Poster Z1 - Angled ridge microfluidic filter for continuous buffer exchange of suspended cells

Friday, October 25, 2024

3:30 PM - 4:30 PM EST

Location: Exhibit Hall E, F & G

Presenting Author(s)

Steven Swingle (he/him/his)

Graduate Research Assistant
Georgia Institute of Technology, United States

Co-Author(s)

Hoseyn A. Amiri, MSc (he/him/his)

Graduate Research Assistant
Georgia Institute of Technology
Atlanta, Georgia, United States

Primary Investigator(s)

TS

Todd Sulchek

Professor
Georgia Institute of Technology, United States

Introduction: Exchanging the buffer of a cell suspension is often a necessary step in cell culture, research, and manufacturing. Typically, this unit operation is performed by pelleting cells in a centrifuge, removing the supernatant, and resuspending the pellet in a new buffer. While simple, this method is prone to cells loss, decreased cell viability, and altered cell behavior from mechanical stress. This method is especially disadvantageous to cell manufacturing workflows as it is discontinuous, difficult to automate, and requires substantial workspace.

Microfluidic buffer exchange (µBE) has many potential benefits over traditional methods as it small, continuous, and relatively easy to automate, and several µBE platforms have been demonstrated. Here, we introduce a novel µBE mechanism (Fig. 1) developed by altering a stiffness-based sorting device from our group. In the new device, a cell suspension is flown parallel to the new buffer, where laminar conditions prevent mixing. A skew constriction smaller than the cell diameter guides cells from the old buffer into the new buffer. We subsequently demonstrate the effectiveness of this method which, once optimized, may have certain advantages over other µBE devices.

Materials and

Methods: Devices were made from PDMS (Sylgard 184) molded onto a silicon wafer coated with SU-8 photoresist patterned using standard photolithographic protocols. ARPE-19 cells were suspended in a PBS and blue food dye (McCormick) solution at a 5000:1 volume ratio, with the dye serving as our model contaminant. Using a syringe pump, (Harvard Apparatus PHD-2000) the cell suspension and fresh PBS were flown into separate device inlets at the same flowrate. Total flowrate through the devices with a ridge angle of 5-, 2-, and 1- degrees were 20, 20, and 40 µL/min, respectively. Cell concentrations of each of the five outlets were measured with either a hemocytometer or a NucleoCounter NC-200. Dye concentration was calculated via the Beer-Lambert Law from spectrophotometer (Cary-5000). Forces experienced by cells at various positions along a device ridge were modeled in COMSOL using a static spherical particle adjacent to a device ridge with varying angles and constriction sizes.

Results, Conclusions, and Discussions: We first needed to adjust device parameters to prevent cells from traveling underneath the angled ridge. To achieve this, we sought to minimize the force ratio (Fr) defined by Fr = Fy/Fx, where Fy and Fx are forces on the cell perpendicular and parallel to the ridge, respectively. Simulations showed that lowering the ridge angle had the greatest impact on minimizing Fr while also minimizing the dependence of Fr on the cell position along the ridge (Fig. 2).

With the insights provided by the simulations, buffer exchanges were performed on devices with 5-,2-, and 1-degree ridge angles. From 0 - 18 minutes, the 5-degree device had 79.5% of cells travel into the target outlet (outlet 5); however, a substantial clog was present beginning at around minute 25, and from 127 – 134 minutes the percentage of cells in the target inlet decreased to 33.3% (Fig. 3A-C). Despite the clogging that occurs in 5-degree ridge devices, the target outlet had a dye concentration 2.55% of that of the inlet cell suspension (Fig. 3D). Clogging was not observed in the devices with a 2- and 1-degree ridge angle, likely due to the lower Fr. For the 1-degree device, the percentage of viable cells traveling to the target outlet was 94.3%, even with a flowrate twice as high as the 5-degree device (Fig. 4).

The presented results demonstrate a proof-of-concept for this novel µBE mechanism. Lowering the ridge angle was a critical device adjustment that eliminated clogging and maximized the number of cells traveled into the target outlet which we believe is due to the resulting small Fr. While cell recovery with this device remains low, we believe this can be improved with better sample collection technique. Additional experiments will be performed to ensure the replicability of our results and to statistically confirm trends suggested by our initial findings. We will also seek to maximize device throughput and minimize operation time by increasing the device flowrates and inlet cell concentration.

Acknowledgements (Optional): This project is supported by National Science Foundation under Grant No. 2134701 and the Georgia Research Alliance based in Atlanta, Georgia.