A portable impedance microflow cytometer for measuring cellular response to hypoxia

Abstract

This article presents the development and testing of a low-cost (<$60), portable, electrical impedance-based microflow cytometer for single-cell analysis under a controlled oxygen microenvironment. The system is based on an AD5933 impedance analyzer chip, a microfluidic chip, and an Arduino microcontroller operated by a custom Android application. A representative case study on human red blood cells (RBCs) affected by sickle cell disease is conducted to demonstrate the capability of the cytometry system. Impedance values of sickle blood samples exhibit remarkable deviations from the common reference line obtained from two normal blood samples. Such deviation is quantified by a conformity score, which allows for the measurement of intrapatient and interpatient variations of sickle cell disease. A low conformity score under oxygenated conditions or drastically different conformity scores between oxygenated and deoxygenated conditions can be used to differentiate a sickle blood sample from normal. Furthermore, an equivalent circuit model of a suspended biological cell is used to interpret the electrical impedance of single flowing RBCs. In response to hypoxia treatment, all samples, regardless of disease state, exhibit significant changes in at least one single-cell electrical property, that is, cytoplasmic resistance and membrane capacitance. The overall response to hypoxia is less in normal cells than those affected by sickle cell disease, where the change in membrane capacitance varies from -23% to seven times as compared with -17% in normal cells. The results reported in this article suggest that the developed method of testing demonstrates the potential application for a low-cost screening technique for sickle cell disease and other diseases in the field and low-resource settings. The developed system and methodology can be extended to analyze cellular response to hypoxia in other cell types.

Keywords: electrical impedance; hypoxia; microfluidics; portable flow cytometer; sickle cell disease; single-cell analysis.

Figure 1.

Device overview of the portable impedance-based flow cytometer prototype. (A) A double layer PDMS device bonded to a glass substrate patterned with Ti/Au electrodes. The electrodes provide connectivity to the portable device. The top view shows the intersection of the gas channel over the point of measurement between the electrodes. The cross-sectional view depicts the gas exchange between two channels to induce cell sickling. The serpentine shape of the gas channel ensures the deoxygenation of cells before they reach the measurement zone. (B) The flow chart of the portable device consisting of all major components used and how information moves between the components. (C) The Android application used to control the portable device operated to continuously scan for a designated length of time and produce a graph of the results.

Figure 2.

Computational domain and boundary conditions for finite element modeling of PBS impedance.

Figure 3.

(A) Representative microscopic image of RBCs under hypoxia travelling through the microfluidic channel to be measured between two microelectrodes. (B) The impedance magnitude response of a single cell passing through the microelectrode pair using the HF2IS. (C) The impedance magnitude response of a single cell passing through the microelectrode pair using the portable device. The gray signal is the raw response recorded. The blue signal is result of passing the raw response through a level 3 wavelet filter with BlockJS denoising in MATLAB. The portable device and benchtop equipment run at a resolution of approximately 220 samples per second.

Figure 4.

(A) The impedance circuit model when a cell is not present. ${\mathrm{Z}}_{\mathrm{d}\mathrm{l}\mathrm{t}}$ refers to the total double layer impedance. ${\mathrm{Z}}_{\mathrm{P}\mathrm{B}\mathrm{S}}$ represents the impedance of the PBS when no cell is being measured. (B) An impedance circuit model of an RBC in the PBS medium between two electrodes. ${\mathrm{Z}}_{\mathrm{d}\mathrm{l}}$ represents the double layer impedance where the electrodes meet the PBS. ${\mathrm{Z}}_{\mathrm{P}\mathrm{B}\mathrm{S}\mathrm{*}}$ and ${\mathrm{Z}}_{\mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l}}$ are the impedance of the PBS while a cell is present in the channel and the measured RBC, respectfully.

Figure 5.

(A) A MATLAB algorithm is used to identify fourteen peaks from the real part of the calculated cell impedance results obtained from the portable device using a minimum threshold. (B) The corresponding imaginary values were obtained by matching the time coordinate from the real plot.

Figure 6.

Changes in magnitude and phase from baseline readings for detected peaks during impedance measurement. (A) and (B) display the scatter plots for AA1 and AA2, respectively. The linear regression of AA samples with a 95% boundary is depicted on all plots. (C-E) show the scatter plots for samples SS1, SS2, and SS3, respectively. Red percentages represent the conformity score of blood samples under Oxy condition. Blue percentages represent the conformity score of blood samples under DeOxy condition. Solid line is the linear regression of two AA samples. The dashed lines show the 95% confidence bonds.

Figure 7.

Effects of hypoxia and SCD on (A) Single-cell internal resistance and (B) single-cell membrane capacitance. The data includes samples AA1 and AA2 merged, SS1, SS2, and SS3. **** represents p ≤ 0.0001, * p ≤ 0.05, and ns for p > 0.05.