Schematic drawings of the perfusion cell culture system (a), the top view (b) and cross-sectional view in the flow direction (c) of the MPA. Plate separation h = 280 μm and well radius is 100 μm. The syringe pump and a syringe containing fresh media were placed inside a fridge (compartment A) at 4 °C. The PDMS MPA was placed inside a gastight environmental chamber (compartment C). Inlet and outlet ports of the flow-cell were connected to a syringe and a waste container, respectively. Air with 5% CO2 was recirculated through a humidifier with a sealed sparging system. The gastight environmental chamber, microscope, and humidifier were surrounded by a purpose-built incubator (compartment B).
KG1a growth curves for (a) cells cultured in IMDM with sodium bicarbonate buffer and 5% CO2 in air or IMDM with HEPES in air, and (b) cells cultured on PDMS or polystyrene substrates.
Media degradation studies. (a) Factorial design experiment to assess the effect of serum, light, and temperature on media degradation (Table I). (b) Normalised plot showing main effects FBS (A), 23 °C versus 37 °C (B), and light (C) and factor interactions (ABC and AC were significant). (c) Degradation of media by tubing. Error bars show standard deviations.
Microwell perfusion culture: (a) Fold expansion versus culture time; Comparison among growth rates in tissue culture dish (solid line), microwell perfusion culture (0.1 μl/min—dashed line and 0.3 μl/min—dotted line) and microwell static culture (dash-dotted line). Error bars show standard error of the mean. (b) Effect of perfusion rates and innoculum on the fold expansion. Cell growth was inhibited at high flow rate. Note, “0.1 μl/min perfusion” indicates the intermittent perfusion for 1 h at 0.3 μl/min followed by 2 h without flow, while “0.3 μl/min perfusion” represents the continuous perfusion at 0.3 μl/min.
(a) Still image from a video showing cell division and movement over 6 days in a microwell. The image stack was thinned and time interval between 2 frames was 6 min. (b) Cell lineages of four individual clones in four wells created by manual processing of image stacks for 100-h cell culture (8000 images). Dots represent cells. (c) Empirical survivor function for cell generation time. LCB and UCB are the 95% lower and upper confidence interval for the Kaplan–Meier estimate. Dash-dotted line is a linear regression model. (a), (b), and (c) are generated from cells cultured with the average perfusion flow rate of 0.1 μl/min. (enhanced online). [URL: http://dx.doi.org/10.1063/1.3669371.1]10.1063/1.3669371.1
Semi-automated cell counting. KG1a cells were randomly distributed between microwells using a flow protocol which deposited 3 cells per well on average. The images were taken on consecutive days from the same position on the microwell array. Green and red circles show automatically segmented wells and cells, respectively. This automated segmentation results were then inspected and corrected manually using an interactive software platform.
Heat map showing the cell number in each well. (a) 0.1 μl/min and (b) 0.3 μl/min.
Media treatment method in the full factorial design.
Tubing specification and media flow rates for biocompatibility studies. The residence time of media inside tubing was 6.6 h.
Calculation of critical perfusion rate for KG1a cells.
Article metrics loading...
Full text loading...