Gastric patient-derived organoids find increasing use in research, yet formal protocols for generating human gastric organoids from single-cell digests with standardized seeding density are lacking. This protocol presents a detailed method for reliably creating gastric organoids from biopsy tissue obtained during upper endoscopy.
Gastric patient-derived organoids (PDOs) offer a unique tool for studying gastric biology and pathology. Consequently, these PDOs find increasing use in a wide array of research applications. However, a shortage of published approaches exists for producing gastric PDOs from single-cell digests while maintaining a standardized initial cell seeding density. In this protocol, the emphasis is on the initiation of gastric organoids from isolated single cells and the provision of a method for passaging organoids through fragmentation. Importantly, the protocol demonstrates that a standardized approach to the initial cell seeding density consistently yields gastric organoids from benign biopsy tissue and allows for standardized quantification of organoid growth. Finally, evidence supports the novel observation that gastric PDOs display varying rates of formation and growth based on whether the organoids originate from biopsies of the body or antral regions of the stomach. Specifically, it is revealed that the use of antral biopsy tissue for organoid initiation results in a greater number of organoids formed and more rapid organoid growth over a 20-day period when compared to organoids generated from biopsies of the gastric body. The protocol described herein offers investigators a timely and reproducible method for successfully generating and working with gastric PDOs.
Organoids are miniature three-dimensional (3D) cellular structures that resemble the architecture and functionality of the organs from which they were derived1,2. These lab-grown models are created by cultivating stem cells or tissue-specific cells in a controlled environment that allows these cells to self-organize and differentiate into various cell types1,2,3. One of the key advantages of organoids is their ability to recapitulate human biology more closely than traditional two-dimensional (2D) cell cultures1,2,3. In particular, human organoids have been shown to maintain the genetic diversity of their tissue of origin3,4,5. Organoids offer a unique opportunity to study human organ development, model diseases, and test potential therapeutics in a controlled laboratory setting. Furthermore, organoids can be derived from individual patient samples, enabling personalized medicine approaches and the potential development of individualized treatments3,6,7.
Researchers have used human gastric organoids to investigate various aspects of gastric biology and pathology. Prominent examples include the use of patient-derived organoids (PDOs) to predict gastric cancer chemotherapy responses8,9,10 and model the epithelial response to Helicobacter pylori infection11,12,13. Human gastric organoids consist of various cell types found in the stomach, including neck cells, pit cells, and other supporting cells11,14. Gastric organoids can either be generated from induced pluripotent stem cells (iPSCs) or stem cells directly isolated from gastric tissue obtained via biopsies or from gastric resection specimens11,14. The isolation of gastric stem cells from gastric tissue is commonly done by isolating and culturing gastric glands or enzymatically digesting tissue samples to liberate single cells9,13,15. Importantly, the differentiation of cells within gastric organoids generated using either of these techniques has been shown to be similar13. The protocol described herein focuses on a single-cell digest.
Organoids represent a scientific innovation that bridges the gap between traditional cell culture and whole organs. As research in the field continues to progress, organoids are poised to contribute to the development of more effective treatments and therapies for a wide range of applications. Given the rising utilization of gastric PDOs, there is a timely need for a standardized approach to their generation. Here, the protocol for generating human gastric PDOs from single cells isolated from benign gastric biopsy tissue acquired during upper endoscopy is described. Importantly and uniquely, a standardized number of single cells is determined for seeding to reliably generate gastric PDOs and allow subsequent characterization. Using this technique, reliable differences in the formation and growth of organoids generated from biopsies of either the gastric body or gastric antrum are demonstrated.
All human tissue utilized in this protocol was collected from individuals who provided informed consent for tissue collection through a gastric tissue collection study approved by the University of Pennsylvania Institutional Review Board (IRB #842961). Participants in this study were required to undergo an upper endoscopy as part of their routine care, be at least 18 years old, and be able to provide informed consent. All research conducted adhered to the guidelines set forth by the University of Pennsylvania.
1. Experimental preparation
2. Isolating single cells from the biopsy tissue
3. Embedding single cells in a basement membrane matrix "dome"
4. Routine passaging of organoids via fragmentation
The subsequent representative results are derived from biopsies taken from the benign epithelium of both the gastric body and gastric antrum regions of the stomachs of five different patients undergoing upper endoscopy. Two to four "domes"/wells were plated and analyzed per patient for both gastric body and antrum biopsies. Organoids were successfully generated from the gastric body and gastric antrum biopsy tissue from all five patients. On average, 41 organoids were analyzed per "dome"/well. All images are z-projections acquired using a confocal microscope, and the quantification of organoid size and sphericity was performed using commercially available image analysis software (see Table of Materials).
Organoids are generally identifiable within 10 days post-seeding of single cells (Figure 3A). By day 20, the organoids are large and typically need to be passaged. While the number of organoids that form post single-cell seeding can be somewhat variable, expected results for the number of organoids formed from body and antral gastric biopsies are shown in Figure 3B. The number of body and antral organoids peaks at day 10 post-seeding. While not significant, the total number of organoids begins to decrease from day 10 to day 15 and day 15 to day 20. There appears to be a subpopulation of small organoids that form by day 10 and then cease growth and die off over the following 10 days, which would explain this trend. Importantly, it is shown that the number of organoids formed from antral biopsy tissue is significantly higher than biopsy tissue from the body at days 10, 15, and 20. The number of antral organoids formed was on average 2-fold higher than the number of organoids formed from the body.
In Figure 3C, representative results for organoid growth after single-cell seeding between day 10 and day 20 are shown. While organoids from both antral and body biopsies saw steady growth from day 10 to 20, organoids generated from antral biopsy tissue displayed a greater growth rate compared to organoids generated from body biopsy tissue. In particular, antral organoids had nearly a 4-fold greater area than body organoids at day 20.
Across different patients, a diversity of organoid morphology is typically observed in any single "dome"/well (Figure 4A). Some organoids are more round or spherical while others displayed a more irregular morphology. However, on average, sphericity, a measurement of how spherical an organoid is (where a score of 1 = a perfect sphere)18, showed little variation within and between organoids generated from body or antral biopsy tissue (Figure 4B). Therefore, although there are differing growth rates, there are typically no significant morphological differences between organoids generated from biopsy tissue of the gastric body or antrum.
By day 20 post-initiation, gastric organoids are typically ready to be passaged. A large organoid size (≥1500 µm in diameter) or a darkened interior (suggestive of extensive cellular turnover) of the organoid are key signs that organoids need to be passaged (Figure 5A). Organoids left to go beyond this point may begin to break down into 2D monolayers (Figure 5B) that do not reliably re-form organoids after passaging, perhaps indicating a loss of viability or stemness. After passaging and reseeding gastric organoids using the fragmentation protocol described herein, the "domes" will contain many organoid fragments (Figure 5C) that will reorganize themselves into many more organoids and grow much more quickly compared to the initial seeding of single cells (Figure 5D). If organoid growth still needs characterization and/or standardization at the time of passaging, gastric organoids can instead be digested to single cells, as previously described19.
Figure 1: Generation of gastric patient-derived organoids. Schematic overview depicting the process of generating gastric patient-derived organoids from biopsies of benign gastric epithelium. Please click here to view a larger version of this figure.
Figure 2: Passaging of gastric patient-derived organoids. Schematic overview illustrating the passaging of gastric patient-derived organoids through fragmentation. Please click here to view a larger version of this figure.
Figure 3: Gastric patient-derived organoid formation and growth. (A) Representative z-projection images displaying the growth of gastric organoids at day 10, 15, and 20 post-single-cell seeding. Images are of organoids generated from gastric body and antrum biopsy tissue from the same patient. Scale bar = 1 mm. (B) Mean (±SD) number of gastric body and antrum organoids at indicated timepoints post-single-cell seeding. (C) Mean (±SD) area (µm2) of gastric body and antrum organoids at indicated timepoints post-single-cell seeding. n = 5 patients per group and timepoint. * = statistically significant difference (p ≤ 0.05) at the indicated timepoint. n.s. = no statistically significant difference at the indicated timepoint. Statistical comparisons conducted via 2-way ANOVA. Please click here to view a larger version of this figure.
Figure 4: Gastric patient-derived organoid morphology. (A) Representative z-projection images of different gastric organoid morphologies from an individual "dome"/well of gastric body-derived organoids at day 15 post-single-cell seeding. Scale bar = 200 µm. (B) Mean (±SD) gastric body and antrum organoid sphericity (where a value of 1 = a perfect sphere). n = 5 patients per group and timepoint. n.s. = no statistically significant difference at any timepoint. Statistical comparisons conducted via 2-way ANOVA. Please click here to view a larger version of this figure.
Figure 5: Gastric patient-derived organoid passaging. (A) Representative image of organoids ready to be passaged at day 20 post-single-cell seeding. (B) Representative image of organoids overdue for passaging at day 25 post-single-cell seeding. (C) Representative image of fragmented organoids after reseeding (Passage 1 – Day 0). (D) Representative image of organoid growth over 5 days after reseeding (Passage 1 – Day 5). All images are z-projections. Scale bars = 1 mm. Please click here to view a larger version of this figure.
Supplementary Table 1: Solutions and media recipes. Please click here to download this File.
Herein, a detailed protocol for reliably generating human gastric organoids from single cells isolated from biopsies of benign epithelium from the gastric body and antrum is outlined. Critical steps in the protocol revolve around timing as well as handling the basement membrane matrix. To preserve viability, it is essential to initiate the protocol as soon as possible after acquiring the biopsy tissue. The aim is to start digesting the biopsy tissue within 30 min of the biopsy being performed. Handling the basement membrane matrix can also be challenging. When thawed on ice, it remains a liquid; however, at temperatures above 4 °C, it polymerizes. Therefore, swiftly transferring the basement membrane matrix from ice to mix with single cells or organoid fragments for plating "domes" must be done promptly, as it begins to polymerize within one minute. Once it has polymerized in a tube, it cannot be aspirated into a pipette tip. If this occurs, the tube can be placed on ice until the Matrigel depolymerizes back into a liquid. When gently pipetting up and down to mix single cells or organoid fragments with the basement membrane matrix, it is also crucial to avoid creating bubbles. While bubbles in a basement membrane matrix "dome" do not seem to hinder organoid formation and growth, they can obstruct visualization. Additionally, after aliquoting the basement membrane matrix/cell mixture into the well(s) of a cell culture plate, the plate must be inverted and placed into an incubator. This step is critical to allow the basement membrane matrix to polymerize into a 3D "dome" shape and prevent the single cells or organoid fragments from sinking to the bottom of the plate.
Using this protocol, gastric organoids can be identified within 10 days of single cell seeding. Experience shows that very few new gastric organoids form beyond day 10, and, in fact, the overall number of organoids may slightly decrease from day 10-20. This is true for organoids generated from both the gastric body and antrum. However, the total number of organoids formed from gastric antral biopsies is significantly higher than from gastric body biopsies. Furthermore, the growth of antral organoids greatly surpasses body organoids between days 10-20 after single cell seeding. This difference may be attributed to variations in Wnt sensitivity. A recent study demonstrated that gastric body PDOs exhibit better growth with lower Wnt activation, whereas gastric antrum PDOs thrive with higher Wnt activation20. The location of gastric biopsies used to generate gastric PDOs is not typically specified in the literature. Such differences should be considered in future studies utilizing gastric PDOs generated from gastric biopsies of different stomach areas.
Another crucial aspect of this protocol is the establishment of a standardized number of single cells to seed per "dome"/well for reliably generating gastric PDOs. While a previous study reported a standardized number of gastric glands to seed for PDO generation, no previous studies using a single cell digest method have mentioned the number of cells seeded or if the number was consistent across different PDO lines. Failing to standardize the number of cells seeded can result in a highly variable number of stem cells being seeded per "dome"/well. Since stem cells are the primary source of gastric organoid formation, this could lead to variable rates of organoid formation and growth. Therefore, utilizing a non-standardized number of single cells could confound interpretations of formation or growth comparisons across different gastric PDO lines. Here, it is demonstrated that standardizing the number of cells seeded to 105 cells per "dome"/well reliably generates gastric PDOs from biopsies of both the body and antral regions of the stomach.
This protocol was optimized for the use of fresh gastric biopsy tissue. Consequently, the success of this protocol may vary when using frozen tissue or tissue preserved by other means. Additionally, there is no data on how long fresh biopsies maintain viability, as the biopsy tissue is processed as soon as possible after removal from a patient's stomach. Presumably, the longer the fresh tissue sits before processing, the fewer viable cells will be isolated.
Passaging gastric PDOs via fragmentation, as described in this protocol, provides an easy method for routine passaging. While gastric PDOs have been successfully passaged up to 4 times using this technique, there have been no attempts to determine how many times they can be passaged while maintaining steady growth and viability. Some reports indicate that the growth of gastric organoids may slow after 5 or more passages21,22, while others have observed reliable growth for up to 10 passages23.
The gastric organoid media used in this protocol contains Wnt-3A, noggin, and R-spondin derived from conditioned media of L-WRN cells. To produce the conditioned media, a protocol described by Miyoshi and Stappenbeck is utilized16. Alternatively, Wnt-3A, noggin, and R-spondin can be purchased separately as recombinant proteins. Purchasing the individual components separately is ideal to avoid potential batch effects that may arise when using conditioned media from L-WRN cells. However, buying the recombinant proteins is expensive and may be cost-prohibitive for researchers who frequently work with organoids.
Given the growing use of gastric PDOs in various applications, it is timely and necessary to establish standardized approaches for generating benign gastric PDOs. The protocol described here offers a reliable method for future investigations utilizing gastric PDOs. In our experience, this protocol has successfully generated organoids from biopsies of benign gastric mucosa over 90% of the time.
The authors have nothing to disclose.
University of Pennsylvania Genomic Medicine T32 HG009495 (KHB), NCI R21 CA267949 (BWK), Men & BRCA Program at the Basser Center for BRCA (KHB, BWK), DeGregorio Family Foundation Grant Award (BWK).
0.25% Trypsin-EDTA | Gibco | 25200-056 | |
A83-01 | R&D Systems | 2939 | |
Advanced DMEM/F12 | Gibco | 12634-010 | |
Amphotericin B | Invitrogen | 15290018 | |
B27 | Invitrogen | 17504044 | |
BZ-X710 | Keyence | n/a | |
cellSens | Olympus | n/a | |
Collagenase III | Worthington | LS004182 | |
Dispase II | Sigma | D4693-1G | |
Dithiothreitol (DTT) | EMSCO/Fisher | BP1725 | |
DPBS | Gibco | 14200-075 | |
Fungin | InvivoGen | NC9326704 | |
Gastrin I | Sigma Aldrich | G9145 | |
Gentamicin | Invitrogen | 1570060 | |
Glutamax | Gibco | 35050-061 | |
hEGF | Peprotech | AF-100-15 | |
HEPES | Invitrogen | 15630080 | |
hFGF-10 | Peprotech | 100-26 | |
L-WRN Cell Line | ATCC | CRL-3276 | |
Matrigel | Corning | 47743-715 | |
Metronidazole | MP Biomedicals | 155710 | |
N2 Supplement | Invitrogen | 17502048 | |
Noggin ELISA Kit | Novus Biologicals | NBP2-80296 | |
Pen Strep | Gibco | 15140-122 | |
RPMI 1640 | Gibco | 11875-085 | |
R-Spondin ELISA Kit | R&D Systems | DY4120-05 | |
Wnt-3a ELISA Kit | R&D Systems | DY1324B-05 | |
Y-27632 | Sigma Aldrich | Y0503 |