pubmed.ncbi.nlm.nih.gov

The Modular µSiM: A Mass Produced, Rapidly Assembled, and Reconfigurable Platform for the Study of Barrier Tissue Models In Vitro - PubMed

The Modular µSiM: A Mass Produced, Rapidly Assembled, and Reconfigurable Platform for the Study of Barrier Tissue Models In Vitro

Molly C McCloskey et al. Adv Healthc Mater. 2022 Sep.

Abstract

Advanced in vitro tissue chip models can reduce and replace animal experimentation and may eventually support "on-chip" clinical trials. To realize this potential, however, tissue chip platforms must be both mass-produced and reconfigurable to allow for customized design. To address these unmet needs, an extension of the µSiM (microdevice featuring a silicon-nitride membrane) platform is introduced. The modular µSiM (m-µSiM) uses mass-produced components to enable rapid assembly and reconfiguration by laboratories without knowledge of microfabrication. The utility of the m-µSiM is demonstrated by establishing an hiPSC-derived blood-brain barrier (BBB) in bioengineering and nonengineering, brain barriers focused laboratories. In situ and sampling-based assays of small molecule diffusion are developed and validated as a measure of barrier function. BBB properties show excellent interlaboratory agreement and match expectations from literature, validating the m-µSiM as a platform for barrier models and demonstrating successful dissemination of components and protocols. The ability to quickly reconfigure the m-µSiM for coculture and immune cell transmigration studies through addition of accessories and/or quick exchange of components is then demonstrated. Because the development of modified components and accessories is easily achieved, custom designs of the m-µSiM shall be accessible to any laboratory desiring a barrier-style tissue chip platform.

Keywords: blood-brain barriers; membranes; modularity; tissue chips; vascular barriers.

© 2022 Wiley-VCH GmbH.

PubMed Disclaimer

Conflict of interest statement

Conflict of Interest

J.L.M. is a cofounder of SiMPore and holds an equity interest in the company. SiMPore is commercializing ultrathin silicon-based technologies including the membranes used in this study. B.E. is an inventor on a provisional U.S. patent application (63/185815) related to the methodology of EECM-BMEC-like cell differentiation.

Figures

Figure 1.
Figure 1.

m-μSiM assembly. A) Fixtures are used to guide components and membrane chip together (left). Fixtures A1 and A2 guide component 1 and membrane assembly, and Fixtures B1 and B2 guide component 1 and 2 assembly. Component 1 is composed of an acrylic top layer with a Transwell-style open well and a PSA sealing layer; Component 2 is composed of a thin bottom channel PSA/PET/PSA layer (“PSA Bottom Channel”) and COP imaging layer (right). B) Assembly is a two-step process. Step 1: Bond Component 1 and Membrane Chip using Fixtures A1/A2. Place the membrane chip on Fixture A1. Place Component 1 inverted over membrane. Use Fixture A2 to press firmly and activate PSA. This irreversibly bonds the membrane to Component 1. Step 2: Bond Components 1 and 2 using Fixtures B1/B2. Place Component 2 in Fixture B1, channel-side up. Place Component 1 with the membrane chip onto Component 2. Use Fixture B2 to press firmly to activate PSA, irreversibly bonding Component 1 and Component 2. C) The modular assembly allows for easy reconfiguration for the application at hand. The example here illustrates the choice of different membrane architectures. The device displayed is a “trench-down”-style device. Component 1’s open well format allows easy cell culture, and access ports provide access to the bottom channel. They are designed to seal-to-fit standard P20 and P200 pipette tips.

Figure 2.
Figure 2.

In situ permeability assay optimization on cell-free devices. A) The imaging plane of a confocal microscope is focused 133 μm below the membrane, within the chip’s trench. Dye diffuses from the well into the trench (left). An example of corresponding image (right) shows a linear region of interest in the center of the membrane (yellow line) where 1D diffusion accurately describes the evolution of fluorescence (see Figure S2, Supporting Information). B) The diffusion coefficient, D, and fluorescence as time reaches infinity, F∞, are solved using 1D Fick’s law describing free diffusion. C) Representative images illustrating 10 kDa Dextran-AF488 diffusion into the trench of an uncoated membrane chip over the course of 10 min. Following the 10-min diffusion, dye from the top well was flushed across the bottom channel to obtain a “Source Intensity” photo. D) Example plots of diffusion across uncoated and collagen/fibronectin-coated chips using 10 kDa Dextran-AF488 (top) and lucifer yellow (LY, bottom). The analytical solutions fit well to the experimental data (normalized root mean square error, NRMSE, <0.1). E) The resulting diffusion coefficients for 10 kDa Dextran-AF488 across uncoated and collagen/fibronectin-coated membranes and LY across collagen/fibronectin-coated membranes. Coating the membrane significantly decreased the apparent diffusion coefficient and larger diffusion coefficients are measured for the smaller molecule, LY. The rapid diffusion through an uncoated membrane was challenging to measure but confirmed a negligible hindrance of the dye by the membrane compared to coated membranes. It overlapped with fluorescence correlation spectroscopy measurements (red bar). N = 3–6 per group. Ordinary one-way ANOVA, p < 0.05. Scalebar = 100 μm.

Figure 3.
Figure 3.

In situ permeability assay optimization using hCMEC/D3. A) A confocal microscope is focused 133 μm below the membrane, within the chip’s trench. Dye diffuses from the well, across an endothelial cell layer, and into the trench. B) Endothelial permeability is calculated assuming constant flux across the monolayer (see the Experimental Section). C) Example plots of the analytical solutions for permeability of 10 kDa Dextran-FITC (top) and 457 Da lucifer yellow (bottom) across an hCMEC/D3 monolayer. The analytical solutions fit well to the experimental data, with low normalized root mean square errors (NRMSE, <0.1).

Figure 4.
Figure 4.

Sampling permeability assay optimization and comparison to in situ method. A) Illustration of the sampling method for collecting dye from the channel. A reservoir pipette tip is added to one port that accesses the bottom channel, and another pipette tip is used to pull media out via reverse pipetting. Media withdrawn is added to a well plate for fluorescence measurements. COMSOL Multiphysics was used to model the B) diffusion or C) flux of lucifer yellow across coated control (B) or cell-seeded devices (C). Time zero illustrates the dye in the bottom channel after 1 h of diffusion or flux, prior to flushing the solution out of the channel. t = 1 s illustrates the flushing process across the channel and out the right port, and t = 4 s shows remaining dye after the flushing is complete. D) COMSOL-generated data were used to determine the volume needed to clear analyte from the channel. The plot shows percent recovery of transported dye, defined as the ratio of total analytes extracted to the total transported, and residual dye left in the chamber in moles (inset). E) hCMEC/D3 permeability to 10 kDa Dextran-FITC (10 kDa Dex) and lucifer yellow (LY). There were no significant differences in measured permeability between the in situ and sampling methods for either dye. Both methods measured significantly higher permeability of hCMEC/D3 to lucifer yellow (LY) compared to 10 kDa Dextran-FITC (10 kDa-Dex). N = 4–5 per group. Two-way ANOVA with Tukey post hoc test, p < 0.05 was used.

Figure 5.
Figure 5.

Establishment of EECM-BMEC-like cell culture in the m-μSiM by a brain barriers laboratory. A) Images of endothelial cells (ECs) at UniBe were acquired on a Nikon E600 Fluorescence microscope using a 10× objective. The field of view (FOV) was centered on the membrane. Images were digitally cropped post-acquisition to better visualize molecular stains. B) EECM-BMEC-like cells expressed key junctional molecules when cultured on m-μSiM devices. C) EECM-BMEC-like cells expressed key cell adhesion molecules upon exposure to proinflammatory stimuli when cultured on m-μSiM devices. Nonstimulated (NS) and stimulated (0.1 ng mL−1 TNFα + 2 IU mL−1 IFNγ) for 16–20 h. Cells have comparable expression patterns to published data from Chamber Slides and Transwells. Scalebar = 100 μm.

Figure 6.
Figure 6.

Interlaboratory reproducibility between nonengineering brain barriers and bioengineering laboratories. A) EECM-BMEC-like cells were cultured in the m-μSiM at both the University of Bern (UniBe) and University of Rochester (UR). Cells were either not stimulated (NS) or stimulated (0.1 ng mL−1 TNFα + 2 IU mL−1 IFNγ) for 16–20 h and stained for ICAM-1. Mean fluorescence intensity was measured and normalized to each laboratory’s respective average of NS mean fluorescence. N = 3 per group. Two-way ANOVA with Tukey post hoc test, p < 0.05 was used. B) EECM-BMEC-like cells were cultured in the m-μSiM at UniBe and UR for 2, 4, or 6 d or in Transwell filters for 6 d, and permeability was measured. There were no significant differences in permeability between labs or culturing platforms upon barrier maturation (6 d in m-μSiM). Red bar: Transwell data were comparable to a previous publication of matching cell culture conditions and assayed for permeability to similar-sized sodium fluorescein (FASEB J 34(12):16693–16715(2020)). N = 4–16 per group. Two-way ANOVA with Tukey post hoc test was used, comparisons only displayed for relevant p < 0.05. Comparisons between laboratories for noncorresponding culture conditions were excluded on the plot.

Figure 7.
Figure 7.

Demonstration of the modular function of m-μSiM. A) Side-by-side coculture was achieved by swapping one window nanoporous (NPN) membranes with two window NPN membranes and by addition of a cell culture insert. EECM-BMEC-like cells were cultured in one chamber and primary human astrocytes (NHA) in the other chamber, with no apparent cross-contamination of cells. Chambers were stained for EECM-BMEC-like cell marker, Claudin-5 (green) and nuclear marker Hoechst (blue). Phase images were acquired on a Nikon Eclipse Ts2 phase contrast microscope and fluorescence images on an Andor Spinning Disc Confocal Microscope. Scalebar = 100 μm. B) Neutrophil migration across EECM-BMEC-like cells cultured on 0.625%, 3 μm dual-scale membranes. Neutrophils were seen migrating across the endothelium (02:32) and through a micropore (17:54), entering the bottom channel (18:14) (time in min:s). Videos were acquired on a Nikon Ti2 Eclipse inverted microscope using a long working distance 40× objective in phase contrast. C) T-cell migration across 1.25%, 3 μm dual-scale m-μSiM was quantified by flow cytometry of CellTracker Green CMFDA-labeled migrated T-cells. Transmigrated T-cells can only access the bottom channel at the membrane window region of the chip. Remaining adhered T-cells were visualized via epifluorescence imaging, paired with phase contrast imaging of the endothelial layer. Images were acquired on a Nikon Eclipse E600 Microscope using a 10× objective. Scalebar = 100 μm.

Similar articles

  • Modelling a Human Blood-Brain Barrier Co-Culture Using an Ultrathin Silicon Nitride Membrane-Based Microfluidic Device.

    Hudecz D, McCloskey MC, Vergo S, Christensen S, McGrath JL, Nielsen MS. Hudecz D, et al. Int J Mol Sci. 2023 Mar 15;24(6):5624. doi: 10.3390/ijms24065624. Int J Mol Sci. 2023. PMID: 36982697 Free PMC article.

  • Peer Play.

    Scott HK, Cogburn M. Scott HK, et al. 2023 Jul 4. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 Jan–. 2023 Jul 4. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 Jan–. PMID: 30020595 Free Books & Documents.

  • Far Posterior Approach for Rib Fracture Fixation: Surgical Technique and Tips.

    Manes TJ, DeGenova DT, Taylor BC, Patel JN. Manes TJ, et al. JBJS Essent Surg Tech. 2024 Dec 6;14(4):e23.00094. doi: 10.2106/JBJS.ST.23.00094. eCollection 2024 Oct-Dec. JBJS Essent Surg Tech. 2024. PMID: 39650795 Free PMC article.

  • Depressing time: Waiting, melancholia, and the psychoanalytic practice of care.

    Salisbury L, Baraitser L. Salisbury L, et al. In: Kirtsoglou E, Simpson B, editors. The Time of Anthropology: Studies of Contemporary Chronopolitics. Abingdon: Routledge; 2020. Chapter 5. In: Kirtsoglou E, Simpson B, editors. The Time of Anthropology: Studies of Contemporary Chronopolitics. Abingdon: Routledge; 2020. Chapter 5. PMID: 36137063 Free Books & Documents. Review.

  • Adenosine Deaminase Deficiency.

    Hershfield M, Tarrant T. Hershfield M, et al. 2006 Oct 3 [updated 2024 Mar 7]. In: Adam MP, Feldman J, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2025. 2006 Oct 3 [updated 2024 Mar 7]. In: Adam MP, Feldman J, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2025. PMID: 20301656 Free Books & Documents. Review.

Cited by

References

    1. Sistare FD, Mattes WB, LeCluyse EL, ILAR J 2016, 57, 186. - PubMed
    1. Kendall LV, Owiny JR, Dohm ED, Knapek KJ, Lee ES, Kopanke JH, Fink M, Hansen SA, Ayers JD, ILAR J 2018, 59, 177. - PubMed
    1. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker HV, Xu W, Richards DR, McDonald-Smith GP, Gao H, Hennessy L, Finnerty CC, Lopez CM, Honari S, Moore EE, Minei JP, Cuschieri J, Bankey PE, Johnson JL, Sperry J, Nathens AB, Billiar TR, West MA, Jeschke MG, Klein MB, Gamelli RL, Gibran NS, Brownstein BH, Miller-Graziano C, Calvano SE, Mason PH, et al., Proc. Natl. Acad. Sci. USA 2013, 110, 3507. - PMC - PubMed
    1. Taylor G, Vaccine 2017, 35, 469. - PMC - PubMed
    1. Veseli BE, Perrotta P, De Meyer GRA, Roth L, Van der Donckt C, Martinet W, De Meyer GRY, Eur. J. Pharmacol 2017, 816, 3. - PubMed

Publication types

MeSH terms

Substances