CoChat

Microbioreactor Arrays for Full Factorial Screening of Exogenous and Paracrine Factors in Human Embryonic Stem Cell Differentiation

Open access

The DOI matches its Crossref record; 35 of the 35 references it cites were found in a registry or an index.

Abstract

Timed exposure of pluripotent stem cell cultures to exogenous molecules is widely used to drive differentiation towards desired cell lineages. However, screening differentiation conditions in conventional static cultures can become impractical in large parameter spaces, and is intrinsically limited by poor spatiotemporal control of the microenvironment that also makes it impossible to determine whether exogenous factors act directly or through paracrine-dependent mechanisms. We detail here the development of a continuous flow microbioreactor array platform that combines full-factorial multiplexing of input factors with progressive accumulation of paracrine factors through serially-connected culture chambers, and further, the use of this system to explore the combinatorial parameter space of both exogenous and paracrine factors involved in human embryonic stem cell (hESC) differentiation to a MIXL1-GFP(+) primitive streak-like population. We show that well known inducers of primitive streak (BMP, Activin and Wnt signals) do not simply act directly on hESC to induce MIXL1 expression, but that this requires accumulation of surplus, endogenous factors; and, that conditioned medium or FGF-2 supplementation is able to offset this. Our approach further reveals the presence of a paracrine, negative feedback loop to the MIXL1-GFP(+) population, which can be overcome with GSK-3β inhibitors (BIO or CHIR99021), implicating secreted Wnt inhibitory signals such as DKKs and sFRPs as candidate effectors. Importantly, modulating paracrine effects identified in microbioreactor arrays by supplementing FGF-2 and CHIR in conventional static culture vessels resulted in improved differentiation outcomes. We therefore demonstrate that this microbioreactor array platform uniquely enables the identification and decoding of complex soluble factor signalling hierarchies, and that this not only challenges prevailing strategies for extrinsic control of hESC differentiation, but also is translatable to conventional culture systems.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1371/journal.pone.0052405

Cites

35 of 35 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Microbioreactor array design and validation. A Microbioreactor array photomask design, with key features marked. B Perspective schematic view showing assembly of glass substrate and PDMS structural layers 1 and 2. Via holes join microchannel structures between layers. C Computational fluid dynamic model of velocity field through centreline planes at nominal operating conditions. Chambers are 1.63 mm in diameter and 250 µm high. D Design normalised concentrations of factors in each column, corresponding to panels E and F. Stock factor and buffer solutions are provided at normalised concentrations of 3 and 0, respectively, to allow for subsequent dilution. E Photograph of microbioreactor array filled with red, yellow and blue food dyes (representing factors A1, B1 & C1, respectively), and mixed with PBS (buffers A0, B0, & C0). F Fluorimetric quantification of soluble factor levels in each column. Stock solution of 40 kDa FITC-dextran was provided at 100 µM, therefore the design concentration levels are 0, 16.7 and 33.3 µM. Bars represent mean ± SD of 2 independently fabricated devices. G Heatmap of number of nuclei in each chamber, with individual numbers marked. Cell numbers do not represent densities used in microbioreactor experiments. H Phase contrast images of hESCs imaged 0 and 2 h after seeding into microbioreactor arrays. Chambers are ∼1.63 mm in diameter.
Figure 1 Microbioreactor array design and validation. A Microbioreactor array photomask design, with key features marked. B Perspective schematic view showing assembly of glass substrate and PDMS structural layers 1 and 2. Via holes join microchannel structures between layers. C Computational fluid dynamic model of velocity field through centreline planes at nominal operating conditions. Chambers are 1.63 mm in diameter and 250 µm high. D Design normalised concentrations of factors in each column, corresponding to panels E and F. Stock factor and buffer solutions are provided at normalised concentrations of 3 and 0, respectively, to allow for subsequent dilution. E Photograph of microbioreactor array filled with red, yellow and blue food dyes (representing factors A1, B1 & C1, respectively), and mixed with PBS (buffers A0, B0, & C0). F Fluorimetric quantification of soluble factor levels in each column. Stock solution of 40 kDa FITC-dextran was provided at 100 µM, therefore the design concentration levels are 0, 16.7 and 33.3 µM. Bars represent mean ± SD of 2 independently fabricated devices. G Heatmap of number of nuclei in each chamber, with individual numbers marked. Cell numbers do not represent densities used in microbioreactor experiments. H Phase contrast images of hESCs imaged 0 and 2 h after seeding into microbioreactor arrays. Chambers are ∼1.63 mm in diameter.CC BY · PubMed Central
Figure 2 Microbioreactor screening of HES3(MIXL1GFP/w) cells undergoing mesendodermal differentiation demonstrates paracrine dependencies. A Screening panel showing array conditions in each column. Numbers indicate concentrations of BIO (µM), BMP-4 (ng/mL) and Activin A (ng/mL). B Confocal images of HES3(MIXL1GFP/w) hESCs expressing GFP and counterstained with Hoechst at experiment endpoint. Medium flow direction was from top to bottom. C Heat maps of total fluorescence intensities in the array (arbitrary units). D Higher-magnification confocal images of individual wells within the microbioreactor array highlighting position-dependent phenotypes, scale bar: 500 µm. E Main effect magnitudes of exogenous factors (BIO, BMP-4, Activin A) and positional dependency (Position) on expression index IMIXL1-GFP. Units represent global standard deviations relative to global mean. F Interaction effect magnitudes of combinations of 2 factors on expression index IMIXL1-GFP. Full dataset is shown in Figures S2, S3. G Traces of expression index IMIXL1-GFP and DNA-normalised expression index IMIXL1-GFP/DNA versus position coordinate for 0 (Column 6) and 1 µM BIO (Column 15) showing extended expression under BIO treatment.
Figure 2 Microbioreactor screening of HES3(MIXL1GFP/w) cells undergoing mesendodermal differentiation demonstrates paracrine dependencies. A Screening panel showing array conditions in each column. Numbers indicate concentrations of BIO (µM), BMP-4 (ng/mL) and Activin A (ng/mL). B Confocal images of HES3(MIXL1GFP/w) hESCs expressing GFP and counterstained with Hoechst at experiment endpoint. Medium flow direction was from top to bottom. C Heat maps of total fluorescence intensities in the array (arbitrary units). D Higher-magnification confocal images of individual wells within the microbioreactor array highlighting position-dependent phenotypes, scale bar: 500 µm. E Main effect magnitudes of exogenous factors (BIO, BMP-4, Activin A) and positional dependency (Position) on expression index IMIXL1-GFP. Units represent global standard deviations relative to global mean. F Interaction effect magnitudes of combinations of 2 factors on expression index IMIXL1-GFP. Full dataset is shown in Figures S2, S3. G Traces of expression index IMIXL1-GFP and DNA-normalised expression index IMIXL1-GFP/DNA versus position coordinate for 0 (Column 6) and 1 µM BIO (Column 15) showing extended expression under BIO treatment.CC BY · PubMed Central
Figure 3 Confirmation and screening of paracrine factors involved in hESC mesendodermal differentiation. A Screening panel showing array conditions, corresponding to B-C. Numbers indicate concentrations of induction-conditioned and factor re-supplemented medium (CM, final % v/v), IWP-4 (µM), and CHIR99021 (CHIR, µM). B Confocal images of HES3(MIXL1GFP/w) hESCs expressing GFP and co-immunostained in situ for NCAM (CD56) at experiment endpoint. Flow direction was from top to bottom. C Heat maps of total fluorescence intensities in the array (arbitrary units). D Selected interaction effect plot showing average effect magnitudes for combinations of CM and Position on expression index IMIXL1-GFP and DNA-normalised expression index IMIXL1-GFP/DNA, highlighting the shift towards earlier rows with CM treatment (legend represents CM concentration in % v/v). Full dataset, Figures S6, S7. E Traces of expression index IMIXL1-GFP and DNA-normalised expression index IMIXL1-GFP/DNA versus position coordinate for 0 (Column 19), 2.5 (Column 20), and 5 µM CHIR (Column 21) showing extended expression under CHIR treatment. F Blocking of MIXL1-GFP induction in static cultures with IWP-4, and improvement and rescue of induction with CHIR. Bars represent mean ± SD of 4 biological replicates over 2 independent experiments. * indicates p<0.05 relative to +None condition.
Figure 3 Confirmation and screening of paracrine factors involved in hESC mesendodermal differentiation. A Screening panel showing array conditions, corresponding to B-C. Numbers indicate concentrations of induction-conditioned and factor re-supplemented medium (CM, final % v/v), IWP-4 (µM), and CHIR99021 (CHIR, µM). B Confocal images of HES3(MIXL1GFP/w) hESCs expressing GFP and co-immunostained in situ for NCAM (CD56) at experiment endpoint. Flow direction was from top to bottom. C Heat maps of total fluorescence intensities in the array (arbitrary units). D Selected interaction effect plot showing average effect magnitudes for combinations of CM and Position on expression index IMIXL1-GFP and DNA-normalised expression index IMIXL1-GFP/DNA, highlighting the shift towards earlier rows with CM treatment (legend represents CM concentration in % v/v). Full dataset, Figures S6, S7. E Traces of expression index IMIXL1-GFP and DNA-normalised expression index IMIXL1-GFP/DNA versus position coordinate for 0 (Column 19), 2.5 (Column 20), and 5 µM CHIR (Column 21) showing extended expression under CHIR treatment. F Blocking of MIXL1-GFP induction in static cultures with IWP-4, and improvement and rescue of induction with CHIR. Bars represent mean ± SD of 4 biological replicates over 2 independent experiments. * indicates p<0.05 relative to +None condition.CC BY · PubMed Central
Figure 4 Interrogation of FGF stimulation and MEK inhibition on hESC mesendodermal differentiation in HES3(MIXL1GFP/w) hESCs. A Screening panel showing array conditions. Numbers indicate concentrations of FGF-2 (ng/mL), PD0325901 (PD; µM), and CHIR99021 (CHIR; µM). B Confocal tile scan image of HES3(MIXL1GFP/w) hESCs expressing GFP and and co-immunostained in situ for NCAM (CD56) and Hoechst at 2.5 d experiment endpoint. Flow direction was from top to bottom. C Heatmaps of total fluorescence intensities in the array (arbitrary units). D Higher-magnification images of cells in highest MIXL1-expressing conditions – Column 21, Row 7– demonstrating efficient induction of MIXL1-GFP and NCAM. Scale bar: 500 µm. E Main effect magnitudes of factors on expression index IMIXL1-GFP. Units represent global standard deviations relative to global mean. F Selected interaction effect plots showing average effect magnitudes for combinations of FGF-2 and Position on expression index IMIXL1-GFP and DNA-normalised expression index IMIXL1-GFP/DNA, highlighting the shift towards earlier rows with FGF-2 treatment (legend represents FGF-2 concentration in ng/mL). G Chemical inhibition of MIXL1-GFP induction with MEK cascade inhibitor PD0325901 and FGF receptor tyrosine kinase inhibitor PD173074. Bars represent mean ± s.d., n = 2−3 from one representative experiment. H-J Addition of FGF-2 or FGF-2+ CHIR99021 improved induction efficiency (H) and median fluorescence intensity (I) in static controls, but without significantly increasing total cell count (J). Bars represent mean ± s.d., n = 4 from 2 independent experiments. * p<0.05 compared to +None condition, one-way ANOVA.
Figure 4 Interrogation of FGF stimulation and MEK inhibition on hESC mesendodermal differentiation in HES3(MIXL1GFP/w) hESCs. A Screening panel showing array conditions. Numbers indicate concentrations of FGF-2 (ng/mL), PD0325901 (PD; µM), and CHIR99021 (CHIR; µM). B Confocal tile scan image of HES3(MIXL1GFP/w) hESCs expressing GFP and and co-immunostained in situ for NCAM (CD56) and Hoechst at 2.5 d experiment endpoint. Flow direction was from top to bottom. C Heatmaps of total fluorescence intensities in the array (arbitrary units). D Higher-magnification images of cells in highest MIXL1-expressing conditions – Column 21, Row 7– demonstrating efficient induction of MIXL1-GFP and NCAM. Scale bar: 500 µm. E Main effect magnitudes of factors on expression index IMIXL1-GFP. Units represent global standard deviations relative to global mean. F Selected interaction effect plots showing average effect magnitudes for combinations of FGF-2 and Position on expression index IMIXL1-GFP and DNA-normalised expression index IMIXL1-GFP/DNA, highlighting the shift towards earlier rows with FGF-2 treatment (legend represents FGF-2 concentration in ng/mL). G Chemical inhibition of MIXL1-GFP induction with MEK cascade inhibitor PD0325901 and FGF receptor tyrosine kinase inhibitor PD173074. Bars represent mean ± s.d., n = 2−3 from one representative experiment. H-J Addition of FGF-2 or FGF-2+ CHIR99021 improved induction efficiency (H) and median fluorescence intensity (I) in static controls, but without significantly increasing total cell count (J). Bars represent mean ± s.d., n = 4 from 2 independent experiments. * p<0.05 compared to +None condition, one-way ANOVA.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Cited by

See all 48

Metadata

Authors
Titmarsh, D. M., Hudson, J. E., Hidalgo, A., Elefanty, A. G., Stanley, E. G., Wolvetang, E. J., & Cooper-White, J. J.
Affiliations
  • The University of Queensland (AU)
  • Royal Children's Hospital (AU)
Corresponding author
Cooper-White, Justin J.
Venue
PLoS ONE
Published
2012-12-26
Volume
7
Issue
12
Pages
e52405
ISSN
1932-6203 (ISSN-L)
OpenAlex
W2065174416
Cited by
48
References
34
Citations by year
2026: 1 · 2023: 1 · 2021: 5 · 2020: 3 · 2019: 5 · 2018: 5 · 2017: 3 · 2016: 7 · 2015: 6 · 2014: 5 · 2013: 7
FWCI
2.31
Type
journal article
Language
en
Open access
Locations
  • PLoS ONE (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Monash University Research Portal (Monash University) (submittedVersion, cc-by)
  • Minerva Access (University of Melbourne) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Figshare (submittedVersion, cc-by)
Funders
Australian National Fabrication Facility
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Pluripotent Stem Cells Research
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Developmental Biology and Gene Regulation
  • Physical Sciences › Engineering › Biomedical Engineering › 3D Printing in Biomedical Research
Keywords
human embryonic stem cell differentiation, paracrine signaling, Mixl1, BMP signaling, activin signaling, Wnt signaling, conditioned medium, FGF2, GSK-3β inhibitors, CHIR99021, secreted frizzled-related proteins, negative feedback loop
MeSH terms
Cell Differentiation, Drug Evaluation, Preclinical, Fibroblast Growth Factors, Humans, Reproducibility of Results, Signal Transduction, Culture Media, Conditioned, Homeodomain Proteins, Bioreactors, Paracrine Communication, Activins, Pluripotent Stem Cells, Microarray Analysis, Wnt Proteins, Embryonic Stem Cells, Bone Morphogenetic Protein 4

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.