566–573.

51. Parker, M. W., and S. C. Feil. 2005. Pore-forming protein toxins: fromstructure to function. Prog. Biophys. Mol. Biol. 88:91–142.

52. Mani, R., M. Tang,., M. Hong. 2006. Membrane-bound dimer struc-ture of a beta-hairpin antimicrobial peptide from rotational-echo dou-ble-resonance solid-state NMR. Biochemistry. 45:8341–8349.

53. Rath, P., O. Bousche, ., K. J. Rothschild. 1991. Fourier transforminfrared evidence for a predominantly alpha-helical structure of themembrane bound channel forming COOH-terminal peptide of colicinE1. Biophys. J. 59:516–522.

54. McClain, M. S., H. Iwamoto, ., T. L. Cover. 2003. Essential role ofa GXXXG motif for membrane channel formation by Helicobacterpylori vacuolating toxin. J. Biol. Chem. 278:12101–12108.

55. Song, L., M. R. Hobaugh, ., J. E. Gouaux. 1996. Structure of staph-ylococcal a-hemolysin, a heptameric transmembrane pore. Science.274:1859–1866.

56. Chalmeau, J., N. Monina, ., V. Noireaux. 2011. a-Hemolysin poreformation into a supported phospholipid bilayer using cell-free expres-sion. Biochim. Biophys. Acta. 1808:271–278.

57. Gennis, R. B., and A. Jonas. 1977. Protein-lipid interactions. Annu.Rev. Biophys. Bioeng. 6:195–238.

58. Zweifach, A., and R. S. Lewis. 1993. Mitogen-regulated Ca2þ currentof T lymphocytes is activated by depletion of intracellular Ca2þ stores.Proc. Natl. Acad. Sci. USA. 90:6295–6299.