INTEGRATED AND STANDALONE LABEL AND REAGENT-FREE MICROFLUIDIC DEVICES AND MICROSYSTEMS FOR DIFFERENTIAL WHITE BLOOD CELL COUNTS

BACKGROUND

1. Technical Field

The present disclosure relates to the field of detection of substances present in biological fluids. More particularly, the present disclosure relates to devices, systems and methods for detection of analytes and substances in biological fluids such as blood.

2. Background of Related Art

Characterization of large quantities of individual particles is highly relevant to multiple fields, particularly blood analysis [1]. Blood is a highly complex fluid consisting of acellular (plasma) as well as diverse cellular components. The latter can cause significant interference when attempting to detect plasma biomarkers A simple cell count is useful, e.g. to diagnose anemia, but the utility of such measurements increases significantly with the ability to also determine size, surface markers, and interior composition. Applications range from CD4 T-cell monitoring in cases of HIV to stem cell characterization in research. The current gold standard for such measurements is bulky benchtop flow cytometers. These rely on a focused stream of blood cells being subjected to multiple analysis methods, involving fluorescent labels for population-specific surface antigens (e.g. CD4, CD8, . . . ), laser light scattering, and absorbance, or impedance measurements. Flow cytometers allow for highly accurate analysis, but rely on labels and complex optics. These factors are some of the major barriers in bringing this technology to the point of care (POC), where it would benefit patients as well as physicians by providing immediate results and increasing accessibility, especially in remote locations [2]. Lab-on-a-chip (LOC) systems have in recent years been shown to provide numerous advantages in clinical diagnostics, including portability, short reaction times, and low sample volumes [2]. These systems aim at bringing tests and procedures currently requiring a centralized laboratory to the POC or even the patient, integrating sample handling, biomarker detection, and readout electronics in a chip-size package.

Differential blood cell counters are representative examples of microfluidic devices that are already commercially available [3, 4]. However, these devices all rely on chemical reagents to enhance differentiation, and to date no portable POC device achieves a full (five-part) white blood cell differential. An intrinsic advantage here is that the current bench top laboratory methods are already based on microfluidics, and researchers have made efforts to translate these to modular LOC approaches [1, 5-11].

Especially impedance cytometry is well suited towards integration in microsystems, as it does not rely on labels or complex optics. To date, only few groups have published LOC-based complete white blood cell differentials based on impedance cytometry, including Holmes et al. [23] and Han et al. [24]. The former's approach, however, is limited due to inclusion also of optical measurements (and thus requiring fluorescent labels and external lasers, lenses, etc.), chemical reagents to enhance white blood cell differentiation, and an exceedingly complex fabrication method. The latter forgo optical measurements, but still rely on chemical reagents and suffer from inadequate differentiation.

Impedance cytometry in its most basic form applies the Coulter principle. As a particle (or cell) of diameter dp passes through an aperture of diameter DA between two chambers, it causes a change in impedance ΔZ measured between two electrodes on either side of the aperture. A first-order approximation for this change is ΔZ=4ρ_md_p³/πD_A⁴, where ρ_mis the resistivity of the electrolyte.

Consequently, this signal can be used to differentiate particles based on their size. This is useful for blood cell differentials, as there are significant differences in geometry—discoid red blood cells with 6-8 μm diameters, compared to spherical white blood cells with diameters ranging from 6-20 μm for the various sub-populations. Red blood cells, outnumbering white blood cells approximately 1000:1, all but prohibit an accurate count of different leukocyte sub-types in whole blood. More elaborate implementations of the impedance method can give additional information about the cells: while direct current (DC) or low frequency alternating current (AC) impedance is sensitive to size, higher frequency AC probes the internal structure of the cell. Leukocyte sub-populations can be distinguished by combining these modes, especially when the much more numerous erythrocytes are lysed by addition of chemical reagents to reduce interference. Recently, an integrated LOC system based on impedance cytometry has been shown to be capable of quantifying the different types of blood cells, including neutrophils [12, 13]. A notable limitation is the reliance on chemicals to achieve erythrocyte lysis and sufficient cell type differentiation (saponin and formic acid, followed by sodium carbonate after a set exposure time).

Extending into the alternating current domain allows for probing more generalized changes in dielectric properties caused by particles within the interaction volume [3].

Multi-frequency impedance cytometry has been presented as an attractive method for multi-dimensional single-cell analysis in LOC systems [4], [5]. However, current implementations still suffer from limited resolution, and employ multi-layer fabrication processes. While flow focusing has been utilized to enhance the performance of coulter counter-type devices, to date no systematic study has been conducted on the interplay between flow ratios, particle sizing sensitivity, and throughput [6]. It is only through such studies, both in models and experiments, that optimal utilization of microsystem capabilities becomes possible.

SUMMARY

The embodiments of the present disclosure provide a novel and non-obvious solution to the problems of mental health treatment as described above by providing a point of care testing (POCT) device that includes a whole blood inlet port in fluidic communication with microchannels extending therefrom.

The embodiments of the present disclosure provide a point of care testing (POCT) device that limits the amount of required chemicals, as additional reagents complicate LOC packaging.

The embodiments of the present disclosure provide a point of care testing (POCT) device that eliminates the need for multi-layer fabrication processes that represent a practical drawback in terms of scale-up.

The embodiments of the present disclosure provide reagent- and label-free assay (only water) in conjunction with impedance cytometry.

Integration of pure water hydrodynamic focusing to enhance signal-to-noise ratio.

Integration of pure water erythrocyte lysis to eliminate background signal and enhance white blood cell differentiation.

Two-layer design with polydimethylsiloxane (PDMS) channels and coplanar gold electrodes on glass for simple, low-cost fabrication.

Consequently, one embodiment of the present disclosure relates to a method of establishing a differential white blood cell count that includes directing at least one stream of deionized water into a microfluidic device containing a sample of whole blood of a subject or a cell-rich fraction of a whole blood sample or a cell-free fraction of whole blood of a subject or combinations thereof to generate a lysate stream of intact white blood cells; directing at least one stream of deionized water into the lysate stream such that the lysate stream with intact white blood cells is forced to flow in a direction of motion by the at least one stream of deionized water to form a virtual non-conductive aperture in a channel of the microfluidic device; and performing impedance cytometry of the lysate stream in the virtual non-conductive aperture via coplanar electrodes to detect the presence of intact white blood cells in the lysate stream.

The method may further include quantitatively differentiating between neutrophils, lymphocytes, monocytes, eosinophils, and basophils in the lysate stream based on the impedance measurements resulting from the performance of the impedance cytometry.

Additionally, the step of directing at least one stream of deionized water into the channel may include symmetrically focusing at least two streams of deionized water orthogonally on opposing sides of the direction of motion of the lysate stream to form the virtual non-conductive aperture.

Yet another embodiment of the present disclosure relates to a method of fabricating a microfluidic device that include forming a layer of material on a substrate and adhering a plurality of pairs of co-planar electrodes on the substrate; and forming a plurality of microchannels in the layer of material. At least one of the microchannels is configured and disposed to receive at least one stream of deionized water to effect lysis of a whole blood sample or of a cell-rich fraction of a whole blood sample to produce a lysate stream. At least one of the microchannels is configured and disposed to receive the lysate stream and to receive at least one focusing flow of deionized water to effect a virtual aperture. At least one the pairs of co-planar electrodes is formed under one of the plurality of microchannels in which is generated the virtual aperture such that impedance cytometry of the lysate stream in the virtual aperture is enabled by application of an electric field to at least two pairs of the plurality of pairs of co-planar electrodes.

The step of adhering a plurality of pairs of co-planar electrodes on the substrate may include applying a chrome adhesive between the plurality of pairs of co-planar electrodes and the substrate.

Still another embodiment of the present disclosure relates to a microfluidic device that includes a layer of material formed over a substrate. A blood separation section is configured and disposed in the layer of material to receive a sample of whole blood of a subject and to separate the whole blood sample into a cell-free fraction and into a cell-rich fraction. An analyte sensor section is configured and disposed in the layer of material to detect an analyte in the cell-free fraction via application of an electrical field and detection of changes in at least one electrical property in the analyte. A cell pre-treatment section is configured and disposed in the layer of material to form a lysate from the cell-rich fraction; and a cell or large particle analyzer section configured and disposed on the layer of material to enable analysis of the lysate from the cell-rich fraction to detect circulating tumor cells or white blood cells including neutrophils, lymphocytes, monocytes, eosinophils, and basophils.

The cell or large particle analyzer section may be configured and disposed on the layer of material to enable analysis of the lysate from the cell-rich fraction to enable a differential white blood cell count via coplanar electrodes formed over the substrate that are configured and disposed to enable impedance cytometry of the white blood cells in the cell or large particle analyzer section.

A further embodiment of the present disclosure relates to a microfluidic device for establishing a differential white blood cell count that includes a substrate. A layer of material is formed over the substrate and a plurality of microchannels is formed in the layer of material. At least one of the plurality of microchannels is configured and disposed to receive a sample of whole blood of a subject or a cell-rich fraction of a whole blood or combinations thereof. At least one of the plurality of microchannels is configured and disposed to receive at least one stream of deionized water to effect lysis of a whole blood sample or of a cell-rich fraction of a whole blood sample to produce a lysate stream. At least one of the plurality of microchannels is configured and disposed to receive the lysate stream and to receive at least one focusing flow of deionized water to effect a virtual aperture. At least one the pairs of co-planar electrodes is formed under one of the plurality of microchannels in which is generated the virtual aperture such that impedance cytometry of the lysate stream in the virtual aperture is enabled by application of an electric field to at least two pairs of the plurality of pairs of co-planar electrodes.

With respect to the at least one of the plurality of microchannels that is configured and disposed to receive the lysate stream and to receive at least one focusing flow of deionized water to effect a virtual aperture, the plurality of microchannels may include at least two deionized water injection channels and a lysate stream channel such that the at least two deionized water injection channels are configured and disposed to symmetrically focus at least two streams of deionized water orthogonally on opposing sides of a direction of motion of the lysate stream in the lysate stream channel.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other advantages will become more apparent from the following detailed description of the various embodiments of the present disclosure with reference to the drawings wherein:

FIG. 1 illustrates a flow chart of an integrated cell-free and cell-rich fraction microfluidic device and testing interface for implementing a method of testing a cell-free fraction and a cell-rich fraction of a whole blood sample of a patient or subject;

FIG. 2A is a perspective view of a microfluidic device that is functionally independent of the integrated microfluidic device of FIG. 1 but is functionally equivalent to the cell-free fraction analysis section of the integrated microfluidic device of FIG. 1;

FIG. 2B is a plan view of the microfluidic device of FIG. 2A;

FIG. 2C is a cross-section view of the microfluidic device of FIG. 2A taken along section line 2C-2C;

FIG. 3 illustrates a microfluidic device that is also physically independent of the integrated microfluidic device of FIG. 1 but which is functionally equivalent to the cell-rich fraction analysis section of the integrated microfluidic device of FIG. 1;

FIG. 4 illustrates an alternate embodiment of the microfluidic device of FIG. 3;

FIG. 5 illustrates one embodiment of a portion of the whole blood analysis section of the microfluidic devices illustrated in FIG. 3 and FIG. 4;

FIG. 6 illustrates a 3-dimensional electrodynamic model that simulates a particle of the lysate stream of FIG. 5 in a section of microfluidic channel which defines first and second vertical channel walls wherein the particle is suspended in the gap between two coplanar electrodes;

FIG. 7 is a plot of virtual aperture width and fluid admittance plotted against flow ratio sample to focus;

FIG. 8 is a plot of the absolute value of the change in impedance |ΔZ| at 200 kHz as a percentage with respect to the empty channel impedance plotted against virtual aperture width;

FIG. 9 is a finite element model simulation of the absolute value of the change in impedance in percent for a cell with given properties represented by a radius, membrane capacitance and cytosol conductivity, and single-parameter variations thereof illustrating three distinct frequency regimes where an increase or decrease significantly alters the signal;

FIG. 10 is a plot of impedance at 200 kHz in ohms corresponding to particles passing between the electrodes as a function of time in seconds;

FIG. 11 is a plot of the average impedance at 200 kHz in percent for separate bead populations as a function of flow ratio sample to focus;

FIG. 12 illustrates a perspective view of one embodiment of the integrated microfluidic device described schematically with respect to FIG. 1

FIG. 12A is a cross-sectional view of the microfluidic device of FIG. 12 taken along section line 12A-12A; and

FIG. 12B is a cross-sectional view of the microfluidic device of FIG. 12 taken along section line 12B-12B.

DETAILED DESCRIPTION

A microfluidic device relying solely on impedance measurements to establish a differential white blood cell_count as disclosed herein introduces a number of improvements over previous designs. The design according to embodiments of the present disclosure employs coplanar electrodes, simplifying device assembly as compared to parallel electrodes not least by reducing the number of physical layers from three to two. Furthermore, the flow channels are defined in polydimethylsiloxane (PDMS) fabricated by established molding techniques. This straightforward approach again eliminates complexity over the use of photolithographically patterned polyimide and micromilled polymethyl methylacrylate (PMMA).

Rather than employing chemical reagents to eliminate erythrocyte interference as well as enhance leukocyte differentiation, pure water is employed. For eventual clinical application, limiting the amount of required chemicals is an important consideration. Exposure of the cell stream to pure water creates a strong osmotic gradient across plasma membranes, causing swelling and ultimately lysis [18], [19]. White blood cells are much more resistant to osmotic gradients than red blood cells, with neutrophils surviving more than three times as long as erythrocytes [20].

To accommodate osmotic lysis on chip, similar to the design employed by Zhan et al. to study the phenomenon, streams of pure water are symmetrically introduced to the sample flow a certain distance prior to the electrodes [19]. The distance and flow speeds are tuned such that the osmotic stress exposure prior to impedance cytometry results in lysis of red, but not white, blood cells. The osmotic swelling experienced by the leukocytes is also expected to heterogeneously affect sub-populations such as to further enhance differences probed by impedance cytometry.

The loss in performance by utilizing coplanar compared to parallel electrodes is about 20%, notably decreasing for increasing cell size [15]. To retain or exceed the performance demonstrated by e.g. Holmes et al. channels with smaller dimensions are employed herein, and thus smaller equivalent aperture DA, than their 40×40 μm². The channel height are comparable to the white blood cell diameters at below 20 μm, thus also reducing the impact of vertical cell position in the flow on the measured signal [15]. Lateral constraint is provided not by the channel itself, but rather by sheath flow focusing. This phenomenon relies on laminar flow and introduction of fluid streams to either side of the sample stream to force central alignment of cells [17]. Providing a virtual aperture, in contrast to physical channel confinement, limits the danger of channel clogging [7], [10]. Although the lysis flows have a similar effect close to their introduction to the main sample flow, that focusing effect wears off over the length of the channel. Separate flows also allow for independent adjustment of parameters for lysis and focusing.

In summary, a microsystem is disclosed that relies on impedance measurements to establish a differential white blood cell count, introducing a number of improvements over previous designs. The microsystem enables a method for separating whole blood into a cellular component for neutrophil counting and an undiluted acellular component for analyte detection.

The overall design, incorporating a main sample flow, pure water lysis flows, focusing flows, and impedance cytometry, is schematically illustrated in FIG. 1 and is described in more detail below. Gold coplanar electrodes may be photolithographically patterned on glass, via a chrome adhesive therebetween, and SU-8 photoresist may be used to create a negative master structure on silicon. Positive PDMS microfluidics can thus be molded and cured, and subsequently bonded to the glass reversibly by simple application of pressure, or permanently by prior application of oxygen plasma. The impedance measurements rely on four sequential sets of coplanar parallel electrode pairs—one for direct current (DC) measurements, one for high-frequency alternating current (AC) measurements, the other two as respective references. The references serve to account for the impedance from the acellular component at both DC and AC frequencies, assuming a cell density resulting in spacing between individual cells larger than the electrode gap. At a known flow rate, the different measurements can be correlated for each cell. Sample flow is provided by pressure actuation from external syringe pumps, connected through capillary tubing. The channels are pre-treated with bovine serum albumin (BSA) protein to reduce sticking of blood cells to the highly hydrophobic PDMS.

Embodiments of the microsystem of the present disclosure enable a decrease in fabrication complexity and in reliance on chemicals through a coplanar electrode design and reliance on pure water to lyse erythrocytes, respectively.

Embodiments of the microsystem of the present disclosure incorporate flow focusing of the white blood cell enhanced fraction via hydrodynamic effects of pure water to create a “virtual aperture” to achieve increased, tunable cell characterization performance and throughput.

The present disclosure of an impedance-based microsystem/microdevice for differential white blood cell counts has the following novel features:

Reagent- and label-free assay (only water).

Integration of pure water hydrodynamic focusing to enhance signal-to-noise ratio.

Integration of pure water erythrocyte lysis to eliminate background signal and enhance white blood cell differentiation.

Two-layer design with PDMS channels and coplanar gold electrodes on glass for simple, low-cost fabrication.

Thus, embodiments of the present disclosure of a microsystem relying on impedance measurements to establish a differential white blood cell count introduce a number of improvements over previous designs, such as a decrease in fabrication complexity and a decrease in reliance on chemicals through a coplanar electrode design and instead reliance on pure water to lyse erythrocytes, respectively. At the same time, by incorporating flow focusing, increased, tunable cell characterization performance and throughput are achieved.

Turning first to FIG. 1, there is illustrated a flow chart of an integrated cell-free and cell-rich fraction microfluidic device and testing interface 100 for implementing a method of testing a cell-free fraction and a cell-rich fraction of a whole blood sample of a patient or subject. More particularly, the integrated device and testing interface 100 includes a patient or subject 102.

As defined herein, cell-free fraction refers to a blood sample from which at least 99% of cellular components such as erythrocytes and leukocytes have been removed from a whole blood sample leaving a plasma of less than 1% cellular composition.

As defined herein, cell-rich fraction refers to a whole blood sample from which 20% or less of plasma volume has been removed, leaving a sample containing 99% of cellular components such as erythrocytes and leukocytes, potentially concentrated with respect to typical whole blood.

As defined herein, white blood cells, also referred to as leukocytes, include neutrophils, lymphocytes, monocytes, eosinophils, and basophils, each of which may exist independently in a whole blood sample or cell-rich fraction.

The method includes extracting or receiving a whole blood sample 104 from the patient or subject 102 and directing the whole blood sample 104 to a blood separation section 1002 of integrated cell-free fraction analysis and cell-rich fraction analysis microfluidic device 1000. The whole blood sample 104 may be directed to the intake of the blood separation section 1002 via generally one micropump 105 that may be externally positioned with respect to the microfluidic device 1000, as shown schematically in FIG. 1, or embedded within the microfluidic device 1000 (not shown). The method includes, via the whole blood sample separation section 1002, separating the whole blood sample 104 into a cell-free fraction 1102 and into a cell-rich fraction 1202. The method includes directing the cell-free fraction 1102 to a cell-free fraction analysis section 1100 of microfluidic device 1000 and directing the cell-rich fraction 1202 to a cell-rich fraction analysis section 1200 of the microfluidic device 1000. As described in more detail below with respect to FIG. 2, the method of testing includes sensing in the cell-free fraction analysis section 1100 an analyte or biomarker 1125 such as, e.g, a drug or pharmaceutical, metabolites, vitamins, viruses, bacteria, hormones, enzymes, inflammatory mediators, chemokines, immunoglobulin isotypes, intracellular signaling molecules, apoptotic mediators, adhesion molecules, and antibodies etc. (Morgan et al., Clinical Immunology 2004 March; 110(3) 252-66) via an analyte or biomarker sensor 1110 and directing analyte or biomarker analysis results 1140 as all or part of point-of-care information 106 provided by the microfluidic device 1000. The method of testing also may include directing the cell-rich fraction 1202 to a cell pre-treatment sub-section 1210 of cell-rich analysis section 1200 wherein pre-treatment may include lysis of the cell-rich fraction 1202, directing the lysate with intact white blood cells 1220 to a cell or large particle analyzer sub-section 1230 and directing impedance cytometry results 1240 as all or part of point-of-care information 106 provided by the microfluidic device 1000. The cell or large particle analyzer sub-section 1230 may detect white blood cells (or leukocytes) which include various sub-types such as neutrophils, lymphocytes, monocytes, eosinophils, and basophils, or circulating tumor cells or both white blood cells and circulating tumor cells. The cell or large particle analyzer sub-section 1230 excludes detection of red blood cells since such cells do not survive the lysis process that occurs in cell pre-treatment sub-section 1210.

The integrated device and testing interface 100 may further include directing the part of point-of-care information 106 to a treatment team of medical professionals or researchers 108 who may direct an adjustment in action plan 110 for the patient or subject 102.

FIGS. 2A, 2B and 2C illustrate in more detail the cell-free fraction analysis section 1100 of microfluidic device 1000 which includes analyte or biomarker sensor 1110 for detecting an analyte or biomarker in cell-free fraction 1102. The analyte or biomarker sensor 1110 includes a counter electrode 1130a, a working electrode 1130b and a reference electrode 1130c wherein the analyte or biomarker. 1125 is sensed or detected on the working electrode 1130b by impedance cytometry that involves imposition of an alternating current to the counter electrode 1130a and working electrode 1130b in the presence of reference or ground electrode 1130c.

Microfluidic device 1101 illustrated in FIGS. 2A, 2B and 2C is functionally independent of integrated cell-free fraction analysis and cell-rich fraction analysis microfluidic device 1000 illustrated in FIG. 1 but is functionally equivalent to the cell-free fraction analysis section 1100 of microfluidic device 1000 that includes analyte or biomarker sensor 1110 for detecting an analyte or biomarker in cell-free fraction 1102, as described above with respect to FIG. 1, with the exception that the cell-rich fraction 1202 that is skimmed from whole blood sample 104 is rejected at whole blood sample rejection outlet 1202′ while cell-free fraction 1102 is analyzed via the analyte or biomarker sensor 1110. Whole blood sample 104 is introduced at whole blood sample inlet port 1112 via generally one micropump such as micropump 105 in FIG. 1. The microfluidic device 1101 includes a plasma skimming module 1114, illustrated in the example shown in FIGS. 2A-2C, as a plurality of parallel channels according to one embodiment of the present disclosure, from which the cell-free fraction 1102 is directed as separated plasma 1116 through separated plasma channels 1116′ that intersect, in the example shown in FIGS. 2A-2C orthogonally, the plasma skimming channels 1114. The separated plasma 1116 is directed to flow through the separated plasma channels 1116′ toward analyte detection microfluidic channel or recess or chamber 1118 (see also FIG. 2C) defined in upper or working surface 1120 where analyte or biomarker 1125 is electrochemically detected by analyte sensor 1110 that may be a 3-electrode electrochemical detector 1130 as shown (see also FIG. 2B). Thus, the plasma skimming module 1114 is configured and disposed to separate plasma 1116 from the whole blood sample 104 prior to entry of the whole blood sample 104 into the detection chamber 1118.

The 3-electrode electrochemical detector 1130 includes linear strip electrode 1130a having an arcuately shaped counter electrode tip 1130a′, linear strip electrode 1130b having an arcuately shaped reference electrode tip 1130b′ and linear strip electrode 1130c having a circularly shaped working electrode tip 1130c′ that is disposed in recess 1118 so that the counter electrode tip 1130a′ and the reference electrode tip 1130c′ are concentrically arranged around the working electrode tip 1130b′. The working electrode tip 1130b′ may be modified with a redox cycling system (not shown) to amplify the electrochemical signal of the analyte or biomarker 1125 that is present in the whole blood sample 104. Other systems or methods of amplifying the electrochemical signal may also be employed and a redox cycling system is one example. The linear strip electrodes 1130a, 1130b and 1130c form connections to external electronics, such as a potentiostat (not shown) for signal detection. Following signal detection by the electrochemical analyte detector 1110 for the presence of analyte or biomarker 1125, the separated plasma 1116 is then drawn out through the separated plasma sample outlet port 1124 such as by application of a vacuum connection, not shown, at whole blood sample rejection outlet 1202′ and at separated plasma sample outlet port 1124 or other means known in the art, such as by application of positive pressure via the micropump 105 (see FIG. 1) at whole blood sample inlet port 1112 as described above.

Referring to FIG. 1, the cell-rich fraction analysis section 1200 of integrated microfluidic device 100 relies on impedance measurements to establish a differential white blood cell count. While parallel electrodes as utilized in the prior art offer advantages in terms of accuracy, they are a significant factor in fabrication complexity. By employing coplanar electrodes, device assembly is simplified at least by reducing the number of physical layers required from three to two. Flow channels are defined in polydimethylsiloxane (PDMS) fabricated by established molding techniques, thereby reducing complexity over the use of photolithographically patterned polyimide and micromilled polymethyl methylacrylate (PMMA). Thus, the microfluidic device 1000 for differential white blood cell counting enables low-cost, scalable fabrication.

The loss in performance by utilizing coplanar compared to parallel electrodes is about 20%, notably decreasing for increasing cell size [15]. To retain or exceed the performance over the prior art, channels with height matched to the size of the particle or cell of interest, and thus smaller equivalent aperture D_A, are employed [13]. The channel height is comparable to the white blood cell diameters at below 20 μm, thus also reducing the impact of vertical cell position in the flow on the measured signal [15]. Lateral constraint is provided not by the channel itself, but rather by sheath flow focusing using pure water. This phenomenon relies on laminar flow and, via at least one micropump (not shown) that is generally external to the microfluidic device 1000. In the embodiments of the present disclosure, the micropump is employed to introduce lysis flow of deionized water to create a lysate stream and to cause flow focusing by introducing fluid streams of deionized water to either side of the sample stream to force central alignment of cells [16, 17].

The micropump 105 is employed to introduce the whole blood sample 104 into the blood separation section 1002. The introduction of the lysis flow and of the focusing flows may be accomplished by either a single external pump or separate dedicated pumps, one for the lysis flow and one for the focusing flow or flows.

Providing a virtual non-conductive aperture limits the danger of channel clogging in contrast to physical channel confinement [7, 10]. While this hydrodynamic focusing effect is well-studied, and is applied in bench-top flow cytometers, application of the hydrodynamic focusing effect in a microfluidic device such as microfluidic device 1000 to enhance differential white blood cell detection performance in an impedance-based lab on a chip (LOC) device represents a novel means for differential white blood cell detection.

For differential white blood cell counting, the high background of red blood cells should be eliminated prior to the impedance cytometer. Rather than employing chemical reagents to eliminate erythrocyte interference, pure water is employed. For clinical application, limiting the amount of required chemicals is an important consideration, as additional reagents complicate LOC packaging. Exposure of the cell stream to pure water creates a strong osmotic gradient across plasma membranes, causing swelling and ultimately lysis [18, 19]. White blood cells are much more resistant to osmotic gradients than red blood cells, with neutrophils surviving more than three times as long as erythrocytes [20]. Although the white blood cells also swell due to the osmotic gradient, they survive intact to a much larger degree than red blood cells, thereby enabling the detection process disclosed herein of impedance cytometry.

To accommodate osmotic lysis in the microfluidic device 1000, streams of pure water are symmetrically introduced to the sample flow a certain distance prior to the impedance cytometry region. In conjunction with the operating characteristics of the whole blood sample micropump 105 and the dedicated lysis injection and flow focusing inject water micropump (not shown) as described above, the distance and flow speeds are tuned such that the osmotic stress exposure prior to measurement results in lysis of red, but not white, blood cells. The osmotic swelling experienced by the white blood cells (i.e., leukocytes) heterogeneously affects sub-populations to further enhance differences probed by impedance cytometry. Microfluidic device 1000 represents a novel application of pure water osmotic lysis in a white blood cell counter to enhance the signal-to-noise ratio.

A microfluidic device design incorporating the features described above is shown in FIG. 3. More particularly, FIG. 3 illustrates a microfluidic device 1201 that is also physically independent of integrated cell-free fraction analysis and cell-rich fraction analysis microfluidic device 1000, and is thus a standalone device with respect to microfluidic device 1000, but which is functionally equivalent to the cell-rich fraction analysis section 1200 of microfluidic device 1000. Therefore, microfluidic device 1201 includes cell pre-treatment sub-section 1210 and cellular analysis sub-section 1230 for establishing a differential white blood cell count in the lysate with intact white blood cells 1220, as described above with respect to FIG. 1, but with the exception that instead of separating the whole blood sample 104 from the patient or subject 102 into cell-free fraction 1102 and cell-rich fraction 1202, the whole blood sample 104 is now pumped, via one or more micropumps such as the generally one micropump 105 that may be externally positioned with respect to the microfluidic device 1201, as described above schematically in FIG. 1, or embedded within the microfluidic device 1201 (not shown), directly into the inlet of whole blood receiving channel 104′.

As defined herein, a microfluidic device according to embodiments of the present disclosure may receive a whole blood sample and separate the whole blood sample into a cell-free fraction and into a cell-rich fraction and subject both the cell-free fraction and the cell-rich fraction to electrically-based analysis techniques.

As defined herein, a microfluidic device according to embodiments of the present disclosure may receive a whole blood sample and separate the whole blood sample into a cell-free fraction and into a cell-rich fraction and subject the cell-rich fraction to a means for causing lysis on the cell-rich fraction to form a lysate stream with intact white blood cells.

As defined herein, a microfluidic device according to embodiments of the present disclosure may receive a whole blood sample and subject the whole blood sample to a means for causing lysis on the whole blood sample to form a lysate stream with intact white blood cells without having first separated the whole blood cells into a cell-free fraction and into a cell-rich fraction.

Additionally, as defined herein, a microfluidic device according to embodiments of the present disclosure may receive a whole blood sample and separate the whole blood sample into a cell-free fraction and into a cell-rich fraction and subject only the cell-free fraction to an electrically-based analysis technique or subject only the cell-rich fraction to an electrically-based analysis technique.

Accordingly, the method of testing also may include directing the cell-rich fraction 1202, or, in the embodiment of the microfluidic device 1201 of FIG. 3, directing the whole blood sample 104 to a cell pre-treatment sub-section 1210 of cell-rich analysis section 1200 wherein pre-treatment may include lysis of the whole blood sample 104, directing the resulting lysate with intact white blood cells 1220 to a lysate analysis or impedance cytometry sub-section 1230 and directing impedance cytometry results 1240 as all or part of point-of-care information 106 provided by the microfluidic device 1000.

Consequently, microfluidic device 1201 in FIG. 3 illustrates in more detail the cell-rich fraction analysis section 1200 of microfluidic device 1000 as formed in cell-rich fraction analysis section microfluidic layer 1201′ (for illustration purposes) as a whole blood sample analysis section 12001. Microfluidic device 1201 includes sample flow channel 1040′ that is configured and disposed to receive a whole blood sample 1040 that is directed into the sample flow channel 1040′ and wherein a first lysis flow channel 1204a′ receives a first lysis flow 1204a and a second lysis flow channel 1204b′ receives a second lysis flow 1204b′ such that the first lysis flow channel 1204a′ and the second lysis flow channel 1204b′ intersect on opposing sides the sample flow channel 1040′ in a quasi-tee or converging forked configuration 1205 to enable mixing of the sample flow 1040 with the first lysis flow 1204a and with the second lysis flow 1204b. The first lysis flow 1204a and the second lysis flow 1204b may be of de-ionized water. The sample mixture 1206 is directed into pre-treatment section 1210 that includes a series of channel loops 1208a . . . 1208n, that are sufficient in number and length to provide sufficient exposure duration time of the sample cells in the sample mixture 1206 to the de-ionized water to cause lysis of the erythrocyte cells such that the sample mixture 1206 emerges from the channel loops 1208a . . . 1208n as a lysate stream 1212 with intact white blood cells which are directed into a lysate channel 1212′.

Thus, the lysate stream 1212 is directed into lysate flow channel 1212′ wherein a first focusing flow channel 1214a′ receives one or more focusing flows, e.g., a first focusing flow 1214a and a second focusing flow channel 1214b′ receives a second focusing flow 1214b′ such that, in a similar manner as with respect to the lysis flow described above, the first focusing flow channel 1214a′ and the second focusing flow channel 1214b′ intersect on opposing sides the lysate flow channel 1212′ in a quasi-tee or forked configuration 1208 to enable the lysate stream 1212 to be directed into a lysate stream channel 1220′ that is configured and disposed in the microfluidic layer 1201′ such that a lysate stream 1212″ with intact white blood cells is directed to flow in a direction of motion, as indicated by arrow A, in the lysate stream channel 1220′. At least two deionized water injection channels 1214a′ and 1214b′ are configured and disposed in the microfluidic device 1000 such that at least two streams of deionized water 1214a and 1214b are directed into the lysate stream channel 1220′ to force the lysate stream 1212″ to flow in the direction of motion A between two streams of deionized water 1214a″ and 1214b″, respectively, to form a virtual non-conductive aperture 1222 in the lysate stream channel 1220′.

In one embodiment, the one or more deionized water injection channels 1214a′ and 1214b′ are configured and disposed to symmetrically focus the two or more streams of deionized water 1214a and 1214b orthogonally on opposing sides of the direction of motion A of the lysate stream 1212″ in the lysate stream channel 1220′.

The microfluidic device 1000 further includes an impedance cytometry section 1230 wherein at least two co-planar electrodes, e.g., electrodes 1230a1, 1230a2 or 1230b1, 1230b2 or 1230c1, 1230c2 or 1230d1, 1230d2, are configured and disposed on a surface 1203 of the microfluidic layer 1201′ such that the white blood cells/leukocytes in the lysate stream channel 1220′ are exposed to an alternating current at at least one frequency emitted from the at least two co-planar electrodes 1230a1, 1230a2 or 1230b1, 1230b2 or 1230c1, 1230c2 or 1230d1, 1230d2. The co-planar electrodes are configured in sequential sets of co-planar parallel electrode pairs 1230a1, 1230a2 followed by 1230b1, 1230b2 followed by 1230c1, 1230c2 followed by 1230d1, 1230d2 that are each positioned orthogonally on the surface 1201 such that the lysate stream channel 1220′ crosses over in an orthogonal manner each of the sequential sets of co-planar parallel electrode pairs.

Thus, the impedance measurements rely on sequential sets of parallel electrode pairs—one for low-frequency measurements, e.g., 1230a1 and 1230a2, and one for high-frequency measurements, e.g., 1230b1 and 1230b2. Two additional pairs of electrodes, e.g., 1230c1, 1230c2 and 1230d1, 1230d2, are included as optional references at the respective frequencies to allow for differential measurements, assuming a cell density resulting in spacing between individual cells larger than the electrode gap (see FIGS. 5 and 6 as described below).

At a known flow rate, the sequential measurements can be correlated for each cell. Sample flow is provided by pressure actuation from external syringe pumps, e.g., one or more micropumps 105 as shown in FIG. 1, connected through capillary tubing. External electronics for signal recording, connected to a microprocessor such as a personal computer (PC) running LabVIEW (National Instruments, Inc., Austin, Tex., USA), may be employed for data acquisition. The external electronics may include impedance recording equipment such as an impedance analyzer or LCR meter (e.g., IET/QuadTech 1910/1920 1 MHz LCR Meter, IET Labs, Inc., Roslyn Heights, N.Y., USA).

The presence of the sequential sets of co-planar parallel electrode pairs 1230a1, 1230a2 followed by 1230b1, 1230b2 followed by 1230c1, 1230c2 followed by 1230d1, 1230d2 enables performing impedance cytometry of the white blood cells/leukocytes in the lysate stream 1212″ in the lysate stream channel 1220′ at the one or more frequencies. For example, as illustrated in FIGS. 5 and 6 and described below, as an intact white blood cell traverses into the gap between each pair of electrodes, co-planar parallel electrode pair 1230a1, 1230a2 and co-planar parallel electrode pair 1230b1, 1230b2 may each be operated at, for example, 100 kilohertz (kHz) and (absolute values of) impedance measurements Z in ohms (Ω) or in percent change in (absolute values of) impedance ΔZ may be taken. These measurements may be repeated by co-planar parallel electrode pair 1230c1, 1230c2 and co-planar parallel electrode pair 1230d1, 1230d2 when the intact white blood cell traverses into the respective gap between each pair of electrodes.

Alternatively, as an intact white blood cell traverses into the gap between each pair of electrodes, co-planar parallel electrode pair 1230a1, 1230a2 may be operated at, for example, 100 kilohertz (kHz) and co-planar parallel electrode pair 1230b1, 1230b2 may be operated at, for example, 500 kilohertz (kHz) and (absolute values of) impedance measurements Z in ohms (Ω) or in percent change in (absolute values of) impedance ΔZ may be taken. These measurements may be repeated by co-planar parallel electrode pair 1230c1, 1230c2 operating at 100 kHz and co-planar parallel electrode pair 1230d1, 1230d2 operating at 500 kHz when the intact white blood cell traverses into the respective gap between each pair of electrodes.

The method includes quantitatively differentiating between neutrophils, lymphocytes, monocytes, eosinophils, and basophils in the lysate stream 1212″ based on impedance measurements resulting from the performance of the impedance cytometry as described above.

Upon flow of the lysate stream 1212″ in the lysate stream channel 1220′ across the sequential sets of co-planar parallel electrode pairs 1230a1, 1230a2 followed by 1230b1, 1230b2 followed by 1230c1, 1230c2 followed by 1230d1, 1230d2, the lysate stream 1212″ is directed to a waste flow outlet 1224.

FIG. 4 illustrates an alternate embodiment of microfluidic device 1201 described above with respect to FIG. 3, and is thus another example of a standalone device with respect to microfluidic device 1000 in FIG. 1. Microfluidic device 1251 illustrates in more detail the cell-rich fraction analysis section 1200 of microfluidic device 1000 as formed in microfluidic layer 1251′ (for illustration purposes) as a whole blood sample analysis section 12002. More particularly, whole blood sample analysis section 12002 is identical to whole blood sample analysis section 12001 in FIG. 3 except that whole blood sample analysis section 12002 includes a common water lysis flow inlet 1204 for the first lysis flow channel 1204a′ that receives first lysis flow 1204a and for the second lysis flow channel 1204b′ that receives second lysis flow 1204b′.

Similarly, whole blood sample analysis section 12002 includes a common water focus flow inlet 1214 for the first focusing flow channel 1214a′ that receives first focusing flow 1214a and for the second focusing flow channel 1214b′ that receives second focusing flow 1214b′.

Additionally, whole blood sample analysis section 12002 formed in microfluidic layer 1251′ further includes the sequential sets of co-planar parallel electrode pairs 1230a1, 1230a2 followed by 1230b1, 1230b2 followed by 1230c1, 1230c2 followed by 1230d1, 1230d2 that are respectively connected to a power supply and impedance recording equipment (not shown), such as an impedance analyzer or LCR meter, (e.g., IET/QuadTech 1910/1920 1 MHz LCR Meter, IET Labs, Inc., Roslyn Heights, N.Y., USA) via connections and pads 1230a10 and 1230a20 for electrodes 1230a1 and 1230a2, respectively, connections and pads 1230b10 and 1230b20 for electrodes 1230b1 and 1230b2, respectively, connections and pads 1230c10 and 1230c20 for electrodes 1230c1 and 1230c2, respectively, and connections and pads 1230d10 and 1230d20 for electrodes 1230d1 and 1230d2, respectively. Although not obvious from FIG. 4 due to the microscopic scale of the electrodes and channels, the electrodes 1230a1, 1230a2, 1230b1, 1230b2, 1230c1, 1230c2, 1230d1, 1230d2 are positioned under lysate stream 1220 in the same sequential manner as displayed in FIG. 3. The connection pads 1230a1, 1230b1, 1230c1 and 1230d1 are disposed on the left side of lysate stream 1212″ with respect to the downstream direction of flow while connection pads 1230a2, 1230b2, 1230c2 and 1230d2 are disposed on the right side of lysate stream 1212″ with respect to the downstream direction of flow.

Again, the impedance measurements rely on sequential sets of parallel electrode pairs—one for low-frequency and one for high-frequency measurements. Two additional pairs of electrodes are included as optional references at the respective frequencies to allow for differential measurements, assuming a cell density resulting in spacing between individual cells larger than the electrode gap. At a known flow rate, the sequential measurements can be correlated for each cell. Sample flow is provided by pressure actuation from external syringe pumps, connected through capillary tubing. External electronics are utilized for signal recording, connected to a PC running LabVIEW for data acquisition.

The lysate stream 1212″ with intact white blood cells is directed to flow in the direction of motion, as indicated by arrow A, in the lysate stream channel 1220′. At least two streams of deionized water 1214a and 1214b are directed into the lysate stream channel 1220′ such that the lysate stream 1212″ is forced to flow in the direction of motion A between two streams of deionized water 1214a″ and 1214b″, respectively, to form the virtual non-conductive aperture 1222 in the lysate stream channel 1220′.

It should be noted that although the foregoing and subsequent description of microfluidic devices 1201 in FIGS. 3 and 1251 in FIG. 4 is presented as representative examples of the lysis sub-section 1210 to form lysate stream 1212 and of the lysate analysis section 1230 that enables analysis of the compressed lysate stream 1212″ from the cell-rich fraction 1202, the integrated microfluidic device 1000 need not be limited to detection of white blood cells but may also be applied to cytometry of other cells such as the known various sub-types of white blood cells and circulating tumor cells and except for red blood cells since the lysis is intended to remove such cells from the lysate stream 1212, Other methods cell detection for the integrated device 1000 may include visual or optical detection via observation of the lysate stream 1212 under a microscope.

Fabrication & Instrumentation

Gold coplanar electrodes were photolithographically patterned on a glass or silicon oxide substrate as one example, and SU-8 photoresist was used to create a negative master structure on silicon. Positive PDMS microfluidics can thus be molded and cured, and subsequently bonded to the glass reversibly by simple application of pressure, or permanently by prior application of oxygen plasma and thus fabricated by standard microfabrication approaches. Gold could conceivably also be another chemically inert conductor. As described above, the impedance measurements rely on the sequential sets of parallel electrode pairs—one for low-frequency and one for high-frequency measurements. Two additional pairs of electrodes are included as optional references at the respective frequencies to allow for differential measurements, assuming a cell density resulting in spacing between individual cells larger than the electrode gap. At a known flow rate, the sequential measurements can be correlated for each cell. Sample flow is provided by pressure actuation from external syringe pumps, connected through capillary tubing. Again, external electronics for signal recording, such as a potentiostat, impedance analyzer, or LCR meter as described above are connected to a PC running LabVIEW for data acquisition.

For the microfluidic layer, a mold was created using SU-8 2015 negative photoresist patterned on silicon using contact photolithography. Using this master, channels were cast from poly(dimethylsiloxane) (PDMS). After thermal curing at 60° C., the PDMS was diced and 2 mm diameter fluidic connections were punched.

Referring to FIGS. 5 and 6, thus the microfluidic devices 1201 and 1251 illustrated in FIGS. 3 and 4, respectively, and as further described below with respect to microfluidic device 1000 in FIGS. 12, 12A and 12B, comprise two physical layers. Four pairs of microelectrodes 1230a1, for impedance measurements 1230a1, 1230a2 followed by 1230b1, 1230b2 followed by 1230c1, 1230c2 followed by 1230d1, 1230d2 each have a width dimension We with gap G between the two electrodes in each pair and are disposed on a lower or first layer of glass or silicon oxide layer via a chrome adhesive therebetween. In one embodiment, the width We is about 25 μm and the gap G is also about 25 μm. The microfluidic channels are formed in an upper or second layer of PDMS wherein the microfluidic channels are formed with a cross-section of 75×20 μm2 (width×height). The lysate stream channel 1220′ has a channel width W of about 75 μm and a height H of about 20 μm. The upper or second layer of PDMS is formed on both the electrodes and the lower or first layer substrate of glass or silicon oxide. The PDMS is plasma-bonded to the glass and also seals and adheres to the gold coplanar electrodes.

FIG. 5 illustrates one embodiment of a portion of the whole blood analysis section 1200 of microfluidic device 1201 as illustrated in FIG. 3 and microfluidic device 1251 as illustrated in FIG. 4. Water focusing flows 1214a and 1214b are directed into the lysate stream 1212 in lysate channel 1212′. Lysate channel 1212′ intersects between first focusing flow channel 1214a′ and second focusing flow channel 1214b′ in a crossed intersection 1205 such that when the two or more streams of deionized water 1214a and 1214b are directed into the lysate stream channel 1220′, a compressed lysate stream 1212″ is forced to flow in the direction of motion A between two streams of deionized water 1214a″ and 1214b″, respectively, to form virtual non-conductive aperture 1222 in the lysate stream channel 1220′.

The lysate stream channel 1220′ is positioned over sequential sets of co-planar parallel electrode pairs 1230a1, 1230a2 followed by 1230b1, 1230b2 such that a detection region 1231 for white blood cells is formed by the gap G between the set of co-planar parallel electrode pairs 1230a1, 1230a2 and by the gap G between the set of co-planar parallel electrode pairs 1230b1, 1230b2. Detection of white blood cells occurs by electric fields from the sequential sets of co-planar parallel electrode pairs 1230a1, 1230a2 and 1230b1, 1230b2 propagating through the virtual aperture 1222 in the gaps G.

Referring also to FIG. 6, in one embodiment, the presence of the sequential sets of co-planar parallel electrode pairs 1230a1, 1230a2 followed by 1230b1, 1230b2 followed by 1230c1, 1230c2 followed by 1230d1, 1230d2 enables performing impedance cytometry of the white blood cells/leukocytes in the lysate stream 1212″ in the lysate stream channel 1220′ at the one or more frequencies. For example, as an intact white blood cell 12120 traverses into the gap G between each pair of electrodes, co-planar parallel electrode pair 1230a1, 1230a2 and co-planar parallel electrode pair 1230b1, 1230b2 may each be operated at, for example, 100 kilohertz (kHz) and (absolute values of) impedance measurements Z in ohms (Ω) or in percent change in (absolute values of) impedance ΔZ may be taken. These measurements may be repeated by co-planar parallel electrode pair 1230c1, 1230c2 and co-planar parallel electrode pair 1230d1, 1230d2 when the intact white blood cell 12120 traverses into the respective gap G between each of those pairs of electrodes.

In still another embodiment, as an intact white blood cell 12120 traverses into the gap G between each pair of electrodes, co-planar parallel electrode pair 1230a1, 1230a2 may be operated at, for example, 100 kilohertz (kHz) and co-planar parallel electrode pair 1230b1, 1230b2 may be operated at, for example, 500 kilohertz (kHz) and (absolute values of) impedance measurements Z in ohms (Ω) or in percent change in (absolute values of) impedance ΔZ may be taken. These measurements may be repeated by co-planar parallel electrode pair 1230c1, 1230c2 operating at 100 kHz and co-planar parallel electrode pair 1230d1, 1230d2 operating at 500 kHz when the intact white blood cell traverses into the respective gap between each pair of electrodes. It is assumed that each electrode pair operates at one specific frequency, but as the intact white blood cell 12120 travels through lysate stream channel 1220′ the cell will experience the particular operating frequency of each pair of electrodes.

For both of the foregoing methods of measuring changes in impedance, the measurements at the coplanar electrode pair 1230b1, 1230b2 and at coplanar electrode pair 1230d1, 1230d2 are considered to be “empty channel” readings since the microfluidic devices 1201 and 1251 should be designed such that statistically it is anticipated that while an intact white blood cell 12120 traverses into the gap G between electrode pair 1230a1, 1230a2, or between electrode pair 1230c1, 1230c2, no particle is anticipated to be present in the gap G between electrode pair 1230b1, 1230b2 or electrode pair 1230d1, 1230d2, respectively while the impedance measurements are being recorded.

Modeling

Extensive use of finite element modeling (FEM) was made, in combination with equivalent circuit modeling, to guide the design process, as described below. Critical parameters such as channel cross-section and electrode gap were chosen based on model optimization. Finite element modeling (FEM) was performed in COMSOL Multiphysics (COMSOL, Inc.; Palo Alto, Calif.), using the MEMS and Microfluidics packages.

The 2D (two-dimensional) hydrodynamic model considered a slow-diffusing species (particles) and a fast-diffusing species (ions) introduced through a center sample channel, focused symmetrically by deionized water (DI-H2O) flows.

A representative simulation is shown in FIG. 5. The main model outputs of interest are the cross-sectional concentration profiles downstream from the flow focusing inlets—the respective full width at half maximum (FWHM) for the ions can be considered equivalent to the VA width. The smaller the dimension of the virtual aperture VA, and thus the dimension of FWHM, the greater the accuracy of the impedance measurements since the impedance measurements mainly only reflect the intact white blood cell 12120 in contrast to the impedance of the two streams of deionized water 1214a″ and 1214b″. As an additional means of increasing accuracy of the impedance measurements, the distance X between the convergent tee 1208 and the upstream edge 1230a1′ of the first electrode 1230a1 is also designed to a minimum value so that the impedance measurements occur at a position before significant divergence of the virtual aperture VA occurs downstream of the final electrode 1230d2. The spacing between the coplanar electrodes 1230a1, 1230a2 followed by 1230b1, 1230b2 followed by 1230c1, 1230c2 followed by 1230d1, 1230d2 generally does not significantly affect the impedance measurements as long as the electrodes are located before any significant divergence of the virtual aperture VA occurs.

FIG. 6 illustrates a perspective sample model geometry which represents the detection portion 1231 identified in FIG. 5. The model output is the change in impedance |ΔZ| measured across electrodes between particle and no-particle conditions.

Referring also to FIG. 5, the 3D electrodynamic model simulates particle 12120 of the lysate stream 1212″ having radius r, conductivity a, and permittivity e, suspended in a section of microfluidic channel, i.e., lysate stream channel 1220′ which defines first vertical channel wall 1226a and second vertical channel wall 1226b, in the gap G between two coplanar electrodes, e.g., electrodes 1230a1 and 1230a2. To approximate the impact of hydrodynamic focusing, two distinct environments (apart from the particle) were incorporated within the segment of lysate stream channel 1220′—electrolyte (compressed lysate stream 1212″) in the center at a certain width Wp corresponding to VA, and deionized water focus flows 1214a and 1214b on opposing sides of the compressed lysate stream 1212″. The deionized water focus flows 1214a and 1214b define respectively deionized water focus flow width W_H20a between channel wall 1226a and the compressed lysate stream 1212″ and deionized water focus flow width W_H20b between channel wall 1226b on the opposite side of the compressed lysate stream 1212″ and of the lysate stream channel 1220′. As indicated above, the model output is the change in impedance |ΔZ| measured across electrodes between particle and no-particle conditions.

Thus the virtual aperture VA represents the cross-sectional area defined by the width W and height H of the lysate stream 1212″, excluding the widths W_H2Oaand W_H2Obof the deionized water focus flows 1214a and 1214b in the channel 1220′.

It should also be noted that although the microfluidic devices 1000, 1201 and 1251 are described and illustrated in FIGS. 1-6 and later in FIGS. 12, 12A, 12B below as configured to receive two deionized water lysis flows 1204a and 1204b, only one lysis flow such as 1204a or 1204b is required, or the microfluidic devices 1000, 1201 and 1251 may be configured to receive additional lysis flow or flows (not shown).

Additionally, although the microfluidic devices 1000, 1201 and 1251 are described and illustrated in FIGS. 1-6 and later in FIGS. 12, 12A, 12B below as configured to receive two deionized water focus flows 1214a and 1214b, only one focus flow such as 1214a or 1214b is required, in which case the virtual aperture VA occurs directly between second vertical channel wall 1226b and focus flow 1214a or between first vertical channel wall 1226a and focus flow 1214b.

Alternatively, the microfluidic devices 1000, 1201 and 1251 may be configured to receive additional focus flow or flows (not shown).

Experiments

The fabricated microfluidic devices were connected to syringes using Tygon tubing (Cole-Parmer; Vernon Hills, Ill., USA). Constant flow was provided through syringe pumps (KDS230 (KD Scientific, Inc.; Holliston, Mass., USA), Genie Plus (Kent Scientific Corporation; Torrington, Conn., USA), NE-300 (New Era Pump Systems, Inc.; Farmingdale, N.Y., USA)). Admittance measurements for model verification were done using a VSP-300 potentiostat (Bio-Logic; Claix, France).

Impedance cytometry data was recorded via LabView utilizing an E4980A Precision LCR Meter (Agilent; Santa Clara, Calif., USA). The background signal was determined through MATLAB (MathWorks, Inc.; Natick, Mass., USA) robust local regression smoothing of the raw data, the signal peaks using a peak finding algorithm. Population averages were calculated using histogram peak fits in OriginPro (OriginLab Corporation; Northampton, Mass., USA).

Prior to use, the LOCs were rinsed with Fetal Bovine Serum (FBS; Life Technologies; Carlsbad, Calif., USA) to reduce PDMS hydrophobicity. Polystyrene particles (r=3 μm and 5 μm; sulfate-type) were purchased from Life Technologies (Carlsbad, Calif., USA) and suspended in phosphate-buffered saline (PBS; 1× from tablet; Sigma-Aldrich; St. Louis, Mo., USA). To reduce settling velocity through density matching, sucrose (Sigma-Aldrich; St. Louis, Mo., USA) was added to 14% w/v. All solutions were based on DI-H2O (ρ=18 Ωcm).

It should be noted that although the impedance cytometry measurements are generally recorded via application of alternating current (AC), it is possible to record impedance cytometry measurements via direct current (DC) although generally the signal-to-noise ratio is reduced as compared to the AC measurements. In the case of DC impedance cytometry, the impedance measurements are a static measurement of resistance R based on R=V/I with respect to time, where V is the applied voltage and I is the measured current.

Results and Discussion

Hydrodynamic Model and Validation

Referring to FIG. 7, the flow rates are the main external parameters to control the VA width, the critical parameter for impedance cytometry performance. To elucidate their correlation, hydrodynamic FEM was utilized to determine the ionic FWHM for a range of flow ratios (FR) of phosphate-buffered saline (PBS)-based sample to DI-H2O focus. In FIG. 7, the left vertical axis is virtual aperture width VA (or Wp) in microns (μm) plotted against horizontal axis of flow ratio sample to focus FR (1:x) where the FEM results are indicated in circles and the experimental results are indicated by crosses. The results are in qualitative agreement with theory [7]. As expected intuitively, the virtual aperture width VA (or Wp) decreases drastically over 80 microns at zero FR to approximately 3-4 microns as flow focusing is introduced, with the effect saturating at high flow ratios. The behavior is independent of flow rate, at least in the laminar flow regime.

To verify these results in the experimental microfluidic device that was utilized, a pure PBS sample flow (10, 20, 50, 100 μl/h) and DI-H2O focus flows (100 μl/h combined) were introduced. To achieve FR=0, PBS was substituted for the deionized water DI-H2O. The left vertical axis is fluid admittance at 200 kHz in microsiemens (μS). The admittance of the fluid across electrodes, which at 200 kHz is dominated by ionic conduction, was measured. Thus, this parameter is expected to linearly correlate with the ionic FWHM. Indeed, the overlaid experimental data as represented by crosses in FIG. 7 aligns well with experimental model results as represented by circles, wherein the fluid admittance exceeds 35 μS at zero FR and decreases to about 5 μS at FR equal to approximately 17.5. The fact that the measured trend is toward a non-zero value at high FR can be attributed to parasitic currents in the real-world instrument-microfluidic device circuit.

Electrodynamic Model

To illustrate the advantages of hydrodynamic focusing in impedance cytometry, electrodynamic FEM is relied upon although analytical modelling may also be employed. In FIG. 8, the vertical axis is the absolute value of the change in impedance |ΔZ| at 200 kHz as a percentage with respect to the empty channel impedance Z plotted against the horizontal axis of virtual aperture width in microns (μm). The cell radius R is 50 μm and the results are plotted for a 25 μm channel width (crosses) and for a 50 μm channel width (boxes) The relative |ΔZ| (i.e., as a percentage of the empty-channel Z) induced by an r=5 μm particle for a range of VA (or Wp widths is displayed. A frequency of f=200 kHz was found to be most sensitive to resistive properties, and thus r, and this was used throughout this work. The plot shows data for channel widths of 25 μm and 50 μm, revealing the signal is independent of the actual channel width (memory constraints prevented simulations for 75 μm width). Therefore, at the chosen f, the VA is expected to function in a manner identical to a physical constriction. As expected, the relative |ΔZ| induced by an r=5 μm particle increases significantly with decreasing VA. Specifically, reducing the aperture from 50 μm to 5 μm enhances the signal 10-fold from |ΔZ|=1.8% to |ΔZ|=18%. At very low VA≦r, the signal saturates, which can be attributed to the fact that in this regime, only part of the particle is in electrolyte and thus contributing to the signal. It is expected that in reality, a boundary layer of PBS would surround the particle, which would alter the results. However, considering the underlying approximation of a well-defined boundary between PBS and DI-H2O in this model, the additional error introduced by the omitted particle boundary layer is likely negligible.

Model Predictions:

Electrodynamic finite element modeling (FEM) effectively illustrates the expected utility of flow focusing in impedance cytometry. The relative ΔZ at 200 kHz induced by an R=5 μm cell, plotted in FIG. 9, is increased up to 10-fold through flow focusing, and independent of the actual channel width.

The signal dependence on different cell parameters is illustrated in FIG. 9 which is a plot of FEM simulation of |ΔZ| for a cell with given r, ∈_mem, and σ_cyt(solid black), where R=5 μm; ∈_mem=11.3 ∈₀; σ_cyt=0.6 S/m (siemens/meter) versus frequency in Hz (plot 90).

As compared to the plot 90, low frequencies up to around 50 kHz, region 901, are most sensitive to cell size, i.e., cell radius R=3.5 μm (plot 91) and radius R=6.5 μm (plot 92), while higher frequencies around 500 kHz, region 902, respond to changes in membrane capacitance ∈_memwhere ∈_memis the cell permittivity of the cellular membrane, ∈₀is permittivity of the vacuum for ∈_mem=5.65 ∈₀(plot 93) and ∈_mem=22.6 ∈₀(plot 94), and to cytoplasm conductivity σ_cytwhere σ_cyt=0.3 S/m (siemens/meter) (plot 95) and σ_cyt=1.2 S/m) (plot 96) around 5 MHz, region 903.

This allows for the critical blood cell type differentiation based on multi-frequency measurements. The observed impact of cell size on the entire frequency range can be corrected for by considering impedance ratios, such as ΔZ_{500 kHz}/ΔZ_{50 kHz}[15]. Preliminary consideration of the ionic double-layer by coupling FEM to a circuit model predicts an overall upward shift of those frequencies of highest sensitivity.

Preliminary Experiments

Preliminary experiments were conducted with polystyrene beads of sizes r=3 μm and 5 μm suspended in buffer solutions in a prototype device incorporating impedance cytometry and flow focusing, as partially depicted in FIGS. 5 and 6 as described above. The prototype device served to validate FEM results, and further to study the interplay of flow focusing and impedance cytometry.

While hydrodynamic focusing has been utilized to enhance the performance of coulter counter-type devices, no systematic study has been conducted on how flow ratios and geometry affect particle sizing sensitivity. Although FEM can give valuable insights into potential trends, these approximations are unlikely to capture the entirety of the system coupling.

FIG. 10 illustrates the impedance-based particle counting principle using the prototype device of FIGS. 5 and 6 with a mixture of both r=3 μm and 5 μm bead populations at a sample flow of 45 μl/h without flow focusing. The MATLAB-processed data shows distinct peaks in impedance |ΔZ| at 200 kHz in ohms (Ω) corresponding to particles passing between the electrodes as a function of time in seconds (s). Furthermore, three distinct populations become apparent, corresponding in order of increasing signal to 3 μm beads (|ΔZ|=94±9Ω), clusters of two 3 μm beads (|ΔZ|=162±23Ω) passing through gap G, and 5 μm beads (|ΔZ|=329±29Ω). While clusters of three or four 3 μm beads are statistically unlikely, their mis-identification as a 5 μm bead cannot be ruled out in this data due to their similar volume. Although the 3 μm bead population signal is well-defined, we note a larger spread in the cluster signals, which is in line with their non-spherical shape—based on orientation relative to the electrodes, the signal magnitude is expected to vary, as the electric field is not isotropic. This may enable more definitive differentiation between clusters of small particles and larger single particles through analysis of the transient signal during passage between the electrodes. The background is random noise from the environment etc.

To determine the impact of hydrodynamic focusing on sensitivity of the device, single-population samples of beads may be utilized, and the ratio of focus flow to sample flow (FR) may be varied while keeping the total flow rate (sample+focus) constant at 45 μl/h. From histograms based on data analogous to that shown in FIG. 10, FIG. 11 is a plot of the average impedance |ΔZ| at 200 kHz in percent (%) for separate bead populations as a function of flow ratio sample to focus FR. The graph indicates up to 276% enhanced size-based differentiation, from Δ |ΔZ|=0.55% to Δ |ΔZ|=1.52%. Underlying this are overall increases in |ΔZ| by 277% and 275% for 3 μm and 5 μm beads, respectively. These numbers highlight the improved sensitivity enabled through the presently disclosed approach.

While the trend agrees with modeling in FIG. 9, the magnitudes are lower than predicted. Specifically, at FR=1:7, modeling predicts VA≈8 μm in FIG. 7, and in consequence |ΔZ|≈15% for r=5 μm beads in FIG. 8. This almost order-of-magnitude difference compared to experimental results warrants further investigation. One potential explanation is the aforementioned model assumption of well-defined boundaries between PBS and DI-H2O in the electrodynamic FEM. However, this is unlikely to be solely responsible for the discrepancy.

Experimental causes such as parasitic capacitances, which become more dominant at high absolute Z (correlating with higher FR), will need to be explored.

Overall, separation efficiency increases with FR; at the same time, the sample throughput (equaling sample input flow rate) in the experimental results decreases (as total flow is kept constant). However, the sample flow rate is inherently independent from FR. In the laminar flow regime, sensitivity and throughput are thus decoupled in the LOC or microfluidic devices according to the present disclosure, enabling tailoring of these parameters to the specific experimental needs.

FIG. 12 illustrates a perspective view of one embodiment of the integrated microfluidic device 1000 that has been described schematically with respect to FIG. 1 above. More particularly, integrated microfluidic device 1000 includes whole blood sample inlet port 104′ wherein the whole blood sample 104 from the patient or subject 102 is directed to blood separation section 1002 of microfluidic device 1000.

The integrated microfluidic device 1000 includes an upper layer or microfluidic layer of material 1232 that incorporates cell-rich fraction analysis section microfluidic layer 1201′ described above with respect to FIG. 3.

Whole blood sample separation section 1002 is configured and disposed in the microfluidic layer of material 1232 to receive the sample 104 of whole blood of a subject via the whole blood sample inlet port 104′ and to separate the whole blood sample 104 into cell-free fraction 1102 and into cell-rich fraction 1202.

In a similar manner as described above with respect to FIGS. 2A, 2B, 2C described above, analyte sensor sub-section 1110 is configured and disposed in the microfluidic layer of material 1232 to detect an analyte 1125 in the cell-free fraction 1100.

The analyte sensor sub-section 1110 includes counter electrode 1130a, a working electrode 1130b and a reference electrode 1130c wherein the analyte or biomarker 1125 is sensed or detected on the working electrode 1130b by impedance cytometry that involves imposition of an alternating current to the counter electrode 1130a and working electrode 1130b in the presence of reference or ground electrode 1130c. As described above with respect to FIGS. 1 and 2, again, the analyte or biomarker 1125 may include, e.g, a drug or pharmaceutical, metabolites, vitamins, viruses, bacteria, hormones, enzymes, inflammatory mediators, chemokines, immunoglobulin isotypes, intracellular signaling molecules; apoptotic mediators; adhesion molecules, and antibodies etc.

The microfluidic device 1000 includes, as previously described above with respect to FIGS. 1 and 3-6, the lysis sub-section 1210 that is configured and disposed on the microfluidic layer of material 1232 to form lysate stream 1212 from the cell-rich fraction 1202, and also includes the lysate analysis section 1230 that is configured and disposed on the substrate 1010 to enable analysis of the compressed lysate stream 1212″ from the cell-rich fraction 1202.

FIGS. 12A and 12B are cross-sectional views of the microfluidic device 1000 taken along section line 12A-12A and section line 12B-12B, respectively, wherein the upper layer or microfluidic layer of material 1232 may be fabricated from PDMS or other suitable materials such as moldable plastic. All of the previously described features of the integrated microfluidic device 1000 of FIGS. 1 and 12, or of the microfluidic device 1101 of FIGS. 2A, 2B, 2C, or of the microfluidic device 1201 of FIG. 3 or microfluidic device 1251 of FIG. 4 and as further described in FIGS. 5 and 6 may be fabricated as described above for microfluidic device 1000.

The co-planar electrodes 1130a, 1130b, 1130c and the co-planar electrodes 1230a1, 1230a2, 1230b1, 1230b2, 1230c1, 1230c2, 1230d1, 1230d2 are disposed on a glass substrate 1236 via a chrome adhesive 1234 applied between the lower surfaces of the co-planar electrodes and the upper surface 1236′ of the glass substrate 1236. The microfluidic device 1000, and correspondingly microfluidic devices 1101, 1201 and 1251, is thus a composite 1010 of the glass substrate 1236 and the microfluidic layer 1232 including the coplanar electrodes 1130 for microfluidic device 1101 or coplanar electrodes 1230a1, 1230a2, 1230b1, 1230b2, 1230c1, 1230c2, 1230d1, 1230d2 and connection pads 1230a10, 1230a20, 1230b10, 1230b20, 1230c10, 1230c20, 1230d10, 1230d20 for microfluidic devices 1201 and 1251, as appropriate, and the chrome adhesive layer 1234.

Those skilled in the art will recognize and understand that the design and usage of the microfluidic devices 1000, 1201 or 1251 are based upon calibration of the particular device to known cell types that have been verified to be present in the lysate stream 1212 via standard laboratory techniques.

As can be appreciated from the foregoing, the embodiments of the impedance-based microdevices described herein for differential white blood cell counts present at least the following novel features:

Reagent- and label-free assay (only water).

Integration of pure water hydrodynamic focusing to enhance signal-to-noise ratio.

Integration of pure water erythrocyte lysis to eliminate background signal and enhance white blood cell differentiation.

Two-layer design with PDMS channels and coplanar gold electrodes on glass for simple, low-cost fabrication.

Although the present disclosure has been described in considerable detail with reference to certain preferred version thereof, other versions are possible and contemplated. Therefore, the spirit and scope of the appended claims should not be limited to the description of the preferred versions contained therein.

While several embodiments of the present disclosure have been shown in the drawings, it is not intended that the disclosure be limited thereto, as it is intended that the disclosure be as broad in scope as the art will allow and that the specification be read likewise. Therefore, the above description should not be construed as limiting, but merely as exemplifications of particular embodiments. Those skilled in the art will envision other modifications within the scope of the claims appended hereto.

LIST OF REFERENCES

[1] K. Kottke-Marchant and B. Davis, Laboratory Hematology Practice (John Wiley & Sons, 2012).

[2] H. Craighead, Nature 442, 387 (2006).

[3] L. Rao, B. A. Ekberg, D. Connor, F. Jakubiak, G. M. Vallaro, and M. Snyder, Clinica Chimica Acta 389, 120 (2008).

[4] J. Nielsen, D. Thode, E. Stenager, K. Ø. Andersen, U. Sondrup, T. N.

Hansen, A. M. Munk, S. Lykkegaard, A. Gosvig, I. Petrov, and P. le Quach, European Neuropsychopharmacology 22, 401 (2012).

[5] H. Andersson and A. van den Berg, Sensors and Actuators B: Chemical 92, 315 (2003).

[6] D. A. Ateya, J. S. Erickson, P. B. H. Jr, L. R. Hilliard, J. P. Golden, and F. S. Ligler, Analytical and Bioanalytical Chemistry 391, 1485 (2008).

[7] K. C. Cheung, M. Di Berardino, G. Schade-Kampmann, M. Hebeisen, A. Pierzchalski, J. Bocsi, A. Mittag, and A. Tárnok, Cytometry Part A 77A, 648 (2010).

[8] H. Morgan, T. Sun, D. Holmes, S. Gawad, and N. G. Green, Journal of Physics D: Applied Physics 40, 61 (2007).

[9] M. E. Piyasena and S. W. Graves, Lab on a Chip (2014), 10.1039/c3lc51152a.

[10] T. Sun and H. Morgan, Microfluidics and Nanofluidics 8, 423 (2010).

[11] H. Zhang, C. H. Chon, X. Pan, and D. Li, Microfluidics and Nanofluidics 7, 739 (2009).

[12] D. Holmes, D. Pettigrew, C. H. Reccius, J. D. Gwyer, C. van Berkel, J. Holloway, D. E. Davies, and H. Morgan, Lab on a Chip 9, 2881 (2009).

[13] X. Han, C. van Berkel, J. Gwyer, L. Capretto, and H. Morgan, Analytical Chemistry 84, 1070 (2012).

[14] D. Holmes, J. K. She, P. L. Roach, and H. Morgan, Lab on a Chip 7, 1048 (2007).

[15] S. Gawad, L. Schild, and P. Renaud, Lab on a Chip 1, 76 (2001).

[16] X. Xuan, J. Zhu, and C. Church, Microfluidics and Nanofluidics 9, 1 (2010).

[17] G.-B. Lee, C.-C. Chang, S.-B. Huang, and R.-J. Yang, Journal of Micromechanics and Microengineering 16, 1024 (2006).

[18] J. Wessels, D. Pals, and J. Veerkamp, Biochimica et Biophysica Acta (BBA)—Biomembranes 291, 165 (1973).

[19] Y. Zhan, D. N. Loufakis, N. Bao, and C. Lu, Lab on a Chip 12, 5063 (2012).

[20] H. P. Ting-Beall, D. Needham, and R. M. Hochmuth, Blood 81, 2774 (1993).

[21] M. Evander, A. J. Ricco, J. Morser, G. T. Kovacs, L. L. Leung, L. Giovangrandi, Lab on a Chip, 2013 Feb. 21; 13(4):722-9. doi: 10.1039/c2lc40896a.

[22] Morgan E¹, Varro R, Sepulveda H, Ember J A, Apgar J, Wilson J, Lowe L, Chen R, Shivraj L, Agadir A, Campos R, Ernst D, Gaur A. Clin Immunol. 2004 March; 110(3):252-66.

[23] D. Holmes, D. Pettigrew, C. H. Reccius, J. D. Gwyer, C. van Berkel, J.

Holloway, D. E. Davies, and H. Morgan, Lab on a Chip 9. 2881 (2009).

[24] X. Han, C. van Berkel, J. Gwyer, L. Capretto, and H. Morgan, Analytical Chemistry 84, 1070 (2012).

INTEGRATED AND STANDALONE LABEL AND REAGENT-FREE MICROFLUIDIC DEVICES AND MICROSYSTEMS FOR DIFFERENTIAL WHITE BLOOD CELL COUNTS

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

International Classifications

Abstract

Description

Claims

CROSS-REFERENCE TO RELATED APPLICATIONS

PCT Information

Provisional Applications (1)