Microbiology is a field of biology that studies the structure, function, and operation of cells. With an understanding of the structure, function, and operation of cells a variety of chemical reactions and processes can be carried out. For example, individual cells may be used to generate cell lines and to aide in the further understanding of mechanisms of cellular function.
The accompanying drawings illustrate various examples of the principles described herein and are part of the specification. The illustrated examples are given merely for illustration, and do not limit the scope of the claims.
Throughout the drawings, identical reference numbers designate similar, but not necessarily identical, elements. The figures are not necessarily to scale, and the size of some parts may be exaggerated to more clearly illustrate the example shown. Moreover, the drawings provide examples and/or implementations consistent with the description; however, the description is not limited to the examples and/or implementations provided in the drawings.
Microbiology is a field of biology that studies the structure, function, and operation of cells. A greater understanding of the different kinds of cells and their function can lead to certain technological innovations that benefit society in countless ways. For example, from cells, certain biologics such as proteins, insulin, other therapeutic drugs, RNA, and DNA may be synthesized.
Accordingly, refinements to cellular processing may enhance the possible uses and reach of cellular analytics. For example, pure cell lines are used to ensure high fidelity production of biologics as well as for understanding the fundamental mechanisms of cellular function. A pure cell line refers to cells that are derived from a single cell with well-defined properties. Accordingly, to generate such pure cell lines, or biologics from a single cell line, a scientist should be confident that 1) the cell line was derived from a single cell and 2) that the cell has certain desired properties. Determining that a single cell, and the pure cell line derived therefrom, has certain desired properties may increase the efficiency of cellular analytics as it may reduce the number of cells to be cultured as those that do not have the desired property may be distinguished from those that do have the desired property thus reducing the amount of quality control needed for the cultured cells.
As a specific example, a scientist may desire to generate chimeric antigen receptor (CAR) T-cells in order to combat cancerous cells in a patient. CAR T-cell therapy changes a patient's T-cells in a way that attacks the cancerous cells. Accordingly, T-cells are processed and manipulated to form CAR T-cells. Were a starting population to include an unknown quantity of T-cells or to include other types of cells, the quantity and quality of produced CAR T-cells would be reduced and/or unknown, which could affect the efficacy of a session of the CAR T-cell therapy.
Accordingly, by using a starting population for which the characteristics are definitively verified, confidence and reliability in the subsequent operations is promoted. Accordingly, the present specification ensures such reproducibility and confidence by automatically sorting cells based on examination of the cells' multiple electrical and mechanical properties and placing individual cells into individual wells for further propagation or analysis. That is, in the above example, T-cells may be placed in individual wells such that a user may know with certainty what cells and what quantity of cells are present in a starting population such that a deterministic operation may be carried out to generate the CAR T-cells. That is, the present specification reduces the unknowns from a starting population such that a scientist may more accurately carry out subsequent chemical operations.
In other words, as can be imagined, uncertainty regarding the characteristics of a starting population introduces uncertainty and inaccuracy for any operation subsequently carried out on that starting population, whether that operation be analysis to identify a particular relationship, cell generation, and/or biologic generation. The present specification generates a starting population where uncertainty is removed by 1) ensuring single cells are used to generate a pure line and 2) ensuring that the pure line is based off a cell with desired properties. Accordingly, there can be greater confidence in the outcome of any subsequent operation.
While some solutions have been presented, they are inadequate for any number of reasons. For example, in some cases cells may be sorted by fluorescence activated cell sorting (FACS) and magnetically activated cell sorting (MACS), both of which rely on labels to identify the cells, a single cell at a time. However, the presence of such labels may interfere with cellular operation and analysis of cell functionalities. Accordingly, the present specification provides or such sorting and singulation without labeling agents, thus preserving the integrity of any subsequent operations.
Some label-free separation methods exist. Examples include density gradient centrifugation separation and filtration-based (size and stiffness based) cell separation. However, these operations are carried out on large quantities of cells and properties of individual cells are not measured.
Label free singulation of cells may include expensive systems involving video detection integrated with dispensing, or dilution of the cell suspension and portioning the suspension into individual volumes (wells). However, with these methods, a large number of wells may end up without cells and just reagent, while other wells may contain more than one cell. Having wells with one more cell may introduce additional quality control operations which is ineffective and costly.
Accordingly, the present system provides automated, low-cost single-cell sorting and dispensing based on a cell's electrical and mechanical properties. The system includes a microfluidic device with integrated pumps, and a plurality of constrictions with integrated sensors. The sensors may be impedance sensing electrodes, but other integrated sensors such as silicon-diode photodetectors may also be used. Each constriction with a sensor forms an independent detection device, each detection device to measure different aspects of the cell properties. Via a controller, each individual detection device then determines a characteristic of the cell found within the detection device. The controller combines the readings from each detection device and decides whether to eject a cell into a well or to eject the cell to a waste receptacle. Direct combination (Boolean “AND”) of the detection device outputs may lead to an exponential increase in the accuracy of obtaining the desired cell.
In one particular example, the system includes a microfluidic chip that includes thermal inkjet (TIJ) dispensing pumps, and a plurality of chambers with a constriction and electrodes, an X-Y stage a control unit to control the stage and the voltages in the microfluidic chip, and a reservoir for cells feeding into the chip. The chambers constrict the cell so that the cell geometry between electrodes is reproducible. In one particular example, a voltage is applied to the electrodes at an appropriate frequency. This generates an electric field. Depending on the electric field and polarizability of the cell membrane, the electric field either penetrates the cell or passes in a thin film of liquid between the cell wall and the constriction. Depending on the elasticity of the cell, which may be a function of the cell microtubule content, membrane content etc., this thin film has a different thickness, and so different impedance. For different cell types, the electric field penetrates the cell membrane at a different frequency. A different field frequency is applied to different chambers, and by comparing the impedance at different chambers, cells can be differentiated on their polarizability, one cell at a time.
Once the cell effective impedance at various frequencies is measured, it is then correlated to both the cell's mechanical properties (deformability during constriction) as well as the membrane electrical properties (membrane capacitance). The controller then either 1) dispenses the cell into an individual well or disposes of the cell and in some cases 2) records the cell properties, which can be further correlated with the cell health (including ability to produce biologics) and other properties analyzed downstream.
The decision to dispense the cell into the well or dispose of the cell is done by consensus of n detection devices. That is, each cell may evaluate whether there is a single cell in the chamber and whether it is of a desired type. Accordingly, if n detection devices agree that the cell is of a desired cell, each with a true positive rate, TP, and a false positive rate, FP, and the frequency of desired cells in the solution is q, then the probability, p, that the cell is actually a desired cell scales exponentially with the number of independent detectors: p(S|W1 . . . WnI)=q(TP/qTP+(1−q)FP)n. Thus, a plurality of detection devices allows for low-cost high-fidelity cell sorting.
Specifically, the present specification describes a cell preparation system. The cell preparation system includes a fluidic channel to transport cells in single file past at least one detection device. The cell preparation system also includes a number of detection devices, each detection device including a constriction to deform a cell found therein and a sensor to measure a state within the constriction. The cell preparation system also includes a controller to analyze outputs from multiple sensors to: verify a single file transport of cells through the system and identify a cell that passes through the fluidic channel.
In some examples, the controller identifies, per detection device and based on output from the sensors, a cell elasticity, a cell size, and a cell effective conductivity and permittivity. The controller may also determine, per detection device, whether an output of a sensor is indicative of a single cell of a desired type.
In some examples, the sensor is an impedance sensor to measure an impedance within a respective constriction. In some examples, the sensor is disposed within the constriction. In other examples, the sensor is disposed outside the constriction.
In some examples, at least one of a constriction shape and an applied electrical field vary for at least one detection device. In some examples, the cell preparation system includes a database. The database is to map, per detection device, sensor outputs to quantity and types of cells and is indexed based on properties of a respective detection device.
The present specification also describes a method. According to the method, different electrical fields are applied to each of a number of detection devices along a fluidic channel. For each detection device, a cell is introduced into a constriction of the detection device and an electrical impedance is determined when the cell is in the constriction. Based on electrical impedance measures from each detection device a single file transport of cells through the system is verified and cells that pass through the fluidic channel are identified.
In one example, based on electrical impedance measures from each detection device, the method includes determining whether to eject the cell onto a target surface or to dispose of the cell to a waste receptacle. In some examples, properties for each cell are recorded.
The present specification also describes a cell preparation system. The cell preparation system includes a signal generator to generate electrical fields to apply to multiple detection devices and at least one fluidic channel to transport cells in single file past multiple detection devices. The cell preparation system also includes a series of detection devices per fluidic channel, each detection device including: a constriction to deform a cell found therein and at least one electrode pair to measure an electrical property within the constriction. The cell preparation system also includes sense circuitry to determine an impedance within each detection device and an ejector to eject cells from the fluidic channel. The cell preparation system also includes a controller to aggregate outputs form the multiple electrode pairs to: verify a single file transport of cells through the system; identify a cell that passes through the fluidic channel; and determining whether to eject the cell onto a target surface or to dispose of the cell to a waste receptacle.
In some examples, the system includes a filter device per detection device to generate an electrical field specific to a respective detection device. The filter device may be selected from the group consisting of a band-pass filter and a lock-in amplifier.
In one example, the sense circuitry includes an isolation component to isolate output from a particular detection device. The isolation component may be selected from the group consisting of a band-pass filter and a switch. The sense circuitry may also include a current meter.
In summary, using such a cell preparation system 1) provides highly accurate cell separation and identification; 2) is low cost; 3) provides for the rapid generation of many singulated cells; 4) avoids additional labelling reagents; 5) provides an integrated design; 6) exhibits a low probability of empty wells; and 7) reduces quality control measures. However, the devices disclosed herein may address other matters and deficiencies in a number of technical areas.
As used in the present specification and in the appended claims, the term “effective conductivity and permittivity” refers to a cells ability to conduct electricity and how a cell polarizes by imposition of an electrical field.
Still further, as used in the present specification and in the appended claims “applying an electrical field” refers to controlling a voltage or controlling a current and measuring the other. For example, an electrical field may be applied by controlling a voltage and measuring a current. In another example, applying an electrical field includes controlling a current and measuring a voltage.
Turning now to the figures,
To carry out such cell preparation, the cell preparation system (100) includes a variety of sub-components. Specifically, the cell preparation system (100) includes a fluidic channel (102) to transport cells in single file past multiple detection devices (104). Accordingly, the fluidic channel (102) may be a microfluidic structure coupled at one end to a cell reservoir and directing cells single-file towards the first detection device (104). Following passage by the multiple detection devices (104), the fluidic channel (102) may direct the cells to another destination, such as an ejection chamber to be expelled onto a target surface such as a well of a titration plate or to be expelled to a waste container.
The cell preparation system (100) also includes a series of detection devices (104). As described above, each detection device (104) includes components that aid in accurately and reliably 1) indicating whether cells are transported single-file and 2) identifying particular cells. That is, each detection device (104) in the series outputs data indicating whether a single cell passed through and that is used to identify the cell that passed through. The controller (100) considers similar information for each of the detection devices (104) and determines from the aggregated information whether a single cell passed by and what the type of cell was. The cell preparation system (100) may include any number of detection devices (104). The greater the number of detection devices (104), the higher the accuracy of the cell preparation system (100).
Each detection device (104) includes multiple components. For example, each detection device (104) includes a constriction (106) to deform the cell. Specifically, the constriction (106) may have a cross-sectional size smaller than the cell so as to deform the cell. Accordingly, as noted above, the constriction (106) may be a microfluidic structure. As will be described below, the degree to which the cells deform as they enter the constriction (106) may be used to identify the particular cell. For example, some cells deform more than others and will more readily deform within the constriction (106). The degree of deformation may therefore be used to identify and classify the cell. The degree of deformation may be determined by the sensors (108) and the controller (110).
The detection device (104) also includes a sensor (108) to measure a state within the constriction (106). The sensor (108) may be of a variety of types. In one example, the sensor (108) is an impedance sensor to measure an electrical conductivity within a respective constriction (106). Such an impedance sensor may include at least one pair of electrodes. These electrodes detect a level of conductivity within the constriction (106). As the contents within the constriction change (106), so does the electrical impedance within the constriction (106). For example, the impedance within the constriction (106) may be different when a cell is present in the constriction (106), as compared to when a cell is not in the constriction (106). Moreover, different types of cells change the impedance by different amounts. Accordingly, the impedance sensors (106) can measure conductivity and an impedance may be determined therefrom. This impedance value can be correlated to determine 1) whether one or multiple cells are found in a constriction (106) and 2) what are the properties of the cell(s) found in the constriction (106).
The cell preparation system (100) also includes a controller (110). The controller (110), which includes hardware components such as a processor and memory, analyzes outputs from multiple sensors (108) to verify a single file transport of cells through the cell preparation system (100) and identifies cells that pass through the fluidic channel (102). That is, as described above, based on the output of not one, but multiple detection devices (104), the controller (110) can determine not only whether cells are being transported in single-file fashion through the fluid channel (102), but can also determine a type of cell, that is its identity. By combining the output of multiple detection devices (104), greater accuracy and reproducibility results. That is, rather than relying on a single detection device (104), which for one reason or another may return skewed results, the controller (110) of the present system relies on output of multiple detection devices (104). The same output from the different detection devices (104) may more assuredly indicate a particular single cell.
Note that in this example, each detection device (110) may apply different electrical fields such that the cell preparation system (100) is not just replicating a detection, but is detecting based on different parameters. That is, a cell is exposed to different experimentational conditions, and when each of the detection devices (104) identifies, based on its specific experimentational conditions, that the cell is of a particular type, the controller (110) may output with a degree of confidence that the cell is of that particular type.
As described above, each detection device (104) includes a constriction (106). For simplicity, a single constriction (106) is indicated with a reference number. However, each detection device (106) may include a constriction (106). Different cells (214) may deform differently when forced through a constriction (106). That is, some cells (214) are more elastic and may more readily deform. The elasticity of a cell (214) may be used to identify and classify a cell (214) and the sensors (
One electrode (212-1) may be coupled to a source (216) while a second electrode (212-2) is coupled to ground. Between these electrodes (216-1, 216-2) a current may be controlled and a voltage measured. In another example, a voltage is controlled and a current measured. Using both these values, the impedance between the electrodes (212), i.e., within the constriction (106) is determined.
The presence of a cell (214) within a constriction (106) will effectively change the electrical conductivity within the constriction (106). Moreover, different types of cells (214) affect the electrical conductivity to different degrees. That is, some cells (214) may have a larger impedance and increase the effective impedance within the constriction (106) to a different degree as compared to other cells (214) that may have a smaller impedance. Accordingly, the controller (110) may rely on a mapping between impedance values and cell (214) types to aide in identifying the cell. Part of the difference in impedance may be attributable to a cell (214) size. Accordingly, the cell (214) size may be determined based on a determined impedance value.
Even further, a particular cell (214) when exposed to different electrical fields may affect the conductivity to differing degrees. For example, a particular cell (214) when exposed to a lower voltage may be insulative, but when exposed to a higher voltage may be conductive. Accordingly, by probing the cell (214) at different voltages/currents the membrane properties, i.e., conductivity and permittivity of the cell, may be determined. Specifically, each of the different sensors (
Even further, a quantity of cells (214) in a constriction (106) may affect the conductivity. That is, more cells (214) may have a larger effect on impedance than a single cell. Accordingly, the controller (110) may map measured impedances with a quantity of cells (214).
In addition to using information from the cell (214) to identify the cell (214), information relating to the impedance of the surrounding fluid may also be used to identify the cell (214). That is, as described above, different cells (214) have different elasticities. Depending on the elasticity of the cell (214), it may deform to different degrees. The more a cell (214) deforms, the less carrier fluid will be present in the constriction (106). Whereas a more rigid cell (214), which deforms less, will result in more carrier fluid present in the constriction (106). The amount of fluid, and in particular the amount of fluid between the cell (214) walls and the walls of the constriction (106), results in different impedance values. Accordingly, the impedance within a constriction (106) can be correlated to a thickness of the film between the cell (214) and the constriction (106) wall. This thickness can then be mapped to a flexibility of the cell (214), and such flexibility used to classify the cell (214).
Note that in these examples, the impedance in the region of the constriction (106) can be used to determine different characteristics of the cell (214) such as cell effective permittivity and conductivity, cell elasticity, and cell count. To determine these values, impedance is measured as a function of constriction geometry and frequency and magnitude of the electric field.
As a specific example, cell effective permittivity and conductivity, cell elasticity, and cell size may be determined based on a correlation between an electrical signal such as impedance and the cell properties. For example, a mammalian cell (214) may have an elasticity of between 1-10 kilopascals (kPa). Cell elasticity may be measured via force microscopy. For example, a cantilever may be pressed with a known pressure into a cell while its under microscopic observation. The slope of the line of the applied pressure vs. deformation (strain) is the elastic modulus.
A mammalian cell (214) may have an electrical conductivity of 0.1-2 Siemens per meter (S/m) and a permittivity of 2-60× vacuum permittivity where vacuum permittivity is 8.85×1012 Farad per meter (F/m). To determine electrical conductivity and permittivity, the cell's (214) motion is observed in an AC electric field at various frequencies. The motion of the cell depends on the electric force applied to the cell (214), which is in turn related to its effective conductivity and permittivity.
A mammalian cell (214) may have a cell size of between 10 and 15 micrometers. However, mammalian cells may be as small as 3 micrometers and as large as 100 micrometers. Cell sizes may be determined by microscopic observation.
Accordingly, a mammalian cell (214) is introduced into the cell preparation system (100), and passes by multiple detection devices (104). However, at the time of introduction, it may be unknown that the cell (214) is a mammalian cell.
As the cell passes through the cell preparation system (100), it passes by a series of detection devices (104). Each detection device (104) includes a constriction (106) and a sensor (
Accordingly, as a cell passes through different detection devices (104), the corresponding electrodes (212) and controller (110) determine an impedance within the constriction (106). Specifically, a voltage is applied to the electrodes (212) by the controller (110). The controller (110) determines a ratio of 1) the voltage applied to the electrodes (212) to 2) the current flowing through the electrodes (212) and determines an impedance therefrom.
These electrical impedance measurements are then correlated to the different characteristics, i.e., cell elasticity, electrical conductivity, and electrical permittivity. That is, there may be a correlation table that maps output impedances to known values of each characteristic. Accordingly, once an output impedance is known, the controller (100) may identify that output impedance in the correlation table and a corresponding elasticity, electrical conductivity, electrical permittivity, and/or cell size may be identified.
As described above, each characteristic, i.e., conductivity, permittivity, size, and elasticity, may be determined by a separate detection device (104). For example, a controller (110) may map an output from a first detection device (104-1) to a cell elasticity and may map an output from a second detection device (104-2) to a cell conductivity.
In the case where a mammalian cell is introduced, output impedances may be determined by the sensor (
Accordingly, once either of these characteristics (i.e., cell effective permittivity and conductivity, cell elasticity, and cell count) is determined, that variable may be marked as known, and the other characteristic may be determined. Accordingly, by probing the cell (214) at different voltages/currents, and frequencies the controller (110) can determine either the dimensions of the film around the cell (214), or the properties of the cell (214) itself. Once the properties of the film around the cell (214) are determined, this is removed as an unknown and other properties of the cell (214), such as the properties of the cell (214) itself can be determined in downstream detection devices (104).
Accordingly, the controller (110) identifies, per detection device (104) and based on an output from the sensors (
The controller (110) also determines, per detection device (104), whether an output of a sensor (
Based on the output of the different detection devices (104), the controller (110) may take a variety of actions. For example, the cell preparation system (100) may include an ejector (218) that expels the cell (214) onto a target surface such as a well of a titration plate. That is, if the controller (110) determines that there is a single cell (214) of a desired type present, the controller (110) may activate the ejector (218) to expel the cell (214). By comparison, if the controller (110) determines that there is not a single cell (214) of the desired type present, the controller (110) dispose of the cell (214). In one particular example, the controller (110) may activate the ejector (218), but the controller (110) or a different processor may move the target surface such that the undesirable cell (214) is ejected onto a different surface, such as a waste receptacle.
The ejector (218) may include a firing resistor or other thermal device, a piezoelectric element, or other mechanism for ejecting fluid from the firing chamber. For example, the ejector (218) may be a firing resistor. The firing resistor heats up in response to an applied voltage. As the firing resistor heats up, a portion of the fluid in the firing chamber vaporizes to form a bubble. This bubble pushes the cell (214) out the opening and onto a surface such as a micro-well plate. As the vaporized fluid bubble collapses, a vacuum pressure along with capillary force within the firing chamber draws another cell (214) into the firing chamber from the fluidic channel (102), and the process repeats. In this example, the ejector (218) may be a thermal inkjet ejector (218).
In another example, the ejector (218) may be a piezoelectric device. As a voltage is applied, the piezoelectric device changes shape which generates a pressure pulse in the firing chamber that pushes a cell (214) out the opening. In this example, the ejector (218) may be a piezoelectric inkjet ejector (218).
According to the method (300), different electrical fields are applied (block 301) to each of multiple detection devices (
Then, for each detection device (
Also, for each detection device (
Based on the electrical impedance measures from each detection device (
Also based on the electrical impedance measures from each detection device (
Still further, other detection devices (
While an example is provided of first determining a thickness of the thin film then determining effective conductivity and permittivity, in some examples, these operations may be reversed. That is, there may be multiple variables of the cell (
Moreover, for each detection device (
Based on electrical impedance measures from each detection device (
Based on electrical impedance measures from each detection device (
The method (400) may include recording (block 407) properties for each cell (
In the example depicted in
In the example depicted in
In the example depicted in
The different constriction (106) shapes may provide further data regarding the elasticity of the cell (
As depicted in
The cell preparation system (100) also includes at least one fluidic channel (102) to transport cells (
In this example, as in others, the cell preparation system (100) includes a series of detection devices (104) per fluidic channel (102), each detection device (104) including a constriction (106) to deform a cell (
In some examples, the cell preparation system (100) includes sense circuitry (924) to determine an impedance within each detection device (104). That is, the sense circuitry (924) may sense the measured voltage/current from the electrodes (212). The sense circuitry (924) may include an isolation component to isolate output from a particular detection device (104), wherein the isolation component is selected from the group consisting of a band-pass filter and a switch. The sense circuitry (924) also includes a current meter.
The cell preparation system (100) may include the ejector (218) to eject cells (
In this example, the cell preparation system (100) includes additional components. For example, the cell preparation system (100) includes a database (1026) to map, per detection device (104), sensor (
In this example, the cell preparation system (100) includes a filter device (1028) per detection device (104) to generate an electrical field specific to a respective detection device (104). That is, the signal generator (922) may generate any number of signals, for example as a step function. The filter device (1028) may gate a subset of the generated signals such that just signals having a particular characteristic are passed to a respective detection device (104). In this fashion, different electrical fields, i.e., those allowed by the filter device (1028), may be applied at the different detection devices (104). The filter device (1028) may be selected form the group consisting of a band-pass filter and a lock-in amplifier.
In summary, using such a cell preparation system 1) provides highly accurate cell separation and identification; 2) is low cost; 3) provides for the rapid generation of many singulated cells; 4) avoids additional labelling reagents; 5) provides an integrated design; 6) exhibits a low probability of empty wells; and 7) reduces quality control measures. However, the devices disclosed herein may address other matters and deficiencies in a number of technical areas.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/US2020/025044 | 3/26/2020 | WO |