Characterization of the phagocytic ability of white blood cells separated using a single curvature spiral microfluidic device

Abstract

The present work describes a microfluidic device developed for separating white blood cells (WBCs) for the Nitroblue Tetrazolium (NBT) bioassay, which quantifies the phagocytic ability of cells. The NBT test requires a small number of phagocytic cells but is highly susceptible to the presence of red blood cells (RBCs). Our inertial microfluidic device can deliver a WBC sample by removing 99.99% of RBCs and subsequently reducing the ratio of RBC to WBC from 848:1 to 2:3. The microdevice operates on a relatively higher hematocrit concentration (1% Hct) of blood. Compared to conventional WBC separation methods, the microdevice’s passive, label-free nature preserves the cell properties of the original sample. A single-turn spiral microfluidic device with a rectangular cross-section is simple to fabricate, cost-effective, and easy to operate. The reported microfluidic device requires only a single drop of whole blood (⁓20 µl) obtained via the finger prick method for efficient phagocytic analysis. Also, the microdevice reported in this study achieves WBC separation in under 10 min, omitting the need for RBC lysis, density gradient centrifugation, or expensive antibodies.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13534-024-00414-y.

Introduction

White Blood Cells (WBCs) are an integral component of the immune system playing a central role in coordinating an effective response to invading pathogens [1]. WBCs serve as essential biomarkers by offering insightful information regarding the immunological well-being of an individual. Therefore, it is necessary to separate WBCs with high purity to ensure the accuracy of disease diagnosis. Conventional methods employed for WBC separation include centrifugation [2], hemolysis [3], and flow cytometry [4], which requires cost-intensive equipment, skilled technicians, multiple steps and reagents. Microfluidics proves a promising alternative that can eliminate the shortcomings of conventional techniques [5, 6]. Passive microfluidic devices used to separate WBCs are based on microfiltration [7, 8] deterministic lateral displacement (DLD) [9, 10], hydrodynamic techniques [11, 12] and inertial focusing techniques [13–15]. Recently, Abhishek et al. [16] used a size-based microfiltration technique to separate specific lymphocytes by utilizing the RBC rosetting mechanism and achieved 70–90% recovery efficiency. Shin et al. [17] employed a degassed microdevice to self-drive the whole blood sample into the device and isolated specific WBCs, such as NK cells, by utilizing a hydrodynamic technique with a combination of a magnetic field and buffer to remove non-targeted cells. In order to manipulate cell trajectories, the inertial focusing technique, which is reliant on channel geometry and cell properties is a good option [18]. Zhang et al. [19] employed a two-stage serpentine channel with a differential inertial focusing technique and achieved 48% WBC purity. Chiu et al. [20] experimented with a spiral channel to separate lymphocytes and granulocytes with 90% and 70% efficiency by infusing a pre-lysed RBC sample and sheath fluid. Wu et al. [21] used a multi-turn spiral channel with a trapezoidal cross-section and isolated WBCs with an RBC removal efficiency of ⁓98% from diluted blood samples. Similarly, Jeon et al. [22] used a multi-dimensional double spiral device with a combination of rectangular and trapezoidal cross sections to separate WBCs with 85% purity and 99.99% RBC removal efficiency after three sample re-circulations. These microfluidic devices help in understanding WBC activation status and biomarker expressions more conveniently [23, 24].

Phagocytosis is a prime function of WBCs, characterized by the formation of phagolysosome for neutralizing pathogens, as shown in Fig. 1 [25]. The detailed phagocytosis process description is provided in Fig. S1. Phagolysosomes have high Reactive Oxygen Species (ROS) levels because NADPH Oxidase (Nicotinamide Adenine Dinucleotide Phosphate Oxidase) enzymes generate superoxide radicals (O₂^–) [26]. However, ROS generation failure causes chronic granulomatous disease (CGD) [27], leukocyte adhesion deficiencies (LAD) [28], cancers [29], and inflammatory diseases. The phagocytic ability of WBCs can be determined by assessing ROS, providing valuable information on host defense mechanisms. WBCs have a high ROS production, typically 5–7 times that of RBCs after stimulation [30]. Therefore, a few pure WBCs are sufficient for phagocytic analysis, but they must be separated from other blood components such as plasma, RBCs, and platelets. RBCs are the most abundant blood cells, measuring 5–8 μm [31], whereas WBCs are fewer, measuring 7–22 μm in diameter [32]. The RBCs to WBCs ratio in healthy human blood is approximately 1000:1 [33]. Therefore, keeping this ratio as low as possible is imperative to achieve an accurate diagnosis in phagocytic analysis.

Fig. 1.

Process of Phagocytosis-showing neutralizing the pathogen

The present work describes an easy-to-use and simple-to-manufacture microdevice based on inertial microfluidics to separate WBCs for their phagocytic activity. This microdevice design implements only a single spiral turn instead of multi-staging or multi-spiral turns. The designed microdevice with the smallest possible curvature radius, allowing cells to focus inertially within the channel. This single spiral turn with a reduction in curvature radius enhanced the centrifugal, inertial, and secondary (Dean drag) forces acting on blood cells, leading to a more efficient separation process. These strong forces quickly shift a large number of RBCs toward the outer channel wall, making it easy to collect at RBC outlets 1 and 2. Another novel feature of our device is the strategically positioned 1st bifurcation, whose location was determined after extensive testing through simulations and experiments. This bifurcation directs roughly half of the inlet RBCs into RBC outlet 1, increasing WBC purity at the WBC outlet during the 2nd bifurcation. The remaining RBCs move into RBC outlet 2 at the 2nd bifurcation. Additionally, our device features a rectangular cross-section, simplifying manufacturing compared to more complex shapes like trapezoidal or U-shaped designs [13, 22]. The numerical simulations and experiments on fluorescent microparticles were used to mimic RBC and WBC dynamics within the microchannel. Furthermore, 1% Hct blood sample was prepared from ⁓20 µl of unprocessed finger prick blood to validate our microfluidic device. This device can potentially be employed in several clinical assays and diagnostic tests.

Methods and materials

Microchannel design

The microchannel incorporates a spiral design with a single inlet and three outlets, as depicted in Fig. 2(a). The channel features a rectangular cross-section with a width of 0.2 mm at spiral section and 0.1, 0.16, and 0.05 mm at outlets 1, 2, and 3, as illustrated in Fig. 2(b). The channel has a depth of 0.05 mm throughout the microdevice and two bifurcations known as the 1st and 2nd bifurcations. The 1st and 2nd bifurcations have a semi-circular curve with a radius of 0.2 mm. The three outputs are labelled in Fig. 2(a-c) as RO1 (RBC Outlet 1), RO2 (RBC Outlet 2), and WO (WBC Outlet). RO1 is tangent to the spiral channel and connected to 1st bifurcation, whereas RO2 and WO branch out from 2nd bifurcation at the end of the spiral channel. Microdevice fabrication and experimental setup details, including the blood dilution procedure, are provided in S1. The final fabricated PDMS (Polydimethylsiloxane) device is shown in Fig. 2(c).

Fig. 2.

(a) Dean Vortices formed before the 1st and 2nd bifurcation within the channel obtained using simulation (b) Microchannel design with dimensions (c) Fabricated PDMS device with working principle. Here, the channel is highlighted by passing a blue colour dye through it

Sample preparation

An experiment using fluorescent microparticles was conducted with blue 6 μm (Polysciences) and green 10 μm (Bangs Laboratories) carboxylate fluorescent microparticles in deionized water, corresponding to RBC and WBC concentrations in 1% Hct blood (100,000/µl and 110/µl, respectively). Human blood (⁓20 µl) was collected via finger-prick from consenting donors, with ethical approval from BITS-Pilani, K K Birla Goa campus. Blood samples were mixed with Ethylenediaminetetraacetic acid (EDTA) in a 25:1 v/v ratio and diluted with normal saline (0.9% NaCl) to achieve various hematocrit levels (0.1%, 0.5%, 1%, and 2% Hct). For instance, a 1% Hct sample was prepared by mixing 20 µL of 50% Hct blood with 980 µL of saline. Dilutions were confirmed using the microhematocrit tube method. Experiments were conducted at room temperature within 24 h of blood collection.

Working of the microdevice

The microdevice was designed to operate in the inertial microfluidics domain, wherein inertial forces cause a secondary flow composed of two counter-rotating vortices in curved channels. This causes a drag force known as Dean Drag to act on the particles/cells traveling in the flow [34, 35]. In addition, the particles also experience a force due to the parabolic shear flow profile and proximity to the channel walls, termed shear-gradient lift (F_L) and wall-induced lift (F_D) force, respectively [36]. These three forces involved in inertial microfluidics depend on particle size, channel dimensions, and flow velocity [37]. The interplay between these forces causes particles to laterally migrate toward certain equilibrium positions. To understand the effect of F_L and F_D in the microchannel, a set of non-dimensional numbers has been defined. The confinement ratio λ = d_p/d_h compares the particle diameter (d_p) with the channel’s hydraulic diameter (d_h) [34]. The channel Reynolds number , where is the Reynolds number of flow that describes the flow through the microchannel, while the particle Reynolds number , accounts for particle behaviour in the channel. Dean number represents the secondary Dean flow’s strength and depends on the curvature ratio δ . An increase in Re_p has been shown to cause lateral migration of particles to their equilibrium position in microchannels [34, 35]. The magnitude of the net inertial lift was shown to scale with the fourth power of the diameter given in Eq. (1) [40].

1

Where ρ is the fluid density, u is the fluid velocity and C_L is a lift coefficient that depends on Re_c and particle location in the channel cross-section. The Dean drag force scales linearly with the particle diameter given in Eq. (2) [40].

2

where µ is the fluid’s dynamic viscosity, and U_D (= ) is the Dean velocity at the particle location [40]. Based on Eq. (1) and Eq. (2), R_f, the ratio of inertial lift to Dean drag forces is proportional to the cube of the particle diameter, allowing the microdevice to separate blood cells based on their size characteristics. Di Carlo’s work showed that particles were shown to focus if they satisfied the condition λ > 0.07, and Dean flow mixing became significant for De > 20 [35].

Figure 2(a) illustrates a pair of counter-rotating vortices along the horizontal plane formed before the 1st and 2nd bifurcations within the microchannel. A schematic representation of the working of microdevice is shown in Fig. 2(c), where RBCs are present closer to the center of the microchannel and focus toward an equilibrium position near the outside wall under the effect of Dean Drag forces. Furthermore, WBCs concentrate closer to the upper and lower walls of the channel and migrate to an equilibrium region between the channel center and the inner wall. The microdevice operates on roughly 1% Hct blood sample at a 0.4 ml/min flow rate. The 1st bifurcation was added along the spiral length, which resulted in the removal of approximately half of the RBCs at RO1, thereby reducing the RBC to WBC ratio. The 2nd bifurcation of the microdevice can then be utilized to separate residual RBCs focused toward the outer wall into RO2 and a few WBCs into WO. The position of 1st and 2nd bifurcations was designed, evaluated using simulation, validated using fluorescent particles and finally tested with a diluted blood sample.

Performance parameters

The performance of the microdevice was evaluated based on: (1) RBC removal efficiency and (2) WBC purity. WBC purity is an important parameter in many bioassays, and getting higher purity is always a challenging due to the presence of a large number of RBCs. If just one RBC follows each WBC during separation, the purity drops to 50%. Equations (3) and (4) are utilized to evaluate the performance of the microdevice [21, 22]. Samples collected at the inlet and outlets of the microdevice were analyzed using a hemocytometer and an auto cell counter (Countess-3 L, Thermo Fisher Scientific). The results are further validated using a flow cytometer (BD FACSMelody).

3

4

Results and discussion

Herein, we analyze the simulation and experimental data obtained using a spiral microdevice. The performance of the microdevice with variations in hematocrit concentration and flow rates is discussed. Also, NBT assay is utilized to quantify the phagocytic ability of isolated WBCs.

Particle simulation and experimental analysis

COMSOL Multiphysics 6.0 was used for simulating the flow of spherical particles with a similar diameter to RBCs (6 μm) and WBCs (10 μm) at a 0.4 ml/min flow rate (S1). The densities for 6 μm and 10 μm particles were taken as 1100 kg/m³ and 1070 kg/m³, with blood sample viscosity equivalent to 1% Hct as 1.1 cP [38]. The Laminar Flow module was used with P2 + P1 discretization to simulate the fluid flow, and particles were enabled through the Particle Tracing module. The particle positions and trajectories in Fig. 3(a) reveal that 10 μm particles mostly enter RO2 and a few enter WO, whereas 6 μm particles showed a slight bias toward the outer wall of the channel.

Fig. 3.

Simulation and experimental results on spiral geometry obtained at a flow rate of 0.4 ml/min with particle concentration equal to nearly 1% Hct. (a) Simulation results on particle positioning and trajectories at 1st and 2nd bifurcations. (b) Experimental results on fluorescent particles showing trajectories at 1st and 2nd bifurcations- 6 μm particles (blue traces) and 10 μm particles (green traces)

Before conducting each experiment, the microdevice was washed using 1x phosphate buffer saline. Fluorescent particle experiments were conducted to validate the performance of microdevice at the obtained flow rate from simulations. Fluorescent microparticles of sizes 6 μm and 10 μm were used to mimic RBCs and WBCs. The flow rate was set to 0.4 ml/min, and images were captured at 1st and 2nd bifurcation of the channel. Figure 3(b) shows the traces of both particles separately. The 6 μm particles (blue) were focused towards outer wall and half of them were removed at RO1 and the remaining at RO2 subsequently (Video S1). However, few 10 μm particles were seen to travel into RO1, with the majority into RO2 and considerable 10 μm particles into WO (Video S2).

Microdevice performance characteristics

The microdevice depicted in Fig. 2 was tested for WBC separation using diluted blood and the results are shown in Fig. 4. Diluted blood samples with varied hematocrits between 0.1 and 2% Hct were infused at flow rates from 0.2 to 0.8 ml/min. We noted that at 0.2 ml/min and 0.6–0.8 ml/min flow rates, many RBCs entered into WO, thereby reducing the RBC removal efficiency and WBC purity, as shown in Fig. 4(a) and 4(b). The device performed best during 1% Hct with a 0.4 ml/min flow rate. At this condition, the RBC removal efficiency was 99.99% with a WBC purity of 51%, indicating that the microdevice reduced RBC to WBC ratio from 848:1 to 1:1. Here, the initial count (before separation) of RBCs and WBCs are 848 and 1, respectively.

Fig. 4.

Analysis of device performance using an auto cell counter at different flow rates. (a) RBC removal efficiency vs. hematocrits-without sample recirculations (b) WBC purity vs. hematocrits–without recirculations (c) Device performance after sample recirculations (d) Device performance for three different samples containing different RBCs to WBC ratios such as 848:1, 769:1, and 815:1 obtained at 0.4 ml/min flow rate with 1% Hct after four sample recirculations

Apart from RBC removal efficiency and WBC purity, the microdevice had 8% WBC separation efficiency (defined as the ratio of WBCs in the WO sample to the total number of WBCs in the inlet sample). To improve the performance of microdevice, the sample collected at RO1 and RO2 was recirculated. The sample infusion for recirculation was done manually by keeping the same flow rate of 0.4 ml/min. Recirculation improved WBC separation efficiency from 8 to 51% at the end of 4th recirculation, as shown in Fig. 4(c). Also, the purity was increased from 51 to 60%, indicating a reduction in the RBC to WBC ratio from 848:1 to 2:3 with an RBC removal efficiency of 99.99%. This reduction in the ratio suggests that behind every two RBCs, there are three WBCs, which is sufficient for bioassays, including phagocytic enzyme-dependent assay. The sample recirculation ended after the 4th recirculation due to the minimal change in WBC purity from the 2nd to the 4th recirculation. The microdevice needs less than 10 min for WBC separation with four-sample recirculation, which is less time intensive than conventional techniques [17]. Furthermore, as shown in Fig. 4(d), the spiral device was examined using three different samples with varying RBC-to-WBC ratios at 1% Hct and 0.4 ml/min flow rate with four recirculations, demonstrating consistency in device performance.

The experimental images captured at 2nd bifurcation (separation zone) of the microdevice without recirculations are illustrated in Fig. 5. It is difficult to visualize or trace the fluorescent-tagged WBCs at high flow rates; therefore, WBC tracing was done at a lower flow rate of 0.005 ml/min, clearly depicting the challenges involved in WBC separation from the crowd of RBCs, as shown in Fig. 5(a). At such a low flow rate, the WBCs were positioned randomly, while many RBCs were seen moving into WO (Video S3).

Fig. 5.

Experimental photographs of WBC separation at 1% Hct without recirculation. (a) Fluorescence image obtained at a low flow rate (0.005 ml/min) shows the positioning of WBCs (b) Bright-field image obtained at 0.4 ml/min shows shifting of all RBCs towards the outer wall of the channel. WBCs are present but not visible due to the bright field view (c) Sequential steps involved in the WBC separation process

Figure 5(b) shows the bright-field image (without recirculation) captured at an actual flow rate of 0.4 ml/min and 1% Hct, showing almost all RBCs going into RO2 (Video S4). The 1st bifurcation removes almost half of the RBCs into RO1 with a few WBCs. As the remaining cells arrived at the 2nd bifurcation, RBCs were removed into the RO2 of the microdevice with a large number of WBCs, and a few WBCs with a low number of RBCs residing close to the inner wall were received at WO. As noted, the behaviour of blood cells (Fig. 5b) is nearly similar to the behaviour of fluorescent particles (at 2nd bifurcation in Fig. 3b) within the microchannel. The sequential steps involved in WBC separation process is shown in Fig. 5(c). The colourless droplet at WO confirms that the WBC sample contains minimal RBCs (Video S5).

Proposed empirical equation

Empirical equations indicate that a microdevice operates optimally at a flow rate of 0.4 ml/min with a Hct range of 0.1–2%. Polynomial curves fit experimental data well, with an R² of 0.9915 for RBC removal efficiency (Eq. 5) and 0.9999 for WBC purity (Eq. 6). At this flow rate, RBC removal efficiency is 99.99% at 0.1% Hct and 99.88% at 2% Hct, while WBC purity is 28.57% at 0.1% Hct and 3.71% at 2% Hct. With sample recirculation, RBC removal efficiency remains 99.99%, and WBC purity improves to 60% at 1% Hct. These equations aid in optimizing microdevice design for point-of-care applications.

5

6

Cell counting

Experiments were repeated three times, and the cells present at the inlet and outlets were counted to test the performance of the microdevice according to parameters, as discussed in Sect. 2.4. Flow cytometry experiments used monoclonal antibodies to differentiate the granulocytes (neutrophils, eosinophils and basophils) from the total WBC population. Fluorescein Isothiocyanate (FITC) conjugated CD66b antibody (positive for granulocytes) and Allophycocyanin (APC) conjugated CD45 antibody (positive for all WBCs) were used. WBCs in inlet-outlet samples were counted based on positive anti-CD45 cells, whereas all granulocytes were counted using positive anti-CD66b and anti-CD45 cells. Upon counting, it was observed that in WO, 45.40% of total cells (RBCs and WBCs) were granulocytes and 15.10% were agranulocytes. The remaining 39.50% cells were RBCs when counted after 4th recirculation. In this study, we aim to separate WBCs from abundant RBCs; hence platelet population was not considered while calculating the WBC purity. Platelets are non-phagocytic cells [39] and smallest in diameter than RBCs and WBCs, making complete removal difficult using microfluidic devices [22, 40].

Phagocytic analysis

The phagocytosis tests were performed on separated WBCs obtained from the spiral microdevice. The quantification of phagocytosis was done using a microscope and spectrophotometer, followed by validation using flow cytometry.

Phagocytosis analysis using a microscope

A 20 µl WO sample was incubated (at 37 °C in the dark for 30 min) with an equal volume of 0.2% NBT solution, one with Phorbol 12-myristate 13-acetate (PMA) and the other without PMA, on a glass slide [21]. These slides were observed under a microscope using safranin as counter-stain. The microscopic test on WBCs is qualitative, and their results related to phagocytosis are shown in Fig. 6. Upon activation with PMA, WBCs produce superoxides that convert yellow-coloured NBT to a blue formazan. Figure 6(a) shows control, wherein no formazan crystals were observed due to the absence of PMA, indicating their inactivated state without any stress. However, conventional centrifugation stresses WBCs [8] and pre-activates them, altering bioassays. Figure 6(b) shows activated WBCs with blue formazan crystals. Upon activation, 93% of WBCs displayed phagocytosis, but no formazan formation indicating, loss of phagocytosis [41].

Fig. 6.

Determination of phagocytic ability of WBCs using a microscope (40x objective) and spectrophotometer. (a) Microscopic observation of WBCs having PMA – (control) (b) Microscopic observation of WBCs having PMA+ (c) Spectrophotometric test for assessing superoxide ions at 620 nm. The top labels on each bar represent the WBC (W) and RBC (R) population in the sample obtained using centrifugation or a spiral microdevice, whereas the middle labels on the bars mentioned for spiral microdevice outputs represents WBC purity such as 60%, 42% and 33%. R1 and R2 show a 2-fold and 3-fold increase in RBCs in the WO sample

Phagocytosis analysis using a spectrophotometer

ROS (i.e., superoxide ions) analysis is a quantitative assessment for phagocytic potential of WBCs. Here, the five different samples (each 100 µl) were taken to study phagocytosis. Two of the samples had WBCs and neutrophils that were separated by density gradient centrifugation, and the third sample was taken from the WO of our spiral microdevice after 4th Recirculation. The remaining two samples were: (1) WO sample with 2-fold increased RBCs (R1-20050), i.e., WO + R1 and (2) WO sample with 3-fold increased RBCs (R2-30500), i.e., WO + R2. This RBC manifestation is done to check the effect of RBCs on ROS production. Initially, NBT without PMA (i.e., control) and PMA was added to above mentioned samples and incubated for 30 min in the dark at 37 °C. The WBC samples treated with PMA will form formazan crystals that were dissolved by adding 120 µl potassium hydroxide and 140 µl dimethyl sulfoxide, respectively [21].

Furthermore, absorbance was taken for each sample at 620 nm to quantify formazan, indicating superoxide ions generation. The samples treated with PMA show higher production of superoxide ions (65%) than the control samples, indicating the cells have phagocytic ability, as shown in Fig. 6(c). Spiral microdevice-extracted WBC samples produce superoxide ions similarly to centrifugation. This implies that fewer RBCs in WO sample (60% WBC purity) had no significant influence on the outcome. The 2-fold increase in RBCs in the WO (42% WBC purity) results in negligible change; however, the 3-fold increase in RBCs (33% WBC purity) generates a significant increase in superoxide ion, which alters the ROS result [30]. It has also been noted that neutrophils (60–70% of total WBC count) alone have stronger phagocytic ability and are predominantly involved in phagocytosis.

Phagocytosis validation using flow cytometer

Further, the ROS generation by WBCs was validated using a flow-cytometry-based DHR (Dihydrorhodamine-123) assay. As we already know, granulocytes (neutrophils) account for most of the phagocytic activity; hence granulocytes are targeted for this assay. The FITC-conjugated CD66b was added to the WO sample treated with PMA to identify granulocytes, followed by DHR dye staining (S1) [24]. The rapid oxidation of the Dihydrorhodamine-123 results in the formation of the fluorescent compound Rhodamine-123, which is then localized in the cell cytoplasm. A flow cytometer recorded the fluorescence intensities produced by positive CD66b cells, and these intensities were used to calculate the stimulation index (SI) [24]. The SI is the ratio of the mean fluorescence intensity of the stimulated sample to the unstimulated sample. The usual stimulation index for adults is 73 to 125, and a result below 30 is abnormal [41]. Figure 7 describes the log shift seen in the histogram of stimulated WO sample over unstimulated WO sample, indicating phagocytic ability with an SI of 93.

Fig. 7.

Flow cytometry DHR assay for determination of ROS. (a) Unstimulated WO sample treated with anti-CD66b (control) (b) PMA-stimulated WO sample treated with anti-CD66b. A shift in the histogram shows that separated WBCs can generate ROS upon stimulation with an SI of 93

Conclusions

This work demonstrated a single curved spiral microfluidic device that separates WBCs. The WBC sample was obtained using an inertial technique from diluted human blood prepared from ⁓20 µl of freshly drawn blood. Experiments with fluorescent particles and numerical simulations were used to find the traces of particles within the microdevice and further validated using blood samples. The performance of the microdevice was investigated at various flow rates and hematocrits to confirm the optimal parameters, such as the flow rate of 0.4 ml/min and hematocrit of 1%. It was also observed that two bifurcations and three outlets remove almost all RBCs to get WBCs. After four recirculations, the microdevice attains a WBC purity of 60%, effectively eliminating 99.99% of RBCs. This level of purity is adequate for studying the phagocytic potential of WBCs. The efficacy of the microdevice is negatively impacted by hematocrits and flow rates above and below the optimal operating values. Phagocytic analysis of isolated WBCs revealed that they had retained their phagocytic capacity. The microdevice reported in this study separates WBCs in less than 10 min without RBC lysis, density gradient centrifugation, or costly antibodies.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

The research was designed and implemented by SM, PJ, and ST, with input from VH. SM conducted the experiments and wrote the manuscript with input from all the authors. The simulation was performed by PJ and confirmed by PD. SB coordinated funding for the research.

Funding

This work was supported by BITS BioCyTiH foundation under grant BBF/BITS (G)/FY2022-23/BCPS-115, covered by National Mission for Interdisciplinary- Cyber Physical System (NMICPS) of Department of Science and Technology, Government of India (D.O: DST/NM-ICPS/MGB/2018).

Declarations

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. All the experiments were conducted following the rules and regulations approved by the Human Ethical Committee, Bits-Pilani, K K Birla Goa campus (EC Registration No.: EC/NEW/INST/2023/3297).

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Consent to publish

The authors affirm that human research participants provided informed consent for publication.

Competing interests

The authors declare that they have no known competing financial interests.