WO1995014962A1 - Potentiometric biosensors, control and applications thereof - Google Patents

Potentiometric biosensors, control and applications thereof Download PDF

Info

Publication number
WO1995014962A1
WO1995014962A1 PCT/IB1994/000408 IB9400408W WO9514962A1 WO 1995014962 A1 WO1995014962 A1 WO 1995014962A1 IB 9400408 W IB9400408 W IB 9400408W WO 9514962 A1 WO9514962 A1 WO 9514962A1
Authority
WO
WIPO (PCT)
Prior art keywords
transducer
sensor
biosensor
redox potential
variation
Prior art date
Application number
PCT/IB1994/000408
Other languages
French (fr)
Inventor
Marco Sartore
Manuela Adami
Claudio Nicolini
Antonio Fanigliulo
Original Assignee
Technobiochip
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from GB939324258A external-priority patent/GB9324258D0/en
Priority claimed from IT93RM000846A external-priority patent/IT1266466B1/en
Priority claimed from GB9411059A external-priority patent/GB9411059D0/en
Priority claimed from GB9411072A external-priority patent/GB9411072D0/en
Priority claimed from GB9414189A external-priority patent/GB9414189D0/en
Application filed by Technobiochip filed Critical Technobiochip
Priority to AU10753/95A priority Critical patent/AU1075395A/en
Priority to EP95901567A priority patent/EP0730760A1/en
Publication of WO1995014962A1 publication Critical patent/WO1995014962A1/en

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/53Immunoassay; Biospecific binding assay; Materials therefor
    • G01N33/543Immunoassay; Biospecific binding assay; Materials therefor with an insoluble carrier for immobilising immunochemicals
    • G01N33/54366Apparatus specially adapted for solid-phase testing
    • G01N33/54373Apparatus specially adapted for solid-phase testing involving physiochemical end-point determination, e.g. wave-guides, FETS, gratings
    • CCHEMISTRY; METALLURGY
    • C12BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12QMEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/001Enzyme electrodes
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D23/00Control of temperature
    • G05D23/19Control of temperature characterised by the use of electric means
    • G05D23/1919Control of temperature characterised by the use of electric means characterised by the type of controller
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D23/00Control of temperature
    • G05D23/19Control of temperature characterised by the use of electric means
    • G05D23/20Control of temperature characterised by the use of electric means with sensing elements having variation of electric or magnetic properties with change of temperature

Definitions

  • Temperature control is very important in many systems, mainly when a temperature sensitive device is being utilized.
  • a silicon-based potentiometric sensor able to detect the H + ions concentration (i.e. the pH) of an electrolyte in contact with it.
  • a cover slip with attached living cells is put closely in front of the sensitive surface of the silicon sensor, in order to obtain a micro-volume nearby the sensor; cells are in contact with fresh medium, by means of a flow mechanism; when the flow is stopped, or some drug is added to the medium, cells acidificate the micro-environment more quickly; monitoring of the acidification rate allows us to understand the "status" of the cells under different conditions, and to predict the effects of drugs on the considered cell system.
  • the senor used is silicon-based, hence it is extremely sensitive to temperature variations; if operating in a non-controlled system, the output signal cannot be unequivocally related to any physical event.
  • extracellular acidification is strongly affected by the temperature at which cells are kept; usually 37°C is a suitable value for optimal cells conditions.
  • Temperature control is needed in very precise measurements. If one just wants to perform basic experiments on the device, acquisitions at room temperature in a normal laboratory are more than satisfactory, and no temperature control system is needed.
  • temperature control is essential.
  • usual temperature control is obtained by means of setups based on a warm fluid flowing around a considered region.
  • a (big) thermostat is kept outside the measuring chamber, and even outside the measuring desk; this system produces a warm fluid circulation in a tube coil, placed just around the measuring chamber.
  • a temperature control device comprises a heating/cooling device in the form of a Peltier cell; a temperature sensor for sensing the temperature in the vicinity of the cell; and control means responsive to the sensed temperature for controlling the heating and cooling activity of the Peltier cell so as to maintain the temperature in the vicinity of the cell at or near a predetermined value.
  • This temperature control device or thermostat is effective, cheap and easy to use. It can be applied in a variety of situations, when a precision temperature control is needed in a relatively small environment.
  • a biosensor including a temperature control device for controlling the temperature in a sample region, the device comprising a heating/cooling device; a temperature sensor for sensing the temperature in the vicinity of the device and control means responsive to the sensed temperature for controlling the heating and cooling activity of the heating/cooling device so as to maintain the temperature in the vicinity of the device at or near a predetermined value.
  • the temperature control device used with this biosensor could take a variety of forms, preferably the heating/cooling device is in the form of a Peltier cell. It has been found that in applications where the temperature of a small environment must be controlled, the use of Peltier elements guarantees high efficiency and uniform temperature distribution. Furthermore, a Peltier-based device enables temperatures to be controlled at target or predetermined values higher or lower than room temperature. Furthermore, the Peltier cell can easily and rapidly be used as a heating or cooling element. Finally, this system is very compact and the driving electronics can be installed in a small box.
  • a number of biosensors are known to employ light-addressable potentiometric sensors (LAPS).
  • LAPS light-addressable potentiometric sensors
  • the principle of operation of a light-addressable potentiometric sensor is described by Hafeman et al, Science (1988) 240:1182 and by Sartore et al, Biosensors & Bioelectronics (1992) 7:57-64.
  • Use of such sensors has allowed the design of biosensors that are capable of measuring either pH variations (if the sensor's surface is coated with a H + sensitive insulator, such as Si 3 N 4 or Ta 2 O 5 ) or redox potential variations (if metal spots are evaporated onto the sensor's surface).
  • Light-addressable potentiometric sensors have also been used to monitor extracellular acidification in micro environments, again through pH variation. Many important metabolic processes of the cells (namely the catabolism of sugars, amino acids and fatty acids) produce H + ions, which are excreted through the cytoplasmic membrane and out of the cell, thereby causing an extracellular pH variation.
  • the use of light-addressable potentiometric sensors to measure this pH variation is described by Owicki et al, Biosensors & Bioelectronics, (1992) 7:255-272; Owicki et al, Proc. Natl. Acad. Sci. (1990) 87:4007-4011; and Wada et al, Journal of Cell Biology, (1991) 115:A2455.
  • the cell medium usually contains a buffer whose capacity masks any pH change due to extracellular acidification, which therefore cannot be measured.
  • the present invention relates to a biosensor, and in particular to a biosensor for measuring pH variations and/or redox potential variations in enzymic reactions, either taking place within the cell and causing extracellular variations, or in isolation.
  • a method of measuring a pH variation or a redox potential variation in an enzymic reaction that generates ions that cause, or whose generation causes, such a variation comprises monitoring the reaction over a period of time using a light-addressable potentiometric sensor that generates a current on the binding of the respective ions thereto; converting current measured over that time to voltage; if desired, calculating the pH variation or the redox potential variation, or the number of the respective ions generated, as a function of the voltage; and comparing the voltage, pH variation or redox potential variation, or number of ions with a precalibrated standard.
  • the method of the invention allows the monitoring of range of enzymic reactions through a variation in pH and/or in redox potential occurring in those reactions.
  • the method of the invention can be applied to determine enzyme and substrate concentrations to low orders of magnitude, for example in the 10 -10 M range.
  • a sensor that is sensitive to redox potential variations, the problem related to the measurement of small pH variations in buffered solutions is overcome.
  • the measured quantity is a surface potential, and therefore buffers, even with high buffer capacity, do not affect the sensitivity of the method.
  • a biosensor can be programmed to carry out either or both of the above methods, and to display the results thereof in a suitable manner. This allows the user to carry out a wide variety of experiments on the same or different biological samples in a simple and efficient manner.
  • a biosensor suitable for carrying out the method of the invention comprises a reaction chamber in which the reaction is to proceed and into which at least one sensor can be immersed.
  • the reaction chamber can be separate from at least one measuring chamber, each having a sensor associated therewith.
  • the biosensor further comprises specific electronic cards and related software programs to acquire the necessary data quickly, and present it in a user- friendly manner.
  • the biosensor is preferably completely automated and driven by a personal computer.
  • the electronics is designed in order to obtain a good general signal-to-noise ratio, hence ensuring reproducible results.
  • the software package typically consists of low-level programs (written, for example, both in C and Assembler), to interact with the data acquisition and controlling cards, and of high-level programs (for example written in C), to obtain, display, save and print data files, and to present a suitable user interface.
  • a very important peculiarity of the biosensor of the invention is the ability to connect different sensors or a plurality of measuring chambers each associated with a different sensor, to the same device; this allows the user to perform a wide variety of experiments, utilizing the same or different biological elements, by simply changing the measuring chamber under consideration and selecting the appropriate acquisition software. This can be done by simply substituting electrical connections and does not require physical movement of the measuring chamber per se.
  • sensors examples include a sensor sensitive to pH variation by the binding thereto of H + ions, and a sensor sensitive to redox potential variation by the binding thereto of ions generated in a redox reaction.
  • An example of the latter kind of sensor is one which has had gold metal evaporated onto its surface.
  • Sensors that are sensitive to inorganic ions are known, and can also be used in the biosensor of the invention.
  • sensors that are sensitive to more than one of the above types of ion can be used.
  • the biosensor of the invention utilizes the fast information recovery from the sensor output signal, and can acquire quantitative data at fraction of a second intervals over period of time; it can therefore be utilized to monitor fast acidification or redox processes.
  • PAB Patentiometric Alternating Biosensor
  • LAPS Light Addressable Potentiometric Sensor, Hafeman et al, 1988
  • YADH Alcohol Dehydrogenase from Yeast, Adami et al, 1994b
  • GST Glutathione-S-Transferase, Antolini et al, 1994
  • a potentiometric sensor based on a silicon chip, able to detect redox potential changes in solution is produced and some of its possible applications are investigated.
  • the redox potential of a solution in contact with the surface of a metal layer deposited on the chip affects the amplitude of a photocurrent signal generated in the silicon by means of a modulated light source.
  • HRP horseradish peroxidase
  • a biosensor capable of 2D acquisition and/or multisensing; the device is potentiometric and is based on a silicon transducer with signal generation caused by light excitation.
  • the system can recover the local spatial information of pH, redox and other significant quantities on a sensing area of about 1cm 2 , which is usually contained in the particular measuring flow chamber specifically designed for the given biosensing application.
  • the bidimensional PAB system allows a fast potentiometric measurement in the 2D space of the transducer surface. Spatial information recovery is achieved by means of local light stimulation of the silicon transducer. In this way the surface potential of the system is locally measured; the surface charge, giving rise to the potential, can be originated in different ways (Delia Ciana et al, 1991), allowing multisensing operations onto a single chip; for instance, the transducer surface can be partially covered by Si 3 N 4/ for pH detection, or by gold, for redox measurements, or by different chemically sensitive sites, such as LaF 3 for F- detection; moreover, specially functionalized sites can be created onto the sensing surface, for instance by chemical interactions between gold spots and specific compounds (i.e. thiols).
  • the system produces a 2D map of the solution pH under investigation; instead, when additional sensitive sites are used, the system can be considered both bidimensional and multisensitive; in other words the local surface potential can still be recovered preserving the spatial information, so obtaining a 2D multisensor.
  • PAB One of the most relevant characteristics of PAB is the versatility of the system.
  • a fifth aspect of the invention provides for the application of PAB system for the investigation of the effects of antineoplastic drugs on two stabilized cell lines (normal mouse fibroblasts 3T6 and transformed HEPG2 hepatomas) and on primary cultures (normal rat hepatocytes); this work shows the feasibility to utilize the PAB biosensor as an innovative test with respect to conventional ones for the estimation of drug efficacy and toxicity.
  • liver is the major site for the uptake of drugs and chemicals, converting them to pharmacologically inactive, active or even toxic metabolites.
  • Biotransformation of xenobiotics and other specific liver functions, mostly performed by parenchimal liver cells (hepatocytes), are difficult to study in the whole organism, because of the influence of other organs, tissues and exogenous/endogenous factors; consequently, in vitro hepatocytes represent a valuable tool in pharmoco-toxicology, since it was demonstrated that they preserve, when in culture, the functional drug metabolising enzymes in culture (Engelmann et al 1985; Dich et al, 1988; Martelli et al, 1988).
  • a possible application of the PAB system as an immunosensor can imply the usage of a modified Langmuir-Blodgett (LB) technique for the production of antibody monolayer.
  • LB Langmuir-Blodgett
  • urease an enzyme
  • the immobilized monolayer was characterised by means of ellipsometric and nanogravimetric techniques in order to evaluate the main physical parameters such as thickness and surface density. Together with the enzymatic activity we also investigated the operative lifetime of the monolayer. The performance of the PAB system utilizing the thin film technology was analyzed.
  • the aim of our work is to study the enzymatic activity of the urease immobilized onto a support (glass or silicon nitride (Si 3 N 4 )) inserted into a microvolume reaction chamber in order to evaluate the application of the LB technology to immunosensors based on PAB.
  • the enzyme used in this experiment was urease, which catalyzes the hydrolysis of the urea.
  • FIG. 1 Schematic view of a typical measuring flow chamber utilizing the thermostat here described.
  • An aluminum plate is in contact with the heating element, a Peltier cell; the temperature sensor is placed just under a micro-volume area (the region to thermostat). Liquid flows in and out by means of two channels, and the measuring area is just nearby the sensitive layer (Si 3 N 4 ) of a silicon-based sensor placed in front of a cover slip.
  • FIG. 2 Control loop of the thermostat. User can set a target temperature T d , continuously compared with the measured temperature to generate an error signal e(t).
  • Non linear control is accomplished by means of a "Bang-Bang" system which drives the Peltier unit either with full power heating or with full power cooling.
  • a temperature sensor closes the loop, and gives the actual temperature value.
  • Figure 3 Theoretical temperature signal vs. time, as derived from equations 7, 10 and 11 (see text).
  • the figure also contains a plot of the input square wave u(t) and of h(t), as expressed in the mentioned formulas. All values are in arbitrary units, and normalized between +1 and -1.
  • Input sequence wave [u(t)], output signal [y(t)] and its deviation from input [h(t)] is shown.
  • FIGS 4A and 4B Hardware circuitry of the thermostat. Terminals referred to as V p should be connected to an external constant-current power supply, regulated for an output voltage as specified in the particular Peltier cell data sheets. The connection to a milli-voltmeter can be also used to interface the circuit to an A/D converter, in order to monitor temperature changes in time by a computer.
  • Figure 5 Computer acquired temperature signal as a function of time.
  • the acquisition interval corresponds to a complete semi-period of the (square-wave) control signal.
  • Target temperature was set to +37°C, and the system kept this value in about 3 minutes after system switch on.
  • a fit with the theoretical expression of y(t) is also shown; data and fitted curve are in considerably good agreement.
  • the control signal frequency was about 5.5 KHz and the resulting value of ⁇ was 22.1 ⁇ s. Variation of measured temperature vs. time for a Peltier cell, and fitting with theoretical expression is shown.
  • Figure 7 Relationship between produced H + moles and pH variations: a region is shown with a direct proportionality between n and ⁇ pH, where a fitting straight line is also shown.
  • Figure 8 The whole system is divided into two main sections: a reaction chamber and a measuring chamber (working volume ⁇ 10 ⁇ l), both kept at 25°C. Conditioning and acquisition electronics are also shown.
  • Figure 9 A typical experiment for the determination of enzyme concentration. Three different zones are visible: a first constant region, a small step and an exponential-like curve; they all are due to the particular configuration of the measuring system, as described in the text.
  • Figure 10 Output signal slope vs. YADH concentration; each data is calculated on the respective third (exponential) portion of the data presented in Figure 9, and the tangent is calculated at the first valid point.
  • FIG 11 Sensor output signal variation during the enzymatic reaction: the figure shows a typical result.
  • Figure 12 Output signal variation vs. EtOH concentrations; each point in the plot is measured as shown in Figure 11.
  • FIG. 13 schematic representation of the measuring system.
  • Cell medium flow is obtained by means of a peristaltic pump, which is connected to a measuring chamber, containing the transducer and a coverslip with cells; the cell medium flows in a very small region between transducer and coverslip.
  • the figure also shows a control electronic block, which provides the system with all the necessary driving signals.
  • Figure 14 output signal during ON-OFF acquisitions.
  • the pump is ON the pH in the measuring chamber is maintained constant, while switching OFF the pump yields to a net pH variation, due to the presence of acidic products in the micro environment.
  • Figure 15 a similar signal as in Figure 9 is here shown, relative to a single OFF time interval.
  • the signal is approximately linear in the considered region, and can be fitted with a straight line, whose slope represents the acidification rate.
  • Figure 17 two images of the 3T6 cell monolayer as usually appears before (A) and after (B) an acquisition session with the presented system. Images have been acquired by a CCD camera with an optical microscope interfaced with a Personal Computer.
  • Fig. 18 Schematic representation of the PAB system.
  • a central control unit drives and controls the solution flowing system and the measuring chambers containing the transducer; the control unit also dislogs with a Personal Computer, utilized to bias the device and acquire and process the experimental data, via appropriate AD/DA interfaces.
  • Fig. 19 Block diagram of the main electronics.
  • the LAPS is biased via a potentiostat and the current signal is converted to voltage, filtered and synchronously demodulated, before being digitized and computer acquired.
  • the figure also shows a digital output interface suitable to drive the flow circuit through an interface; in addition a temperature control scheme is provided.
  • Fig. 20 Schematic circuit of the synchronous demodulation technique utilized, me input signal, after a current to voltage conversion and a filtering, is multiplied by the reference signal coming from the LED driver; the output of these blocks are then filtered and amplified before being fed into the circuit for calculation of the modulus.
  • Fig. 21 Software organization of the PAB system.
  • a shell offers a suitable user interface, and allows parameters input, data display, saving and printing; the usual procedure consists in the acquisition of a complete characteristic curve (current versus bias potential) and in the consequent determination of the inflection point; our software then bias the device and continuously acquires the digitized signals.
  • Fig. 22 Schematic of a very simple measuring chamber; a well in the upper part confines the measuring solution in a region delimited by an 0-ring, directly pressed against the LAPS surface. Counter and Reference electrodes are dipped in the measuring solution from the top, while a LED confined in the lower part just near the back of the chip provides the light excitation.
  • Fig. 23 Measuring flow chamber suitable for enzymic experiments.
  • the lower side is quite similar to that of Fig. 22; the upper part contains an inlet channel towards a well able to contain a membrane; usually some biological specie is entrapped in the membrane at this level; the reaction products flow then in a small volume region near the sensitive surface, through a channel as tiny as possible, in order to reduce the delay time of the sensor response.
  • the solution flows away through an outlet channel, contacting the reference electrode.
  • Part A shows the immunoreaction scheme, where the antibody is fixed onto a substrate and the antigen with. enzyme is injected in the solution to compete with the free antigen; the binding of enzyme-linked molecules is detectable after the production of electrons.
  • Part B shows a typical acquisition curve (I/V characteristic) obtained utilizing Horse Radish Peroxidase as the labelling enzyme, 2,4 dichlorophenossyacetic acid (2,4-D) as antigen and a monoclonal antibody against the pesticide 2,4-D; the right-band curve is before binding, the leftmost after binding.
  • Fig. 25 Measuring flow chamber for cellular acidification on experiments.
  • the counter electrode is inserted inside the inlet channel, while the reference electrode is in the outlet.
  • the LAPS is separated from the cells immobilized onto a glass coverslip by a teflon spacer, whose thickness defines the micro volume of the measuring chamber.
  • Thermal control is achieved by a Peltier cell connected to an appropriate circuitry and fixed on the top of the chamber; heating is obtained near the cells by a metal cylinder (usually aluminum) .
  • the right part of the figure shows the top view of the gasket, utilized to prevent medium leakage, of the coverslip where cells are grown and of the teflon spacer, giving a indication of the flow circuit inside the "sandwich".
  • Fig. 26 Two-chips measuring chamber suitable for differential measurements.
  • the flowing solution appears at the first chip, then passes through a removable reaction chamber, before reaching a second chip.
  • both chip should be affected by the same noise, as the particular design ensures for both the same environmental conditions, hence a differential acquisition scheme detects only the true reaction signal.
  • Fig. 27 Schematic representations of the PAB system in a redox configuration.
  • the measuring chamber is shown in detail. Biasing of the transducer is achieved by a potentiostat , while signal measurements are performed by a synchronous demodulation technique.
  • Fig. 28 Different curves of chips with and without metal layer to a solution containing ImM Fe(II), ImM Fe(III) and lOOmM sodium citrate buffer pH 6.0
  • Fig. 29a Biphasic response curves of a chip with a gold layer covering part of the sensing surface; buffered solution containing the redox pair in different ratios were used. From left to right the curves correspond to solutions of Fe(II): Fe(III) r)tios ranging from 1:100 to 100:1.
  • Fig. 29b Biphasic response curves of a chip with partial gold covering of its exposed surface to solutions at two different pH values: at left pH 6 ; at right pH 9.
  • Fig. 30 Bias potential values corresponding to the inflection points at different redox pair ratios plotted versus the logarithm of the ratios themselves for chips with gold and chromium layer.
  • Fig. 31 Variation of the logarithm of the redox pair concentration ratio when Fe(II) is oxidized to Fe(III).
  • the experimental conditions are: reaction chamber volume 10l, 0.8mM initial Fe(II) concentration. 0.4mM initial Fe(III) concentration.
  • the curve can be approximated to a straight line in the y-interval between +0.3 and -0.3.
  • Fig. 32 Calibration curve for the HRP enzyme in solution.
  • the enzymatic activity units are here defined as moles of TMB oxidized per minute, corresponding, in turn, to half of the moles of Fe(II) converted to Fe(III).
  • Fig. 33 Calibration curve for the HRP enzyme immobilized on activated membrane.
  • Fig. 34 Monitoring of Alcohol Dehydrogenase activity in redox configuration by the use of Diaphorase as auxiliary enzyme.
  • Fig. 35 Block diagram of the PAB system.
  • the transducer is biased by a potentiostat, and the alternating current photogenerated by LEDs is converter into voltage and amplified; after filtering, the RMS values of the signal are extracted by synchronous demodulation, and acquired by a DAC card into a Personal Computer.
  • a temperature control system based on a Peltier cell is provided, in order to stabilize the environmental conditions of the measurement.
  • Fig. 36 Schematic structure of the transducer-optical fibres system. The distance of adjacent light spots is related to the wafer thickness, because the main process in the signal generation is the diffusion of charge carries.
  • Part A represents a simple equivalent circuit of the three-electrode system; Zrc is the impedance between counter and reference electrodes, and Zwr is that between working and reference electrodes; changes in these impedances cause signal amplitude variations, as depicted in Part B.
  • the normalization procedure described in the text avoids the eventual artefacts due to the monitoring of such impedance variations.
  • Fig. 38 Schematic representation of a measuring flow chamber of the 2D PAB system. Part A is connected to B after the eventual insertion of some biological element (i.e. membrane) into the small volume region F. Medium flows from inlet C, passes through the sensing spots H and G, and exits through outlet E, after filing the electrodes region D. An array of optical fibres I is fixed at a given distance from the chip backside by the holder L.
  • some biological element i.e. membrane
  • Fig. 39 Hardware circuit for the optical fibres array selection.
  • the driving signal (usually a sine- or a square- wave) is fed to a single selected LED at a time, passing through the operational amplifier and the switching transistor.
  • the remaining blocks are used to digitally connect a desired LED to the drive electronics; in particular two 4-bit words are the row-column addressed for the CD 4028 multiplexers, which activate corresponding digital switches. (Drawing missing).
  • Fig. 40 Schematic representation of the transducer sensing surface utilized in the presented experiments. Two gold spots have been evaporated onto the silicon nitride surface, and four light spots are created in order to obtain two redox sensitive and two pH sensitive sites.
  • Fig. 41 Typical acquisition results after the insertion of a Potassium Ferrocianide-Ferricianide solution into the measuring chamber. Data corresponding to coordinates (1,1) and (2,2) are relative to redox, while those corresponding to coordinates (1,2) and (2,1) refer to pH. (Drawing missing) .
  • Fig. 42 Series of measurements as the one shown in Fig. 41. Data from A to E show redox pair concentration variations range from 1:100 to 100;1 at pH 7; data from F to L range back from 100:1 to 1:100 at pH 9. It is evident that variations in the redox concentrations selectively affect the only redox-specific spots, while the pH variation is specifically sensed by the silicon nitride regions. (Drawing missing).
  • Fig. 43 shows the dose/effect trend of ara-C on S-phase 3T6 cells acquired with PAB system
  • Fig. 44 shows a comparison between PAB results and results of the Trypan Blue Test
  • Fig. 45 shows rat hepatocyte acidification curves acquired with PAB.
  • Fig. 46 shows monolayer surface density v. surface pressure obtained by means of nanogravimetry
  • Fig. 47 shows pH variation due to urease activity monitored by means of PAB system: the enzyme was immobilized on silanized glass: the regression line helps in evaluating the initial alcalinization rate of the reaction; and
  • Fig. 48 shows activity of urease monolayer immobilized of silanized glass during repeated assays (Deposition Pressure 20mN/m) .
  • the system utilizes a flow chamber 1, as depicted in Figure 1; a silicon chip 2 is put in front of a cover slip 3, and the medium flows in between.
  • the cover slip (containing cells) occupies a well 4 in an aluminium plate 5, which is in contact with the heating element (the Peltier cell) 6.
  • a temperature sensor 7 (AD 590, by Analog Devices) is placed just under the cover slip 3.
  • the other side of the Peltier cell 6 is connected to a convection heat sink 8.
  • Another typical control strategy utilizes a PID (Proportional Integral Derivative) algorithm; in the latter case the control signal is generated by adding three terms, one proportional, the second related to the first derivative and the last related to the integral of the error signal.
  • PID Proportional Integral Derivative
  • the Bang-Bang control technique is much easier and immediate than the PID one, as it is not necessary to compute (either by software or by hardware) the proportionality factors at each control step.
  • an hardware solution for the control system (as the one here presented) only requires a comparator-based stage in the circuit.
  • the performance here requested (thermal stability within 0.1°C around the desired temperature, and the latter ranging between +20°C and +45 °C) can be fully obtained without a PID control.
  • the desired temperature T d can be manually set; it, is continuously subtracted from the actual measured temperature y(t) by a subtracter 10, hence generating an error signal e(t) used to set the desired temperature, while the two lateral 500 ⁇ variable resistors are trimmers used to set the minimum and maximum allowed temperature; in order to set these two values one should turn the potentiometer first completely to the left, then adjusting the left trimmer until the desired minimum value appears on the connected digital display; then one should turn the potentiometer completely to the right, and adjust the right trimmer for the maximum value.
  • the second signal to the comparator comes from the Operational Amplifier CA313024, used as a driver for the temperature sensor (AD 590).
  • the 10K ⁇ trimmer on the right allows to adjust the driver in order to sense the correct temperature value; to set the temperature measuring section it is necessary to utilize at least a 0.1°C precision thermometer and to perform several steps of temperature monitoring near the two desired extremes with both methods, then adjusting the trimmer in order to read the true value on the milli-voltmeter.
  • a switch 25 connects either the pre-set desired temperature signal, or the actual measured temperature signal to a milli-voltmeter (digital display); usually one sets the switch to the T d position, selects the desired temperature, and then switch to the monitoring position in order to read the actual temperature while thermostating process is on.
  • the signal to the display can be directly connected (with no additional circuitry) to an A/D converter input, if computer monitoring of the temperature is needed.
  • the error e(t) is greater than 0 if the actual temperature is lower than the desired one, lower in the opposite case: e(t)>0 if T d >y(t)
  • the signal e(t) is used as input to a non linear stage 11, which produces a control signal u(t), which, power-amplified 12, corresponds to the driving signal of the Peltier cell 6.
  • a control signal u(t) When the control signal u(t) is low, one side of the Peltier element 6 is heating, the opposite is cooling (and vice-versa when the control signal is high); fast transitions of the control signal allows the temperature control around the target value.
  • the feedback system is completed by a temperature measuring block 7. Temperature sensing is accomplished by using a temperature probe connected to an appropriate driving system, whose output is y(t).
  • a typical set up as the one here described takes some time to reach a stable state corresponding to the desired temperature; the elapsed interval from the system power-up to the stable situation depends upon several factors: the selected current in the Peltier power supplying circuit, the temperature probe location with respect to the heating element, the type and shape of the heat sink, the system to thermostat.
  • the driving control signal is a square wave with a 50% duty cycle.
  • the first term depends on time through the exponential factor exp(-t/ ⁇ ), and represents the initial transient response following the system power-on; this term expires with the time constant ⁇ .
  • the second term represents the steady-state portion of the response to the square wave, as its dependence on t is expressed by exponential terms like exp[-(t-na)/ ⁇ ], related to the number of sweeps before the actual one.
  • the circuit implementing the Bang-Bang control here described is quite simple ( Figure 4) .
  • the heating-cooling technique is achieved by feeding the Peltier element 6 with a given current flowing in one sense or in the opposite one; this can be easily obtained by applying a fixed voltage V p to the unit in order to heat, and the opposite -V p in order to cool.
  • the circuitry of the control system is depicted in Figure 4.
  • the Peltier element 6 connected to the external power supply generating V p by means of power transistors pairs (2N3055) 13 used as switching elements; if the right pair or the left pair is on, current flows into the Peltier cells in one sense or in the opposite one, yielding the heating or cooling of a given cell side.
  • transistors 17,18 are used to drive red and green LEDs 19,20 indicating to the user when the system is heating and when it is cooling.
  • the non-linear control signal is generated by an AD 845 (Analog Devices) Operational Amplifier 21, here utilized as a zero-hysteresis comparator.
  • AD 845 Analog Devices Operational Amplifier 21
  • This integrated circuit is a suitable comparator because its slew-rate is quite high (100 V/ ⁇ s).
  • the operational amplifier 21 continuously compares the two signals fed into its inverting and non-inverting input pins, hence generating a +V sat /-V sat output if the actual temperature is lower/higher than the desired one.
  • the central variable resistors 23 is a multi-turn potentiometer
  • the described thermostat has been realized and tested with a 3x3 cm, 65 W, Peltier cell.
  • the system was supplied by a commercial +.12 V power supply; in addition a selectable constant-current power supply was connected to the terminals V p in Figure 4, i.e. to the Peltier element through the transistors pairs.
  • this external power supply was 8.5 V, and 1.5 A.
  • the measuring chamber to thermostat is a plexiglass-aluminum cylinder of about 5 cm base diameter, and 5 cm high.
  • the Peltier cell 6 as connected to the aluminum part 5 depicted in Figure 1, and the system was tested at room temperature for target temperatures ranging between +20°C and +45°C. After switching on the thermostat, the output of the comparator is at high/low level depending upon the desired temperature as compared to the actual one; then a heating/cooling process starts, and no commutation arises until the temperature becomes higher/lower than the desired one. At this point a transition in the control system causes an inverse cooling/heating process with respect to the prior one (the heated side is now cooled and vice versa).
  • the system reaches and keeps the target temperature.
  • the actual temperature has been computer- acquired, after the initial transient expired and the target temperature was kept; values have been fit with the theoretical expression of y(t) presented before.
  • Figure 5 shows the result; for a sake of clarity, the figure corresponds exactly to a low-level semi-period of the control signal square wave when the target temperature was set at 37 °C.
  • the total semi-period is 90 ⁇ s, which implies a commutation frequency of the Bang-Bang control system of about 5.5 KHz, as can also be easily observed by connecting an oscilloscope to the comparator output; the behaviour of the actual temperature vs. time has an exponential nature, as predicted by the expression of y(t). Moreover, fitting with y(t) gives an estimation of , which results to be about 22 ⁇ s.
  • the described thermostat can be easily applied in different research areas, when a precision temperature control is needed in a small environment; the application here reported makes use of a flow chamber, and the "sensitive region" to control is about 8 ⁇ 8 cm; by using bigger Peltier elements it is possible to control, by means of the very same circuitry, a bigger area and a different system.
  • PAB Potentiometric Alternating Biosensor
  • the enzymic reaction under consideration is that of YADH on NAD + and C 2 H 5 OH.
  • the transducer output signal is a photocurrent, converted to a voltage ⁇ by a current to voltage converter with a usual gain of 10 6 ; this voltage is directly proportional to the actual pH of the solution in contact with the sensing area, see Figure 6.
  • the sensitivity ⁇ of the transducer corresponds to the bias voltage shift to apply to the system in order to get the same ⁇ after a pH unit step; to this sensitivity directly corresponds a y-axis sensitivity, which also depends upon the maximum curve slope ⁇ ; a variation ⁇ of the output signal can be expressed in terms of ⁇ pH, as follows:
  • the first portion of the curve can be approximated very well with a straight line, by means of a least squares fitting; the fitting line is also shown in Figure 7.
  • the portion utilized for the fitting corresponds to a (pH-pK a ) ranging from +0.3 to -0.5.
  • the total pH variation caused by YADH is comprised in the mentioned interval.
  • the pH variation is directly proportional to the number n of moles produced by the reaction.
  • a buffer is preferably chosen with a pK a value very close to the optimal enzyme pH range, and with a molarity which is a good compromise between the need for a pH-controlled environment and the need for detection of small pH variations.
  • the enzyme reaction is monitored by the system depicted in Figure 8. It is divided into two main sections: a reaction chamber and a measuring chamber, with a working volume of about 10 ⁇ l, which contains the silicon transducer, its driving light source and two electrodes (a Pt counter electrode and a Calomel reference electrode (AMEL 303/scg/6J)) connected to a potentiostat. The whole system is thermostated at 25 °C.
  • the system is biased through a Personal Computer AT bus 80486 processor, by means of a digital-to-analog card directly connected to the potentiostat.
  • the photocurrent signal is derived from an ohmic contact on the chip back side, and it is converted into a voltage by a current- to-voltage amplifier; then it is fed into a commercial Lock-In Amplifier (EG&G mod. 5210).
  • the output signal (previously referred to as T) is then computer acquired by means of a 16 bit analog-to-digital conversion card (National Instruments Mod. AT-MIO-16X). Appropriate software written under National Instruments LabWindows ambient allows fast signal acquisition and processing.
  • ⁇ vs. bias potential curves (as depicted in Figure 6) have been acquired, in order to determine both the sensitivity of the actual chip and its maximum slope, necessary to define a system sensitivity in terms of ⁇ , as already stated.
  • the signal ⁇ is recorded vs. time at a fixed bias potential; the total acquisition time interval can be set by the user: for the experiments here reported it was always set to 22 min.
  • NAD + Li was utilized instead of NAD + because the former does not cause any pH variation during the preparation of the solution.
  • the YADH (Sigma, cat.n.A7011) concentration was varied between 1.4 ⁇ 10 -8 M and 7. 1 ⁇ 10 -10 M, while the concentrations of substrates were kept constant.
  • formula (10) can be used to fit our experimental data, as shown in the previous Figure 9.
  • Figure 10 shows a set of experimental data at different YADH concentrations; each point is the slope of the tangent line to the exponential fitting function of the signal; fitting is performed only on the third (exponential) portion of the data, and the tangent is calculated at the first valid point .
  • the EtOH concentration was varied between 0.8 and 352 mM, and the same kind of acquisition (output signal vs. time) has been performed as previously described.
  • Figure 12 represents data values at different EtOH concentrations; each point in the plot is measured as described above, and as shown in Figure 11.
  • the biosensor here described has been tested to monitor enzyme reactions in order to detect both enzyme and substrate concentrations .
  • the sensor seems to be more accurate and useful for enzyme determination than for substrate detection, as appears from the presented results.
  • Figure 13 shows a schematic arrangement of the experimental setup. Inside the measuring flow chamber, a coverslip with living cells is placed in close proximity to the transducer, in order to obtain a microvolume chamber: in this tiny environment it is possible to detect even the pH variation related to the cellular activity.
  • the experiments here reported have been carried out with a volume of about 16 ⁇ l, obtained by interposing a thin spacer (typically 50-250 ⁇ m thick) between coverslip and transducer.
  • a suitable hydraulic circuit based on an electronic-controlled pump, continuously refreshes the cell medium into the microvolume chamber. To ensure an optimal condition for the biological sample, both the nutrient medium and the measuring chamber have been thermostated at 37°C.
  • a light addressable sensor should be biased by an external voltage source a counter electrode.
  • a modulated light source shining on the silicon chip causes an alternating current (photocurrent) to flow in the circuit; the current is measured, usually after a conversion to voltage, a filtration and a rectification.
  • Output voltage (V out ) versus bias potential (V bias ) characteristics curves have a sigmoidal shape, as V out varies from a high level (corresponding to silicon inversion condition), through a falling edge (corresponding to the depletion of majority carriers in the silicon), to a low level (corresponding to silicon accumulation condition).
  • V out versus V bias characteristics shift along the V bias axis accordingly with the pH variations of the solution in contact with the insulator surface and with the chip sensitivity.
  • Monitoring of the relative position, on the V bias axis, of the curve inflection point is a good method to follow pH variations.
  • Our transducer is biased by means of two electrodes (an Ag/AgCl reference electrode and a Pt counter electrode) connected to a potentiostat circuit; the modulation of light is obtained by driving an infrared LED with a sinusoidal waveform.
  • the output photocurrent is converted into voltage (with an usual gain of 106) and filtered.
  • a computer interfaced AD/DA acquisition card (National Instruments AT-MIO-16X) allows both to set the bias voltage and to acquire the output signal, by means of specific software programs.
  • the mouse fibroblast line 3T6 has been utilized in our study. All culturing was under standard conditions; 3T6 cells were in fact grown routinely in monolayer cultures at 37°C and 5% C0 2 ; culture medium consisted of RPMI 1640 (SIGMA, catalog number R- 6504) supplemented of antibiotic and 10% Fetal Calf Serum (Boheringher Mannhein, catalog number 210463).
  • the transducer is usually prepared by cleaning the sensitive surface (Si 3 N 4 layer) with isopropanol and double distilled water to remove impurities. Then the wafer is placed into the measuring chamber together with the spacer and a coverslip with cells. After these preliminary steps, the medium starts to flow into the chamber, feeding the cells. At this stage it is necessary to acquire the I/O sigmoidal characteristic (V out versus V biaa ) in order to evaluate its inflection point and its slope ct; these parameters will be used to define the proportionality factor between the read voltage and the actual ⁇ pH, as will be shown.
  • a software program allows the user to select one of several acquisition procedures; in every acquisition session the system is biased at a value corresponding to the inflection point, and the V out signal is continuously recorded.
  • the peristaltic pump is automatically switched ON and OFF at desired fixed time intervals.
  • the acidification rate can be expressed as:
  • ⁇ V out (expressed in mV) is the variation of V out during the considered time interval and ⁇ t is the considered time interval (expressed in sec).
  • Acidification data are presented as plots of V out as a function of time; by means of the above formula, V out values can be easily converted into pH units.
  • the device finalized to detect the pH variations of a microenvironment containing living cells, represents a valid tool to monitor the metabolic activity and eventually the physiological and pathological alterations of the cells. Because this potentiometric system detects pH changes and not directly the number of H + effectively excreted, it is important to define the mathematical expression representing the relationship between the protons produced by the cells and the resulting pH variation in the given
  • ß v is the buffer capacity of the cell medium
  • R is the proton generation rate in mol/sec
  • V is the volume in litres.
  • the informative portions of the plot of Figure 14 correspond to the OFF intervals.
  • the pH signal decreases linearly, and this variation, of the order of 10 -2 -10 -3 V/min (depending on the total acquisition time), represents the change in the acidification rate during the experiments.
  • the decreasing in the acidification rate may be due to a slowing down of the metabolic activities.
  • the PAB system integrates mechanics, hardware and software aspects, and consists in a complete stand-alone device for biosensing purposes; the block diagram is visible in Fig.18. As it clearly appears, the system is driven by a computer; this choice allows continuous measurements in time, and the automatic recording of data. As the interfacing electronics is standard and based on a 16 bit AT-BUS, a normal Personal Computer can be utilized, without any particular hardware modification.
  • the computer "talks" with the electronics through a digital interface, opportunely studied to guarantee high speed and fast acquisitions; in particular an analog-to-digital board converts the measuring signal, while a digital-to-analog board gives the proper supply to bias the transducer; in addition a digital output card provides eight digital signals to be used as switches for external controls.
  • the main block is the central Control Unit, consisting of several cards devoted to a precise signal amplification and conditioning, to the rectification and finally to the control of several parts of the system.
  • This unit interfaces directly with the solution pumping system and with the measuring chambers .
  • the latter always contain one or more transducers, directly biased and interfaced to the control unit.
  • FIG. 19 A schematic block diagram of the electronics is shown in Fig. 19. Several circuits are necessary to drive the LAPS
  • the amount of circuitry is bigger than in the case of ISFETs, where a few operational amplifiers can fully drive the transducer.
  • the picture shows the main blocks used to drive the transducer, but also some additional circuit devoted to the control of a pump and of the thermal regulation of the measuring environment.
  • the transducer is connected to a current-to-voltage
  • the measuring signal is an alternating current of the order of several micro amperes, usually generated by means of modulated light impinging the silicon surface.
  • the voltage-converted signal is then filtered and measured as RMS values; as the driving signal is available, our system utilizes a synchronous demodulation technique to recover the information.
  • a schematic diagram of this circuit is shown in Fig. 20. Here both the filtered input and the reference signals are fed in two multipliers; the first output consists of the product of the input signal with the reference component in phase with it, while the second output is derived by the 90° shifted reference component.
  • the output of this block is then a DC signal corresponding to the RMS of the actual (alternating) current signal, in a suitable form to be computer acquired after digitalization.
  • the generally used procedure to detect these fluctuations of the signal with respect to the applied potential involves the determination of the bias voltage corresponding to the inflection point of the characteristic; this value can be successfully used as an indicator of the actual surface potential.
  • a suitable electronics must be provided in order to achieve rather fast voltage scans, and the corresponding signal acquisitions; then the inflection point can be computed either by hardware circuitry or by software. In any case, the information can be retrieved only after a complete voltage scan, the global signal acquisition and the relative signal processing
  • the first circuit allows a proper biasing and control of the electrolyte-transducer system; the input consists of the digitally-converted bias potential, coming from the computer settings; a controlling electrode (Pt wire) and a reference electrode (SCE) are connected to this block.
  • the functional feature of the potentiostat is to maintain a controlled potential among the two electrodes and the transducer, which is considered as the working electrode, regardless any impedance fluctuation among them.
  • Thermal control is a must for bi ⁇ sensing purposes; it is definitely necessary when a biological sample is utilized, as in the case of cells (to maintain the population in living conditions) or in the case of enzymes (to guarantee the good working temperature).
  • Our system is equipped with a temperature control system, consisting of a circuit implementing a non-linear control algorithm and whose heating/cooling element is a Peltier cell; of course, the setup depends upon the application and varies accordingly to the actual measuring chamber; in fact the Peltier cell is always fixed inside a chamber, together with a temperature sensor utilized in the feedback circuit.
  • a temperature control system consisting of a circuit implementing a non-linear control algorithm and whose heating/cooling element is a Peltier cell; of course, the setup depends upon the application and varies accordingly to the actual measuring chamber; in fact the Peltier cell is always fixed inside a chamber, together with a temperature sensor utilized in the feedback circuit.
  • the last block is the interface with the flowing circuit; a flow of a measuring solution, of a given compound (i.e.
  • peristaltic pump which allows the solutions to flow from a beaker to the measuring chamber; the pump is completely automated, in the sense that start and stop commands come from the computer. This allows a very precise flow control in time, which is for example very useful in acidification experiments.
  • the interface with the motor pump has been designed by means of opto-couplers, which allows a good controlling technique and avoid
  • the complete hardware system described above practically consists of an interface board to plug into a PC bus, a connection cable, and an external circuit connected to the pump, to the electrodes and to the measuring chambers.
  • FIG. 21 Another important portion of the system is the set of program to drive the circuitry, to acquire the data and to perform the necessary processing.
  • the block diagram of software is shown in Fig. 21; it represents the main procedures for a typical data acquisition session.
  • a shell provides a user-friendly interface, and allows both parameters input and a complete data acquisition control.
  • the acquisition procedure depends upon the particular experiment, but in general a surface potential measurement is requested versus time, either pH or redox; basically, one can even acquire simple characteristic curves, for instance to investigate some surface property or to test a particular sensitive layer; in this case the user connects a standard (static) measuring chamber, biases the transducer and records the corresponding current values. This procedure can be easily performed, as it is the basis for further and more complicate acquisitions.
  • the program in this case constitutes the main portion of the system, as it performs the work that specific electronics usually does.
  • the pH or redox detection method implies the acquisition of a characteristic curve (current signal versus bias voltage); then the program automatically computes the bias potential corresponding to the inflection point, which s a good indication of the actual surface potential. Once the V bias is known, the system is biased at the inflection point
  • the actual temperature monitoring is also possible, by an additional AD related software.
  • the system can be equipped with a variety of measuring chambers; the purpose is to allow different experiments with the same device.
  • a standard static cell able to contain a solution in contact with the transducer; the system is depicted in Fig. 22.
  • the solution is confined by an O-ring pressed against the chip, and the two electrodes are dipped from the top.
  • This chamber is the simplest example, and it is useful for standard measurements; in addition it can be utilized for experiments on the transducer, as it can be thermostated and generally guarantees the maintaining of conditions constant in time.
  • Fig. 23 shows a flow cell suitable for enzymic applications; the solution flows through a reaction chamber where a membrane can be easily located; for instance, the membrane could be a PALL ImmunodyneTM for the immobilization of proteins (enzymes or antibodies).
  • the picture also shows the transducer and the related light emitting diode; again the measuring volume is delimited by a small O-ring, and a Pt wire is placed just in the proximity of the sensitive surface; the reference electrode can be located in the outlet channel.
  • the membrane can also be placed in the small volume near the chip, but in this case the eventual remotion implies opening the entire chamber.
  • Fig. 9 shows a typical acquisition for the determination of enzyme concentration; the signal decay is due to the H + production after the reaction took place, yielding to an acidification of the solution in contact with the transducer. Curves as the previous one allow the calculation of calibration curves, as the one shown in Fig. 10, which evidence the good sensitivity of the system, since it was possible to
  • potentiometric measurements is the overcoming of buffering capacity problems, as the surface potential is affected by electrons instead of protons; in order to create a surface charge layer just on the transducer, we evaporated a metal spot on the silicon nitride. Practically, we are using a competitive method and a configuration based on (monoclonal) antibodies entrapped in a membrane; the competition appears between the free antigen and the antigen bound to the HRP molecules. As soon as the enzyme substrates are injected into the reaction chamber, we obtain a signal variation responsive to the immune complex formation. A sketch of the described method is visible in Fig. 24A. A typical signal variation after the immunocomplex formation is shown in Fig. 24B.
  • FIG. 25 Another exciting application available to this system concerns measurements of extracellular acidification; hence, a specific micro-volume flow chamber able to contain a cell population has been designed and realised and is visible in Fig. 25.
  • a specific micro-volume flow chamber able to contain a cell population has been designed and realised and is visible in Fig. 25.
  • One important feature of the chamber is the
  • a good solution consists of a
  • a related chamber has been designed and is shown in Fig. 26; two chips are fixed in the lower side of the chamber and the flow, circuit sequentially contact both the
  • a reaction chamber (which can be a commercial membrane-containing teflon system) can be easily connected and removed in the area between the two measuring areas; in this case the first chip gives a signal related to the solution, while the second one measures the effects of the reaction.
  • the chips are very close to each other and fixed in the very same environment and with the same
  • LAPS generation technique of LAPS can be utilized in conjunction with different ion sensitive electrodes, yielding to a multi parameter detection within the same device; in addition we proved that in principle a standard MOS transistor can be utilized instead of the LAPS chip, by inducing an alternating current through a time varying small signal, and connecting some ion selective electrode; in addition to the previously mentioned advantage, here it is possible to think also to some integration of the biosensor, as standard processes are being utilized.
  • the invention provides a novel Potentiometric Alternating Biosensor (PAB) capable of functioning as an integrated system under different conditions and of being utilized with a variety of biological sensing elements for the analysis of a wide range of biological samples.
  • the transducing element is the so-called LAPS which actually might be regarded as one of the most reliable transducers used in the biosensor engineering.
  • transducers of this type could be considered already optimized for their specific applications due to the availability of the information arising from the sound knowledge of the modem semiconductor science .
  • the data acquisition and the control systems of the PAB were completely re-designed to better adapt the electronics to the general requirements of the biosensor and of the interfacing.
  • These sub-systems consist of several blocks interfaced by means of AD/DA converters to a usual IBM PC or compatible.
  • the set-up of the electronic cards has been specifically studied in order to minimize the noise and maximize the operation speed. Another important feature is the
  • a completely original software package has been written for the proper system control, for signal measuring and for data archivation and visualization.
  • the software runs on a DOS operating system, and allows the user to manage different types of signal acquisition.
  • the main feature of the low- level software is the interrupt-based structure which has confirmed to be the most reliable and to allow a good computer interfacing independently on the system clock.
  • controlled by the same computer unit permits to automate completely the operations with flow-through chambers, especially the start/stop operations, by means of respective electronic interfaces.
  • the PAB system can be thermostated by means of a purpose-designed temperature controlling system, based on a Peltier cell, which has an elevated precision and reliability.
  • a transducer whose surface is sensitive to many types of ion, and not only to hydrogen as in the case of silicon nitride transducers, may be used, for example, a gallium arsenide-based transducer; in addition to the GaAs property of a very high charge mobility, in fact, it is possible to build extremely sensitive layers by LB technology.
  • one of the advantages of the PAB lies in the possibility of uniting the pH-sensitive and redox-sensitive elements on a single sensor chip.
  • the transducer is essentially a heterostructure made of silicon ("n"- or "p”- type), silicon dioxide and silicon nitride (Sartore et al, 1992a; Sartore et al 1992b; Bousse et al. 1994).
  • the insulator is pH-sensitive , due to the proton binding capacity of its groups (essentially Si-O and Si-NH 2 ) over a large pH range (2-12), with an theoretical Nernstian response (if hysteresis and drift phenomena are not considered).
  • the sinusoidally-modulated LED illuminates the back-side of this modified structure, it produces an alternating photocurrent: its shape depends on the ratio between the metal layer size and the light spot size. If the light source illuminate both a silicon nitride region and a metal region, a biphasic response is obtained, due to the different surface potentials for the two zones (see Fig. 28, curve B). In this case, an infrared LED illuminates an area that is, approximately, in an equal percent, silicon nitride and metal layer in contact with the electrolyte solution.
  • the liquid phase contains ImM Potassium Ferricyanide, ImM Potassium Ferrocyanide and 100mM sodium citrate buffer, pH 6.
  • the A and C curves of Fig. 28 represent the two "extreme
  • the A curve is obtained with a metal spot that covers completely the nitride surface; the C curve, vice versa, represents the situation in which only the silicon nitride is in contact with the electrolyte.
  • the first inflection point of the biphasic response is due to the redox potential while the second one depends upon the pH of the electrolyte solution.
  • the ratio in the redox pair concentration an almost Nernstian shift along the bias potential axis is obtained, but this shift affects only the first portion of the bi-phasic response as depicted in Fig. 29a.
  • the H of the solution containing the redox pair only the second portion of the characteristic shifts, as represented in Fig. 29b.
  • the used metal are Chromium (a pad about 500-1000 A thick, evaporated by Balzers MlO metal evaporator) and Gold (a pad about 1000 A thick, evaporated by the same instrument, and on a previous thin layer of Chromium in order to increase the adhesion) .
  • Chromium a pad about 500-1000 A thick, evaporated by Balzers MlO metal evaporator
  • Gold a pad about 1000 A thick, evaporated by the same instrument, and on a previous thin layer of Chromium in order to increase the adhesion
  • characteristic curve is indicative of the actual surface potential, as already reported in Adami et al 1994A and B.
  • the second reaction is spontaneously occurring in the presence of the redox pair.
  • the pFe depends, as defined, on the conversion of Fe(II) into Fe(III); a theoretical
  • Fig. 31 The pFe dependence on the conversation of Fe(II) into Fe(III) is shown in Fig. 31; in the interval between +0.3 and -0.3 of the pFe axis the curve can be well approximated with a straight line: its slope is the proportionality factor between the pFe variations and n, the number of nanomoles of Fe transformed.
  • the enzymatic activity can be expressed in units defined as micromoles of TMB oxidized per minute, equivalent to half of the micromoles of Fe converted. So, it is possible to correlate the output signal of the sensor with the enzymatic concentration, if the specific enzymatic activity is known (Bousse et al, 1992; Adami et al 1994b) .
  • the calibration curve for the HRP enzyme in solution has been obtained, as shown in Fig. 32: the lowest detection limit in enzyme concentration is 5 ⁇ 10 -11 M.
  • the enzymatic activity determined with the procedure above described is in good agreement with the value coming from usual spectrophotometric assays.
  • it is essential to determine the activity of the immobilized enzyme.
  • the same configuration as depicted in Fig. 27 has been used, by simply inserting in the measuring chamber a membrane (Pall Biodyne B) of about 4x4 mm on which the enzyme has been immobilized.
  • the relative calibration curve has been obtained, as shown in Fig. 33: the lowest detection limit, in this case, is of about 2.5 ⁇ 10 -8 M.
  • the loss of activity due to the immobilization is considerable but does not prevent from the applicability of this system to an immunoenzymatic assay.
  • Diaphorase an auxiliary enzyme, Diaphorase, which can reduce Fe(III) in the presence of NADH
  • Fig. 34 shows the transducer output signal vs. time; this is the basic data by which calibration curves are determined, and shows the signal variation during the enzymatic reaction.
  • the detection of the ADH activity in a redox configuration presents several advantages with respect to a pH sensitive configuration, namely:
  • a high buffer concentration can be used, stabilizing the enzyme activity
  • thermodynamic equilibrium is not reached because of coenzyme recycling.
  • FIG. 1 A block diagram of a bidimensional PAB system is shown in Figure 1.
  • PAB Patentiometric Alternating Biosensor
  • the transducer consists of a light-addressable silicon chip, which provides regions properly modified and functionalized to yield sensitivity to either pH or redox potentials.
  • the system is electronically controlled and driven, and it is connected to a common personal Computer; ad hoc software drives an array of light sources, allowing 2D signal
  • the system is here utilized with a specific measuring chamber to monitor biological events related to in vivo cell metabolism.
  • Several other chambers have been designed to monitor different phenomena such as enzymatic activity (Y-ADH, Urease, HRP) and antigen-antibody binding.
  • the system can provide multiparameter information related to the specific distinct local modifications made on the sensing surface of the transducer.
  • the quantity which strongly affects the choice of dimensions in the design of this 2D system is the minority carriers diffusion length, defined as:
  • the signal acquired in a certain area is inevitably conditioned by the surface potential of adjacent zones; in the case of a multisensor, this can mean for example that a pH measurement can be affected by the redox potential of an area close to the pH sensitive one.
  • a preliminary setup was based on a single optical fibre scanned along the chip backside; at each step position a new value was acquired; this setup has the advantage of a uniform optical excitation of the chip, as the light source maintains the same features even if moved across the transducer; every acquired signal can be directly related to the other ones because all of them derive from the very same light
  • the final version of the system is based on an optical fibre array fixed in the close proximity of the transducer
  • the normalization procedure of the acquired signals is a very efficient method to avoid a lot of artefacts.
  • a potentiostat is utilized, as visible in Fig. 35, the system is composed by three electrodes; a (Pt) counter electrode, a (Calomel) reference electrode and the chip surface, which is considered the working electrode; this setup can be easily modelled with a couple of impedances, as it is very well explained in Bard et al, 1980, and depicted in Fig. 37A.
  • a significative number of artefacts during a measurement can be due to an undesired variation of such impedances, which can be due to temperature oscillations, to the medium conductivity, or even to more macroscopic reasons, such as bubbles in the hydraulic circuit.
  • the I-V characteristic curve of the transducer (whose inflection point is useful for the measurements, as already reported in Adami et al, 1994B) is globally distorted and varies in amplitude, as depicted in Fig. 37B.
  • the same effect takes place when the light source is positioned at different distances from the chip backside, and again when considering the I-V curves of single sensing spots lighted by fibres positioned in different locations with respect to the chip, or well positioned but driven by LEDs with different optical powers.
  • the normalization method here utilized considers the signal nearby the inflection point of the characteristic curve and the maximum amplitude of it, i.e. the value corresponding to biasing the device in inversion of majority carriers. Then any "true" data is the ratio between the inflection point value and the maximum one; this represents an easy way to overcome the above mentioned underived effects.
  • the design of a suitable measuring chamber for the 2D system combines the need of a relatively wide sensing area (with different sensing spots) and the corresponding backside positioning of the optical fibres array; in this sense a big effort should be spent in order to obtain a perfect alignment between the sensing surface spots and the corresponding fibres .
  • the chamber design allows the positioning of membranes in the close proximity of the sensing surface, and a cell population can either be grown directly on the chip or on a cover slip fixed in front of the transducer, at a very small distance (usually 50-200 ⁇ m) . With these peculiarities the chamber directly allows biosensing applications, such as enzymatic or cellular measurements .
  • FIG. 38 A schematic of the flow chamber is presented Fig. 38.
  • the chamber is made of two parts (indicated A and B) to be connected after the eventual insertion of a biological layer in the microvolume region F.
  • a medium flow comes from inlet C and fills the measuring chamber F, allowing a
  • the medium flow Before leaving the chamber through the outlet E, the medium flow enters the area indicated with D, where both a counter and a reference electrode can be placed (alternatively the counter electrode can be accommodated even within the inlet connection).
  • An array of optical fibres I is fixed at a predetermined distance from the transducer backside by an attachment L; this array is connected on the opposite side to a corresponding array of infrared LEDs (Light Emitting Diodes), driven by a multiplexing hardware.
  • the driving electronics of the 2D system requires a single light modulating source and two 4-bit (or one 8-bit) digital ports.
  • each fibre is connected to a LED which is addressed digitally.
  • the modulating signal is fed into a switching transistor through an operational amplifier, in order to ensure the proper current to the LEDs without affecting the local oscillator.
  • the driving transistor is connected to a series of digitally-controlled switches, in order to address a single LED at a time.
  • the selection of a given light emitting diode is performed by sending the corresponding digital address to a couple of multiplexers, namely the CD 4028. Up to 10 rows and 10 columns can be addressed by the proposed scheme, which can be connected either to a serial port (by an appropriate
  • DIO Digital Input Output
  • the driving software can be easily integrated within the single-channel acquisition program already described (Adami et al, 1994B); in fact it is necessary to repeat acquisitions at the different sensing locations.
  • the program Before acquiring the signal, the program sends two 4-bit words to the hardware circuitry in order to select a single LED, so enabling a given sensing spot; then the corresponding data is acquired, and the process re-starts.
  • a bidimensional pattern has been acquired at different measuring conditions.
  • a Si/SiO 2 /Si 3 N 4 chip has been partially covered with gold in two distinct areas; the gold was evaporated onto the chip by means of a Balzer M10 metal evaporator; for a better results, a thin layer of Chromium was previously deposited onto the silicon nitride.
  • the chip has been addressed by 4 optical fibres focused on the transducer backside at locations corresponding to the front side sensing regions; the
  • the chip was inserted into the measuring chamber, and then the measuring solutions were injected. Solutions containing different ratios of Potassium Ferrocyanide and Potassium Ferricyanide (namely 1:100, 1:10, 1:1, 10:1, 100:1) at pH 7 and 9 have been prepared.
  • Fig. 41 shows a typical acquisition when a solution
  • Fig. 41 shows the possibility to obtain bidimensional patterns relative to different sensitivities; one can very either the pH or the redox compounds ratio, causing a variation only of the specific local signal.
  • Fig. 42 is a collection of 2D patterns of the same type of the previous Fig. 41, varying both pH and redox pair
  • patterns A to E refer to pH 7 solutions at redox pair ratios range from 1:100 to 100:1
  • patterns F to L refer to pH 9 solutions at the same redox pair ratios, precisely ranging back from 100:1 to 1:100; it is possible to observe a very specific variation corresponding to a change in the measuring solution.
  • Biosensor for In Vitro Drug Screening Toxicity Testing in Cancerous Hepatocytes
  • the Potentiometric Alternating Biosensor (PAB) system has been utilized to monitor the effects of two antineoplastic drugs, Cytosine arabinoside and Mitoxantrone, on two distinct cell lines, namely on established cell line (3T6 mouse fibroblast) in different phases of cell cycle and primary culture (rat hepatocytes) in resting GO cells.
  • PAB Potentiometric Alternating Biosensor
  • An ad hoc microvolume flow chamber has been designed and produced; the chamber is equipped with inlet and outlet circuits and with a fixed transducer (Si/SiO 2 /Si 3 N 4 chip), facing a cover slip on which cells grow; the transducer allows monitoring of pH in the microenvironment where the cells are placed; the system is used in the pH-sensitive configuration and a single measuring spot has been used for the experiments, warranting an accurate determination of the change in the extracellular acidification rate resulting from drug administration.
  • a fixed transducer Si/SiO 2 /Si 3 N 4 chip
  • hepatocarcinomas which includes drugs as mitoxantrone, adriamycin and cis-platinum.
  • the cellular line chosen for experiment with Ara-C is 3T6 (Swiss albino mouse embryo, fibroblast) purchased from ATCC (American Type Culture Collection) . Conventional culture procedures have been followed using culture medium RPMI 1640 (Sigma).
  • the cells have been plated onto a glass support.
  • the glass supports have been
  • rat hepatocytes The mitoxantrone toxicity tests, we have chosen primary cultures of rat hepatocytes because they are easy to obtain and, like all hepatocytes, they maintain, in the first hours in vitro, their metabolic skills practically unchanged with respect to the in vivo situation. Hepatocytes were isolated from liver of Sprague-Dawley albino rats (200-250g) by in situ collagenase perfusion according to Williams (1977).
  • Isolated cells were suspended in Williams E. Medium (WME), supplemented with 10% fetal bovine serum and genamicin (50 g/ml) at the concentration at 5 ⁇ 10 5 hepatocytes/ml. Aliquots of this suspension were plated as follows: a) 1 ⁇ 10 5 cells were plated on the glass support (coated with collagen) for the measurement with PAB; b) 6 ⁇ 10 5 cells were plated on 35 mm dis es, coated with collagen, for conventional toxicity tests.
  • Ara-C is a synthetic nucleotide, which differs from the natural ones (cytidine and deoxycytidine) for substitution of ribose and deoxyribose with arabinose.
  • Constant levels of drug can be maintained by means of continuous i.v. administration.
  • Ara-C is principally used to induce regression of the acute proliferation hemopaties of the granulocytic series in the adult.
  • the secondary application is in the other
  • Mitoxantrone is a potent inhibitor of RNA and DNA synthesis. It intercalates on DNA, inducing cross links intra- and inter-strand, especially on GC base pairs. It interacts with the cellular membranes changing their functions.
  • administration route is the intravenous one. According to the autopic observations, the largest residual amount of mitoxantrone can be found in liver. In fact, in addition to renal excretion, the hepatobiliar pathway is largely involved in drug removal. There are evidences that the drug undergoes hepatic metabolisation; in fact four different metabolites have been isolated from urine.
  • the drug has been used since 1984 for the treatment of hepatocarcinomas, the dose of intravenous administration being 12-14 mg/m 2 in bolo. Since 1986, also the
  • the assay times are the following:
  • the tests have been carried out on primary culture of rat hepatocytes and on HEPG2 cells, with exposure times of 30 minutes and of 30 minutes + 20 hours of incubation in absence of drug, with doses of 2.5 - 5 - 10 - 20 - 40 - 80 g/ml for hepatocytes and 5 - 10 - 20 g/ml for HEPG2 cells (dosage selected after the tests on hepatocytes) .
  • Exposure time 30 minutes, cell line: rat hepatocytes; the cells showed a continuous distribution and a polygonal form similar to the one of the controls at doses 2.5 - 5 - 10 and 20 g/ml, even if a certain detachment of cells was observed with increasing doses.
  • the cells were losing the polygonal form and looked damaged, while at 80 g/ml the appearance of the cells was better: the phenomenon could be attributed to an edema of the cells caused by loss of normal permeability.
  • Exposure time 30 minutes + 20 hours
  • cell line rat
  • hepatocytes the comparison with the control cultures, which maintained the polygonal form, resulted to be strongly disadvantageous for all the doses.
  • the hepatocytes treated with 2.5 - 5 - 10 g/ml doses maintained the polygonal form even if not on a continuous layer and many detached cells were observed.
  • the cells assumed a round-shaped aspect; at 40 g/ml they were nearly completely detached and therefore dead for more that 50%.
  • 80 g/ml even if granulose and without evident margins, cells were still adherent to the support.
  • Exposure time 30 minutes, cell line: HEPG2; cells did not reach confluence, not even in checking dishes; cell
  • Exposure time 30 minutes + 20 hours, cell line: HEPG2; all the cultures appeared suffering compared to the control; we have observed a maximum cell detachment at a dose of 10 g/ml.
  • the cells presented several cytoplasmic vacuoles and extended cytoplasm.
  • the still viable hepatocytes showed, at all doses, to have incorporated mitoxantrone, in order to metabolise it.
  • the measurements with PAB are based on the monitoring of extracellular acidification; the relationship between production of acidic metabolites and rate of extracellular acidification is the following:
  • dn/dt is the generation rate of H + ions, due to acidic dissociation of excreted metabolites
  • ß is the buffering capacity of the examined solution
  • V is the volume of the chamber; in order to improve the sensitivity of the method, we can lower both volume and buffering capacity of the solution.
  • hepatocytes and HEPG2 cells we have used a 251-reaction chamber and a low buffered solution (phosphate buffer 1 mM, pH 7.4, added with NaCl to a final concentration of 100 mM): the short times (about 10 minutes of each measurement) of cell exposure to this non-specific medium practically does not influence their response.
  • phosphate buffer 1 mM, pH 7.4, added with NaCl to a final concentration of 100 mM phosphate buffer 1 mM, pH 7.4, added with NaCl to a final concentration of 100 mM
  • S is the sensitivity (50mV/pH, in our case) and a is the slope of the I-V characteristic curve of the transducer.
  • Figure 43 shows the relationship between extracellular acidification of 3T6 and drug (Ara-C) dose: low doses give no evident effect on the cell population while, at doses greater than 40 ⁇ g/ml, the action of the antimetabolite Ara-C on the cells synchronized is evident and consists of a linear decrease in extracellular acidification rates.
  • Figure 45 shows some of the acidification data acquired with PAB system on rat hepatocytes (30 min + 20 h treatment) .
  • mitoxantrone for instance, could be better understood: this drug has been studied in primary cultures of hepatocytes obtained from rat, rabbit and humans (Richard et al, 1991); variability in the metabolic pattern between the different species was observed.
  • the scheme of the PAB system is shown in figures 1, 8, 27 and 35.
  • a sinusoidally-modulated IR radiation lights the back side of the sensor chip a sinusoidal photocurrent is obtained, and its profile mainly depends on the chemical reaction which takes place in the measuring chamber.
  • the urease (Urea aminohydrolase, 61000 unit per gr.) from jack beans was purchased from Sigma and the 3-glycidoxypropyltriethoxysilane (GOPTS) from Aldrich.
  • the monolayer of urease was formed spreading 0.2 ml of 1 mg/ml enzyme solution at the water-air interface of the Langmuir-Blodgett trough (MDT Corp- Russia) whose dimensions are
  • Carbonate buffer (pH 8.6) was used as subphase.
  • the enzyme monolayer was transferred onto the activated support by Langmuir-Schaefer technique. After the incubation at temperature of 4°C for 4 hours the monolayer was washed by water flow and dried with nitrogen. The obtained enzyme monolayer was characterized by measuring the thickness and the surface density by means of ellipsometric and
  • the thickness of the monolayer measured was about 45 ⁇ , that also corresponds, in order of magnitude, to the dimensions of the urease molecule. Therefore, both these measurements confirmed that the obtained monolayer was densely packed.
  • reaction volume 25 ⁇ l
  • [UREA] 100mM
  • Figure 47 shows one of the curves acquired.
  • the pump When the pump is ON the products of the enzymic reaction flow away from the chamber and the dynamic equilibrium condition is reached: in this way the pH is nearly constant (first part of the curve).
  • the pump When the pump is OFF, the flow stops and the products of the reaction accumulate in a standing volume: this causes an increase of the pH (second part of the curve).
  • the "initial alcalinization rates" As the slope of the regression line in the first relevant points of the enzymatic reaction curve (see Figure 47). By repeating several times the measurements, always in the same conditions, it was possible to evaluate the stability of the monolayer.
  • LAPS light-addressable potentiometric sensor
  • PAB a newly designed potentiometric alternating biosensor system
  • PAB a newly designed potentiometric alternating biosensor system
  • Gavazzo P. Paddeu S., Sartore M. and Nicolini C. "Study of the relationship between extra cellular acidification and cell viability by a silicon based sensor”. Sensors and Actuators (1993, in press).
  • Nicolini C., Adami, M. Zunino M. and Sartore, M. A newly designed silicon-based cell biosensor for in vivo drug screening. Part 1 : a 2D system for simultaneous pH and redox determination.
  • LAPS light addressable potentiometric sensor

Abstract

The present invention relates generally to a potentiometric alternating biosensor system (PAB) which can use a silicon based light-addressable transducer to measure parameters such as pH and redox potential. A first aspect relates to temperature control of a PAB system. A second aspect relates to a biosensor for measuring pH and/or redox potential variations in enzymic reactions associated with cells. A third aspect relates to a silicon transducer redox potential biosensor having deposited thereon a metal layer over at least a portion of the surface, and its use in a PAB to monitor enzymes catalyzing redox reactions. A fourth aspect relates to a 2D PAB system capable of simultaneous pH and redox potential measurement. A fifth aspect relates to use of a PAB system for measurement of extracellular acidification to monitor cell metabolism and its use for testing anti-cancer activity/toxicity of compounds on cells. A sixth aspect relates to a PAB system capable of acting as an immunosensor in an immunoassay by virtue of the application to the transducer of an immobilized monolayer of an enzyme.

Description

Potentiometric Biosensors, Control and Applications Thereof
Temperature control is very important in many systems, mainly when a temperature sensitive device is being utilized. We have developed a silicon-based potentiometric sensor able to detect the H+ ions concentration (i.e. the pH) of an electrolyte in contact with it. One of the most exciting applications of such a sensor is the monitoring of the metabolic activity of a cell population; in this case a cover slip with attached living cells is put closely in front of the sensitive surface of the silicon sensor, in order to obtain a micro-volume nearby the sensor; cells are in contact with fresh medium, by means of a flow mechanism; when the flow is stopped, or some drug is added to the medium, cells acidificate the micro-environment more quickly; monitoring of the acidification rate allows us to understand the "status" of the cells under different conditions, and to predict the effects of drugs on the considered cell system.
In our application precision temperature control is necessary at least for three reasons :
the sensor used is silicon-based, hence it is extremely sensitive to temperature variations; if operating in a non- controlled system, the output signal cannot be unequivocally related to any physical event.
cells behaviour depends on temperature; the whole population must be at the same temperature if quantitative monitoring is performed.
extracellular acidification is strongly affected by the temperature at which cells are kept; usually 37°C is a suitable value for optimal cells conditions.
Temperature control is needed in very precise measurements. If one just wants to perform basic experiments on the device, acquisitions at room temperature in a normal laboratory are more than satisfactory, and no temperature control system is needed.
In applications such as cellular metabolism detection or enzymatic activity monitoring, temperature control is essential. In the biosensors field, usual temperature control is obtained by means of setups based on a warm fluid flowing around a considered region. In this case a (big) thermostat is kept outside the measuring chamber, and even outside the measuring desk; this system produces a warm fluid circulation in a tube coil, placed just around the measuring chamber.
The main difficulties utilizing these systems derive from low efficiency, rather high response time and presence of big mechanical apparatus (with related practical problems).
In accordance with one aspect of the present invention, a temperature control device comprises a heating/cooling device in the form of a Peltier cell; a temperature sensor for sensing the temperature in the vicinity of the cell; and control means responsive to the sensed temperature for controlling the heating and cooling activity of the Peltier cell so as to maintain the temperature in the vicinity of the cell at or near a predetermined value.
This temperature control device or thermostat is effective, cheap and easy to use. It can be applied in a variety of situations, when a precision temperature control is needed in a relatively small environment.
As mentioned above, the area of biosensors is one in which temperature control is important. We therefore provide, in accordance with this aspect of the present invention, a biosensor including a temperature control device for controlling the temperature in a sample region, the device comprising a heating/cooling device; a temperature sensor for sensing the temperature in the vicinity of the device and control means responsive to the sensed temperature for controlling the heating and cooling activity of the heating/cooling device so as to maintain the temperature in the vicinity of the device at or near a predetermined value.
Although the temperature control device used with this biosensor could take a variety of forms, preferably the heating/cooling device is in the form of a Peltier cell. It has been found that in applications where the temperature of a small environment must be controlled, the use of Peltier elements guarantees high efficiency and uniform temperature distribution. Furthermore, a Peltier-based device enables temperatures to be controlled at target or predetermined values higher or lower than room temperature. Furthermore, the Peltier cell can easily and rapidly be used as a heating or cooling element. Finally, this system is very compact and the driving electronics can be installed in a small box.
A number of biosensors are known to employ light-addressable potentiometric sensors (LAPS). The principle of operation of a light-addressable potentiometric sensor is described by Hafeman et al, Science (1988) 240:1182 and by Sartore et al, Biosensors & Bioelectronics (1992) 7:57-64. Use of such sensors has allowed the design of biosensors that are capable of measuring either pH variations (if the sensor's surface is coated with a H+ sensitive insulator, such as Si3N4 or Ta2O5) or redox potential variations (if metal spots are evaporated onto the sensor's surface). It is also possible to produce a sensor having a specific sensitivity to inorganic ions, by covering the sensor surface with PVC/ionophore membranes, as described by Adami et al, Sensor & Actuators B, (1992) 7:343-346.
The use of light-addressable potentiometric sensors for monitoring enzyme reactions through pH variations has been investigated, as described in Bousse et al, Sensors & Actuators B, (1990) 1:555-560. However, there are practical problems underlying their use for this purpose. Every enzyme needs a certain pH-controlled environment to work properly, and this is obtained using a buffer. Unfortunately the buffering capacity can easily overcome the effect of the enzymic reaction, in terms of pH variation, which therefore cannot be measured.
Light-addressable potentiometric sensors have also been used to monitor extracellular acidification in micro environments, again through pH variation. Many important metabolic processes of the cells (namely the catabolism of sugars, amino acids and fatty acids) produce H+ ions, which are excreted through the cytoplasmic membrane and out of the cell, thereby causing an extracellular pH variation. The use of light-addressable potentiometric sensors to measure this pH variation is described by Owicki et al, Biosensors & Bioelectronics, (1992) 7:255-272; Owicki et al, Proc. Natl. Acad. Sci. (1990) 87:4007-4011; and Wada et al, Journal of Cell Biology, (1991) 115:A2455.
However, again, there is a practical problem in this application, in that the cell medium usually contains a buffer whose capacity masks any pH change due to extracellular acidification, which therefore cannot be measured.
In a second aspect, the present invention relates to a biosensor, and in particular to a biosensor for measuring pH variations and/or redox potential variations in enzymic reactions, either taking place within the cell and causing extracellular variations, or in isolation.
According to the present invention, a method of measuring a pH variation or a redox potential variation in an enzymic reaction that generates ions that cause, or whose generation causes, such a variation, comprises monitoring the reaction over a period of time using a light-addressable potentiometric sensor that generates a current on the binding of the respective ions thereto; converting current measured over that time to voltage; if desired, calculating the pH variation or the redox potential variation, or the number of the respective ions generated, as a function of the voltage; and comparing the voltage, pH variation or redox potential variation, or number of ions with a precalibrated standard.
The method of the invention allows the monitoring of range of enzymic reactions through a variation in pH and/or in redox potential occurring in those reactions. The method of the invention can be applied to determine enzyme and substrate concentrations to low orders of magnitude, for example in the 10-10M range. By using a sensor that is sensitive to redox potential variations, the problem related to the measurement of small pH variations in buffered solutions is overcome. In this case the measured quantity is a surface potential, and therefore buffers, even with high buffer capacity, do not affect the sensitivity of the method.
A biosensor can be programmed to carry out either or both of the above methods, and to display the results thereof in a suitable manner. This allows the user to carry out a wide variety of experiments on the same or different biological samples in a simple and efficient manner.
As will be apparent from the following, a biosensor suitable for carrying out the method of the invention comprises a reaction chamber in which the reaction is to proceed and into which at least one sensor can be immersed. Alternatively, the reaction chamber can be separate from at least one measuring chamber, each having a sensor associated therewith. The biosensor further comprises specific electronic cards and related software programs to acquire the necessary data quickly, and present it in a user- friendly manner. The biosensor is preferably completely automated and driven by a personal computer.
The electronics is designed in order to obtain a good general signal-to-noise ratio, hence ensuring reproducible results.
The software package typically consists of low-level programs (written, for example, both in C and Assembler), to interact with the data acquisition and controlling cards, and of high-level programs (for example written in C), to obtain, display, save and print data files, and to present a suitable user interface.
A very important peculiarity of the biosensor of the invention is the ability to connect different sensors or a plurality of measuring chambers each associated with a different sensor, to the same device; this allows the user to perform a wide variety of experiments, utilizing the same or different biological elements, by simply changing the measuring chamber under consideration and selecting the appropriate acquisition software. This can be done by simply substituting electrical connections and does not require physical movement of the measuring chamber per se.
Examples of different kinds of sensors include a sensor sensitive to pH variation by the binding thereto of H+ ions, and a sensor sensitive to redox potential variation by the binding thereto of ions generated in a redox reaction. An example of the latter kind of sensor is one which has had gold metal evaporated onto its surface. Sensors that are sensitive to inorganic ions are known, and can also be used in the biosensor of the invention.
Alternatively sensors that are sensitive to more than one of the above types of ion can be used.
The biosensor of the invention utilizes the fast information recovery from the sensor output signal, and can acquire quantitative data at fraction of a second intervals over period of time; it can therefore be utilized to monitor fast acidification or redox processes.
All the above aspects, together with the compactness and the modularity of the device, and with the capability to sense both small pH and redox potentials, and potentially other quantities, show that this system can be regarded as a complete biosensor system.
Our biosensor system, called PAB (Potentiometric Alternating Biosensor) offers the possibility to connect different measuring chambers to the same device, allowing the user to perform a wide variety of experiments, by simply changing the reaction chamber and selecting the appropriate software. The used transducer is based on LAPS (Light Addressable Potentiometric Sensor, Hafeman et al, 1988) technology, hence it is basically sensitive to pH variations of the electrolyte solution in contact with the sensitive surface. The system has been tested with cells (Gavazzo et al. 1994) and with enzymes producing a pH variation, such as YADH (Alcohol Dehydrogenase from Yeast, Adami et al, 1994b) or GST (Glutathione-S-Transferase, Antolini et al, 1994). In a third aspect of the invention, we used a particular flow-through chamber containing a modified silicon based transducer in order to detect the redox potential of a redox pair in contact with the sensitive layer.
A potentiometric sensor, based on a silicon chip, able to detect redox potential changes in solution is produced and some of its possible applications are investigated. The redox potential of a solution in contact with the surface of a metal layer deposited on the chip affects the amplitude of a photocurrent signal generated in the silicon by means of a modulated light source.
We investigated the behaviour of the structure at different ratios of the redox pair concentration, in order to obtain a calibration curve. The same measurements have been performed with different metal layers, of different sized, in order to find a configuration suitable for a biosensing purpose.
An enzymatic application has been shown with HRP in solution and then immobilized on an activated membrane. For these studies a micro-volume reaction chamber has been set up, with a microchannel system near the sensitive area.
The choice of HRP is linked to the widespread use of this enzyme as label in immunoassays, therefore giving the possibility to use this system as an immunosensor . Anyway, other enzymes can be used and another type of assay is proposed, using Diaphorase together with Alcohol Dehydrogenase.
The biosensors field is presently growing in the direction of multisensors and applications have been already presented where 2D information was recovered through an array of sensing sites (for example see Lundstrom et al, 1993). The improvement being made by going 2D in gas sensitive field effect structures is readily apparent if we compare the original demonstration during 1975 (Stiblert et al, 1975; Lundstrom et al, 1975) of MOS-FETs to the recent production of response patterns for molecules and odours (Lundstrom et al, 1993). Similarly striking improvements can be expected by applying multiparametric determination to the potentiometric alternating biosensor recently introduced by our group (Adami et al 1994); this is achieved through the deposition on the silicon transducer of gold spots (redox sensitive regions with and without thiols monolayers with different chain lengths and tail groups, i.e. with varying functionalities (Tangvall et al, 1992. Bertilsson and Liedberg, 1993)). The parallel and simultaneous acquisition of many signals relative to a given measuring situation is indeed interesting for the possibility to obtain recognition systems able to work even in unfavourable conditions, likewise in the presence of electrical noise or of undesirable contributions.
This applies as well to analytical devices capable of selective and precise measurements of cell metabolism at high resolution (Adami et al 1994A and Gavazzo et al 1994). In a fourth aspect of the invention there is provided a biosensor capable of 2D acquisition and/or multisensing; the device is potentiometric and is based on a silicon transducer with signal generation caused by light excitation. The system can recover the local spatial information of pH, redox and other significant quantities on a sensing area of about 1cm2, which is usually contained in the particular measuring flow chamber specifically designed for the given biosensing application.
It is our opinion that the surface biology in complex but biologically relevant systems may be conveniently studied through at 2D PAB addressing the desired cell surface property with the thiols properly functionalized.
The bidimensional PAB system allows a fast potentiometric measurement in the 2D space of the transducer surface. Spatial information recovery is achieved by means of local light stimulation of the silicon transducer. In this way the surface potential of the system is locally measured; the surface charge, giving rise to the potential, can be originated in different ways (Delia Ciana et al, 1991), allowing multisensing operations onto a single chip; for instance, the transducer surface can be partially covered by Si3N4/ for pH detection, or by gold, for redox measurements, or by different chemically sensitive sites, such as LaF3 for F- detection; moreover, specially functionalized sites can be created onto the sensing surface, for instance by chemical interactions between gold spots and specific compounds (i.e. thiols).
When the transducer surface is completely covered by Si3N4 the system produces a 2D map of the solution pH under investigation; instead, when additional sensitive sites are used, the system can be considered both bidimensional and multisensitive; in other words the local surface potential can still be recovered preserving the spatial information, so obtaining a 2D multisensor.
One of the most relevant characteristics of PAB is the versatility of the system.
The basic principles of cell metabolism monitoring with PAB can be illustrated as follows: every perturbation of the steady-state condition yields fluxes of biomolecules through the cell itself with cause changes in the extracellular acidification (Gavazzo et al, 1993) the stronger alteration, the bigger the pH variation.
This aspect of our work used stabilized cell lines (HeLa, 3T6) of which we have monitored extracellular acidification in different conditions; the use of these lines allowed us to test the performances of our system as a device for investigation of cellular events: PAB was found to be sensitive to variations in cellular status due to the different cell cycle phase and to drugs's action (Gavazzo et al 1993; Nicolini et al 1994).
A fifth aspect of the invention provides for the application of PAB system for the investigation of the effects of antineoplastic drugs on two stabilized cell lines (normal mouse fibroblasts 3T6 and transformed HEPG2 hepatomas) and on primary cultures (normal rat hepatocytes); this work shows the feasibility to utilize the PAB biosensor as an innovative test with respect to conventional ones for the estimation of drug efficacy and toxicity.
In order to test the feature of PAB, we have chosen 3T6 cell line for investigation of Ara-C (cytosine arabinoside) effects and rat hepatocytes and HEPG2 as cellular target for the estimation of the toxicity-efficacy of Mitoxantrone, a drug extensively utilized in the clinics to treat liver carcinomas.
The liver is the major site for the uptake of drugs and chemicals, converting them to pharmacologically inactive, active or even toxic metabolites. Biotransformation of xenobiotics and other specific liver functions, mostly performed by parenchimal liver cells (hepatocytes), are difficult to study in the whole organism, because of the influence of other organs, tissues and exogenous/endogenous factors; consequently, in vitro hepatocytes represent a valuable tool in pharmoco-toxicology, since it was demonstrated that they preserve, when in culture, the functional drug metabolising enzymes in culture (Engelmann et al 1985; Dich et al, 1988; Martelli et al, 1988).
A large variety of drugs, having different therapeutic actions and exhibiting in vivo either a simple or a complex metabolism, has been tested with hepatocyte cultures and the major metabolite or even the complete metabolite's pattern has been determined.
Our British Patent Application No. 9324257.4 describes the PAB. Our British Patent Application No. 9324227.7 describes the production of thermally stable protein layers. These two concepts are combined here.
A possible application of the PAB system as an immunosensor can imply the usage of a modified Langmuir-Blodgett (LB) technique for the production of antibody monolayer. According to a sixth aspect of the invention, we immobilized a thin film monolayer of an enzyme (urease) to simulate the final step of an immunoenzymatic assay. In order to ensure a strong link between the support of the LB film, we used the method of substrate vacuum silanization for the covalent immobilization of the monolayer.
The immobilized monolayer was characterised by means of ellipsometric and nanogravimetric techniques in order to evaluate the main physical parameters such as thickness and surface density. Together with the enzymatic activity we also investigated the operative lifetime of the monolayer. The performance of the PAB system utilizing the thin film technology was analyzed.
The aim of our work is to study the enzymatic activity of the urease immobilized onto a support (glass or silicon nitride (Si3N4)) inserted into a microvolume reaction chamber in order to evaluate the application of the LB technology to immunosensors based on PAB. The enzyme used in this experiment was urease, which catalyzes the hydrolysis of the urea.
An example of a temperature control device and biosensor according to the first aspect of the invention will now be described with reference to the accompanying drawings, in which:
Figure 1: Schematic view of a typical measuring flow chamber utilizing the thermostat here described. An aluminum plate is in contact with the heating element, a Peltier cell; the temperature sensor is placed just under a micro-volume area (the region to thermostat). Liquid flows in and out by means of two channels, and the measuring area is just nearby the sensitive layer (Si3N4) of a silicon-based sensor placed in front of a cover slip.
Figure 2 : Control loop of the thermostat. User can set a target temperature Td, continuously compared with the measured temperature to generate an error signal e(t). Non linear control is accomplished by means of a "Bang-Bang" system which drives the Peltier unit either with full power heating or with full power cooling. A temperature sensor closes the loop, and gives the actual temperature value.
Figure 3: Theoretical temperature signal vs. time, as derived from equations 7, 10 and 11 (see text). The figure also contains a plot of the input square wave u(t) and of h(t), as expressed in the mentioned formulas. All values are in arbitrary units, and normalized between +1 and -1. Input sequence wave [u(t)], output signal [y(t)] and its deviation from input [h(t)] is shown.
Figures 4A and 4B: Hardware circuitry of the thermostat. Terminals referred to as Vp should be connected to an external constant-current power supply, regulated for an output voltage as specified in the particular Peltier cell data sheets. The connection to a milli-voltmeter can be also used to interface the circuit to an A/D converter, in order to monitor temperature changes in time by a computer.
Figure 5: Computer acquired temperature signal as a function of time. The acquisition interval corresponds to a complete semi-period of the (square-wave) control signal. Target temperature was set to +37°C, and the system kept this value in about 3 minutes after system switch on. A fit with the theoretical expression of y(t) is also shown; data and fitted curve are in considerably good agreement. The control signal frequency was about 5.5 KHz and the resulting value of τ was 22.1 μs. Variation of measured temperature vs. time for a Peltier cell, and fitting with theoretical expression is shown.
The second aspect of the invention will be described by way of example only with reference to the accompanying drawings, in which:
Figure 6: Voltage Ψ vs. bias potential curves corresponding to two different pH values in the solution in contact with the transducer.
Figure 7: Relationship between produced H+ moles and pH variations: a region is shown with a direct proportionality between n and ΔpH, where a fitting straight line is also shown.
Figure 8: The whole system is divided into two main sections: a reaction chamber and a measuring chamber (working volume ≈ 10 μl), both kept at 25°C. Conditioning and acquisition electronics are also shown.
Figure 9: A typical experiment for the determination of enzyme concentration. Three different zones are visible: a first constant region, a small step and an exponential-like curve; they all are due to the particular configuration of the measuring system, as described in the text.
Figure 10: Output signal slope vs. YADH concentration; each data is calculated on the respective third (exponential) portion of the data presented in Figure 9, and the tangent is calculated at the first valid point.
Figure 11: Sensor output signal variation during the enzymatic reaction: the figure shows a typical result. The net variation in the output signal is essentially the difference between the baseline signal level (at to = 5 min) and the steady level (at to ≈ 22 min).
Figure 12: Output signal variation vs. EtOH concentrations; each point in the plot is measured as shown in Figure 11.
Figure 13: schematic representation of the measuring system. Cell medium flow is obtained by means of a peristaltic pump, which is connected to a measuring chamber, containing the transducer and a coverslip with cells; the cell medium flows in a very small region between transducer and coverslip. The figure also shows a control electronic block, which provides the system with all the necessary driving signals.
Figure 14: output signal during ON-OFF acquisitions. When the pump is ON the pH in the measuring chamber is maintained constant, while switching OFF the pump yields to a net pH variation, due to the presence of acidic products in the micro environment.
Figure 15: a similar signal as in Figure 9 is here shown, relative to a single OFF time interval. The signal is approximately linear in the considered region, and can be fitted with a straight line, whose slope represents the acidification rate.
(Figure 16: not included).
Figure 17: two images of the 3T6 cell monolayer as usually appears before (A) and after (B) an acquisition session with the presented system. Images have been acquired by a CCD camera with an optical microscope interfaced with a Personal Computer.
Fig. 18: Schematic representation of the PAB system. A central control unit drives and controls the solution flowing system and the measuring chambers containing the transducer; the control unit also dislogs with a Personal Computer, utilized to bias the device and acquire and process the experimental data, via appropriate AD/DA interfaces.
Fig. 19: Block diagram of the main electronics. The LAPS is biased via a potentiostat and the current signal is converted to voltage, filtered and synchronously demodulated, before being digitized and computer acquired. The figure also shows a digital output interface suitable to drive the flow circuit through an interface; in addition a temperature control scheme is provided.
Fig. 20: Schematic circuit of the synchronous demodulation technique utilized, me input signal, after a current to voltage conversion and a filtering, is multiplied by the reference signal coming from the LED driver; the output of these blocks are then filtered and amplified before being fed into the circuit for calculation of the modulus.
Fig. 21: Software organization of the PAB system. A shell offers a suitable user interface, and allows parameters input, data display, saving and printing; the usual procedure consists in the acquisition of a complete characteristic curve (current versus bias potential) and in the consequent determination of the inflection point; our software then bias the device and continuously acquires the digitized signals.
Fig. 22: Schematic of a very simple measuring chamber; a well in the upper part confines the measuring solution in a region delimited by an 0-ring, directly pressed against the LAPS surface. Counter and Reference electrodes are dipped in the measuring solution from the top, while a LED confined in the lower part just near the back of the chip provides the light excitation.
Fig. 23: Measuring flow chamber suitable for enzymic experiments. The lower side is quite similar to that of Fig. 22; the upper part contains an inlet channel towards a well able to contain a membrane; usually some biological specie is entrapped in the membrane at this level; the reaction products flow then in a small volume region near the sensitive surface, through a channel as tiny as possible, in order to reduce the delay time of the sensor response. The solution flows away through an outlet channel, contacting the reference electrode.
Fig. 24: Part A shows the immunoreaction scheme, where the antibody is fixed onto a substrate and the antigen with. enzyme is injected in the solution to compete with the free antigen; the binding of enzyme-linked molecules is detectable after the production of electrons. Part B shows a typical acquisition curve (I/V characteristic) obtained utilizing Horse Radish Peroxidase as the labelling enzyme, 2,4 dichlorophenossyacetic acid (2,4-D) as antigen and a monoclonal antibody against the pesticide 2,4-D; the right-band curve is before binding, the leftmost after binding.
Fig. 25: Measuring flow chamber for cellular acidification on experiments. The counter electrode is inserted inside the inlet channel, while the reference electrode is in the outlet. The LAPS is separated from the cells immobilized onto a glass coverslip by a teflon spacer, whose thickness defines the micro volume of the measuring chamber. Thermal control is achieved by a Peltier cell connected to an appropriate circuitry and fixed on the top of the chamber; heating is obtained near the cells by a metal cylinder (usually aluminum) .
The right part of the figure shows the top view of the gasket, utilized to prevent medium leakage, of the coverslip where cells are grown and of the teflon spacer, giving a indication of the flow circuit inside the "sandwich".
Fig. 26: Two-chips measuring chamber suitable for differential measurements. Here the flowing solution appears at the first chip, then passes through a removable reaction chamber, before reaching a second chip. Of course, only the second LAPS senses the effects of the reaction that took place in the chamber; both chip should be affected by the same noise, as the particular design ensures for both the same environmental conditions, hence a differential acquisition scheme detects only the true reaction signal.
The third aspect of the invention will be described by way of example only with reference to the accompanying drawings, in which:
Fig. 27. Schematic representations of the PAB system in a redox configuration. The measuring chamber is shown in detail. Biasing of the transducer is achieved by a potentiostat , while signal measurements are performed by a synchronous demodulation technique.
Fig. 28. Different curves of chips with and without metal layer to a solution containing ImM Fe(II), ImM Fe(III) and lOOmM sodium citrate buffer pH 6.0 A: chip with a gold layer covering the entire surface in contact with the solution. B: chip with a gold layer covering only a portion of the exposed surface. C: chip without metal layer (Si3N4 sensing surface).
Fig. 29a. Biphasic response curves of a chip with a gold layer covering part of the sensing surface; buffered solution containing the redox pair in different ratios were used. From left to right the curves correspond to solutions of Fe(II): Fe(III) r)tios ranging from 1:100 to 100:1. Fig. 29b. Biphasic response curves of a chip with partial gold covering of its exposed surface to solutions at two different pH values: at left pH 6 ; at right pH 9.
Fig. 30. Bias potential values corresponding to the inflection points at different redox pair ratios plotted versus the logarithm of the ratios themselves for chips with gold and chromium layer.
Fig. 31. Variation of the logarithm of the redox pair concentration ratio when Fe(II) is oxidized to Fe(III). The experimental conditions are: reaction chamber volume 10ℓ, 0.8mM initial Fe(II) concentration. 0.4mM initial Fe(III) concentration. The curve can be approximated to a straight line in the y-interval between +0.3 and -0.3.
Fig. 32. Calibration curve for the HRP enzyme in solution. The enzymatic activity units are here defined as moles of TMB oxidized per minute, corresponding, in turn, to half of the moles of Fe(II) converted to Fe(III).
Fig. 33. Calibration curve for the HRP enzyme immobilized on activated membrane.
Fig. 34. Monitoring of Alcohol Dehydrogenase activity in redox configuration by the use of Diaphorase as auxiliary enzyme.
The fourth aspect of the invention will be described by way of example only with reference to the accompanying drawings, in which:
Fig. 35. Block diagram of the PAB system. The transducer is biased by a potentiostat, and the alternating current photogenerated by LEDs is converter into voltage and amplified; after filtering, the RMS values of the signal are extracted by synchronous demodulation, and acquired by a DAC card into a Personal Computer. A temperature control system based on a Peltier cell is provided, in order to stabilize the environmental conditions of the measurement. Fig. 36. Schematic structure of the transducer-optical fibres system. The distance of adjacent light spots is related to the wafer thickness, because the main process in the signal generation is the diffusion of charge carries.
Fig. 37. Part A represents a simple equivalent circuit of the three-electrode system; Zrc is the impedance between counter and reference electrodes, and Zwr is that between working and reference electrodes; changes in these impedances cause signal amplitude variations, as depicted in Part B. The normalization procedure described in the text avoids the eventual artefacts due to the monitoring of such impedance variations.
Fig. 38. Schematic representation of a measuring flow chamber of the 2D PAB system. Part A is connected to B after the eventual insertion of some biological element (i.e. membrane) into the small volume region F. Medium flows from inlet C, passes through the sensing spots H and G, and exits through outlet E, after filing the electrodes region D. An array of optical fibres I is fixed at a given distance from the chip backside by the holder L.
Fig. 39. Hardware circuit for the optical fibres array selection. The driving signal (usually a sine- or a square- wave) is fed to a single selected LED at a time, passing through the operational amplifier and the switching transistor. The remaining blocks are used to digitally connect a desired LED to the drive electronics; in particular two 4-bit words are the row-column addressed for the CD 4028 multiplexers, which activate corresponding digital switches. (Drawing missing).
Fig. 40. Schematic representation of the transducer sensing surface utilized in the presented experiments. Two gold spots have been evaporated onto the silicon nitride surface, and four light spots are created in order to obtain two redox sensitive and two pH sensitive sites.
Fig. 41. Typical acquisition results after the insertion of a Potassium Ferrocianide-Ferricianide solution into the measuring chamber. Data corresponding to coordinates (1,1) and (2,2) are relative to redox, while those corresponding to coordinates (1,2) and (2,1) refer to pH. (Drawing missing) .
Fig. 42. Series of measurements as the one shown in Fig. 41. Data from A to E show redox pair concentration variations range from 1:100 to 100;1 at pH 7; data from F to L range back from 100:1 to 1:100 at pH 9. It is evident that variations in the redox concentrations selectively affect the only redox-specific spots, while the pH variation is specifically sensed by the silicon nitride regions. (Drawing missing).
The fifth aspect of the invention will now be described by way of example only with referenced to the accompanying drawings, in which:
Fig. 43 shows the dose/effect trend of ara-C on S-phase 3T6 cells acquired with PAB system;
Fig. 44 shows a comparison between PAB results and results of the Trypan Blue Test; and
Fig. 45 shows rat hepatocyte acidification curves acquired with PAB.
The sixth aspect of the invention will now be described by way of example only with reference to the accompanying drawings, in which:
Fig. 46 shows monolayer surface density v. surface pressure obtained by means of nanogravimetry;
Fig. 47 shows pH variation due to urease activity monitored by means of PAB system: the enzyme was immobilized on silanized glass: the regression line helps in evaluating the initial alcalinization rate of the reaction; and
Fig. 48 shows activity of urease monolayer immobilized of silanized glass during repeated assays (Deposition Pressure 20mN/m) .
Temperature control device
The system utilizes a flow chamber 1, as depicted in Figure 1; a silicon chip 2 is put in front of a cover slip 3, and the medium flows in between. The cover slip (containing cells) occupies a well 4 in an aluminium plate 5, which is in contact with the heating element (the Peltier cell) 6. A temperature sensor 7 (AD 590, by Analog Devices) is placed just under the cover slip 3. The other side of the Peltier cell 6 is connected to a convection heat sink 8.
In order to control the temperature of an environment like a micro-volume flow chamber we found suitable to use a Bang-Bang control algorithm. The method implies, in general, a full-on / full-off input to the actuator of the considered system; in our case this means to feed full-power heating or full-power cooling to the Peltier element.
Another typical control strategy utilizes a PID (Proportional Integral Derivative) algorithm; in the latter case the control signal is generated by adding three terms, one proportional, the second related to the first derivative and the last related to the integral of the error signal.
The Bang-Bang control technique is much easier and immediate than the PID one, as it is not necessary to compute (either by software or by hardware) the proportionality factors at each control step. In addition, an hardware solution for the control system (as the one here presented) only requires a comparator-based stage in the circuit. Finally, the performance here requested (thermal stability within 0.1°C around the desired temperature, and the latter ranging between +20°C and +45 °C) can be fully obtained without a PID control.
A sketch of the control system loop is shown in Figure 2. The desired temperature Td can be manually set; it, is continuously subtracted from the actual measured temperature y(t) by a subtracter 10, hence generating an error signal e(t) used to set the desired temperature, while the two lateral 500 Ω variable resistors are trimmers used to set the minimum and maximum allowed temperature; in order to set these two values one should turn the potentiometer first completely to the left, then adjusting the left trimmer until the desired minimum value appears on the connected digital display; then one should turn the potentiometer completely to the right, and adjust the right trimmer for the maximum value.
The second signal to the comparator comes from the Operational Amplifier CA313024, used as a driver for the temperature sensor (AD 590). The 10KΩ trimmer on the right allows to adjust the driver in order to sense the correct temperature value; to set the temperature measuring section it is necessary to utilize at least a 0.1°C precision thermometer and to perform several steps of temperature monitoring near the two desired extremes with both methods, then adjusting the trimmer in order to read the true value on the milli-voltmeter.
Finally, a switch 25 connects either the pre-set desired temperature signal, or the actual measured temperature signal to a milli-voltmeter (digital display); usually one sets the switch to the Td position, selects the desired temperature, and then switch to the monitoring position in order to read the actual temperature while thermostating process is on.
The signal to the display can be directly connected (with no additional circuitry) to an A/D converter input, if computer monitoring of the temperature is needed.
The error e(t) is greater than 0 if the actual temperature is lower than the desired one, lower in the opposite case: e(t)>0 if Td>y(t)
e(t)<0 if Td<y(t)
The signal e(t) is used as input to a non linear stage 11, which produces a control signal u(t), which, power-amplified 12, corresponds to the driving signal of the Peltier cell 6. When the control signal u(t) is low, one side of the Peltier element 6 is heating, the opposite is cooling (and vice-versa when the control signal is high); fast transitions of the control signal allows the temperature control around the target value.
The feedback system is completed by a temperature measuring block 7. Temperature sensing is accomplished by using a temperature probe connected to an appropriate driving system, whose output is y(t).
A typical set up as the one here described takes some time to reach a stable state corresponding to the desired temperature; the elapsed interval from the system power-up to the stable situation depends upon several factors: the selected current in the Peltier power supplying circuit, the temperature probe location with respect to the heating element, the type and shape of the heat sink, the system to thermostat.
After the stable state is reached, every variation of the temperature above or below the desired value causes an immediate transition of the control signal from high to low level or vice versa. In this situation the variation of the control signal is nearly periodical, and the time interval the control signal is at high level is ideally equal to the time it is at low level; any discordance with this description is due to noise, either environmental or thermal. Then, as soon as the temperature reaches the selected value, the driving control signal is a square wave with a 50% duty cycle.
The situation can be analyzed by supposing that the application of a step unit function u(t)=l(t) to the Peltier element yields to a response in the measured temperature signal y(t) of exponential nature: y{t) = {l - e-t/τ) [1]
The respective Laplace transforms of the input and output signals are: [2 ]
[3 ] and the system transfer function is
[4]
In order to explore the depicted situation, let us suppose now to apply a (normalized) input square wave as the control signal to the Peltier element, starting from time t=O, as expressed before:
[5] where a is the semi-period of the square wave ranging between -1 and +1
The Laplace transform of [5] is:
[6]
In this case the output temperature signal is given by:
[7] where the function H(s) is expressed as follows:
[8]
The problem is now to find the inverse Laplace transform of H(s); this can be accomplished using the convolution theorem, and the result is:
[9] where n= [t/a] is the number of input square wave sweeps before the actual one. The first term depends on time through the exponential factor exp(-t/τ), and represents the initial transient response following the system power-on; this term expires with the time constant τ. The second term represents the steady-state portion of the response to the square wave, as its dependence on t is expressed by exponential terms like exp[-(t-na)/τ], related to the number of sweeps before the actual one.
The steady-state part of the response varies with respect to the parity of n, so that we can consider the two cases when n=O and n=1 to successfully describe the system response:
[10]
[11]
The desired description of the output signal of a Peltier element when the input driving signal is a square wave has thus been found. The function hss (t), together with the input square wave and the total output signal y(t), is shown in Figure 3.
The circuit implementing the Bang-Bang control here described is quite simple (Figure 4) . The heating-cooling technique is achieved by feeding the Peltier element 6 with a given current flowing in one sense or in the opposite one; this can be easily obtained by applying a fixed voltage Vp to the unit in order to heat, and the opposite -Vp in order to cool.
The circuitry of the control system is depicted in Figure 4. The Peltier element 6 connected to the external power supply generating Vp by means of power transistors pairs (2N3055) 13 used as switching elements; if the right pair or the left pair is on, current flows into the Peltier cells in one sense or in the opposite one, yielding the heating or cooling of a given cell side.
The couple of transistors 14, 15; 14', 15' before each final pair provides the necessary base current to the finals, when a high level is present on the control signal line 16. Two additional
Figure imgf000027_0001
Figure imgf000027_0002
Figure imgf000027_0003
Figure imgf000027_0004
Figure imgf000027_0005
Figure imgf000028_0001
Figure imgf000028_0002
Figure imgf000028_0003
Figure imgf000028_0004
Figure imgf000028_0005
transistors 17,18 are used to drive red and green LEDs 19,20 indicating to the user when the system is heating and when it is cooling.
The non-linear control signal is generated by an AD 845 (Analog Devices) Operational Amplifier 21, here utilized as a zero-hysteresis comparator. This integrated circuit is a suitable comparator because its slew-rate is quite high (100 V/μs).
The operational amplifier 21 continuously compares the two signals fed into its inverting and non-inverting input pins, hence generating a +Vsat/-Vsat output if the actual temperature is lower/higher than the desired one.
If the output value is +Vsat then the left-hand-side NPN-type 2N2219 transistor 15 will switch on, while if the output value is -Vsat the right-handside PNP-type BC320 15' transistor will switch on.
One of the compared signals comes from the resistors 22 to the left of the comparator 21: here the central variable resistors 23 is a multi-turn potentiometer
Example
The described thermostat has been realized and tested with a 3x3 cm, 65 W, Peltier cell. The system was supplied by a commercial +.12 V power supply; in addition a selectable constant-current power supply was connected to the terminals Vp in Figure 4, i.e. to the Peltier element through the transistors pairs. As our Peltier cell data sheets suggest we set this external power supply to 8.5 V, and 1.5 A. The measuring chamber to thermostat is a plexiglass-aluminum cylinder of about 5 cm base diameter, and 5 cm high.
The Peltier cell 6 as connected to the aluminum part 5 depicted in Figure 1, and the system was tested at room temperature for target temperatures ranging between +20°C and +45°C. After switching on the thermostat, the output of the comparator is at high/low level depending upon the desired temperature as compared to the actual one; then a heating/cooling process starts, and no commutation arises until the temperature becomes higher/lower than the desired one. At this point a transition in the control system causes an inverse cooling/heating process with respect to the prior one (the heated side is now cooled and vice versa). Monitoring of the actual temperatures shows in this situation a series of over- and under-shoots around the desired value; we observed that the number of over-/under-shoots depends on the relative positioning of the temperature probe with respect to the Peltier cell, and also on the current value of the external power supply; in particular higher selected current values cause more over-/under-shoots.
In any case, after a time depending upon the mentioned parameters, and in our case not greater than 5 minutes, the system reaches and keeps the target temperature.
We observed that below a certain current value (in our case 1.5 A) it was not possible to reach 45 °C as a target temperature, and that beyond another current value (in our case 3 A) the system showed a precision less than the claimed ±0.1°C; however, current values up to 3 A did not produce system instability.
As an example, the actual temperature has been computer- acquired, after the initial transient expired and the target temperature was kept; values have been fit with the theoretical expression of y(t) presented before. Figure 5 shows the result; for a sake of clarity, the figure corresponds exactly to a low-level semi-period of the control signal square wave when the target temperature was set at 37 °C.
The total semi-period is 90 μs, which implies a commutation frequency of the Bang-Bang control system of about 5.5 KHz, as can also be easily observed by connecting an oscilloscope to the comparator output; the behaviour of the actual temperature vs. time has an exponential nature, as predicted by the expression of y(t). Moreover, fitting with y(t) gives an estimation of , which results to be about 22 μs.
The described thermostat can be easily applied in different research areas, when a precision temperature control is needed in a small environment; the application here reported makes use of a flow chamber, and the "sensitive region" to control is about 8×8 cm; by using bigger Peltier elements it is possible to control, by means of the very same circuitry, a bigger area and a different system.
Finally, the total cost of the device is very low, allowing the possible users to get a precision system very cheaply.
Potentiometric Alternating Biosensor (PAB)
In the following, the embodiment of the invention to monitor pH variations in enzymic reactions in isolation is described first, and then the embodiment to monitor intracellular reactions, e.g. metabolic processes, by way of extracellular acidification is described. Finally, the biosensor, referred to in the following as a Potentiometric Alternating Biosensor (PAB) system, of the invention is described.
Enzymic reactions in isolation
For present purposes, the enzymic reaction under consideration is that of YADH on NAD+ and C2H5OH.
In order to reveal the enzymatic activity by means of the sensor output signal, we must define the correlation between H+ production, ΔpH in low buffered solution and sensor response.
The transducer output signal is a photocurrent, converted to a voltage Ψ by a current to voltage converter with a usual gain of 106; this voltage is directly proportional to the actual pH of the solution in contact with the sensing area, see Figure 6. The sensitivity σ of the transducer corresponds to the bias voltage shift to apply to the system in order to get the same Ψ after a pH unit step; to this sensitivity directly corresponds a y-axis sensitivity, which also depends upon the maximum curve slope α; a variation ΔΨ of the output signal can be expressed in terms of ΔpH, as follows:
Figure imgf000032_0002
hence providing a direct and univocal relationship between read voltage Ψ and pH variations.
At this point we need to know the correlation between the number of H+ moles produced and the obtained pH variation. This relationship depends upon the working volume V and the volumetric buffer capacity ßv, as expressed by the following formula:
Figure imgf000032_0001
As ßv depends on pH, the complete equation relating pH and n is as follows (1)
Figure imgf000032_0003
where c is the molar buffer concentration, V is the volume expressed in litres and xo is equal to 2.303 (pKa-pHo), where pHo is the initial pH, before the H+ production by the enzymatic reaction and pKa is related to the buffer specie. The above function is plotted in Figure 7, for a 1 mM buffer concentration and a volume of 10 ml; the value of pKa is 8.4 and that of pHo is 8.7 (the same values have been utilized in our experiments, as described later).
The first portion of the curve can be approximated very well with a straight line, by means of a least squares fitting; the fitting line is also shown in Figure 7. The portion utilized for the fitting corresponds to a (pH-pKa) ranging from +0.3 to -0.5. The total pH variation caused by YADH is comprised in the mentioned interval.
Considering the selected region, formula 1 can be rewritten as:
Figure imgf000033_0002
and in this case the pH variation is directly proportional to the number n of moles produced by the reaction.
It is finally possible to univocally correlate the sensor output signal T with n by means of the proportionality factors contained in (1) and (4).
As mentioned above, when measuring pH variations the production of H* ions is typically masked by the pH in the reaction chamber that is maintained constant by the presence of a buffer. In order to by pass this problem, a buffer is preferably chosen with a pKa value very close to the optimal enzyme pH range, and with a molarity which is a good compromise between the need for a pH-controlled environment and the need for detection of small pH variations.
The enzymatic reaction we are considering for our sensor is :
Figure imgf000033_0001
The YADH optimal working pH is around 8.8, so we decided to use TAPS buffer (pKa = 8.4).
The theoretical study proved that a value of 8.7 for pHo is the most suitable in this case and that a molarity of 1 mM is the best choice for a good ΔpH detection and for a stable working condition.
The enzyme reaction is monitored by the system depicted in Figure 8. It is divided into two main sections: a reaction chamber and a measuring chamber, with a working volume of about 10 μl, which contains the silicon transducer, its driving light source and two electrodes (a Pt counter electrode and a Calomel reference electrode (AMEL 303/scg/6J)) connected to a potentiostat. The whole system is thermostated at 25 °C.
The system is biased through a Personal Computer AT bus 80486 processor, by means of a digital-to-analog card directly connected to the potentiostat.
The photocurrent signal is derived from an ohmic contact on the chip back side, and it is converted into a voltage by a current- to-voltage amplifier; then it is fed into a commercial Lock-In Amplifier (EG&G mod. 5210). The output signal (previously referred to as T) is then computer acquired by means of a 16 bit analog-to-digital conversion card (National Instruments Mod. AT-MIO-16X). Appropriate software written under National Instruments LabWindows ambient allows fast signal acquisition and processing.
Before starting every experiment with enzyme, Ψ vs. bias potential curves (as depicted in Figure 6) have been acquired, in order to determine both the sensitivity of the actual chip and its maximum slope, necessary to define a system sensitivity in terms of Ψ, as already stated.
In a typical experiment, the signal Ψ is recorded vs. time at a fixed bias potential; the total acquisition time interval can be set by the user: for the experiments here reported it was always set to 22 min.
A baseline was acquired before every experiment with enzyme; the acquisition was carried out in the following conditions: TAPS 1 mM, NaCl 100 mM, EtOH (concentration depending on the experiment), NAD+Li ImM (every compound was purchased from Sigma). In particular, both in this case and during experiments with enzyme NAD+Li was utilized instead of NAD+ because the former does not cause any pH variation during the preparation of the solution.
Two kinds of experiments have been carried out, for the determination of enzyme and substrate concentrations. Experiments for the determination of enzyme concentration
In these experiments, the YADH (Sigma, cat.n.A7011) concentration was varied between 1.4×10-8M and 7. 1×10-10M, while the concentrations of substrates were kept constant.
After recording the baseline curve, we introduced a certain amount of enzyme in the reaction chamber and recorded data as those visible in Figure 9. Here it is possible to define three different zones which depend upon the physical construction of the experimental setup: a first constant region, a small step and an exponential-like curve. The constant part is the signal acquired before the introduction of the enzyme, and it is constant because no pH variation occurs, as predicted. A dead volume between the reaction and the measuring chambers causes the enzyme-containing solution to come into the sensing region when the enzyme reaction already started; the transducer then senses a pH step depending on the dead volume. The third portion of the curve represents the true enzyme reaction monitoring, and quantitative data have been calculated based on the analysis of this region.
It can be stated that:
Figure imgf000035_0001
and that the reaction kinetic is described by the following equation:
Figure imgf000035_0002
If we consider the initial part of the reaction, still far from the equilibrium point, then the second term in the above formula can be neglected. Under this hypothesis, the reaction kinetic can be described as follows:
Figure imgf000035_0003
We used a rather high concentration of EtOH, which can be considered constant during the initial part of the reaction; it follows that the above formula can be rewritten as:
Figure imgf000036_0001
where the term k* includes both the kinetic constant and the concentration of EtOH. The last equation can be easily integrated:
Figure imgf000036_0002
Figure imgf000036_0003
Figure imgf000036_0004
Considering eqs . (3) and (5), which contain a correlation between reaction kinetic and pH variation, formula (10) can be used to fit our experimental data, as shown in the previous Figure 9.
Figure 10 shows a set of experimental data at different YADH concentrations; each point is the slope of the tangent line to the exponential fitting function of the signal; fitting is performed only on the third (exponential) portion of the data, and the tangent is calculated at the first valid point .
Experiments for the determination of substrate concentration
These experiments have been carried out at a concentration of YADH of 7×10-8 M, constant for every experiment; this rather high value was chosen in order to reach quickly the final equilibrium.
The EtOH concentration was varied between 0.8 and 352 mM, and the same kind of acquisition (output signal vs. time) has been performed as previously described.
In these experiments the total acquisition time was 22 min; an example of such acquisitions is shown in Figure 11, relative to an EtOH concentration of 352 mM. This curve is used to calculate the net variation in the output signal, after reaching the equilibrium; the measured value is essentially the difference between the baseline signal level (at t0 ≈ 5 min) and the steady level (at tf ≈ 22 min).
Figure 12 represents data values at different EtOH concentrations; each point in the plot is measured as described above, and as shown in Figure 11.
The biosensor here described has been tested to monitor enzyme reactions in order to detect both enzyme and substrate concentrations .
The sensor seems to be more accurate and useful for enzyme determination than for substrate detection, as appears from the presented results.
In fact, despite the unfavourable equilibrium constant of the catalyzed reaction, the sensitivity of the sensor for YADH (in solution) is good. Detection of 7×10-10 M of enzyme could allow the use of our system in immune-enzymatic assays, so offering a wider range of possible applications, as described by Colston et al, Biosensors & Bioelectronics (1993) 8:117-121.
Extracellular Acidification
Data Acquisition system
Figure 13 shows a schematic arrangement of the experimental setup. Inside the measuring flow chamber, a coverslip with living cells is placed in close proximity to the transducer, in order to obtain a microvolume chamber: in this tiny environment it is possible to detect even the pH variation related to the cellular activity. The experiments here reported have been carried out with a volume of about 16 μl, obtained by interposing a thin spacer (typically 50-250 μm thick) between coverslip and transducer. A suitable hydraulic circuit, based on an electronic-controlled pump, continuously refreshes the cell medium into the microvolume chamber. To ensure an optimal condition for the biological sample, both the nutrient medium and the measuring chamber have been thermostated at 37°C.
As already reported, a light addressable sensor should be biased by an external voltage source a counter electrode. A modulated light source shining on the silicon chip causes an alternating current (photocurrent) to flow in the circuit; the current is measured, usually after a conversion to voltage, a filtration and a rectification.
Output voltage (Vout) versus bias potential (Vbias) characteristics curves have a sigmoidal shape, as Vout varies from a high level (corresponding to silicon inversion condition), through a falling edge (corresponding to the depletion of majority carriers in the silicon), to a low level (corresponding to silicon accumulation condition).
Vout versus Vbias characteristics shift along the Vbias axis accordingly with the pH variations of the solution in contact with the insulator surface and with the chip sensitivity. Monitoring of the relative position, on the Vbias axis, of the curve inflection point is a good method to follow pH variations.
Our transducer is biased by means of two electrodes (an Ag/AgCl reference electrode and a Pt counter electrode) connected to a potentiostat circuit; the modulation of light is obtained by driving an infrared LED with a sinusoidal waveform. The output photocurrent is converted into voltage (with an usual gain of 106) and filtered. A computer interfaced AD/DA acquisition card (National Instruments AT-MIO-16X) allows both to set the bias voltage and to acquire the output signal, by means of specific software programs.
The whole system is placed into a Faraday box in order to minimize the effects of external electromagnetic interferences. Cell Culture
The mouse fibroblast line 3T6 has been utilized in our study. All culturing was under standard conditions; 3T6 cells were in fact grown routinely in monolayer cultures at 37°C and 5% C02; culture medium consisted of RPMI 1640 (SIGMA, catalog number R- 6504) supplemented of antibiotic and 10% Fetal Calf Serum (Boheringher Mannhein, catalog number 210463).
For experimental purposes the cells were subcultured in Petri dishes on negatively charged polystyrene coverslips and were allowed to reach complete confluency.
Microscopic analysis
Every coverslip utilized for extracellular acidification experiments has been observed by a Zeiss Axioplan Microscope
(Zeiss, Oberoken, FRG) to get information about the density and the morphology of the cells. These observations have been performed using a phase contrast illumination and an objective lens 20x magnification; the images, acquired using a CCD camera
(Photometrix L.T.D., Tuxon, AZ) and computer digitized, have been subtracted of their background to get a better resolution.
A common "Trypan Blue exclusion" test has been utilized to estimate the percentual viability of cells after the acidification assay. Nevertheless it should be considered that dead cells mainly detach from the monolayer under the mechanical action of the medium flow and are therefore lost during the acquisition experiments.
Data acquisition procedure
The transducer is usually prepared by cleaning the sensitive surface (Si3N4 layer) with isopropanol and double distilled water to remove impurities. Then the wafer is placed into the measuring chamber together with the spacer and a coverslip with cells. After these preliminary steps, the medium starts to flow into the chamber, feeding the cells. At this stage it is necessary to acquire the I/O sigmoidal characteristic (Vout versus Vbiaa) in order to evaluate its inflection point and its slope ct; these parameters will be used to define the proportionality factor between the read voltage and the actual ΔpH, as will be shown.
Then a signal acquisition versus time can start. A software program allows the user to select one of several acquisition procedures; in every acquisition session the system is biased at a value corresponding to the inflection point, and the Vout signal is continuously recorded. In addition, the peristaltic pump is automatically switched ON and OFF at desired fixed time intervals.
When the pump is ON, fresh medium is continuously injected into the measuring chamber which contains the cells; in this condition the metabolic products flow away and a pH approximately constant is maintained inside the chamber. Depending on the medium buffering strength and composition, however, pH can slightly vary even in this case; this effect has been observed with different media. As soon as the pump is switched OFF, cells excrete their acidic products in a standing medium, so causing a pH variation. This acidification is transduced by the sensor into a corresponding variation of the output signal Vout.
Given the sensitivity S (expressed in mV/pH) of the transducer and the previously determined slope α , the acidification rate can be expressed as:
L W r
Figure imgf000040_0001
where ΔVout (expressed in mV) is the variation of Vout during the considered time interval and Δt is the considered time interval (expressed in sec).
Acidification data are presented as plots of Vout as a function of time; by means of the above formula, Vout values can be easily converted into pH units. The device, finalized to detect the pH variations of a microenvironment containing living cells, represents a valid tool to monitor the metabolic activity and eventually the physiological and pathological alterations of the cells. Because this potentiometric system detects pH changes and not directly the number of H+ effectively excreted, it is important to define the mathematical expression representing the relationship between the protons produced by the cells and the resulting pH variation in the given
environment; the link between pH variations and number of protons is as follows:
Figure imgf000041_0001
where ßv is the buffer capacity of the cell medium, R is the proton generation rate in mol/sec and V is the volume in litres.
By the combination of (2) with the Handerson-Hasselbach equation it is possible to derive :
Figure imgf000041_0002
where c is the medium concentration and x=2.303 (pKA-pH), pKA being the acid dissociation constant.
Both a theoretical calculation mid an experimental titration on the RPMI medium, utilized in our experiments, show that βv is constant in the pH range between 7 and 7.4 (which corresponds to our working pH range) and has a value of 2.5 mM.
It is necessary to underline that the medium flowing in the microvolume chamber has no added NaHCO3 (normally added to the medium to obtain pH stability), because its very high buffer capacity at the physiological pH could consistently reduce the detection of the excreted H+.
In Figure 14 a typical acidification acquisition performed alternating flow ON and flow OFF steps is shown. This plot refers to a 3T6 experiment in the mentioned conditions . When the peristaltic pump is stopped, a dramatic increase of H+ concentration in the extracellular environment induces an abrupt variation in the output voltage signal. The same signal behaviour as the one of Figure 14 has been observed, in the same conditions, with a very high reproducibility. As one can observe, during OFF time intervals the pH signal is quite noisy and slightly increasing; the effect can be ascribed to a non perfect medium thermal condition between pump and chamber or the high variety of components present in the medium. Similar acquisitions with other mammalian cells and different medium show a better signal stability during OFF intervals. In any case, the informative portions of the plot of Figure 14 correspond to the OFF intervals. Here the pH signal decreases linearly, and this variation, of the order of 10-2-10-3 V/min (depending on the total acquisition time), represents the change in the acidification rate during the experiments.
Recording of acidification rates has been performed on many 3T6 samples; a significant set of measurements can be found in Table 1, below, where the acidification rate is expressed in terms of different units; experimental time is intended starting from the positioning of the coverslip into the measuring chamber. In our opinion, it is important to refer to this event, rather than to the effective acquisition beginning, because cells can be damaged during this period of time, in which they are forced to live without fresh medium. In particular, it is very important to perform all the mechanical actions (coverslip insertion, tube connections, electrode positioning, ...) as soon as possible. Our setup, actually, allows a rather fast cell positioning, of the order of 2-3 min; this time is comparable with a usual OFF interval, and does not cause significant alterations to the sample .
Values in Table 1 were derived from the Vout signal; in fact, it is possible to calculate the acidification rate by a linear regression of the signal in the flow OFF regions, as visible in Figure 15. The ΔpH values are obtained by formula (11), considering the actual chip sensitivity [in our case S = 50.9 mV/pH] and the maximum slope of the characteristic [in our case α= - 2 . 916] .
Finally, from pH it is possible to calculate the number of protons effectively excreted from the cells, taking into account the buffer capacity of the cell culture medium (ßv) and the chamber volume (V).
For the reported experiment R varies from 5.21×1013 [protons/sec] in the first acquisition to 1.24×1013 [protons/sec] after 1% hours. In order to better understand the possible reasons of this decreasing we analyzed by means of an optical microscope, before and after the experiments, the cells utilized in the acidification assay. What is immediately evident from the pictures of Figure 17 is a decrease in the cellular area, averaging from 650 μm2 (Figure 17-A) to 350 μm2 (Figure 17-B) and a change of the cellular morphology, that from polygonal and fibroblast-like becomes tendentially more round shaped. If we consider that the spherical morphology generally determines for the cells a reduction in the available surface for exchanging material with the extracellular environment and represents a protection of its functions in unfavourable circumstances, then the decreasing in the acidification rate, at least partially, may be due to a slowing down of the metabolic activities.
On the other hand a consistent number of cells detach from the monolayer during the medium flowing, while a percent of cells (about 5%) of the final monolayer are dead, as evidentiated from the Trypan blue exclusion test.
All these factors can justify the reduction of the resulting acidification rate, that generally varies from 0.1 ΔpH/min to 0.05 ΔpH/min during about 45 min. We cannot calculate the weight of every individual factor affecting the acidification rate during the acquisition, while it is possible to estimate the number of protons initially (i.e. relative to the first data acquisition) excreted from the single cell in the time unit. In this case, considering the number of cells (8×104) in the chamber, the volume of the measuring chamber itself, the buffer capacity and the acidification rate, by formula (12) we can estimate this number around 6.5×108 protons per cell per second.
Finally, considering all the above results we can conclude that our system, devoted to the analysis of cellular metabolic activity, seems to be suitable for this type of measurements. In fact cells are allowed to live for a long time (3-4 hours), once located in the device, and only a few percent of them die. The measuring chamber, the software and the electronics design allows the very small pH changes, due to cell metabolism, to be correctly recorded.
Figure imgf000045_0001
The PAB system integrates mechanics, hardware and software aspects, and consists in a complete stand-alone device for biosensing purposes; the block diagram is visible in Fig.18. As it clearly appears, the system is driven by a computer; this choice allows continuous measurements in time, and the automatic recording of data. As the interfacing electronics is standard and based on a 16 bit AT-BUS, a normal Personal Computer can be utilized, without any particular hardware modification.
Appropriate software programs have been written both for signal acquisition and processing, and for controlling the device.
The computer "talks" with the electronics through a digital interface, opportunely studied to guarantee high speed and fast acquisitions; in particular an analog-to-digital board converts the measuring signal, while a digital-to-analog board gives the proper supply to bias the transducer; in addition a digital output card provides eight digital signals to be used as switches for external controls.
The main block is the central Control Unit, consisting of several cards devoted to a precise signal amplification and conditioning, to the rectification and finally to the control of several parts of the system. This unit interfaces directly with the solution pumping system and with the measuring chambers . The latter always contain one or more transducers, directly biased and interfaced to the control unit.
A schematic block diagram of the electronics is shown in Fig. 19. Several circuits are necessary to drive the LAPS
transducer, and to measure the output signal; in this respect, the amount of circuitry is bigger than in the case of ISFETs, where a few operational amplifiers can fully drive the transducer.
The picture shows the main blocks used to drive the transducer, but also some additional circuit devoted to the control of a pump and of the thermal regulation of the measuring environment.
The transducer is connected to a current-to-voltage
converter; in fact, the measuring signal is an alternating current of the order of several micro amperes, usually generated by means of modulated light impinging the silicon surface. The voltage-converted signal is then filtered and measured as RMS values; as the driving signal is available, our system utilizes a synchronous demodulation technique to recover the information. A schematic diagram of this circuit is shown in Fig. 20. Here both the filtered input and the reference signals are fed in two multipliers; the first output consists of the product of the input signal with the reference component in phase with it, while the second output is derived by the 90° shifted reference component. These signals, low-pass filtered to cut the undesired high
frequency components, are fed into a circuit for the
calculation of the modulus, after a proper amplification.
The output of this block is then a DC signal corresponding to the RMS of the actual (alternating) current signal, in a suitable form to be computer acquired after digitalization.
Acquisition of this signal versus the bias potential yields to the well known sigmoidal characteristic curves, as already reported (Hafeman et al., 1988 and Sartore et al., 1992a); the surface potential detection method is always related to the shift of the mentioned curve on the Vbias axis after a modification in the surface charge; in case of pH variations detection, it is possible to define a sensitivity of the transducer in terms of the shift amplitude per pH unit; on the other hand, in case of redox potential measurements, a similar definition can be expressed in terms of the redox pair components.
The generally used procedure to detect these fluctuations of the signal with respect to the applied potential involves the determination of the bias voltage corresponding to the inflection point of the characteristic; this value can be successfully used as an indicator of the actual surface potential. When applying the described technique, a suitable electronics must be provided in order to achieve rather fast voltage scans, and the corresponding signal acquisitions; then the inflection point can be computed either by hardware circuitry or by software. In any case, the information can be retrieved only after a complete voltage scan, the global signal acquisition and the relative signal processing
(usually involving the second derivative computation); the computational time needed to perform all the steps is the limiting factor in the acquisition velocity.
Our system, as also visible in Fig. 19, does not present any special circuit for the above mentioned signal acquisition and processing; in fact, we only partially utilize the described method, as it will be detailed in the next software section. The implemented method allows a faster information detection and, at the same time, greatly simplifies the hardware .
Coming back to Fig. 19, we can now focus on the remaining blocks: the potentiostat, the thermal control and the interface for the flowing mechanism. The first circuit allows a proper biasing and control of the electrolyte-transducer system; the input consists of the digitally-converted bias potential, coming from the computer settings; a controlling electrode (Pt wire) and a reference electrode (SCE) are connected to this block. The functional feature of the potentiostat is to maintain a controlled potential among the two electrodes and the transducer, which is considered as the working electrode, regardless any impedance fluctuation among them.
Thermal control is a must for biσsensing purposes; it is definitely necessary when a biological sample is utilized, as in the case of cells (to maintain the population in living conditions) or in the case of enzymes (to guarantee the good working temperature). Anyway, in addition to the
preservation of the optimal biological conditions temperature control is necessary because of the use of a silicon
transducer, extremely sensitive to thermal variations.
Finally, a good thermostat system is necessary to obtain true reproducible results.
Our system is equipped with a temperature control system, consisting of a circuit implementing a non-linear control algorithm and whose heating/cooling element is a Peltier cell; of course, the setup depends upon the application and varies accordingly to the actual measuring chamber; in fact the Peltier cell is always fixed inside a chamber, together with a temperature sensor utilized in the feedback circuit.
The last block is the interface with the flowing circuit; a flow of a measuring solution, of a given compound (i.e.
enzyme substrates), or of cell medium is usually utilized in the standard experiments. Our system makes use of a
peristaltic pump which allows the solutions to flow from a beaker to the measuring chamber; the pump is completely automated, in the sense that start and stop commands come from the computer. This allows a very precise flow control in time, which is for example very useful in acidification experiments. In order to minimize the noise in the signal conditioning circuits, already presented, the interface with the motor pump has been designed by means of opto-couplers, which allows a good controlling technique and avoid
interference problems, as the motor and driving circuit grounds are physically separated.
The complete hardware system described above practically consists of an interface board to plug into a PC bus, a connection cable, and an external circuit connected to the pump, to the electrodes and to the measuring chambers.
Another important portion of the system is the set of program to drive the circuitry, to acquire the data and to perform the necessary processing. The block diagram of software is shown in Fig. 21; it represents the main procedures for a typical data acquisition session.
A shell provides a user-friendly interface, and allows both parameters input and a complete data acquisition control. Of course, the acquisition procedure depends upon the particular experiment, but in general a surface potential measurement is requested versus time, either pH or redox; basically, one can even acquire simple characteristic curves, for instance to investigate some surface property or to test a particular sensitive layer; in this case the user connects a standard (static) measuring chamber, biases the transducer and records the corresponding current values. This procedure can be easily performed, as it is the basis for further and more complicate acquisitions.
If a measurement versus time is requested, as in enzymic or cellular applications, then a particular algorithm runs.
Referring to the above the program in this case constitutes the main portion of the system, as it performs the work that specific electronics usually does.
The pH or redox detection method implies the acquisition of a characteristic curve (current signal versus bias voltage); then the program automatically computes the bias potential corresponding to the inflection point, which s a good indication of the actual surface potential. Once the Vbias is known, the system is biased at the inflection point
corresponding to the middle of a characteristic region which can be approximated by a straight line, as we have already proved (Adami et al., 1992b). Due to this fact, if a small surface potential variation occurs, the characteristic curve shift can be regarded as a line shift; then the current value corresponding to the previous bias voltage can be linearly related to the surface potential change, as it can be, first of all, linearly related to the transducer sensitivity. This argumentation shows that quantitative measurements can be performed by biasing the system at a given voltage and by simply reading the current values during time; each value can be converted into the corresponding pH or redox units by a linear formula depending upon the chip sensitivity and the characteristic curve slope at the inflection point. The program essentially implements the described procedure; in particular, a complete characteristic curve is acquired and data are stored in memory; then a digital filtering
procedure, based on the Savitsky-Golay algorithm (Press et al., 1990) applies; finally data are fitted with a particular theoretical formula, and the bias potential at which the inflection point occurs is computed. The procedure is time consuming, but in most of cases it is applied only once; in fact the system is biased at the value just found, and the current signal is then acquired continuously, and related to the surface potential as previously stated. This procedure allows an information detection every fraction of a second; hence data acquisition can be performed to monitor fast pH or redox changes. The described method can be applied only when no impedance variation among the electrodes arises during the measurement; this can happen, for instance, if a membrane is pressed against the transducer.
The remaining blocks of Fig. 21 just show the described program procedures: inflection point calculation, consequent system biasing, and finally acquisition in time.
With regard to dialogue with the electronics, we utilize an interrupt-based structure; this guarantees faster
communication from/to the circuits, and allows a fast data acquisition. Basically, an interrupt occurs every time a data conversion is performed, the control then passes to a specific routine which acquires the data and resets the program.
When a communication with the flowing circuit (pump on/oft) is requested during the acquisition, proper procedures interact with the hardware.
The actual temperature monitoring is also possible, by an additional AD related software.
The system can be equipped with a variety of measuring chambers; the purpose is to allow different experiments with the same device.
Among the chambers already realized, first of all is a standard static cell able to contain a solution in contact with the transducer; the system is depicted in Fig. 22. The solution is confined by an O-ring pressed against the chip, and the two electrodes are dipped from the top. This chamber is the simplest example, and it is useful for standard measurements; in addition it can be utilized for experiments on the transducer, as it can be thermostated and generally guarantees the maintaining of conditions constant in time.
Fig. 23 shows a flow cell suitable for enzymic applications; the solution flows through a reaction chamber where a membrane can be easily located; for instance, the membrane could be a PALL Immunodyne™ for the immobilization of proteins (enzymes or antibodies).
The picture also shows the transducer and the related light emitting diode; again the measuring volume is delimited by a small O-ring, and a Pt wire is placed just in the proximity of the sensitive surface; the reference electrode can be located in the outlet channel. The membrane can also be placed in the small volume near the chip, but in this case the eventual remotion implies opening the entire chamber.
The chamber above described has been utilized for
measurements of an enzymatic reaction catalyzed by alcohol dehydrogenase from yeast (YADH) (Adami et al., 1993b). Fig. 9 shows a typical acquisition for the determination of enzyme concentration; the signal decay is due to the H+ production after the reaction took place, yielding to an acidification of the solution in contact with the transducer. Curves as the previous one allow the calculation of calibration curves, as the one shown in Fig. 10, which evidence the good sensitivity of the system, since it was possible to
discriminate enzyme concentrations of up to 10-10M.
Encouraged by these results, we are trying to use enzymes like YADH, or a redox enzyme like Horse Radish Peroxidase (HRP) for immunoenzymatic assays: the enzyme is regarded as a label to detect the reaction between antibodies and antigens. By using a redox enzyme the big advantage in this
potentiometric measurements is the overcoming of buffering capacity problems, as the surface potential is affected by electrons instead of protons; in order to create a surface charge layer just on the transducer, we evaporated a metal spot on the silicon nitride. Practically, we are using a competitive method and a configuration based on (monoclonal) antibodies entrapped in a membrane; the competition appears between the free antigen and the antigen bound to the HRP molecules. As soon as the enzyme substrates are injected into the reaction chamber, we obtain a signal variation responsive to the immune complex formation. A sketch of the described method is visible in Fig. 24A. A typical signal variation after the immunocomplex formation is shown in Fig. 24B.
Another exciting application available to this system concerns measurements of extracellular acidification; hence, a specific micro-volume flow chamber able to contain a cell population has been designed and realised and is visible in Fig. 25. One important feature of the chamber is the
possibility to insert glass coverslips, routinely utilized by biologists; other methods imply fixing the cells inside a porous membrane, but this method is not as standardized as the usage of coverslips. Another flexibility offered by our measuring chamber is the possibility to vary the microvolume, i.e. the distance between the transducer and the cells; this is accomplished by using different teflon spacers in between. The set-up mechanism is very easy: the user should simply take the coverslip containing living cells and insert it into the appropriate guides visible in the figure; then one should insert the desired spacer (usually ranging between 50 and 400 μm) and close the chamber. It is quite necessary to thermostat the system a few minutes before starting this set-up procedure, in order to offer a good thermal condition to the cell population, and to start the medium flow
immediately after the set-up is complete. In usual
applications the entire procedure takes a few seconds to be carried out, hence no particular perturbation is caused to the cells, and quantitative measurements are possible.
The usual experiment consists of several flow on/flow off steps and in the consequent signal acquisition; when the flow is stopped, acidification effects are detectable. A typical result is shown in Fig. 14; data are relative to a CHO confluent cell population, at 37°C and utilizing a standard F12 culture medium without hydrogen carbonate, in order to minimize the buffering capacity effects.
In a recent study carried out on 3T6 cells (Gavazzo et al, 1993) we have been able to estimate the number of protons derived from the acidification effects, quantized in 6.5×110 protons per cell per second. In addition a detailed evaluation of the extracellular acidification as related to the cells viability has been performed.
A more recent application makes use of microorganisms; in this case and in many other applications it is preferable to have a quantitative method to take into account eventual noise or possible artefacts deriving from the particular biological setup; tins can be accomplished using a
differential scheme. The solution must utilize the above described electronics for both signals, in order to ensure the same electrical characteristics and the same gains for the two measurements. A good solution consists of a
digital-driven switch to alternatively select the two signals. A related chamber has been designed and is shown in Fig. 26; two chips are fixed in the lower side of the chamber and the flow, circuit sequentially contact both the
transducers; a reaction chamber (which can be a commercial membrane-containing teflon system) can be easily connected and removed in the area between the two measuring areas; in this case the first chip gives a signal related to the solution, while the second one measures the effects of the reaction. As the chips are very close to each other and fixed in the very same environment and with the same
connections, the eventually present noise should affect both chips in the same way, hence a differential measurement should eliminate noise effects . The two chips used in this configuration must be carefully chosen, and must present electrical properties as close as possible (mainly drift and sensitivity), in order to have the same response in time. Anyway, even by utilizing a geometrically modified chip, a similar chamber can be designed using a single LAPS, so reducing the problems connected to the different features.
Regarding the transducers, several studies have been
performed in order to optimize the operating features; it has been shown that the frequency and the wavelength of the emanating light of the Light Emitting Diode (LED) used for illumination of the LAPS chip influences the sensitivity and stability of the response .
The optimization of the PAB system operational parameters lead to a significant increase in the signal stability, yielding a drift always lower than 0.01 pH per minute. The sensitivity of the system with such drift was approximately 52 mV per unit pH.
Some modifications of the native device are possible, and will be further investigated; in particular, as we have already shown (Adami et al., 1992a), the same signal
generation technique of LAPS can be utilized in conjunction with different ion sensitive electrodes, yielding to a multi parameter detection within the same device; in addition we proved that in principle a standard MOS transistor can be utilized instead of the LAPS chip, by inducing an alternating current through a time varying small signal, and connecting some ion selective electrode; in addition to the previously mentioned advantage, here it is possible to think also to some integration of the biosensor, as standard processes are being utilized.
In conclusion, the invention provides a novel Potentiometric Alternating Biosensor (PAB) capable of functioning as an integrated system under different conditions and of being utilized with a variety of biological sensing elements for the analysis of a wide range of biological samples. The transducing element is the so-called LAPS which actually might be regarded as one of the most reliable transducers used in the biosensor engineering.
The structural physical parameters of transducers of this type could be considered already optimized for their specific applications due to the availability of the information arising from the sound knowledge of the modem semiconductor science .
The data acquisition and the control systems of the PAB were completely re-designed to better adapt the electronics to the general requirements of the biosensor and of the interfacing. These sub-systems consist of several blocks interfaced by means of AD/DA converters to a usual IBM PC or compatible. The set-up of the electronic cards has been specifically studied in order to minimize the noise and maximize the operation speed. Another important feature is the
utilization of a separate "measuring box!' that incorporates all the necessary components and can be easily interfaced with the available PC. This implies that, instead of purchasing a dedicated processing unit, a connection with the conventional PC is sufficient.
A completely original software package has been written for the proper system control, for signal measuring and for data archivation and visualization. The software runs on a DOS operating system, and allows the user to manage different types of signal acquisition. The main feature of the low- level software is the interrupt-based structure which has confirmed to be the most reliable and to allow a good computer interfacing independently on the system clock.
The fact that the flow circuit of the PAB system is
controlled by the same computer unit permits to automate completely the operations with flow-through chambers, especially the start/stop operations, by means of respective electronic interfaces.
Different measuring chambers can be connected to the same device rendering the system suitable for numerous
applications: some chambers for standard solution
measurements, a mini-volume chamber for enzymic applications, a micro-volume flow-through chamber for cellular experiments. As an example, the latter has a completely new flow circuit which allows the realization of controlled and variable micro volumes, simply by inserting different Teflon spacers.
The PAB system can be thermostated by means of a purpose-designed temperature controlling system, based on a Peltier cell, which has an elevated precision and reliability.
A transducer whose surface is sensitive to many types of ion, and not only to hydrogen as in the case of silicon nitride transducers, may be used, for example, a gallium arsenide-based transducer; in addition to the GaAs property of a very high charge mobility, in fact, it is possible to build extremely sensitive layers by LB technology.
As it has been noticed previously, one of the advantages of the PAB lies in the possibility of uniting the pH-sensitive and redox-sensitive elements on a single sensor chip.
Moreover, the potentiality of utilizing several LEDs at different wavelengths addressing regions covered by
biological sensing elements with various specificity (e.g. different enzymes) renders the PAB system extremely
interesting for combined assays of complex samples.
Silicon Transducer Redox Potential Biosensor The system is shown generally in Figure 27. The transducer is essentially a heterostructure made of silicon ("n"- or "p"- type), silicon dioxide and silicon nitride (Sartore et al, 1992a; Sartore et al 1992b; Bousse et al. 1994). The insulator is pH-sensitive , due to the proton binding capacity of its groups (essentially Si-O and Si-NH2) over a large pH range (2-12), with an theoretical Nernstian response (if hysteresis and drift phenomena are not considered).
Redox potential measurements can be obtained by the
deposition of metal layers over the silicon nitride surface; when the solution in contact with the silicon structure contains a redox pair, the metal changes its potential toward a value stated by the Nerst equation and determined by the ratio of the concentrations of the solution species (Bard et al, 1980) . This process is due to the initial differences in the electron affinities of the two phases that are put in contact, the meal layer and the electrolyte solution; from this interfacial discontinuity an exchange of electrons between metal and solution originates and continues until the resulting potential change reaches the equilibrium point, i.e. the electron affinities of the two conductive phases are equal. So the metal, in some way, adequates its
characteristics to the solution and reflects the peculiarity of the liquid phase. By our configuration we measure a signal that responds to changes in redox potential of the electrolyte .
When the sinusoidally-modulated LED illuminates the back-side of this modified structure, it produces an alternating photocurrent: its shape depends on the ratio between the metal layer size and the light spot size. If the light source illuminate both a silicon nitride region and a metal region, a biphasic response is obtained, due to the different surface potentials for the two zones (see Fig. 28, curve B). In this case, an infrared LED illuminates an area that is, approximately, in an equal percent, silicon nitride and metal layer in contact with the electrolyte solution. The liquid phase contains ImM Potassium Ferricyanide, ImM Potassium Ferrocyanide and 100mM sodium citrate buffer, pH 6. The A and C curves of Fig. 28 represent the two "extreme
conditions" of the situation above described: the A curve is obtained with a metal spot that covers completely the nitride surface; the C curve, vice versa, represents the situation in which only the silicon nitride is in contact with the electrolyte. By a comparison of the three curves, it is possible to say that the first inflection point of the biphasic response is due to the redox potential while the second one depends upon the pH of the electrolyte solution. So, by changing the ratio in the redox pair concentration, an almost Nernstian shift along the bias potential axis is obtained, but this shift affects only the first portion of the bi-phasic response as depicted in Fig. 29a. Vice versa, by changing the H of the solution containing the redox pair, only the second portion of the characteristic shifts, as represented in Fig. 29b.
This phenomenon has been investigated, by depositing a metal spot of different sizes (2-5mm in diameter) over an
illuminated area of about 5mm in diameter; the used metal are Chromium (a pad about 500-1000 A thick, evaporated by Balzers MlO metal evaporator) and Gold (a pad about 1000 A thick, evaporated by the same instrument, and on a previous thin layer of Chromium in order to increase the adhesion) . A good sensitivity can be obtained in a structure with a metal spot of about 5mm in diameter; in Fig. 30 the calibration curve of this system is shown, obtained with the metals above
indicated.
As in the case of silicon nitride sensing surfaces, also when metal spots are used the inflection point of the I-V
characteristic curve is indicative of the actual surface potential, as already reported in Adami et al 1994A and B.
The sensitivity and the linearity of the response are, in both cases, in good agreement with the Nernst theory and permit to establish a proportionality between the output signal and a parameter, pFe, defined as Log
( [FeII) 1/[Fe(III)] ) . With the system above described it is possible to evaluate the concentration of enzymes catalyzing redox reactions, such as HRP (Horse Radish Peroxidase, purchased from Sigma) that oxidizes TMB (Tetramethylbenzidine, purchased from Sigma) reducing H2O2; the reactions taking place are as follows:
Figure imgf000060_0001
The second reaction is spontaneously occurring in the presence of the redox pair. The pFe depends, as defined, on the conversion of Fe(II) into Fe(III); a theoretical
prediction of the pFe behaviour can be estimated taking into account the concentrations and the buffering capacity of the solutions; the theoretical approach already published in the case of ADH experiments by Adami et al 1994a can be full applied also in this case.
The pFe dependence on the conversation of Fe(II) into Fe(III) is shown in Fig. 31; in the interval between +0.3 and -0.3 of the pFe axis the curve can be well approximated with a straight line: its slope is the proportionality factor between the pFe variations and n, the number of nanomoles of Fe transformed. Moreover, the enzymatic activity can be expressed in units defined as micromoles of TMB oxidized per minute, equivalent to half of the micromoles of Fe converted. So, it is possible to correlate the output signal of the sensor with the enzymatic concentration, if the specific enzymatic activity is known (Bousse et al, 1992; Adami et al 1994b) .
The calibration curve for the HRP enzyme in solution has been obtained, as shown in Fig. 32: the lowest detection limit in enzyme concentration is 5×10-11 M. The enzymatic activity determined with the procedure above described is in good agreement with the value coming from usual spectrophotometric assays. Considering the purpose of an immunoenzymatic application, it is essential to determine the activity of the immobilized enzyme. The same configuration as depicted in Fig. 27 has been used, by simply inserting in the measuring chamber a membrane (Pall Biodyne B) of about 4x4 mm on which the enzyme has been immobilized. The relative calibration curve has been obtained, as shown in Fig. 33: the lowest detection limit, in this case, is of about 2.5×10-8 M. The loss of activity due to the immobilization is considerable but does not prevent from the applicability of this system to an immunoenzymatic assay.
Also enzymes which do not produce directly a redox potential change in the solution can be used with this configuration; as an example, we tried to determine ADH activity
(immobilized on a membrane) by using an auxiliary enzyme, Diaphorase, which can reduce Fe(III) in the presence of NADH
(all the chemicals have been purchased from Sigma):
Figure imgf000061_0001
An example of this assay, Fig. 34 shows the transducer output signal vs. time; this is the basic data by which calibration curves are determined, and shows the signal variation during the enzymatic reaction. The detection of the ADH activity in a redox configuration, presents several advantages with respect to a pH sensitive configuration, namely:
i) the membrane buffer capacity does not interfere with the redox measurements;
ii) a high buffer concentration can be used, stabilizing the enzyme activity;
iii) thermodynamic equilibrium is not reached because of coenzyme recycling. 2 Dimensional PAB System
A block diagram of a bidimensional PAB system is shown in Figure 1. Based on the existing silicon sensors technology, a significant upgrading of the biosensor system, named PAB (Potentiometric Alternating Biosensor) , is here presented. The transducer consists of a light-addressable silicon chip, which provides regions properly modified and functionalized to yield sensitivity to either pH or redox potentials. The system is electronically controlled and driven, and it is connected to a common personal Computer; ad hoc software drives an array of light sources, allowing 2D signal
acquisition and processing up to the production of response images or 2D histograms. The system is here utilized with a specific measuring chamber to monitor biological events related to in vivo cell metabolism. Several other chambers have been designed to monitor different phenomena such as enzymatic activity (Y-ADH, Urease, HRP) and antigen-antibody binding.
The system can provide multiparameter information related to the specific distinct local modifications made on the sensing surface of the transducer.
In such a bidimensional system we need to define the smallest sensing region of the transducer, or the minimum distance between adjacent acquisition spots; this value is a
measurement of the spatial resolution for a given transducer.
The quantity which strongly affects the choice of dimensions in the design of this 2D system is the minority carriers diffusion length, defined as:
L=√Dτ where D is the diffusion coefficient and τ is the minority carriers lifetime; in particular depends upon the chip used and strongly varies from wafer to wafer (Bousse et al 1994 and Sartore et al 1993). Depending on the values of these parameters, the charge carriers can diffuse rapidly and can reach positions quite far from the generation area; this effect is undesired, as it affects the spatial resolution.
From a more practical point of view, in the design of the 2D system it should be simply avoided that the carriers
generated inside the silicon diffuse to the adjacent sensing spots; as the diffusion in the silicon bulk can be considered isotropic, the carriers reaching the space-charge region just under the insulator are only a fraction of these generated in the bulk; the remaining portion of carriers diffuse to neighbour areas; referring to the symbols defined in Fig. 36, it is necessary to satisfy the condition: Wspots > Wsi.
Practically it is suitable to have Wspots = 2Wsi at least.
If the above condition is not verified, then the signal acquired in a certain area is inevitably conditioned by the surface potential of adjacent zones; in the case of a multisensor, this can mean for example that a pH measurement can be affected by the redox potential of an area close to the pH sensitive one.
A preliminary setup was based on a single optical fibre scanned along the chip backside; at each step position a new value was acquired; this setup has the advantage of a uniform optical excitation of the chip, as the light source maintains the same features even if moved across the transducer; every acquired signal can be directly related to the other ones because all of them derive from the very same light
excitation. Unfortunately this setup has the big
disadvantage to need an XY mover, which can cause
displacement errors and increase the total cost of the device; this solution also requires specific hardware and software to drive at least two movers.
The final version of the system is based on an optical fibre array fixed in the close proximity of the transducer
backside; each fibre is addressed at a given time interval and the corresponding surface potential is measured. This solution does not guarantee "a priori" identical light excitation at different locations, but avoids movable parts and allows a precise detection in a fixed number of sites; moreover, the very small differences in the optical power from point to point does not affect the quality of
measurements, because the acquired signals are easily normalized. Our actual setup utilizes a 5x5 fibre array for an active transducer area of lcm x lcm.
The normalization procedure of the acquired signals, in addition to recover from different optical fibre positioning, is a very efficient method to avoid a lot of artefacts. In fact, as a potentiostat is utilized, as visible in Fig. 35, the system is composed by three electrodes; a (Pt) counter electrode, a (Calomel) reference electrode and the chip surface, which is considered the working electrode; this setup can be easily modelled with a couple of impedances, as it is very well explained in Bard et al, 1980, and depicted in Fig. 37A. A significative number of artefacts during a measurement can be due to an undesired variation of such impedances, which can be due to temperature oscillations, to the medium conductivity, or even to more macroscopic reasons, such as bubbles in the hydraulic circuit.
As a direct consequence of electrode impedance variations, the I-V characteristic curve of the transducer (whose inflection point is useful for the measurements, as already reported in Adami et al, 1994B) is globally distorted and varies in amplitude, as depicted in Fig. 37B. The same effect takes place when the light source is positioned at different distances from the chip backside, and again when considering the I-V curves of single sensing spots lighted by fibres positioned in different locations with respect to the chip, or well positioned but driven by LEDs with different optical powers.
The normalization method here utilized considers the signal nearby the inflection point of the characteristic curve and the maximum amplitude of it, i.e. the value corresponding to biasing the device in inversion of majority carriers. Then any "true" data is the ratio between the inflection point value and the maximum one; this represents an easy way to overcome the above mentioned underived effects.
Measuring chamber
The design of a suitable measuring chamber for the 2D system combines the need of a relatively wide sensing area (with different sensing spots) and the corresponding backside positioning of the optical fibres array; in this sense a big effort should be spent in order to obtain a perfect alignment between the sensing surface spots and the corresponding fibres .
For biosensor purposes it is necessary to connect some biological sensitive layer to the chemical transducer, the chamber design allows the positioning of membranes in the close proximity of the sensing surface, and a cell population can either be grown directly on the chip or on a cover slip fixed in front of the transducer, at a very small distance (usually 50-200μm) . With these peculiarities the chamber directly allows biosensing applications, such as enzymatic or cellular measurements .
A schematic of the flow chamber is presented Fig. 38.
The chamber is made of two parts (indicated A and B) to be connected after the eventual insertion of a biological layer in the microvolume region F. A medium flow comes from inlet C and fills the measuring chamber F, allowing a
multidetection through the sensing sites H, formed onto the transducer surface G. Before leaving the chamber through the outlet E, the medium flow enters the area indicated with D, where both a counter and a reference electrode can be placed (alternatively the counter electrode can be accommodated even within the inlet connection). An array of optical fibres I is fixed at a predetermined distance from the transducer backside by an attachment L; this array is connected on the opposite side to a corresponding array of infrared LEDs (Light Emitting Diodes), driven by a multiplexing hardware.
Hardware and software
The driving electronics of the 2D system requires a single light modulating source and two 4-bit (or one 8-bit) digital ports.
Usually 10 KHz, full positive, sine or square wave is utilized to generate carriers into the silicon by light excitation; each fibre is connected to a LED which is addressed digitally. The modulating signal is fed into a switching transistor through an operational amplifier, in order to ensure the proper current to the LEDs without affecting the local oscillator.
In a possible scheme of the 2D driver, (Fig. 39), the driving transistor is connected to a series of digitally-controlled switches, in order to address a single LED at a time.
The selection of a given light emitting diode is performed by sending the corresponding digital address to a couple of multiplexers, namely the CD 4028. Up to 10 rows and 10 columns can be addressed by the proposed scheme, which can be connected either to a serial port (by an appropriate
interface) or to a DIO (Digital Input Output) card of a common Personal Computer.
The driving software can be easily integrated within the single-channel acquisition program already described (Adami et al, 1994B); in fact it is necessary to repeat acquisitions at the different sensing locations.
Before acquiring the signal, the program sends two 4-bit words to the hardware circuitry in order to select a single LED, so enabling a given sensing spot; then the corresponding data is acquired, and the process re-starts. In order to show the application possibilities of the 2D system here described, a bidimensional pattern has been acquired at different measuring conditions.
For a sake of clarity the experiment makes use of sample solutions injected into the measuring chamber, without any addition of biologically-active components; in this fashion the system is utilized as a chemical sensor; of course biosensors applications are very easy to be performed, as already stated.
A Si/SiO2/Si3N4 chip has been partially covered with gold in two distinct areas; the gold was evaporated onto the chip by means of a Balzer M10 metal evaporator; for a better results, a thin layer of Chromium was previously deposited onto the silicon nitride. The chip has been addressed by 4 optical fibres focused on the transducer backside at locations corresponding to the front side sensing regions; the
surfacial pattern of the chip is visible in Fig. 40.
The chip was inserted into the measuring chamber, and then the measuring solutions were injected. Solutions containing different ratios of Potassium Ferrocyanide and Potassium Ferricyanide (namely 1:100, 1:10, 1:1, 10:1, 100:1) at pH 7 and 9 have been prepared.
Fig. 41 shows a typical acquisition when a solution
containing the redox pair in a ratio of 1:10 in concentration of pH 7 is injected into the measuring chamber; the "bar chart" directly corresponds to the chip shape shown in the schematics of Fig. 40, i.e. the higher bars, whose
coordinates are (1,1) and (2,2), are relative to the gold spots (redox measurements) and the lower bars, whose
coordinates are (1,2) and (2,1), refer to the Si3N4 sites (pH measurements).
The data contained in Fig. 41 shows the possibility to obtain bidimensional patterns relative to different sensitivities; one can very either the pH or the redox compounds ratio, causing a variation only of the specific local signal.
In order to validate the above sentence, we introduced into the flow chamber different solutions and measured the corresponding output signals.
Fig. 42 is a collection of 2D patterns of the same type of the previous Fig. 41, varying both pH and redox pair
concentrations; in particular ten 2D patterns are shown and indicated with letters A through L; patterns A to E refer to pH 7 solutions at redox pair ratios range from 1:100 to 100:1, while patterns F to L refer to pH 9 solutions at the same redox pair ratios, precisely ranging back from 100:1 to 1:100; it is possible to observe a very specific variation corresponding to a change in the measuring solution.
Biosensor for In Vitro Drug Screening: Toxicity Testing in Cancerous Hepatocytes
The Potentiometric Alternating Biosensor (PAB) system has been utilized to monitor the effects of two antineoplastic drugs, Cytosine arabinoside and Mitoxantrone, on two distinct cell lines, namely on established cell line (3T6 mouse fibroblast) in different phases of cell cycle and primary culture (rat hepatocytes) in resting GO cells.
An ad hoc microvolume flow chamber has been designed and produced; the chamber is equipped with inlet and outlet circuits and with a fixed transducer (Si/SiO2/Si3N4 chip), facing a cover slip on which cells grow; the transducer allows monitoring of pH in the microenvironment where the cells are placed; the system is used in the pH-sensitive configuration and a single measuring spot has been used for the experiments, warranting an accurate determination of the change in the extracellular acidification rate resulting from drug administration.
New insights into normal vs. abnormal cell growth and differentiation and into in vitro drug toxicity and efficacy can be obtained with the presented system, towards effective human cancer treatment. The effect of increasing drug concentrations on cellular metabolism is here compared with the results coming from conventional tests (optical
microscopy, Neutral Red and Trypan Blue assays).
As regards the experiments on 3T6 cell line, our goal was to verify, by means of PAB, the possibility to monitor the action of an antimetabolite on a specific cell cycle phase, which is the one of DNA synthesis (S phase). Therefore, we have chosen Cytosine arabinoside (l--D-arabinofuranosyl cytosine - Ara-C), which is, in the form of triphosphate nucleotide, a powerful inhibitor of DNA polymerase.
As regards the experiments of hepatocytes and hepatomas, in order to choose the proper drug we have focused our efforts on chemotherapy against hepatocarcinomas, which includes drugs as mitoxantrone, adriamycin and cis-platinum.
Among these, we have chosen mitoxantrone for the following reasons:
1) it is used in primary liver tumours;
2) it is used in non-associative therapeutic protocols;
3) it presents one of the best therapeutic effectiveness;
4) it is one of the most used drugs in clinics;
5) it is administered in bolus, way of administration for which a better comparison with in vitro exposition appears possible;
6) it has a very short half-life period, therefore it allows short duration exposures;
7) it does not have a specific action on a single phase of the cellular cycle, even if proliferating cells are more sensitive than the quiescent ones.
For both drugs, we have compared PAB results with those obtained with classical cytotoxicity tests (Trypan blue, Neutral Red and optical microscopy observations). Materials and methods
Cellular targets
The cellular line chosen for experiment with Ara-C is 3T6 (Swiss albino mouse embryo, fibroblast) purchased from ATCC (American Type Culture Collection) . Conventional culture procedures have been followed using culture medium RPMI 1640 (Sigma).
The cells have been plated onto a glass support. In order to avoid detachment of cells due to shear stress of the flow in the measuring chamber, the glass supports have been
previously treated with collagen (Sigma C3511).
The mitoxantrone toxicity tests, we have chosen primary cultures of rat hepatocytes because they are easy to obtain and, like all hepatocytes, they maintain, in the first hours in vitro, their metabolic skills practically unchanged with respect to the in vivo situation. Hepatocytes were isolated from liver of Sprague-Dawley albino rats (200-250g) by in situ collagenase perfusion according to Williams (1977).
Isolated cells were suspended in Williams E. Medium (WME), supplemented with 10% fetal bovine serum and genamicin (50 g/ml) at the concentration at 5×105 hepatocytes/ml. Aliquots of this suspension were plated as follows: a) 1×105 cells were plated on the glass support (coated with collagen) for the measurement with PAB; b) 6×105 cells were plated on 35 mm dis es, coated with collagen, for conventional toxicity tests.
Dynamics and kinetics of Ara-C
Ara-C is a synthetic nucleotide, which differs from the natural ones (cytidine and deoxycytidine) for substitution of ribose and deoxyribose with arabinose.
It has a cytotoxic effect due to a phase-specific activity on the S phase of cell cycle.
It is quickly metabolised and it is not effective if
administered per os (only 20% of the dose is absorbed along the gastrointestinal tract). Constant levels of drug can be maintained by means of continuous i.v. administration.
Clinical use of Ara-C
Ara-C is principally used to induce regression of the acute proliferation hemopaties of the granulocytic series in the adult. The secondary application is in the other
proliferative forms of the leucocytes series in both the adult and child.
Dynamics and kinetics of mitoxantrone
Mitoxantrone is a potent inhibitor of RNA and DNA synthesis. It intercalates on DNA, inducing cross links intra- and inter-strand, especially on GC base pairs. It interacts with the cellular membranes changing their functions.
It is scarcely absorbed per os, the recommended
administration route is the intravenous one. According to the autopic observations, the largest residual amount of mitoxantrone can be found in liver. In fact, in addition to renal excretion, the hepatobiliar pathway is largely involved in drug removal. There are evidences that the drug undergoes hepatic metabolisation; in fact four different metabolites have been isolated from urine.
Clinical use of mitoxantrone
The drug has been used since 1984 for the treatment of hepatocarcinomas, the dose of intravenous administration being 12-14 mg/m2 in bolo. Since 1986, also the
intraarterious administration has been employed at the dose of 6-14 mg/m2. A large bibliography exists on administration modes and side effects. Estimation of the appropriate dose for "in vitro" testing
The rationale on which we based our choice of mitoxantrone doses is the following: average clinical dose: 12 mg/m2
average adult body area: 1.72-1.80 m2
total dose of drug: 20 mg/individual
Since 20 mg are often administered in bolo into the hepatic artery, their distribution could be considered in about 500 ml of blood (hepatic artery capacity - 7% of the total);
since in vitro there is no binding to blood elements, we could approximate this situation to a distribution in about 1 litre of plasma (culture medium), therefore a dose in vitro approximable to the therapeutic one could be 20 mg/l (= 20 g/ml) . Therefore, the following concentrations below and above the therapeutic one have been chosen:
2.5 - 5 - 10 - 20. - 40 - 80 g/ml
Since the half-life of distribution in vivo is about 30 minutes and considering the enhanced "accessibility" of the isolated hepatocytes, a reasonable exposure time was
estimated to be 30 minutes.
Since the toxic effects of the interaction with DNA,
responsible for the therapeutic action, can become effective within a certain time after exposure, it seems feasible to verify the response after a period of "recovery" following the exposure; it was decided to evaluate the effect 20 hours after stopping the incubation with mitoxantrone.
The assay times are the following:
30 minutes of incubation with the drug;
30 minutes of incubation +20 hours in absence of drug.
As regards Ara-C, for the evaluation of the right dose, we did not follow the scheme applied for mitoxantrone, since the cellular systems involved in the two cases are not comparable at all . The choice of 3T6 cells was due to the need of a well stabilised line on which to make preliminary studies. Therefore, we have tested several doses in the 0-300 g/ml range and several exposure times.
Toxicity test for the comparison with PAB system
The more used and better known cytotoxicity tests were chosen to be compared with PAB measurements:
1) optical microscopy evaluation of the morphological alterations induced in the culture by chemotherapeutic treatment. The observation of the culture, if carried out by well-trained personnel, correlates in a satisfactory way with the more sensitive evaluations of toxicity;
2) Trypan Blue exclusion test, which evidences macroscopic damages in the cellular membranes;
3) Neutral Red test (Babach and Borenfreund, 1987), which is based on the accumulation of the supra vital dye in the lysosomes.
Results
Conventional tests
The tests have been carried out on primary culture of rat hepatocytes and on HEPG2 cells, with exposure times of 30 minutes and of 30 minutes + 20 hours of incubation in absence of drug, with doses of 2.5 - 5 - 10 - 20 - 40 - 80 g/ml for hepatocytes and 5 - 10 - 20 g/ml for HEPG2 cells (dosage selected after the tests on hepatocytes) .
The results obtained are the following:
A) Optical Microscopy
Exposure time: 30 minutes, cell line: rat hepatocytes; the cells showed a continuous distribution and a polygonal form similar to the one of the controls at doses 2.5 - 5 - 10 and 20 g/ml, even if a certain detachment of cells was observed with increasing doses. At 40 g/ml, the cells were losing the polygonal form and looked damaged, while at 80 g/ml the appearance of the cells was better: the phenomenon could be attributed to an edema of the cells caused by loss of normal permeability.
Exposure time: 30 minutes + 20 hours, cell line: rat
hepatocytes; the comparison with the control cultures, which maintained the polygonal form, resulted to be strongly disadvantageous for all the doses. The hepatocytes treated with 2.5 - 5 - 10 g/ml doses maintained the polygonal form even if not on a continuous layer and many detached cells were observed. At 20 g/ml the cells assumed a round-shaped aspect; at 40 g/ml they were nearly completely detached and therefore dead for more that 50%. At 80 g/ml, even if granulose and without evident margins, cells were still adherent to the support.
Exposure time: 30 minutes, cell line: HEPG2; cells did not reach confluence, not even in checking dishes; cell
morphology was comparable at all doses even if, increasing the dose (especially at 20 g/ml), we could observe several detached cells, especially among the piled groups (major exposure to chemotherapeutic).
Exposure time: 30 minutes + 20 hours, cell line: HEPG2; all the cultures appeared suffering compared to the control; we have observed a maximum cell detachment at a dose of 10 g/ml.
At all doses, the cells presented several cytoplasmic vacuoles and extended cytoplasm.
Trypan Blue test
The estimated percentages of hepatocytes viability are reported in Table I. From the analysis of the viability data, it has been possible to obtain the LC50 (Lethal concentration for 50% of
hepatocytes) values reported in Table II.
At 30 min. + 20 h, the still viable hepatocytes showed, at all doses, to have incorporated mitoxantrone, in order to metabolise it.
The results of the Trypan Blue exclusion test for hepatoma cell line expired to mitoxantrone are reported in Table III.
From data analysis, we obtained significantly lower values for the 50% lethal concentration of mitoxantrone, as
reporting in Table IV.
Neutral Red test
This test allows to directly calculate the concentration of chemotherapeutic able to kill 50% of the exposed cells. This concentration is expressed as NR50. The results obtained with mitoxantrone are shown in Table V.
Extracellular acidification
The measurements with PAB are based on the monitoring of extracellular acidification; the relationship between production of acidic metabolites and rate of extracellular acidification is the following:
Figure imgf000075_0001
where dn/dt is the generation rate of H+ ions, due to acidic dissociation of excreted metabolites, ß is the buffering capacity of the examined solution and V is the volume of the chamber; in order to improve the sensitivity of the method, we can lower both volume and buffering capacity of the solution. To monitor extracellular acidification of 3T6, rat
hepatocytes and HEPG2 cells, we have used a 251-reaction chamber and a low buffered solution (phosphate buffer 1 mM, pH 7.4, added with NaCl to a final concentration of 100 mM): the short times (about 10 minutes of each measurement) of cell exposure to this non-specific medium practically does not influence their response.
We have monitored the pH variations in the chamber due to the stopping of medium entry for 2 minutes : by means of the following equation it is possible to transform the transducer signal (dVout/dt) in acidification rate:
Figure imgf000076_0001
where S is the sensitivity (50mV/pH, in our case) and a is the slope of the I-V characteristic curve of the transducer.
In Table VI, data are shown concerning acidification rates of 3T6 cells synchronised in S phase and exposed to increasing Ara-C concentrations for an incubation time of 5 hours.
The measurements with PAB have been performed immediately after treatment with the antineoplastic drug: the total assay time was 10 minutes for each dose.
Figure 43 shows the relationship between extracellular acidification of 3T6 and drug (Ara-C) dose: low doses give no evident effect on the cell population while, at doses greater than 40 μg/ml, the action of the antimetabolite Ara-C on the cells synchronized is evident and consists of a linear decrease in extracellular acidification rates.
In Table VII, data concerning the treatment of hepatocytes with mitoxantrone (30 mins + 20 h) are summarised: two series of measurements (acidification rates) are reported for each drug dose. A comparison between the data obtained with PAB and ones obtained with the Trypan Blue standard test is presented in Figure 44; the dose/effect trend observed with the PAB system is fully confirmed by the results obtained with Trypan Blue.
Figure 45 shows some of the acidification data acquired with PAB system on rat hepatocytes (30 min + 20 h treatment) .
Since the results obtained with rat hepatocytes and 3T6 fibroblast were very promising, we confirmed with PAB the results of conventional tests on HEPG2 cells treated with mitoxantrone.
Results are presented in Table VIII.
Discussion
The use of PAB system as a new tool in toxicological studies seems to be very promising.
The experiments with stabilized lines (3T6, HEPG2) and primary cultures (rat hepatocytes) have confirmed that with this biosensor it is possible to monitor the metabolic status of cells and to correlate small variations of this status with the action of specific agents.
A remarkable feature of PAB system is the possibility to monitor in continuum the effect of a drug on a cell
population, while other tests measure only the end point of an experiment (i.e. the death of cells).
The specific features of the new anticancer drug
mitoxantrone, for instance, could be better understood: this drug has been studied in primary cultures of hepatocytes obtained from rat, rabbit and humans (Richard et al, 1991); variability in the metabolic pattern between the different species was observed.
Further studies on human hepatocytes will be carried out with PAB to verify the obtained results can differentiate the interspecific variability of mitoxantrone metabolic pattern.
Figure imgf000079_0001
Figure imgf000079_0002
Figure imgf000080_0001
Figure imgf000080_0002
Figure imgf000081_0001
Figure imgf000081_0002
Figure imgf000082_0001
Figure imgf000082_0002
PAB System with Immobilized Enzyme Monolayer Sensor
The scheme of the PAB system is shown in figures 1, 8, 27 and 35. When a sinusoidally-modulated IR radiation lights the back side of the sensor chip a sinusoidal photocurrent is obtained, and its profile mainly depends on the chemical reaction which takes place in the measuring chamber.
When a Si/SiO2/Si3N4 chip is used in the PAB, the system is sensitive to pH variations. By means of evaporation of metal spots on the Si3N4 surface, it is possible to make the device sensitive to redox potential variations. Further strategies of modification of chip sensitive layer allow the usage of PAB system in different configurations.
Materials and monolayer preparation
The urease (Urea aminohydrolase, 61000 unit per gr.) from jack beans was purchased from Sigma and the 3-glycidoxypropyltriethoxysilane (GOPTS) from Aldrich. The monolayer of urease was formed spreading 0.2 ml of 1 mg/ml enzyme solution at the water-air interface of the Langmuir-Blodgett trough (MDT Corp-Russia) whose dimensions are
120×240×30 mm. Carbonate buffer (pH 8.6) was used as subphase.
The vacuum silanization of substrate with GOPTS
(glycidoxypropyltriethoxysilane) is based on the method developed by M. Malmquist et al, 1989. We activated both the glass and then the silicon nitride surface of the transducer in order to choose the best measuring configuration.
The enzyme monolayer was transferred onto the activated support by Langmuir-Schaefer technique. After the incubation at temperature of 4°C for 4 hours the monolayer was washed by water flow and dried with nitrogen. The obtained enzyme monolayer was characterized by measuring the thickness and the surface density by means of ellipsometric and
nanograviometric technique respectively. Characterization of monolayer
The surface density of the monolayer versus deposition pressures has been measured (see figure 46) and helped us in estimating the corresponding value of area per molecule. For a surface pressure in the range of 20-30 mN/m we found agreement between the experimental results and the
theoretical value obtained by evaluating the cross-section of the enzyme molecule from its molecular weight (15000-10000 Å2) .
In the same region of surface pressure the thickness of the monolayer measured was about 45 Å, that also corresponds, in order of magnitude, to the dimensions of the urease molecule. Therefore, both these measurements confirmed that the obtained monolayer was densely packed.
PAB measurements
In our measurements we used the pH sensitive configuration of PAB, with both the reaction chamber and the solution
thermostated at 37°C. First we monitored the activity with the urease in solution, in order to estimate the lowest enzyme concentration detectable by the system and the limit that we achieved was 10-10 M. Afterwards the activity of urease, immobilized over silanized support (glass or silicon nitride), was monitored under the following operative conditions: reaction volume = 25 μl; [UREA] = 100mM;
phosphate buffer 1 mM - pH 6.5; [NaCl] - 100 mM; surface pressure = 20 mN/m.
Figure 47 shows one of the curves acquired. When the pump is ON the products of the enzymic reaction flow away from the chamber and the dynamic equilibrium condition is reached: in this way the pH is nearly constant (first part of the curve). When the pump is OFF, the flow stops and the products of the reaction accumulate in a standing volume: this causes an increase of the pH (second part of the curve). In order to have an analytical correlation between the immobilized enzyme and the acquired curve, we evaluated the "initial alcalinization rates" as the slope of the regression line in the first relevant points of the enzymatic reaction curve (see Figure 47). By repeating several times the measurements, always in the same conditions, it was possible to evaluate the stability of the monolayer. In Figure 48 are shown the values of the initial alcalinization rate measured in repeated assays on the same support. The obtained results showed that the immobilized urease monolayer maintains the activity, although a small decrease was observed. This effect is probably due to two reasons: (a) the detaching of a small quantity of enzyme with the solution flows through the chamber; (b) the denaturation of a small amount of enzyme.
It is to be understood from the foregoing description that the present invention relates to all of the described aspects both taken alone and in any combination of described features and/or aspects. The appended claims are therefore to be construed in their broadest sense to cover individual aspects of the invention and combined aspects of the invention.
All references shown herein are to be considered as
incorporated by reference.
References: Temperature Control
M. Sartore, M.Adami and C.Nicolini, Biosensors and
Bioelectronics, 7-1, 57 (1992).
D.Hafeman, W.Parce and H.McConnell, Science, 240, 1182
(1988).
M.Adami, M. Sartore, E.Baldini, A.Rossi and C.Nicolini,
Sensors and Actuators B, 9, 25 (1992).
X.Zhu, E . Krochmann and J.Chen, Review of Scientific
Instruments, 63, 1999(1992).
G.Hostetter, C. Savant and R.Stefani, "Design of feedback control systems" 2nd ed., Saunders College Publishing, USA, 1989.
References: Potentiometric Alternating Biosensor
Adami, M. , Sartore, M. , Rapallo, A. and Nicolini, C. (1992a). Possible developments of a potentiometric biosensor. Sensors and Actuators B, 7, 343-346.
Adami, M. , Sartore, M. , Baldini, E., Rossi, A. and Nicolini, C. (1992b). New measuring principle for LAPS devices.
Sensors and Actuators B, 9, 25-3 1.
Adami, M. , Piras, L.,. Lanzi, M. , Fanigliulo, A., Vakula, S. and Nicolini; C. (1993b). Monitoring of enzymatic activity and quantitative measurements of substrates by means of a newly designed silicon-based potentiometric sensor. Sensors and Actuators B, in press.
Gavazzo, P., Paddeu, S., Sartore, M. and Nicolini, C. (1993) Study of the relationship between extracellular acidification and cell viability by a silicon based sensor. Sensors and Actuators, in press.
Hafeman, D.G., Parce, J.W. and McConnell, H.M. (1988). Light addressable potentiometric sensor for biochemical systems . Science, 240, 1182.
Press, W.H., Flannery, B.P., Teukolsky, S.A. and Vetterling, W.T. (1990). Numerical Recipes in C. The art of scientific computing, Cambridge University Press, Cambridge.
Sartore, M., Adami, M. and Nicolini, C. (1992a). Computer simulation and optimization of light addressable
potentiometric sensor. Biosensors and Bioelectronics, 7, 57-64.
References: Silicon Transducer Redox Potential Biosensor
Adami, M., Piras, L., Lanzi, M., Fanigliulo, A., Vakula, S. and Nicolini, C. (1994b). Monitoring of enzymatic activity and quantitative measurements of substrates by means of a newly designed silicon-based potentiometric sensor, Sensors and Actuators B, in press.
Antolini, F., Paddeu, S., Adami M., and Nicolini C. (1994). Langmuir-Blodgett film of Glutathione-S-Transferase: film characterization and its possible PAB application for pesticides detection, Biosensors and Bioelectronics,
submitted.
Bard, A.J., Faulkner, L.R., (1980). Elecrochemical methods. Fundamentals and Applications, Wiley International Edition, John Wiley & Sons, New York.
Bousse, L., Owicki, J.C. and Parce, J.W., (1992). Biosensors with microvolume reaction chambers, in Chemical Sensors Technology, 4, 145-166, Yamauchi S., Elsevier Science
Publishers.
Bousse, L., Mostarshed, S., Hafeman, D., Sartore, M., Adami M. and Nicolini, C. (1994). Investigation of carrier transport through silicon wafer by AC photocurrent
measurements. Journal of Applied Physics, 75 (8). Gavazzo P., Paddeu, S., Sartore, M. and Nicolini, C. (1994). Study of the relationship between extracellular acidification and cell viability by a silicon based sensor. Sensors and Actuators B, in press.
Hafeman, D., Parce, W. and McConnell, H. (1988), Light addressable potentiometric sensor for biochemical systems. Science, 240, 1182-1185.
Sartore, M., Adami, M., Nicolini, C. (1992a). Computer simulation and optimization of light-addressable
potentiometric sensor. Biosensors and Bioelectronics, 7, 57-64.
Sartore, M., Adami, M., Nicolini, C, Bousse, L. , Mostarshed, S. and Hafeman, D. (1992b). Minority carrier diffusion length effect on light-addressable potentiometric sensor (LAPS) devices. Sensors and Actuators A, 32, 431-436.
References: 2D PAB System
Adami, M., Piras, L. , Lanzi, M., Fanigliulo, A., Vakula S. and Nicolini, C. (1994A). Monitoring of enzymatic activity and quantitative measurements of substrates by means of a newly designed silicon-based potentiometric sensors, Sensors and Actuators B, in press.
Adami, M., Satore, M. and Nicolini, C. (1994B) . PAB: a newly designed potentiometric alternating biosensor system,
Biosensors and Bioelectronics, in press.
Bousse, L., Mostarshed, S., Hafeman, D., Sartore, M., Adami, M., Nicolini, C. (April 1994). Minority carriers diffusion length effects by AC photovoltage measurements. Journal of Applied Physics.
Bard, A.J., Faulkner, L.R. (1980). Electrochemical methods: fundamentals and applications, Wiley International Edition, John Wiley and Sons, New York. Bertilsson, L. and Liedberg, B. (1993). Infrared study of thiol monolayer assemblies on gold: preparation,
characterization and functionalization of mixed monolayers, Langmuir, 9,141-149.
Delia Ciana, L., Chinaglia, B., Adami, M., Satore, M.,
Nicolini, C. Improvements in Chemical-sensitive
semiconductor devices (CSSDs), European Patent No.
91113155.5, (1991).
Gavazzo, P., Paddau, S., Satore, M. and Nicolini, C. (1993). Study of the relationship between extra cellular
acidification and cell viability by a silicon based sensor. Sensors and Actuators, in press.
Lessin, S.R., Abraham, S.R. and Nicolini C. (1982).
Biophysical identification and sorting of high metastatic variants from B16 melanoma tumour. Cytometry, 2, 407.
Lundstrom, I., Svensson, C, Spetz, A., Sundgren, H. and Winquist, F. (1993) . From hydrogen sensors to olfactory images - twenty years with catalytic field-effect devices. Sensors & Actuators B, n. 13-14, 16-23.
Nicolini, C. (1975). The discrete phases of the cell cycle: autoradiographic, physical and chemical evidences, Journal National Cancer Inst., 55, 821.
Nicolini, C. (1976). The principles and methods of cell synchronization in cancer chemotherapy. Biochemica
Biophysica Acta, 458, 243.
Sartore, M., Adami, M., Nicolini, C, Bousse L., Mostarshed, S. and Hafeman, D. (1992). Minority carrier diffusion length effect on light addressable potentiometric sensor (LAPS) devices. Senors and Actuators A. 32, 431-436.
Stiblert, L., Svensson, C, (1975). A hydrogen leak detector using a palladium gate MOS transistor, Review of Scientific Instruments, 46, 1206-1208.
Tengvall, P., Lestelius, M., Liedberg, B. and Lundstrom. I. (1992). Plasma protein and antisera interactions with L-Cysteine and 3-Mercaptoproprionic acid monolayers on gold surfaces, Langmuir, 9, 1236-1238.
References: Toxicity Testing in Cancerous Hepatocytes
Adami, M., Sartore M. and Nicolini C. "PAB: a newly designed potentiometric alternating biosensor system" . Biosensors and Bioelectronics, (1994, in press).
Adami, M., Piras L, Lanzi M., Fanigliulo A., Vakula S. and Nicolini C, "Monitoring enzymatic activity and quantitative measurements of substrates by means of a newly designed silicon-based potentiometric sensor" . Sensors and Actuators B., (1993, in press).
Babich H. and Borenfreund, E. "Structure-activity
relationship (SAR) models established in vitro with the neutral red cytotoxicity assays". Toxicology in Vitro 1: 3-9, 1987.
Delia Ciana et al , "Improvements in Chemical-sensitive semiconductor devices (CSSDs, European Patent No. 91113155.5. (1991).
Dich J., Vind, C. and Grumet N. "Long-term cultures of hepatocytes: effects of hormones on enzyme activities and metabolic capacity". Hepatology 8: 39-45, 1988.
Dollery, C. Therapeutic Drugs Vol .2 Churchill Livinstone, Edinburgh, 1991 pp. M218-M222.
Dubrowsky T., Vakula S. and Nicolini C, "Preparation and immobilization of L&B films of antibodies conjugated to enzymes for potentiometric sensor application" . Sensors and Actuators (1994, in press). Englemann G.L., Richardson, A.G. and Fierer, J.A.
"Maintenance and induction of cytochrome P450 in cultured rat hepatocytes". Arch. Biochem. Biophys., 238: 359-367, 1985.
Gavazzo P., Paddeu S., Sartore M. and Nicolini C. "Study of the relationship between extra cellular acidification and cell viability by a silicon based sensor". Sensors and Actuators (1993, in press).
Martelli, A., Robbiano L., Gassaniga, G.M. and Brambilia G. "Comparative study of DNA damage and repair induced by ten N- nitroso compounds in primary cultures of human and rat hepatocytes", Cancer Research 48, 4144-4152, 1988.
Nicolini C., Adami, M. Zunino M. and Sartore, M. "A newly designed silicon-based cell biosensor for in vivo drug screening. Part 1 : a 2D system for simultaneous pH and redox determination".
Sartore, M., Adami, M. , Nicolini C, Bousse L., Mostarshed S. and Hafeman D. "Minority carrier diffusion length effect on light addressable potentiometric sensor (LAPS) devices".
Sensors and Actuators A, 32, 1992, 431-436.
Williams, G.M. "Detection of chemical carcinogens by
unscheduled DNA synthesis in rat liver primary cultures". Cancer Res. 37: 1845-1851, 1977.
REFERENCES : Enzyme Monolayer Immunosensor
C. Nicolini, M.Adami, F. Antolini, F. Beltram, M. Sartore and S. Vakula, Biosensors: a step to bioelectronics, Physics World, May (1992), 30-34.
C. Nicolini, M. Adami and M. Sartore, PAB: a newly designed potentiometric alternating biosensor for biomedical and environmental applications. (1993) Patent pending.
M. Adami, M. Sartore and C. Nicolini, PAB: a newly designed potentiometric alternating biosensor for biomedical and environmental applications, Biosensors and Bioelectronics, (1994) in press.
D.G Hafeman, J.W.Parce and H.M. McConnell, Light Addressable Potentiometric Sensor for biochemical systems, Science, 240, (1988), 1182-1185.
P. Gavazzo, S. Paddeu, M. Sartore and C. Nicolini, Study of the relationship between extracellular acidification and cell viability by a silicon based sensor. Sensors and Actuators B, 18-19, (1994) 368-372.
C. Nicolini, P. Accossato, A. Fanigliulo, M. Lanzi, F. Mattioli and A. Martelli, A newly designed silicon-based cell biosensor for a real-time drug screening. Part.II: toxicity testing in normal versus cancer liver cells, Biosensors & Bioelectronics, in press.
M. Adami, L. Piras, M. Lanzi, A. Fanigliulo, S. Vakula and C. Nicolini; Monitoring of enzymatic activity and quantitative measurements of substrates by means of a newly designed silicon-based potentiometric sensor. Sensors and Actuators B, 18-19, (1994) 178-182. F. Antolini, S. Paddeu, M. Adami and C. Nicolini, Langmuir-Blodgett film of Glutathione S-Transferase: deposition, characterisation and possible application for pesticides detection, Biosensors & Bioelectronics, in press.
G. Roberts, Langmuir-Blodgett films, Plenum Press, New York (1990).
C. Nicolini, V. Erokin, F. Antolini, P. Catasti and P. Facci, Thermal stability of protein secondary structure in Langmuir-Blodgett films, Biochimica et Biophysica. Acta, 1158, (1994), 273-278.
C. Nicolini, V. Erokin, F.Antolini and P. Facci, A new general procedure for protein thermal stabilization by Langmuir-Blodgett technique, (1993), Patent pending.
M. Adami, M. Sartore, A. Rapallo and C. Nicolini, Possible developments of a potentiometric biosensor, Sensors and Actuators B , 7, (1992), 343-346.
M. Malmqvist and G. Olofsson, Method of silanization of surfaces, US Patent 4,833,093 (1993).
I. Langmuir, VJ. Schaefer, Activities of urease and pepsin monolayers, J. Amer. Chem. Soc, 60 (1938) 1351-1360.
T. B. Dubrovsky, M. V. Demcheva, A. P. Savitsky, E. Yu. Mantrova, A. I. Yaropolov, V. V. Savransky, L. V. Belovolova, Fluorescent and phosphorescent study of Langmuir- Blodgett antibody films for application to immunosensors, Biosensors & Bioelectronics, 8 (1993) 377-385. P. Facci, V. Erokhin and C. Nicolini, Nanogravimetric gauge for surface density measurements and deposition analysis of Langmuir-Blodgett films, Thin Solid Films, G.Z. Sauerbray, Z.Phys. 178, (1964) 457-462.
ELM. Chandler et al., J. Immunol. Meth., 53,(1982) 187.
R.A.Williams and H.V.Blanch, Covalent immobilization of protein monolayers for biosensor applications, Biosensors and Bioelectronics, 9 (1994) 159-167.

Claims

1. A temperature control device comprising a heating/cooling device in the form of a Peltier cell; a temperature sensor for sensing the temperature in the vicinity of the cell; and control means responsive to the sensed temperature for controlling the heating and cooling activity of the Peltier cell so as to maintain the temperature in the vicinity of the cell at or near a predetermined value.
2. A biosensor including a temperature control device for controlling the temperature in a sample region, the device comprising a heating/cooling device; a temperature sensor for sensing the temperature in the vicinity of the device and control means responsive to the sensed temperature for controlling the heating and cooling activity of the heating/cooling device so as to maintain the temperature in the vicinity of the device at or near a predetermined value.
3. A biosensor according to claim 2, wherein the
heating/cooling device is in the form of a Peltier cell.
4. A sensor according to claim 2 or claim 3 , wherein the sample region defines a micro volume.
5. A sensor according to any of claims 2 to 4, wherein the sample region is formed by part of a channel through which the sample flows .
6. A sensor according to any of claims 2 to 5, wherein the sensor is a potentiometric sensor.
7. A sensor according to any of claims 2 to 6, wherein the sensor is adapted to detect the pH of an electrolyte.
8. A device or sensor according to any of the preceding claims, wherein the control means includes a comparator for comparing the sensed temperature with a predetermined value and generates a control signal to the heating/cooling device to cause the device to heat or cool the vicinity of the device depending on whether the sensed temperature is less than or more than the predetermined value respectively.
9. A device or sensor according to claim 8, wherein the control means generates a fixed value control signal
independent on the outcome of the comparison.
10. A device or sensor according to claim 8 or claim 9, wherein the comparator continuously compares the sensed temperature with the predetermined value.
11. A method of measuring a pH variation or a redox potential variation in an enzymic reaction that generates ions that cause, or whose generation causes, such a
variation, the method comprising monitoring the reaction over a period of time using a light-addressable potentiometric sensor that generates a current on the binding of the respective ions thereto; converting current measured over that time to voltage; if desired, calculating the pH
variation or the redox potential variation and/or the number of respective ions generated as a function of the voltage; and comparing the voltage, pH variation or redox potential variation, or number of ions with a precalibrated standard.
12. A method according to claim 11 for calculating the concentration of enzyme, wherein the precalibrated standard is a plot of the voltage, pH variation or redox potential variation, or number of ions generated against concentrations of enzyme at a constant concentration of substrate.
13. A method according to claim 11 for calculating the concentration of substrate, wherein the precalibrated standard is a plot of the voltage, pH variation or redox potential variation, or number of ions generated against concentrations of enzyme at a constant concentration of substrate.
14. A biosensor for carrying out a method according to any preceding claim, comprising a reaction chamber and in communication therewith at least one first light-addressable potentiometric sensor that each of which generates a current on the binding of the respective ions thereto, and each is programmed with a precalibrated standard and to convert current to voltage; if desired, to calculate the pH variation and/or the redox potential variation and/or the number of respective ions generated as a function of the voltage; to compare the voltage, pH variation and/or redox potential variation, or number of ions with the precalibrated standard; and to display that data.
15. A biosensor according to claim 14, wherein each at least one potentiometric sensor is arranged in a measuring chamber that communicates with the reaction chamber via a conduit.
16. A biosensor according to claim 14, wherein the reaction chamber has an inlet and an outlet for reactants to flow into and out of the chamber, and a switch to control the flow of the reactants.
17. A biosensor according to any of claims 14 to 16 , which comprises two potentiometric sensors , one to which H+ ions bind and the other to which ions, whose generation cause a redox potential, bind.
18. A biosensor according to any of claims 14 to 16, wherein the at least one potentiometric sensor binds both H+ ions and ions whose generation cause a redox potential.
19. A biosensor according to any of claims 14 to 18, further comprising at least one second light-addressable sensor that binds the same ions as at least one first sensor, and that is programmed to subtract data acquired by each second sensor from that acquired by each respective first sensor, to reduce background noise.
20. A biosensor according to any of claims 14 to 19 that is enclosed within a housing.
21. A method or biosensor according to any of claims 11 to 20 including a temperature control device according to claim 1.
22. A transducer comprising a heterostructure of silicon ("n" or "p" type), silicon dioxide and silicon nitride and having at least one metal layer deposited over at least a portion of the silicon nitride surface, the
transducer being for use in a redox potential biosensor.
23. A redox potential biosensor including a transducer as defined in claim 22.
24. Use of a redox potential biosensor according to claim 23 to measure the concentration/activity of the enzyme Horse Radish Peroxidase (HRP) or the enzymes Alcohol
Dehydrogenase (ADH) together with Diaphorase when catalyzing a redox reaction.
25. A transducer comprising a light-addressable silicon substrate which has two or more front side surface regions modified and functionalized by virtue of a surface coating such that each region can act as a potentiometric sensor for local spatial measurement of pH or redox potential or specific chemical entity.
26. A transducer according to claim 25 which is light-addressed by means of two or more optical fibres arranged at predetermined locations in the form of an array fixed in close proximity to the transducer backside surface.
27. A transducer according to claim 25 or 26 in which the transducer front side surface is wholly or partly covered with a layer of Si3N4 for pH measurement.
28. A transducer according to claim 25 or 26 in which the transducer front side surface is wholly or partly covered with a layer of gold for redox potential measurements.
29. A transducer according to any of claims 25 to 28 which is multisensitive and bidimensional by virtue of having at least two regions for each of both pH and redox potential measurement.
30. A potentiometric bidimensional biosensor including a transducer as defined in any of claim 25 to 29.
31. Use of a biosensor according to claim 30 to measure pH and/or redox potential and/or other chemical entity at more than one region of a transducer as defined in any of claims 25 to 29, thereby to produce a bidimensional spatial distribution of measurements.
32. Use of a PAB system including a potentiometric pH sensor to monitor cellular metabolism by measurement of extracellular acidification.
33. Use according to claim 32 to monitor cellular status at one or more phases of the cell metabolic cycle.
34. Use according to claim 32 or 33 to monitor a change in extracellular acidification rate resulting from drug administration to the cell.
35. Use according to claim 34 to determine efficacy of drug administration.
36. Use according to claim 34 to determine toxicity of drug administration.
37. Use according to claim 34 wherein the drug
administered is antineoplastic.
38. Use according to claim 37 wherein the drug is selected from (i) cytosine arabinoside (ara-C), and (ii) Mitoxantrone.
39. Use according to any of claims 32 to 38 wherein the cells are selected from (i) a cell line of mouse fibroblasts, and (ii) a primary culture of rat hepatocytes.
40. A light-addressable silicon transducer adapted for use as an immunosensor by virtue of the application to the front side surface thereof of an immobilized thin film monolayer of an enzyme.
41. A potentiometric alternating biosensor for pH and or redox potential measurement including a transducer as defined in claim 40.
42. Use of a biosensor according to claim 41 in an immunoenzymatic assay.
43. Use according to claim 42 for measurement of the enzymatic activity of urease.
PCT/IB1994/000408 1993-11-25 1994-11-25 Potentiometric biosensors, control and applications thereof WO1995014962A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
AU10753/95A AU1075395A (en) 1993-11-25 1994-11-25 Potentiometric biosensors, control and applications thereof
EP95901567A EP0730760A1 (en) 1993-11-25 1994-11-25 Potentiometric biosensors, control and applications thereof

Applications Claiming Priority (10)

Application Number Priority Date Filing Date Title
GB939324258A GB9324258D0 (en) 1993-11-25 1993-11-25 Biosensor and its use
GB9324258.4 1993-11-25
ITRM93A000846 1993-12-22
IT93RM000846A IT1266466B1 (en) 1993-12-22 1993-12-22 Automatic temperature control device
GB9411059A GB9411059D0 (en) 1994-06-02 1994-06-02 Biosensor and its use
GB9411072.3 1994-06-02
GB9411059.0 1994-06-02
GB9411072A GB9411072D0 (en) 1994-06-02 1994-06-02 Biosensor and its use
GB9414189A GB9414189D0 (en) 1994-07-14 1994-07-14 Sensor
GB9414189.2 1994-07-14

Publications (1)

Publication Number Publication Date
WO1995014962A1 true WO1995014962A1 (en) 1995-06-01

Family

ID=27517218

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB1994/000408 WO1995014962A1 (en) 1993-11-25 1994-11-25 Potentiometric biosensors, control and applications thereof

Country Status (3)

Country Link
EP (1) EP0730760A1 (en)
AU (1) AU1075395A (en)
WO (1) WO1995014962A1 (en)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996002830A1 (en) * 1994-07-14 1996-02-01 Technobiochip Biosensor and method and instrument for deposition of alternating monomolecular layers
GB2316196A (en) * 1996-05-22 1998-02-18 Univ Paisley Temperature changing and controlling apparatus
EP0851229A1 (en) * 1996-12-23 1998-07-01 Bayer Corporation Use of thermochromic liquid crystals in reflectometry based diagnostic methods
WO2001057243A1 (en) * 2000-02-03 2001-08-09 Danfoss A/S Biological microsensor
WO2002035219A2 (en) * 2000-10-24 2002-05-02 Forschungszentrum Jülich GmbH Measuring device for detecting one-dimensional or multidimensional distribution of a chemical or biochemical component
US6780296B1 (en) 1999-12-23 2004-08-24 Roche Diagnostics Corporation Thermally conductive sensor
DE10309349B4 (en) * 2003-03-03 2005-11-10 Micronas Holding Gmbh Device for analyzing an analyte
WO2007053105A1 (en) * 2005-10-31 2007-05-10 Senzime Point Of Care Ab A biosensor apparatus for detection of thermal flow
WO2010097266A1 (en) * 2009-02-26 2010-09-02 Siemens Aktiengesellschaft Method and arrangement for inhibiting a chemical reaction of substances in a liquid before a measurement
IT202100015608A1 (en) * 2021-06-15 2021-09-15 Claudio Gianotti Thermoelectric device for achieving human normothermia
US11448580B2 (en) * 2018-01-16 2022-09-20 Southeast University Biodetector based on interference effect of thin film with ordered porous nanostructures and method for using same to detect biomolecules

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4020830A (en) * 1975-03-12 1977-05-03 The University Of Utah Selective chemical sensitive FET transducers
US4454007A (en) * 1983-01-27 1984-06-12 E. I. Du Pont De Nemours And Company Ion-selective layered sensor and methods of making and using the same
US4653056A (en) * 1985-05-01 1987-03-24 Spectra-Physics, Inc. Nd-YAG laser
US4713165A (en) * 1986-07-02 1987-12-15 Ilex Corporation Sensor having ion-selective electrodes
EP0318273A2 (en) * 1987-11-25 1989-05-31 Northeastern University Integrated temperature control/alignment system for high performance capillary electrophoretic apparatus
US4894339A (en) * 1985-12-18 1990-01-16 Seitaikinouriyou Kagakuhin Sinseizogijutsu Kenkyu Kumiai Immobilized enzyme membrane for a semiconductor sensor
EP0382426A2 (en) * 1989-02-06 1990-08-16 Applied Biosystems, Inc. Micro-preparative electrophoresis apparatus
US5157352A (en) * 1991-11-04 1992-10-20 Electronic Instrumentation And Technology Inc. Bias current control for operational amplifier current/voltage converters
US5288388A (en) * 1987-12-09 1994-02-22 Societe Anonyme: Societe Lyonnaise D'electronique Appliquee Homogeneous potentiometer sensor

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4020830A (en) * 1975-03-12 1977-05-03 The University Of Utah Selective chemical sensitive FET transducers
US4020830B1 (en) * 1975-03-12 1984-09-04
US4454007A (en) * 1983-01-27 1984-06-12 E. I. Du Pont De Nemours And Company Ion-selective layered sensor and methods of making and using the same
US4653056A (en) * 1985-05-01 1987-03-24 Spectra-Physics, Inc. Nd-YAG laser
US4653056B1 (en) * 1985-05-01 1990-07-24 Spectra Physics
US4894339A (en) * 1985-12-18 1990-01-16 Seitaikinouriyou Kagakuhin Sinseizogijutsu Kenkyu Kumiai Immobilized enzyme membrane for a semiconductor sensor
US4713165A (en) * 1986-07-02 1987-12-15 Ilex Corporation Sensor having ion-selective electrodes
EP0318273A2 (en) * 1987-11-25 1989-05-31 Northeastern University Integrated temperature control/alignment system for high performance capillary electrophoretic apparatus
US5288388A (en) * 1987-12-09 1994-02-22 Societe Anonyme: Societe Lyonnaise D'electronique Appliquee Homogeneous potentiometer sensor
EP0382426A2 (en) * 1989-02-06 1990-08-16 Applied Biosystems, Inc. Micro-preparative electrophoresis apparatus
US5157352A (en) * 1991-11-04 1992-10-20 Electronic Instrumentation And Technology Inc. Bias current control for operational amplifier current/voltage converters

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
L. BOUSSE ET AL.: "Biosensors for Detection of Enzymes Immobilized in Microvolume Reaction Chambers", SENSORS AND ACTUATORS, B CHEMICAL, vol. B1, 1990, pages 555 - 600 *

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996002830A1 (en) * 1994-07-14 1996-02-01 Technobiochip Biosensor and method and instrument for deposition of alternating monomolecular layers
GB2316196A (en) * 1996-05-22 1998-02-18 Univ Paisley Temperature changing and controlling apparatus
GB2316196B (en) * 1996-05-22 2000-07-19 Univ Paisley Temperature control apparatus and methods
EP0851229A1 (en) * 1996-12-23 1998-07-01 Bayer Corporation Use of thermochromic liquid crystals in reflectometry based diagnostic methods
US7510643B2 (en) 1999-12-23 2009-03-31 Roche Diagnostics Operations, Inc. Sensor system
US6780296B1 (en) 1999-12-23 2004-08-24 Roche Diagnostics Corporation Thermally conductive sensor
WO2001057243A1 (en) * 2000-02-03 2001-08-09 Danfoss A/S Biological microsensor
WO2002035219A2 (en) * 2000-10-24 2002-05-02 Forschungszentrum Jülich GmbH Measuring device for detecting one-dimensional or multidimensional distribution of a chemical or biochemical component
WO2002035219A3 (en) * 2000-10-24 2002-12-19 Forschungszentrum Juelich Gmbh Measuring device for detecting one-dimensional or multidimensional distribution of a chemical or biochemical component
DE10309349B4 (en) * 2003-03-03 2005-11-10 Micronas Holding Gmbh Device for analyzing an analyte
WO2007053105A1 (en) * 2005-10-31 2007-05-10 Senzime Point Of Care Ab A biosensor apparatus for detection of thermal flow
US7947223B2 (en) 2005-10-31 2011-05-24 Senzime Ab Biosensor apparatus for detection of thermal flow
WO2010097266A1 (en) * 2009-02-26 2010-09-02 Siemens Aktiengesellschaft Method and arrangement for inhibiting a chemical reaction of substances in a liquid before a measurement
DE102009010639B4 (en) 2009-02-26 2020-07-02 Boehringer Ingelheim Vetmedica Gmbh Method and arrangement for inhibiting a chemical reaction of substances in a liquid before a measurement
US11448580B2 (en) * 2018-01-16 2022-09-20 Southeast University Biodetector based on interference effect of thin film with ordered porous nanostructures and method for using same to detect biomolecules
IT202100015608A1 (en) * 2021-06-15 2021-09-15 Claudio Gianotti Thermoelectric device for achieving human normothermia

Also Published As

Publication number Publication date
EP0730760A1 (en) 1996-09-11
AU1075395A (en) 1995-06-13

Similar Documents

Publication Publication Date Title
Fuchs et al. In-line analysis of organ-on-chip systems with sensors: Integration, fabrication, challenges, and potential
Niwa et al. Concentration of extracellular L-glutamate released from cultured nerve cells measured with a small-volume online sensor
Baumann et al. Microelectronic sensor system for microphysiological application on living cells
Wolf et al. Monitoring of cellular signalling and metabolism with modular sensor-technique: The PhysioControl-Microsystem (PCM®)
Porterfield Measuring metabolism and biophysical flux in the tissue, cellular and sub-cellular domains: recent developments in self-referencing amperometry for physiological sensing
Thévenot et al. Electrochemical biosensors: recommended definitions and classification
Kukla et al. Multienzyme electrochemical sensor array for determination of heavy metal ions
Eklund et al. A microphysiometer for simultaneous measurement of changes in extracellular glucose, lactate, oxygen, and acidification rate
JPS6123957A (en) Detection and discrimination of optical responding redox
Yotter et al. Sensor technologies for monitoring metabolic activity in single cells-part II: nonoptical methods and applications
Lowe Biosensors
Flampouri et al. Spheroid-3D and monolayer-2D intestinal electrochemical biosensor for toxicity/viability testing: Applications in drug screening, food safety, and environmental pollutant analysis
WO1995014962A1 (en) Potentiometric biosensors, control and applications thereof
Scheller et al. Biosensors: Fundamentals, applications and trends
Arbault et al. Oxidative stress in cancer prone xeroderma pigmentosum fibroblasts. Real-time and single cell monitoring of superoxide and nitric oxide production with microelectrodes
Ehret et al. Multiparametric microsensor chips for screening applications
JPH04506898A (en) Piezoelectric cell growth biosensing method and system
Roda et al. Chemiluminescent low-light imaging of biospecific reactions on macro-and microsamples using a videocamera-based luminograph
Abdel-Hamid et al. Development of a flow-through immunoassay system
Bäcker et al. Electrochemical sensor array for bioprocess monitoring
Adami et al. PAB: a newly designed potentiometric alternating biosensor system
Ino et al. Combination of double-mediator system with large-scale integration-based amperometric devices for detecting NAD (P) H: quinone oxidoreductase 1 activity of cancer cell aggregates
Zhang et al. Correlation between cell growth rate and glucose consumption determined by electrochemical monitoring
US20050076943A1 (en) Thermoelectric sensor
Lammers et al. Thermal biosensors in biotechnology

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AM AT AU BB BG BR BY CA CH CN CZ DE DK EE ES FI GB GE HU JP KE KG KP KR KZ LK LT LU LV MD MG MN MW NL NO NZ PL PT RO RU SD SE SI SK TJ TT UA US UZ VN

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): KE MW SD SZ AT BE CH DE DK ES FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN ML MR NE SN TD TG

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 1995901567

Country of ref document: EP

ENP Entry into the national phase

Ref country code: US

Ref document number: 1996 649627

Date of ref document: 19960723

Kind code of ref document: A

Format of ref document f/p: F

WWP Wipo information: published in national office

Ref document number: 1995901567

Country of ref document: EP

REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

NENP Non-entry into the national phase

Ref country code: CA

WWW Wipo information: withdrawn in national office

Ref document number: 1995901567

Country of ref document: EP