Nirav Patel1, Srinivasan Ramachandran1, Rustam Azimov2, Bruce L Kagan2, Ratnesh Lal1. 1. Departments of Bioengineering and Mechanical & Aerospace Engineering, University of California at San Diego , La Jolla, California 92093, United States. 2. Department of Psychiatry, David Geffen School of Medicine, Semel Institute for Neuroscience & Human Behavior, University of California , Los Angeles, California 90095, United States.
Abstract
Tau is a microtubule associated protein implicated in the pathogenesis of several neurodegenerative diseases. Because of the channel forming properties of other amyloid peptides, we employed planar lipid bilayers and atomic force microscopy to test tau for its ability to form ion permeable channels. Our results demonstrate that tau can form such channels, but only under acidic conditions. The channels formed are remarkably similar to amyloid peptide channels in their appearance, physical and electrical size, permanence, lack of ion selectivity, and multiple channel conductances. These channels differ from amyloid channels in their voltage dependence and resistance to blockade by zinc ion. These channels could explain tau's pathologic role in disease by lowering membrane potential, dysregulating calcium, depolarizing mitochondria, or depleting energy stores. Tau might also combine with amyloid beta peptides to form toxic channels.
Tau is a microtubule associated protein implicated in the pathogenesis of several neurodegenerative diseases. Because of the channel forming properties of other amyloid peptides, we employed planar lipid bilayers and atomic force microscopy to test tau for its ability to form ion permeable channels. Our results demonstrate that tau can form such channels, but only under acidic conditions. The channels formed are remarkably similar to amyloid peptide channels in their appearance, physical and electrical size, permanence, lack of ion selectivity, and multiple channel conductances. These channels differ from amyloid channels in their voltage dependence and resistance to blockade by zinc ion. These channels could explain tau's pathologic role in disease by lowering membrane potential, dysregulating calcium, depolarizing mitochondria, or depleting energy stores. Tau might also combine with amyloid beta peptides to form toxic channels.
Tau is a microtubule associated protein
genetically linked to dementia.
It is found in the neurofibrillary tangles of Alzheimer’s disease
where it appears aggregated as paired helical filaments.[1−4] Tau is also associated with other neurodegenerative diseases such
as fronto-temporal dementia.[5−8] Tau is intrinsically disordered proteins that is
capable of aggregation, especially in the presence of lipid membranes
and cause their disruption.[9−13] The binding is mediated by short amphipathic helices located in
the microtubule binding repeat regions of tau.[10,14−16] Although the role of tau in causing neurodegeneration
remains uncertain, it has been demonstrated to show toxicity to neurons
in vitro.[17] Tau has also been shown to
spread from neuron to neuron in the brain, exhibiting a prion-like
contagion.[18,19]Amyloid peptides and proteins associated with Alzheimer’s
disease, prion diseases, Parkinson’s disease, and others have
been demonstrated to form nonspecific ion channels in lipid membranes.[20−22] These channels have similar properties that suggest a common mechanism
of action in causing cellular pathology. Tau’s propensity to
aggregate in the presence of membranes and cause membrane disruption
led us to examine the effects of tau-441, an isoform of humantau
(2N4R, hereafter tau) on planar lipid bilayers.[23] In this report, we present the structure and activity of
tau in planar lipid bilayers as examined by electrical conductance
and high resolution atomic force microscopy studies.
Methods
Planar Lipid
Bilayer Electrical Recording
Phospholipid
bilayer membranes were formed as previously described.[24] Briefly, bilayer membranes were formed by placing
a bubble of lipid dissolved in solvent onto the end of the Teflon
tube approximately 300 μm in diameter. The chamber design allow
rapid introduction of solution into immediate proximity with the membrane
in a volume of 50 μL. For this, 1% agar salt bridges with 1
M KCl were used to connect the Ag/AgCl electrodes (E-207, Warner Instruments)
to the solutions. Voltage clamp conditions were employed in all experiments.
The side that tau was added (cis side) was taken
as ground. All voltages given refer to the voltage of the trans side. Current was recorded with an “Axopatch
C1” amplifier, and stored by “DataTrax” system
for later analysis. Membrane capacitance and resistance were monitored
frequently to ensure the integrity of the membranes.POPE, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine, and POPG, 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol),
lipids were purchased from Avanti Polar Lipids and stored at −20
°C. Lipids were dissolved in n-heptane to a
final concentration of 1.5%. Tau was dissolved in double distilled
water at 1 mg/mL concentration and stored frozen in aliquots at −20
°C. All other chemicals and tau were purchased from Sigma (St.
Louis, MO).
AFM Imaging
For AFM experiments,
tau was suspended
in MES buffer (50 mM 2-(N-morpholino) ethanesulfonic
acid, 100 mM NaCl, 0.5 mM EGTA, pH 6.8) at 1 mg/mL concentration and
stored in 10 μL aliquots at −20 °C for one time
use in experiments to avoid multiple freeze thaw cycles. 1, 2-Dioleoyl-sn-glycero-3-phosphoserine (DOPS) phospholipids were purchased
from Avanti Polar Lipids, Inc. (Alabaster, AL). Lipid vesicles were
prepared by drying the chloroform stock solution and suspending the
phospholipids in a buffer solution (50 mM MES, 100 mM NaCl, 1 mM MgCl2). The lipid solution was sonicated for 5 min to form vesicles,
and tau aliquots were added to lipid vesicles to form proteoliposomes
at a 1:100 protein–lipid ratio. The mixture was gently vortexed
and sonicated in an ice bath for 15 min.AFM samples of tau
peptide on mica were prepared by depositing 20 μL of protein
solution (0.1 mg/mL) onto freshly cleaved mica discs (12 mm diameter).
After 2 min, samples were rinsed with 1 mL of Milli-Q water, and dried
under a stream of inert gas. AFM samples of tau peptide on planar
lipid bilayers were prepared by depositing 50 μL of proteoliposome
solution onto freshly cleaved mica. Samples were incubated for 10
or 100 min and gently rinsed thrice with phosphate-buffered saline
(PBS). All samples were imaged immediately after preparation. AFM
imaging was conducted on a Multimode AFM equipped with a Nanoscope
IIIa controller (Bruker, Santa Barbara, CA). Experiments were conducted
in air or in liquid conditions in PBS using cantilevers with nominal
spring constants of 0.02 N/m. Image analysis was performed using Bruker
Nanoscope Analysis software.
Results
Ion Channel
Activity
Figure A shows the current flow across the phospholipid
bilayer upon addition of tau (32–43 μM) to one side of
the membrane (cis side) when the membrane voltage
was held at −50 mV (voltages refer to the trans side). In the absence of peptide, baseline current flow was indistinguishable
from zero (∼0.05 pA that gives ∼1 pS of conductance),
indicating that the membrane by itself is impermeable to ions in the
solution. Only stable, low conductance, low noise membranes were used
for experiments. Channel behavior exhibited voltage dependence, with
more channel openings and closings at negative voltage than at positive
voltage (Figure B).
This voltage-dependent behavior was reversible and reproducible.
Figure 1
(A) Single
channel currents induced by tau in bilayer membrane
at −50 mV; 0.43 μM tau in 10 mM Tris-citrate (pH 5.2)
was added to the lipid bilayer made up of 2:1 POPE/POPG in 100 mM
NaCl. (B) Voltage dependence of currents induced by tau.
(A) Single
channel currents induced by tau in bilayer membrane
at −50 mV; 0.43 μM tau in 10 mM Tris-citrate (pH 5.2)
was added to the lipid bilayer made up of 2:1 POPE/POPG in 100 mM
NaCl. (B) Voltage dependence of currents induced by tau.The induced currents were due to the formation
of single channels
in the membrane. Figure shows the opening and closing of single channels in the membrane.
The single channel conductance shown were not unique as multiple single
channel conductance transitions were observed (35, 53, and 121 pS),
a phenomenon common to the channel forming amyloid peptides. This
may reflect aggregation of tau monomers into aggregates of varying
number, size, and single channel conductance. Current induced only
at acidic pH 5.2, there was no activity registered at pH 7.4. A reversal
potential of 7 to 11 mV was recorded in a 10-fold gradient of NaCl,
indicating little ion selectivity between negatively and positively
charged ions. However, tau channels are not blocked by Zn2+ ions (Figure ) unlike
channels formed by other amyloid peptides, even at higher concentrations
of Zn2+ (1 mM). Channels could last for several hours in
the membrane indicating that they were irreversibly associated with
the membrane.
Figure 2
Heterodisperse single channel conductances of tau channels.
(A)
Single channel conductance of 35 pS. (B) Heterogeneous conductances
of 121, 35, and 53 pS recorded, suggesting multiple subunits continue
to form channel structures with different single channel conductances.
Figure 3
Unlike other amyloid beta channels, tau channels
are not blocked
by Zn2+ ions even at higher concentrations (1 mM).
Heterodisperse single channel conductances of tau channels.
(A)
Single channel conductance of 35 pS. (B) Heterogeneous conductances
of 121, 35, and 53 pS recorded, suggesting multiple subunits continue
to form channel structures with different single channel conductances.Unlike other amyloid beta channels, tau channels
are not blocked
by Zn2+ ions even at higher concentrations (1 mM).
Tau Forms Porelike Structures
Tau protein predominantly
forms globular oligomers in the presence of a negatively charged surface.
Contact mode AFM images of tau on mica (Figure A) show a population of tau monomers and
oligomers. The diameter and height of tau proteins on mica is 28.6
(±6.4) and 1.21 (±0.44) nm, respectively. We investigated
the structure of aggregated tau in reconstituted into lipid membranes.
Tapping mode AFM images reveal that tau predominantly forms globular
oligomers on DOPS planar bilayers (Figure C).
Figure 4
(A) AFM images of tau protein on mica reveal
monomers and oligomers
distributed on the surface. (Height scale is 2 nm). (B) DOPS bilayers
in the absence of tau protein reveal uniformly smooth bilayers with
little deviation in the surface height. (C) Tau oligomers on DOPS
lipid bilayer. Tau predominantly forms globular oligomers on the bilayer
surface. Few porelike structures are observed within the bilayer (red
circles). (D, E) Plots of line traces indicated in the AFM image intersect
a globular oligomer (D) and porelike structure (E) (indicated by asterisks).
(A) AFM images of tau protein on mica reveal
monomers and oligomers
distributed on the surface. (Height scale is 2 nm). (B) DOPS bilayers
in the absence of tau protein reveal uniformly smooth bilayers with
little deviation in the surface height. (C) Tau oligomers on DOPSlipid bilayer. Tau predominantly forms globular oligomers on the bilayer
surface. Few porelike structures are observed within the bilayer (red
circles). (D, E) Plots of line traces indicated in the AFM image intersect
a globular oligomer (D) and porelike structure (E) (indicated by asterisks).High-resolution AFM imaging of
tau interacting with lipid bilayers
reveals the presence of porelike structures within the bilayer (Figure ). Porelike structures
were identified by the presence of subunits forming a circular group
with a small height increase respective to the bilayer surface.[25] Observed structures were heterogeneous in size
and shape. The protrusion height of these structures is approximately
1.5–1.8 nm from the surface of the surrounding bilayer. The
inner diameter of these structures is approximately 2.0–3.9
nm, and average outer diameter is 15.1–20.5 nm. For comparison,
the globular oligomers of tau have an average height of 4.33 (±1.68
nm) and an average diameter of 14.6 (±4.4 nm). The three-dimensional
structures of observed porelike structures reveal four to five subunits
circularly arranged around a central channel. Few observed structures
have a distinct central depression, likely due to features being obscured
by the AFM tip morphology and local movements of subunits within the
lipid bilayer.
Figure 5
Three-dimensional representation of porelike structures
of tau
protein in a lipid bilayer.
Three-dimensional representation of porelike structures
of tau
protein in a lipid bilayer.
The Porelike Structures Are Distinct from Annular Protofibrils
of Tau
The structure of the porelike structures formed by
tau is distinct from the annular protofibrils as revealed by AFM imaging
(data not shown). The average height of annular protofibrils is 3.60
± 0.77 nm, which is comparable to that of globular oligomers
of tau and distinctly greater than that of the porelike structures.
The average inner and outer diameter of an annular protofibril is
16.5 (±9.6) nm and 52.9 (±23.3) nm, respectively.
Discussion
Both electrical conductance and high resolution AFM imaging demonstrate
that tau forms conducting ion channels in planar lipid bilayers. The
data is remarkably similar to that previously reported for other amyloid
channels. Specifically, tau channels are irreversibly associated with
the membrane, heterodisperse, that is, exhibiting multiple single
channels conductances and relatively nonselective for common physiologic
ions. These channels differ from amyloid channels in at least 2 properties.
First, tau channels require acidic pH for formation. This may be explained
by increased aggregation of tau at acidic pH and may reflect the kinetics
of the aggregation process. Further experiments are needed to elucidate
this. Second, tau channels are not blocked by Zn2+. This
might reflect a difference in the structure of tau channels from other
amyloid peptides. The tau protein (2N4R) that we used is far larger
(MW 45 900) than most of the amyloid peptides shown to form
channels. The sheer bulk of the protein may be responsible for the
lack of Zn2+ block. Again, further investigations are required
to understand this.The channels observed in our experiments
are due to tau. Control
experiments without tau showed no changes in membrane conductance.
Experiments with tau at physiologic pH also showed no channels, but
when the pH was lowered sufficiently, channels formed. Raising the
pH after channels formation did not cause the channels to close or
disappear, indicating that acidic pH is likely required for tau aggregation
and/or membrane insertion. Channels were stable structures with long
lifetimes and relative permanence in the membrane. This argues strongly
for these channels representing a tau protein mediated, rather than
a lipid defect.The multiplicity of single channel conductances
observed in the
tau channels suggests that multiple species of tau oligomers are forming
channels. This is characteristic of channel forming molecules that
aggregate. Alternately, the multiple conductance levels observed could
be conformations of a single (or several) molecular entities. Further
work is required to resolve this issue.Furthermore, the multiplicity
of channel conductance is supported
by AFM results (Figure ). The observed channel size is smaller than expected considering
the molecular weight of tau compared to that of other amyloids.[20,25,26] The size similarity of tau and
other amyloid structures can be explained by a greater fraction of
aggregated tau being embedded within the lipid bilayer. For comparison,
connexin hemichannels, acetylcholine receptor channels (AChR), Na
and K channels, for example, all with similar molecular weights, are
only 8–12 nm in outer diameter with an ionic pore of 1–3
nm.[27−29]The tau channels reported here could play a
role in the pathophysiology
of Alzheimer’s and other dementias. These channels could depolarize
membranes, disrupt Ca2+ regulation, depolarize mitochondria,
and deplete cellular energy stores. These effects could impair synaptic
function including long-term potentiation and lead to the deficits
in memory that are characteristic of Alzheimer’s and other
neurodegenerative diseases. Tau channels might also interact with
amyloid beta channels. There is strong evidence for a role of amyloid
beta channels in Alzheimer’s pathophysiology. Tau channels
could exacerbate the effects of amyloid beta channels. This notion
is supported by the results with transgenic mice which show that mice
containing amyloid beta and tau develop more severe memory problems
at an earlier stage than mice with either peptide alone.[30]Our past results do not address possible
interactions of tau and
amyloid beta. There is evidence that tau is present in amyloid plaques
and that amyloid beta and tau may bind when in the appropriate β
sheet conformation. This amyloid “synergy” may lead
to new pathways for drug development to arrest the relentless progression
of these devastating neurological diseases. Blockers of amyloid beta
channels can prevent neurotoxicity in vitro. Future investigations
of tau channels should search for protective blockers that might be
therapeutic lead compounds. The multiplicity of single channels conductances
observed in the tau channels suggests that multiple species of tau
oligomers are forming channels. This phenomenon has been observed
with almost all other amyloid channels and reflects the necessity
of aggregation/dysregulation as a prerequisite for channel formation.
The heterodisperse nature of channels may underlie the failure of
anti-amyloid drugs to help patients with Alzheimer’s disease.
If many species are toxic, a single drug may be insufficient to stop
the disease process as it only targets one species out of many.
Authors: P Poorkaj; T D Bird; E Wijsman; E Nemens; R M Garruto; L Anderson; A Andreadis; W C Wiederholt; M Raskind; G D Schellenberg Journal: Ann Neurol Date: 1998-06 Impact factor: 10.422
Authors: M Hutton; C L Lendon; P Rizzu; M Baker; S Froelich; H Houlden; S Pickering-Brown; S Chakraverty; A Isaacs; A Grover; J Hackett; J Adamson; S Lincoln; D Dickson; P Davies; R C Petersen; M Stevens; E de Graaff; E Wauters; J van Baren; M Hillebrand; M Joosse; J M Kwon; P Nowotny; L K Che; J Norton; J C Morris; L A Reed; J Trojanowski; H Basun; L Lannfelt; M Neystat; S Fahn; F Dark; T Tannenberg; P R Dodd; N Hayward; J B Kwok; P R Schofield; A Andreadis; J Snowden; D Craufurd; D Neary; F Owen; B A Oostra; J Hardy; A Goate; J van Swieten; D Mann; T Lynch; P Heutink Journal: Nature Date: 1998-06-18 Impact factor: 49.962
Authors: H Eric Feinstein; Sarah J Benbow; Nichole E LaPointe; Nirav Patel; Srinivasan Ramachandran; Thanh D Do; Michelle R Gaylord; Noelle E Huskey; Nicolette Dressler; Megan Korff; Brady Quon; Kristi Lazar Cantrell; Michael T Bowers; Ratnesh Lal; Stuart C Feinstein Journal: J Neurochem Date: 2016-04-20 Impact factor: 5.372
Authors: Pu Chun Ke; Ruhong Zhou; Louise C Serpell; Roland Riek; Tuomas P J Knowles; Hilal A Lashuel; Ehud Gazit; Ian W Hamley; Thomas P Davis; Marcus Fändrich; Daniel Erik Otzen; Matthew R Chapman; Christopher M Dobson; David S Eisenberg; Raffaele Mezzenga Journal: Chem Soc Rev Date: 2020-07-07 Impact factor: 54.564