WO2016020561A1

WO2016020561A1 - Bioimpedance measurement system for wirelessly monitoring cell cultures in real time, based on an oscillation test using integrated circuits

Info

Publication number: WO2016020561A1
Application number: PCT/ES2015/000101
Authority: WO
Inventors: José Andrés MALDONADO JACOBI; Alberto Yufera Garcia
Original assignee: Universidad De Sevilla
Priority date: 2014-08-06
Filing date: 2015-07-30
Publication date: 2016-02-11
Also published as: ES2562810A1; ES2562810B1

Abstract

The invention relates to a novel bioimpedance measurement system for wirelessly monitoring cell cultures in real time. The system uses electrodes as bioimpedance sensors and implements a "biological oscillator" circuit for measurement with integrated circuits. The invention uses oscillation parameters (frequency, amplitude, phase etc.) as empirical markers for carrying out a suitable interpretation in terms of identification of cell size, cell count, cell growth, growth rate, etc.

Description

Title

Bioimpedance measurement system for real-time and wireless monitoring of cell cultures based on an oscillation test using integrated circuits

Object of the invention

The object of the present invention relates to a new bioimpedance measurement system for real-time and wireless monitoring of cell cultures. The system uses electrodes as bioimpedance sensors and implements a "biological oscillator" measuring circuit with integrated circuits. It is proposed to use the oscillation parameters (frequency, amplitude, phase, etc.) as empirical markers to perform a proper interpretation in terms of identification of cell size, cell count, cell growth, growth rate, etc.

The invention is framed within the measure of electrical impedance of biological material. It also refers to an electronic sensor device and all the circuitry necessary to carry out said measurement.

State of the art

Characterizing in detail the number of cells in a culture at a specific time, as well as measuring the rate of cell proliferation, has wide involvement in biomedicine, both technically and biologically. The density of the cells in the dish or culture vessel will affect the success of many different techniques, including transfection, infection, reaction to many chemicals, drugs, motility, etc. The knowledge in time of the cell density of cell cultures and its temporal evolution can help improve the quality of experiments performed by professionals in biomedical laboratories, as well as reduce material and human resources costs. From a biological point of view, it is a powerful tool for the study of a cell: excessive proliferation is the hallmark of a cancer cell and any physical or chemical treatment that reduces the speed of cell growth leads to potential cancer therapy; The lack of proliferation usually indicates excessive cell death due to extrinsic or intrinsic mechanisms, or the activation of aging cell pathways (the cellular senescence). Thus, understanding the processes that reduce or accelerate the proliferation rate of a crop is a powerful research tool.

However, despite its importance, biologists cannot monitor cell proliferation in detail. At this level, two important technical problems have to be addressed. First, the protocols for estimating the number of cells in a conventional culture dish are highly invasive, and profoundly affect the biology of the cells. In most cases, longer cells cannot be used for subsequent experiments. Second, scientists can only take snapshots of the crop at specific points. Thus, for the majority of researchers using mammalian cells, the only alternative is to establish parallel experiments and reconstitute the progression of a culture using such experiments. In many cases this methodology leads to retrospective analysis, and the understanding of how cells have behaved can only be done at the end of the experiment, as an aggregate of several similar sub-experiments.

Many biological parameters and processes can be detected and controlled by measuring their bioimpedance, with the advantage of being a non-invasive and relatively cheap technique. The growth of a cell, changes in cell composition or changes in cell location are just a few examples of processes that can be detected by micro-electrodes through impedance changes [1-4]. This technique (Electrical Cell-substrate Impedance Spectroscopy, ECIS) was invented by Ivar Giaever and Charles Keese in 1986 [1], registering in a patent an apparatus for cell culture monitoring, based on a series of wells where the culture is performed cell, each of them with an array of micro-electrodes through which an alternating current is introduced, measuring the resulting electrical impedance. This initial patent [5] was completed with a series of related patents, applied to the subject of the study of cell mobility [6] or metastatic activity of cancer cells [7].

Another bioimpedance measurement system was registered in 2005 by B. Rubinsky et al. [8]. This system uses two electrodes between which a potential difference is applied and a dielectric membrane with micro-holes through which it is forced The passage of electric current. In Spain, methods have also been recorded for the simultaneous determination and visualization of electrical bioimpedance signals in biological material at various frequencies [9], using a treatment of the excitation and response signal as two independent functions in the time domain, and applying signal processing techniques (cross correlation and Fourier transformation) to obtain better results.

In general, for the problem of measuring a given impedance Z _x , of magnitude Z _xo and phase Φ, several methods have been described, which require excitation and processing circuits. The excitation is usually implemented with alternating current (AC), while the processing is based on the principle of coherent demodulation [10] or synchronous sampling [11-12]. In both, the circuit processing must be synchronized with the excitation signals, as a requirement for the technique to work, obtaining the best noise performance when the appropriate filter functions (High-Pass (HP) or Low-Pass are incorporated) (LP)).

This paper presents a new impedance measurement system for biological samples useful for obtaining information from a cell culture in real time and wirelessly. It is based on the use of a two-dimensional array of electrodes as bioimpedance sensors, integrated circuits for the implementation of the measurement circuit (oscillator) and on the use of the resulting electrical signals (oscillations) as parameters of interpretation of the culture status.

The idea is conceptually different from the previously reported methods, since it does not use excitation signals. It is about converting the "biological circuit" (formed by cell culture, electrodes and a few added electronic components) into an oscillator. Any modification of the biological part (for example a change in the number of cells) will produce changes in both the frequency and amplitude of the oscillations and this alteration will become observable to draw conclusions about the behavior of the culture in real time.

For all these reasons, any technical input that facilitates the monitoring of cell progression, especially real-time monitoring, has an immediate impact on the biological sciences; more particularly those that reduce the negative effects on the cells and allow to continuously measure the number of cells in a plate. Thus, it is worth developing measurement techniques that allow researchers to monitor the evolution of their experiments in real time, with a simple and autonomous assembly, as is the case with this invention.

RELEVANT DOCUMENTS

[I] I. Giaever et al., Use of Electric Fields to Monitor the Dynamical Aspect of Cell Behavior in Tissue Culture, IEEE Transaction on Biomedical Engineering, vol BME-33, No. 2, pp: 242-247, Feb. 1986 .

[2] S. M. Radke and E. C. Alocilja, Design and Fabrication of a Microimpedance Biosensor for Bacterial Detection, IEEE Sensor Journal, vol 4, n ° 4, pp: 434-440, Aug. 2004.

[3] D. A. Borkholder: Cell-Based Biosensors Using Microelectrodes, PhD Thesis, Stanford University. Nov. 1998.

[4] A. Yúfera et al., A Tissue Impedance Measurement Chip for Myocardial Ischemia Detection. IEEE transaction on Circuits and Systems: Part I. vol.52, no .: 12, pp: 2620-2628. Dec. 2005.

[5] I. Giaever, C. R. Keese, Cell substrate electrical impedance sensor with multiple electrode array, US 5, 187,096, Feb. 16, 1993.

[6] I. Giaever, C. R. Keese, Electrical wounding assay for cells in vitro, US 7,332,313 Feb. 19, 2008.

[7] I. Giaever, C. R. Keese, Real-time impedance assay to follow the invasive activities of metastatic cells in culture, US 7,399,631, July 15, 2008.

[8] B. Rubinsky, Y. Huang, Cell viability detection using electrical measurements, US 6,927,049 B2, Aug. 9, 2005.

[9] P. Owen Whiters, Method and apparatus for displaying bio-impedance at multiple frequencies, ES 2 1 18 133 T3.

[10] J. J. Ackmann, Complex bioelectric impedance measurement system for the frequency range from 5-1 MHz, Annals of Biomedical Engineering 21 (1993) 135-146.

[I I] R.Pallas, J.G. Webbster, Bioelectric impedance measurements using synchronous sampling, IEEE Transactions on Biomedical Engineering 40 (8) (1993) 824-829.

[12] M. Min, A. Kink, R. Land and T. Parve, Method and device for measurement of electrical bioimpedance, US 7,706,872 B2, Apr 27, 2010.

[13] X. Huang et al., Simulation of Microelectrode Impedance Changes Due to Cell Growth, IEEE Sensors Journal, vol.4, n ° 5, pp: 576-583. 2004 [14] P. Daza, A. Olmo, D. Cañete, and A. Yúfera. Monitoring Living Cell Assays with Bio-lmpedance Sensors. Sensors and Actuators B: Chemical. Elsevier, pp: 605-610: vol. 176. January. 2013. [15] G. Huertas et al. Oscillation-Based Test in Mixed-Signal Circuits (Frontiers in Electronic Testing). Springer 2006

[16] J. Gibson. Nonlinear Automatic Control. McGraw-Hill Japan, 1963.

[17] P. Fleischer, A. Ganesan and K. Laker, A Switched Capacitor Oscillator with Precision Amplitude Control and Guar-anteed Start-Up, IEEE Journal of Solid-State Circuits, Vol. SC-20, No. 2 , April 1985.

[18] N. Joye, et al., An Electrical Model of the Cell-Electrode Interface for High-density Microelectrode Arrays, IEEE EMBS, pp: 559-562. 2008

Description of the content of the figures

Figure 1. Bio-electronic measurement system. (a) Electric model of an electrode. (b) Electrical model of an electrode partially covered by cells (ff = A / A _Ci where A is the area of the electrode and Ac is the area of the electrode covered by cells), (c) Circuit for cell-electrode bioimpedance testing. Values for an electrode of size 50x50 μηι ² : R _s = 5.4 kQ, Ζ (ω) = C, || R _P , with C | = 0.37nF and R _p = 25ΜΩ. R _gap = 75kQ.

Figure 2. Block diagram of the proposed measurement system object of this invention.

Figure 3. Dependence of the oscillation parameters (f _osc , a _osc ) with the overlap of the cell area with the electrode defined by the filling factor, ff. The approximate sensitivity is Ο.Ι δΗζ / μιη ² for f _0S c, using a square microelectrode 50μιη on the side.

Figure 4. (a) Relationship between the amplitude of the secondary output and the frequency of oscillation, (b) Dependence of the amplitude of the secondary output (A _0SC 2) with the cell-electrode overlap defined with ff. The approximate sensitivity is 0.2mV ^ m ² for the secondary output, V _out2 , using a 50μιη side electrode. Figure 5. Schematic example of the filter used in the invention. Figure 6. Schematic example of the high pass filter plus comparator used in the invention. Figure 7. Schematic example of the AGC used in the invention.

Figure 8. Block diagram of the overall system used in the invention. Figure 9. Multiplexing diagram.

Description of the invention

The impedance of a two electrode system (Figure 1 (a)) has been widely estimated [3, 13, 14]. Particularly, in this work, a micro-electrode that can be totally or partially covered by cells in the culture has been considered. The fill factor (ff) represents the amount of cell electrode area (A) covered. Varying from ff = 0, if the presence of any cell is not detected, up to ff = 1, with the entire area occupied by cells.

Our objective is, using the electric model for bioimpedance (Figure 1 (b)) previously reported [3, 13, 14], to obtain the area of overlap of the cells with the electrode (A _c ) using the measurements made with the circuitry proposal.

The magnitude and phase response for _ce ii _-e iectrode (s) can be obtained from the assembly of Figure 1 (c). The R _in resistor allows the current flowing through the electrode-cell system to be maintained at adequate signal levels (1-20μΑ to protect the cell and 10-50mV for restrictions on electrode modeling [3]).

The system object of the present invention carries out the measurements of the bio-electronic system (Z _ce n-eiectrode), which we will now call CCUT (Cell-Culture Under Test, that is, cell culture under test). This bio-system is transformed into a robust oscillator, adding some components (Figure 2). To force oscillations, a positive feedback loop has been implemented. The technique requires accurately predicting the oscillation parameters (frequency and amplitude), either analytically or through simulations [15-17]. It is also necessary to avoid the dependence of these parameters with the saturation characteristic of the active elements, as in the case of common oscillators One solution to this problem is to use a nonlinear element (a simple comparator) by closing a feedback loop to ensure self-maintained oscillations [15-17]. This non-linear element also provides precise control of the amplitude of the oscillations. On the other hand, we need to ensure that this system complies with a (relatively) simple set of oscillation conditions that can give information about the oscillation parameters. A simple way to achieve this objective is to use a bandpass filter in the loop, as we propose in the general circuit block in Figure 2.

For simplicity, we consider the case of a second-order bandpass filter and a comparator with saturation levels ± V _ref . This closed loop system verifies the required premises: it is autonomous, the nonlinearity is separable and independent of the frequency and the linear transfer function contains enough low pass filtering to neglect high order harmonics at the comparator output. In this way, the equations that govern the oscillation conditions can be easily handled.

By properly choosing the bandpass filter, the closed loop system of Figure 2 can be forced to oscillate and its first order description equation (DF) [15], N (a) + 1 / H (s) = 0 , has an oscillatory solution (or _OS c, to _osc), with or _osc the oscillation frequency and the oscillation amplitude _osc. The DF function will in this case be N (a), while H (s) is the transfer function of the open loop system.

Mathematically, the characteristic equation is:

(1) where, N (a) is the DF of the comparator and H (s) the modified system.

The general BP function will be given by,

where ω ₀ *, Q * and ki ^{* are} the BPF parameters. The transfer function V _0U tA ¡ _n of the bio-electronic system of Figure 1 (c) we will call it, H _z (s),

Where the constant parameters (ω ₀ , Q and k ₀ , ki, k ₂ ) are directly related to the electrode size, technology and biological material (ff). The expression of the total function will be given by,

To force the oscillations, a pair of complex conjugate poles of the global system must be located on the imaginary axes. The way to determine the oscillation conditions (gain, frequency and amplitude) is to solve equation (1).

This is equivalent to finding a solution to the set of equations.

Being the coefficients given by the equations,

_{B =} ^ _{+ + N (fl∞c) fc fc2} I

Q

(6)

There is a swing solution. The main oscillation parameters are a function of the area of the cell culture occupied. This dependence is shown in Figure 3, for the frequency and amplitude of the oscillations. The output of the biological filter (the input to the comparator in Figure 2) is approximately sinusoidal due to the characteristic band pass of the overall structure. This fact allows us to use the linear approximation established by the method of the function of descriptive [15-17] for the linear treatment of the non-linear element. The transfer function of the comparator, according to this, will be given by:

N ( _MC ) = - ^

(7)

Where, as noted, V _ref is the reference voltage of the comparator (whose value can be swept to obtain appropriate signal levels) and _osc is the amplitude of the oscillations.

Note that for our example (Figure 3), the oscillation frequency increases monotonously in the range [7560, 7920] Hz (0.16Ηζ / μιη ² of the electrode area occupied by cells) and the oscillation amplitudes [0, 40] mV , as the overlap of the cell area in the electrode grows. Because the output signal level V _out (see Figure 2) is very small due to the restrictions of the electrode model, we consider the secondary signal (V _out2 in Figure 2) as a potential output (whose value is related to the frequency oscillation), achieving an improvement in the dynamic range.

If f (t) is the square output of the comparator, it can be expressed in terms of its Fourier series development:

4 1 1

f (t) = - V _ref [sen (ca _osc t) + - sen (3 « _OIC + - sen (5w _osc t) + ...]

π 3 5

On the other hand, the gain of the bandpass filter with respect to the oscillation frequency is:

Result the following expression of the amplitude of the secondary output:

We show in Figure 4 (a), how is this dependence and in Figure 4 (b) how the dynamic range is improved if we use the secondary output.

The reasons that make this invention attractive are: the concept is very simple, avoids the need for complex and expensive equipment for the generation of stimuli (in fact, it does not require any external input signal) and for the interpretation of the response (the measures to be taken are relatively simple [fifteen]). These characteristics are what open the door to extend the concept to real-time monitoring as we propose here.

Embodiment of the invention

The schematic of the main system (oscillator for measurements) was previously shown in Figure 2. It is composed, as we said, of three fundamental blocks: a bandpass filter (BP), the bioimpedance block and a comparator. It also requires some kind of start-up mechanism. Band pass filter:

An in-depth theoretical study of our "biological oscillator" (when considering the particular case of a gold micro-electrode, square, 50 μιη on the side, which may be partially or totally covered with cells in the culture) reveals that the Optimal frequency for the bandpass filter in order to achieve a proper oscillation mode is around 8kHz. However, in order to obtain greater flexibility, due to the nature of different kinds of electrodes, an adjustable filter has been implemented. A state variable filter was chosen because with this topology the resonance frequency, ω ₀ , and the quality factor, Q can be adjusted independently. Figure 5 shows the implementation of the complete filter.

By simulation the possible resistance values can be obtained and, with it, the values of ω ₀ (after selecting the values of R _m and C in Figure 5). These values have to be chosen to have a frequency range from 6.5kHz to 13kHz and an almost linear variation over the entire range. This is achieved by taking R _m = 10kQ and C = 1.2nF.

Comparator

Figure 6 shows the comparator used. It is composed of three blocks, the comparator itself, a high pass filter and an amplifier in a non-inverting configuration. The bioimpedance output signal is first filtered to reduce low frequency noise (mainly 50Hz of the mains) with a high pass filter Sallen-Key adjustable to 1 kHz and with variable Q to allow accurate selection of the phase deviation in the resonance frequency of the bandpass filter. Then, the signal is amplified x20 with a non-inverting configuration and passes to the comparator where thanks to the 6mV hysteresis noise cancellation is achieved. Also, its rapid response guarantees phase deviation low. In addition, due to its enable input we can keep the system in standby while changing the bioimpedance or while the system is in its normal mode. Automatic gain control (AGC):

This element (Figure 7) is included because of the advantage of guaranteeing a constant output voltage for different occupancy levels. The AGC allows a higher level of tension to be used through bioimpedance at a low level of occupancy and a lower level of tension at high levels of occupancy. This makes the system less sensitive to noise and solves the problems we face with signals of tens of millivolts obtained for the lowest occupancy levels. With the use of the AGC the information on the occupancy level resides in the voltage applied to the programmable gain amplifier, which we will call Vg from now on. This component is composed of three blocks: a variable gain amplifier, a precision inverter rectifier and a non-inverting amplifier (Figure 7).

Digital circuit:

The system object of the present invention is composed of electrodes, on which cell culture is carried out, the circuitry necessary to construct the oscillator and measure the impedance, a radio frequency transmitter circuit for wireless data transmission, and a micro-controller To process the data. The scheme of the system architecture is shown in Figure 8. The digital part is composed of a micro controller responsible for activating the bio-oscillator, choosing the multiplexed output (Figure 9) (if we particularize the system for measurement to through electrodes, 50μιη x 50μιη) and interact with users. The MCU needs to have an ADC of at least 6 bits to have a 1.56% accuracy when measuring Vg, timer with external trip and counter, multiplier module, serial port, usb port and enough general purpose inputs and outputs to connect buttons, a display, a temperature and humidity sensor, a piezoelectric buzzer and other devices or peripherals that are considered necessary or important. The MCU activates the bio-oscillator, chooses the desired cell by means of the multiplexer, measures the oscillation frequency of the system and the value of the voltage at Vg in the AGC module and calculates the value of the fill factor. This is shown on the display or the system can be configured to send it via its serial port to a wireless connection (Bluetooth for example) to a device that accepts that connection (mobile phone, computer ...) The system also measures and sends information on temperature and humidity, on battery status, elapsed time, as well as any other information deemed important. The wireless connection also allows the user to configure the device.

A piezoelectric buzzer acts as an alarm and alerts the user when situations occur that he himself has established. A USB connection can be used to access the data record of experiments performed and also to update the system firmware. The system is also provided with a display that allows you to display information about the fill factor and other data if the wireless connection cannot be used.

All this allows the wireless monitoring of the cell culture, without the need to extract the samples from the incubator or to interfere with the processes of the cell culture. Similarly, the radio frequency signal transmitter and receiver circuit allows the wireless programming of the bio-oscillator and the control and measurement parameters described above can be remotely established. This radio frequency signal transmitter and receiver circuit may be implemented so that the data is transmitted at a frequency of 2.4 Ghz or other available bands, and so that it is compatible with 802.11, 802.15 or similar standards. This monitoring system allows to obtain wirelessly the measurements that show the evolution of cell culture over time, without the need to perform a visual inspection of the culture, with the consequent saving of time and with the possibility of implementing automatic alarm signals before changes Unexpected Similarly, automation in obtaining information in digital form allows further processing of the data for a more advanced study of crop evolution.

Claims

Bioimpedance measurement system for real-time and wireless monitoring of cell cultures consisting essentially of a set of electrodes as bioimpedance sensors where the "biological circuit" is converted into an oscillator and electrical signals (oscillations) are used as Interpretation parameters of the state of the crop. The system is composed of:

a) micro-electrodes as bioimpedance sensors

b) that are part of a closed loop bioimpedance measuring oscillator circuit,

c) which is connected to a radio frequency signal transmission and reception circuit and

d) a micro-controller for monitoring and interpretation of data.

Bioimpedance measurement system for real-time and wireless monitoring of cell cultures according to claim 1, characterized in that the micro-electrodes are implemented through CMOS processes.

3. Bioimpedance measurement system for real-time and wireless monitoring of cell cultures according to claim 1, characterized by the use of a bandpass filter and a closed loop comparator for the construction of the oscillator that generates the signals of information of the state of the cell culture, an AGC block for the control of the appropriate signal levels and for the interpretation of the measurements, a high pass filter to eliminate noise signals that could invalidate the data, a micro-controller for the information monitoring and processing and temperature and humidity sensors for process control.

4. Bioimpedance measurement system for real-time and wireless monitoring of cell cultures according to claim 1, characterized in that the radio frequency signal transmitter and receiver circuit is compatible with 802.1 1, 802.15 or similar standards.