We worked with
engineered membrane protein channels and explored their applications in
biotechnology. The engineered channels can be used to make single
molecule biosensors for the detection of, for example, pharmaceuticals1,
metal ions3 and cellular second messengers5; to control the ion flow by
reversing the ion selectivity of pores2, and to transport neutral
molecules by electroosmosis6. Engineered channels equipped with
non-covalent adapters can form new supramolecular structures4. Protein
pores with new functions are potentially applicable in medical
diagnosis, bio-warfare agent detection, pharmaceutical screening and
environmental processing.
Keywords:
ion channels,
transmembrane protein pore, nanopore, bilayer, single molecule, single
channel recording, patch cramming, stochastic sensing, biosensor,
adapter, electroosmosis, ion selectivity, molecular device
 Ion
channels are macromolecular nanopores in cell membranes. In response to
physical stimuli, they can open and close to selectively transport ions
and molecules across lipid bilayers, thereby acting to sense, amplify, and
modulate environment
signals, e.g.
pH, membrane potential,
transmitters and ions. We were inspired to ask: Can these natural
nanopores
acquire new functions
by protein
engineering. In
particular, we
are
investigating stochastic sensing by using single engineered ion channels
as the sensor
elements. In principle, if an analyte
transports
through or binds to the lumen of
a channel, it will
characteristically modulate the ion
flux through the channel to the discrete conductance states. The Analyte
can be identified by measuring the amplitude and the residence time, and
quantified by
counting the frequency
of the binding events.
 alpha-hemolysin
(aHL,
figure at right) is a bacterial toxin that forms a heptameric transmembrane
pore. With the
help of three-dimensional crystal structure, this
protein nanopore has been made with diverse functional properties by using
genetic engineering and targeted chemical modification. These engineered pores will
find broad applications in medical diagnosis, drug delivery, molecular
separation, and environmental detection.
 Detection of
organic analytes
1 Cyclic
molecules such as cyclodextrins (CD, oligo-saccharides comprising glucose
units), can be non-covalently lodged in the lumen of the pore and produce
partial conductance block. Since cyclodextrins bind a variety of small
organic molecules in their hydrophobic interiors, therefore organic
molecules produce further block once encapsulated in the cyclodextrin
cavity. The characteristic residual current and the residence time of the
blockade can be used to identify and quantify the concentration of
analytes in mixture. In this case, cyclodextrins act as non-covalent
molecular adapters. Using adapter-mediated protein sensors, a variety of
therapeutic drugs can be recognized and their concentrations quantified. The protein pore sensor working on this principle is
programmable due to the diversity of adapter candidates, e.g. cyclic peptides.
 Detection of
metal ions 3 The aHL
pore can be engineered for quantification of nano-molar traces of divalent
metal cations M(II) in mixtures. This time,
four histidine residues that were introduced near the trans mouth
of the lumen in only one subunit among seven enable the quantification of
three or four M(II) simultaneously. The sensor element can be permanently
calibrated without a detailed understanding of the kinetics of interaction
of the metal ions with the engineered pore.
 Detection of
IP3 5
Arginine rings were introduced in lumen of the pore to bind
phosphate esters, e.g. second messenger inositol 1,4,5-triphosphate, in
the nano-molar range. A great future
challenge for this sensor will be the incorporation within a micro-pipette
and detection of the second messenger within cells in real time.
 Trapping
single organic molecules in a nanocavity4
The aHL
protein pore can be further engineered to accommodate two different
cyclodextrin adapters simultaneously within the lumen of the
transmembrane
b-barrel. The non-covalent self-assembled adapters form a cavity within the
barrel with a dimension of ~4.4 nm3. Analysis of
single-channel recordings reveals that
individual charged organic
molecules can be pulled into the cavity
by an electrical potential. Once trapped, an organic molecule shuttles
back and forth between the adapters for hundreds of
milliseconds. This technology for monitoring single molecule kinetics in a
confined nano-scale space enables the fabrication of multianalyte
sensors and provides a means to study of nano-chemical events, e.g.
chemical reactions in a confined space.
 Controlling
the ion selectivity of protein pore2 Ion selectivity is an
essential property of many protein channels in living cells that generate
electromotive forces required for electrical signaling. We have developed
a new way to reversibly modulate the ion selectivity of a protein pore by
using adapters as modular selectivity filters.
By reducing the dimensions of the channel lumen from "wide" to
"mid-sized," the adapters dominate the charge selectivity of
the pore. bCD
prefers anions in its hydrophobic cavity, thereby making the pore
considerably anion-selective. By contrast, the negatively charged sulfate
bCD
rejects anions, allowing cations to pass in preference, turning
aHL
into a cation-selective pore. The approach is versatile because
various adapters can be used to program the same protein. It may be
possible to use a similar approach to alter the activity of
other proteins (e.g., to modify the active site of an enzyme)
and to introduce greater selectivity by using an adapter with a
ring of carbonyl groups similar to those found in the ionophore
valinomycin or potassium channels. The molecular adapters generate
permeability ratios (PK+/PCl-)
over a 200-fold range and should be useful in the de novo
design of membrane channels for fundamental studies of ion permeation.
 Driving
neutral molecules by electroosmosis 6
In an ion-selective protein pore, there
exists a water stream that accompanies the net ion flow, and becomes a
force to drive neutral molecules into the pore. The electrostatic
environment in the lumen determines the direction of the voltage-sensitive
water stream, resulting in the voltage-sensitivity of adapter affinity.
The electroosmotic flow using the protein nanopores maybe significant in
new ways to deliver neutral substrates into cells, such as drug molecules.
Prospects We
are using using engineered transmembrane channels and pores as sensor elements to explore stochastic
sensing, a technique in which the interactions of analyte molecules with a
single receptor are monitored. A single protein molecule provides both a
binding site for the analyte and “read-out”. Single protein pore sensors
have many advantages over traditional multi-element sensors: digital
output, high sensitivity, a rapid and continuous response, wide dynamic
range, the ability to identity and quantify several analytes
simultaneously, high S/N at low concentration, and no need to be high
selective because of the distinctive signature of each analyte.
These results demonstrate the diversity of new properties that can be
engineered into ion channels. We are therefore inspired to think of its
many potential applications. The protein pore sensor may find applications
where fast time response, trace analyte detection and high throughput are
required. Single protein pores equipped with multiple binding sites might
be used to examine the binding kinetics of multivalent proteins. In basic
science, analytes, such as the second messenger IP3 might be
detected inside cells by incorporating a responsive pore into a
microeletrode that is inserted into the cell under examination. From a
wider point of view, nanotubes or any tubules with diameters of less than
2 nm might be used as sensor elements as alternatives to protein-based
pores, if the chemistry of the lumen can be manipulated appropriately.
Single-molecule fluorescence or force measurement may be combined with
pico-Ampere current detection or used as alternative detection methods for
stochastic sensing.
|
