Biotechnology Journal
Biotechnol. J. 2015, 10, 1546–1554
DOI 10.1002/biot.201500035
www.biotechnology-journal.com
Research Article
Clonal analysis of individual human embryonic stem cell differentiation patterns in microfluidic cultures Darek J. Sikorski1,2,3, Nicolas J. Caron3, Michael VanInsberghe1, Hans Zahn1, Connie J. Eaves4,5, James M. Piret2,3 and Carl L. Hansen1,6 1 Centre
for High-Throughput Biology, University of British Columbia, Vancouver, BC, Canada of Chemical and Biological Engineering, University of British Columbia, Vancouver, BC, Canada 3 Michael Smith Laboratories, University of British Columbia, Vancouver, BC, Canada 4 Terry Fox Laboratory, British Columbia Cancer Agency, Vancouver, BC, Canada 5 Department of Medical Genetics, University of British Columbia, Vancouver, BC, Canada 6 Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada 2 Department
Heterogeneity in the clonal outputs of individual human embryonic stem cells (hESCs) confounds analysis of their properties in studies of bulk populations and how to manipulate them for clinical applications. To circumvent this problem we developed a microfluidic device that supports the robust generation of colonies derived from single ESCs. This microfluidic system contains 160 individually addressable chambers equipped for perfusion culture of individual hESCs that could be shown to match the growth rates, marker expression and colony morphologies obtained in conventional cultures. Use of this microfluidic device to analyze the clonal growth kinetics of multiple individual hESCs induced to differentiation revealed variable shifts in the growth rate, area per cell and expression of OCT4 in the progeny of individual hESCs. Interestingly, low OCT4 expression, a slower growth rate and low nuclear to cytoplasmic ratios were found to be correlated responses. This study demonstrates how microfluidic systems can be used to enable large scale live-cell imaging of isolated hESCs exposed to changing culture conditions, to examine how different aspects of their variable responses are correlated.
Received Revised Accepted Accepted article online
21 JAN 2015 04 APR 2015 05 JUN 2015 17 JUN 2015
Supporting information available online
Keywords: Clonal analysis · Heterogeneity · Human embryonic stem cell (hESC) · Microfluidic cell culture See accompanying commentary by Fernandes DOI 10.1002/biot.201500307
1 Introduction Human embryonic stem cells (hESCs) are defined by their ability to generate precursors of all three germ layers as well as undifferentiated progeny with the same potentialiCorrespondence: Dr. Carl L. Hansen, Centre for High-Throughput Biology, University of British Columbia, 2125 East Mall, Vancouver, BC, Canada V6T 1Z4. E-mail: [email protected] Abbreviations: BSA, bovine serum albumin; CT, cycle threshold; DIC, differential interference contrast; DMEM/F12, Dulbecco’s modified eagle medium: nutrient mixture F-12; ESC, embryonic stem cell; FBS, fetal bovine serum; hESC, human embryonic stem cell; NTC, no template control; PBS, phosphate buffered saline; PDMS, polydimethylsiloxane; PI, propidium iodide; ROCK inhibitor, Y-27632 Rho-associated kinase inhibitor
1546
ties [1]. The self-renewal property of hESCs allows them to serve as an essentially unlimited source of cells for research and therapeutic purposes. hESCs and their derivatives also display heterogeneity in their outputs as they expand in microenvironments or specialized niches that induce their differentiation [2, 3, 4], and this heterogeneity is also seen when hESCs are induced to differentiate in vitro. Heterogeneity within hESC cultures also affects the fate of their progeny [5, 6, 7], and this can result in the manifestation of diverse self-renewal and differentiation capacities at a clonal level [8, 9, 10]. For clinical applications, such heterogeneity is a significant issue, as incompletely differentiated cells carry a risk of tumorigenesis [11]. Thus there is a real need for an improved understanding of the determinants of hESC heterogeneity, and analytical methods to analyze their extent and control [12, 13, 14, 15].
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Biotechnology Journal
Biotechnol. J. 2015, 10, 1546–1554 www.biotecvisions.com
www.biotechnology-journal.com
The very low survival and plating efficiency of isolated hESCs (1 mm2 in area) [43, 44, 45, 46, 47, 48, 49, 50, 51], with limited reports of single hESC colony cultures in 99% positive) and Tra-1-60 (>99%) (Supporting information, Fig. S4). We next used RT-qPCR to assess OCT4 transcript levels in extracts of individual colonies generated on the microfluidics chip. All chambers on the device were lysed simultaneously followed by the separate elution of each chamber into individual micro-centrifuge tube reactions. A total of 96 chambers were eluted (across two experiments) consisting of 18 NTCs (including empty, dead and re-eluted chambers), 41 chambers perfused exclusively with mTESR1, as well as 24 and 13 chambers exposed to 10% FBS for 110 and 134 h, respectively.
Figure 3. Heterogeneity in the growth rates of individually assessed colonies of CA1S cells. (A) Colony stained with Hoechst 33342 after 190 h of growth in mTeSR1. (B) Colony stained with Hoechst 33342 after 190 h of incubation when exposed to 10% FBS for the last 110 h. (C) Colony stained with Hoechst 33342 after 165 h of incubation when exposed to 10% FBS for the last 134 h. Scale bar for (A)–(C): 200 μm. (D) Distribution of growth rates for all colonies generated in mTESR1. Each point shows the data for a single colony and the solid and dashed green lines show the mean ± the 95% confidence intervals. Histograms showing the distribution of growth rates for each condition projected onto the axis. Growth rates for each clone were calculated by enumerating the cells at the start (by DIC) and the nuclei (stained by Hoechst 33342) at the end of the experiment.
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1551
Biotechnology Journal
Biotechnol. J. 2015, 10, 1546–1554 www.biotecvisions.com
www.biotechnology-journal.com
Figure 4. Heterogeneity in OCT4 transcript levels and relation to different culture conditions. (A) Distribution of OCT4 transcript levels (data pooled from two experiments). Each point shows the result for a single colony and the green lines show the mean ± the 95% confidence intervals. (B) Normalized OCT4 transcript level and growth rates for each colony. (C) Normalized OCT4 transcript levels and area per cell for each colony. (D) Growth rate and area per cell for each colony.
OCT4 and GAPDH transcript levels in the cells recovered from each chamber of each device were then measured by RT-qPCR, with GAPDH levels and Hoechst 33342 cell counts used for normalization. The magnitude of the decrease in mean OCT4 expression (1.00 ± 0.38 for mTESR1 compared to 0.73 ± 0.47 and 0.34 ± 0.56-fold expression for cells in FBS for 110 and 134 h, respectively) was similar to previous reports [31, 58], as with the distribution of growth rates. However, our analysis revealed that individual clones exhibited considerable heterogeneity in OCT4 expression (Fig. 4A), and the emergence of OCT4 negative clones as the cells differentiated. The bimodal distributions of clones exposed to 110 and 134 h of 10% FBS were also significantly different than those only exposed to mTESR1 (p
Biotechnol. J. 2015, 10, 1546–1554
DOI 10.1002/biot.201500035
www.biotechnology-journal.com
Research Article
Clonal analysis of individual human embryonic stem cell differentiation patterns in microfluidic cultures Darek J. Sikorski1,2,3, Nicolas J. Caron3, Michael VanInsberghe1, Hans Zahn1, Connie J. Eaves4,5, James M. Piret2,3 and Carl L. Hansen1,6 1 Centre
for High-Throughput Biology, University of British Columbia, Vancouver, BC, Canada of Chemical and Biological Engineering, University of British Columbia, Vancouver, BC, Canada 3 Michael Smith Laboratories, University of British Columbia, Vancouver, BC, Canada 4 Terry Fox Laboratory, British Columbia Cancer Agency, Vancouver, BC, Canada 5 Department of Medical Genetics, University of British Columbia, Vancouver, BC, Canada 6 Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada 2 Department
Heterogeneity in the clonal outputs of individual human embryonic stem cells (hESCs) confounds analysis of their properties in studies of bulk populations and how to manipulate them for clinical applications. To circumvent this problem we developed a microfluidic device that supports the robust generation of colonies derived from single ESCs. This microfluidic system contains 160 individually addressable chambers equipped for perfusion culture of individual hESCs that could be shown to match the growth rates, marker expression and colony morphologies obtained in conventional cultures. Use of this microfluidic device to analyze the clonal growth kinetics of multiple individual hESCs induced to differentiation revealed variable shifts in the growth rate, area per cell and expression of OCT4 in the progeny of individual hESCs. Interestingly, low OCT4 expression, a slower growth rate and low nuclear to cytoplasmic ratios were found to be correlated responses. This study demonstrates how microfluidic systems can be used to enable large scale live-cell imaging of isolated hESCs exposed to changing culture conditions, to examine how different aspects of their variable responses are correlated.
Received Revised Accepted Accepted article online
21 JAN 2015 04 APR 2015 05 JUN 2015 17 JUN 2015
Supporting information available online
Keywords: Clonal analysis · Heterogeneity · Human embryonic stem cell (hESC) · Microfluidic cell culture See accompanying commentary by Fernandes DOI 10.1002/biot.201500307
1 Introduction Human embryonic stem cells (hESCs) are defined by their ability to generate precursors of all three germ layers as well as undifferentiated progeny with the same potentialiCorrespondence: Dr. Carl L. Hansen, Centre for High-Throughput Biology, University of British Columbia, 2125 East Mall, Vancouver, BC, Canada V6T 1Z4. E-mail: [email protected] Abbreviations: BSA, bovine serum albumin; CT, cycle threshold; DIC, differential interference contrast; DMEM/F12, Dulbecco’s modified eagle medium: nutrient mixture F-12; ESC, embryonic stem cell; FBS, fetal bovine serum; hESC, human embryonic stem cell; NTC, no template control; PBS, phosphate buffered saline; PDMS, polydimethylsiloxane; PI, propidium iodide; ROCK inhibitor, Y-27632 Rho-associated kinase inhibitor
1546
ties [1]. The self-renewal property of hESCs allows them to serve as an essentially unlimited source of cells for research and therapeutic purposes. hESCs and their derivatives also display heterogeneity in their outputs as they expand in microenvironments or specialized niches that induce their differentiation [2, 3, 4], and this heterogeneity is also seen when hESCs are induced to differentiate in vitro. Heterogeneity within hESC cultures also affects the fate of their progeny [5, 6, 7], and this can result in the manifestation of diverse self-renewal and differentiation capacities at a clonal level [8, 9, 10]. For clinical applications, such heterogeneity is a significant issue, as incompletely differentiated cells carry a risk of tumorigenesis [11]. Thus there is a real need for an improved understanding of the determinants of hESC heterogeneity, and analytical methods to analyze their extent and control [12, 13, 14, 15].
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Biotechnology Journal
Biotechnol. J. 2015, 10, 1546–1554 www.biotecvisions.com
www.biotechnology-journal.com
The very low survival and plating efficiency of isolated hESCs (1 mm2 in area) [43, 44, 45, 46, 47, 48, 49, 50, 51], with limited reports of single hESC colony cultures in 99% positive) and Tra-1-60 (>99%) (Supporting information, Fig. S4). We next used RT-qPCR to assess OCT4 transcript levels in extracts of individual colonies generated on the microfluidics chip. All chambers on the device were lysed simultaneously followed by the separate elution of each chamber into individual micro-centrifuge tube reactions. A total of 96 chambers were eluted (across two experiments) consisting of 18 NTCs (including empty, dead and re-eluted chambers), 41 chambers perfused exclusively with mTESR1, as well as 24 and 13 chambers exposed to 10% FBS for 110 and 134 h, respectively.
Figure 3. Heterogeneity in the growth rates of individually assessed colonies of CA1S cells. (A) Colony stained with Hoechst 33342 after 190 h of growth in mTeSR1. (B) Colony stained with Hoechst 33342 after 190 h of incubation when exposed to 10% FBS for the last 110 h. (C) Colony stained with Hoechst 33342 after 165 h of incubation when exposed to 10% FBS for the last 134 h. Scale bar for (A)–(C): 200 μm. (D) Distribution of growth rates for all colonies generated in mTESR1. Each point shows the data for a single colony and the solid and dashed green lines show the mean ± the 95% confidence intervals. Histograms showing the distribution of growth rates for each condition projected onto the axis. Growth rates for each clone were calculated by enumerating the cells at the start (by DIC) and the nuclei (stained by Hoechst 33342) at the end of the experiment.
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1551
Biotechnology Journal
Biotechnol. J. 2015, 10, 1546–1554 www.biotecvisions.com
www.biotechnology-journal.com
Figure 4. Heterogeneity in OCT4 transcript levels and relation to different culture conditions. (A) Distribution of OCT4 transcript levels (data pooled from two experiments). Each point shows the result for a single colony and the green lines show the mean ± the 95% confidence intervals. (B) Normalized OCT4 transcript level and growth rates for each colony. (C) Normalized OCT4 transcript levels and area per cell for each colony. (D) Growth rate and area per cell for each colony.
OCT4 and GAPDH transcript levels in the cells recovered from each chamber of each device were then measured by RT-qPCR, with GAPDH levels and Hoechst 33342 cell counts used for normalization. The magnitude of the decrease in mean OCT4 expression (1.00 ± 0.38 for mTESR1 compared to 0.73 ± 0.47 and 0.34 ± 0.56-fold expression for cells in FBS for 110 and 134 h, respectively) was similar to previous reports [31, 58], as with the distribution of growth rates. However, our analysis revealed that individual clones exhibited considerable heterogeneity in OCT4 expression (Fig. 4A), and the emergence of OCT4 negative clones as the cells differentiated. The bimodal distributions of clones exposed to 110 and 134 h of 10% FBS were also significantly different than those only exposed to mTESR1 (p
