WO2014177670A1

WO2014177670A1 - Polymeric device for electrophysiological recordings

Info

Publication number: WO2014177670A1
Application number: PCT/EP2014/058945
Authority: WO
Inventors: Simone TANZI; Rafael J. TABORYSKI
Original assignee: Sophion Bioscience AS
Current assignee: Sophion Bioscience AS
Priority date: 2013-05-01
Filing date: 2014-05-01
Publication date: 2014-11-06
Anticipated expiration: 2015-11-01

Abstract

The invention provides a microfluidic chip, suitable for recording ion-currents across the cell membrane of a living biological cell. The microfluidic chip is formed of a thermoplastic polymer. A method for the fabrication of such a chip via injection-moulding is also provided, as well as a microfluidic device comprising said microfluidic chip.

Description

POLYMERIC DEVICE FOR ELECTROPHYSIOLOGICAL RECORDINGS

FIELD OF THE INVENTION

The present invention relates to a microfluidic chip, suitable for recording ion-currents across the cell membrane of a living biological cell. A method for the fabrication of such a chip via injection-moulding is also provided, as well as a microfluidic device comprising said microfluidic chip.

BACKGROUND OF THE INVENTION

Microfluidic analyses of biological systems are widely used in medical and biological research in order to assess the mutual effects of various combinations of reagents and samples on living biological cells. So-called microtitre plates have been developed, which are flat plates with a plurality of wells used as small reactor chambers. Such microtitre plates have become a standard tool in analytical research and clinical diagnostic testing laboratories.

In particular, with chips to be used in patch clamp techniques, good adhesion of the cell to the chip is required, so that a high-resistance seal (in the order of GOhms) can be obtained between the chip and the cell membrane (a "gigaseal").

Microfluidic chips are often fabricated from micromachining glass or silicon, or by casting polydimethylsiloxane (PDMS). However, fabrication of the requisite microfluidic channels in these materials is expensive and time-consuming. PDMS has the additional drawback that softeners often migrate or leach from the PDMS. Furthermore, after use, microfluidic chips have proved difficult to clean to the high level of cleanliness required for re-use.

US2002/0110847 describes a method for measuring a state variable of a biological cell.

Other microfluidic systems are described in US2007/0155016, WO 99/19056, US

2008/0305011, WO 01/94939 and US 2003/0153067.

S. Tanzi et al. J. Micromech. Microeng 22 (1012) 115008 discloses single-cell capture devices produced by injection moulding. However, this document does not disclose that gigaseals can be produced using these devices. In fact, the cell-capture devices in this document function as cell-capture devices, not patch-clamp devices. It is also noted that the roughness measured in this document is measured at the bonding surfaces; the roughness in the channels is "somewhat higher".

The present invention provides the advantage of enabling cost-effective, reproducible production of microfluidic chips with the required characteristics (i.e. capable of obtaining a Gigaseal) by means of injection-moulding only one polymer part comprising microfluidic channels. Use of commercial injection-moulding techniques could lower the price of microfluidic chips, making them disposable rather than re-useable devices.

SUMMARY OF THE INVENTION

In a first aspect the present invention relates to a method for the fabrication of a microfluidic cell-capture chip, said microfluidic chip comprising a predetermined pattern of microfluidic channels, said method comprising the steps of: a. providing a shaped silicon master comprising said predetermined pattern of

microfluidic channels; b. depositing a layer of silicon oxide on top of said shaped silicon master; c. depositing a metal seed layer onto the layer of silicon oxide produced in step b.; d. electroforming a metal shim onto said metal seed layer, such that said metal shim comprises a relief pattern corresponding to said predetermined pattern of microfluidic channels in said shaped silicon master; e. removing the shaped silicon master from the metal shim, a nd; f. injection-moulding a microfluidic chip from a thermoplastic polymer using said metal shim, such that the predetermined pattern of microfluidic channels are moulded into said microfluidic chip.

By means of the method of the present invention, relatively expensive microfabrication steps are only carried out once on a single silicon master and substantially identical high-quality polymer chips can be produced cheaply, efficiently and in high quantity. The invention also provides a microfluidic cell-capture chip formed of a thermoplastic polymer, said microfluidic chip comprising at least one extracellular microfluidic channel, at least one intracellular microfluidic channel and at least one shallow microfluidic channel extending between said extracellular and said intracellular microfluidic channels, in which all microfluidic channels have a surface roughness below 5nm, suitably below 3nm and wherein at least one shallow microfluidic channel comprises a straight extracellular region which extends a distance of 5 to 50 μιη in a straight line from the point at which said shallow channel meets said extracellular channel. A microfluidic device comprising (a) the microfluidic cell-capture chip according to the invention, and (b) a lid formed of a thermoplastic polymer bonded to said microfluidic chip is also provided.

LEGENDS TO THE FIGURES

Figures la-c illustrate the principle of the microfluidic cell-capture device: la) Cells move in extracellular channel, lb) One single cell will be trapped at the aperture formed by the shallow channel by application of a negative pressure difference across the shallow channel. lc) The effect of a test substance on the trapped cell is studied.

Figure 2: Various atomic force microscope images of various microfluidic cell-capture chips. 2a) Injection-moulded polymer cell-capture chip according to the invention, including TEOS deposition step during metal shim fabrication. 2b) Injection-moulded polymer cell-capture chip fabricated without TEOS deposition step during metal shim fabrication (i.e. according to S. Tanzi et al. J. Micromech. Microeng 22 (1012) 115008).

Figure 3 is a three-dimensional illustration of a microfluidic device according to the invention, showing intra- and extracellular microfluidic channels, a shallow microfluidic channel connecting these channels and a cell trapped at the aperture formed by the shallow microfluidic channel in the wall of the extracellular microfluidic channel. Figure 3 also shows the lid and microfluidic chip.

Figures 4a - f illustrate the injection moulding process of the invention. Figures 5a-f illustrate how the shaped silicon master is formed. Figure 6 illustrates how the microfluidic device is set up.

Figure 7: a) A family of Na⁺ activation currents in response to depolarization pulses from -90 to 70 mV. Protocol used for determining both activation and inactivation is also shown. The membrane potential was held at a holding potential of -90 mV, subsequently shifted to potentials ranging from -90 to 70 mV for 1000 ms, and finally to 10 mV. To the left the raw current responses are shown and to the right the data after leak subtraction, b) A family of Na-i- inactivation currents from the same recording upon the step to lOmV. To the left the raw current responses are shown and to the right the data after leak subtraction, c) The resulting I-V relationship for peak Na-i- channel currents. The activation threshold was -50 mV, and the maximal current amplitude was obtained at -30 mV for the polymer device. The threshold was -30 mV with a maximum at -lOmV for the QPatch™. d) The inactivation graph for the Na-i- channel. At potentials more negative than -80mV the channels were

predominantly closed, whereas at potentials above -30mV they were predominantly inactivated for the QPatch™. In the case of polymer device the channels were predominantly inactivated at potentials above -50mV.

Figure 8: Activation currents for Navl.7 channels for determining the lidocaine inhibition on Navl.7 channels. The graph shows values of the currents amplitude of peak 1 and 2 before and after ΙΟΟμΜ lidocaine was applied to the cell. Peak 1 isn't affected and peak 2 is reduced by 50%. Figure 9: Concentration-response relationships of lidocaine inhibition on Navl .7 channels. Data were fitted with a Hill type equation. IC50 was 119 ±11μΜ (Hill coefficient n = 1) for the polymer device and 152 ±12μΜ (Hill coefficient n = 0.9) for the QPatch™.

DETAILED DISCLOSURE OF THE INVENTION

Definitions The terms "aperture" and "patch hole" are used interchangeably in this text to indicate the opening formed by the shallow microfluidic channel at the junction of the extracellular microfluidic channel.

Microfluidic cell-capture chip The invention provides a microfluidic cell-capture chip. The cell-capture chip is injection- moulded from a hard thermoplastic polymer; i.e. the chip is constituted by thermoplastic polymer. Suitable thermoplastic polymers are selected from polyethylene polymers, polypropylene polymers, polystyrene polymers, polyacrylate polymers, polyoxymethylene polymers, polyamide polymers, polycarbonate polymers or polyetheretherketone polymers and cyclic olefin polymers, and blends and co-polymers thereof. Of these, preferred polymers are acyclic olefin polymer (COC), a polyethylene polymer, a polypropylene polymer, a polystyrene polymer or a polyacrylate polymer. The thermoplastic polymer is preferably a cyclic olefin polymer, due to their high chemical resistance, low water absorption and their previous use in biological applications. By the term "hard" is meant that the microfluidic chip has a Young's modulus greater than 1000 MPa.

The microfluidic cell-capture chip of the invention is substantially planar with two opposing surfaces. Typically, the cell-capture chip is disc-shaped, with a thickness of 0.5-3 mm, typically 2mm, and a diameter of 50-150mm, typically circa 100mm. However, the skilled person will be able to design cell-capture chips with other overall shapes, while remaining within the scope of the invention.

The microfluidic cell-capture chip of the invention has an arrangement of channels moulded into one surface thereof. The relative "depth" of the channels, and the "bottom" and "sides" or "sidewalls"" of the channels, are described relative to this surface. This means that a "deep" channel will extend further into the material of the chip from said surface than a "shallow" channel. According to the invention, the cell-capture chip comprises at least one extracellular microfluidic channel, at least one intracellular microfluidic channel and at least one shallow microfluidic channel extending between said extracellular and said intracellular microfluidic channels. Therefore, the microfluidic chip comprises one extracellular and one intracellular microfluidic channel per shallow microfluidic channel. Depending on the layout of the microfluidic chip, the chip may comprise more than one intracellular channels per extracellular channel, for example 2, 3 or 4 intracellular channels per extracellular channel.

The extracellular microfluidic channel typically has a depth of 30-100 μιη, typically circa 50 μιη and a width of 50-400 μιη, typically circa 200 μιη. The intracellular microfluidic channel typically has a depth of 30-100 μιη, typically circa 50 μιη and a width of 50-400 μιη, typically circa 200 μιη. These deep microfluidic channels preferably have a tapered cross-sectional profile, allowing easy de-moulding in the injection-moulding process. The shallow microfluidic channel typically has a depth of 1.5-3 μιη , typically circa 2 μιη and a width of 1.5-3 μιη , typically circa 2 μιη. The dimensions of the shallow microfluidic channel at the junction with the extracellular microfluidic channel are smaller than the size of the cells to be captured. The shallow microfluidic channel typically has a shape, in the plane of the cell-capture chip, which is wider at the junction with the intracellular channel and tapers to junction with the extracellular channel. This causes a reduction of the total hydraulic resistance of the shallow channel, making cell capture easier. In addition, the shallow microfluidic channel may comprise one or more support structures which extend from the bottom of said channel to prevent collapse of the shallow channel. The shallow microfluidic channel may have a semi-circular cross-sectional profile, or a rectangular profile with rounded corners.

The extracellular and intracellular microfluidic channels are arranged such that liquid, optionally a cell suspension, can be passed through each channel independently. The arrangement of pumps, liquid reservoirs, connectors, valves etc. required to achieve this can be selected by the skilled person. As the microfluidic chips of the present invention are injection-moulded, the scope for connectors etc. is large. The microfluidic chip of the invention typically comprises Luer-type connectors on the surface opposite that containing the channels.

Figures la-c illustrate the principle of the microfluidic cell-capture chip 100, and shows an extracellular microfluidic channel 120, an intracellular microfluidic channel 140, and a shallow microfluidic channel 130 extending between them. A suspension of cells 121 is introduced into the extracellular channel 120 (Figure la). One single cell 121a will be trapped at the aperture formed by the shallow channel 130 by application of a negative pressure difference across the shallow channel 130 (Figure lb). The cell is perforated, and a test substance is introduced into the extracellular channel 120, and its effect on the captured cell is studied. Cell perforation often occurs spontaneously upon cell capture; alternatively combinations of suction pulses (e.g. -30mbar to -400mbar) and electrical pulses can be used to perforate the cells.

Thermoplastic polymers generally have low surface energy. To ensure that a gigaseal is achieved between the microfluidic cell-capture chip, the microfluidic channels firstly have a surface roughness below 5nm, suitably below 3nm. Surface roughness is measured by atomic force microscopy. Secondly, in the microfluidic chips according to the invention, the at least one shallow microfluidic channel comprises a straight extracellular region which extends a distance of 5 to 50 μιη in a straight line from the point at which said shallow channel meets said extracellular channel. This straight extracellular region of the shallow microfluidic channel typically extends in a direction perpendicular to the extracellular channel. This allows the captured cell to extend into the shallow microfluidic channel a distance which is sufficient for establishing a gigaseal. This straight extracellular region in the shallow channel of the microfluidic chip is seen most clearly in Figure 2f.

Microfluidic device The invention also provides a microfluidic device 10 comprising (a) the microfluidic cell- capture chip 100 according to the invention, and (b) a lid 200 formed of a thermoplastic polymer bonded to said microfluidic chip. This is illustrated in more detail in Figure 3. The lid is bonded so as to substantially cover the channels moulded in the microfluidic chip, and preferably covers the entire surface of the microfluidic chip containing said channels. Secure bonding between the lid 200 and the microfluidic chip 100 is required to prevent leakages. The lid 200 may be bonded to the microfluidic chip 100 by means of an adhesive, , by partially melting lid 200 and/or chip 100, by laser bonding, ultrasonic welding, UV-assisted thermal bonding or solvent assisted thermal bonding.

In a particular embodiment the lid 200 and chip 100 can be bonded together via UV-assisted thermal bonding where both parts are exposed to UV radiation. In order to ensure uniform pressure all over the chip surface, an aluminum holder was built in which the Luer fitting protrusions could be inserted during the bonding process. In addition, the lid 200 may be covered with a combination of a 300 μιη thick nickel disc and a thin PDMS layer with the purpose of compensating possible non-uniformities in flatness. In a similar manner to the microfluidic chip 100, the lid 200 is suitably extruded from a hard thermoplastic polymer; i.e. the lid 200 is constituted by thermoplastic polymer. Suitable thermoplastic polymers for the lid 200 are selected from the same thermoplastic polymers given above for the microfluidic chip 100. By the term "hard" is meant that the microfluidic chip has a Young's modulus greater than 1000 MPa. Preferably, the lid 200 and the microfluidic chip 100 are formed from the same thermoplastic polymer. The lid 200 may comprise one or more connectors, in addition to, or as an alternative to the connectors in the microfluidic chip 100, which are arranged such that liquid, optionally a cell suspension, can be independently passed through the microfluidic channels of the microfluidic chip 100. As shown in Figure 3, the microfluidic device 10 is preferably used with the microfluidic chip 100 uppermost, and the lid 200 underneath. The shallow microfluidic channel is therefore located in the lower portion of the extracellular/intracellular microfluidic channels, which increases the likelihood of cells becoming attached thereto.

The microfluidic device 10 according to the invention may additionally comprise at least one extracellular faradaic electrode connected to said extracellular microfluidic channel, and at least one intracellular faradaic electrode connected to said intracellular microfluidic channel (not shown in the Figures). An electrical circuit can therefore be established between said extracellular and said intracellular faradaic electrodes, across said shallow microfluidic channel. Electrophysiological measurements of a trapped cell can be made using these electrodes, as set out below.

Fabrication of microfluidic cell-capture chip

The microfluidic cell-capture chips of the invention comprising a predetermined pattern of microfluidic channels are injection-moulded from thermoplastic polymer and have very low surface roughness. They are fabricated via a novel method which allows very high surface smoothness to be achieved. Suitably, the microfluidic cell-capture chip comprises at least one extracellular microfluidic channel, at least one intracellular microfluidic channel and at least one shallow microfluidic channel extending between said extracellular and said intracellular microfluidic channels.

This method is illustrated generally in Figures 4a - 4f. Fabrication begins by providing a shaped silicon master comprising a predetermined pattern of microfluidic channels (Figure

4a). The shaped silicon master is formed by various photolithographic and etching processes; details of its formation are provided below. The shaped silicon master is essentially a copy of the microfluidic chip to be produced.

In step b. (Figure 4b) a layer of silicon oxide is deposited on top of said shaped silicon master. The channels therefore become coated with silicon oxide. The layer of silicon oxide is preferably deposited on top of said shaped silicon master in this step using a tetraalkyl orthosilicate, preferably a tetraethyl orthosilicate (TEOS), process. Alternatively, a layer of silicon oxide can be deposited on top of the shaped silicon master by thermal oxide deposition, sputtering of Si oxide, and by application of silicon nitride.

A metal seed layer is then deposited onto the layer of silicon oxide produced in step b. (Fig. 4c). The metal seed layer preferably comprises an adhesion layer on the silicon master and a high electrical conductivity layer on top of said adhesion layer. The adhesion layer is preferably Ti, but may also be Cr, Ni, NiCr alloy or NiV alloy. The high electrical conductivity layer is preferably Au, but may also be Ag, Cu or Pt. The high electrical conductivity layer may be omitted completely; in which case a thicker layer of Cr, NiCr of NiV is preferred as the adhesion layer. The metal seed layer therefore most preferably comprises an adhesion layer of Ti on the silicon master and a high electrical conductivity layer of Au on top of said adhesion layer.

A preferred method for depositing the metal seed layer is sputtering, as this provides an even seed layer on the bottom and sidewalls of the microfluidic channels. Other deposition methods for the metal seed layer include CVD, ALD, E-beam evaporation and thermal evaporation .

Once the metal seed layer is formed, a metal shim is electroformed onto the metal seed layer (Figure 4d). Electroforming allows the formation of relatively thick deposits which are self- supporting and can be removed from the template upon which they are formed. Typically, the metal shim has a thickness of 150-1000 μιη.

Electroforming is a highly accurate deposition method, which results in the metal shim comprising a relief pattern corresponding to said predetermined pattern of microfluidic channels in said shaped silicon master. In a preferred embodiment, the metal shim is a nickel shim, which provides good hardness and chemical resistance. Once the nickel shim is formed, the shaped silicon master is removed from the metal shim (Figure 4e). Preferably, the shaped silicon master is removed from the metal shim together with the silicon oxide layer and the metal seed layer. Removal of the shaped silicon master from the metal shim preferably takes place by dissolving the shaped silicon master in Si- etchants based on aqueous alkaline solutions (e.g. an KOH solution with a pH of 10-14). Alternatively, it may be possible to simply delaminate the silicon master from the metal shim. As seen in Figure 4e, the nickel shim is used in a process of injection-moulding a microfiuidic chip from a thermoplastic polymer. This provides a predetermined pattern of microfiuidic channels moulded into said microfiuidic chip. In the injection-moulding step, two moulds are provided - a first mould with Leur connectors and a second mould containing the

microstructured nickel shim. The two moulds are closed together to form a cavity into which thermoplastic polymer is injected. Pulling apart the moulds provides the microfiuidic chip of the invention moulded from thermoplastic polymer.

Suitable thermoplastic polymers for use in the injection-moulding are selected from a cyclic olefin polymer (COC), a polyethylene polymer, a polypropylene polymer, a polystyrene polymer or a polyacrylate polymer, preferably a cyclic olefin polymer.

The step of depositing a layer of silicon oxide on top of said shaped silicon master (e.g. via TEOS deposition) gives the channels their high smoothness. To illustrate the importance of this step, Figure 2a shows a detail of a microfiuidic chip fabricated according to the method of the invention, showing smooth channel surfaces. In contrast, Figure 2b shows a detail of a microfiuidic chip in which the fabrication method did not include the silicon oxide (e.g. TEOS) deposition step; i.e. according to J. Micromech. Microeng 22 (1012) 115008. Figures 2a and 2b are given at the same magnification. It is clear that silicon oxide deposition prior to metal seed layer deposition provides the very smooth channel surfaces in the microfiuidic chip which are required for establishing a gigaseal. Previous attempts were made to capture and bind cells also to the device in J. Micromech. Microeng 22 (1012) 115008 where channels were somewhat larger, ie. 5 μιη by 4 μιη. Cells could only be captured with very gentle suction, while larger suctions would make the cells to go through the patching channel. When channels with a depth of 2 microns (2 um by 2 um approximately) were tested, a small increase in the resistance up to 20-50 MOhms could be observed. Howerer, a gigaseal could not be obtain with any of the devices presented in J. Micromech. Microeng 22 (1012) 115008.

Fabrication of shaped silicon master

The shaped silicon master used in forming the metal shim is fabricated using a combination of photolithography, deposition and etching steps. As shown in Figure 5f and Figure 4a, the shaped silicon master comprises at least one extracellular microfiuidic channel, at least one intracellular microfiuidic channel and at least one shallow microfiuidic channel extending between said extracellular and said intracellular microfiuidic channels. The shaped silicon master is formed as set out in the following. To begin, a silicon wafer is provided. The silicon wafer is disc-shaped, with a thickness of 300-600 μιη and a diameter of 50-150mm, typically circa 100mm. However, the skilled person will be able to use silicon wafers with other overall shapes, while remaining within the scope of the invention. Silicon wafers suitable for use in the present invention are commercially available; e.g. _1 0 0_, single side polished, 525 μιη thick silicon wafers from Topsil semiconductor materials A/S.

Various preparatory steps may be performed on the silicon wafer. For example, native oxide may be removed (a preferred method for which is soaking for 1 minute in an HF bath and rinsed in water). As shown in Figure 5a, a first photolithographic process is performed on said silicon wafer, in which said at least one shallow microfluidic channel(s) is developed in a first photoresist. This is illustrated as the shallow Y-form in Figure 5a. The first photolithographic process is carried out using standard methodology, and can be adapted by the skilled person as desired. In a typical first photolithographic process, photoresist is spin-coated onto the silicon wafer. A preferred photoresist is AZ5214E. The photoresist has a thickness of about 1.5 μιη. The photoresist is dried (e.g. by baking at 90°C for 90 seconds), and then exposed using a Mask Aligner MA6 (exposure wavelength 365 nm) from Karl Suss for 6.7 sec. 800ml of "Developer AZ 351" is mixed with 4000ml water to form a water bath having a temperature of 22°C. The photoresist is developed by dipping the coated silicon wafer in this bath for 63 seconds. The silicon wafer coated with photoresist is etched to form the at least one shallow microfluidic channel(s) in the silicon wafer, and any remaining first photoresist from the first photolithographic process is removed (Figure 5b). A preferred method for the etching is Reactive Ion Etching (RIE).

In a particular embodiment of the invention, the silicon wafer was etched (RIE) by using a BOSCH process in a Pegasus system from SPTS Technologies. The process time was 24.2 seconds corresponding to 11 cycles. Every cycle had a deposition phase (1 second) and an etching phase (1.2 sec). The process started with etching phase and ended with deposition phase. The gas flow was C4F8/SF6/02 = 150/275/5 seem. Coil power was 2000W(deposition phase) and 2500 W(etching phase). Platen power was 0W(deposition phase) and 35

W(etching phase). Platen temperature was 0°C.

Any remaining first photoresist from the first photolithographic process is then removed. In a particular embodiment of the invention photoresist was stripped by using a PVA 300 Plasma asher from Tepla. The process time was 20 min. The gas flow was 02/N2 = 240/ 70 ml/min and the power 1000 W.

Once the at least one shallow microfluidic channel is formed in the silicon wafer, optional steps may be included before proceeding.

An RCA clean was performed on the silicon wafer. This process is used for cleaning wafers before further processing in a furnace, it comprises three steps: RCA1 removes organic films, RCA2 removes alkali ions and metal hydroxides, hydrofluoric acid HF removes the thin oxide film grown during RCA1 and RCA2. RCA 1 contains ammonia (25%) : hydrogen peroxide: DI water (1 : 1 : 5). RCA 2 contains hydrochloride (37%) : hydrogen peroxide: DI water (1 : 1 : 5). A buffered aqueous solution (HF solution, 5%).

Subsequently, a first silicon oxide layer is formed on at least a portion of the silicon wafer, wherein said portion includes said at least one shallow microfluidic channel(s), see Figure 5c. That is, the first silicon oxide layer is deposited so that it at least covers the shallow microfluidic channel(s).

To achieve this aim, silicon wafers were oxidized for 40 min in a Tempress horizontal furnace at 1050 _°C with an oxygen flow of 5 slm (standard liters per minute) and 20 min annealing at the same temperature, with a nitrogen flow of 3 slm. During heating and cooling, a nitrogen flow of 3 slm is used. Thickness of the oxide layer was 50-65nm. A second photolithographic process is performed on the silicon wafer, in which said extracellular and said intracellular microfluidic channels are developed in a second photoresist (Figure 5d), such that said at least one shallow microfluidic channel extends between said extracellular and said intracellular microfluidic channels. The materials and methods used in the second photolithographic process are essentially the same as those for the first photolithographic process, although this may be adapted as required by the skilled person.

The first silicon oxide layer is then etched to form the extracellular and intracellular microfluidic channels in the silicon oxide layer and then further etched to form extracellular and intracellular microfluidic channels in the silicon wafer (Figure 5e).

To remove the oxide, the silicon wafer was etched in an Advanced Oxide Etch (AOE) from SPTS Technologies. Process time was 25 seconds. The gas flow was C₄F₈/H₂ = 5/4 seem. Coil power was 1300 W, platen power was 200 W, platen temperature was 0°C, pressure was 4 mTorr.

The silicon wafer was etched by using a ramped BOSCH process in a Pegasus system from SPTS Technologies. The process time was 9 minutes and 53 seconds corresponding to 215 cycles. Every cycle had a deposition phase (ramped from 2 to 1 seconds) and a etching phase (ramped from 2.4 to 0.1 seconds). The process started with etching phase and ended with deposition phase. The gas flow was C₄F₈/SF₅/0₂ = 150/275/5 seem. Coil power was 2000W (deposition phase) and 2500 W(etching phase). Platen power was 0W (deposition phase) and 35 W (etching phase). Platen temperature was 0°C. As before, any remaining second photoresist from the second photolithographic process is then removed.

To remove defects at the bottom of said extracellular and said intracellular microfluidic channels, the silicon wafer may be wet etched at this point. The silicon wafer was dipped in a polysilicon etch mixture of HN0₃, BHF and H₂0 in the ratio 20: 1 : 20 for 8 minutes at room temperature and then rinsed in water.

Any remaining first silicon oxide layer is stripped from said silicon wafer to provide the shaped silicon master. The shaped silicon master was dipped in a buffered HF solution for 3 minutes at room temperature and then rinsed in water.

To smooth the sidewalls of the extracellular microfluidic channels, the method according to the invention may further comprise the steps of: forming a second silicon oxide layer on at least a portion of said shaped silicon master, and stripping said silicon oxide layer from said shaped silicon master.

As an example of these additional steps; the shaped silicon master was oxidized for 150 min in a Tempress horizontal furnace at 1100 _°C with an oxygen flow of 5 slm (standard liters per minute) and 20 min annealing at the same temperature, with a nitrogen flow of 3 slm. During heating and cooling, a nitrogen flow of 3 slm is used. The thickness of the oxide layer was circa 200nm. The shaped silicon master was dipped in a buffered HF solution for 15 minutes at room temperature and then rinsed in water EXAMPLES

Formation of shaped silicon master

A shaped silicon master was formed by the fabrication process outlined above, which is illustrated generally in Figures 5a - 5f. Injection-moulding of chips

Microfluidic chips were injection moulded with an ENGEL Victory 80/45 Tech injection moulder, with the nickel shim installed into the mould. Nozzle temperature was 280°C, mould temperature was 130°C, with a demoulding temperature below 60°C. A holding pressure of 1700 bar was used. The holding pressure decreased from 1700 bar to 0 bar in 0.75 seconds. Chips were moulded from COC TOPAS grade 5013 (glass transition temperature, Tg of 135°C) from TOPAS Advanced Polymers GmbH.

Formation of microfluidic device

A microfluidic device was formed from the injection-moulded microfluidic chip by placing a lid on the chip and sealing it. A 100 μιη thick extruded polymer film TOPAS 5013F-04 from Advanced Polymers Extrusion Lab, was used as a cover lid. Both the microfluidic chip and the lid were exposed to a UV radiation for 30 s. The exposure was performed using a DIMAX mercuryUVbulb F/5000 lamp emitting over the full unfiltered Hg line spectrum and the bonding by using a P/O/Weber press with decoupled internal temperature control of both plates. The microfluidic chip was placed in a customized aluminum holder in which the Luer fitting protrusions could be accomodated. Seven such microfluidic chips could be placed in the holder at the same time. The lid was covered with a combination of a 300 μιη thick nickel disc and a thin PDMS layer with the purpose of compensating possible non-uniformities in flatness. A piston force of 40 kN applied for 10 minutes at 115°C.

Set-up of the microfluidic device Figure 6 illustrates schematically the set-up used of the microfluidic device for the recordings. The microfluidic device 10 is connected to two intracellular solution reservoirs 11, 12 and then to the pressure controller 13. Ag/CI electrodes 14, 15 are connected to the headstage of the amplifier 16. Syringes 17, 18 are used respectively to translate cells from the inlet to the patch zone and to apply lidocaine from one of the lateral apertures.

Validation of the microfluidic device with living cells (HEK Nav 1.7):

The microfluidic device was mounted into an aluminium box and positioned on the stage of an Olympus 1X70 inverted microscope. Prior to experiments the device was primed with the two electrolyte solutions. First the intracellular channels (140 in fig. l) together with the shallow patching channels (130 in fig. l) were primed with intracellular solution and then the extracellular channel (120 in fig. l) with the extracellular solution. The two Luer ports connected to the shallow patching channel (130 in fig. l) were connected to two reservoirs filled up with intracellular solution and from them to a custom-made pressure controller build from a piezo-valve terminal from Festo and controlled with Labview software (National Instruments). The reservoirs avoided air bubbles when applying suction to the patching channel. External Ag/AgCI electrodes positioned across the recording channel ensured electrical connection. Electrodes were electrically connected to the amplifier head stage mounted on the customized aluminum box. A typical chip resistance of 8-10 ΜΩ for the patching channel was shown by applying a 10 mV test square pulse for 10 ms. The cells were introduced into the inlet port and translated in the extracellular channel (120 in fig.l) towards the shallow patching channel (130 in fig. 1) with the flow from a syringe connected to the port at the other end. Before trapping the cell, positive pressure of 3-5 mbar was applied to the shallow patch channel (130 in fig. 1) to prevent contamination of the aperture. A cell was captured to the patch hole after applying suction (negative pressure of 400 mbar) to the shallow patching channel (130 in fig.l). The amplifier offset potential was zeroed prior to patching the cell and the holding potential held at -90mV. All recordings were carried out using a HEKA Patch Clamp EPC 9 amplifier (HEKA Electronics) at room temperature. Pulse software (v 8.53, HEKA Electronics) was used for data acquisition. The cells used for the experiments were Human Embryonic Kidney 293 cells (HEK293) expressing the subtype of the voltage gated sodium channel Na_v 1.7 and they were purchased from Scottish Biomedical Ltd. The cells were grown and maintained under standard culture conditions at 37°C and 5% C0₂. The cell concentration in the suspension was 2-3 mill/ml. The intracellular electrolyte solution contained (in mM) : 135 CsF, 1/5 EGTA/CsOH, 10 mM HEPES and 10 NaCI. The pH was adjusted to 7.3 with KCI and osmolarity to 320mOsm with sucrose. The extracellular electrolyte solution contained (in mM) : 1 CaCI₂, 1 MgCI₂, 5 HEPES, 3 KCI, 140 NaCI, 0.1 CdCI₂ and 20 TEA-CI. The pH was adjusted to 7.3 with NaOH and osmolarity to 320mOsm with sucrose. All chemicals were purchased from Sigma Aldrich. Both solutions were stored in the fridge at 4°C and vacuum degassed for 20 minutes before use. Lidocaine hydrochloride monohydrate (Sigma Aldrich L5647) was dissolved in DMSO to give a 100 mM stock solution kept in the freezer. Subsequent dilutions were performed in extracellular electrolyte solution.

Once a cell was trapped, the suction of -400mbar was kept on for about 20 seconds and then reduced to -30mbar. The resistance across the aperture was continuously monitored by applying a 10 mV pulse for 10 ms. At this stage the resistance was typically between 100 and 200 ΜΩ. Some of the cells showed whole-cell immediately after patching. Otherwise the whole-cell was achieved by applying suction pulses (-30mbar to -400mbar) and electrical pulses combined.

A subtype of the voltage gated sodium channel, Navl.7 was tested on Human Embryonic Kidney (HEK) cells. Experiments were designed to explore the current-voltage (IV)- relationship for activation and inactivation and the sensitivity of Navl.7 channels to

Lidocaine. Reduction in the transition to slow inactivation in Navl.7 voltage-gated sodium channels was successfully shown.

Both IVs and Lidocaine dose response curves obtained from the injection-moulded polymer device are comparable with data obtained using the commercially-available QPatch™ system from Sophion Bioscence.

Figure 7 shows the current-voltage relationships for activation and inactivation. Experiments were carried out with the polymer device and with the QPatch™. The two systems shared the cells, the electrolyte solutions and the compounds. Data were leak subtracted in order to compensate for the capacitance (Fig. 7a-b). Activation currents were obtained after the start of depolarization pulses from -90mV to 70mV while the holding potential was -90mV. Steady state inactivation currents were investigated at + 10mV after conditioning potentials ranging from -90mV to 70mV for 1000 ms and holding potential held at -90mV. The resulting current amplitude represents the portion of sodium channels in the inactivated state. The currents were plotted as a function of the voltage. The activation threshold was -50 mV, and the maximal current amplitude of 1.6 nA was obtained at -30 mV. At positive potentials the current amplitude was gradually reduced to 0.2 nA as the electrochemical driving force vanished. The activation threshold was found at -30 mV with a maximal current amplitude of 1.6 nA registered at -lOmV with the QPatch™ system. The I-V relationships showed a good agreement, beside a 20 mV shift to depolarized potentials of the data obtained with our device, see Fig. 7c. The inactivation graph in Fig. 7d shows the gradual transition from a state where the sodium channels were predominantly closed to a state where they were predominantly inactivated, above -50mV. The cut off at -90mV doesn't show at which potential the transition starts, presumably at about -lOOmV. The transition from closed state to inactivated state happened between -80 and -30mV with the QPatch™ system. As mentioned for the activation currents, the IV relationship obtained from the polymer device results shifted by 20 mV to depolarized potentials.

Figure 8 and figure 9 illustrate the study of sensitivity of sodium channels to lidocaine.

Lidocaine inhibition of whole-cell sodium currents was explored in voltage-clamped mode by application of lidocaine concentrations ranging from 50 μΜ to 1 mM. Lidocaine ability to bind to sodium channels is state dependent, lidocaine binds to the Na-ι- channel in the inactivated state only. To explore state-dependency the cell was depolarized twice at +0mV for 100 ms and 20 ms from a holding potential of - lOOmV with a temporal separation of 15ms.

Depolarizations were repeated every 5s. Lidocaine response was studied at the start of the second pulse after the resting interval, during which only a portion of the sodium channels were able to recover from inactivation. An example of the current-time relationship for the peak sodium currents recorded in response to the first (empty dots) and the second (full dots) depolarization is shown in Fig. 8. Lidocaine inhibited the channel in the inactivated state (peak 2) which effect is reversible when the compound is removed from the cell proximity. Inhibition is plotted against lidocaine concentration in Fig. 9. Complete inhibition was observed at ImM lidocaine. At 30μΜ lidocaine, 20% inhibition of sodium currents was observed. The amplitudes of sodium currents immediately prior to lidocaine application were set to 100%. Lidocaine was applied from one of the lateral apertures located 1.2 mm from the cell being patched. It was also showed that the inhibition produced by lidocaine could be reversed 40-60 seconds after the lidocaine perfusion was stopped, see Fig. 9. The half- blocking concentration IC₅₀ was 119 ±11μΜ. This value show an excellent agreement with IC₅₀ of 152 ±11μΜ found for the QPatch™. The Hill coefficients were similar for the two concentration-response relationships. Although the invention has been described with reference to a number of embodiments and figures, the scope of the invention should not be limited to these embodiments, and the skilled person can combine embodiments with their common general knowledge to arrive at alternative embodiments of the invention. In particular, the skilled person will know which materials will be most suitable for particular components, and which deposition methods will be most suitable for particular deposition methods. The true scope of the invention is defined in the herein-appended claims.

Claims

1. A method for the fabrication of a microfluidic cell-capture chip, said microfluidic chip comprising a predetermined pattern of microfluidic channels, said method comprising the steps of: a. providing a shaped silicon master comprising said predetermined pattern of

microfluidic channels; b. depositing a layer of silicon oxide on top of said shaped silicon master; c. depositing a metal seed layer onto the layer of silicon oxide produced in step b.; d. electroforming a metal shim onto said metal seed layer, such that said metal shim comprises a relief pattern corresponding to said predetermined pattern of microfluidic channels in said shaped silicon master; e. removing the shaped silicon master from the metal shim, and; f. injection-moulding a microfluidic chip from a thermoplastic polymer using said metal shim, such that the predetermined pattern of microfluidic channels are moulded into said microfluidic chip.

2. A method according to claim 1, wherein said microfluidic cell-capture chip comprises at least one extracellular microfluidic channel, at least one intracellular microfluidic channel and at least one shallow microfluidic channel extending between said extracellular and said intracellular microfluidic channels.

3. The method according to any one of the preceding claims, wherein the layer of silicon oxide is deposited on top of said shaped silicon master in step b. using a tetraalkyl orthosilicate, preferably a tetraethyl orthosilicate, process.

4. The method according to any one of the preceding claims, wherein the microfluidic chip has a Young's modulus greater than 1000 MPa.

5. The method according to any one of the preceding claims, wherein the metal seed layer comprises an adhesion layer on the silicon master and a high electrical conductivity layer on top of said adhesion layer.

6. The method according to any one of the preceding claims, wherein the shaped silicon master is removed from the metal shim by dissolving the shaped silicon master in an aqueous alkaline solution.

7. The method according to any one of the preceding claims, wherein the microfluidic chip is injection-moulded from a thermoplastic polymer selected from a cyclic olefin polymer (COC), a polyethylene polymer, a polypropylene polymer, a polystyrene polymer or a polyacrylate polymer, preferably a cyclic olefin polymer.

8. The method according to any one of claims 2-7, wherein said microfluidic cell-capture chip comprises at least one extracellular microfluidic channel, at least one intracellular microfluidic channel and at least one shallow microfluidic channel extending between said extracellular and said intracellular microfluidic channels; whereby said shaped silicon master is formed by providing a silicon wafer; performing a first photolithographic process on said silicon wafer, in which said at least one shallow microfluidic channel(s) is developed in a first photoresist; etching said silicon wafer to form said at least one shallow microfluidic channel(s) the silicon wafer;

IV removing any remaining first photoresist from the first photolithographic process; v forming a first silicon oxide layer on at least a portion of said silicon wafer, wherein said portion includes said at least one shallow microfluidic channel(s); performing a second photolithographic process on said silicon wafer, in which said extracellular and said intracellular microfluidic channels are developed in a second photoresist, such that said at least one shallow microfluidic channel extends between said extracellular and said intracellular microfluidic channels; etching said first silicon oxide layer to form said extracellular and said intracellula microfluidic channels in the silicon oxide layer; viii. etching said silicon wafer to form said extracellular and said intracellular microfluidic channels in the silicon wafer; removing any remaining second photoresist from the second photolithog

process; x. stripping any remaining first silicon oxide layer from said silicon wafer to provide said shaped silicon master.

9. The method according to claim 8, further comprising the step of: wet etching said silicon master to remove defects at the bottom of said extracellular and said intracellular microfluidic channels, after step ix.

10. The method according to any one of claims 8-9, further comprising the steps of: xi. forming a second silicon oxide layer on at least a portion of said shaped silicon

master, and xii. stripping said silicon oxide layer from said shaped silicon master to smooth the sidewalls of said extracellular microfluidic channels wherein said steps xi. and xii. are carried out after step x.

11. A microfluidic cell-capture chip formed of a thermoplastic polymer, said microfluidic chip comprising at least one extracellular microfluidic channel, at least one intracellular microfluidic channel and at least one shallow microfluidic channel extending between said extracellular and said intracellular microfluidic channels, in which all microfluidic channels have a surface roughness below 5nm, suitably below 3nm and wherein at least one shallow microfluidic channel comprises a straight extracellular region which extends a distance of 5 to 50 μιη in a straight line from the point at which said shallow channel meets said extracellular channel.

12. The microfluidic cell-capture chip according to claim 11, wherein the microfluidic chip is formed of a thermoplastic polymer selected from a cyclic olefin polymer (COC), a polyethylene polymer, a polypropylene polymer, a polystyrene polymer or a polyacrylate polymer, preferably a cyclic olefin polymer.

13. A microfluidic device comprising (a) the microfluidic cell-capture chip according to any one of claims 11-12, and (b) a lid formed of a thermoplastic polymer bonded to said microfluidic chip.

14. The microfluidic device according to claim 13, said microfluidic device additionally comprising at least one extracellular faradaic electrode connected to said extracellular microfluidic channel, and at least one intracellular faradaic electrode connected to said intracellular microfluidic channel, such that an electrical circuit can be established between said extracellular and said intracellular faradaic electrodes, across said shallow microfluidic channel.

15. The method according to any one of claims 2-10, or the microfluidic device according to any one of claims 11-13, wherein said shallow microfluidic channel has a depth of 1.5-3 μιη, such as circa 2 μιη and a width of 1.5-3 μιη, such as circa 2 μιη.

16. The method according to any one of claims 2-10, or the microfluidic device according to any one of claims 11-13, wherein the at least one shallow microfluidic channel comprises a straight extracellular region which extends a distance of 5 to 50 μιη in a straight line from the point at which said shallow channel meets said extracellular channel