Affinity-Bead-Mediated Enrichment of CD8+ Lymphocytes from Peripheral Blood Progenitor Cell Products Using Acoustophoresis

Abstract

Acoustophoresis is a technique that applies ultrasonic standing wave forces in a microchannel to sort cells depending on their physical properties in relation to the surrounding media. Cell handling and separation for research and clinical applications aims to efficiently separate specific cell populations. Here, we investigated the sorting of CD8 lymphocytes from peripheral blood progenitor cell (PBPC) products by affinity-bead-mediated acoustophoresis. PBPC samples were obtained from healthy donors (n = 4) and patients (n = 18). Mononuclear cells were labeled with anti-CD8-coated magnetic beads and sorted on an acoustophoretic microfluidic device and by standard magnetic cell sorting as a reference method. CD8 lymphocytes were acoustically sorted with a mean purity of 91% ± 8% and a median separation efficiency of 63% (range 15.1%⁻90.5%) as compared to magnetic sorting (purity 91% ± 14%, recovery 29% (range 5.1%⁻47.3%)). The viability as well as the proliferation capacity of sorted lymphocytes in the target fraction were unimpaired and, furthermore, hematopoietic progenitor cell assay revealed a preserved clonogenic capacity post-sorting. Bead-mediated acoustophoresis can, therefore, be utilized to efficiently sort less frequent CD8+ lymphocytes from PBPC products in a continuous flow mode while maintaining cell viability and functional capacity of both target and non-target fractions.

Keywords: CD8 lymphocytes; PBPC; acoustophoresis; cell sorting; magnetic-beads; peripheral blood progenitor cells; ultrasound.

Figure 1

Schematic picture of the acoustophoretic chip design. The sample consisting of a mixture of bead-bound cells (red) and unbound cells (white) is injected into the pre-focusing channel through the sample inlet. In a first step particles are pre-focused into two parallel bands (A) using a 5 MHz piezoceramic transducer that drives a full wavelength resonance across the channel width, with two pressure nodes that superimpose with a half-wavelength resonance in the vertical direction. Following the flow direction, the particles are then bifurcated to each side of the wash buffer inlet where Ficoll is infused. Due to the pre-focusing step, cells enter the separation channel close to the channel walls and are prealigned in the width and height dimension of the channel, ensuring an identical starting position for the separation procedure and thereby maximizing the resolution of the separation. In the main separation channel, acoustic forces in an ultrasonic standing wave field with a pressure node in the center of the channel induce movement of cells and particles depending on their acoustophysical properties. Bead-bound cells are forced into the Ficoll buffer (B) and can be collected through the target outlet while unbound cells stay close to the channel wall and can be collected through the waste outlet.

Figure 2

Frequency of CD8+ cytotoxic T cells in pre-sorted peripheral blood progenitor cell (PBPC) products and CD8+ purities following acoustic and magnetic separation post-sorted samples are shown. Both, acoustic and magnetic separation allowed effective enrichment of CD8+ cells. Data are presented as individual data points (triangles, circles, and quadrants) and corresponding means ± SD, n = 22.

Figure 3

Flow cytometry analysis of the distribution of leukocyte subpopulations in the non-target fraction (side outlet) of acoustically sorted samples. Comparison of the mean (±SD) percentages of CD3+/CD4+, CD3+/CD8+, CD19+/CD3−, CD34+ and CD56+/CD3− cells in pre-sorted PBPC and post-sorted non-target samples (n = 3). Due to the removal of CD8+ cells from the sample and collection in the target fraction, a relative decrease of CD3+/CD8+ cells is observed in the non-target fraction compared to the input PBPC sample.

Figure 4

CD3/CD28-mediated T cell proliferation of acoustically and magnetically sorted CD8+ cytotoxic T cells. Cells were stimulated in the presence of anti-CD3/CD28 and proliferation was measured on days 2, 3 and 4 of culture using CFSE staining (n = 3). For each day the relative number of proliferating cells (a) as well as the proliferation index, i.e., the average number of cell divisions all responding cells have undergone, are presented (b).

Figure 5

Hematopoietic progenitor cell colony-forming ability. The mean (±SD) number of granulocyte macrophage colony-forming units (CFU-GM) as well as erythroid burst-forming units (BFU-E) is shown for cells from pre-sorted (PBPC) and acoustically post-sorted (non-target) fractions (n = 4).

Figure 6

Purity and separation efficiency of acoustically sorted CD8+ cells in relation to the central buffer density. (A) Schematic drawing of the microfluidic device showing the effect of the high density central buffer. Bead-labeled cells (orange) and unlabeled cells (white) enter into the main separation channel. High density buffer infused in the central inlet (yellow) creates a barrier across which bead-labeled cells can be acoustically moved into the center stream and collected in the central outlet while unlabeled cells are not able to move into the high density buffer and thus exit the separation channel through the side outlet. Arrows indicate the flow direction; (B) Ficoll was diluted in PBS in different concentrations and used as central buffer. The purity of the acoustic separation as well as the separation efficiency was determined (n = 3); (C) The image sequence of the central outlet illustrates the effect on the acoustic separation at increasing levels of Ficoll in the central buffer. The increase of the central buffer density creates a barrier through which bead-labeled cells (orange) are able to move whereas unlabeled cells (white) are not and remain at the sidewall to be collected through the side outlet.