WO2008102121A1 - Formation of lipid bilayers - Google Patents

Abstract

A method for forming a lipid bilayer across an aperture, comprises: (a) providing a cell having a chamber adjacent to a septum comprising a membrane having an aperture capable of supporting a lipid bilayer; (b) depositing one or more lipids on an internal surface of the chamber; (c) introducing an aqueous solution into the chamber to cover the aperture and the internal surface and to form an interface between the solution and lipids; and (d) moving the interface past the aperture at least once to form a lipid bilayer across the aperture.

Description

FORMATION OF LIPID BILAYERS

The invention relates to the formation of lipid bilayers. In particular, the invention relates to the formation of a lipid bilayer across an aperture.

Lipid bilayers are models of cell membranes and serve as excellent platforms for a range of experimental studies. For example, lipid bilayers can be used for in vitro investigation of membrane proteins by single-channel recording. Alternatively, lipid bilayers can be used as biosensors to detect the presence of a range of substances. In particular, lipid bilayers can used to detect the presence of membrane pores or channels or can be used in stochastic sensing in which the response of a membrane protein to a molecule or physical stimulus is used to perform sensing of that molecule or stimulus.

Lipid bilayers are commonly formed by the method of Montal and Mueller (Proc. Natl. Acad. Sci. USA., 1972; 69: 3561-3566), in which a lipid monolayer is carried on aqueous solution/air interface past either side of an aperture which is perpendicular to that interface. The lipid is normally added to the surface of an aqueous electrolyte solution by first dissolving it in an organic solvent and then allowing a drop of the solvent to evaporate on the surface of the aqueous solution on either side of the aperture. Once the organic solvent has evaporated, the solution/air interfaces on either side of the aperture are physically moved up and down past the aperture until a bilayer is formed.

The method of Montal & Mueller is popular because it is a cost-effective and relatively straightforward method of forming good quality lipid bilayers that are suitable for protein pore insertion.

Other common methods of bilayer formation include tip-dipping, painting bilayers and patch-clamping of liposome bilayers.

Tip-dipping bilayer formation entails touching the aperture surface (for example, a pipette tip) onto the surface of a test solution that is carrying a monolayer of lipid. Again, the lipid monolayer is first generated at the solution/air interface by allowing a drop of lipid dissolved in organic solvent to evaporate at the solution surface. The bilayer is then formed by the Langmuir- Schaefer process and requires mechanical automation to move the aperture relative to the solution surface.

For painted bilayers, a drop of lipid dissolved in organic solvent is applied directly to the aperture, which is submerged in an aqueous test solution. The lipid solution is spread thinly over the aperture using a paintbrush or an equivalent. Thinning of the solvent results in formation of a lipid bilayer. However, complete removal of the solvent from the bilayer is difficult and consequently the bilayer formed by this method is less stable and more prone to noise during electrochemical measurement.

Patch-clamping is commonly used in the study of biological cell membranes. The cell membrane is clamped to the end of a pipette by suction and a patch of the membrane becomes attached over the aperture. The method has been adapted for producing lipid bilayers by clamping liposomes which then burst to leave a lipid bilayer sealing over the aperture of the pipette. The method requires stable, giant and unilamellar liposomes and the fabrication of small apertures in materials having a glass surface.

These common methods of forming lipid bilayers are complicated and time consuming. For instance, it is normally necessary to wait for the evaporation of an organic solvent in which the lipids are dissolved before a bilayer can be formed. There is therefore a need for simple and rapid methods of forming a lipid bilayer that do not involve the use of organic solvents.

In one aspect, the present invention provides a method for forming a lipid bilayer across an aperture, comprising:

(a) providing a cell having a chamber adjacent to a septum comprising a membrane having an aperture capable of supporting a lipid bilayer;

(b) depositing one or more lipids on an internal surface of the chamber;

(c) introducing an aqueous solution into the chamber to cover the aperture and the internal surface and to form an interface between the solution and lipids; and

(d) moving the interface past the aperture at least once to form a lipid bilayer across the aperture.

In another aspect, the invention provides a device for forming a lipid bilayer comprising,

(a) a cell having a chamber adjacent to a septum comprising a membrane having an aperture capable of supporting a lipid bilayer; and

(b) one or more lipids deposited on an internal surface of the chamber, wherein the cell comprises an inlet for introducing an aqueous solution into the chamber having lipid deposited therein.

The inventors have shown that a lipid bilayer will form across an aperture following the deposition of lipids on a surface adjacent to the aperture. They have shown that an aqueous solution can be used to collect the lipids from the surface and form a lipid/solution interface. The lipid bilayer forms across an aperture as the interface passes the aperture.

Advantageously, the lipids can be dried. The inventors have also shown that a lipid bilayer will form across an aperture following the rehydration of dried lipids. They have shown that an aqueous solution can be used to rehydrate the lipids and form a lipid/solution interface. The lipid bilayer forms across an aperture as the interface passes the aperture.

The invention has several advantages. The invention allows the formation of a lipid bilayer in the absence of large amounts of organic solvent. This means that a lipid bilayer can be formed rapidly because it is not necessary to wait for evaporation of the organic solvent before the lipid bilayer can be formed.

In addition, this means that the cell in the device of the invention can be made from materials that may be sensitive to organic solvents. For instance, organic-based adhesives can be used to construct the cell and screen-printed conductive silver/silver chloride paste can be used to construct electrodes within the cell. This means that the device can be cheaply manufactured in a straightforward manner. In addition, the use of organic solvent-sensitive polymers to construct the membrane comprising the aperture facilitates manufacture of the device.

As lipid bilayers are preferably formed from dried lipid, this allows the lipid to be stably stored in the cell until it is needed to form a lipid bilayer. This also avoids the need for wet storage of lipid in the device prior to use. Dry storage of lipids means that the device has a long shelf life.

The invention generally concerns the formation of a lipid bilayer across an aperture. A lipid bilayer is formed from two opposing layers of lipids. The two layers of lipids are arranged such that their hydrophobic tail groups face towards each other to form a hydrophobic interior. The hydrophilic head groups of the lipids face outwards towards the aqueous environment on each side of the bilayer. The bilayer may be present in a number of lipid phases including, but not limited to, the liquid disordered phase (fluid lamellar), liquid ordered phase, solid ordered phase (lamellar gel phase, interdigitated gel phase) and planar bilayer crystals (lamellar sub-gel phase, lamellar crystalline phase).

The lipid bilayer can be formed from one or more lipids. The lipid bilayer can also contain additives that affect the properties of the bilayer. In many applications, the lipid bilayer has one or more membrane proteins inserted therein. Certain lipids, additives and proteins that can be used in accordance with the invention are discussed in more detail below.

The lipid bilayer is formed inside a cell. In general, any cell can be used. The cell may be any shape or size. The cell may be a conventional electrophysiology cell or a specially-constructed cell, such as a biosensor chip.

The cell comprises an internal chamber. The chamber may be any size and shape. The volume of the chamber is typically 0.1 μl to 10ml. The chamber is adjacent to a septum. In a preferred embodiment, the cell comprises a septum which divides the cavity into two chambers. The two chambers may have equal volumes or different volumes.

The septum comprises a membrane. The membrane can be made from any material including, but not limited to, a polymer, glass and a metal. The membrane is preferably made from a material that forms a barrier to the flow of ions from the chamber. Suitable materials include, but are not limited to, polycarbonate (PC), polytetrafluoroethylene (PTFE), polyethylene, polypropylene, nylon and polyethylene naphthalate (PEN), polyvinylchloride (PVC), polyacrylonitrile (PAN), polyether sulphone (PES), polyimide, polystyrene, polyvinylfluoride (PVF), polyethylene telephthalate (PET), aluminized PET, nitrocellulose, polyetheretherketone (PEEK) and fluoroethylkene polymer (FEP). The membrane is preferably made from polycarbonate or PTFE.

The membrane is sufficiently thin to facilitate formation of the lipid bilayer across an aperture as described below. Typically the thickness will be in the range of IOnm to lmm. The membrane is preferably 0.1 μm to 25μm thick.

The membrane is preferably pre-treated to make the lipids and the aperture more compatible such that the lipid bilayer forms more easily that it would in the absence of pre- treatment. The membrane is preferably pre-treated to increase its affinity to lipids. The inventors have shown that pre-treatment of the membrane to increase its affinity to lipids allows the lipid bilayer to form more easily. In particular, it allows the formation of a lipid bilayer across the aperture following a single pass of the lipid/solution interface. The removal of the need to move the lipid/solution interface back and forth past the aperture means that the method of the invention is simplified. It also means that there is no need for fluidics control in the device of the invention. Hence, the cost and size of the device of the invention are reduced. The inventors have also shown that pre-treatment of the membrane to increase its affinity to lipids results in the formation of a lipid bilayer with increased stability. This means that the method of the invention can be used to form stable lipid bilayers. It also means that the device of the invention can be used in situations where the lipid bilayer is likely to encounter mechanical or other forces. For instance, the device of the invention can be used as a hand-held device.

Any treatment that modifies the surface of the membrane surrounding the aperture to increase its affinity to lipids may be used. The membrane is typically pre-treated with long chain organic molecules in an organic solvent. Suitable long chain organic molecules include, but are not limited to, n-decane, hexadecane, hexadecance mixed with one or more of the lipids discussed below, iso-eicosane, octadecane, squalene, fluoroinated oils (suitable for use with fluorinated lipids), alkyl-silane (suitable for use with a glass membrane) and alkyl-thiols (suitable for use with a metallic membrane). Suitable solvents include, but are not limited to, pentane, hexane, heptane, octane, decane, iso-ecoisane and toluene. The membrane is typically pre-treated with from 0.1% (v/v) to 50% (v/v), such as 0.3%, 1% or 3% (v/v), hexadecane in pentane. The volume of hexadecane in pentane used is typically from 0.1 μl to lOμl. The hexadecane can be mixed with one or more lipids. For instance, the hexadecane can be mixed with any of the lipids discussed below. The hexadecane is preferably mixed with diphantytanoyl-sn-glycero-3-phosphocoline (DPhPC). Preferably, the aperture is treated with 2μl of 1% (v/v) hexadecane and 0.6mg/ml lipid, such as DPhPC, in pentane.

Some specific pretreatments are set out in Table 1 by way of example and without limitation. Table 1:

The precise volume of pretreatment substance required depends on the pretreatment both the size of the aperture, the formulation of the pretreatment, and the amount and distribution of the pretreatment when it dries around the aperture. In general increasing the amount of pretreatment (i.e. by volume and/or by concentration) improves the effectiveness, but too much pretreament can block the aperture. As the diameter of the aperture is decreased, the amount of pretreatment required also decreases. The distribution of the pretreatment can also affect effectiveness, this being dependent on the method of deposition, and the compatibility of the membrane surface chemistry.

The relationship between the pretreatment and the ease and stability of bilayer formation is therefore complex, depending on a complex cyclic interaction between the aperture dimensions, the membrane surface chemistry, the pretreatment formulation and volume, and the method of deposition. The temperature dependent stability of the pretreated aperture further complicates this relationship. However, the pretreatment may be optimised by routine trial and error to enable bilayer formation immediately upon first exposure of the dry aperture to the lipid monolayer at the liquid interface.

Although the pretreatment provides a beneficial effect, it is not essential.

The septum preferably further comprises a support sheet on at least one side of the membrane. The septum preferably comprises a support sheet on both sides of the membrane. The support sheet may be of any material. Suitable materials include, but are not limited to, Delrin® (polyoxymethylene or acetal homopolymer), Mylar® (biaxially-oriented polyethylene terephthalate (boPET) polyester film), polycarbonate (PC), polyvinylchloride (PVC), polyacrylonitrile (PAN), polyether sulphone (PES), polysulphone, polyimide, polystyrene, polyethylene, polyvinylfluoride (PVF), polyethylene telephthalate (PET), polytetrafluoroethylene (PTFE), polyetheretherketone (PEEK) and fluoroethylkene polymer (FEP).

The membrane has an aperture which is capable of supporting a lipid bilayer. The septum typically has one aperture but can have more than one aperture. A lipid bilayer will form across each of the apertures in the membrane. If the membrane is made from a material that forms a barrier to the flow of ions, the aperture allows the movement of ions between from the chamber. The aperture may be any size and shape which is capable of supporting a lipid bilayer. The aperture preferably has a diameter in at least one dimension which is 20μm or less. The inventors have shown that this preferred size of aperture results in the formation of a lipid bilayer with increased stability. This means that the method of the invention can form stable lipid bilayers and that the device of the invention can be used in situations where the lipid bilayer is likely to encounter mechanical or other forces. For instance, it can be used as a hand-held device. The preferred size of aperture also allows the lipid bilayer to form more easily. In particular, it allows the formation of a lipid bilayer across the aperture following a single pass of the lipid/solution interface and removes the need to move the lipid/solution interface back and forth past the aperture.

The aperture may be created using any method. Suitable methods include, but are not limited to, spark generation and laser drilling.

Preferred combinations of membrane and aperture for use in accordance with the invention are shown in the Table 2 which sets out in the first column the thickness and material of the membrane and in the second column the diameter and method of forming the aperture. Table 2

One or more lipids are deposited on an internal surface of the chamber. The lipids can be deposited on one or more of any of the internal surfaces of the chamber. If the cell has two chambers, one or more lipids are deposited on an internal surface of one or both chambers. The lipids can be deposited on one or more of any of the internal surfaces of one or both chambers. The lipids can be deposited on one or both sides of the septum and on the membrane and/or the support sheet. The lipids are deposited in such a manner that the aqueous solution covers the lipids and the apertures as discussed in more detail below. The lipid can be deposited on the septum and/or one or more internal walls of the chamber but are preferably deposited on the septum. The lipids can be deposited on one or both sides of the septum and on the membrane and/or the support sheet. The lipids are deposited in such a manner that the aqueous solution covers the lipids and the apertures as discussed in more detail below.

Any method may be used to deposit the lipids on an internal surface of the chamber. Suitable methods include, but are not limited to, evaporation or sublimation of a carrier solvent, spontaneous deposition of liposomes or vesicles from a solution and direct transfer of the dry lipid from another surface. Cells having lipids deposited on an internal surface may be fabricated using methods including, but not limited to, drop coating, various printing techniques, spin-coating, painting, dip coating and aerosol application.

The lipids are preferably dried. Even when dried to a solid state, the lipids will typically contain trace amounts of residual solvent. Dried lipids are preferably lipids that comprise less than 50wt% solvent, such as less than 40wt%, less than 30wt%, less than 20wt%, less than 15wt%, less than 10wt% or less than 5wt% solvent.

Any lipids that form a lipid bilayer may be deposited. The lipids deposited in the cell are chosen such that a lipid bilayer having the required properties, such surface charge, ability to support membrane proteins, packing density or mechanical properties, is formed. The lipids can comprise one or more different lipids. For instance, the lipids can contain up to 100 lipids. The lipids preferably contain 1 to 10 lipids. The lipids may comprise naturally-occurring lipids and/or artificial lipids.

The lipids typically comprise a head group, an interfacial moiety and two hydrophobic tail groups which may be the same or different. Suitable head groups include, but are not limited to, neutral head groups, such as diacylglycerides (DG) and ceramides (CM); zwitterionic head groups, such as phosphatidylcholine (PC), phosphatidylethanolamine (PE) and sphingomyelin (SM); negatively charged head groups, such as phosphatidylglycerol (PG); phosphatidylserine (PS), phosphatidylinositol (PI), phosphatic acid (PA) and cardiolipin (CA); and positively charged headgroups, such as trimethylammonium-Propane (TAP). Suitable interfacial moieties include, but are not limited to, naturally-occurring interfacial moieties, such as glycerol-based or ceramide- based moieties. Suitable hydrophobic tail groups include, but are not limited to, saturated hydrocarbon chains, such as lauric acid (n-Dodecanolic acid), myristic acid (n-Tetradecononic acid), palmitic acid (n-Hexadecanoic acid), stearic acid (n-Octadecanoic) and arachidic («- Eicosanoic); unsaturated hydrocarbon chains, such as oleic acid (cis-9-Octadecanoic); and branched hydrocarbon chains, such as phytanoyl. The length of the chain and the position and number of the double bonds in the unsaturated hydrocarbon chains can vary. The length of the chains and the position and number of the branches, such as methyl groups, in the branched hydrocarbon chains can vary. The hydrophobic tail groups can be linked to the interfacial moiety as an ether or an ester.

The lipids can also be chemically-modified. The head group or the tail group of the lipids may be chemically-modified. Suitable lipids whose head groups have been chemically-modified include, but are not limited to, PEG-modified lipids, such as l,2-Diacyl-sn-Glycero-3- Phosphoethanolamine-N -[Methoxy(Polyethylene glycol)-2000]; functionionalised PEG Lipids, such as l,2-Distearoyl-sn-Glycero-3 Phosphoethanolamine-N-[Biotinyl(Polyethylene Glycol)2000]; and lipids modified for conjugation, such as l,2-Dioleoyl-sn-Glycero-3- Phosphoethanolamine-N-(succinyl) and 1 ,2-Dipalmitoyl-sn-Glycero-3 -Phosphoethanolamine-N- (Biotinyl). Suitable lipids whose tail groups have been chemically-modified include, but are not limited to, polymerisable lipids, such as l,2-bis(10,12-tricosadiynoyl)-sn-Glycero-3- Phosphocholine; fluorinated lipids, such as l-Palmitoyl-2-(16-Fluoropalmitoyl)-sn-Glycero-3- Phosphocholine; deuterated lipids, such as l,2-Dipalmitoyl-D62-sn-Glycero-3-Phosphocholine; and ether linked lipids, such as l,2-Di-O-phytanyl-sn-Glycero-3-Phosphocholine.

The lipids typically comprise one or more additives that will affect the properties of the lipid bilayer. Suitable additives include, but are not limited to, fatty acids, such as palmitic acid, myristic acid and oleic acid; fatty alcohols, such as palmitic alcohol, myristic alcohol and oleic alcohol; sterols, such as cholesterol, ergosterol, lanosterol, sitosterol and stigmasterol; lysophospholipids, such as l-Acyl-2-Hydroxy-sn- Glycero-3-Phosphocholine; and ceramides. The lipid preferably comprises cholesterol and/or ergosterol when membrane proteins are to be inserted into the lipid bilayer.

The lipid bilayer is formed by introducing an aqueous solution into the chamber. The aqueous solution covers both the internal surface on which the lipids are deposited and the aperture. The chamber may be completely filled with the aqueous solution or may be partially filled with the aqueous solution, as long as the both the lipids and the aperture are covered with the aqueous solution. If the cell has two chambers, one chamber may be completely filled, while the other is only partially filled.

The aqueous solution may cover the lipids and the aperture in any order but preferably covers the lipids before the aperture. The inventors have shown that covering the lipids before the aperture allows the lipid bilayer to form more easily. In particular, it allows the formation of a lipid bilayer across the aperture following a single pass of the lipid/solution interface. The removal of the need to move the lipid/solution interface back and forth past the aperture means that the method of the invention is simplified. It also means that there is no need for fluidics control in the device of the invention, thereby reducing its cost and size.

The design of the chamber and the position of the lipids may be chosen to determine the order in which the aqueous solution covers the lipids and aperture. For instance, if the lipids are to be covered first, a chamber is provided in which the lipids are positioned along the flow path between the point at which the aqueous solution is introduced to the chamber and the aperture.

Any aqueous solution that collects the lipids from the internal surface and allows the formation of a lipid bilayer may be used. The aqueous solution is typically a physiologically acceptable solution. The physiologically acceptable solution is typically buffered to a pH of 3 to 9. The pH of the solution will be dependent on the lipids used and the final application of the lipid bilayer. Suitable buffers include, but are not limited, to phosphate buffered saline (PBS), N-2- Hydroxyethylpiperazine-N'-2-Ethanesulfonic Acid (HEPES) buffered saline, Piperazine-l,4-Bis-2- Ethanesulfonic Acid (PIPES) buffered saline, 3-(n-Morpholino)Propanesulfonic Acid (MOPS) buffered saline and Tris(Hydroxymethyl)aminomethane (TRIS) buffered saline. By way of example, in one implementation, the aqueous solution may be 1OmM PBS containing l.OM sodium chloride (NaCl) and having a pH of 6.9.

The introduction of the aqueous solution collects the lipids from the internal surface. The immiscibility of the rehydrated lipids and the aqueous solution allows the formation of an interface between the lipids and the solution. The interface can be any shape and size. The interface typically separates a layer of lipids from the aqueous solution. The layer of lipids preferably forms on the top of the solution. The layer of lipid typically separates the solution from any air in the chamber(s).

The lipid bilayer is formed as the interface moves past the aperture. The interface moves past the aperture in such a way that the layer of lipids contacts the membrane material surrounding the aperture and a lipid bilayer is formed. The interface can be at any angle relative to the membrane as it moves past the aperture. The interface is preferably perpendicular to the membrane as it moves past the aperture.

The interface may move past the aperture as many times as is necessary to form the lipid bilayer. The interface moves past the aperture at least once. The interface can move past the aperture more than once, such as twice, three times or more. The interface can move past the aperture on one side or on both sides of the membrane.

If the aqueous solution covers the internal surface on which the lipids are deposited before the aperture, the lipid bilayer may form as the interface moves past the aperture as the chamber fills. Hence, if the lipid bilayer can be formed by a single pass of the interface past the aperture, the step of moving the interface past the aperture may be performed by the filling of the chamber.

In other embodiments, it will be necessary to move the interface back and forth past the aperture. For instance, if the aqueous solution covers the aperture before the lipids or covers the aperture and lipids simultaneously, it may be necessary to move the interface back and forth past the aperture.

In a preferred embodiment, the cell has two chambers and one of the chambers contains a gel. The chamber is typically filled with the gel such that the gel contacts the membrane. The presence of the gel contacting the membrane facilitates the formation of the lipid bilayer by physically supporting the bilayer. The presence of the gel allows the lipid bilayer to form more easily. In particular, it allows the formation of a lipid bilayer across the aperture following a single pass of the lipid/solution interface and removes the need to move the lipid/solution interface back and forth past the aperture. It also means that there is no need for fluidics control in the device of the invention, thereby reducing its cost and size. The presence of the gel also results in the formation of a lipid bilayer with increased stability. This means that the method of the invention can be used to form stable lipid bilayers. It also means that the device of the invention can be used in situations where the lipid bilayer is likely to encounter mechanical or other forces. For instance, it can be used as a hand-held device.

In another embodiment, there can remain a gap between the gel and the membrane. The presence of the gap means that a wider variety of materials can be used to make the gel, including ionically non-conductive materials.

The gel is preferably a hydrogel. The gel is typically ionically conductive. Suitable ionically conductive gels include, but are not limited to, agarose, polyacrylamide gel, Gellan gel and carbomer gel. However, if there is a gap present between the gel and the aperture, the gel can be ionically non-conductive.

The invention preferably also involves inserting membrane proteins into the lipid bilayer once it has been formed. The membrane proteins are deposited within the chamber and spontaneously insert into the lipid bilayer following the introduction of the aqueous solution. The inventors have shown that membrane proteins will spontaneously insert into the lipid bilayer following their removal from an internal surface of the chamber by the aqueous solution. This avoids the need to actively insert the membrane proteins into the lipid bilayer by introducing the proteins into the solution surrounding the bilayer or physically carrying the protein through the solution to the bilayer. Again, this simplifies the method of the invention as well as removes the need for wet storage of the proteins and the need for automation within a device of the invention.

In one embodiment, the gel described above comprises one or more membrane proteins. The membrane proteins can be deposited on the surface of the gel and/or can be present within the body of gel. Once the lipid bilayer has formed, the membrane proteins move from the gel and spontaneously insert themselves into the lipid bilayer. The gel can comprise one or more different membrane proteins.

In another embodiment, one or more membrane proteins are deposited on an internal surface of the chamber. The aqueous solution collects the membrane proteins from the surface and allows them to insert into the lipid bilayer. The membrane proteins may be deposited anywhere within the cell such that, once they have been collected from the surface, they can diffuse to and spontaneously insert into the lipid bilayer. The membrane proteins can be deposited on the same or different internal surface as the lipids. The lipids and the membrane proteins may be mixed together. The membrane proteins can be deposited on the septum and/or one or more internal walls of the chamber, but are preferably deposited on the septum. They may be deposited on one or both sides of the septum and on the membrane or the support sheet.

The lipids, the aperture and the membrane proteins may be covered by the aqueous solution in any order, although as already discussed the aqueous solution preferably covers the lipids first. The design of the cell and the position of the membrane proteins may be chosen to determine the order in which the aqueous solution covers the lipids, the aperture and the membrane proteins.

Any method may be used to deposit the membrane proteins on an internal surface of the cell. Suitable methods include, but are not limited to, drop coating, various printing techniques, spin-coating, painting, dip coating and aerosol application.

The membrane proteins are preferably dried. Even when dried to a solid state, the membrane proteins will typically contain trace amounts of residual solvent. Dried membrane proteins are preferably membrane proteins that comprise less than 20wt% solvent, such as less than 15wt% , less than 10wt% or less than 5wt% solvent.

In a further embodiment, the gel comprises one or more membrane proteins and one or more membrane proteins are deposited on an internal surface of one or both chambers.

Any membrane proteins that insert into a lipid bilayer may be deposited. The membrane proteins may be naturally-occurring proteins and/or artificial proteins. Suitable membrane proteins include, but are not limited to, β-barrel membrane proteins, such as non-constitutive toxins, porins and relatives and autotransporters; membrane channels, such as ion channels and aquaporins; bacterial rhodopsins; G-protein coupled receptors; and antibodies. Examples of non-constitutive toxins include hemolysin and leukocidin, such as Staphylococcal leukocidin. Examples of porins include maltoporin, OmpG, OmpA and OmpF. Examples of autotransporters include the NaIP and Hia transporters. Examples of ion channels include the NMDA receptor, the potassium channel from Streptomyces lividans (KcsA), the bacterial mechanosensitive membrane channel of large conductance (MscL), the bacterial mechanosensitive membrane channel of small conductance (MscS) and gramicidin. Examples of G-protein coupled receptors include the metabotropic glutamate receptor. The membrane protein can also be the anthrax protective antigen.

The membrane proteins preferably comprise α-hemolysin or a variant thereof. The α- hemolysin pore is formed of seven identical subunits (heptameric). The polynucleotide sequence that encodes one subunit of α-hemolysin is shown in SEQ ID NO: 1. The full-length amino acid sequence of one subunit of α-hemolysin is shown in SEQ ID NO: 2. The first 26 amino acids of SEQ ID NO: 2 correspond to the signal peptide. The amino acid sequence of one mature subunit of α-hemolysin without the signal peptide is shown in SEQ ID NO: 3. SEQ ID NO: 3 has a methionine residue at position 1 instead of the 26 amino acid signal peptide that is present in SEQ ID NO: 2.

A variant is a heptameric pore in which one or more of the seven subunits has an amino acid sequence which varies from that of SEQ ID NO: 2 or 3 and which retains pore activity. 1, 2, 3, 4, 5, 6 or 7 of the subunits in a variant α-hemolysin may have an amino acid sequence that varies from that of SEQ ID NO: 2 or 3. The seven subunits within a variant pore are typically identical but may be different.

The variant may be a naturally-occurring variant which is expressed by an organism, for instance by a Staphylococcus bacterium. Variants also include non-naturally occurring variants produced by recombinant technology. Over the entire length of the amino acid sequence of SEQ ID NO: 2 or 3, a variant will preferably be at least 50% homologous to that sequence based on amino acid identity. More preferably, the subunit polypeptide is at least 80%, at least 90%, at least 95%, at least 98%, at least 99% homologous based on amino acid identity to the amino acid sequence of SEQ ID NO: 2 or 3 over the entire sequence.

Amino acid substitutions may be made to the amino acid sequence of SEQ ID NO: 2 or 3, for example a single amino acid substitution may be made or two or more substitutions may be made. Conservative substitutions may be made, for example, according to the Table 3. Amino acids in the same block in the second column and preferably in the same line in the third column may be substituted for each other: Table 3:

Non-conservative substitutions may also be made at one or more positions within SEQ ID NO: 2 or 3, wherein the substituted residue is replaced with an amino acid of markedly different chemical characteristics and/or physical size. One example of a non-conservative substitution that may be made is the replacement of the lysine at position 34 in SEQ ID NO: 2 and position 9 in SEQ ID NO: 3 with cysteine (i.e. K34C or K9C). Another example of a non-conservative substitution that may be made is the replacement of the asparagine residue at position 43 of SEQ ID NO: 2 or position 18 of SEQ ED NO: 3 with cysteine (i.e. N43C or Nl 7C). The inclusion of these cysteine residues in SEQ ED NO: 2 or 3 provides thiol attachment points at the relevant positions. Similar changes could be made at all other positions, and at multiple positions on the same subunit.

One or more amino acid residues of the amino acid sequence of SEQ ED NO: 2 or 3 may alternatively or additionally be deleted. Up to 50% of the residues may be deleted, either as a contiguous region or multiple smaller regions distributed throughout the length of the amino acid chain.

Variants can include subunits made of fragments of SEQ ID NO: 2 or 3. Such fragments retain their ability to insert into the lipid bilayer. Fragments can be at least 100, such as 150, 200 or 250, amino acids in length. Such fragments may be used to produce chimeric pores. A fragment preferably comprises the β-barrel domain of SEQ ID NO: 2 or 3.

Variants include chimeric proteins comprising fragments or portions of SEQ ID NO: 2 or 3. Chimeric proteins are formed from subunits each comprising fragments or portions of SEQ ID NO: 2 or 3. The β-barrel part of chimeric proteins are typically formed by the fragments or portions of SEQ ID NO: 2 or 3.

One or more amino acid residues may alternatively or additionally be inserted into, or at one or other or both ends of, the amino acid sequence SEQ ID NO: 2 or 3. Insertion of one, two or more additional amino acids to the C terminal end of the peptide sequence is less likely to perturb the structure and/or function of the protein, and these additions could be substantial, but preferably peptide sequences of up to 10, 20, 50, 100 or 500 amino acids or more can be used. Additions at the N terminal end of the monomer could also be substantial, with one, two or more additional residues added, but more preferably 10, 20, 50, 500 or more residues being added. Additional sequences can also be added to the protein in the trans-membrane region, between amino acid residues 119 and 139 of SEQ ID NO: 3. More precisely, additional sequences can be added between residues 127 and 130 of SEQ ID NO: 3, following removal of residues 128 and 129. Additions can be made at the equivalent positions in SEQ ID NO: 2. A carrier protein may be fused to an amino acid sequence according to the invention.

Standard methods in the art may be used to determine homology. For example the UWGCG Package provides the BESTFIT program which can be used to calculate homology, for example used on its default settings (Devereux et al (1984) Nucleic Acids Research 12, p387-395). The PILEUP and BLAST algorithms can be used to calculate homology or line up sequences (such as identifying equivalent residues or corresponding sequences (typically on their default settings)), for example as described in Altschul S. F. (1993) J MoI Evol 36:290-300; Altschul, S.F et al (1990) J MoI Biol 215:403-10. Software for performing BLAST analyses is publicly available through the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/).

The membrane proteins can be labelled with a revealing label. The revealing label can be any suitable label which allows the proteins to be detected. Suitable labels include, but are not limited to, fluorescent molecules, radioisotopes, e.g. ¹²⁵1, ³⁵S, en∑ymes, antibodies, polynucleotides and linkers such as biotin.

The membrane proteins may be isolated from an organism, such as Staphylococcus aureus, or made synthetically or by recombinant means. For example, the protein may be synthesized by in vitro translation transcription. The amino acid sequence of the proteins may be modified to include non-naturally occurring amino acids or to increase the stability of the proteins. When the proteins are produced by synthetic means, such amino acids may be introduced during production. The proteins may also be modified following either synthetic or recombinant production.

The proteins may also be produced using D-amino acids. In such cases the amino acids will be linked in reverse sequence in the C to N orientation. This is conventional in the art for producing such proteins.

A number of side chain modifications are known in the art and may be made to the side chains of the membrane proteins. Such modifications include, for example, modifications of amino acids by reductive alkylation by reaction with an aldehyde followed by reduction with NaBH₄, amidination with methylacetimidate or acylation with acetic anhydride.

Recombinant membrane proteins can be produced using standard methods known in the art. Nucleic acid sequences encoding a protein can be isolated and replicated using standard methods in the art. Nucleic acid sequences encoding a protein can be expressed in a bacterial host cell using standard techniques in the art. The protein can be introduced into a cell by in situ expression of the polypeptide from a recombinant expression vector. The expression vector optionally carries an inducible promoter to control the expression of the polypeptide.

The lipid bilayer may be used for a variety of purposes. The lipid bilayer may be used for in vitro investigation of membrane proteins by single-channel recording. The lipid bilayer may be used as a biosensor to detect the presence of a range of substances. The lipid bilayer may be used to detect the presence or absence of membrane pores or channels in a sample. The presence of the pore or channel may be detected as a change in the current flow across the lipid bilayer as the pore or channel inserts into the lipid bilayer. The lipid bilayer preferably contains membrane protein and is used to detect the presence or absence of a molecule or stimulus using stochastic sensing. The lipid bilayer may be used for a range of other purposes, such as studying the properties of molecules known to be present (e.g. DNA sequencing or drug screening), or separating components for a reaction.

To allow further understanding, embodiments of the invention will now be described by way of a non-limiting example with reference to the drawings, in which:

Fig. 1 is a view of an example of a device of the invention;

Fig. 2 is a schematic diagram of an electrical circuit that can be used with the device of the invention;

Fig. 3 is a graph of the current response to a 2OmV 50Hz alternating current (a.c.) waveform in the absence of the high resistance electrical sealing of the aperture by a bilayer;

Fig. 4 is a graph of the characteristic square wave capacitive current in response to a 2OmV amplitude triangular waveform at 50Hz indicative of bilayer formation across the aperture;

Fig. 5 is a graph of the stepwise increase of 6OpA direct current (d.c.) as α-hemolysin pores automatically insert into the bilayer formed by the Montal and Mueller method; Fig. 6 is a graph of the characteristic interruptions in the current caused by single molecules of γ-cyclodextrin transiently binding to the α-hemolysin pores;

Fig. 7 is a graph of the current response to an applied potential before, during and following spontaneous bilayer formation and pore insertion in accordance with the invention in a standard two-chamber research cell;

Fig. 8 shows an expanded view of the current (1 second full scale) during the final minute of recording shown in Fig. 7;

Fig. 9 is a graph of the current recorded over the duration of a single test using a specially constructed cell;

Fig. 10 is a graph of the characteristic square wave indicative of bilayer formation in accordance with the invention in a specially constructed cell;

Fig. 11 is a graph of the current response to a 2OmV 50Hz a.c. waveform in the absence of the high resistance electrical sealing of the aperture by the bilayer in the specially constructed cell;

Fig. 12 is a graph of step increases in the current ~100pA as α-hemolysin pores automatically insert into the bilayer in the specially constructed cell; and

Fig. 13 is a graph of the characteristic interruptions in the current caused by single molecules of γ-cyclodextrin transiently binding to the α-hemolysin pores in the specially constructed cell;

Fig. 14 is a perspective view of a sensor system;

Fig. 15 is a perspective view of a cell of the sensor system;

Fig. 16 is a cross-sectional of the cell, taken along line ITl-III in Fig. 2;

Fig. 17 is a perspective view of a support sheet of the cell in isolation;

Fig. 18 is a perspective view of a body of the cell in isolation with a first arrangement for an inlet;

Fig. 19 is a perspective view of a cover sheet of the cell in isolation with a second arrangement for an inlet;

Fig. 20 is a cross-sectional view of the cell similar to that of Fig. 3 but showing introduction of a sample;

Fig. 21 is an expanded, partial cross-sectional view of a cell containing gel with a gap between the gel and an aperture;

Fig. 22 is an expanded perspective view of the connector portion of the reader unit;

Fig. 23 is a perspective view of a rigid metal body connected to the reader unit;

Fig. 24 is a cross-sectional view of the rigid metal body, taken along line XII-XII in Fig. l i;

Fig. 25 is a cross-sectional view of the cell contained in a Faraday cage;

Figs. 26 to 28 are diagrams of various forms of the electrical circuit in the reader unit; and

Fig. 29 is a flow chart of the operation of the reader unit; and Fig. 30 is a graph of a bias voltage applied to the reader unit; and

Figs. 31 to 35 are graphs of the current signal generated in the cell during operation.

In all of the graphs, the x-axis shows time in ms, the top portion of the y-axis shows the current in pA, and the bottom portion of the y-axis shows potential in mV.

A device 130 in accordance with the invention is illustrated in Fig. 1. The device 130 includes an electrophysiology cell 101 which is of a conventional type and construction for the performance of stochastic sensing using a membrane protein inserted in a lipid bilayer.

The electrophysiology cell 101 comprises two chambers body portions 102 having constructions which are mirror images of each other. The chamber body portions 102 may be made from Delrin® (polyoxymethylene or acetal homopolymer). The chamber body portions 102 each define a chamber portion 103 having an opening in the upper surface 104 of the respective chamber body portion 102. The chamber portions 103 each have a volume of a few millilitres, for example 1.5 ml. The chamber portions 103 have no wall on a side surface 105 of the respective chamber body portion 102. To form a chamber body, the two chamber body portions 102 are assembled together with their side surfaces 105 facing one another so that the respective chamber portions 103 are aligned and together form a chamber. The chamber body portions 102 may be attached by any suitable means, typically a clamp or an adhesive.

The electrophysiology cell 101 further comprises a membrane 106 made of polycarbonate or any other suitable polymer. Each face of the membrane 106 may be pre-treated in a conventional manner, for example with 10% (VTV) hexadecane in pentane. The membrane 106 is positioned between the facing side surfaces 105 of the two chamber body portions 102, for example by adhering both chamber body portions 102 to the membrane 106. Accordingly, the membrane 106 forms a wall which divides the chamber formed by the two chamber portions 103.

The membrane 106 has an aperture 107 which is aligned with the chamber portions 103 when the electrophysiology cell is assembled. The membrane 106 is sufficiently thin to facilitate formation of a lipid bilayer, for example being 25 μm thick. The aperture 107 may in general be of any shape or size which is capable of supporting the lipid bilayer, but preferably has a diameter in one dimension of 20μm or less. The cell 101 comprises inlets for introducing an aqueous solution into each chamber portion 103, namely the openings in the upper surface 104 of each chamber body portion 102.

The device 130 further comprises lipids 108 deposited in each chamber portion 103 of each one of the chamber body portions 102. The shape of the patch of lipids 108 deposited in each chamber portion 103 may vary.

The electrophysiology cell 101 may be used to form a lipid bilayer in accordance with the method of the invention. For example, an aqueous solution may be introduced into both chamber portions 103 simultaneously via openings in the upper surface 104 of each chamber body portion 102. The aqueous solution will cover the lipids 108 deposited in each chamber portion 103 and a lipid/solution interface will form with a layer of lipid resting on top of the solution. As more aqueous solution is introduced, the interface will rise within both chamber portions 103 and move past the aperture 107 on both sides of the membrane 106 thereby forming a lipid bilayer across the aperture 107. In this example, the lipid 108 is covered by the aqueous solution before the aperture 107 is covered.

The electrophysiology cell 101 can further includes respective electrodes (not shown in Fig. 1) provided in each chamber portion 103 of each one of the chamber body portions 102. The electrodes may be Ag/AgCl electrodes. The electrodes may form part of an electrical circuit 120 which is capable of measuring an electrical signal across the lipid bilayer. A suitable electrical circuit 120 is illustrated schematically in Fig. 2 and is of a conventional type for performing stochastic sensing by detecting the current flowing across the lipid bilayer.

The electrodes 109 are connected to an amplifier 121 such as a patch-clamp amplifier (eg an Axopatch 200B supplied by Axon Instruments) which amplifies the current signal output from the electrodes 109.

The current signal output by the amplifier 121 is supplied through a low-pass filter 122, such as a Bessel filter (eg with characteristics 80dB/decade with a corner frequency of 2kHz).

The current signal output by the low-pass filter 122 is supplied to an A/D converter 123, such as a Digitata 1320 A/D converter supplied by Axon Instruments. The A/D converter 123 might typically operate with a sampling frequency of 5kHz. The A/D converter 123 converts the current signal into a digital signal which is then supplied to a computer 124 for analysis. The computer 124 may be a conventional personal computer running an appropriate program to store the current signal and display it on a display device.

As an alternative, the invention may be applied to a device which is the cell of the sensor system described in detail below.

For comparative purposes, a bilayer was first formed using the common Montal and Mueller method. Bilayer formation was performed using a standard two-chamber research cell. The research cell is typical of those used in laboratory bilayer tests and comprises two Delrin (acetal homopolymer) blocks, each machined to create an open-sided 700ul chamber with appropriate access portals. The blocks are clamped together on either side of a polymer film which thereby separates the two chambers. The only electrical connection between the two chambers is by ionic conduction of the electrolyte solution through a small aperture created in the polymer film.

In order to facilitate bilayer formation, it is first necessary to apply a chemical surface treatment (commonly called the "pre-treatment") to either side of the aperture. 2-5 ul of 10% hexadecane dissolved in pentane was applied to either side of a dry aperture having a diameter of approximately 50μm. The pentane was allowed to evaporate.

Once the pre-treatment on the apertures had dried, both chambers of the research cell were filled with electrolyte solution comprising 1OmM Phosphate Buffered Saline (PBS) solution at pH 7.2, spiked with IM NaCl. A lOμl drop of l^-diphytanoyl-sn-glycero-S-phosphocholine lipid dissolved in pentane (10mg/ml) was then carefully applied to the surface of the solution in both chambers of the cell, and left to stand at room temperature for 15 minutes to allow the pentane to evaporate. Bilayers were subsequently formed by sequentially lowering and raising the air/solution interface past either side of the aperture, as described in Montal and Mueller, Proc. Natl. Acad. Sci. USA., 1972; 69: 3561-3566.

An electrical potential difference was applied across the membrane between the chambers of the test cell using Ag/ AgCl electrodes, one immersed in each chamber. Control of the applied potential and recording of the subsequent current was carried out using a current amplifier (MultiClamp700B from Axon Instruments with a CV 7B/BL headstage), coupled to a data acquisition system (DigiData 1322A also from Axon Instruments). The headstage and the test cell were housed in a Faraday cage to prevent interference from external electromagnetic noise. The DigiData 1322A is interfaced to a computer using pClamp version 9.2 software, and data acquired at 4 kHz, with a 2 kHz Bessel filter.

Formation of a bilayer across the aperture was confirmed by creation of a high resistance sealing of the aperture (>10GΩ), by measurement of the capacitance of the high resistance seal and by the subsequent successful insertion of α-hemolysin (α-HL) pores into the bilayer which resulted in a fixed current flow which is identical for each pore.

An electrical potential difference of +10OmV was applied between the two chambers once the electrolyte solution has been added, and a recorded current <10pA is consistent with bilayer formation. Evidence that the bilayer has sealed the aperture was provided by measurement of a predictable capacitive current in response to applying an alternating current (a.c.) potential perturbation. The current response to the applied a.c. waveform when the aperture was not sealed is given in Fig. 3. The squareware current response to a 2OmV amplitude triangular waveform at 50Hz in the presence of a lipid bilayer across the aperture is presented in Fig. 4.

Wild-type α-HL pores were injected into the bulk of the test solution. Confirmation that the high resistance seal across the aperture was caused by a lipid bilayer was provided by the successful insertion of pores in the bilayer. This insertion was seen as a stepwise increase in the direct current (d.c.) across the bilayer and is presented in Fig. 5.

Finally, confirmation that the stepwise increase in current is specifically due to insertion of the wild-type α-HL pores into the bilayer was provided by the addition of α-cyclodextrin, a well- characterised analyte that transiently binds to α-HL pores. A characteristic interruption in the current through the pores of approximately 60% was recorded as single molecules of α-cyclodextrin transiently bound to the pores, with a spread in binding durations in the range -100ms. This is presented in Fig. 6.

In the following Examples, a lipid bilayer was formed in accordance with the invention. The research test cell and apparatus described above was used to investigate the use of lipid dried to the base of the cell chambers and α-HL dried on the membrane around the aperture.

Two different polymer films were used to create the membrane separating the two chambers of the test cells; a 6μm thick biaxial polycarbonate film, and a 5μm thick polytetrafluoroethylene (PTFE) film: both films were sourced from Goodfellow Cambridge Ltd.

For each of these two polymer films, apertures were created by one of two different methods: sparking and laser drilling.

Laser-drilled membranes were produced using an Excimer laser at the UK Laser Micromachining Centre, Bangor, Wales. The laser-drilled holes used in these experiments were in the range of 5-30μm in diameter with a tapered morphology in cross section. Holes of this size allow a stable lipid bilayer to e formed more easily.

Spark-generated holes also in the range of 5-30μm in diameter were produced using a spark- generating device. The four polymer film/aperture combinations that were used are summarised below in Table 4. Table 4:

Holes of this size allow a stable lipid bilayer to be formed more easily.

Part of the aperture construction involved a chemical surface treatment to facilitate the bilayer formation process. This involved application of 2μl of 1% hexadecane in pentane to either side of the aperture. This was then allowed to evaporate. Pre-treatment also allows the easy formation of a stable lipid bilayer.

In addition to the chemical treatment during preparation of the aperture, lμl of 0.17mg/ml wild-type α-HL was subsequently applied to the aperture and allowed to dry at room temperature.

The test cells were then loaded with 20ul of the lipid solution (10mg/ml of 1,2-diphytanoyl- sn-glycero-3-phosphocholine in pentane) applied to the base of each chamber and stored at room temperature to allow the pentane to evaporate, leaving dry lipid coated on the base of each chamber.

The dry research cells, already loaded with lipid and α-HL, were re-hydrated by injecting a test solution (1OmM Phosphate Buffered Saline solution, 1.0M NaCl, and 0.25mM g-cyclodextrin, at pH 6.9) into the base of each chamber of the cell, raising the lipid/solution interface past the aperture only once on either side sequentially.

The electrical potential difference was applied across the membrane using Ag/ AgCl electrodes, as in the traditional set up, and data recorded at a sampling rate of 250μs per point using the equipment described previously.

The first recorded evidence of spontaneous bilayer formation and pore insertion upon rehydration of a test cell, which had been pre-loaded with dried lipid and protein pores, is presented in Fig. 7. Hence, a lipid bilayer can be formed by one pass of the lipid/solution interface past aperture if the solution covers the dried lipid before it covers the aperture, the aperture has a diameter of less than 20μm and the membrane has been pre-treated to increase its affinity to lipids. This removes the need to move the interface back and forth past the aperture.

Further testing was performed to confirm bilayer formation and pore insertion. Hence, pores will spontaneously insert into the lipid bilayer if they are deposited in dried form on an internal surface of the cell. This avoids the need to actively insert the pores into the lipid bilayer. The bilayer formation and pore insertion were consistent with the results for the traditional Montal and Mueller method that are described above and shown in Figs 4, 5 and 6.

The results for the second membrane in Table 2, with a pre-treatment of 2μl of 1% hexadecane in pentane, are shown in Fig. 7. Fig. 7 shows the current response (pA, upper portion) and the applied potential (mV, lower portion) recorded over a period of 180 seconds. Over the first 40 seconds of recording the cell is dry and the Faraday cage is open. Solution is injected on either side of the membrane just prior to the first marker on the plot (approximately 45 seconds). After a period of fluctuation as the electrodes are wetted and the Faraday cage is closed, the applied potential is then increased to +10OmV.

The current remains at <10pA, consistent with the GΩ seal of a bilayer, and then rises in a single step to approximately 9OpA, consistent with insertion of an α-HL pore and confirming that the aperture was blocked with a lipid bilayer (as described above for Fig. 5). The stepwise fluctuation in the current is from binding events with cyclodextrin causing transient partial blockage of the pore, and confirms that the current is due to an α-HL pore in the bilayer (as described above for Fig. 6). After -70s a second pore inserts into the bilayer.

Fig. 8 shows an expanded view of the current (1 second full scale) during the final minute of recording, again illustrating the characteristic step-like profile of the analyte binding events.

The results presented therefore illustrate that bilayer formation and subsequent pore insertion is possible directly upon re-hydration of the dry test cell with test solution using lipid and α-HL dried in the test cell. By this method the bilayers can be formed on the first exposure of the aperture to the solution/air interface carrying the lipid and a variety of apertures can be used including different membrane materials and aperture formation methods. Although the results are not presented here, bilayers have been formed on all the membrane/aperture combinations presented in Table 2 above.

A cell having a much smaller scale that the two-chamber research cell used above was constructed. The cell contained two cylindrical chambers having a cross-sectional diameter of 12mm and a length of 2mm. The volume of each chamber was approximately 56μl. Two alternative membrane materials were tested: a 6um thick biaxial polycarbonate film, and a 5μm thick PTFE film (Goodfellow Cambridge Ltd.)- Apertures were formed in the centre of the membrane by one of two different methods described above, sparking and laser drilling. The laser-drilled holes used in these experiments were lOμm in diameter with a tapered morphology in cross section. Spark generated apertures in the 5μm PTFE film membranes were approximately lOμm diameter circular holes, whereas for the 6μm polycarbonate film the sparked apertures were elliptical with dimensions approximately 20μm by 30μm. Holes of this size allow a stable lipid bilayer to be easily formed.

The apertures then received a chemical pre-treatment to facilitate the bilayer formation process. This consisted of 2μl of 1% hexadecane in pentane applied to either side of the aperture by capillary pipette. Pre-treatment also allows the easy formation of a stable lipid bilayer.

Once the pentane solvent had evaporated a lμl drop of aqueous protein solution (0.017 mg/ml w.t. α-HL) was applied near to one side of the aperture and dried.

The interior of each chamber was then coated with 4μl of lOmg/ml diphantytanoyl-sn- glycero-3-phosphocholine (DPhPC) dissolved in pentane.

The lipid re-hydrated by injecting test solution (1OmM Phosphate Buffered Saline solution, 1.0M NaCl, and 0.25mM g-cyclodextrin, at pH 6.9) into each chamber.

Control of the applied potential between the cell Ag/AgCl electrodes and recording of the subsequent current was with the same equipment described above. An electrical potential difference of +10OmV was applied between the two chambers after the test solution had been added, and a measured current <10pA was consistent with bilayer formation. Formation of a bilayer across the aperture was confirmed as discussed above. Hence, a lipid bilayer can be formed by one pass of the lipid/solution interface past the aperture if the solution covers the dried lipid before the aperture, the aperture has a diameter of less than 20μm and the membrane has been pre-treated to increase its affinity to lipids.

Fig. 9 shows a typical current trace recorded over the entire duration of one test. Figs 10, 11, 12 and 13 show expanded areas of Fig. 9. Each Fig. contains two graphs: the upper plot shows the current response to the applied potential, which is shown in the lower plot.

Prior to arrow 1 in Fig. 9, a 50Hz triangular a.c. potential waveform of 2OmV amplitude is applied between the electrodes, which are initially dry. When the test solution is then injected into each chamber of the cell, and the Faraday cage is closed, a square-wave capacitive current response is recorded with amplitude ~330pA, as seen in Fig. 10, indicating bilayer formation across the aperture. When the a.c. potential waveform is then replaced by a d.c. potential of +10OmV (after arrow 1 in Fig. 9), a constant current of <10pA is recorded, confirming that the aperture is sealed with >10 GΩ resistance as would be expected with a bilayer.

In the period immediately prior to arrow 2 in Fig. 9, the bilayer is deliberately broken by 'zapping' with a 50ms potential pulse of IV d.c. applied on top of 50Hz triangular a.c. waveform. The potential pulse is sufficient to permanently disrupt the high resistance electrical seal of the aperture, causing the current to go off scale, as seen in Fig. 11 (recorded between arrow 2 and arrow 3 in Fig. 9).

Beyond arrow 3 in Fig. 9, a new bilayer is formed using the Montal and Mueller method by lowering and then raising the solution/air interface carrying the lipid monolayer past the aperture. The square wave capacitive current is restored as the new bilayer forms. The potential waveform is then turned off and +10OmV d.c. applied, which results in approximately lOOpA step increases in the current as α-HL protein pores automatically insert into the bilayer. An expanded view showing the current as the pores insert into the bilayer is presented in Fig. 12 (after arrow 4 in Fig. 9). Again, pores will spontaneously insert into the lipid bilayer if they are deposited in dried form on an interface of the cell. This avoids the need to actively insert the pores into the lipid bilayer.

The γ-cyclodextrin in the test solution binds stochastically to the α-HL pores causing characteristic interruptions in the pore current, seen as approximately 6OpA step drops in the current which last 50-500ms. This is presented in Fig. 13.

The alternative device mentioned above which is the cell of a sensor system and to which the present invention may alternatively be applied will now be described. The sensor system is also described in a co-pending application being filed simultaneously with this application [J A Kemp & Co Ref: N.99663A; Oxford Nanolabs Ref: ONL IP 002] which is incorporated herein by reference. All the teachings of that application may be applied equally to the present invention.

A sensor system 1 is shown in Fig. 14 and comprises a cell 2 and an electrical reader unit 3 which may be connected together. In use, sensing using a lipid bilayer is formed in the cell 2 and an electrical current signal across the bilayer is monitored and interpreted by the reader unit 3. The sensor system 1 has been designed for use outside of a laboratory setting. Some examples include use in medicine for point of care testing (POCT), use in environmental protection for a field-based test for pollutants, use for counter bioterrorism for the detection of explosives and chemical and biological agents at the "point of terror". Nonetheless, some of features of the sensor system 1 also make it advantageous for laboratory use.

The cell 2 has a construction allowing it to be mass-produced at a low cost, allowing it to be a disposable product. The cell 2 is easily connected and replaced in the reader unit 3. The reader unit 3 is sufficiently small to be hand-held and portable.

The cell 2 is shown in Figs. 15 and 16 and will now be described in detail. The cell 2 has a layered construction formed from a stack of layers fixed together.

The cell 2 comprises a membrane 10 having an aperture 11 across which a lipid bilayer is supported in use. Although only a single aperture 11 is used in many applications, there may be plural apertures 11. The membrane 10 may be made of any material capable of supporting lipid bilayer across the aperture 11. Some examples include but are not limited to: a biaxial polycarbonate, PTFE, polyethylene, polypropylene, nylon, PEN, PVC, PAN, PES, polyimide, polystyrene, PVF, PET, aluminized PET, nitrocellulose, PEEK, or FEP. One factor in the choice of the material of the membrane 10 is the affinity to the lipid which affects the ease of bilayer formation. However the material of the membrane 10 has less significance when a pretreatment is used as described below. The choice of the material of the membrane 10 also affects the ease of formation of the aperture 11.

Similarly, the thickness of the membrane 10 is made sufficiently small to facilitate formation of the lipid bilayer across the aperture, typically being at most 25μm, preferably being at most lOμm thick, for example 5μm or 6μm. The thickness of the membrane 10 is typically at least 0.1 μm. The aperture 11 may in general be of any shape or size which it is capable of supporting a lipid bilayer, although it preferably has a restricted size as discussed further below.

The thickness of the membrane 10 is also dependent on the size of the aperture 11. As the aperture 11 decreases in size, the membrane 10 also needs to decrease in thickness in order to assist the formation of a lipid bilayer. Typically the thickness of the membrane 10 is no more than the minimum diameter of the aperture 11. Another factor is the electrical resistance of the membrane 10 which changes with the thickness. It is desirable that the resistance of the membrane 10 is sufficiently high relative to the resistance of the ion channel in a membrane protein inserted in the membrane 10 that the current flowing across the membrane 10 does not mask the current through the ion channel.

The membrane 10 is supported by two support sheets 12, provided on opposite sides of the membrane 10 and fixed thereto. As described further below, the membrane 10 and the support sheets 12 together form a septum 17. The support sheets 12 each have a window 13 which is aligned with the aperture 11 in the membrane 10 but is of larger size than the aperture 11 in order that the support sheets 12 do not interfere with the formation of a lipid bilayer across the aperture 11. The support sheets 12 have the function of supporting and strengthening the membrane 10 and may be made of any material suitable for achieving this purpose. Suitable materials include, but are not limited to: Delrin® (polyoxymethylene or acetal homopolymer), a polyester, eg Mylar® (biaxially-oriented polyethylene terephthalate (boPET) polyester film), PC, PVC, PAN, PES, polysulphone, polyimide, polystyrene, polyethylene, PVF, PET, PTFE, PEEK, or FEP

The support sheets 12 are typically thicker than the membrane 10, having a thickness typically at least 0.1 μm, preferably at least lOμm. The support sheets 12 are thinner than the bodies 14 described below, having a thickness typically at most lmm, preferably at most 0.5mm.

The cell 1 further comprises two bodies 14 each fixed to one of the support sheets 12. The bodies 14 are each formed from a sheet of material having an aperture 15 extending therethrough. The apertures 15 in the bodies 14 are of larger area, parallel to the membrane 10, than the windows 13 in the support sheets 12 and are aligned therewith. Thus, the apertures 15 in the bodies 14 each define a respective chamber 16, the two chambers 16 being separated by the septum 17 formed by the membrane 10 and the support sheets 12 together, and the aperture 11 in the membrane 10 opening into each of the chambers 16.

The thickness of each body 14 is greater than the thickness of the support sheets 12 and is chosen to provide a desired volume for the chambers 16. In general, the bodies 14 may have any thickness, but typically the thickness of each body 14 is in the range from 1 μm to 3mm. Typically, for use in a disposable portable sensing system, the chambers 16 have a volume of 0.1 μl to 250μl. However, a restricted thickness can be advantageous as described further below. The bodies 14 may be formed of any suitable material, for example silicone rubber.

The chambers 16 are closed by means of a respective closure sheet 18 which is fixed to the outer surface of the respective body 14 covering the aperture 15 formed therein. The closure sheet 18 may be formed from any material, but may for convenience be the same material as the support sheets 12.

The septum 17 including the membrane 10 is not electrically conductive and is designed to have a high electrical resistance. Consequently, in use, the only significant electrical connection between the two chambers 17 is by ionic conduction of an electrolyte solution in the chambers 17 through the aperture 11 in the membrane 10. Formation of a lipid bilayer across the aperture 11 blocks the aperture 11 creating a high-resistance electrical seal between the chambers 17. Insertion of a membrane protein which is an ion channel, for example a pore, restores the electrical connection between the two chambers 17 but only by ionic conduction through the membrane protein. Subsequently, binding events between an analyte and a membrane protein cause a characteristic interruption of the current flowing between the chambers under an applied electrical potential difference.

In order to detect and monitor such electrical signals, each of the chambers 16 is provided with an electrode 20 formed as part of a layer 23 of conductive material deposited on the surface of the respective support sheet 12 which is internal to the chamber 16. In particular, the electrodes 20 are illustrated in Fig. 17 which shows one of the support sheets 12 as viewed from the side internal to the adjacent chamber 16. In Fig. 17, the positions of the aperture 15 in the body 14 and the aperture 11 in the membrane 10 are shown in dotted outline. The conductive material of the electrodes 20 may be for example Ag/ AgCl.

As shown in Fig. 17, the support sheets 12 each include a protruding portion 21 which extends beyond the periphery of the body 14. The layer 23 of conductive material which is deposited on the support sheet 12 to form the electrode 20 extends from the chamber 16 across the support sheet 12 to the protruding portion 21. Accordingly each layer 23 of conductive material forms not only an electrode 20 but also a contact 24 which is exposed on a connector portion 22, and a track 25 which electrically connects the contact 24 and the electrode 20. As described further below, the two protruding portions 21 of the two support sheets 12 together form a connector portion 22 for connecting the cell 2 to the reader unit 3, and the electrical signal received by the electrodes 20 in each chamber 16 is supplied to the reader unit 3 via the contacts 24.

In use, a sample solution is introduced into the chamber 16 on one side of the membrane 10. The chamber 16 which receives the sample solution will now be referred to as the test chamber 16-1 and the other chamber will now be referred to as the secondary chamber 16-2, although in many embodiments both chambers 16 will be identical in size and construction.

To allow introduction of the sample solution, the test chamber 16-1 may be provided with an inlet 30 or 32 using either one of the following two alternative arrangements.

In the first inlet arrangement, the inlet 30 is formed in the body 14 as shown in Fig. 18. In particular, the inlet 30 is formed in one of the surfaces of the body 14 which may in general be either the inner or outer surface as a channel extending from the periphery of the body 14 to the aperture 15. The sample may be injected through the inlet 13, for example using a pipette or syringe. To allow exhaust of air in the chambers 16 displaced by the sample, the test chamber 16-1 is further provided with an exhaust outlet 31 having an identical construction to the inlet 30.

In the second inlet arrangement, the inlet 32 is formed in the closure sheet 18 as illustrated in Fig. 19. In particular, the inlet 32 is formed as a hole extending through the closure sheet 18 and aligned with the aperture 15 in the body 14 which defines the test chamber 16-1, as shown in dotted outline in Fig. 19. To allow exhaust of air in the chambers 16 displaced by the sample, the test chamber 16-1 is further provided with an exhaust outlet 33 having an identical construction to the inlet 32.

Such an inlet 30 or 32 may be provided with a closure, or may be omitted altogether by making a portion of the cell 2 of a material which allows penetration by a syringe for filling the test chamber 16-1.

As a result of the design of the electrode 20 as shown in Fig. 17, the electrode 20 is arranged in the flow path between the inlet 30 or 32 and the aperture 11. In other words, when an aqueous solution is introduced into the test chamber 16-1 through the inlet 30 or 32 it contacts the electrode 20 before reaching the aperture 11. This means that the electrode 20 is wetted before the lipid bilayer is formed, the formation of the bilayer being described in more detail below. When the electrode 20 is wetted, there can occur a pertubation in the potential across the electrodes 20 between the two chambers 16, derived from the reader unit 3. If this occurs before the lipid bilayer is formed, then this causes no difficulty. However if the aqueous solution was to contact the electrode 20 after reaching the bilayer, such a pertubation in the potential across the electrodes could occur after the lipid bilayer is formed and risk rupturing the lipid bilayer.

The secondary chamber 16-2 may, in use contains a buffer solution or a gel. The cell 2 may be supplied to users with the secondary chamber 16-2 already containing the buffer solution or gel. In this case, the secondary chamber 16-2 does not need an inlet 30 or 32 as described above. Alternatively the cell 2 may be supplied with the secondary chamber 16-2 empty. In this case, the user must introduce a buffer solution or gel into the secondary chamber 16-2. To facilitate this the secondary chamber 16-2 may also be provided with an inlet 30 or 32 as described above.

Thus the chambers 16 are closed except for an inlet 30 or 32 if provided. This contrasts with a conventional laboratory apparatus in which chambers on either side of an aperture are formed as recesses in a molded block which are open to the atmosphere. Use of closed chambers 16 has the advantage of reducing evaporation from the contents of the chambers 16. This in turn reduces the cooling of the contents which is important to maintain appropriate temperatures in the case of many membrane proteins which may be inserted in the bilayer.

An internal surface of the test chamber 16-1 has a lipid deposited thereon. When the sample is inserted into the test chamber 16-1, the sample rehydrates the lipids and forms a lipid/solution interface between the sample and the air in the test chamber 16-1. This interface is subsequently moved across the aperture 11, either once or repeatedly, in order to form the lipid bilayer across the aperture 11.

In general, the lipid may be applied to any internal surface of the test chamber 16-1. The lipid may be deposited on the septum 17 during manufacture after the septum 17 has been constructed by fixing together the membrane 10 and the support sheets 12 but before assembly of the septum 17 into the remainder of the cell 2. Alternatively the lipid may be deposited on the internal walls of the chamber 16 formed by the aperture 15 in the body 14 or the closure sheet 18, either before or after the body 14 is fixed to the closure sheet 18, but before assembly to the septum 17.

The deposition may be achieved by coating the septum 17 with a solution of the lipid dissolved in an organic solvent such as pentane and then subsequently allowing evaporation of the solvent, although other techniques could equally be applied.

In general the chambers 16 may be of any size. However, particular advantage is achieved by restricting the depth of the test chamber 16-1 in the direction perpendicular to the septum 17. This depth is controlled by selection of the thickness of the body 14. In particular, the depth is restricted to a level at which the surface tension of a sample solution introduced into the test chamber 16-1 prevents the liquid from flowing across the test chamber 16-1 and instead contains the liquid in part of the test chamber 16-1 across its area parallel to the septum 17. In this state, the liquid interface with the air in the chamber 16 extends across the depth of the chamber 16, perhaps with some meniscus forming depending on the relative pressures of the liquid and the air.

This effect is illustrated in Fig. 20 which shows a cell 2 in which the liquid sample 40 has been introduced into one side of the test chamber 16-1 through the inlet 30 or 32 (although for simplicity the inlet 30 or 32 is not shown in Fig. 20). As can be seen, instead of the liquid sample 40 falling under gravity to the lowest possible level in the chamber 16, surface tension holds the liquid interface 41 with the air in the chamber 16 extending across the depth of the chamber 16 between the septum 17 and the closure sheet 18. Thus, the interface 41 is generally perpendicular to the septum 17 and the aperture 11 except for the formation of a meniscus.

By applying pressure at the inlet 30 or 32 to introduce more liquid or to withdraw the liquid, the interface 41 may be moved in the direction of the arrow A along the chamber parallel to the septum 17 and hence across the aperture 11. Once the liquid sample 40 has rehydrated the dried lipid inside the chamber 16 the liquid interface 41 will support a layer of the lipid. Thus, such movement of the liquid interface 41 across the aperture 11 in the membrane 10 may be used to form a lipid bilayer.

A particular advantage of such a restricted depth for the chamber 16 is that the above- described effect of surface tension occurs irrespective of the orientation of the cell 2. Although the cell 2 is illustrated in Fig. 14 with the aperture 11 extending horizontally, the same effect occurs regardless of the orientation of the cell 2. Thus the above-described process of forming a lipid bilayer across the aperture 11 may be carried out with the cell 2 in any orientation. This reduces the degree of care needed by the user and enhances the ability to use the sensor system outside of a laboratory setting.

The cell 2 is easy to manufacture simply by cutting and affixing together the individual layers of the cell 2. For convenience the layers of the cell 2 are affixed by adhesive, although in principle some form of mechanical fixing could be used. Conveniently due to the use of a layered construction plural cells 2 or parts thereof may conveniently be manufactured together from a large sheet and subsequently cut out. As a result of these points, the cell 2 is capable of mass production at relatively low cost.

By way of example and without limitation, one particular manufacturing method will now be described in detail.

Firstly, a template for plural cells 2 is inkjet printed onto the release paper of adhesive- coated polyester A4 sized cards from which six rows of sixteen support sheets 12 are to be formed. The cards were Mylar polyester sheet (DuPont) of thickness 250μm with a 467MP self-adhesive coating of thickness 50μm on one side. With the release-paper facing upwards, 4mm diameter holes are punched in the cards on the template to provide the windows 13 of each support sheet 12 and any burring of the edges of the punched holes removed using a scalpel blade.

The layers 23 of conductive material are then stencil screen-printed onto the cards using a 60/40 composition silver/silver chloride paste (Gwent Electronic Materials Ltd.), and left overnight to dry at room temperature. The registration and electrical resistance of the layers 23 of conductive material is checked and the surface of the cards covered with a sheet of A4 paper, to keep the surface clean in subsequent stages of sensor production.

With the release paper side facing upwards, the cards are then cut using a guillotine lengthwise into the six rows of support sheets 12. In this example the membranes 10 are formed from either a 6μm thick biaxial polycarbonate film or a 5μm thick PTFE film (Goodfellow Cambridge Ltd.). Prior to use the apertures 11 are formed as discussed below. The membrane 10 around the apertures 11 then receives a chemical pretreatment to facilitate the bilayer formation process. In this case, the pretreatment consists of 2μl of 1% hexadecane in pentane applied to either side of the aperture by capillary pipette.

Once the pentane solvent had evaporated a lμl drop of aqueous protein solution (0.017 mg/ml w.t. α-HL) was applied near to one side of the aperture and dried.

Next the films are cut into strips, cleaned on both sides by rinsing with ethanol, and gently air-dried.

A tape-laying jig with a rubber coated veneer roller is used to roll the membrane film strips evenly over the self-adhesive of one half of the card rows. Care is taken to ensure that the film above the punched holes in the card remained flat and free from creases.

To complete the septums 17, the other half of the card rows are stuck back to back to sandwich the membrane film strips, with the punched holes carefully aligned on either side with the apertures 11. Then the strips are cut using a guillotine into septums 17 for individual cells 2.

In this example the body 14 is formed from a 2mm thick solid silicone rubber sheet with self-adhesive coating on both sides. A large such sheet is cut into A4 sized sheets. An array of 12mm diameter circular apertures 15 for respective cells 2 are formed by removal of the material of the sheet, in particular by hollow punching the spacer sheets. Chamber volumes as low as 56μl have been produced by punching 6mm diameter holes through the 2mm thick spacer material.

The individual chambers 16 are then closed by sticking an A4 sized card of plain 250μm thick Mylar polyester sheet (DuPont), which ultimately forms the closure sheets 18, to one side of the silicone rubber sheet. This sheet is then cut using a guillotine lengthwise into rows having the desired width of the body 14. Channels of width lmm, to form the inlet 30 and exhaust gas outlet 31 are then cut in the silicone rubber sheet material (but not through the backing card).

The interior of each chamber 16 is then coated with a solution of 4μl of lOmg/ml DPhPC lipid dissolved in pentane. The rows of lipid-loaded chambers are cut using a guillotine into individual chambers 16 according to the template and then bonded symmetrically to each side of the individual septums 17 to form cells 2.

The size and formation of the aperture 11 in the membrane 10 will now be considered further.

In general, the aperture 11 may be of any size capable of supporting a lipid bilayer. By way of comparison, the diameter of an aperture in a conventional laboratory apparatus is typically in the order of 30μm to 150μm and an aperture 11 of such a size may used in the present cell 2.

However, it has been appreciated that particular advantage may be achieved by restricting the size of the aperture 11. In particular, this has been found to increase the mechanical stability of the bilayers formed. The increased stability reduces the number of passes of the liquid interface supporting the lipid past the aperture necessary to allow formation of the bilayer. Furthermore, the increased stability increases the robustness of the bilayer and reducing the chances of the bilayer rupturing. This is of particular advantage when the sensor system 1 is used outside a laboratory setting where it may be subject to external forces.

The increased stability achieved by restricting the size of the aperture 11 has been experimentally demonstrated as follows.

A number of actual membranes 10 which have been tested are listed in Table 2 above

The apertures 11 which are sparked-generated were produced by a spark generating device which comprises an adjustable high voltage generator that charges a storage capacitor, with feedback control. The storage capacitor is then switched to discharge into a high voltage transformer coil to rapidly produce a large potential difference between the points of two electrodes attached to the transformer output. Dielectric breakdown between the electrode points results in a spark. The energy of the spark is controlled by switching the value of the storage capacitor (33nF- 30OnF), by adjusting the capacitor charging voltage (200nV-500V), and by changing the distance between the output electrode points.

The polymer film from which a membrane 10 is subsequently cut is mounted flat on the sparking platform and the two output electrodes of the sparking device are positioned opposite each other, above and below the film.

To form apertures 11 of small diameter the spark energy is minimised by choosing the lowest storage capacitor and lowest charging potential that can create a spark that penetrates through the film, and by controlling the dielectric resistance between the two electrodes. For example, decreasing the thickness of the membrane film enabled the use of lower energy sparks and produced smaller apertures, such that it was possible to create apertures in the range 5μm- lOμm diameter in PTFE film of 5μm thickness. Further control of the aperture 11 diameter could easily be introduced through limiting the sparking energy by gating the discharge after detecting the onset of dielectric breakdown.

The laser-generated apertures 11 were produce by laser drilling.

The morphology of the aperture 11 can been seen to vary with the material of the membrane 10 and method used to form the bilayer. For example, with biaxial polycarbonate film, the spark generated apertures 11 were elliptical while the laser drilled apertures 11 were mostly circular. Similarly the spark generated apertures 11 generally had a uniform cross-section while the laser drilled apertures 11 generally a cross section which tapered through the thickness of the membrane 10.

The regularity of the inside edge of the aperture 11 is also sensitive to the material of the membrane 10, the thickness of the membrane 10, and the method of formation of the aperture 11. This is expected to impact on the stability of bilayer formation at the aperture. However in all cases irrespective of the method of formation of the aperture 11, it is apparent that restricting the diameter of the aperture 11 results in increasing the stability of the bilayers, in fact to a dramatic degree. For example with an aperture 11 of diameter lOμm the cell 2 can firmly knocked against the table or disconnected from the reader unit 3 and carried by hand without breaking the bilayer. This is of significant advantage in the context of use of the sensor system 1 outside the laboratory setting.

For these reasons it is preferred that the aperture 11 has a restricted diameter, say of 20μm or less in at least one dimension. The aperture 11 may have such a restricted diameter in all dimensions, but the advantage of increased stability is achieved provided the aperture 11 is relatively small in one dimension, even if the aperture 11 is longer in another dimension.

The work described above demonstrates that apertures 11 of small diameter may be formed using cheap off-the-shelf materials and processes adaptable for mass production. Nonetheless, the choice of materials for the membrane 10 and methods capable of generating the apertures 11 is considerably more extensive than those considered above.

As mentioned above, in one type of cell 2, the secondary chamber 16-2 may contain a gel 50 as shown for example in the cell 2 of Fig. 20. In particular, the gel 50 extends across the aperture 11 in the membrane 10. The presence of the gel acts to physically support a lipid bilayer formed across the aperture 11. As a result, the gel 50 assists the formation of the lipid bilayer and furthermore provides the lipid bilayer with increased stability. Both of these advantages are significant in the context of using the sensor system 1 in a non-laboratory setting, because it makes the sensor system 1 easier to use and also more robust against external forces of the type which may disturb the sensor system 1 in normal use. In addition, the gel 50 may act as a matrix for controlling the supply of molecules to the lipid bilayer.

In order to support the lipid bilayer, the gel 50 may fill the secondary chamber 16-2 such that the gel 50 contacts the membrane 10. This case is illustrated in Fig. 20. In this case, the gel 50 may directly support the lipid bilayer formed across the aperture 11. This is preferred in order to improve bilayer formation and stability.

However, in an alternative illustrated in Fig. 21, there may remain a gap 51 between the gel 50 and the membrane 10. In this case, the gel 50 may still support the lipid bilayer formed across the aperture 11 by acting through a solution occupying the gap 51 , although this effect will reduce as the size of the gap 51 increases. The presence of the gap 51 means that a wider variety of materials can be used to make the gel 50, including ionically non-conductive materials.

The gel 50 may be ionically conductive and indeed this is necessary if the gel 50 directly contacts the lipid bilayer. In this case the gel 50 may be for example a hydrogel. Suitable ionically conductive gels include, but are not limited to, agarose polyacrylimide gel, Gellan™ gel or Carbomer™ gel. Particular gels which have been used are 5% agarose doped with NaCl or Signa Gel (Parker Laboratories Inc.). In one case agarose gel 50 was made using 1OmM PBS to which IM NaCl had been added. The gel 50 was melted and then injected in the chamber 16 where it solidified upon cooling.

It has been discovered that when one chamber 16 of the cell 2 is filled with a gel 50, formation of a lipid bilayer was possible by moving the liquid interface 41 carrying a lipid monolayer past the aperture 11 on only one side of the aperture 11, as opposed to both sides of the aperture 11 as more commonly performed in the Montal & Muller method. Further, bilayers could be formed with or without pretreatment of the membrane 10 by this method. However considerably more attempts were required without the pretreatment. Pretreating only the top side of the membrane 10 was found to be sufficient for reproducible bilayer formation. Being able to apply the pretreatment to only one side of the membrane 10 greatly simplifies the manufacturing process.

The cell 2 may be provided to the user with the secondary chamber 16-2 already containing the gel 50. This improves the ease of use of the cell 2 because no filling the secondary chamber 16-

2 is necessary by the user.

Each of the features described above of (1) restricting the size of the aperture 11, (2) use of a pretreatment and (3) use of a gel 50 assist the formation of a lipid bilayer across the aperture 11 in the membrane 10. In particular, this reduces the number of times in which the interface 41 carrying a lipid monolayer must be moved past the aperture 11 in order to form the bilayer. This improves the ease of use of the cell 2.

In fact, in actual embodiments of the cell 2 employing each of features (1) to (3) there has been demonstrated reliable formation of lipid bilayer on a single pass of the liquid interface 41 pass the aperture 11. This is of significant advantage because it means that the lipid bilayer may be formed across the aperture 11 simply on insertion of the test solution 40 into the cell 2, for example using a pipette or a syringe. This means that the user does not need to repeatedly move the liquid interface 41 back and fourth across the aperture 11 whilst monitoring the formation of the lipid bilayer, and so the required user skill level is greatly reduced. Furthermore, it is not necessary to employ any complicated fluidics control to so move the liquid interface 41.

The reader unit 3 will now be described in detail.

The reader unit 3 has a connector portion 60 which is arranged to make a physical connection with the connector portion 24 of the cell 2. The connector portion 60 of the reader unit

3 is visible in Fig. 14 but is shown in expanded form in Fig. 22. In particular, the connector portion 60 consists simply of a pair of blocks 61 which are separated by a spacing designed to provide a tight fit for the connector portion 24 of the cell 2. Thus, the connector portion 24 of the cell 2 may be plugged into the connector portion 60 in between the blocks 61 by insertion of the cell 2 in the direction of arrow B, thereby providing mating between the connector portions 24 and 60.

In addition, respective contacts 62 and 63 are provided on each of the facing surfaces of the block 61 or the connector portions 60. The contacts 62 and 63 are simply pieces of metal, typically gold-plated to assist formation of good electrical contact. The contacts 62 and 63 may be sprung. When the connector portion 24 of the cell 2 is plugged into the connector portion 60 of the reader unit 3, the contacts 24 of the cell 2 make an electrical connection with the contacts 62 and 63 of the reader unit 3. The reader unit 3 includes an electrical circuit 90 described further below which is connected to the contacts 62 and 63. hi this manner, the connection together of the cell 2 in the reader unit 3 allows the electrical signal generated between the chambers 16 to be supplied from the electrodes 20 to the reader unit 3.

There will now be described some alternatives for providing the cell 2 with a Faraday cage to produce electrical interference from ambient electrical magnetic radiation with the electrical signals generated in the cell 2 when it is connected to the reader unit 3. Two alternative approaches are as follows.

The first approach uses a rigid metal body 70 as the Faraday cage. The rigid metal body has an internal cavity 71 sufficient to accommodate the cell 2. At one end 72, the rigid metal body 70 is open and connected to the body 73 of the reader unit 3 so that the cavity 71 is aligned with the connection portions 60. In this way, the cell 2 is accommodated inside the cavity 71 when it is connected to the reader unit 3, as shown in Fig. 24.

However, rather than entirely enclosing the cell 2, the rigid metal body 70 has an aperture 74 facing the connector portion 60. The aperture 74 is of sufficient size to allow passage of the cell 2 when the cell 2 is connected to the reader unit 3. Therefore, an individual cell 2 may be connected to the reader unit 3 and replaced by another cell 3 by insertion through the aperture 74 without removal of the rigid metal body 70. It has been appreciated that surprisingly the presence of the aperture 74 does not prevent the operation of the rigid metal body 70 as a Faraday cage. In particular, this is because the aperture 74 may be of sufficiently small size that any electrical interference caused by electro magnetic radiation penetrating the aperture 74 is at a sufficient high frequency that it does not significantly degrade the quality of the electrical signal of interest. In particular, the aperture 74 of the rigid metal body 70 may have a maximum dimension (horizontally in Fig. 23) of 50mm or less, preferably 20mm or less.

The rigid metal body 70 also has a sample introduction hole 76 which is aligned with the inlet 30 or 32 when the cell 2 is connected to the reader unit 3. The sample introduction hole 76 allows the sample to be introduced into the cell 2 after the cell 2 has been connected to the reader unit 3. The sample introduction hole 76 is smaller than the aperture 74, typically having a maximum dimension of 5mm or less. Thus the sample introduction hole 76 is also of sufficiently small size that any electrical interference caused by electro magnetic radiation penetrating the sample introduction hole 76 is at a sufficient high frequency that it does not significantly degrade the quality of the electrical signal of interest.

The second alternative approach is to provide a Faraday cage 75 fixed around the periphery of the cell 2, for example as shown in Fig. 25. In this case, the Faraday cage 75 entirely encloses the cell 2, except for the connector portion 24 which protrudes out of the Faraday cage 75. In this case, the Faraday cage 75 may be formed by a solid metal body. Alternatively, the Faraday cage 75 may be formed by a metal foil which has the advantage of being easy to manufacture, for example simply by adhering the metal foil to the exterior of the cell 2.

It is noted that the provision of a Faraday cage attached around the exterior of the cell 2 is equally applicable to other types of electrical sensor cell which are operative to detect an analyte by measurement of an electrical signal developed in the cell.

The reader unit 3 houses an electrical circuit 90 which will now be described in detail. The primary function of the electrical circuit 90 is to measure the electrical current signal developed across the electrodes 20 to provide a meaningful output to the user. This may be simply an output of the measured signal or may involve further analysis of the signal.

The electrical circuit 90 may take various different forms and some possible circuit designs are shown in Figs. 14 to 16. In each design there are some common elements as follows.

The two contacts 62 and 63 of the connector portion 60 will be referred to as a reference contact 62 and a working contact 63. Although the electrodes 62 and 63 are physically the same, in operation the reference contact 62 provides a bias voltage potential relative to the working contact 63, whilst the working contact 63 is at virtual ground potential and supplies the current signal to electrical circuit 90.

A possible alternative which is not illustrated would be for the reference contact 62 to be held at ground and working contact 63 to be offset by the bias voltage.

The reader circuit 90 has a bias circuit 91 connected to the reference contact 62 and arranged to apply a bias voltage which effectively appears across the two contacts 62 and 63 and hence across the electrodes 20 of a cell 2 connected to the reader unit 3. The bias circuit 91 may take different forms as described below.

The reader circuit 90 also has an amplifier circuit 92 connected to the working contact 63 for amplifying the electrical current signal the electrodes 20 of the cell 2 and appearing across the two contacts 62 and 63. In each design of the electrical circuit 90, the amplifier circuit 92 consists of a first amplifier stage 93 and a second amplifier stage 94.

The first amplifier stage 93 is connected to the working electrode 63 and arranged to convert the current signal into a voltage signal in a first stage amplifier. It may comprise an electrometer operational amplifier configured as an inverting amplifier with a high impedance feedback resistor, of for example 500MΩ, to provides the gain necessary to amplify the current signal which typically has a magnitude of the order of tens to hundreds of picoamps.

The second amplifier stage 94 is connected to the output of the first amplifier stage 93 and arranged to amplify and filter the voltage signal voltage. The second amplifier stage 94 provides sufficient gain to raise the signal to a sufficient level for processing in the microcontroller 95 described below. For example with a 500MΩ feedback resistance in the first amplifier stage 93, the input voltage to the second amplifier stage 94, given a typical current signal of the order of 10OpA, will be of the order of 5OmV, and in this case the second amplifier stage 94 must provide a gain of 50 to raise the 5OmV signal range to 2.5V. If the signal contains frequencies beyond the bandwidth limit of the first stage then analogue filtering is provided in the second amplifier stage 94 to increase gain at frequencies beyond the first stage bandwidth limitation. The filtering results in a combined first and second stage frequency response with constant gain beyond the first stage limitation.

To save power, the analogue circuitry in the bias circuit 91 and the amplifier circuit 92 is shutdown when not being used. Each power rail is connected to bipolar PNP switching transistors for low leakage switching of the analogue circuitry.

Typically the signal will be unipolar, but if bipolar current signals are required the gain of the second amplifier stage 94 can be halved and a DC offset applied to the inverting input of the second amplifier stage 94 equal to half reference voltage value of the microcontroller 95.

The first design of the electrical circuit 90 shown in Fig. 26 and will now be described. This design is intended for a stand-alone battery-operated reader unit 3 with PC connectivity. In this case, the bias circuit 91 and the amplifier circuit 92 are connected to a microcontroller 95. The microcontroller 95 has a power control circuit 96 which supplies power from a battery. The microcontroller 95 incorporates an analog-to-digital converter 97 which receives the output of the amplifier circuit 92 and converts it into a digital signal. The analog-to-digital converter 97 may be of a successive approximation type or of a voltage-to-frequency type, both resulting in a digital word for each conversion. A sampling rate is chosen that is at least twice the bandwidth of the signal at the output of the second amplifier stage 94 to prevent aliasing.

In this case the analog-to-digital converter 97 is embedded on the same silicon die as the microcontroller 95, but it could alternatively be a separate circuit element.

The microcontroller 95 incorporates a microprocessor 98 which runs code to process and analyse the digital signal. The microcontroller 95 has a display 99 which is conveniently an LCD display, and on which the microcontroller causes display of the signal itself or other analysis results such as temporal results of the signal analysis.

The microcontroller 95 receives commands from a keypad 100. Of course other input and output devices could be used in addition to, or instead of, the display 99 and keypad 100, for example LEDs used as indicators or an audio generator 105.

The microcontroller 95 also has an interface 101 to provide data communication with another digital device, for example a computer. The interface 101 may be of any type, for example a UART interface. This allows the received signal to be supplied to another device for display, storage and/or further analysis.

The microcontroller 95 is connected to the bias circuit 91 as follows. The microcontroller 95 has a PWM generator 102 which generates a PWM (pulse width modulation) voltage waveform, that is a digital signal with fixed frequency but varying duty cycle. The PWM generator 102 is of conventional construction. Generally, an internal timer is set running to generate the PWM signal frequency and a register is loaded with the count at which the PWM output is switched and a comparator detects when the count is reached.

The bias circuit 91 includes a low-pass filter 103 connected to low-pass filter the PWM signal output by the PWM generator 102. The duty cycle of the PWM signal varies with time so that the output of the low-pass filter is the desired analog signal, which is the average voltage over one period of the PWM cycle. The PWM generator 102 built in this manner has a resolution equivalent to the smallest duty cycle change possible with the microcontroller 95. Bipolar outputs can be achieved by using a pair of PWM signals each connected to one of a pair of low pass filters 103 and one fed to the positive input and the other the negative input of a summing amplifier, this being shown in Fig. 26.

The bias circuit 91 further includes an output amplifier 104 for amplifying the output of the low-pass filter 103. In the case described above that a bipolar output is required, the output amplifier 104 is a summing amplifier arranged to subtract the output of one of the pair of low pass filters 103 from the other.

For systems requiring multiple or arrayed cells 2, the microcontroller 95 can be chosen with an embedded analogue multiplexer . In this case multiple analogue input circuits are required and the output of each second amplifier stage 94 is sampled by the analog-to-digital converter 97 through the multiplexer.

The second design of the electrical circuit 90 is shown in Fig. 27 and will now be described. This design is intended for a reader unit 3 which is a derivative of a standard Personal Digital Assistant (PDA) architecture. The second design is identical to the first design except that the microcontroller 95 interfaces with a PDA device 106 which is a conventional PDA. This allows the reader unit 3 to take advantage of the existing functionality of PDAs. The PDA device 106 may have input/output facilities based on a variety of protocols, such as universal connectors, Secure Digital cards (SD), Compact Flash cards (CF, CF2), MultiMedia cards (MMC), memory stick cards or SIM card. Such functionality may be used to provide a framework for the reader unit 2 to provide the functions of a large interactive display with key or touch entry and a rechargeable power source.

In this case, one option is for the connector portion 60, the amplifier circuit 92, the bias circuit 91 and the microcontroller 95 to be mounted within an electrical assembly shaped to fit in an SD card slot or other card format slot. This allows the reader unit 2 to be formed by an existing PDA device with the assembly fitted in a card slot.

The third design of the electrical circuit 90 is shown in Fig. 28 and will now be described. This design is intended for a reader unit 3 which is based on a data acquisition card 107 to be plugged into a computer 108 such as a desktop or laptop. This design is the simplest in terms of hardware development requiring only three amplifier stages and the data acquisition card. In this case the amplifier circuit 92 is arranged as described above, but the bias circuit 91 is simply formed by an inverting amplifier 109 supplied with a signal from a digital-to-analog converter 110 which may be either a dedicated device or a part of the data acquisition card 107 and which provides a voltage output dependent on the code loaded into the data acquisition card 107 from software.

The third design of the electrical circuit 90 shown in Fig. 28 may be modified to provide a multi-port reader system connected through a fast transport interface such as the Universal Serial Bus or Ethernet for the purpose of analysing many cells at once. In work involving drug-screening or an industrial manufacturing environment there is a need for multiple readers connected to a central computer for research, analysis and quality control. In this case the data acquisition card 107 is modified to provide the transport interface allowing multiple data streams into the computer.

The electrical circuit 90 may provide analysis of the received signal. Such analysis may be performed, for example, by programming one of the microprocessors in the electrical circuit, for example the microprocessor 98 in the microcontroller 95 or the PDA device 106 in the above described designs of the electrical circuit. In particular the analysis may involve interpretation of the electrical signal. As already described, the electrical signal is characteristic of the physical state of the cell 2. Accordingly, the state of the cell 2 can be detected from the electrical signal by the electrical circuit 90.

For example, when the cell 2 is used as described above, the following states each have a characteristic electrical signal which may be detected by the electrical circuit 90:

1) the chambers 16 in the cell 2 being dry;

2) the chambers 16 in the cell 2 containing an aqueous solution without a lipid bilayer being formed across the aperture 11 in the membrane 10;

3) a lipid bilayer being formed across the aperture 11 in the membrane 10 without a membrane protein being inserted therein;

4) a lipid bilayer being formed across the aperture 11 in the membrane 10 with a membrane protein being inserted therein without an analyte binding to the membrane protein; and

5) a lipid bilayer being formed across the aperture 11 in the membrane 10 with a membrane protein being inserted therein with an analyte binding to the membrane protein.

Such states may be detected based on predetermined thresholds or adaptive thresholds, which may be derived from scientific study of the membrane protein and physical system being used in the cell 2. On detection of such a state, the electrical circuit 90 then produces an output indicative of the detected state, for example by displaying the detected state on the display 99 or some other audio and/or visual output, or by outputting a signal indicative of the detected state, for example to a computer device connected thereto.

By detecting the continuous sequence of states (1) to (5) in order, the reader unit 2 may also monitor the correct performance of the sensing process to check and ensure that the cell 2 is operating correctly from the moment it is connected to the reader unit 3 until the end of the measurement assay. The reader unit 3 may apply a bias potential and continuously monitor the resultant signal. If the signal falls outside the expected levels showing a proper progress through the states (1) to (5), the reader unit 3 may output a signal reporting an error mode, or alternatively may perform an automated remediation.

As each state is detected the time duration of the state will be stored for subsequent or continuous statistical analysis. This may provide further information. For example, signals derived from single molecule binding events in or near multiple membrane protein channels will result in a time-varying current based on the number of binding events.

Another example is where the membrane protein includes a tether. Signals derived from either single or multiple binding events to either single or multiple tethers attached to single or multiple membrane protein channels will appear as noisy signals which become less noisy when the tether or tethers are bound to a target analyte. Each tether will have a binding site for the target analyte. These signals will be analysed with an algorithm to detect the reduction in noise and as each event is detected the time duration of the event or the time course of noise reduction will be stored for subsequent or continuous statistical analysis.

There will now be described an actual example of the algorithm used to monitor of the state of the cell 2 in the case using the membrane protein α-HL to sense the presence of the analyte γ-cyclodextrin. The electrical circuit 90 performs the process as shown in Fig. 29.

In an initialisation step Sl performed before connection of the cell 2 to the reader unit 3, the electrical circuit 17 applies a bias voltage as shown in Fig. 30 having a waveform which is a 50Hz triangular AC signal with 2OmV amplitude, superimposed on +10OmV DC potential.

In step S2 it is detected whether the received signal is representative of a current and impedance within the respective limits for the reader unit 3 in the absence of the cell 2. In the absence of the cell 2, the contacts 62 and 63 of the reader unit 3 behaves as a capacitor and produce a square wave current response to the applied triangular AC potential, as shown in Fig. 31. In particular the square wave has a 2OpA amplitude centred on OpA. This waveform is characteristic of normal operation of the electrical circuit 90 and so in step S2 it is detected whether this waveform is produced, within a reasonable margin. If not, then in step S3, the electrical circuit 90 outputs a signal indicate indicative of a circuit error. Otherwise in step S4, the user connects a cell 2 to the reader unit 3. The electrical circuit 90 may for example await a user input to indicate this.

Subsequently in step S5, there is detected state (1) that the chambers 16 in the cell 2 are dry. In this case, the expected signal is the same as that detected in step S2 except that the insertion of the cell 2 causes an increase, for example the order of 25%, in the amplitude of the resultant squarewave, for example to provide an amplitude of 27pA. If state (1) is not detected, then in step S6 and there is output an error signal indicating malfunctioning of the cell 2.

Otherwise, in step S7 there is output a signal indicating state (1) and in step S 8 the electrical circuit 90 changes the bias potential by removing the DC component, but maintaining the AC voltage of the waveform shown in Fig. 30. In step S9, the user introduces the test solution into the cell 2.

In this particular implementation, state (2) is not detected, but in step SlO there is detected state (3) of the lipid bilayer being formed across the aperture 11, as follows. In the absence of a lipid bilayer, the aperture 11 provides a conductive path between the electrodes 20 and so the cell 2 provides a current response. Typically the current saturates the amplifier, for example as shown in the typical response shown in Fig. 32.

In contrast, formation of the lipid bilayer prevents flow of ionic current through the aperture 11 and so the cell 2 provides a capacitive response. As a result, the resultant current signal is a squarewave as shown in Fig. 33 typically having an amplitude of around 25OpA centred on OpA. State (3) is detected in step SlO by detecting a current signal showing this capacitive response. Typically the DC resistance is greater than 10GΩ.

If state (3) is not detected, then in step Sl 1 the detected current is compared to a threshold and then depending on whether the threshold is exceed or not there is output one of two possible error signals in steps S12 and S 13 which indicate the absence of bilayer formation.

However, if state (3) is detected in step SlO, then in step S 14 there is output a signal indicating that state (3) has been detected and in step Sl 5 the bias voltage is changed by removing the AC waveform and instead applying a DC waveform.

In step S 16 there is detected state (4) of a membrane protein being inserted into the lipid bilayer formed across the aperture 11. This is detected by detection of the predictable step increases in the DC current response which occurs on insertion of the membrane protein due to the ionic current flowing through the ion channel. This is shown in Fig. 34 which shows the current increasing by a step of the order of 95pA on insertion of single α-HL membrane protein. In this example, one such insertion occurs at around 0.1 minutes and a second insertion occurs at around 1.7 minutes. Since the electrical composition of the solution and the bias potential are known, the total current reflects the total number of membrane proteins inserted and this information may be determined and subsequently used to calibrate the assay calculations.

If state (4) is not detected within a reasonable period then there is output in step S 17 an error signal indicating failure of insertion. Otherwise, in step Sl 8 there is output a signal indicating that state (4) has been detected.

Thereafter, in step S 19 there is detected state (5) of an analyte binding to the membrane protein. This may be detected as follows. When the analyte binds to the membrane protein this temporarily interrupts the ironic current passing through the ion channel causing a characteristic step decrease in the current. Prior knowledge of the analyte binding characteristics (eg current deflection and distribution in event duration) allows the electrical circuit 90 to identify the relevant binding events. An example of the current is shown in Fig. 35. The analyte γ-cyclodextrin causes a decrease in the current of the order of 6OpA. Four such binding events are evident in Fig. 35. The electrical circuit 90 detects these characteristic changes as binding events. A signal indicative of this is output in step S20. To detect successive binding events, steps S19 and S 20 are repeated.

Finally in step S21 the concentration of the analyte γ-cyclodextrin is calculated based on the kinetics of the measured analyte binding.

Claims

1. A method for forming a lipid bilayer across an aperture, comprising:

(a) providing a cell having a chamber adjacent to a septum comprising a membrane having an aperture capable of supporting a lipid bilayer;

(b) depositing one or more lipids on an internal surface of the chamber;

(c) introducing an aqueous solution into the chamber to cover the aperture and the internal surface and to form an interface between the solution and lipids; and

(d) moving the interface past the aperture at least once to form a lipid bilayer across the aperture.

2. A method according to claim 1 , wherein the lipids are dried.

3. A method according to claim 2, wherein the dried lipids comprises less than 50wt% solvent.

4. A method according to any one of the preceding claims, wherein the aqueous solution covers the internal surface before it covers the aperture.

5. A method according to any one of the preceding claims, wherein the internal surface on which the lipids are deposited is on the septum.

6. A method according to any one of the preceding claims, wherein the aperture has a diameter in at least one dimension which is 20μm or less.

7. A method according to any one of the preceding claims, further comprising pre- treating the membrane to increase its affinity to lipids.

8. A method according to claim 7, comprising pre-treating the membrane with hexadecane.

9. A method according to any one of the preceding claims, wherein the cell has two chambers.

10. A method according to claim 9, wherein the method comprises depositing the lipids on an internal surface of both chambers.

11. A method according to claim 9 or 10, wherein the aqueous solution is introduced into one chamber in step (c) and the other chamber comprises a gel.

12. A method according to claim 11 , wherein the gel is a hydrogel.

13. A method according to claim 11 or 12, wherein the gel comprises one or more membrane proteins.

14. A method according to any one of the preceding claims, wherein step (b) further comprises depositing one or more membrane proteins on the same or different internal surface, step (c) further comprises introducing the aqueous solution into the chamber to cover the membrane proteins and the method further comprises allowing the membrane proteins to insert into the lipid bilayer.

15. A method according to claim 14, wherein the internal surface on which the one or more membrane proteins are deposited is on the septum.

16. A method according to claim 14 or 15, wherein the membrane proteins are dried.

17. A method according to claim 16, wherein the membrane proteins comprise less than 20wt% solvent.

18. A method according to any one of claims 13 to 17, wherein the one or more membrane proteins comprise α-hemolysin or a variant thereof.

19. A method according to any one of the preceding claims, wherein the septum further comprises a support sheet on at least one side of the membrane.

20. A method according to any one of the preceding claims, wherein the membrane is made from polycarbonate (PC), polytetrafluoroethylene (PTFE), polyethylene, polypropylene, nylon and polyethylene naphthalate (PEN), polyvinylchloride (PVC), polyacrylonitrile (PAN), polyether sulphone (PES), polyimide, polystyrene, polyvinylfluoride (PVF), polyethylene telephthalate (PET), aluminized PET, nitrocellulose, polyetheretherketone (PEEK) or fluoroethylkene polymer (FEP).

21. A method according to any one of the preceding claims, wherein the lipids comprise diphantytanoyl-sn-glycero^-phosphocholine.

22. A device for forming a lipid bilayer comprising:

(a) a cell having a chamber adjacent to a septum comprising a membrane having an aperture capable of supporting a lipid bilayer; and

(b) one or more lipids deposited on an internal surface of the chamber, wherein the cell comprises an inlet for introducing an aqueous solution into the chamber having lipid deposited therein.

23. A method according to claim 22, wherein the lipids are dried.

24. A method according to claim 23, wherein the dried lipids comprises less than 50wt% solvent.

25. A device according to any one of claims 22 to 24, wherein the inlet, internal surface and aperture are arranged in such a manner that the aqueous solution covers the internal surface before it covers the aperture.

26. A device according to any one of claims 22 to 25, wherein the internal surface on which the lipids are deposited is on the septum.

27. A device according to any one claims 22 to 26, wherein the aperture has a diameter in at least one dimension which is 20μm or less.

28. A device according to any one of claims 22 to 27, wherein the membrane has a pre- treatment to increase its affinity to lipids.

29. A device according to claim 28, wherein the membrane has a hexadecane pre- treatment.

30. A device according to any one of claims 22 to 29, wherein the cell has two chambers.

31. A device according to claim 30, wherein the inlet opens into one chamber and the other chamber comprises a gel.

32. A device according to claim 31 , wherein the gel is a hydrogel.

33. A device according to claim 31 or 32, wherein the gel comprises one or more membrane proteins.

34. A device according to any one of claims 22 to 33, further comprising one or more membrane proteins deposited on the same or different internal surface.

35. A device according to claim 34, wherein the internal surface on which the one or more membrane proteins are deposited is on the septum.

36. A device according to claim 34 or 35, wherein the membrane proteins are dried.

37. A device according to claim 36, wherein the dried membrane proteins comprise less than 20wt% solvent.

38. A device according to any one of claims 33 to 37, wherein the one or more membrane proteins comprise α-hemolysin or a variant thereof.

39. A device according to any one of claims 22 to 38, wherein the septum further comprises a support sheet material on at least one side of the membrane.

40. A device according to any one of claims 22 to 39, wherein the membrane is made from polycarbonate (PC), polytetrafluoroethylene (PTFE), polyethylene, polypropylene, nylon and polyethylene naphthalate (PEN), polyvinylchloride (PVC), polyacrylonitrile (PAN), polyether sulphone (PES), polyimide, polystyrene, polyvinylfiuoride (PVF), polyethylene telephthalate (PET), aluminized PET, nitrocellulose, polyetheretherketone (PEEK) or fluoroethylkene polymer (FEP).

41. A device according to any one of claims 22 to 40, wherein the lipids comprise diphantytanoyl-sn-glycero-3-phosphocholine.