Optimization of Mechanical Tissue Dissociation Using an Integrated Microfluidic Device for Improved Generation of Single Cells Following Digestion

Abstract

The dissociation of tissue and cell aggregates into single cells is of high interest for single cell analysis studies, primary cultures, tissue engineering, and regenerative medicine. However, current methods are slow, poorly controlled, variable, and can introduce artifacts. We previously developed a microfluidic device that contains two separate dissociation modules, a branching channel array and nylon mesh filters, which was used as a polishing step after tissue processing with a microfluidic digestion device. Here, we employed the integrated disaggregation and filtration (IDF) device as a standalone method with both cell aggregates and traditionally digested tissue to perform a well-controlled and detailed study into the effect of mechanical forces on dissociation, including modulation of flow rate, device pass number, and even the mechanism. Using a strongly cohesive cell aggregate model, we found that single cell recovery was highest using flow rates exceeding 40 ml/min and multiple passes through the filter module, either with or without the channel module. For minced and digested kidney tissue, recovery of diverse cell types was maximal using multiple passes through the channel module and only a single pass through the filter module. Notably, we found that epithelial cell recovery from the optimized IDF device alone exceeded our previous efforts, and this result was maintained after reducing digestion time to 20 min. However, endothelial cells and leukocytes still required extended digestion time for maximal recover. These findings highlight the significance of parameter optimization to achieve the highest cell yield and viability based on tissue sample size, extracellular matrix content, and strength of cell-cell interactions.

Keywords: dissociation; microfluidics; single cell analysis; tissue engineering; tissue processing.

FIGURE 1

Schematic representation of the Integrated Disaggregation and Filtration (IDF) device. Large aggregates containing high extracellular matrix (ECM) content, such as digested tissue, are exposed to stepwise increases in shear stress throughout the branching channel array as the width narrows from 1 mm to 125 µm. Cell aggregates are held together via through cell-cell (dark orange perimeter) and cell-ECM (green fibers) interactions. As ECM is digested by collagenase, the channel array gradually reduces aggregate size via hydrodynamic shear forces. The smallest channels and nylon mesh membranes then break down the cell-cell interactions that hold together the smallest aggregates and clusters. Channels, cell aggregates, and membrane pore sizes are not shown to scale.

FIGURE 2

Optimization of the branching channel array using cell aggregates. MCF-7 cells were passed through the channels at different flow rates for either 10 or 20 passes. Results are shown for (A) total single cell, (B) aggregate, and (C) live single cell yields, which are all normalized to the control that did not pass through the device. Also presented is (D) single cell viability. Strong effects were observed for each metric above 40 ml/min, but higher pass number did not substantially influence results. Data are presented as mean values ± SEM from at least three independent experiments. Two-sided t test was used for statistical testing. Stars indicate p < .05 and double stars indicate p < .01 relative to the unprocessed control.

FIGURE 3

Evaluation of different dissociation formats using cell aggregates. MCF-7 cells were passed through the channel module 10 times and filter module once (10C, 1F), both channel and filter modules 10 times (10C + F), or the filter module 10 times (10F) at different flow rates. Results are shown for (A) total single cell, (B) aggregate, and (C) live single cell yeilds, which are all normalized to the unprocessed control. Also presented is (D) single cell viability. Optimal results were observed at 40 ml/min using multiple filter module passes. Data are presented as mean values ± SEM from at least three independent experiments. Two-sided t test was used for statistical testing. Stars indicate p < .05 and double stars indicate p < .01 relative to the unprocessed control.

FIGURE 4

Evaluation of dissociation formats using murine kidney. Kidneys were harvested, minced, and digested for the indicated time periods. Samples were then passed through the channel module 10 times and filter module once (10C, 1F), simultaneously through both modules 10 times (10C + F), or sequentially through both modules 10 times (10C + 10F) at 40 ml/min and the resulting cell suspensions were analyzed using flow cytometry. Controls were pipetted/vortexed and passed through a cell strainer. Results are shown for viable and single (A) total cells, (B) EpCAM + epithelial cells, (C) endothelial cells, and (D) leukocytes. Optimal results were attained using the single filter pass at each digestion time. Data are presented as mean values ± SEM from at least three independent experiments. Two-sided t test was used for statistical testing. Stars indicate p < .05 and double stars indicate p < .01 relative to the control at the same digestion time.

FIGURE 5

Final optimization of murine kidney. Kidneys were harvested, minced, and digested for the indicated time periods. Samples were then passed through the channel module for the indicated number of times and filter module once (10C, 1F) and resulting cell suspensions were analyzed using flow cytometry. Controls were pipetted/vortexed and passed through a cell strainer. Results are shown for viable and single (A) total cells, (B) EpCAM + epithelial cells, (C) endothelial cells, and (D) leukocytes. Similar epithelial yields were obtained after 20 min digestion time using 20 passes and 60 min digestion time using 10 passes. However, maximal endothelial and leukocyte yields required 60 min digestion time and 10 passes. Data are presented as mean values ± SEM from at least three independent experiments. Two-sided t test was used for statistical testing. Stars indicate p < .05 and double stars indicate p < .01 relative to the control at the same digestion time.