This protocol describes rectal organoid morphology analysis (ROMA), a novel diagnostic assay for cystic fibrosis (CF). Morphological characteristics, namely the roundness (circularity index, CI) and the presence of a lumen (intensity ratio, IR), are a measure of CFTR function. Analysis of 189 subjects showed perfect discrimination between CF and non-CF.
Diagnosis of cystic fibrosis (CF) is not always straightforward, especially when sweat chloride concentration is intermediate and/or less than two disease-causing CFTR mutations can be identified. Physiological CFTR assays (nasal potential difference, intestinal current measurement) have been included in the diagnostic algorithm but are not always readily available or feasible (e.g., in infants). Rectal organoids are 3D structures that grow from stem cells isolated from crypts of a rectal biopsy when cultured under specific conditions. Organoids from non-CF subjects have a round shape and a fluid-filled lumen, as CFTR-mediated chloride transport drives water into the lumen. Organoids with defective CFTR function do not swell, retaining an irregular shape and having no visible lumen. Differences in morphology between CF and non-CF organoids are quantified in the 'Rectal Organoid Morphology Analysis' (ROMA) as a novel CFTR physiological assay. For the ROMA assay, organoids are plated in 96-well plates, stained with calcein, and imaged in a confocal microscope. Morphological differences are quantified using two indexes: The circularity index (CI) quantifies the roundness of organoids, and the intensity ratio (IR) is a measure of the presence of a central lumen. Non-CF organoids have a high CI and low IR compared to CF organoids. ROMA indexes perfectly discriminated 167 subjects with CF from 22 subjects without CF, making ROMA an appealing physiological CFTR assay to aid in CF diagnosis. Rectal biopsies can be routinely performed at all ages in most hospitals and tissue can be sent to a central lab for organoid culture and ROMA. In the future, ROMA might also be applied to test the efficacy of CFTR modulators in vitro. The aim of the present report is to fully explain the methods used for ROMA, to allow replication in other labs.
Cystic fibrosis (CF) is an autosomal recessive disease caused by mutations in the CF transmembrane conductance regulator (CFTR) gene. The CFTR protein is a chloride and bicarbonate channel, ensuring hydration of several epithelia1. CF is a high-burden, life-shortening, multi-system disease, manifesting primarily as a respiratory disease, but also affecting the gastrointestinal tract, pancreas, liver, and reproductive tract2.
Disease-causing CFTR mutations lead to a decrease in the amount or function of CFTR, in turn causing mucus dehydration. More than 2,000 variants in the CFTR gene have been described3, of which only 466 have been thoroughly characterized4.
A diagnosis of CF can be made when either sweat chloride concentration (SCC) is above the threshold of 60 mmol/L or when two disease-causing CFTR mutations (according to the CFTR2 database) are identified4,5. In subjects with only intermediately elevated (30-60 mmol/L) SCC, which occurs in about 4%-5% of sweat tests6, and CFTR mutations of varying or unknown clinical consequence, the diagnosis cannot be confirmed nor ruled out, even when they have CF compatible symptoms or a positive neonatal screening test. For these cases, second-line physiological CFTR assays (nasal potential difference (NPD) and intestinal current measurements (ICM)) have been included in the diagnostic algorithm. These tests are not readily available at most centers nor feasible at all ages, especially in infants5.
Rectal organoids are 3D structures grown from Lgr5(+) adult intestinal stem cells from intestinal crypts obtained through rectal biopsy7. Organoids are being used increasingly in biomedical research, such as testing modulator treatment in CF8. A viable biopsy can be obtained by either suction or forceps biopsy, a procedure that causes only minimal discomfort and is safe even in infants, with low complication rates9. The crypts isolated from the rectal biopsies are enriched in stem cells, and under specific culturing conditions, these self-organize into rectal organoids. The morphology of these organoids is determined by the expression and function of CFTR, located at the apical membrane of epithelial cells. Functional CFTR allows chloride and water to enter the organoid lumen, thereby inducing swelling of non-CF organoids. CF organoids do not swell and have no visible lumen10,11.
Rectal organoid morphology analysis (ROMA) allows the discrimination between CF and non-CF organoids based on these differences in organoid morphology. Non-CF organoids are more round and have a visible lumen, while the opposite is true for CF organoids. For this assay, patient-specific organoids are plated in 32 wells of a 96-well plate. After 1 day of growing, the organoids are stained with calcein green and imaged in a confocal microscope. The non-CF organoids show a more circular shape and a less fluorescent central part, as the lumen contains fluid and calcein stains only cells. These differences in morphology are quantified using two ROMA indexes: the circularity index (CI) quantifies the roundness of organoids, while the intensity ratio (IR) is a measure of the presence or absence of a central lumen. In this report, we describe in detail the protocol to obtain these discriminative indexes, to allow replication of the technique.
For all procedures involving human tissue, approval by the Ethics Committee Research UZ/KU Leuven (EC research) was acquired. All research was performed with informed consent and/or assent from parents, representatives, and/or patients.
NOTE: All procedures involving rectal biopsies and organoids should be performed in a laminar flow to protect the researcher from any biological hazard and to minimize the risk of contamination of the cultures. As for any lab procedure, researchers should at all times wear lab coats, gloves, and safety goggles to manipulate samples.
1. Rectal biopsy, isolation of adult stem cells from crypts, and organoid culture
2. Organoid plating for ROMA (day 1)
3. Organoid imaging using confocal microscopy (day 2)
4. Image analysis (Figure 5)
5. Measure the indices in the imaging software (Figure 6)
Organoids from 212 subjects were collected during routine clinical visits. No adverse events occurred during or after the rectal biopsy procedure. Organoids were imaged by one researcher blinded to subject characteristics such as genotype and clinical information. Due to low-quality images, 23 subjects were excluded. Examples of successful and failed organoid cultures and image acquisition can be seen in Figure 2.
Organoids of 167 subjects with CF and two disease-causing CFTR mutations (as defined by the CFTR2 database4) and 22 non-CF subjects were analyzed. The mean amount of organoids per culture was 1,519 (about 40-50 organoids per well). The mean amount of organoids per culture included for analysis was 77% (the number of structures ≥60 µm divided by the number of structures ≥40 µm, corresponding to the fraction of organoids large enough to reflect typical CF or non-CF morphology).
The IR and CI discriminated (p < 0.001) between organoids from subjects with and without CF (Table 1). With linear discriminant analysis, perfect discrimination (AUC = 1) was obtained between CF and non-CF, not only when using data from all 32 wells (Figure 8), but also when eight wells were chosen randomly for each culture (Figure 9).
Figure 10 depicts histograms showing the distribution of values for circularity, the intensity of the central part of the organoid, and the intensity of the whole organoid for four illustrative cultures (two CF, two non-CF).
Figure 1: Images of well-grown and viable organoids. (A) Organoids from a person without CF and (B) organoids from a person with CF. Both cultures were grown for 7 days after the previous splitting. Images were made using a brightfield microscope with a 5x objective. Please click here to view a larger version of this figure.
Figure 2: Illustration of organoid imaging in the confocal microscope. (A) Good quality CF organoids; (B) good quality non-CF organoids; (C) density of plating too low: not enough organoids for representative imaging; (D) density of plating too high: overlapping of organoids prevents adequate calcein staining and assessment of morphology; (E) intensity too low due to either a problem with calcein staining or the master gain setting being too low; (F) intensity too high due to the master gain setting being too high: small lumens may be masked by the overexposed fluorescence signal; (G) dead and burst organoids, where separate cells can be seen and morphology cannot be assessed anymore; (H) differentiated organoids where stem cell status is lost, showing up as thick structures often with high fluorescence signals in the middle of the organoid structures, not reflecting typical CF or non-CF morphology; (I) background signal too high, either due to calcein staining having been performed too long ago with diffusion into the background or due to the master gain setting being too high; (J) insufficient mechanical splitting of organoids, leaving them too big for the assay, not staining well, and not adequately reflecting CF or non-CF morphology. Please click here to view a larger version of this figure.
Figure 3: Focusing on organoids and acquiring pictures. (A) Choose the Acquisition tab. Click on the Live button below for real-time imaging. (B) Use live cell imaging settings with emission at 488 nm. (C) Image at a resolution of 1024 pixels x 1024 pixels and a depth of 16 bits per pixel. Choose the unidirectional imaging parameter. (D) Adjust the master gain for the optimal intensity of fluorescence to optimize imaging of organoid characteristics such as shape and presence or absence of a lumen. (E) Save single positions (x, y, and z) for each of the 32 wells per organoid culture. (F) Click on the Start Experiment button to run the pre-defined protocol and acquire images according to the parameters chosen. (G) Images can be saved upon completion, or an autosave can be set up with the Autosave button. Please click here to view a larger version of this figure.
Figure 4: Exporting images for analysis. (A) Choose the Processing tab, and click on the Batch button to export images of multiple organoid cultures in one procedure. (B) Click on the Add button and select the saved images needed for analysis. (C) Select image export as the method. (D) Export images as TIFF files, do not convert to 8 bits, and do not compress or resize. Export original data without burn-in graphics. Select the 32 wells of the 96-well plate with the scene parameter. Do not re-tile. (E) Click on the Apply button to extract using the chosen parameters. Please click here to view a larger version of this figure.
Figure 5: Analysis of organoid images in the imaging software. (A) Right-click the grey bar above the results section (red asterisk) to set up parameters for object measurement. Add circularity and mean intensity. (B) Click on File > Import/Export > Create ND file from file sequence to combine the TIFF files for one culture into one ND file. (C) Select the desired file and confirm; 32 pictures will be combined (red asterisk). (D) Click Recalibration > Recalibrate Document. (E) Click on Pixel Size in the pop-up window. (F) Input 2.5 µm as the size of 1 pixel. (G) The picture will now be recalibrated to micrometers instead of pixels (red asterisk). (H) Click Binary > Define Threshold. (I) Size selection can be performed in the pop-up window. The minimum size is 40 µm for organoid counting and 60 µm for organoid morphology analysis. Always turn off the Smooth and Clean function, turn on the Fill Holes function, and input Separate x3. Apply to all frames. (J) The software will show delineation of the defined structures. Click on the Update ND Measurement button to analyze and get the chosen parameters from step A as output. (K) Click the downward facing arrow next to the export button and select data to the clipboard. (L) Click on the Export button to copy the data output to the clipboard. Data can now be pasted to a spreadsheet. (M) Click on the Reset Data button to empty the results section before a new measurement is performed. (N) When erosion has to be performed for intensity measurement for calculation of the intensity ratio, click Binary > Erode. The Remove Objects Touching Borders function can be found in the same drop-down menu. (O) Select the matrix shown in the figure for erosion, and choose the desired count (1 pixel or 2.5 μm for removal of the halo surrounding organoids, 10 pixels or 25 μm for removal of the cellular border surrounding a lumen if present). Please see the Supplemental File for full-screen panels of this figure. Please click here to view a larger version of this figure.
Figure 6: Images of rectal organoids from people without (upper panels) and with CF (lower panels). Illustration of the methods to calculate the two indexes, IR (intensity ratio; central panel) and CI (circularity index; right-hand panel), used to quantify morphological differences between rectal organoids of subjects with and without CF. IR measures the presence or absence of a central lumen, calculated in three steps: (I) calculate the global fluorescence intensity of the organoids: erode 1 pixel (2.5 μm) to remove the surrounding 'halo' around each structure, and measure the mean fluorescence intensity of the remaining whole organoid; (II) calculate the central fluorescence intensity of the organoids: erode 10 pixels (25 µm) around each structure to remove the cellular border from the organoids and measure the mean fluorescence intensity of the remaining structure; (III) IR is equal to , and is higher in CF than in non-CF organoids. CI quantifies the roundness of the organoids, defined as , which is lower in CF than in non-CF organoids. CF: cystic fibrosis; IR: intensity ratio; CI: circularity index. This figure has been reprinted with permission from Cuyx et al.13. Please click here to view a larger version of this figure.
Figure 7: Example of a spreadsheet for calculation of CI and IR. The output (circularity and mean intensity, as defined in Figure 5) is copied into the spreadsheet for both erosion steps. The imaging software automatically adds the mean values for each parameter. These means can be copied into a cell of choice in the spreadsheet and then used for the calculation of CI and IR. Please click here to view a larger version of this figure.
Figure 8: Intensity ratio (IR) and circularity index (CI) values of each subject according to disease status, pancreatic status, and sweat chloride concentration. The line represents the optimal discrimination line obtained by linear discriminant analysis. CF: cystic fibrosis; PS: pancreatic sufficient; PI: pancreatic insufficient; SCC: sweat chloride concentration. This figure has been reprinted with permission from Cuyx et al.13. Please click here to view a larger version of this figure.
Figure 9: Calculation of ROMA indexes. Calculation of ROMA indexes using eight wells at random per subject and using the same statistical methodology was comparable to the results using 32 wells. Again, perfect discrimination was obtained. CI: circularity index; IR: intensity ratio. This figure has been reprinted with permission from Cuyx et al.13. Please click here to view a larger version of this figure.
Figure 10: Histograms illustrating the distribution of values measured in each single organoid in a given culture. The circularity parameter, the intensity parameter for the central part of the organoid, and the intensity parameter for the whole organoid are depicted (columns). Results in two CF and two non-CF cultures are shown (rows). Please click here to view a larger version of this figure.
Supplemental File: Full-resolution presentation of Figure 5. Please click here to download.
CF | Non-CF | p-value | |
n | 167 | 22* | |
IR | 1.11 (0.93–1.34) | 0.76 (0.61–0.88) | <0.001 |
CI | 0.59 (0.49–0.70) | 0.79 (0.73–0.84) | <0.001 |
Age (years) | 18 (0–60) | 44 (0–77) | <0.001 |
Gender | 85 male (51%) 82 female (49%) |
11 male (50%) 11 female (50%) |
>0.999 |
SCC (mmol/l) (n = 164) | 97.61 (36–160) | ||
SCC low (<87 mmol/L) or high (≥87 mmol/L) | 41 low (25%) 123 high (75%) |
||
Pancreatic status (n = 165) | 28 PS (17%) 137 PI (83%) |
Table 1: Baseline characteristics of the subjects and indexes calculated using rectal organoid morphology analysis (ROMA). n or mean and range. CF: cystic fibrosis; IR: intensity ratio; CI: circularity index; SCC: sweat chloride concentration; PI: pancreatic insufficient; PS: pancreatic sufficient. *seven carriers, three non-carriers, two autosomal dominant polycystic kidney disease, six ulcerative colitis, one polyp screening, three healthy controls included in a study about inflammatory bowel disease. This table has been reprinted with permission from Cuyx et al.13.
We provide a detailed protocol for rectal organoid morphology analysis (ROMA). The two indexes calculated with ROMA, IR, and CI, distinguished organoids from subjects with CF from those without CF with perfect accuracy. ROMA could thus function as a novel physiological CFTR assay complementary to SCC and other currently available tests13,14,15.
The protocol is dependent on the use of intestinal organoids, which have a round shape and central lumen when CFTR is functional, as described before10,11. Organoids are used increasingly in the assessment of novel CF treatments (e.g., in the FIS assay8). The ROMA protocol can be integrated with the FIS assay protocol, as for the FIS assay 32 wells are incubated overnight without correctors. These wells can be imaged after the addition of calcein green but before the addition of forskolin and/or potentiators. This way, one 96-well plate can be used for both diagnostic and personalized therapeutic research for each specific patient. Apart from diagnosis, ROMA might also provide information in the characterization of variants of unknown significance.
The biggest practical hurdle of this protocol would probably be the startup of intestinal organoid cultures. However, in most general hospitals, rectal suction biopsies can be obtained according to a standardized protocol and with low complication rates, even in infants9. Generating organoids from biopsies requires only the presence of intestinal crypts, while for ICM, full-thickness biopsies of higher quality are necessary12,15. Biopsies can be transported to a central lab for generating an organoid culture, and subsequent ROMA using the standardized and semi-automated protocol described herein9,12. As reduction from 32 wells to eight wells for ROMA showed no difference in the classification of cases as CF or non-CF, plating eight wells for analysis would be sufficient, thus reducing the cost.
For further validation, ROMA will have to be performed on subjects with equivocal diagnoses. Organoids can be stored in a biobank for later research into personalized medicine for those diagnosed with CF16. ROMA could play a role in a personalized medicine approach as well. For example, IR could detect the appearance of a central lumen after incubating organoids of subjects with residual CFTR function but before stimulation with a potentiator and forskolin.
The authors have nothing to disclose.
We thank the patients and parents who participated in this study. We thank Abida Bibi for all culturing work with the organoids. We thank Els Aertgeerts, Karolien Bruneel, Claire Collard, Liliane Collignon, Monique Delfosse, Anja Delporte, Nathalie Feyaerts, Cécile Lambremont, Lut Nieuwborg, Nathalie Peeters, Ann Raman, Pim Sansen, Hilde Stevens, Marianne Schulte, Els Van Ransbeeck, Christel Van de Brande, Greet Van den Eynde, Marleen Vanderkerken, Inge Van Dijck, Audrey Wagener, Monika Waskiewicz, and Bernard Wenderickx for logistic support. We also thank the Mucovereniging/Association Muco, and specifically Stefan Joris and Dr. Jan Vanleeuwe, for their support and financing. We thank all collaborators from the Belgian Organoid Project: Hedwige Boboli (CHR Citadelle, Liège, Belgium), Linda Boulanger (University Hospitals Leuven, Belgium), Georges Casimir (HUDERF, Brussels, Belgium), Benedicte De Meyere (University Hospital Ghent, Belgium), Elke De Wachter (University Hospital Brussels, Belgium), Danny De Looze (University Hospital Ghent, Belgium), Isabelle Etienne (CHU Erasme, Brussels, Belgium), Laurence Hanssens (HUDERF, Brussels), Christiane Knoop (CHU Erasme, Brussels, Belgium), Monique Lequesne (University Hospital Antwerp, Belgium), Vicky Nowé (GZA St. Vincentius Hospital Antwerp), Dirk Staessen (GZA St. Vincentius Hospital Antwerp), Stephanie Van Biervliet (University Hospital Ghent, Belgium), Eva Van Braeckel (University Hospital Ghent, Belgium), Kim Van Hoorenbeeck (University Hospital Antwerp, Belgium), Eef Vanderhelst (University Hospital Brussels, Belgium), Stijn Verhulst (University Hospital Antwerp, Belgium), Stefanie Vincken (University Hospital Brussels, Belgium).
1.5 mL microcentrifuge tubes | Sorenson | 17040 | |
15 mL conical tubes | VWR | 525-0605 | |
24 well plates | Corning | 3526 | |
96 well plates | Greiner | 655101 | |
Brightfield microscope | Zeiss | Axiovert 40C | |
Centrifuge | Eppendorf | 5702 | |
CO2 incubator | Binder | CB160 | |
Computer | Hewlett-Packard | Z240 | |
Confocal microscope | Zeiss | LSM 800 | |
Laminar flow hood | Thermo Fisher | 51025413 | |
Material for organoid culture as detailed in previous protocol10 | |||
Micropipettes (20, 200, and 1000 µL) | Eppendorf | 3123000039, 3123000055, 3123000063 | |
Microsoft Excel | Microsoft | Microsoft Excel 2019 MSO 64-bit | Spreadsheet software |
NIS-Elements Advanced Research Analysis Imaging Software | Nikon | v.5.02.00 | Imaging software |
Pipette tips (20, 200, and 1000 µL) | Greiner | 774288, 775353, 750288 | |
Zeiss Zen Blue software | Zeiss | v2.6 | Imaging software |