Cell selection

Abstract

The disclosure concerns the use of di-electric spectroscopy for the selection of cells, typically cells engineered by recombinant methods, but also including antibody producing hybridoma cell lines, that express a polypeptide, such as an antibody, CAR-T cells, induced pluripotent stem (IPS) cells and cells derived from IPS cells.

Description

[0001] CELL SELECTION

[0002] Field of the Invention

[0003] The disclosure concerns the use of di-electric spectroscopy for the selection of cells, typically cells engineered by recombinant methods, but also including antibody producing hybridoma cell lines, that express a polypeptide, such as an antibody; CAR-T cells, induced pluripotent stem (IPS) cells and cells derived from IPS cells; and host cells expressing viral polypeptides.

[0004] Background to the Invention

[0005] The expression of recombinant proteins including monoclonal antibodies provide therapeutic agents such as therapeutic antibodies, pharmaceutical proteins, such as protein hormones, and antigens for vaccines. Often bacterial systems are the preferred host for protein expression due to the low maintenance and ease of handling. However, the inability of bacterial expression systems to provide accurate posttranslational modifications for therapeutic agents, has led to the development of alternative expression systems such as yeast, insect cells, plant and mammalian expression systems. Approximately 70% of recombinant protein pharmaceuticals and most proteins used for vaccination, human therapy or diagnostics are currently produced in mammalian cells, such as Chinese Hamster Ovary cells (CHO). However, the cost of maintaining mammalian cell culture for the expression of therapeutic proteins, which includes the production of monoclonal antibodies by hybridomas, is high when compared to bacterial expression systems.

[0006] The demand to manufacture therapeutic proteins in biological systems has increased substantially over the last decade and systems are required which facilitate this process efficiently. Therapeutic agents such as therapeutic antibodies, pharmaceutical proteins, such as protein hormones, and antigens for vaccines are obtained by recombinant protein expression in various different cell types such as bacterial or mammalian cells. Bacterial systems are often preferred due to the low maintenance, however, alternative expression systems such as yeast, insect cells, plant and mammalian expression systems are essential when proteins require posttranslational modifications.

[0007] Expression of therapeutic proteins is challenging due to the protein characteristics such as the requirement for posttranslational modification and the complex requirement for culturing cells. The cell culturing process includes cell line selection and media optimisation on a small scale and once the successful production of the descried protein is achieved, the system requires scale up to allow production of the therapeutic protein to industrial levels.

[0008] For protein expression on an industrial scale several optimisation steps are required starting with the transfection of the cells with a vector encoding the protein to be expressed, culturing the so modified cell and selecting the cells using selectable markers. Once a stable cell line expressing the protein of interest is identified the scale up process can begin. The identification of suitable candidates expressing the protein in a desired amount can include the screening of thousands of cell lines using tests such as ELISAs which is very labour and cost intensive. Moreover, even if the cells express the desired protein in a small scale, the same cell may not perform sufficiently when scaled up and there is no easy way of testing which cells perform efficiently under different conditions.

[0009] Analytical methods for the measurement of cells are known in the art. For example, Fluorescence Activated Cell Sorting (FACS) is an example of such a method. The technique uses flow cytometry combined with fluorescence dyes typically delivered to cells using antibody conjugates comprising a fluorescent dye that binds to a cell protein, for example a cell membrane located receptor, characteristic of the cell. The stained cells are then passed through a flow cell, where they are illuminated by a laser beam. As the cells pass through the laser, their fluorescence is detected and analysed in real-time. Based on the fluorescence properties, the cells can be sorted and separated for further analysis.

[0010] An alternative, label free method for analysing cells utilises impedance or dielectric spectroscopy to measure characteristics of a cell population. Impedance or dielectric spectroscopy involves applying an electrodynamic field to a solution containing cells and measuring the changes to the field due to the presence of the cell e.g., caused by the complex electrical permittivity of a cell. When a cell is measured by a measurement apparatus, the fluid medium surrounding the cell is also measured. The effect of measurement of the surrounding fluid medium is addressed by carrying out two measurements at the same time, namely a first measurement of a cell and fluid medium holding the cell, and a second measurement of fluid medium without a cell. The difference between the first and second measurements is then determined to provide a measurement of the cell. The measurement apparatus therefore comprises a first measurement component which makes the first measurement and a second measurement component which makes the second measurement, the first and second measurement components operative at the same time. Where measurement involves applying a stimulus to a sample and measuring a response to the stimulus, such as in impedance or dielectric spectroscopy, the stimulus is usually of considerably greater amplitude than the amplitude of the response to the stimulus. If the first and second measurements are misaligned in time, then parts of the stimuli are present in the difference between the first and second measurements with the parts of the stimuli considerably greater in amplitude than amplitude of responses to the stimuli. Furthermore, the parts of the stimuli occupy the available voltage range of processing electronics to much greater extent than the responses to the stimuli. There is thus reduction in utilisation of the available voltage range of the processing electronics by the responses to the stimuli.

[0011] An example of the use of impedance or dielectric spectroscopy is disclosed in WO2015 / 001355, the content of which is incorporated by reference in its entirety, which describes a spectroscopy system that allows label-free cell analytics for the detection and analysis of cells. The system enables the user to identify cells based on their intrinsic dielectric profile, providing them with a unique identifier for subsequent downstream analysis and characterisation, without cell labelling or lasers.

[0012] This disclosure relates to a method for the analysis of cells expressing polypeptides, typically therapeutic polypeptides such as antibodies, that facilitates identification of high titre expressing cell lines, for example recombinant cell lines or hybridomas using impedance or dielectric spectroscopy. A problem associated with the production of recombinant polypeptides, and monoclonal antibodies from isolated hybridoma cell lines, is that there is variation in the yield obtained from the cell line or hybridoma. The present disclosure provides a rapid and simple method to assess polypeptide yield from recombinant cell lines or hybridomas to identify high yielding cell lines when compared to low or medium cell line producers in a label free environment. The disclosure also relates to the analysis of other cells such as CAR-T cells, pluripotent embryonic stem cells, induced pluripotent stem cells and cells expressing viral proteins such as adeno-associated virus (AAV) and lentiviral proteins.

[0013] Statements of Invention

[0014] According to an aspect of the invention there is provided a method to identify an eukaryotic cell clone transfected or transduced with a nucleic acid molecule adapted to express at least one recombinant polypeptide or peptide comprising the steps of: i) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to a non-transfected eukaryotic cell; ii) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit and obtaining a first frequency value; iii) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to said eukaryotic cell clone of i) above wherein said eukaryotic cell clone is transfected with a nucleic acid molecule adapted to express at least one recombinant polypeptide or peptide; iv) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit and obtaining a second frequency value, and v) comparing the first and second frequency value to obtain a frequency difference wherein said frequency difference is indicative of said eukaryotic cell expressing said polypeptide or peptide.

[0015] In a preferred method of the invention said nucleic acid molecule encodes an antibody or antibody fragment.

[0016] In a preferred method of the invention said antibody or antibody fragment is a therapeutic antibody or antibody fragment.

[0017] In an alternative preferred method of the invention said nucleic acid molecule encodes a pharmaceutically active polypeptide or peptide.

[0018] In a preferred method of the invention said nucleic acid molecule encodes one or more viral polypeptides.

[0019] In a preferred method of the invention said one or more viral polypeptide is selected from the group: an adenoviral polypeptide, an adeno-associated viral (AAV) polypeptide and a lentiviral polypeptide.

[0020] Preferably said one or more adeno-associated viral polypeptides are from an AAV selected from the group: AAV2, AAV3, AAV6, AAV13; AAV1 , AAV4, AAV5, AAV6, AAV9 and AAVrhIO. In a preferred method of the invention said nucleic acid molecule encodes a chimeric T cell receptor.

[0021] In a preferred method of the invention said nucleic acid molecule encodes one or more pluripotency associated genes.

[0022] Preferably, said nucleic acid molecule encoding one or more pluripotency associated genes is transiently expressed.

[0023] In a preferred method of the invention said eukaryotic cell is a mammalian cell.

[0024] Preferably, said mammalian cell is a Chinese Hamster Ovary cell. Alternatively, said mammalian cell is a HEK cell.

[0025] In a preferred method of the invention said mammalian cell is a somatic cell.

[0026] In a preferred method of the invention said eukaryotic cell is a fungal cell.

[0027] In a preferred method of the invention said eukaryotic cell is a plant cell.

[0028] In a preferred method of the invention said eukaryotic cell is an insect cell.

[0029] According to a further aspect of the invention there is provided the use of impedance or dielectric spectroscopy for the identification and analysis of a mammalian cell transfected with a nucleic acid molecule encoding a recombinant polypeptide or peptide.

[0030] In a preferred embodiment of the invention said recombinant polypeptide or peptide is an antibody or antibody fragment expressed by the mammalian cell at a high antibody titre.

[0031] High antibody titre refers to antibody yields in the range of between 4-10 g / l, more preferable between 5-8g / l. Moreover, the identified cells show stable expression of the antibodies.

[0032] In a preferred embodiment of the invention said recombinant polypeptide or peptide is a pharmaceutically active polypeptide or peptide.

[0033] In a preferred embodiment of the invention said recombinant polypeptide is a viral polypeptide. In a preferred embodiment of the invention said recombinant polypeptide is a chimeric T cell receptor.

[0034] In a preferred embodiment of the invention said mammalian cell is a stem cell.

[0035] In a preferred embodiment of the invention said stem cell is a pluripotent stem cell, for example an embryonic stem (ES) cell or embryonal germ (EG) cell or an induced pluripotent stem cell.

[0036] In a preferred method of the invention said stem cell is a lineage restricted stem cell, e.g., a multipotent stem cell.

[0037] Preferably said lineage restricted stem cell is selected from the group: haemopoietic stem cells e.g. macrophage; neural stem cells; bone stem cells; muscle stem cells; mesenchymal stem cells; trophoblastic stem cells; epithelial stem cells (derived from organs such as the skin, gastrointestinal mucosa, kidney, bladder, mammary glands, uterus, prostate and endocrine glands such as the pituitary); endodermal stem cells (derived from organs such as the liver, pancreas, lung and blood vessels); muscle cell e.g. cardiomyocyte.

[0038] In a preferred method of the invention said multipotent stem cell is a haemopoietic stem cell.

[0039] In a preferred method of the invention said haemopoietic stem cell is a monocyte capable of differentiation into a macrophage. Preferably, said haemopoietic stem cell is obtained from a human subject.

[0040] In an alternative embodiment of the invention said polypeptide or peptide is a pharmaceutically active polypeptide or peptide and not an antibody.

[0041] In a further alternative embodiment of the invention said recombinant cell line is adapted to produce an antigenic polypeptide or peptide for use in a vaccine.

[0042] The cell expressing the polypeptide or peptide may be carried in a fluent material. The predetermined, electrodynamic field may be applied to a fluent material, e.g., a fluent material that carries cells. The electric stimulus signal and hence the pre-determined, electrodynamic field may comprise a signal packet, with at least one signal packet being applied to each cell to be measured. The at least one stimulation electrode and the at least one sensing electrode may be arranged for impedance or dielectric spectroscopy. At least one stimulation electrode and at least one sensing electrode may be arranged to form a measurement capacitor, with the fluent material and any cell therein being located in the electric field between the electrodes.

[0043] The measurement capacitor may have an operative surface configured to electrically couple with the fluent material and any cell, e.g., such that the fluent material and any cell forms a dielectric for the measurement capacitor. The operative surfaces of the measurement capacitor may be exposed to the fluent material, with no intermediate layer. The operative surfaces of the measurement capacitor may conduct into the fluent material.

[0044] The electrodes may be generally in cuboidal, and the operative surfaces of the electrodes may be square or rectangular. The operative surfaces of the electrodes may be flat. The at least one stimulation electrode and the at least one sensing electrode may be provided on an exposed surface of a semiconductor and may be formed in an integrated circuit formed by a semiconductor fabrication process, such as CMOS.

[0045] The electrodes of the measurement capacitor may be disposed on or towards a side of the flow passageway, such as in a direction perpendicular to a direction of flow of the fluent material. The electrodes of the measurement capacitor may be disposed side by side. Such a same side disposition may be appropriate where the cell stimulation apparatus is comprised in a planar semiconductor integrated circuit, such as a CMOS integrated circuit.

[0046] Where the at least one stimulation electrode and the at least one sensing electrode are arranged with their operative surfaces in a generally planar arrangement, side-by-side, the fluent material and any cell therein may flow over the operative surfaces of the electrodes.

[0047] The electrodes of the measurement capacitor may take the form of an electrode array, which is formed on a surface of a semiconductor device. The apparatus may comprise a plurality of electrode arrays, i.e., a plurality of measurement capacitors. The electrodes may be formed by conductive plates, e.g., metal plates, which are exposed to the flow of fluent material. The fabrication process for the electrode array may lack a polyimide layer deposition step, such that no polyimide top layer is present. The electrode array may have a passivation layer, e.g., a silicon nitride layer, but may be without any passivation layer over the electrodes, such that the electrodes conduct into the fluent material. The hydrophilic nature of the passivation layer provides for maximum exposure, with the openings in the passivation layer between the electrodes and the fluent material enabling conduction between the electrodes and the fluent material. The size of the electrodes is selected depending on the size of the cells being measured. The sensing circuit may comprise a high impedance input to thereby provide for a significant sensed signal. More specifically the sensing circuit may comprise an impedance buffer, such as a field-effect transistor (FET). A FET may provide a capacitive load, e.g., for a sensing electrode of the sensing apparatus, which may be selected to be as small as possible. Alternatively, or in addition, the sensing circuit may provide one of a voltage signal and a current signal as the output signal.

[0048] Alternatively, or in addition, the sensing circuit may be configured to amplify the input signal. The measurement apparatus may be configured to compare the electric stimulus signal with the electric response signal, e.g., as sensed by the sensing apparatus. Comparison of the electric stimulus signal with the electric response signal may comprise cross-correlation of the electric stimulus signal with the electric response signal. The measurement apparatus may be operative to determine a time delay between application of the electric stimulus signal and the response as provided by at least one cell in the fluent material. The measurement apparatus may be further operative to determine a transfer function or at least to approximate a transfer function for the at least one cell in dependence on the time delay. The determined or approximated transfer function may then provide for characterisation of the at least one cell.

[0049] The pre-determined, electrodynamic field applied to the cell according to the invention may comprise at least one corresponding frequency component. The sensing apparatus may be configured to sense a frequency component comprised in the response electrodynamic field, sensed by the cell sensing apparatus.

[0050] The electric stimulus signal may comprise a pseudo-random binary sequence having a length and data rate that provide the pre-determined, electrodynamic field with a range of frequencies. The measurement apparatus may be configured to generate the pseudo-random noise signal. The pseudo-random noise signal may be generated from a pseudo-random binary sequence.

[0051] The pseudo-random binary sequences are generated with a deterministic algorithm. The pseudo-random binary sequence nevertheless exhibits statistical behaviour similar to a truly random sequence. The pseudo-random binary sequence may be a maximum length sequence, a so-called “m-sequence”, which may be generated by a linear feedback shift register. Where a binary sequence is generated by a linear feedback shift register, the output eventually repeats itself. An m-sequence is the longest possible non-repeating sequence for a given number of shift registers.

[0052] The pseudo-random noise signal may therefore comprise a m-sequence. M-sequences exhibit a flat power spectral density across a desired bandwidth of operation. Furthermore, m- sequences may be readily provided for by way of standard digital circuitry and thus may be suited to implementation in an integrated circuit formed by a semiconductor fabrication process, such as CMOS.

[0053] The stimulation circuit may be operative to generate a stimulation signal in the form of an m- sequence by way of a linear feedback shift register or otherwise as would be within the ordinary design skills of the person skilled in the art. For example, the stimulation circuit may comprise memory storing the m-sequences. Alternatively, an m-sequence may be provided by an external signal generator.

[0054] The length of each m-sequence is 2n- 1 , where n is the number of registers in the linear- feedback shift register. The range of frequencies provided by the pseudo-random binary sequence may be determined by the length of each of the pseudo-random binary sequence, determined in numbers of bits, and the bit rate of the electric stimulus signal.

[0055] The range of frequencies of the pseudo-random binary sequence is determined by the binary signal achievable with the length of the pseudo-random binary sequence and the bit rate of the electric stimulus signal. The highest frequency of the range of frequencies may be an alternating sequence of bits, immediately adjacent to each other, which will have a frequency of half of the bit rate. The lowest frequency may be a constant value across the entire length of the sequence, which will have a frequency of the bit rate divided by the length of the sequence.

[0056] The measurement apparatus may comprise processing apparatus, which may include the stimulation circuit and / or the sensing circuit. The processing apparatus and / or the sensing circuit may be configured to receive the electric response signal and to convert the electric response signal to digital form. The processing apparatus and / or the sensing circuit may therefore have an analogue input and a digital output, such that the output is a digital response signal.

[0057] The processing apparatus and / or the sensing circuit may therefore comprise an analogue-to- digital converter and whatever signal conditioning circuitry may be required, such as an amplifier and an anti-aliasing filter. The processing apparatus may be constituted by any suitable electronic apparatus, such as a separate analogue-to-digital converter circuit, a separate amplifier circuit and a separate filter circuit or a configurable integrated circuit, such as an FPGA, or a dedicated integrated circuit, such as an Application Specific Integrated Circuit (ASIC), comprising such circuits.

[0058] The same clock may be used for the generation of the electric stimulus signal and the sample rate of the processing apparatus. This ensures that the electric stimulus signal does not drift in time relative to the digital response signal. The sample rate of the processing apparatus may, however, be different to the bit rate of the electric stimulus signal, as discussed in more detail below. The sample rate of the processing apparatus and the bit rate of the electric stimulus signal may also be expressed in terms of the sample period of the processing apparatus and the bit period of the electric stimulus signal. Where the clock used for the generation of the electric stimulus signal and the clock used for sampling are different, the respective clocks are preferably derived from the same clock.

[0059] The digital response signal may be stored in memory. The processing apparatus and / or the sensing circuit may comprise a processor and may be operative to decode the received output signal, which may be m-sequence encoded data.

[0060] The impulse response may be calculated by application of a Hadamard transform, such as the Fast Hadamard transform. This provides for rapid calculation of the impulse response. The fast Hadamard transform is given by: is the estimated output spectrum of the system under test, m is the sequence order, H is the Hadamard matrix, q is the measured m-sequence encoded response, and and are the encode and decode matrices for transforming m-sequence data into the correct order for use with the Hadamard matrix. In one form, and are equal to each other.

[0061] Alternatively, analysis apparatus may be configured to decode the received output signal, which may be m-sequence encoded data, for example by cross-correlation. The analysis apparatus may be constituted by any suitable electronic apparatus, e.g., a general-purpose computer, such as a Personal Computer (PC).

[0062] The analysis apparatus may be further operative to perform a Fourier Transform, such as a Fast Fourier Transform (FFT), on the decoded output signal to thereby provide frequency domain data. The frequency domain data may be displayed for user interpretation.

[0063] The measurement apparatus may be operative and perhaps also configured to be label free. The biological sensing apparatus may therefore operate on fluent material lacking any label, such as a fluorochrome or microbeads.

[0064] The measurement apparatus may be operative and perhaps also configured for sensing of microbiological samples. The measurement apparatus may be configured for a cell size of cell or a range of sizes of cells in respect of a dimension of at least one of the stimulation apparatus and the sensing apparatus, such as a size of at least one electrode. More specifically, a dimension of at least one of the stimulation apparatus and the sensing apparatus may correspond to a size of cell or a range of sizes of cells.

[0065] The semiconductor fabrication process may be a planar semiconductor fabrication process. Alternatively, or in addition, the semiconductor fabrication process may be a metal-oxide semiconductor process, such as a CMOS process. Alternatively, or in addition, the semiconductor fabrication process may be a submicron semiconductor fabrication process, such as a 0.18 micron CMOS process and perhaps a high voltage 0.18 micron CMOS process.

[0066] The fluent material may be substantially liquid, e.g., at room temperature. The fluent material may therefore comprise a liquid which carries the cells. More specifically, the fluent material may comprise charge carriers, such as salt molecules. For example, the fluent material may comprise phosphate buffered saline (PBS).

[0067] The measurement apparatus may comprise, or be adapted to be operative with, a flow apparatus that provides for flow of the fluent material and the cells in a flow passageway. In use, a sample of fluent material, such as a cell sample, is introduced into a flow apparatus with the flow apparatus being configured to contain and provide for flow of the fluent material through the flow passageway. The flow apparatus may, for example, define an open-ended fluid passageway which contains the fluent material and allows for flow such as when flow is created by way of a pump. Alternatively, or in addition, the flow apparatus may be configured to actuate flow of itself. More specifically the flow apparatus may be configured to draw the fluent material through the flow apparatus by way of capillary action.

[0068] The flow apparatus may define a main channel through which the fluent material flows when in use. The sensing apparatus may be disposed relative to the main channel so as to provide for sensing of cells present in the main channel. The sensing apparatus may be disposed on at least one of first and second opposite sides of a flow of fluent material. Thus, for example, components such as sensing electrodes of the sensing apparatus may be disposed on one side of the flow of fluent material. According to another example, components of the cell sensing apparatus may be disposed on both sides of the flow of fluent material.

[0069] The flow apparatus may comprise a sample inlet which is configured to receive a sample of fluent material, e.g., by way of injection, which is to be subject to measurement, the sample inlet being in fluid communication with the main channel. The flow apparatus may comprise a sample outlet at an opposite end of the flow apparatus from the sample inlet, the sample outlet being in fluid communication with the main channel. The sample outlet may provide for flow of fluent material from the main channel. The flow apparatus may comprise at least one further inlet disposed laterally of the sample inlet. More specifically, the flow apparatus may comprise first and second further inlets with the first inlet disposed laterally on one side of the sample inlet and the second inlet disposed laterally on another opposite side of the sample inlet. The at least one further inlet may be in fluid communication with the main channel. In use, a sheath fluid, such as phosphate buffered saline (PBS), may be received by the at least one further inlet to thereby provide for a flow of sheath fluid in the main channel, the flow of sheath fluid being lateral of a flow of fluent material. The flow of sheath fluid may provide for registration of the cell comprising fluent material with the cell sensing apparatus and may also help preserve the integrity of the flow of fluent material as it progresses though the flow apparatus.

[0070] The flow apparatus may be formed from glass and / or at least in part from a polymer, such as poly(methyl methacrylate) (PMMA). The stimulation and sensing apparatus may be formed separately from the flow apparatus. The stimulation and sensing apparatus may be disposed relative to the flow apparatus by attaching the stimulation and sensing apparatus and the flow apparatus to each other. The stimulation and sensing apparatus and the flow apparatus may be adhered together, for example by appropriate chemical or physical bonding, or may be mechanically attached to each other, e.g., by way of a fastener apparatus comprising a silicone gasket layer, which may be releasable. Appropriate chemical or physical bonding may include plasma bonding.

[0071] The cell measurement apparatus may be configured to be operable as a flow cytometer. The measurement apparatus may further comprise control apparatus. The control apparatus may be constituted by any suitable electronic apparatus, such as a microprocessor or a configurable electronic circuit, such as a Field Programmable Gate Array (FPGA).

[0072] The measurement apparatus may comprise flow inducing apparatus, i.e., a pump, which is operative to induce a flow of fluent material through the flow apparatus. The flow inducing apparatus may be controlled, for example in respect of a rate of flow of fluent material through the flow apparatus, in dependence on an output from the sensing apparatus. The control apparatus may be operative to receive an output from the sensing apparatus and to provide an output to the flow inducing apparatus in dependence thereon.

[0073] The sensing apparatus as described elsewhere herein may be operative to provide for determination of a rate of flow of fluent material through the flow apparatus, the rate of flow being received by the control apparatus. Where the sensing apparatus comprises plural spaced apart sensing electrodes with each sensing electrode being operative to sense cells, the rate of flow may be determined in dependence on the separation between the sensing electrodes being known and a time between sensing of a cell by different sensing electrodes. Alternatively, or in addition, characterisation of at least one cell as described elsewhere herein may be compared with a predetermined criterion and the flow inducing apparatus may be controlled in dependence on the comparison. For example, characterisation of the at least one cell may comprise a level of confidence value which is compared with a predefined value. More specifically, if the level of confidence value is below the predefined value the flow inducing apparatus may be operative to reduce a rate of flow of the fluent material to thereby provide for improved characterisation.

[0074] The processing apparatus may provide an output signal, e.g., a digital output signal. The output signal may be stored in memory of the measurement apparatus. The measurement apparatus may comprise an output for sending the output signal to analysis apparatus, which may or may not be integrated with the measurement apparatus.

[0075] The analysis apparatus may be configured to make determinations with regards to cells comprised in the fluent material in dependence on at least one output from the cell sensing apparatus. For example, the analysis apparatus may be operative to make determinations in dependence on electric field measurements made by the cell sensing apparatus after analogue to digital conversion. Determinations may be made in respect of the like of the density of cells comprised in the fluent material, differentiation of one form of cell from another, for example in relation to expression of endogenous or recombinant expression and characteristics of cells such as in respect their expression of said endogenous or recombinant expression. The analysis apparatus may be constituted by any suitable electronic apparatus, e.g., a general-purpose computer, such as a Personal Computer (PC), an embedded microprocessor, a configurable electronic circuit, such as a FPGA or the like. Further embodiments of this aspect of the present invention may comprise one or more further features of the first aspect of the present invention.

[0076] The sensor chip may be configured to be mounted within a bioreactor, e.g., from microbioreactors to large bioreactors. Microbioreactors are particularly suitable for use in personalised medicine, e.g., small batches may be required on a per patient basis.

[0077] According to a further aspect of the invention there is provided a method to screen cell lines for differences in expression of a polypeptide or peptide comprising the steps of: i) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to a first cell expressing a polypeptide or peptide; ii) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit; iii) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to a second different cell expressing the same polypeptide or peptide; and iv) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit, and v) comparing the response of said first cell and said second different cell as a measure of expression of said polypeptide or peptide by each cell.

[0078] According to a further aspect of the invention there is provided a method to monitor expression of a recombinant antibody by a mammalian cell in cell culture comprising the steps of: i) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to a cell expressing said recombinant antibody; ii) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit and obtaining a first frequency value at a first time point; iii) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to said cell expressing said recombinant antibody; iv) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit and obtaining a second frequency value at a second or further time point; and v) comparing the first and second or further frequency value to obtain a frequency difference wherein said frequency difference is indicative of said mammalian cell expressing said recombinant antibody as a measure of expression during cell culture.

[0079] In a preferred method of the invention the stability of expression is measured during cell culture.

[0080] In a preferred method of the invention wherein said mammalian cell is a Chinese Hamster Ovary cell. In an alternative preferred method of the invention said mammalian cell is a HEK cell.

[0081] Specific Embodiments

[0082] Recombinant Antibodies

[0083] Antibodies include polyclonal, monoclonal antibodies, humanised and chimeric antibodies and derived fragments comprising complementarity determining regions (CDRs). Chimeric antibodies are recombinant antibodies in which all of the V-regions of a mouse or rat antibody are combined with human antibody C-regions. Humanised antibodies are recombinant hybrid antibodies which fuse the complementarity determining regions from a rodent antibody V- region with the framework regions from the human antibody V-regions. The C-regions from the human antibody are also used. The CDRs are the regions within the N-terminal domain of both the heavy and light chain of the antibody to where the majority of the variation of the V- region is restricted. These regions form loops at the surface of the antibody molecule. These loops provide the binding surface between the antibody and antigen.

[0084] Antibodies from non-human animals provoke an immune response to the foreign antibody in the human and its removal from the circulation. Both chimeric and humanised antibodies have reduced antigenicity when injected to a human subject because there is a reduced amount of rodent (i.e. foreign) antibody within the recombinant hybrid antibody, while the human antibody regions do not elicit an immune response. This results in a weaker immune response and a decrease in the clearance of the antibody. This is clearly desirable when using therapeutic antibodies in the treatment of human diseases. Humanised antibodies are designed to have less “foreign” antibody regions and are therefore thought to be less immunogenic than chimeric antibodies.

[0085] Various fragments of antibodies are known in the art. A Fab fragment is a multimeric protein consisting of the immunologically active portions of an immunoglobulin heavy chain variable region and an immunoglobulin light chain variable region, covalently coupled together and capable of specifically binding to an antigen. Fab fragments are generated via proteolytic cleavage (with, for example, papain) of an intact immunoglobulin molecule. A Fab2 fragment comprises two joined Fab fragments. When these two fragments are joined by the immunoglobulin hinge region, a F(ab’)2 fragment results. A Fv fragment is multimeric protein consisting of the immunologically active portions of an immunoglobulin heavy chain variable region and an immunoglobulin light chain variable region covalently coupled together and capable of specifically binding to an antigen. A fragment could also be a single chain polypeptide containing only one light chain variable region, or a fragment thereof that contains the three CDRs of the light chain variable region, without an associated heavy chain variable region, or a fragment thereof containing the three CDRs of the heavy chain variable region, without an associated light chain moiety; and multi specific antibodies formed from antibody fragments, this has for example been described in US patent No 6,248,516. Fv fragments or single region (domain) fragments are typically generated by expression in host cell lines of the antibody fragment . These and other immunoglobulin or antibody fragments are within the scope of the invention and are described in standard immunology textbooks such as Paul, Fundamental Immunology or Janeway et al. Immunobiology (cited above). Molecular biology now allows direct synthesis (via expression in cells or chemically) of these fragments, as well as synthesis of combinations thereof. A fragment of an antibody or immunoglobulin can also have bispecific function as described above.

[0086] Pharmaceutically Active Polypeptides

[0087] Examples of pharmaceutical proteins include “cytokines”. Cytokines are involved in a number of diverse cellular functions. These include modulation of the immune system, regulation of energy metabolism and control of growth and development. Cytokines mediate their effects via receptors expressed at the cell surface on target cells. Examples of cytokines include the interleukins such as: IL1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32 and 33. Other examples include growth hormone, leptin, erythropoietin, prolactin, tumour necrosis factor [TNF] granulocyte colony stimulating factor (GCSF), granulocyte macrophage colony stimulating factor (GMCSF), ciliary neurotrophic factor (CNTF), cardiotrophin-1 (CT-1), leukemia inhibitory factor (LIF) and oncostatin M (OSM), interferon a, interferon p, interferon E, interferon K and co interferon.

[0088] Examples of pharmaceutically active peptides include GLP-1 , anti-diuretic hormone; oxytocin; gonadotropin releasing hormone, corticotrophin releasing hormone; calcitonin, glucagon, amylin, A-type natriuretic hormone, B-type natriuretic hormone, ghrelin, neuropeptide Y, neuropeptide YY3-36, growth hormone releasing hormone, somatostatin; or homologues or analogues thereof.

[0089] The term “chemokine” refers to a group of structurally related low-molecular weight factors secreted by cells having mitogenic, chemotactic or inflammatory activities. They are primarily cationic proteins of 70 to 100 amino acid residues that share four conserved cysteine residues. These proteins can be sorted into two groups based on the spacing of the two amino-terminal cysteines. In the first group, the two cysteines are separated by a single residue (C-x-C), while in the second group they are adjacent (C-C). Examples of members of the 'C-x-C chemokines include but are not limited to platelet factor 4 (PF4), platelet basic protein (PBP), interleukin-8 (IL-8), melanoma growth stimulatory activity protein (MGSA), macrophage inflammatory protein 2 (Ml P-2), mouse Mig (m119), chicken 9E3 (or pCEF-4), pig alveolar macrophage chemotactic factors I and II (AMCF-I and -II), pre-B cell growth stimulating factor (PBSF),and IP10. Examples of members of the 'C-C group include but are not limited to monocyte chemotactic protein 1 (MCP-1), monocyte chemotactic protein 2 (MCP-2), monocyte chemotactic protein 3 (MCP-3), monocyte chemotactic protein 4 (MCP-4), macrophage inflammatory protein 1 a (MIP-1-a), macrophage inflammatory protein ip (MIP-1-P), macrophage inflammatory protein 1-y (MIP-1-y), macrophage inflammatory protein 3 a (MIP- 3-a, macrophage inflammatory protein 3 p (MIP-3-P), chemokine (ELC), macrophage inflammatory protein-4 (MIP-4), macrophage inflammatory protein 5 (MIP-5), LD78 p, RANTES, SIS-epsilon (p500), thymus and activation-regulated chemokine (TARC), eotaxin, I-309, human protein HCC-1 / NCC-2, human protein HCC-3.

[0090] Several growth factors have been identified which promote / activate endothelial cells to undergo angiogenesis. These include vascular endothelial growth factor (VEGF A); VEGF B, VEGF C, and VEGF D; transforming growth factor (TGFb); acidic and basic fibroblast growth factor (aFGF and bFGF); and platelet derived growth factor (PDGF). VEGF is an endothelial cell-specific growth factor which has a very specific site of action, namely the promotion of endothelial cell proliferation, migration and differentiation. VEGF is a complex comprising two identical 23 kD polypeptides. VEGF can exist as four distinct polypeptides of different molecular weight, each being derived from an alternatively spliced mRNA. bFGF is a growth factor that functions to stimulate the proliferation of fibroblasts and endothelial cells. bFGF is a single polypeptide chain with a molecular weight of 16.5Kd. Several molecular forms of bFGF have been discovered which differ in the length at their amino terminal region. However, the biological function of the various molecular forms appears to be the same.

[0091] CAR-T Cell Therapy

[0092] Chimeric antigen receptors (CAR) are engineered fusion proteins designed on T cells (CAR- T) to target antigens on e.g. cancer cells. This therapy is termed CAR-T therapy and has been used effectively to treat patients with relapsed B cell lymphomas, B cell acute lymphoblastic leukaemia and multiple myeloma. However, several factors have a direct impact on positive patient outcomes including the quality and variability of donor apheresis product or T-cells, the developed CAR-T manufacturing procedure, product efficacy and the health status and conditioning of the patient. To enhance therapy efficacy, researchers are continually investigating ways to optimise CAR binding domains enabling enhanced function and reduce immune response to the CAR-T therapy by designing low immunogenic CAR-T constructs. In addition, gaining a better understanding of manufacturing parameters is key to ensure optimal T cell composition (comprising of less-differentiated, naive T cells or central memory T cells), optimal process control (presence of cytokines or inhibitors at key procedural stages) and optimised I shorter culture times to reduce T cell exhaustion and enhance potency. A key step in the manufacture of CAR-T therapy is ensuring that the genetic material which recognises the tumour (the CAR), has been introduced correctly to the T cell. This is known as transduction efficiency and has a direct impact on functional characteristics of the engineered effector cells such as proliferation, persistence, antitumour activity and safety of the CAR-T therapy. Flow cytometry analysis is the standard method for detection of transduction efficiency. However, time from sampling point and result post assay may result in a delayed response limiting the ability to make key processing decisions in real time. This study investigated the use of dielectric spectroscopy in the method of the invention for the determination of transfection efficiency without the requirement of labels or labour-intensive assay procedures.

[0093] Pluripotent Stem Cells

[0094] Pluripotent stem cells have the ability to differentiate into each differentiated or partially differentiated cell-type found in an organism. They include embryonic stem cells, embryonic germ cells, lineage restricted stem cells (also called multipotent stem cells) and induced pluripotent stem cells (cells, typically fibroblasts), that have been genetically engineered to a de-differentiated state and is then differentiated to an alternative cell type.

[0095] A Typical example includes the differentiation of pluripotent stem cells or haemopoietic cells into differentiated immune cells such as macrophages. WO2024 / 038182, the content of which is incorporated by reference in its entirety, discloses engineered stem cells that are modified to reduce or exclude expression of HLA genes (HLA1 and / or HLA2) thereby forming hypoimmunogenic stem cells that are differentiated into hypoimmunogenic macrophages useful in the treatment of inflammatory conditions. The differentiation of stems cells to macrophages is known and disclosed in WO2021 / 240162 the content of which is incorporated by reference in its entirety. A further example is provided by WO2024 / 074376, the content of which is incorporated by reference in its entirety, which discloses monocyte derived macrophages that are engineered to express cytokines such as IL- 10 and / or proteases such as matrix metalloprotease 9 (MMP9) or are derived from IPS cells similarly modified to express IL-10 and / or MMP9. WO2024 / 068728, the content of which is incorporated by reference in its entirety, discloses engineered macrophages expressing MMP9 and / or MMP12.

[0096] Viral Polypeptides and Gene Therapy

[0097] A number of viruses are commonly used as vectors for the delivery of exogenous genes. Commonly employed vectors include recombinantly modified enveloped or non-enveloped DNA and RNA viruses, for example from the family of baculoviridiae, parvoviridiae, picornoviridiae, herpesveridiae, poxviridae, adenoviridiae, picornnaviridiae or retroviridae. Chimeric vectors may also be employed which exploit advantageous elements of each of the parent vector properties. Such viral vectors may be wild-type or may be modified by recombinant DNA techniques to be replication deficient, conditionally replicating or replication competent. Conditionally replicating viral vectors are used to achieve selective expression in particular cell types while avoiding untoward broad-spectrum infection.

[0098] Adeno-associated virus (AAV) vectors are known in the art and offer, when compared to retroviral or lentiviral vectors, a variety of advantages such as their mild immune response, capability to infect a broad range of cells and that the desired DNA is not integrated into the genome resulting in potential disruption and knock out of other genes but is stored extrachromosomal in the cell. AAV comprises single-stranded DNA genome of approximately 4.8 kilobases (kb) comprising three genes with coding sequences flanked by inverted repeats which are required for genome replication and packaging. Uses of AAVs and modified AAV vectors are known in the art and disclosed in WO2019 / 032898, W02020041498 or WO2019 / 028306.

[0099] In the context of this invention the term “signature” refers to and is equal to the frequency value obtained using the method of the invention.

[0100] Throughout the description and claims of this specification, the words “comprise” and “contain” and variations of the words, for example “comprising” and “comprises”, means “including but not limited to”, and is not intended to (and does not) exclude other moieties, additives, components, integers or steps. “Consisting essentially” means having the essential integers but including integers which do not materially affect the function of the essential integers.

[0101] Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

[0102] Features, integers, characteristics, compounds, chemical moieties or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith.

[0103] An embodiment of the invention will now be described by example only and with reference to the following figures:

[0104] Figure 1 is a block diagram representation of biological measurement apparatus;

[0105] Figure 2 is a representation of a flow apparatus that is part of the biological measurement apparatus embodiment;

[0106] Figure 3 is a representation of an electrode array that is part of the biological measurement apparatus embodiment;

[0107] Figure 4 is a circuit representation of stimulation apparatus that is part of the biological measurement apparatus embodiment;

[0108] Figure 5 is a circuit representation of sensing apparatus that is part of the biological measurement apparatus embodiment;

[0109] Figure 6. A. A comparative frequency response showing the mean difference of each cell line against the control line in high (first peak) and low frequencies (right peak), p-values of the statistical comparisons of the different samples to a control sample. B. p-values less than 5% or 0.05 are used to provide statistical significance;

[0110] Figure 7A: x-axis - time of high and low frequency response, y-axis - % shift of test signature vs control signature. Signatures of control cells from Days 1 - 6 were mostly similar to one another. B. Stable AuraCyt signatures were demonstrated for Parental cells during this project, as indicated by values greater than 5% with 4 exceptions highlighted in red;

[0111] Figure 8. The signatures of clone 3 and clone 1 were statistically significantly different to the parental line (red and green spikes, p<5%). Clone 7 was also significantly different to the parental line (yellow spike - p=0.026). Clone 6, Clone 9 and Clone 14 did not significantly differ from the parental line (light blue, green, and purple spikes), x-axis - time of high and low frequency response, y-axis - % shift of test signature vs control signature;

[0112] Figure 9. A. Consistent AuraCyt signatures demonstrating detection of high producing Clone 1 , when compared to the control. The signatures of medium producer Clone 6 and low producer Clone 14 did not differ from the parental line, x-axis - time of high and low frequency response, y-axis - % shift of test signature vs control signature B. Scatterplot displaying visible shift in the scatterplots of the parental line signature (yellow) and the high producer line signature (red,); and low (green);

[0113] Figure 10. Plot displaying high producers Clone 1 , Clone 2, Clone 3 and Clone 4. All high producers were compared against the parental control and were shown to be significantly different without any exceptions, x-axis - time of high and low frequency response, y-axis - % shift of test signature vs control signature;

[0114] Figure 11. Flow cytometry analysis for CAR-CD34 expression on investigation day 6 (a) transduced T cell population and (b) non-transduced T cell population;

[0115] Figure 12 (a) PCA scatter plot generated by AuraCyt™ analysis plotting the dimensionally reduced data generated for day 0 non-transduced / non-activated T cells (Tcells DO); day 6 non-transduced / activated T cells (Tcells D6) and day 6 CAR-CD34 T cells (CAR-T D6). (b) A PCA scatter plot plotting the dimensionally reduced data generated for day 0 non- transduced / non-activated T cells (Tcells DO), day 8 non transduced / activated T cells (T cells D8) and day 8 CAR-CD34 T cells (CAR-T cells D8);

[0116] Figure 13 Hotelling’s T2 statistical analysis comparing each control and test condition. The values were generated to a 95% confidence interval, where p<0.05 indicates a significant difference. DO I D4 / D6 / D8 pertains to experimental days 0,4,6, and 8 respectively. DO NT = non transduced / non activated cells. D4, D6 and D8 NT= non-transduced / activated cells. D4, D6 and D8 CAR-T = CAR-CD34 transduced / activated cells;

[0117] Figure 14. A line graph demonstrating the relative effect size shift (to a 0.95 confidence interval) observed between AuraCyt™ signatures produced by non-transduced / activated (NT) and CAR-T (CAR-CD34) cells throughout the culture expansion process (days 4, 6 and 8) when compared against the Day 0 non-activated / non-transduced T cell profile. Figure 15. Dimensionally reduced output from AuraCyt technology assessed with the standard flow cytometry platform FCS express. Heat map plots generated for the (a) day 0 non- activated / non-transduced T cells (b) day 4 non-transduced / activated T cells, (c) day 6 nontransduced / activated T cells and (d) day 8 non-transduced / activated T cells, (e) day 4 CAR CD34 transduced / activated T cells (f) day 6 CAR CD34 transduced / activated T cells (g) day 8 CAR CD34 transduced / activated T cells. T cell populations are plotted against the first coefficient for the phase and magnitude shift in the high frequency signal received during analysis on the AuraCyt™ platform (Col5 / Col1). All plots are gated against the gate generated around the day 0 non-activated I non-transduced T cell population; and

[0118] Figure 16. Dimensionally reduced output from AuraCyt technology assessed with the standard flow cytometry platform FCS express. Heat map plots generated for (a) day 4 non-transduced / activated T cells, (b) day 6 non-transduced / activated T cells and (c) day 8 non-transduced / activated T cells, (d) day 4 CAR CD34 transduced / activated T cells (e) day 6 CAR CD34 transduced / activated T cells (f) day 8 CAR CD34 transduced / activated T cells. Plots (a) and (d) are gated against the activated non transduced cells for day 4. Plots (b) and (e) are gated against the activated non transduced cells for day 6. Plots (c) and (f) are gated against the activated non transduced cells for day 8. T cell populations are plotted against the first coefficient for the phase and magnitude shift in the high frequency signal received during analysis on the AuraCyt™ platform (Col5 / Col1).

[0119] MATERIALS AND METHODS

[0120] Detailed Description of Device

[0121] A block diagram representation of biological measurement apparatus 10 is shown in Figure 1. The biological analysis apparatus 10 comprises flow apparatus 30, stimulation apparatus 12, sensing apparatus 13, control and processing apparatus 14, and analysis apparatus 16. The stimulation apparatus 12 and sensing apparatus 13 receive, via the flow apparatus 30, a flow of analyte in the form of a fluent material in which cells are suspended.

[0122] The flow of analyte is directed by the flow apparatus 30 through the stimulation apparatus 12 and sensing apparatus 13 where it is subject to stimulation and sensing, as described in detail below, before exiting 20 from the measurement apparatus 10.

[0123] The control and processing apparatus 14 controls the application of stimulation signals to the analyte by the stimulation apparatus 12, and processes signals sensed by the sensing apparatus 13. Processing comprises amplification of sensed signals, analogue to digital conversion of sensed signals and storage of converted sensed signals. Although not shown in Figure 1, the measurement apparatus 10 further comprises a pump which is operative to push or draw analyte through the stimulation apparatus 12 and sensing apparatus 13, via the flow apparatus 30.

[0124] The analysis apparatus 16 is operative to make at least one analytical determination in dependence on the stored converted sensed signals. The analysis apparatus 16 is also operative to provide for supervisory control of the control and processing apparatus 14, e.g. in respect of a change in the form of control of the biological sensing apparatus 12 exercised by the control and processing apparatus 14.

[0125] The control and processing apparatus 14 is constituted by any suitable electronic apparatus, such as a separate analogue-to-digital converter circuit, a separate amplifier circuit and a separate electronic memory circuit, or a configurable integrated circuit, such as a System-on- a-Chip (SOC), comprising the digital circuits and an ASIC comprising the analogue circuits. The analysis apparatus 16 is constituted by any suitable electronic apparatus, e.g. a general purpose computer, such as a PC, an embedded microprocessor, a configurable electronic circuit, such as an FPGA, or the like. The control and processing apparatus 14 and analysis apparatus 16 are constituted apart from each other, e.g. as separate modules, or constituted together, e.g. in a same integrated circuit or same general purpose computer.

[0126] The flow apparatus 30 receives the analyte 18 and provides for flow of the analyte before the analyte exits 20 from the flow apparatus. With reference to Figure 2, the measurement apparatus also comprises a two-dimensional array of electrodes 32, which comprise stimulation electrodes of the stimulation apparatus 12 and sensing electrodes of the sensing apparatus 13. The flow apparatus 30 and the array of electrodes 32 are disposed in relation to each other such that the array of electrodes 32 is above a main channel of the flow apparatus 30.

[0127] The control and processing apparatus 14 is electrically coupled to the array of electrodes 32. The control and processing apparatus 14 is operative to provide for biological cell stimulation, sensing and actuation by way of the array of electrodes 32.

[0128] The arrangement of the array of electrodes is shown in more detail in Figure 3.

[0129] The array of electrodes 32 and the control and processing apparatus 14 are constituted by a CMOS process such as a 0.35 micron CMOS process. The array of electrodes 32 and the control and processing apparatus 14 are both comprised in a CMOS ASIC. Each electrode in the array 32 is 18 microns by 18 microns with a 2 micron gap between electrodes whereby the array pitch is 20 microns. The electrodes are surrounded by a busbar that prevents capacitive coupling to the silicon, and which is grounded.

[0130] The thickness and permittivity of the standard polyimide top layer of the ASIC provides insufficient capacitance for proper engagement of the electrodes 32 with the analyte, and hence the fabrication process lacks a polyimide layer deposition step so no polyimide top layer is present. The hydrophilic nature of the silicon nitride layer provides for maximum exposure. However, the silicon nitride layer may be removed over the electrodes, such that the electrodes conduct into the analyte. The ASIC is disposed, as is mentioned above, relative to the flow apparatus 30 such that the array of electrodes 32 engages with the analyte flowing through the flow apparatus 30.

[0131] The control and processing apparatus 14 of the ASIC comprises binary to decimal decoders and memory for row and column addressing of the array of electrodes 32, global configuration logic and bias circuitry for the sensor output signal paths. The global configuration logic is operative to provide for the like of memory resetting and the gating of control signals with respect to a global reset signal to ensure all control lines power up in a known state.

[0132] The measurement apparatus 10 further comprises a Printed Circuit Board (PCB) which supports and provides for electrical connectivity for electrical circuits which support the ASIC. The electrical circuits comprised in the PCB includes an SOC which is configured to provide various digital functions including the generation of stimulus signals, addressing of individual electrodes in the array of electrodes 32 and communication with a Universal Serial Bus (USB) module.

[0133] The SOC 44 is operative to generate a stimulation signal in the form of an m-sequence that is stored in memory and output one bit at a time. In particular, in this embodiment, the stimulation signal consists of a higher frequency m-sequence, which is generated from a linear feedback shift register having 5 registers, and a lower frequency m-sequence, which is generated from a linear feedback shift register having 7 registers. The higher frequency m-sequence therefore has a length of 31 bits and the lower frequency m-sequence has a length of 127 bits. The bit rate of the higher frequency m-sequence of the stimulation signal is 950Mbps, and the bit rate of the lower frequency m-sequence of the stimulation signal is 50Mbps.

[0134] The PCB also includes input signal conditioning circuitry which is configured to receive stimulus signals from the SOC or from the external (un-illustrated) signal generator and provide for programmable gain amplification of the voltage swing of the stimulus signals.

[0135] In addition, the PCB includes output signal conditioning circuitry which performs a variety of functions including fixed gain, low distortion amplification of sensed single ended signals followed by programmable gain amplification or attenuation of such initially amplified signals under the control of a PC.

[0136] The output signal conditioning circuitry also includes an analogue-to-digital converter. The sample rate of the analogue-to-digital converter is 200MHz. The analogue-to-digital converter is configured to sample the higher frequency m-sequence of the stimulation signal in four interleaved passes, providing a sample length of 124 bits. The theoretical bandwidth of this m-Sequence measurement is therefore 1 ,9GHz - 30.65MHz. Due to the x4 oversample ratio, the flat region of this bandwidth only extends to 475MHz, 1 / 4 of the theoretical bandwidth, but that through post processing, the higher frequency regions above this point are also accessible.

[0137] The analogue-to-digital converter is configured to sample the lower frequency m-sequence of the stimulation signal at 200MHz, which is four times greater than the bit rate of the lower frequency m-sequence of the stimulation signal. However, the analogue-to-digital converter is configured to discard samples that align with a transition point of the m-sequence, and to average the remaining 3 samples to provide the digital output.

[0138] Because the stimulus for both high and low frequency sections of the stimulus signal are from the same source, the low frequency stimulus signal must align to the sampling method. In this case, each low frequency sample period (20ns) is made up of 4 analogue-to-digital converter samples (5ns each), as well as 19 high frequency stimulus samples (~1.052ns each); 19*1.052ns = 20ns.

[0139] To stop significant contamination from one stimulus pattern to the next, a gap of at least one m-sequence period is maintained between the two sections of the capture.

[0140] In this specific implementation, the entire stimulus package, which includes both the high and low frequency signals, are stored in a single section of memory. This is then streamed out, bit by bit, at the high frequency data rate (950MHz in this example). The low frequency signal is created by padding out the (in this case 7-bit) M-Sequence signal by a factor of 19, so each bit lasts for 19 high frequency periods.

[0141] As described above, the PCB comprises a USB module. The USB module provides for communication with a PC running software operative to perform the functions of the analysis apparatus 16 of Figure 1. More specifically, the PC is operative to configure the ASIC 42 and the circuits comprised in the PCB. In addition, the PC receives real-time sensed data or blocks of data which have been acquired and stored locally from the SOC.

[0142] The PC is operative to decode the received m-sequence encoded data by cross-correlation to provide the impulse response. The PC is further operative to perform a Fast Fourier Transform (FFT) on the decoded data to thereby provide frequency domain data. The frequency domain data is then displayed for user interpretation.

[0143] The PC is also operative to count biological cells present in the analyte and to determine a density of cells present in the analyte in dependence on the flow rate and volume of the flow apparatus with the count and density information being displayed to the user.

[0144] A representation of a flow apparatus 30 comprised in the biological sensing apparatus of Figure 2 is shown in detail in Figure 2. The flow apparatus 30 of Figure 2 is formed from glass and has a length of about 25mm and a width of about 10mm. The flow apparatus 30 comprises a main channel 34 through which the analyte flows. The array of electrodes 32 is disposed above the main channel 34 so that the electrodes 32 engage with the analyte as the analyte flows through the main channel.

[0145] As is described above, the array of electrodes 32 is comprised in a CMOS ASIC. The CMOS ASIC and the flow apparatus 30 are releasably attached to each other by way of a fastener apparatus comprising a silicone gasket layer, such that a proper relative disposition of electrodes and main channel is achieved.

[0146] The flow apparatus 30 also comprises a sample inlet 40 which receives the analyte, e.g. by way of injection, and a sample outlet 50 at an opposite end of the flow apparatus from the sample inlet 40. The sample inlet 40 and the sample outlet 50 are each in fluid communication with the main channel 34. In addition, the flow apparatus comprises first and second further inlets 42, 44. The first further inlet 42 is disposed laterally on one side of the sample inlet 40 and the second further inlet 44 is disposed laterally on the other opposite side of the sample inlet. Each of the first and second further inlets 42, 44 are in fluid communication with the main channel 34.

[0147] In use, a sheath fluid is received by each of the first and second further inlets 42, 44 to thereby provide for a flow of sheath fluid in the main channel, the flow of sheath fluid being lateral of a flow of analyte received by the sample inlet 40. The flow of sheath fluid provides for registration of the biological cell comprising analyte with the array of electrodes 32 and also helps preserve the integrity of the flow of analyte as it progresses though the main channel.

[0148] Electric field stimulation and electric field sensing can both be accomplished within a CMOS ASIC of the form described above. More specifically the array of electrodes 32 is used for both electric field stimulation and electric field sensing, with different sets of electrodes being used for stimulation and sensing. A stimulation cell 100 configured for single ended operation is shown in Figure 3. The stimulation cell 100 of Figure 3 comprises a single electrode 102, which is comprised in the array of electrodes 32. A stimulation signal is applied to the stimulation electrode 102.

[0149] The stimulation cell 100 further comprises a multiplexer 112 which provides for one of two states selected in accordance with a state selection bit 118. The stimulation cell 100 also comprises a memory bit 114 that stores the state of the first state selection bit 118. The memory bits 114 are constituted in Static Random Access Memory (SRAM).

[0150] The multiplexer of Figure 3 provides for one of two states. To provide for one of the two states, the electrode is addressed with the address-sensitive first state selection bit 118 and then the first state selection bit 118 is stored as the memory bit 114.

[0151] In a first state when the state selection bit 118 is at zero, the electrode 102 is connected by way of a switch to common ground potential. In a second state when the first state selection bit 118 is at one, the electrode 102 is configured for stimulation whereby the electrode receives a stimulation input from a signal bus 124. The signal bus 124 is electrically connected to the part of the control and processing apparatus 14 which is operative to generate stimulation signals, as described above.

[0152] Each stimulation electrode 102 in the array of electrodes 32 comprises the multiplexer and memory circuitry shown in Figure 4.

[0153] A sensing cell 200 configured for single ended operation is shown in Figure 5. The sensing cell 200 of Figure 4 comprises a single electrode 202, which is comprised in the array of electrodes 32. The sensing cell 200 also comprises an output buffer 215 and an output pin 230.

[0154] The electrode 202 is configured for sensing the electrodynamic response field, whereby the electrode 202 is connected to a sensor output pin 230 via the output buffer 215, which is addressed by second state selection bit 225. The sensor output pin 230 is electrically connected to the part of the control and the processing circuitry 34 that is operative to process sensed signals, as described above.

[0155] The stimulus electrodes 102 and the sensing electrodes 202 are arranged in a rectangular array of electrodes 32, which is shown schematically in Figure 3.

[0156] The array of electrodes 32 comprises two columns aligned with the flow of analyte, with a first column consisting of stimulus electrodes 102 and a second column consisting of sensing electrodes 202. The rectangular array of electrodes 32 may comprise a transverse arrangement of multiple pairs of the two columns of electrodes, for example an array of 32 columns and 8 rows, with alternating rows of stimulus electrodes 102 and sensing electrodes 202, as shown in Figure 3.

[0157] In the embodiment of Figure 3, the main channel of the flow apparatus 30 has a width that is sufficient for five columns of electrodes to be disposed within the main channel, with the remainder of the electrodes of the array 32 being outside the main channel of the flow apparatus. The electrodes that are active, in use, are selected to be in the columns, eg two columns, that are generally central in the main channel of the flow apparatus, where there is laminar flow.

[0158] The electrodes that are configured for use in Figure 3 are the stimulation electrodes 102a and 102b, and the sensing electrodes 202a and 202b. In particular, a stimulation signal is provided to stimulation electrode 102a, and the same stimulation signal or a complementary signal of oppositive polarity is provided to stimulation electrode 102b. The stimulation electrode 102a forms a capacitor with the adjacent sensing electrode 202a, with the fluent material and any cell therein being located within the electrical field between these electrodes 102a, 202a. Similarly, the stimulation electrode 102b forms a capacitor with the adjacent sensing electrode 202b, with the fluent material and any cell therein being located within the electrical field between these electrodes 102b, 202b.

[0159] The output signals from the sensing circuit associated with the sensing electrodes 202a, 202b are compared, e.g., by subtraction, such that a differential measurement is made. In particular, the concentration of cells and the flow rate is controlled such that each cell in the flow passes over the first pair of electrodes 102a, 202a and then the second pair of electrodes 102b, 202b. The differential measurement will be zero until a cell passes over the first pair of electrodes 102a, 202a or the second pair of electrodes 102b, 202b, when a differential signal will be output.

[0160] The distance between the first pair of electrodes 102a, 202a or the second pair of electrodes 102b, 202b is selected to be sufficient that the respective electrical fields are separated, such that a cell isn’t detected by both pairs of electrodes at the same time.

[0161] The pump of the measurement apparatus 10 of Figure 1 is controlled in dependence on at least one of: a rate of flow of analyte through the measurement apparatus 10; and a level of confidence of characterisation of the analyte flowing through measurement apparatus 10. Considering rate of flow of analyte further, the separation between pairs of electrodes in the array is known and the time of travel of biological cells between pairs of electrodes is determined by the control and processing apparatus 14.

[0162] The control and processing apparatus 14 is then operative to determine the speed of movement of biological cells through the measurement apparatus 10. The control and processing apparatus 14 is then operative to control the pump in dependence on the determined speed. For example, if the determined speed is below a predetermined value the control and processing apparatus 14 is operative to increase the flow rate by controlling the pump. Considering level of confidence of characterisation of the analyte further, the control and processing apparatus 14 is operative to characterise biological cells and to determine a level of confidence of the characterisation. The control and processing apparatus 14 is further operative to compare the determined level of confidence with a predetermined level and then to control the pump in dependence thereon. If the determined level of confidence is below the predetermined level the control and processing apparatus 14 is operative to reduce the rate of flow by controlling the pump to thereby provide for improved characterisation of the cells.

[0163] Cell Culture Protocols

[0164] Table 1 discloses 15 CHO cell lines (1 parental (control) line and 14 mAb producer clones comprising of high, medium and low mAb producers, see Table 1 for more details). The parental line is a non-transfected line whereas all producer clones have been transfected with the gene of interest for mAb production. All lines were grown under standard cell culture procedures.

[0165] Testing using Prototype Sensor

[0166] Three phases were performed to complete the testing strategy:

[0167] Phase 1 - The control line was tested in triplicate on 2 different instruments every passage day. In addition, single samples of each clone were tested on 2 different instruments every passage day. Each test / control (passing QC test) generated a signature or frequency value (Days 1-4).

[0168] Table 1. List of all cell lines tested during the Collaborator 1 project and their respective titre and status classification information, as provided by Collaborator 1.

[0169]

[0170] Each test and control generated a signature or frequency value. (Days 5-7).

[0171] To test the sensor, cells were washed in Buffer P as shown in Appendix 3, then loaded onto the instrument by taking a 150ul sample at 1x106cells / ml.

[0172] Buffer P Formulation

[0173] Buffer P was formulated as shown in Table 2 below, prepared fresh on a daily basis prior to testing.

[0174] Table 2. List of reagents used for Buffer P formulation

[0175] All reagents were within expiry date.

[0176] Buffer P metrics: pH, conductivity and osmolality

[0177] The pH, osmolality and conductivity of Buffer P were measured after formulation to ensure that the buffer was within optimal parameters (data not shown). All buffer metrics were within accepted range with the exception of pH which was within 6.0 - 6.5. The optimal pH for cell maintenance is 7 - 7.4, however a balance was required to maintain cell viability and conductivity below 125 uS / cm, which translates to 0.5% ionic content. 0.5% PBS Mg+ / Ca+ was identified as essential to preserve cell viability throughout testing whilst maximising sensor signal output, allowing for no further pH buffering agents to be added to Buffer P. Following an initial drop in cell numbers upon wash in Buffer P, viable cell concentration remains stable over the course of one hour.

[0178] Table 3. List of equipment, consumables and reagents utilised in this study.

[0179] All equipment had been calibrated or within warranty and operating optimally. All reagents were within expiry date.

[0180] Data collection, Quality Control and Statistical Analysis.

[0181] Signatures or frequency values were generated using the technology parameters described above.

[0182] Sensor in-house software allows us to check the quality of data collected in each test. The QC pipeline performs 12 different tests, assessing different parameters of the data in each sample. Each test has “pass-fail” result, which has been set based on extensive testing and analysis of parameters governing the elements which constitute ‘an acceptable test’. Failed test conditions are: Impulse Drift more than 5% and Background vs Time more than 1000. Sources of variation / test failure can include a bubble in the sample, a small clump or higher ionic content due to residual media ions.

[0183] After QC assessment, sensor signatures were generated for each high, medium, low clones and parental line. Signals collected by the sensor were processed using algorithms to create unique signatures. As multiple frequencies are used to assess each cell, a ‘point’ in multidimensional space is generated per cell. For ease of interpretation, the output is flattened into 2-dimensional scatterplot.

[0184] Sensor signatures were compared against one another to determine statistical significance using a comparison algorithm to assess high dimensional data sets (modified T-test for high dimensional data). Test samples are assessed against controls to generate a differential impulse response plot (Figure 6).

[0185] CAR-T Cell culture and transduction

[0186] Donor T cells were seeded into static T-175 flasks using RPMI media + 10% FBS + 2mM L- glutamine and cultured for 3 days prior to commencing the investigation. Day 0 was the time point prior to the day of activation. On day 1 of the investigation, 30 lll / mL IL-2 was added to the culture media to activate donor T cells. On day 2, the culture was divided to accommodate two conditions; a control population and a transduced test population. The control population was non-transduced and maintained in static T-25 flasks throughout the duration of the experiment. The test population was transduced with a CAR-CD34 lentiviral vector in retronectin coated 6-well plates and then reseeded into a T-175 flask.

[0187] Test and control populations were held in the appropriate flasks during a pre-expansion phase (day 3 and day 4) then expanded using key cytokines at key points of the manufacturing process from day 5 to experiment completion (day 8). Flow cytometry analysis was performed on day 6 using standard flow cytometry procedures to confirm the surface expression of CAR- CD34. Cell viability and concentration were analysed daily using the NC-3000 automated cell counter.

[0188] Testing using AuraCyt™ Technology (Dielectric spectroscopy)

[0189] All testing was performed on the Celledonia™ instrument using AuraCyt™ technology. At each testing timepoint (days 0, 4,6 and 8) 1 ml of sample was removed from each culture vessel for each condition (test I control), centrifuged for 5 mins at 400 ref, the supernatant was removed with the resulting cell pellet diluted to 100,000 cells / ml in PBS prior to loading onto Celledonia™. Samples were run through Celledonia™ in triplicate. Beads were run at the beginning and end of each experimental day to confirm that the instrument was performing within optimal parameters throughout the duration of the experiment.

[0190] Data collection and statistical analysis using Maestro

[0191] AuraCyt™ data was presented in 2D PCA scatterplots using the dimensional reduction algorithms built into the analysis software. A multivariate hotelling T2-testwas then performed to determine statistically significant similarities or differences when comparing test and control populations throughout the duration of the experiment. A statistical power analysis was performed to set the power of the test to ensure differences between events of the same cell type were not found to be significant. This resulted in the data being analysed in blocks of 200 events. In addition, effect size and 95% confidence intervals were calculated to quantify shifts between samples of each test and control condition.

[0192] Data Analysis using FCS express

[0193] Dielectric spectroscopy using AuraCyt™ data was assessed using standard flow cytometry analysis software, FCS express. To enable this, phase and magnitude shifts of the signal received from the cells (as they pass over the sensor) were calculated for both the high and low frequency signals generated. These shifts were then plotted as four separate 3’dorder polynomial curves in AuraCyt™ those being the high frequency phase shift curve, the high frequency magnitude shift curve, the low frequency phase shift curve and low frequency magnitude shift curves.

[0194] The four coefficients from these curves were then exported from AuraCyt™ software and imported into FCS express software. Heatmap plots were generated for each of the groups against two of the chosen coefficients. Gates were drawn around baseline profiles and compared against subsequent samples analysed to track the change in the samples analysed between different conditions over time.

[0195] Example 1 QC analysis of the data

[0196] All data collected over the testing period was subjected to extensive QC. Only events between 200-600 seconds of testing were analysed. Any tests that failed criteria specified in section

[0197] 2.5 were excluded from analysis. The list of excluded tests is summarised in Table 5:

[0198] Table 1. Summary of all samples excluded from analysis due to failed QC

[0199] Example 2 Stable sensor signatures were demonstrated for parental cells

[0200] To demonstrate that a consistent control was generated to enable comparisons between the parental line and producer clones, sensor signatures were generated from Day 1 to Day 6 for the parental cells. Day 7 testing data was not included due to limited data.

[0201] For all comparisons (with only 4 exceptions as highlighted by the red boxes, Figure 7B), parental signatures were significantly similar (p>5% or 0.05) which positively affirms the generation of robust sensor signatures.

[0202] The signature of the parental control line remained stable over time, with only 4 exceptions that most likely attributed to the instrument- instrument and test-test variation present in Prototype Alpha. This may be due to the signal-to-noise ratio limitations of the current system which can be mitigated with instrument compensation systems.

[0203] This provided confidence that the parental line is an acceptable and stable control against which all producer line sensor signatures can be tested.

[0204] Example 3 Phase 1 : Sensor can detect differences between producer clone status

[0205] Phase 1 data were pooled from two separate instruments, 1 replicate and 4 test days (i.e. Modules 12 and 14 and replicates from test days 1-4, refer to section 2.3 for further detail).

[0206] Figure 8 shows a subset of comparisons between the parental line and two high producers (Clone 3 and Clone 1), two medium producers (Clone 7 and Clone 6), and two low producers (Clone 9 and 14). The full dataset is shown in Table 2.

[0207] The signature or frequency value of both high producers were significantly different to the parental line (p<0.05). Medium producer Clone 7 was also significantly different to the parental line (p<0.05). However, the signatures of Clone 6, Clone 9and Clone 14 were similar to the control (p>0.05).

[0208] Greatest differences were observed in high frequencies indicating that cellular complexity rather than cell membrane factors were important in the differentiation between control and high producing CHO.

[0209] Example 4 Objective 2: Phase 2: sensor can reliably detect response, y between producer clone status

[0210] Figure 9A displays Phase 2 data for clones. Clone 1 , Clone 6, Clone 14 and parental line Control tested in triplicate to increase robustness of Phase 1 results. Data were pooled from two separate instruments, 3 replicates and 3 test days (i.e. Modules 12 and 14 and replicates from test day 5, 6 & 7, refer to section 2.3 for further detail). This plot displays the mean difference of each cell clone against the control line.

[0211] The high producer line Clone 1 is significantly different to the parental line (parent, p=0.00472 - orange line). However, both the medium and low producer lines (Clone 6 and D11 respectively, p>0.05 for both comparisons) show no significant difference when compared against the control (Green line = medium producer I yellow line = low producer). This is consistent with phase 1 testing (refer to section 2.3 for further detail).

[0212] Figure 9 B displays visible shift in the scatterplots of the parental line signature (yellow) and the high producer line signature (red, C136).

[0213] Example 5 Sensor can reliably identify all high producing mAb CHO clones tested in this project

[0214] All high producers were compared against the parental control (Figure 10) and were shown to be significantly different (p<0.05).

[0215] High producing clones were statistically different to all other medium and low clones (p<0.05) data not shown. This positively affirms that the sensor can reliably identify high producer clones. Example 6 Direct correlation of AuraCyt analysis against titre p-values lower than 5% (<0.05) correlated with high antibody production in 100% of cases (green). One medium and one low producer CHO clones (green) were significantly different to the control. Most medium and low producing clones appear similar to the parental control (p>5%).

[0216] Table 2. Correlation of AuraCyt p value of each producer compared against the control, correlated to Final Titre, p value less than 5% indicates significant differences to the parental line (highlighted in green). The signature of the parental line remained stable over time, with only 4 exceptions that most likely attributed to the instrument- instrument and test-test variation present in prototype sensor. This may be due to the signal-to-noise ratio limitations of the current system which can be mitigated with instrument compensation systems. However, the consistency of this profile made it an ideal control against which producer lines were compared.

[0217] Dielectric spectroscopy can detect statistically significant differences in signatures of all high producing CHO clones when compared to the parental control with direct correlation against reported titres.

[0218] The greatest differences were observed in high frequencies indicating that differences were due to cellular complexity rather than cell membrane factors. This may be important in the differentiation between control, low, medium and high producing CHO and the rapid selection of high producing clones enhancing the efficiency of the cell line development phase and development reducing the cost of production.

[0219] Generally medium and low producing clones were significantly similar to the parental control which may allow these clones to be excluded quickly rather than integrating into prolonged testing strategies.

[0220] Only one medium and one low producing clone were significantly different to the parental control.

[0221] Example 7

[0222] CAR-T transduction using the lentiviral vector was deemed successful as confirmed by the expression of the CAR (CAR CD34) by flow cytometry on Day 6 where 44% of the transduced cells were positive for the CD34 marker. This confirms expression of the CD34 CAR on the cell surface. As expected, the non-transduced cells expressed only 0.38% CD34 (Figure 11).

[0223] Example 8

[0224] To assess if AuraCyt™ could identify CAR-T CD34 cells by detecting this shift in expression based on label-free cell capacitance measurements, both test and control cells were tested using AuraCyt™ on days 0, 4, 6 and 8 and the resulting PCA plots are shown in Figure 12 (data for Day 4 is not shown). From this PCA display, there are clear shifts when comparing non-activated (T cells DO) vs activated cells (T cells D6 or D8) regardless of the CAR expression. This may be indicative of T cell activation and expansion when 30 lll / mL IL-2 was added to the culture media.

[0225] An even greater shift was displayed when comparing non transduced cells (T-cells DO / D6 / D8) with transduced cells (CAR T cells D6 / D8). This shows that AuraCyt™ can detect and differentiate between expanding cells and CAR-T CD34 expressing cells.

[0226] Example 9

[0227] To demonstrate if the observed shifts were significant, the Hotelling’s T2 statistic was applied to the data to compare all test and control conditions against one another. As shown by Figure 13, all direct condition comparisons were significantly different from one another (p<0.05). However, this statistic alone does not measure the size or direction of the differences observed between the test and control populations over time.

[0228] Example 10

[0229] To investigate this further, the relative effect size (to a 0.95% confidence interval) was calculated for non-transduced and transduced CAR-T CD34 cells at each time point (day 4, day 6 and day 8) where each condition was compared to the day 0 control (non-transduced I non activated). By plotting effect size against test days (Figure 14), both non-transduced and CAR-T CD34 populations changed significantly (p<0.05) and increasingly from the control population over the duration of the experiment. This indicates that the process of T cell activation with IL-2 is reflected by an increase in effect size over time. In addition, when using day 0 as the control, effect size was greater when comparing CAR-T cells with the nontransduced cells. This indicates that transduction of the CAR-CD34 gene and subsequent expression of the CAR could be identified and differentiated from cell activation using AuraCyt™ technology.

[0230] Taken collectively, AuraCyt™ analysis has the ability to qualitatively identify cell populations as they progress through a manufacturing event using PCA scatter plots (Figure 12). These qualitative outcomes can be quantified with significance using the Hotelling’s T statistic (Figure 13) alongside measurements of effect size and confidence intervals (Figure 14) to provide an understanding of key manufacturing inflection points that can be used to track the manufacture of therapeutic cells. Tracking cells in real time during the manufacture of CAR-T therapies has proven challenging to researchers due to labour intensive procedures associated with flow cytometry and the variable nature of donor starting material. However, combining the dimensionally reduced output from AuraCyt™ technology with the standard flow cytometry platform FCS express enables the generation of heatmaps that can be gated using the identified key manufacturing inflection points with the ability to generate a system to track cells as they progress through a manufacturing event.

[0231] Example 11

[0232] Figure 15 shows how both activated non-transduced and activated CAR-CD34 cells were gated against the day 0 control (non-activated I non-transduced). The day 0 gate was then applied to subsequent timepoints (days 4, 6 and 8) for both the non-transduced and CAR- CD34 cell conditions. Both cell conditions shifted from the control gate over time with the greatest shifts observed by the CAR-CD34 cells where <2% of the cells remained within the control gate from day 6 onwards. A distinct shift was also observed by the non-transfected cells, but this was less marked where approximately 20% of the cells remained within the gate by day 8. Using these known shifts, a defined gating strategy could be generated to develop an assay for the successful monitoring of T cell activation.

[0233] Example 12

[0234] To refine this gating strategy further and to differentiate between activated non-transduced cells and activated cells expressing the CAR, gates were drawn around the activated nontransduced cells and applied to the CAR-CD34 cells (Figure 16). Gates were generated independently for each test day to eliminate the effect of activation. On day 4, >75% of both non-transduced and transduced cells fall within the same gate. This may indicate that although activated, the CAR-CD34 cells are not showing population differences that could be attributed to CAR expression. By day 6, CAR-CD34 cells show a clear shift from the non-transduced gate (<10% of transduced cells remain within the gate) which may be attributable to CAR expression. This hypothesis is supported by the expression of CD34 on the surface of the CAR-CD34 cells on day 6 as shown by Figure 11 (44% positive for CD34 expression). By day 8, there was a slight increase in the number of CAR-CD34 cells falling within the nontransduced gate (37.18%). Further study is required to investigate this observation, but this increase may be attributed factors other than CAR expression and may correlate to further key inflection points of the CAR-T manufacturing process such as exhaustion or reduced expansion rates. From the data presented, AuraCyt™ can identify and differentiate transduced cells from controls with the potential to monitor CAR-T manufacturing events. This technology can monitor cells at the single cell level without the use of lasers or labels enabling real-time decision making for process development such as enhancing CAR performance in optimised growth or manufacturing conditions. As donor cell starting material is variable, AuraCyt™, in conjunction with supporting software platforms such as FCS express enables the tracking of subtle cell changes enabling actionable insights for adaptive control within the manufacturing space. This allows the researcher to pinpoint critical material attributes and critical process parameters which are key when enhancing product quality and efficacy.

Claims

CLAIMS1. A method to identify a eukaryotic cell clone transfected or transduced with a nucleic acid molecule adapted to express at least one recombinant polypeptide or peptide comprising the steps of: i) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to a non-transfected eukaryotic cell; ii) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit and obtaining a first frequency value; iii) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to a eukaryotic cell clone of i) above wherein said eukaryotic cell clone is transfected with a nucleic acid molecule adapted to express at least one recombinant polypeptide or peptide; iv) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit and obtaining a second frequency value, and v) comparing the first and second frequency value to obtain a frequency difference wherein said frequency difference is indicative of said eukaryotic cell expressing said polypeptide or peptide.

2. The method according to claim 1 wherein said nucleic acid molecule encodes an antibody or antibody fragment.

3. The method according to claim 2 wherein said antibody or antibody fragment is a therapeutic antibody or antibody fragment.

4. The method according to claim 1 wherein said nucleic acid molecule encodes a pharmaceutically active polypeptide or peptide.

5. The method according to claim 1 wherein said nucleic acid molecule encodes one or more viral polypeptides.

6. The method according to claim 5 wherein said one or more viral polypeptide is selected from the group: an adenoviral polypeptide, an adeno-associated viral (AAV) polypeptide and a lentiviral polypeptide.

7. The method according to claim 6 wherein said one or more adeno associated viral polypeptides from an AAV selected from the group: AAV2, AAV3, AAV6, AAV13; AAV1 , AAV4, AAV5, AAV6, AAV9 and AAVrhIO.

8. The method according to claim 1 wherein said nucleic acid molecule encodes a chimeric T cell receptor.

9. The method according to claim 1 wherein said nucleic acid molecule encodes one or more pluripotency associated genes.

10. The method according to claim 9 wherein said nucleic acid molecule encoding one or more pluripotency associated genes is transiently expressed.

11. The method according to any one of claims 1 to 10 wherein said eukaryotic cell is a mammalian cell.

12. The method according to claim 11 wherein said mammalian cell is a Chinese Hamster Ovary cell.

13. The method according to claim 11 wherein said mammalian cell is a HEK cell.

14. The method according to claim 11 wherein said mammalian cell is a somatic cell.

15. The method according to any one of claims 1 to 3 wherein said eukaryotic cell is a fungal cell.

17. The method according to any one of claims 1 to 3 wherein said eukaryotic cell is a plant cell.

18. The method according to any one of claims 1 to 3 wherein said eukaryotic cell is an insect cell.

19. Use of impedance or dielectric spectroscopy for the identification and analysis of a mammalian cell transfected with a nucleic acid molecule encoding a recombinant polypeptide or peptide.

20. The use according to claim 19 wherein said recombinant polypeptide or peptide is an antibody or antibody fragment expressed by the mammalian cell at a high antibody titre.

21. The use according to claim 19 wherein said recombinant polypeptide or peptide is a pharmaceutically active polypeptide or peptide.

22. The use according to claim 19 wherein said recombinant polypeptide or peptide is a viral polypeptide.

23. The use according to claim 19 wherein said recombinant polypeptide is a chimeric T cell receptor.

24. The use according to claim 19 wherein said recombinant polypeptide is a polypeptide that induces pluripotency in said mammalian cell25. A method to monitor expression of a recombinant antibody by a mammalian cell in cell culture comprising the steps of: i) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to a cell expressing said recombinant antibody; ii) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit and obtaining a first frequency value at a first time point; iii) providing an electric stimulus signal to at least one stimulus electrode, such that the at least one stimulus electrode generates a pre-determined, electrodynamic field that is applied to said cell expressing said recombinant antibody; iv) sensing a response in the electrodynamic field by means of at least one sensing electrode and providing a corresponding electric response signal to a sensing circuit and obtaining a second frequency value at a second or further time point; and v) comparing the first and second or further frequency value to obtain a frequency difference wherein said frequency difference is indicative of said mammalian cell expressing said recombinant antibody as a measure of expression during cell culture.

26. The method according to claim 25 wherein the stability of expression is measured during cell culture.

27. The method according to claim 25 or 26 to wherein said mammalian cell is a Chinese Hamster Ovary cell.

28. The method according to claim 25 or 26 wherein said mammalian cell is a HEK cell.

Images