The protocol combines human intestinal organoid technology with single cell transcriptomic analysis to provide significant insight into previously unexplored intestinal biology.
Single cell transcriptomics has revolutionized our understanding of the cell biology of the human body. State-of-the-art human small intestinal organoid cultures provide ex vivo model systems that bridge the gap between animal models and clinical studies. The application of single cell transcriptomics to human intestinal organoid (HIO) models is revealing previously unrecognized cell biology, biochemistry, and physiology of the GI tract. The advanced single cell transcriptomics platforms use microfluidic partitioning and barcoding to generate cDNA libraries. These barcoded cDNAs can be easily sequenced by next generation sequencing platforms and used by various visualization tools to generate maps. Here, we describe methods to culture and differentiate human small intestinal HIOs in different formats and procedures for isolating viable cells from these formats that are suitable for use in single-cell transcriptional profiling platforms. These protocols and procedures facilitate the use of small intestinal HIOs to obtain an increased understanding of the cellular response of human intestinal epithelium at the transcriptional level in the context of a variety of different environments.
The small intestinal epithelium has two distinct zones: the crypt that houses the intestinal stem cell (ISC) and the villus, which is comprised of differentiated cells of the secretory and absorptive lineages. Adding to this complexity is the regional specificity of the epithelium that provides unique functional properties between the regions of the small intestine. Pioneering work established culture conditions in which both the human small intestinal crypt and villus zones can be generated ex vivo from surgical tissues or tissue biopsies1. These cultures are bridging the gap between animal studies and clinical trials and are revealing previously unrecognized cell biology, biochemistry, and physiology of the gastrointestinal tract. The HIOs are propagated as 3D spherical structures using a media with growth factors that promote stem cell viability and extracellular support matrices. These conditions result in a crypt-like HIO model that consists mostly of progenitors and stem cells. Removal of the growth factors promotes ISC differentiation (villus-like model) and production of mature intestinal epithelial cells (goblet, enteroendocrine, tuft, enterocyte) in appropriate ratios along with cell proliferation and differentiation, polarization, barrier integrity, regional specific features, and appropriate physiological responses2. HIO cultures are genetically stable and can be propagated indefinitely in their crypt-like state. Easy access to the apical surface of these cultures is provided by culture conditions that allow growth in a monolayer format3. HIOs also allow considerations of host individual variability such as genetics, age, sex, ethnicity, and disease status to be included in biological analyses. Analytic and functional assessment tools are identical to those used in approaches centered on transformed cell lines and include a variety of molecular techniques such as flow cytometry, microscopy, transcriptomics, proteomics, and metabolomics.
Single cell transcriptomics is revolutionizing our understanding of the biology and physiology of the small intestine by providing insight into the individual contributions of each cell type to a biological process. Pioneering work using this technology has provided a landscape of the cell types present in the native human intestine4,5,6. Single cell transcriptional profiling platforms allow exploration of the transcriptional landscape of individual cells, allowing cell heterogeneity to come to the forefront of scientific exploration. In some single cell transcriptional profiling platforms, microfluidic partitioning and barcoding are used to generate cDNA libraries from cellular polyadenylated mRNAs obtained from up to 10,000 cells per sample. On this platform, droplets containing single cells, barcoded oligonucleotides, reverse transcription reagents, and oil form a reaction vesicle that results in all cDNAs from a single cell having the same barcode. The barcoded cDNAs can then be efficiently sequenced using next generation sequencing. Data generated can be handled through software and visualization tools, which convert the barcoded sequences into visualization maps and single cell transcriptional profiles. Cell populations can be identified using publicly available databases of human small intestinal epithelium4,5,6. Although many studies have utilized this platform to interrogate murine intestinal organoids at the single cell level, the analysis of single cell transcriptional responses of HIOs has lagged behind7,8,9,10,11,12.
Here, we provide a step-by-step guide to isolating viable cells from small intestinal HIOs for processing on single cell transcriptional profiling platforms (Figure 1). We provide culturing and differentiation guidelines along with media components that have been optimized for epithelial differentiation. We outline cell recovery methods for three different culture formats: three dimensional (3D) and monolayer cultures either on plastic or on membrane cell culture inserts. We provide sample clustering data obtained using open-source software to derive differentially expressed genes in each cluster.
The organoid lines used here were obtained from the Texas Medical Center Digestive Disease Center GEMS Core. Briefly, to initially establish organoid lines, donor tissue samples were washed and enzymatically digested to release the intestinal crypts. Crypts were embedded in a basement membrane and cultured in a medium. The Institutional Review Board at Baylor College of Medicine approved the study protocol to obtain tissue samples from which organoid lines were established, and informed consent was obtained from all donors to establish organoid lines from the donated tissue.
1. Passaging of 3D HIOs to expand for differentiation
2. HIO differentiation: 3D format
3. HIO differentiation: Monolayer format
4. Preparation of single cell suspensions from differentiated 3D HIOs for single cell transcriptomics
5. Preparation of single cell suspensions from HIOs differentiated on membrane cell culture insert
6. Preparation of single cell suspensions from HIOs differentiated as 96-well monolayers
Single-cell suspensions were pooled from 2-3 wells of membrane cell culture insert, monolayer, and 3D HIOs to ensure sufficient cell yield and reduce well-to-well variation. Single cell libraries were prepared using reagents specific to the single cell transcriptional profiling platform. and sequenced with paired end reads on a next generation sequencing platform, 30,000 reads/cell. Reads were mapped, counted, and analyzed using analytical tools for single cell genomics. Low-quality cells with more than 20% mitochondrial read, or less than 200 counted genes were excluded from the analysis. To assess the quality of the cells, we plotted the number of genes, UMIs, and percent mitochondrial reads per cell (Figure 3); high-quality cells contain approximately 3500 genes, 15,000 UMIs, and 10% mitochondrial reads. Following filtering, we obtained 4000 to 9500 single cells depending on the HIO growth format. Unsupervised clustering and UMAP dimension reduction identified distinct clusters representing expected cell types, including absorptive enterocytes and secretory cells (Figure 4).
Figure 1: Outline of protocol steps. Schematics outlining the major steps in plating HIOs in different formats to single cell data workflow. This image was created using Biorender. Please click here to view a larger version of this figure.
Figure 2: Bright field images. HIOs ready for processing were imaged using a bright field microscope at 5X magnification. Left, 3D organoids; middle, 96-well plate; right, transwell. Scale bars = 200 µm. Please click here to view a larger version of this figure.
Figure 3: Quality control of data. Example metrics of analyzed single cells isolated from jejunal HIOs grown as monolayers on 96-well plates. Violin plots show the number of genes (nFeature_RNA), UMIs (nCount_RNA), and percent mitochondrial reads (percent.mt) per cell. Please click here to view a larger version of this figure.
Figure 4: UMAPs of analyzed single cells. (A) A total of 9952 single cells were isolated from differentiated jejunal HIOs grown in 96-well plates. (B) A total of 5623 single cells were isolated from differentiated jejunal HIOs grown on membrane cell culture inserts. (C) A total of 4402 single cells were isolated from differentiated ileal 3D HIOs. Please click here to view a larger version of this figure.
WRNE Growth Medium | Differentiation medium | ||
Component | Volume | Final Concentration | WRNE medium without L-WRN conditioned medium, Nicotinamide and SB202190 |
*Y-27632 (5 mM) optional | 200 µL | 10 µM | |
A 83-01 (500 µM) | 100 µL | 500 nM | |
Advanced DMEM/F12 | 45 mL | ||
B27 | 2 mL | ||
EGF (50 µg/mL) | 100 µL | 50 ng/mL | |
Gastrin (10 µM) | 100 µL | 10 nM | |
GlutaMax (100X) | 1 mL | ||
HEPES 1 M | 1 mL | ||
L-WRN Con- Media | 50 mL | ~50% | |
N2 | 1 mL | ||
N-Acetylcysteine (500 mM) | 100 µL | 500 µM | |
Nicotinamide 1 M | 1 mL | 10 mM | |
SB202190 (10 mM) | 100 µL | 10 µM | |
Total Volume | 100 mL | ||
CMGF- Medium | |||
Advanced DMEM/F12 | 500 mL | ||
Glutamax | 5 mL | 100X | |
1 M Hepes | 5 mL |
Table 1: Composition of various media used.
Using single cell genomics platforms, complex biological systems, such as tissue derived HIO cultures that model the intestinal epithelium, can be broken down to yield individual cellular contributions to overall biological response4,5,6. Cellular heterogeneity and rare cell populations can also be identified and interrogated. Cellular input needs to be optimized to maximize output using single cell transcriptomic-based platforms. Here we describe protocols that will yield highly viable and good quality single cell suspensions that will increase the likelihood of obtaining downstream high quality transcriptomic data. Familiarity with the protocol is of utmost importance, and it is recommended that several practice runs be performed before proceeding with library preparation. Reproducible achievement of viable single cell suspensions is suggested prior to using the cells for library preparation. The rapidness at which the protocol is implemented is of utmost importance as speed is directly proportional to viability outcomes. Based on this, limiting the number of samples handled at one time is advised. Interfacing with the entity that will prepare the libraries prior to initiating these protocols for assistance with troubleshooting is recommended.
HIO viability directly influences the DNA library sent for sequencing. Cell viability of the HIO depends on several steps in the preparation of the single cell suspension, including dissociation, centrifugation, buffers, pipetting, and counting accuracy. The goal of preparing the HIOs for the platform is to balance the maintenance of cell viability with sample quality. Critical to obtaining high quality data is the minimization of cellular aggregates, dead cells, non-cellular nucleic acids, and biochemical inhibitors of reverse transcription. Dying cells increase nucleic acid noise and contribute to cell clumping, leading to data and platform failures7,8,9,10,11,12. Filtering with an appropriate filter size can help remove large aggregates and improve data quality.
Several factors need to be considered to minimize HIO cell death when preparing single cell suspensions. Obtaining at least 80% viable cells is necessary, and >90% viability is even more desirable. Samples with high numbers of non-viable cells can be improved using strategies to remove dead cells, such as column-based microbeads, although these require more initial cells and larger sample volumes due to loss during handling. HIOs are particularly sensitive to buffers and will undergo cell death if left too long at RT in certain buffers. Organoid media results in the least amount of cell loss and aggregate formation. Pipetting can also affect HIO viability during dissociation. Regular bore tips can shear organoids, helping to dissociate into single cell suspensions; however, pipetting too rapidly with a regular bore tip can cause HIO cell damage from the shearing forces. Slow pipetting with wide bore tips minimizes stress on the HIO cells and greatly improves viable cell numbers downstream. Centrifugation, washing, and resuspending are other factors that can influence the quality of HIO single cells suspensions. Swinging bucket centrifugation results in gentler pelleting of HIO cells, resulting in lower numbers of dead cells while minimizing cell loss. Finally, counting accuracy to determine concentration is essential to ensure the maximum number of cells reach the platform for library synthesis. Attention to these details will result in the minimization of stress on the HIO cells during the isolation process and ensure the highest probability of data return after processing.
Although single cell transcriptomics is a powerful tool to dissect biological profiles and responses at the single cell level4,5,6, there are several limitations that need to be considered when applying the technology to analyze human intestinal organoid cultures. First, there can be significant transcriptional variation because of technical variability introduced during library preparation and sequencing or due to batch effects resulting from sample preparation. In addition, sequencing can result in coverage bias where there is uneven sequencing depth across genes or cells. This causes challenges when detecting lowly expressed genes or identifying rare cell populations. Second, the identification of cell types can be difficult due to the cellular and transcriptional heterogeneity of the organoid cultures and a less differentiated cellular profile when compared to the native intestinal epithelium. Coupling transcriptional responses to functional phenotypes is also a challenge and requires additional validation using other approaches or assays. Third, this approach does not include consideration of spatial context or temporal dynamics of the cells. The dissociation procedure removes the ability to infer cell-cell interactions from the transcriptional data. It only provides a single snapshot in time of the cellular state rather than a biologic continuum. Finally, single cell transcriptional approaches can be costly and require specialized computational tools and bioinformatics expertise. However, this approach still remains a dominant state-of-the-art tool for understanding cellular heterogeneity and identifying novel cell types.
The authors have nothing to disclose.
The authors acknowledge U19 AI157984, U01 DK103168, U19 AI144297, P30 DK56338, P01 AI057788, U19 AI116497 grants and NASA Cooperative Agreement Notice/TRISH NNX16AO69A.
[Leu15]-Gastrin I | Sigma-Aldrich | G9145 | 10 nM |
0.05% Trypsin-EDTA | Invitrogen | 25300054 | |
0.4% Trypan blue | Millipore-Sigma | T8154 | |
0.5 M EDTA | Corning | 46-034-CI | |
1x PBS Ca- Mg- | Corning | 21-040-CM | |
24 mm Transwell | Costar | 3412 | |
24 well Nunclon delta surface tissue culture dish | Thermo Scientific | 142475 | |
40 µm cell strainer | Falcon | 352340 | |
40 µm Flowmi tip strainer | SP Bel-Art Labware | H13680-0040 | |
70 µm Flowmi tip strainer | SP Bel-Art Labware | H13680-0070 | |
96 well plate | Corning | 3595 | |
A-83-01 | Tocris | 2939 | 500 nM |
Accutase | StemCell Technologies | 7920 | |
Advanced DMEM/F12 | Invitrogen | 12634-028 | |
B27 supplement | Invitrogen | 17504-044 | 1X |
Chromium Next GEM Single Cell 3’ GEM, Library & Gel Bead Kit v3.1 | 10x Genomics | PN-1000128 | |
Collagen IV | Sigma-Aldrich | C5533-5MG | 33 µg/mL |
Corning Cell Recovery Solution | VWR | 354253 | |
DPBS (Mg2-, Ca2-) | Invitrogen | 14190-136 | 1X |
GlutaMAX-I | Invitrogen | 35050-061 | 2 mM |
HEPES 1M | Invitrogen | 15630-080 | 10 mM |
L-WRN conditioned media | ATCC | CRL-3276 | |
Matrigel, GFR, phenol free | Corning | 356231 | |
mouse recombinant EGF | Invitrogen | PMG8043 | 50 ng/mL |
N2 supplement | Invitrogen | 17502-048 | 1X |
N-Acetylcysteine | Sigma-Aldrich | A9165-5G | 500 µM |
Nicotinamide | Sigma-Aldrich | N0636 | 10 mM |
SB202190 | Sigma-Aldrich | S7067 | 10 µM |
Transwell | Corning | 3413 | |
Y27632 | Stem Cell Technologies | 72308 | 10 µM |
.