2022 Americas Building Biology in 3D Symposium

This two-day event addressed the successes and limitations of using 3D systems in discovery and applied research while acknowledging the need for improvements to ensure widespread adoption.

Topics included:

  • current and near-future enabling technologies,
  • applications of such systems in high-throughput screening,
  • advances in imaging and analysis of data generated
  • and the expansion into novel model systems.

Presentations are only published with the permission of the presenters.

Nila C. Wu

PhD Candidate

University of Toronto

Nila is a PhD candidate at the Institute of Biomedical Engineering at the University of Toronto. She received her BSc in Cell Biology at McGill University in Montreal. Under the supervision of Professor Alison McGuigan, Nila’s research focuses on developing an easy-to-use, higher-throughput 3D-engineered tumor model and complementary image-based assays, to interrogate the dynamics of patient-derived organoids upon standard-of-care chemotherapy treatment, and the impact of the tumor microenvironment on treatment response. Nila was recently awarded the PRiME Fellowship, where she is co-supervised by Dr. Cheryl Arrowsmith from the Toronto Structural Genomics Centre, and has received support from NSERC CREATE graduate scholarships, the Barbara and Frank Milligan Graduate Fellowship, and the BME Doctoral Completion Award. Nila is an active member of the graduate student community and is Programming Co-chair for the student and young investigator section at the upcoming TERMIS-Americas Conference.

Shannon Mumenthaler, PhD

Faculty and Chief Translational Research Officer

Lawrence J Ellison Institute for Transformative Medicine

Shannon Mumenthaler, Ph.D., is an Assistant Professor of Medicine and Biomedical Engineering at the University of Southern California. She is also Chief Translational Research Officer for the Lawrence J Ellison Institute for Transformative Medicine of USC, which is a translational institute that bridges research and innovation by bringing together researchers, patients, and physicians dedicated to improving health of the human condition. Dr. Mumenthaler’s research program is centered around the development and utilization of physiologically relevant, organ-dependent tumor models that allow for the characterization of colorectal cancer cellular dynamics, and serve as a platform for testing specific therapeutic modalities to prevent or delay tumor progression. Specifically, she is integrating biologically-inspired 3D model systems (i.e., organoids and organs-on-chips) with dynamic imaging and computational approaches to provide new insights into the significance of the physical and cellular microenvironment on tumor progression.

Sarah M. Moss, M.S.

Staff Scientist II

Advanced Solutions Life Sciences

Sarah Moss is a Staff Scientist at Advanced Solutions Life Sciences. She joined the Advanced Solutions teams in early 2018 and has since contributed to various projects related to 3D tissue design and automation. Specifically, her focus is on studying isolated human microvessels and developing strategies for their incorporation into various 3D tissue environments. She previously completed her graduate work at The Ohio State University with a focus in microfluidics and cancer biology.

Fabio Stossi, PhD

Assistant Professor

Baylor College of Medicine

Dr Fabio Stossi, Associate Professor in the Department of Molecular and Cellular Biology, Baylor Collge of Medicine, Houston, TX; is a native of Milan, Italy. He completed his BS and PhD studies at Universita’ degli Studi di Milano, in Pharmaceutical Chemistry and Technology, and in Endocrinology and Metabolism. He then moved to the US as a postdoc in Dr. Benita S. Katzenellenbogen’s laboratory at University of Illinois at Urbana-Champaign, where he became interested in gene transcription and its modulation by steroid receptors, particularly focusing on estrogen receptor in breast cancer. He joined Dr. Michael A. Mancini’s group at Baylor College of Medicine, as an Assistant Professor, to explore the use of imaging technoques and high content analysis in steroid receptor biology. He is currently Associate Professor, Technical Director of the Integrated Microscopy Core and group leader for imaging in the GCC Center for Advanced Microscopy and Image Informatics. His interests are imaging and analysis of single cell gene transcription, development of novel analysis methods for single cell measurements, and assay development in environmental toxicology.

Timothy Spicer, Ph.D.

Co-Director, Scripps Molecular Screening Center

Scripps Research-Florida

Tim Spicer joined Scripps early on at its onset in Jupiter Florida and has been working there for 10 years. Prior to that he was employed at Bristol-Myers Squibb Co. for nine years as a Research Scientist in the Department of Lead Discovery and Profiling, Discovery Technologies and Infectious Diseases. Tim received an MS in microbiology at the SUNY Health Science Center at Syracuse. He has authored 75 publications and 4 patents. Tim works across multiple organizations and currently co-directs the Scripps Molecular Screening Center. He leads a team that implements assays and supports screening on fully automated platforms. He has experience developing anti-microbial, GPCR, kinase, protease, nuclear receptor, and/or ion channel assays in a variety of HTS-compatible formats (reporter-gene, TR-FRET, HTRF, fluorescence, luminescence, and absorbance). He is recognized at Scripps and by collaborators as a leader in evaluating new assay technologies for the purpose of making UHTS the standard.

Greg Luerman, PhD

Vice President, Research & Partnerships

Curi Bio

Hans Clevers, MD, PhD

Head of Pharma Research and Early Development

F.Hoffmann-La Roche Ltd

Hans Clevers obtained his MD degree in 1984 and his PhD degree in 1985 from the University Utrecht, the Netherlands. His postdoctoral work (1986-1989) was done with Cox Terhorst at the Dana-Farber Cancer Institute of the Harvard University, Boston, USA. From 1991-2002 Hans Clevers was Professor in Immunology at the University Utrecht and, since 2002, Professor in Molecular Genetics. From 2002-2012 he was director of the Hubrecht Institute in Utrecht. From 2012-2015 he was President of the Royal Netherlands Academy of Arts and Sciences (KNAW). From June 2015-2019 he was director Research of the Princess Máxima Center for pediatric oncology. As of March 18th, 2022 Hans Clevers is the Head of Pharma Research and Early Development and a Member of the Enlarged Corporate Executive Committee of F.Hoffmann-La Roche Ltd , in Basel Switzerland.

Jason Yim

PhD student

MIT

Jason Yim is a first year PhD student at MIT studying Computer Science and advised by Regina Barzilay and Tommi Jaakkola. His research interests are in using machine learning to understand the natural sciences, particularly biology. Jason graduated with a Bachelor of Sciences in Computer Science and Applied Mathematics from Johns Hopkins University in 2018 and was employed by DeepMind between 2018-2021, working on topics ranging from AI for medical imaging and protein structure modeling.

Allysa Stern, PhD

Scientist II, Product Applications

Cell Microsystems

Dr. R. Allysa Stern, M.S., Ph.D., is an accomplished cell biologist with extensive experience using cell-based assays to study mechanisms of disease and toxicity. She received her Ph.D. in Physiology from North Carolina State University with a concentration on cellular physiology and disease modeling. She has years of experience developing novel 2D and 3D in vitro models using a variety of cell types from multiple tissues, including primary hepatocytes and stem cells. As a Product Applications Scientist at Cell Microsystems, she leverages her expertise in advanced imaging and cellular and molecular biology techniques for new product and assay development and customer support.

Stefan N. Lukianov, BS, BS, MS, AM

Founder/CEO

Salve Therapeutics

Stefan N. Lukianov, AM MS is the first-time JHU student founder of Salve Therapeutics.  He has undergraduate degrees from the University of Maine and masters from the University of Pittsburgh and Harvard University in the biomedical sciences.  He has worked in reputable labs at Boston Children’s Hospital, Brigham and Women’s Hospital, McLean Hospital and the UPMC Hillman Cancer Center. He also has extensive experience in science journalism, having published and edited for ASBMB Today, ACS C&EN Show Daily, the Harvard Medical Student Review and MIT Science Policy Review.  Stefan also loves teaching and has held a diverse array of private and public education roles at various grade levels in STEM fields.

Betty Li, PhD

Facility Manager

Sunnybrook Research Institute

Betty is currently the facility manager for the High Content Cellular Analysis (HiCCA) Lab at Sunnybrook Research Institute in Toronto. She graduated with a Ph.D. in Biomedical Engineering from the University of Toronto. Her Ph.D. was in developing a novel 3-dimensional invasion assay in a microfluidic platform called Digital Microfluidics.

Hong Wang, M.Sc

Scientist

Boehringer Ingelheim Pharmaceutical Inc.

Key:

Complete
Failed
Available
Locked
Opening Session
04/26/2022 at 9:00 AM (EDT)  |  90 minutes
04/26/2022 at 9:00 AM (EDT)  |  90 minutes --Welcome and Opening Remarks-- --Keynote: Nate Coussens, Ph.D., Frederick National Laboratory for Cancer Research Patient-Derived Three-Dimensional Tumor Models for High-Throughput Screening of Oncology Compounds--
Enabling Technologies
Capturing biological complexity in a tumor-on-a-chip model
Open to view video.
Open to view video. The highly complex and evolving nature of cancer makes it challenging to study. Here we describe a microfluidic organ-on-chip platform, incorporating tissue-tissue interfaces and physical forces, to support novel interrogations of colorectal cancer progression. We will cover advancements made through combining organ-on-chip models with high content imaging and mass spectrometry-based metabolomics to provide measurements that capture spatial and temporal dynamics. This work reveals important interactions between colorectal cancer cells and their microenvironment, which can be used to prevent or delay cancer progression.
3D Microgels to Quantify Tumor Cell Properties and Therapy Response Dynamics
Open to view video.
Open to view video. Background Tumors contain heterogeneous and dynamic populations of cells that cannot all be targeted by traditional chemotherapies. There is a need therefore, to develop novel treatment strategies that target diverse tumor cell properties. Identifying successful strategies is challenging however, as current approaches have relied on unrepresentative culture models, such as cell line monolayers, and with treatment response being assessed using whole-well, endpoint assays that do not capture tumor heterogeneity or the reactive nature of the disease. Here, we report an in vitro culture platform using micro-molded hydrogels (microgels), the Gels for Live Analysis of Compartmentalized Environments (GLAnCE) platform in a 96-well plate format (96-GLAnCE), to address these limitations. Methodology In 96-GLAnCE, cells embedded in gel are micro-molded in a custom-made plate bottom to generate uniform, thin microgels, that are advantageous for image acquisition using widefield microscopy, as they lack the typical meniscus associated with other 3D-ECM culture methods. Upon final assembly with one microgel per well, images are acquired using a commercial automated widefield microscope, and continuous readouts of tumor cell response are measured. As a result of the design and seeding method, 96-GLAnCE requires only 3 µL of reagent per microgel, which enables the incorporation of precious cancer patient-derived organoids (PDOs). Results We first validated the design of 96-GLAnCE by demonstrating lower variation compared to other 3D-ECM models (gel domes and gel plugs) for cellular image-based analysis. Next, we used a cancer cell line and PDOs to demonstrate two use cases for conducting a longitudinal assay in 96-GLAnCE. First, we model increased cancer cell growth in the direct presence of the tumour stromal cell population, cancer-associated fibroblasts. Second, we model post-treatment cancer growth (regrowth) in response to standard of care chemotherapy treatment, and showcase discrepancies captured in our system that are overlooked in traditional endpoint drug response assays. Moreover, we reveal subpopulations of drug-treated cells that dominate during regrowth. Conclusion & Significance 96-GLAnCE enables spatial and temporal resolution of engineered tumor cultures by using longitudinal automated imaging and patient-derived samples, in a 3D extracellular matrix (ECM) environment. Here we used 96-GLAnCE to quantify both tumor cell growth and regrowth after treatment, to distinguish heterogeneous cellular responses in the system, either by cell type or subpopulations. 96-GLAnCE is a versatile and robust platform, that provides an additional tool for pre-clinical anti-cancer drug discovery for the identification of novel targets with translatable clinical significance.
Robotic Automation of Organoid Culture and Analysis
Open to view video.
Open to view video.
High Throughput Advanced Cellular Models
Imaging estrogen receptors activity and development of single cell quality control pipelines
Open to view video.
Open to view video. Phenotypic heterogeneity is a central dogma in biology, yet its measurement, reproducibility and use in high throughput screening is underutilized, with preferences given to bulk analysis of cell populations. We used the Estrogen Receptor (ER) pathway as a test case to develop novel quality control pipelines to qualify single cell high throughput experiments, and used heterogeneity metrics to measure effects of common environmental toxicants. We also analyzed transcriptional responses of cells to hormones, highlighting heterogeneity not only at the single cell, but also at the single allele level. Interestingly, we identified molecular mechanisms that the cell utilizes to control the heterogeneity of transcriptional responses to a stimulus. Expansion of these methods to 3D complex models and other assays is currently underway. Collectively, these high throughput-amenable approaches and pipelines are poised to greatly aid in the understanding of the role of phenotypic heterogeneity in biological responses, and serve as a robust model for next-generation drug screening and toxicological assays.
Drug Response Profiling Using 3D Tumor Models
Open to view video.
Open to view video.
Modeling Contractile Diseases Using Scalable 3D Engineered Muscle Tissues for Drug Discovery
Open to view video.
Open to view video. Author Block Bonnie Berry – Director, Stem Cells & Tissue Engineering, Curi Bio; Kevin Gray – Sr. Product Development Engineer, Curi Bio; Samir Kharoufeh – Production Manager, Curi Bio; Shawn Luttrell – Sr. Research Scientist, Curi Bio; Jesse Macadangdang – Sr. Research Scientist, Curi Bio; Christal Worthen – Director of Therapeutic Assays, Curi Bio Model systems that accurately recapitulate healthy and diseased function in a dish are critical for the development of novel therapeutics. For cardiac and skeletal muscle diseases, direct assessment of contractile output constitutes the most reliable metric with which to assess overall tissue function, as other ‘proxy’ measurements are poor predictors of muscle strength. 3D engineered muscle tissues (EMTs) derived from iPSCs hold great potential for modeling contractile function. However, the bioengineering strategies required to generate these predictive models are oftentimes out of reach for most investigators. Here, we have developed an automation-friendly platform and device that utilizes 3D EMTs in conjunction with a label-free magnetic sensing array. The platform enables robotically-assisted, highly reproducible fabrication of 3D EMTs using virtually any cell source, and is coupled with a highly parallel direct measurement of contractile strength. This approach enables the stratification of healthy and diseased muscle phenotypes and facilitates dose-dependent compound safety and efficacy screening for evaluation of a drug’s effect on contractile output. We will present a 3D model of Duchenne muscular dystrophy (DMD) that utilizes skeletal muscle EMTs formed from an isogenic pair of healthy and diseased cells. These constructs achieve robust twitch and tetanic responses upon stimulation. We then use these models to compare functional metrics of contractility such as force and fatigability over weeks or months. Our platform also lends itself to the utilization of other optically-based assays such as calcium detection, which is a critical phenotype in diseases such as DMD where calcium handling is misregulated. We will also present data showing both acute and chronic drug toxicity in both cardiac and skeletal muscle EMTs including a drug (BMS-986094) that failed clinical trials due to unanticipated cardiotoxicity. These data demonstrate a first-and-only commercial platform for high-throughput assessment of 3D skeletal and cardiac muscle contraction with potential for widespread adoption within the drug development field.
Point-of-use, automated fabrication of a customizable, 3D human vascularized liver tissue
Open to view video.
Open to view video.
Advances in Imaging and Analysis
Artificial intelligence to predict eye disease from 3D ocular images
Open to view video.
Open to view video. Artificial intelligence (AI) through deep learning has made tremendous progress in building automated systems for understanding complex, structured data. This is fortunate because as medical imaging modalities become more advanced, so does the data they generate; to the point where humans cannot be relied on to make critical decisions when we cannot fully understand the data. In this talk, I will discuss one such instance of attempting to use AI for understanding complex 3D data and decision making based on our recent work "Predicting conversion to wet age-related macular degeneration using deep learning". I will discuss challenges and methods of building AI-based learning systems on real life clinical data.
Image, Identify, and Isolate Single Organoids with the CellRaft AIR System
Open to view video.
Open to view video. Author Block Lexi Land – Research Associate, Product Applications, Cell Microsystems, Inc.; Keith Williams – Software Engineer, Cell Microsystems, Inc.; Brandon Thompson – Senior Product Development Engineer, Cell Microsystems, Inc.; Rob McClellan – Director of Software Systems Engineering, Cell Microsystems, Inc.; Steven Gebhart – Director of Engineering, Cell Microsystems, Inc.; Jessica Hartman – Senior Director of Product Applications, Cell Microsystems, Inc. Abstract: Organoids are valuable multi-cellular 3D tissues that are gaining traction in research and drug discovery pipelines due to their ability to accurately model the cellular structure, complexity, and pathophysiology of in vivo organs. However, traditional culture methods of organoids present challenges in imaging and analysis of heterogenous organoid populations, including multifocal imaging requirements due to overlapping structures and unreliable imaging over time, which makes monitoring phenotypic changes of individual structures difficult. The 3D CytoSort® Array and CellRaft AIR® System are uniquely suited to address these challenges and provide a one instrument solution for imaging, analysis, and retrieval of single, intact organoids for downstream applications. The 3D CytoSort Array is a custom tissue culture dish comprised of thousands of encoded microwells, each containing an optically transparent, releasable microscale growth surface (CellRaft®), which enables the culture of hundreds of spatially segregated organoids on a single array. The CellRaft AIR System performs rapid, automated full array scans of organoids grown on the CytoSort Array in brightfield and three-channel fluorescence. Serial imaged scans are stored in a central database, allowing for a complete growth record of every organoid. We have developed sensitive brightfield algorithms that can detect both single cells and organoids as well as advanced organoid analysis tools for identifying phenotypic and morphologic characteristics such as organoid diameter, circularity, dimensionality, and fluorescence intensity. These tools allow users to build populations and select organoids based on desired size, morphology, viability, and fluorescent marker expression. CellRafts containing organoids of interest can be selected for z-stack acquisition for visualization through the full height of the 3D structure. Using live cell staining protocols for cell surface markers, we have imaged through the full height of organoids up to 600 microns in diameter without fixation, clearing steps, or dissociation. In addition to single organoid imaging and analysis elements, the CellRaft AIR System can isolate single, viable organoids for downstream applications after phenotypic characterization- a feature that is unique to this platform. We have demonstrated the utility of the technology in several applications, including clonal organoid development, drug toxicity screening, and iPSC-derived organoids. Using the 3D CytoSort Array, iPSCs were seeded directly onto the array in dilute ECM. Shortly after cell seeding, an initial scan of the array was performed to identify CellRafts containing single or small clusters of iPSCs. By serially scanning the array over 12-30 days, we monitored and assessed phenotypic changes of hundreds of individual organoids throughout the process of tissue-specific differentiation, including choroid plexus and kidney. At desired timepoints throughout the differentiation process, organoids can be isolated for further characterization or endpoints such as tissue-specific drug screening. The three key advantages of the CellRaft technology – the ability to image, identify, and isolate organoids using a single platform – enables more phenotypic evaluation organoids and provides a user-friendly solution to single organoid workflows that are not possible using other technologies.
VirCAD©: an HPC-powered AI/ML BioCAD Platform for Viral Bioengineering
Open to view video.
Open to view video. Current gene therapies are mostly limited to plasmid-based and Adeno-associated virus variants with inefficient response rates and limited use. Better viral delivery methods would expand the available pharmacopeia to produce precision medicines for intractable and incurable genetic diseases. We are building VirCAD© (Virus Computer-Aided Design), an HPC-powered, cloud-accessible, AI/ML-capable bioCAD platform to mine, design, model and test new viral drugs for improved cell and gene therapies, vaccines, oncolytics, antibiotics and other categories. These modalities will treat the many inherited and acquired genetic disorders afflicting patients due to their therapeutic specificity, efficiency, and customizability.
Translational Models
Development of new dyes to probe cell physiology of 3D cultures in high content screening
Open to view video.
Open to view video. High-content imaging approaches, in combination with the use of perturbing agents such as small molecules or CRISPR-driven gene edition, have helped identify new therapeutic compounds [1]. Together with recent advances in image-analysis methods [2], the use of high-content screens is accelerating and therefore making it easier to identify such compounds. However, large-scale high-content screens of live cells have yet to be widely adopted due to the lack of suitable probes compatible with cell growth, as well as limitations of the number of probes that can be used to identify important cellular structures and cell behaviours. This gap is especially evident with 3D structures. Here we present novel fluorescent non-toxic dyes that display color and pattern changes in response to different physiological states and cellular subtypes. Their unique emission spectra pattern also allows for generation of ratiometric images that can label and identify different cellular components with usage of one single dye. The dyes are non-cytotoxic and do not affect proliferating cells. This advantage makes them uniquely ideal to stain 3D cultures to track growth or death, and compatible with large-scale screens. The dyes have helped provide unparalleled insight into cells cultured in 2D as well as in 3D, and in as diverse as patient derived cancer organoids and stem cell derived hair follicles. The rich information they provide facilitates unbiased quantitative phenotypic analysis at larger scale, and ultimately pave the way for more discoveries of new therapeutic agents. [1] Oppermann, S., Ylanko, J., Shi, Y., Hariharan, S., Oakes, C. C., Brauer, P. M., ... & Andrews, D. W. (2016). High-content screening identifies kinase inhibitors that overcome venetoclax resistance in activated CLL cells. Blood, The Journal of the American Society of Hematology, 128(7), 934-947. [2] Mergenthaler, P., Hariharan, S., Pemberton, J. M., Lourenco, C., Penn, L. Z., & Andrews, D. W. (2021). Rapid 3D phenotypic analysis of neurons and organoids using data-driven cell segmentation-free machine learning. PLoS computational biology, 17(2), e1008630.
Human NASH model Development in LiverChips
Open to view video.
Open to view video. We have developed several tailored multicellular primary human hepatocyte and non-parenchymal cell models in a 3D scaffold under continuous perfusion (LiverChips), which can be utilized to investigate mechanisms of NALFD and NASH. Using TGFb to promote fibrosis in LiverChips, we have demonstrated increased matrix production by mRNA expression, soluble pro-collagen1 production, and collagen deposition assessed by confocal microscopy. Furthermore, TGFb diminished hepatocyte health and induces an inflammatory phenotype. Using a RNA based cell-type deconvolution analysis, we observed a dramatic reduction in the proportion of hepatocytes, with a corresponding increase in activated stellate cells and endothelial cells. Concurrent treatment with an ALK5 inhibitor prevented hepatocellular injury and inflammation and reduced pro-collagen1 production. To simulate NAFLD and NASH conditions, LiverChips were treated with oleate (OA) and palmitate (PA) for up to 14 days. NGS analysis revealed both time and treatment related gene expression changes. The most dramatic responses to OA:PA were observed after 14days treatment and included reduced metabolism with enhanced fibrosis and EMT related hallmarks. An in silico treatment analysis was performed and showed strong recapitulation of NASH F4 human disease with respect to extracellular matrix organization, chemokine receptors, sphingolipid metabolism, and limited recapitulation of immune system components. Overall, our LiverChip models represent a promising complex in vitro system for studying the pathomechanisms of NAFLD and NASH. They have high potential to support the functional evaluation of small molecules, siRNAs, and antibodies to support novel therapeutic concepts.
Keynote
Organoids to model human diseases
Open to view video.
Open to view video.