Smart interface materials integrated with microfluidics for on-demand local capture and release of cells

Abstract

Stimuli responsive, smart interface materials are integrated with microfluidic technologies creating new functions for a broad range of biological and clinical applications by controlling the material and cell interactions. Local capture and on-demand local release of cells are demonstrated with spatial and temporal control in a microfluidic system.

Copyright © 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Figures

Figure 1

Local capture and release of cells in microchannels integrated with smart interface materials. Schematic description of: A) local cooling of channels for local cell release, and B) local warming of channels for local cell capture with thermoelectric module. C) The schematic representation of the cross-section of channels in the absence of heating elements. D) When the whole channel surface was warmed to 37 °C, target cells were captured everywhere on the channel surface. E) Channels can be locally warmed to 37 °C, which enables local capture of cells in a selected region. F) Thermoresponsive channels can be locally cooled using thermoelectric modules to facilitate on-demand local release of a selected set of captured cells in microchannels. G) Local temperature control in thermoresponsive channels. The middle channel on the microchips was stained with a temperature responsive dye to monitor temperature change of the device. A typical image of a chip with a local temperature change in one channel shows the full spectrum of colors. Black area indicates the channel on the microchip that was stained with thermoresponsive dye. The white dashed rectangle shows the area of interest used in image processing. The thermoelectric unit was located below the channel, which resulted in a local temperature control and color change in the channel. H) Temperature responsive dye was responsive in the range of 32 °C to 41 °C, and displayed green color at 37 °C, i.e., the temperature at which cells were captured. The dye displayed black color at temperatures below 32 °C, at which release of cells in the channels was achieved. I) Baseline RGB values represent the colors displayed by the thermosensitive dye, which were used in image processing to quantify temperature distribution in channels.

Figure 2

Local cell capture in microchannels. A) Temperature distribution in control and local capture channels. In local capture channels, temperature was controlled locally, enabling local capture of cells in zone 1 only. B) Typical images of cells in channels in zones 1–4 before and after capture of cells (cells were marked with circles). Images indicated a significant difference in number of cells captured in zone 1 compared to zones 2–4. Cells captured in zones 1–4 in control channels displayed a typical distribution pattern for cell capture in microchannels. C) Quantitative analysis of cell capture in zones 1–4, in control and local capture channels. A statistically significant, greater number of cells were captured in zone 1 (79% ±4%, n = 4 channels) compared to zone 1 of control channel and zones 2–4 of local capture channel. Each zone corresponded to 4.3 mm of the 25 mm full channel length. D) Bright field and CD4 immunofluorescent stained locally captured cells indicating the capture specificity of the channels. Capture specificity of the channels was quantified to be 92% (±2%, n = 4 channels). (Brackets connecting individual groups indicate statistically significant difference. Non-parametric Mann-Whitney U test, p

Figure 3

Local release of captured cells in microchannels. A) Temperature distribution in channels during capture and for subsequent local release. Temperature was controlled locally, which resulted in release of captured cells in zone 1 only. B) Typical images of cells within channels in zones 1–4, before and after release of captured cells (cells were marked with circles). The cells were released from zone 1. Images indicated a significant difference in number of cells remaining in zone 1 after release. On the other hand, the number of cells remained similar before and after release in zones 2–4. C) Quantitative analysis of cell numbers in channels in zones 1–4 before and after release. A statistically significant number (85% ± 4%, n = 4 channels) of captured cells was released locally in zone 1. Each zone corresponded to 4.3 mm of the 25 mm full channel length. (Brackets connecting individual groups indicate statistically significant difference. Non-parametric Mann-Whitney U test, p

Figure 4

Release of locally captured cells in microchannels. A) Temperature change in channels before and after local capture/release. Temperature was controlled locally, which resulted in local capture and release of cells in zone 1 only. B) Typical images of cells in channels in zones 1–4 before and after local release of locally captured cells (cells were marked with circles). The cells were captured specifically in zone 1, followed by on-demand release from the same area. Images indicated a significant difference in number of cells in zone 1 after capture and release. As designed, the number of cells remained similar before and after capture/release steps in zones 2–4. C) Quantitative analysis of cell numbers in channels in zones 1–4 before and after release. A statistically significant number (93% ± 2%, n = 4 channels) of captured cells was released locally in zone 1, at which local temperature control was performed. Each zone corresponded to 4.3 mm of the 25 mm full channel length. (Brackets connecting individual groups indicate statistically significant difference. Non-parametric Mann-Whitney U test, p

Figure 5

Local capture and release of cells in zones close to the middle of channels. Temperature was controlled locally, which resulted in local capture and release of cells in zone 2 only. A) A statistically significant number (65% ± 8%, n = 4 channels) of captured cells was released locally in zone 2, at which temperature control was performed. B) Release of locally captured cells in zone 2. A statistically significant number (86% ± 7%, n = 4 channels) of captured cells was released locally in zone 2, at which local capture and release was performed. These results indicated the capability of controlling local capture and release of cells towards the middle of the channels. Each zone corresponded to 4.3 mm of the 25 mm full channel length. (Brackets connecting individual groups indicate statistically significant difference. Non-parametric Mann-Whitney U test, p

Figure 6

Theoretical understanding of local cell release in microchannels and comparison to experimental results. A) Schematic for 2D computational modeling of fluid flow and heat transfer inside a microchannel. Thermoelectric module was placed onto locations: x1 = 1 mm for zone 1, and x1 = 6 mm for zone 2. Top surface was 3.5 mm thick polymethyl-methacrylate (PMMA) and bottom surface was 0.65 mm thick polystyrene. To model heat loss from top and bottom surfaces, thermal conductivity of PMMA was set as, k = 0.19 W/m.K, and thermal conductivity of polystyrene was set as, k = 0.08 W/m.K. Channel height is H = 80 μm, and length is L = 25 mm. BC stands for boundary condition. Width of inlet flow was w2 = 250 μm, and width of thermoelectric module was w1 = 4.3 mm. Steady-state surface temperature (°C) distribution as a function of channel location (mm) were plotted in the presence of thermoelectric module at: B) x1 = 1 mm (zone 1), and C) x1 = 6 mm (zone 2). Results from local release experiments were compared to the results from computational model for (D) zone 1, and E) zone 2.