Droplet-based microfluidic assay development for cytotoxicity evaluation of Natural Killer cells 

This short review article is originally based on the research paper entitled “Droplet-Based Cytotoxicity Assay: Implementation of Time-Efficient Screening of Antitumor Activity of Natural Killer Cells” authored by Silvia Antona, Ilia Platzman, and Joachim P. Spatz. The research paper was originally published in the journal – ACS Omega.

It explores a droplet-based microfluidic approach to create an efficient system of measuring and evaluating cytotoxic ability of NK-92 cells towards target cells. The precise control of droplet size and pulseless flow control was achieved with the help of pressure-driven flow controlled microfluidics.

The researchers were able to explore other efficiency parameters of NK-92 cells like killing and serial killing ability towards 3 target cell types.

Abstract

Natural killer cells play a key role for the immune system. Mostly owing to their rapid cytotoxicity against infectious pathogens, hematologic malignancies, and solid tumors, NK cells are considered to be the ideal contenders for cell-based immunotherapy.

The heterogeneity in their cytotoxic behavior presents a disadvantage, even after years of research and progress. The objective to screen the intrinsic diversity of NK cells is performed with the help of droplet-based microfluidic technology.

Pressure-driven flow controlled microfluidics is used to fabricate a time-efficient single-cell cytotoxic assay. In this study, NK-92 cells are encapsulated with hematological tumor cell lines in water-in-oil droplets of different sizes and eventually their cytotoxic activity is measured.

Afterwards, the impact of droplet-based confinement on NK cytotoxicity is explored with the help of precise control of the droplet volume. This strategic refinement of droplet size enables the researchers to make this process time-efficient when compared to cytotoxicity assays based on flow cytometry.

Furthermore, the potential of NK-92 cells to destroy numerous target cells, in series is studied. This section of the study helps build knowledge around the serial killing process dynamics of NK-92 cells.

Introduction

Natural killer (NK) cells are crucial members of the human immune system. They are highly cytotoxic immune effectors that are responsible for quick cytotoxicity against several infectious pathogens, hematologic malignancies, and a number of solid tumors without prior target recognition [1]. Cytotoxicity in general, is defined as the quality that results in being destructive for cells. It is used to describe the processes of immune cells, that is Natural Killer Cells. Due to these properties of NK cells, they have become of valuable interest recently as possible substitutes for autologous T lymphocytes in cancer immunotherapy [2]. They present the potential to overcome some of the drawbacks related to T-cell based immunotherapy.

Even though NK cells have great potential, NK-cell based immunotherapy still poses some key challenges. One of these drawbacks is the absence of screening and sorting mechanisms for rapid and accurate ex vivo preselection of highly cytotoxic immune cells within a heterogeneous population. For Natural Killer cells, 50% of them do not kill any target cells, and moreover only 7-10% of them show the potential of killing multiple target cells at a time. Current cytotoxic studies are conducted in bulk and provide average results for an entire population of cells. Time lapse analysis of individual NK cells engaging with their respective target cells is difficult to accomplish in bulk analysis due to continuous movement of immune cells. Fluorescence-activated cell sorting (FACS) is generally used as the ideal procedure but necessary labelling procedures make it complicated and affecting possible in vivo applications.

With the objective to devise a more efficient method, droplet-based microfluidics and pressure driven flow control has emerged as a powerful tool to evaluate single-cell genetics and behaviour [3].

To discover more tips and tricks about droplet-based microfluidics, please check our new droplet userguide!

Aim & objectives

In this study, a droplet-based microfluidic approach is employed to develop a systematic cytotoxicity assay for Natural Killer (NK) cells. The following objectives and concepts are also explored:

• To utilize droplet-based microfluidics for fabrication of a time efficient single and serial NK cell cytotoxicity assay.
• Droplet size regulation to identify the effect of a compartment size on the cytolytic activity of NK cells
• To identify highly cytotoxic Natural Killer cells among a heterogeneous population.
• To record and dynamically analyze single and serial cytotoxic events caused by NK-92 cells over a 12 hour period.
• To demonstrate how droplet encapsulation presents a boost in NK-92 cytotoxicity against several target cells.

Images of the pressure-driven flow controller along with the experimental setup for this study are displayed below:

Materials & methods

The chip used for this study was a droplet-based microfluidic device poly(dimethylsiloxane) (PDMS) were prepared by photolithography and soft lithography methods as described previously.

For a detailed understanding of the microfluidic device design and fabrication process used for this study, here is a link to the original research paper.

Key findings

NK-92 cytotoxicity assay inside encapsulation

A single-cell cytotoxicity assay is developed inside droplet-based compartments. This is performed with the aim of overcoming the spatial-related issues for standard bulk live imaging, and temporal-based issues for flow cytometry assays.

Different confinement sizes were used to assess whether encapsulation size has an effect on dynamics of cytolytic activity of NK-92 single and serial cytotoxicity, against 3 hematological malignant cell lines (K562, Jurkat, and KG1a). Figure 2 shows a theoretical representation of co encapsulation within droplets of different diameters.

Altogether, 3 confinement sizes were selected (∼67, 85, and 105 μm) with factors such as cell viability, effector-target proximity and immune cell motility taken into consideration for the selection. The droplet size was controlled with the help of a pressure driven flow controller that can tune the aqueous and oil phase inlets of the microfluidic device(Table 1 & Figure 3A,C).

Next, to explore the dynamics of NK-92 effector cells, the droplets containing NK-92 effector cells were examined. (Figure 3D). Following droplet generation and co encapsulation of effector and target cells, droplets were enclosed in a microfluidic observation chamber (Figure 3B) for cytotoxic assessment.

Upon sedimentation, NK-92 cells showed compartemnt size-dependent motility behaviour around target cells. This behaviour can be characterized as a movement towards target cells, and alternation of high, low or no motility while changing their morphology and transiently scanning the target cells.

Flow cytometry NK-92 cytotoxicity assay

In the second stage of this study, with the objective to evaluate the strength and sensitivity of the droplet-based cytotoxicity assay, the assays performed in ∼67 μm droplets are compared to the flow cytometry measurements between NK-92 and target cells. It was found that, for all three target cell lines, the percentage of killing events after a 3 hour duration were higher than in bulk flow cytometry measurements.

For the K562, and Jurkat target cells, the values are observed to taper and converge over time. Only in the case of KG1a target cells, after 12 hours, the number of killing events are observed to be higher in bulk than in droplets. These findings clearly show faster cytotoxic dynamics inside droplets in the early stages of the experiment.

Assessment of early-stage serial killing potential of NK-92 cells inside encapsulated droplets

In the final stage of this experiment, the effect of confinement dynamics is assessed for NK-92 cells towards killing multiple target cells. Droplets with an approximate size of 85 and 105 μm consisting of a single effector cell with 2 or 3 target cells are fabricated. The 85 μm droplet size is expected to create a better environment for interaction with at least 2 to 3 target cells.

The heatmaps that are displayed in Figure 4A-H show data of serial killing events inside individual droplets of 2 different sizes. For each droplet size, a minimum of 75 and 21 droplets were studied for 1:2 and 1:3 effector-to-target cell ratios respectively.

It was found that a reduction in droplet size from 105 to 85 μm for droplets containing a 1:2 effector/ target cell ratio resulted in an increase of two consecutive killing events. The increase is quantified to be from 37% to 55% (Figure 4A, B) and 5% to 37% (Figure 4C,D) for K562 and Jurkat cells respectively. For droplets with 3 target cells, the encapsulation size reduction resulted in an increase of serial killing events from 28% to 37% for K562 (Figure 4E,F); and from 3% to 14% in the case of Jurkat cells (Figure 4G,H). The data is summarized in Figure 4I and 4J.

Conclusion

To summarize this research paper, a droplet-based microfluidic assay is designed and developed for evaluation of NK cell cytotoxicity heterogeneity against hematological cancer cells. In fact, the direct relationship between confinement size and cytotoxic behaviour of individual NK-92 cells against K562 and Jurkat target cells. This microfluidic cytotoxic assay also helped identify individual cells with the ability to kill multiple targets. The confinement size is also found to have a huge effect on these factors. The pressure-driven flow controller, Elveflow OB1 MK3+ makes regulating the droplet size a seamless process, and thus the experiment can be easily replicated. The real-world applications for this system are found to be cell-based immunotherapy, and other biological and medical research settings.

These exciting results could be achieved by the researchers by utilizing pressure-driven flow controlled microfluidics as the droplet generation technique. For an in-depth analysis of how droplet-based microfluidics was employed to perform this promising study, the complete paper is available at Ilia Platzman et al.

References