Long-Term Three-Dimensional High-Resolution Imaging of Live Unlabeled Small Intestinal Organoids Using Low-Coherence Holotomography

doi:10.21203/rs.3.rs-4354176/v1

Download PDF

Article

Long-Term Three-Dimensional High-Resolution Imaging of Live Unlabeled Small Intestinal Organoids Using Low-Coherence Holotomography

https://doi.org/10.21203/rs.3.rs-4354176/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 01 Oct, 2024

Read the published version in Experimental & Molecular Medicine →

You are reading this latest preprint version

Organoids, which are miniature, in vitro versions of organs, present significant potential for studying human diseases and elucidating underlying mechanisms. To fully appreciate and understand the complex structure and dynamic biological phenomena of organoids, live imaging techniques play a crucial role in the field of organoid research. However, challenges in live, unlabeled high-resolution imaging of native organoids are prevalent, primarily due to the complexities of sample handling and optical scattering inherent in three-dimensional (3D) structures. Additionally, conventional imaging methods fall short in capturing the real-time dynamic processes of growing organoids. In this study, we introduce low-coherence holotomography as an advanced, label-free, quantitative imaging modality, designed to overcome related technical obstacles for long-term live imaging of 3D organoids. We demonstrate its efficacy by capturing high-resolution morphological details and dynamic activities within mouse small intestinal organoids at subcellular resolution. Moreover, our approach facilitates the distinction between viable and non-viable organoids, significantly enhancing its utility in organoid-based research. This advancement underscores the critical role of live imaging in organoid studies, offering a more comprehensive understanding of these complex systems.

Biological sciences/Biological techniques/Imaging

Biological sciences/Developmental biology/Organogenesis

Organoids, three-dimensional (3D) multicellular structures that closely emulate organ architecture and function, are cultivated in vitro from both pluripotent and adult stem cells^1–3. The development of organoids has transformed them into essential resources for a broad spectrum of biological research targeting various organs, such as the intestine^4,5, lung^6,7, liver^8–10, kidney^11,12, and brain^13,14. These three-dimensional organ models surpass the conventional two-dimensional cell culture methods by closely replicating the natural architecture and organization of cells as found in living organisms. Organoids are formed when progenitor cells, sourced from an organ's adult stem cells or induced pluripotent stem cells, undergo differentiation and organization into structures that resemble organ function and structure, mirroring the organ development process in nature. This methodology ensures that organoids retain the cellular and molecular properties more faithfully compared to traditional cell cultures. By closely simulating conditions within a living organism, organoids have become pivotal in disease modeling, facilitating the exploration of disease mechanisms and the evaluation of new drugs and therapeutic approaches³.

Imaging plays a crucial role in unraveling the complexity of organoids, revealing their morphology and confirming their accurate representation of in vivo counterparts. The use of thin sectioning along with traditional staining has been widely employed to examine tissue architecture in 2D and to analyze the distribution of single and multiple stained biomarkers. Although brightfield microscopy remains instrumental for organoid visualization, this methodology falls short in capturing their 3D complexity. More advanced modalities, including confocal, multiphoton, and light-sheet microscopy, offer improved 3D imaging but come with their own caveats—namely, the necessity for fluorescent labeling of samples and potential phototoxic effects^15,16. Consequently, there is an ongoing quest for advanced imaging technologies that can provide high-resolution views, deep tissue penetration, and real-time observation of dynamic cellular changes of organoids, all without requiring time consuming sample preparation, labeling, and imaging¹⁷.

In recent years, quantitative phase imaging (QPI) has gained recognition as a technique for the label-free imaging of live biological samples^18–22. Within this realm, holotomography (HT)—a 3D extension of QPI—offers a unique advantage by enabling real-time capture of cellular dynamics in organoids without the drawbacks of phototoxicity or photobleaching. In this work, we employ low-coherence HT for the sustained, label-free monitoring of organoids, shedding light on their developmental trajectories and pharmacological responses. Utilizing mouse small intestinal organoids (sIOs) as our organoid imaging model system, we achieved time-lapse imaging for a duration exceeding 120 hours. This approach allowed us to observe growth patterns, capture detailed subcellular structures with a lateral resolution of 155 nm and an axial resolution of 947 nm, and quantitatively evaluate the organoids' responsiveness to drug treatments. Low-coherence HT was uniquely effective in distinguishing between viable and non-viable cells within the organoids, and offered unparalleled details in depicting 3D morphological shifts following drug exposure. The method further allowed quantitative measurements of organoid volume, protein concentration, and dry mass, setting a new standard for comprehensive, rigorous statistical assessments for the biological studies of organoids.

Organoid culture

Mouse sIOs, derived from primary crypts isolated from the mouse small intestine, were kindly provided by the Center for Genome Engineering (Institute for Basic Science, Korea). Organoid culture media was either manually prepared or purchased. For sIO culture media preparation, advanced DMEM/F12 (Gibco, USA) was supplemented with 0.1M HEPES (Gibco, USA), 1X Glutamax (2mM L-alanyl-L-glutamine dipeptide) (Gibco, USA), 100U/mL Penicillin-Streptomycin (Gibco, USA), 1X serum-free B27 supplement (Gibco, USA), 1.25mM N-Acetyl-L-cysteine (NAc) (Sigma-Aldrich, USA), 0.05 µg/mL mouse EGF (Gibco), 0.1 µg/mL human Noggin (Peprotech, USA), and 0.1 µg/mL human R-Spondin-1 (Peprotech, USA). Alternatively, the commercially available culture medium, IntestiCult (STEMCELL, USA), was prepared according to the manufacturer’s instructions. The culture medium was replaced every 2–3 days, and organoids were passaged weekly. For passaging, both the Matrigel dome and the organoids were mechanically dissociated into individual crypt domains. These individual crypts were then mixed with fresh Matrigel at a split ratio of 1:8, kept at temperatures below 4℃, and distributed in 15 µL dome shapes on a 48-well plate. To polymerize the Matrigel dome, the plate was placed upside-down in an incubator for 10 minutes at 37℃ with 5% CO₂. Once the Matrigel solidified, 250 µL of fresh culture medium was added to each well.

Imaging platform

Holotomograms of sIOs were obtained using a low-coherence HT system (HT-X1, Tomocube Inc., Korea). HT-X1 employs incoherent 450-nm LED light for illumination, enabling the imaging of the RI of thick specimens with reducing speckle noise. The sample undergoes axial scanning over a range of 140 µm, with the transmitted intensity for each illumination pattern being measured at intervals of 947 nm. Subsequently, deconvolution and stitching techniques are utilized to reconstruct the RI of the samples within the desired field of view²³. Low-coherence HT has a theoretical resolution of 155 nm and 947 nm in the lateral and axial directions, respectively. An integrated stage top incubation chamber ensured stable physiological conditions of temperature, humidity, and CO₂ concentration during long-term imaging over several weeks. All motorized microscopic operations were controlled and monitored by an operating software TomoStudio X (Tomocube Inc., Korea).

Sample preparation for imaging

The HT-X1 is adaptable with a wide variety of commercial imaging dishes, multi-well plates, and custom imaging vessels, as long as they have a bottom thickness of #1.5H. In this study, organoids were imaged using a coverslip-bottomed imaging dish (Tomocube Inc., Korea). Prior to imaging, the sIOs underwent mechanical dissociation as part of the organoid passaging procedure. The dissociated individual crypts were mixed with fresh Matrigel, then distributed as 15 µL domes onto the coverslip-bottom imaging dish. This was followed by a polymerization process. Once the Matrigel domes solidified, they were topped up with 3mL of culture medium and the imaging dish was positioned in the chamber of the HT-X1.

For correlative fluorescent organoid imaging, a staining solution is prepared using Calcein-AM (ThermoFisher, USA) diluted 1:200 to achieve a 5 µM concentration, and Hoechst (ThermoFisher, USA) diluted 1:1000 to reach a 1 µg/mL concentration. Both are in 3 mL of DPBS (Sigma Aldrich, USA). Following this, the culture medium surrounding the sIOs in the imaging dish is gently removed. Subsequently, sIOs are rinsed with DPBS. Once the rinse is removed, the staining solution is applied to the sIOs, ensuring uniform coverage. The organoids are then incubated for 1.5 hours to ensure optimal staining. After this duration, the excess staining solution is gently aspirated, followed by a final rinse with DPBS to remove any residual dye.

Data acquisition

The center of the Matrigel dome was found manually with the help of a brightfield preview image scanned over the area of 4 mm × 4 mm. Randomly selected organoids within the preview were imaged with a field of view (FOV) of 160 µm × 160 µm and a depth range up to 140 µm. For the organoids exceed the size of FOV, multiple holotomograms were acquired to construct a stitched volume. The stitching process was automatically done by the software TomoStudio X (Tomocube Inc, Korea). In the case of time-lapse imaging, the humidity was maintained by replenishing the incubation chamber with deionized H₂O 5 mL/day. Temperature and CO₂ concentration were automatically adjusted to 37℃ and 5% respectively by the chamber controller unit.

Image processing

Each stack of raw holotomograms was processed with Matlab (Mathwork, USA). First, we calculated gradient image from raw image by applying Sobel filter for each XY-slice. Stacks were cropped with a range along z-axis determined by thresholding mean intensity of gradient, representing the existence of in-focus features within each slice. Generated stack of gradient image then applied by depth color coding. Finally, each stack was projected in single plane by maximum intensity-based projection method. Applied colormap was expressed as a color bar in each produced image.

Feature extractions

After obtaining the organoid mask using the open-source, user-interactive segmentation toolkit, ilastik²⁴, it was processed in Matlab to determine the volume through the built-in function 'regionprops3'. The number of voxels was converted to volume by adjusting for both lateral and axial resolution. Protein density was derived from the organoid's RI, with an increment value of 0.185 ng/µm used in this study. The protein mass of the organoids was then calculated by multiplying the volume by the protein density. Histograms of the organoids were generated with the Matlab built-in 'histogram' function. The region of interest for the histogram was specifically designated using the organoid mask to exclude non-organoid regions.

Statistical analysis

Statistical analyses were performed using the Matlab by a two-tailed, unpaired Student’s t-test. A p-value of < 0.05 was considered statistically significant and data are presented as the mean ± s.e.m.

Low-coherence HT reveals distinct structures of native sIOs

To comprehensively examine and continuously observe live sIOs, we developed protocols that encompass initial steps such as subculturing, followed by imaging, segmentation, and detailed morphological and quantitative analyses (Fig. 1a). The first step involves splitting mature sIOs, and subculturing these fragments within Matrigel domes that mimic the extracellular matrix, targeting 5–10 organoids per dome. Following a five-day incubation, the organoid samples are relocated to coverslip-bottom dishes for imaging. Low-coherence HT is then employed, using four distinct patterned beams to illuminate the samples²³. These raw images undergo a deconvolution algorithm for reconstruction. As organoids grow beyond the lateral field of view (160 µm × 160 µm) of the imaging device, tiled image capturing becomes necessary, followed by post-acquisition stitching. The reconstructed images show the distribution of the refractive index (RI) within the sIOs. In cases where multiple organoids are present volumetrically, and out-of-focus beams interfere with the illumination, aberrations are inevitable. To counteract these aberration-induced artifacts, which obscure organoid boundaries, a Sobel filter is applied. This results in synthetically generated images with clearly delineated boundaries. Further segmentation and labeling of organoid outlines are accomplished using the open-source, user-interactive segmentation toolkit, ilastik²⁴. To track the axial growth of sIOs, color-coding along the axial direction is employed, with images projected in a consistent orientation. Within these segmented images, parameters such as volume, protein concentration, and mass are quantitatively assessed, enabling comprehensive, longitudinal tracking of organoid features.

With our 3D reconstruction images of sIOs, we identified two types of sIOs based on their morphology: enterocystic sIOs (Fig. 1b) and multilobular sIOs (Fig. 1c). In the early stages of organoid growth, enterocystic sIOs are commonly observed, while in the later stages, the emergence of multilobular structures becomes more prevalent, a change primarily due to the process of crypt budding. These observations are consistent with previous studies on long-term imaging of sIOs growth^25,26. Through optical sectioning of reconstructed HT images, we elucidate the complex internal architecture of sIOs including goblet cells, Paneth cells, and enteroendocrine cells, each exhibiting typical morphologies consistent with previously reported electron microscopic studies^27,28 (Fig. 1d). Additionally, observed cellular polarity provides valuable insights into the apical lumen and basal environments. Wide-field stitched images further reveal macroscopic features such as brush border formation and the accumulation of exfoliated cells within the apical lumen (Fig. 1e).

Low-coherence HT captures rapid dynamic changes of sIOs

To systematically examine sIO development, Matrigel-embedded organoids were subjected to hourly imaging over a span of 120 hours. From the initial enterocystic stage, we noted phenomena such as symmetry breaking and protrusion, eventually leading to the budding of crypts (Fig. 2a). The axial progression of cyst formation and crypt budding was vividly captured using projected depth-color-coded images. By applying segmented masks to Sobel-filtered images, we tracked organoid size over total imaging period, revealing an exponential growth trend (Fig. 2b).

Furthermore, we segmented and tracked the volume of the lumens within the sIOs, documenting an increase in volume corresponding with organoid growth. Given the linear relationship between RI and protein density in biological samples^29,30, we calculated the protein density within the sIOs (Fig. 2c). Our findings indicate that the total protein concentration within the sIOs remained relatively constant in comparison to their luminal space, suggesting minimal changes in compositional makeup. As the sIOs enlarged, we noted a reduction in lumen volume, a phenomenon likely due to the dilution of protein density by the increased presence of empty space from exfoliated cells within the lumen. By integrating volume and protein density measurements, we successfully determined the dry protein mass of the sIOs and their luminal contents (Fig. 2d).

Benefiting from the rapid acquisition capabilities of low-coherence HT, we were able to capture the dynamic behavior of live sIOs embedded in Matrigel. While specific dividing cell types were not identified, key mitotic events were observed, including apical cytokinesis and cell interspersion, followed by basal reattachment (Fig. 2e). These observations correlate well with established mitotic features from prior studies³¹. Moreover, we documented epithelial cell exfoliation and subsequent cellular apoptosis and chromatin condensation (Fig. 2f). While previous studies were limited to observing intestinal cell exfoliation^32,33, our study expanded upon this by capturing the entire process, including intercellular migration and apoptosis of exfoliated cells. This supports the long-debated notion that the exfoliation removes senescent or damaged enterocytes³⁴.

Observing drug responses in organoids with low-coherence HT

To evaluate cell death and viability within the organoids, sIOs were treated with cisplatin, and morphological changes were closely observed. We cultured the sIOs in media containing varying concentrations of cisplatin to ascertain the optimal concentration for discernible toxic effects. A concentration of 10 µM was identified as optimal, a finding corroborated using bright-field microscopy. Dimethyl sulfoxide (DMSO) served as a vehicle and thus a positive control. Post-treatment, the samples were placed in the imaging system's incubator and were imaged at 10-minute intervals over a 48-hour span (Fig. 3a). Post-cisplatin treatment, we observed that the branched crypts underwent noticeable shrinkage and displayed an increasing number of dissociated dead cells. In contrast, the vehicle-treated organoids showed enhanced crypt budding and growth (Fig. 3b). The observed reduction in organoid volume can likely be attributed to cellular shrinkage and the exfoliation and dissociation of dead cells, as previously reported³⁵. Quantitative tracking of organoid size further supported these observations (Fig. 3c). While the volume of vehicle-treated sIOs exhibited an increase, those treated with cisplatin showed a decrease. In terms of protein density, the vehicle-treated group remained constant, whereas a significant decrease was observed in the cisplatin-treated group after 24 hours (Fig. 3d). Subsequently, when assessing the overall protein mass, an increase was noted in the vehicle-treated sIOs, in contrast to a marked decrease in the cisplatin-treated group (Fig. 3e).

Quantification of viability using low-coherence HT reveals a decrease in the RI

To further elucidate the marked decline in protein density observed in the cisplatin-treated group, we investigated the RI distribution within the sIOs. Initially, our aim was to perform a cell-by-cell analysis by segmenting individual cells using single-cell masks. However, this approach was impeded by indistinct cell boundaries and cell clumping. To circumvent this challenge, we partitioned the holotomograms of the sIOs into multiple cubic sections, each measuring 50 µm × 50 µm laterally and 10 µm axially. This strategy ensured that each section contained an entire single enterocyte for more accurate analysis (Fig. 4a). We calculated protein density of each cubic section, and compared between organoid of cisplatin-treated sIO and vehicle-treated sIO after 24 hours from treatment (Fig. 4b). The cisplatin-treated group showed lower protein density (mean ± std; 0.104 ± 0.045 ng/µm³), while vehicle-treated group showed higher (0.133 ± 0.037 ng/µm³). We hypothesized the lower protein density originate from the increased number of dead cells within the cisplatin-treated sIO since the region including dead cells exhibited low RI (Fig. 4a). Previous research reported that, during cell death, a decrease in the phase of scattered beams from cells, which also indicates a decrease in RI and this back up our hypothesis³⁶. To validate our hypothesis, we stained sIOs with Hoechst and Calcein-AM and take their fluorescent images with low-coherence HT (Fig. 4c). Since Hoechst stain whole nucleus of cells in any live/dead state while Calcein-AM can stain only live cells, we could identify live cell dominant regions and dead cell dominant regions. Then we compared protein density of co-stained regions (live cells dominant) and Hoechst-only stained regions (dead cells dominant). The histograms of protein density revealed different distribution between co-stained regions and Hoechst-only regions (Fig. 4d). The dead cell dominant regions showed lower protein density (mean ± std; 0.101 ± 0.048 ng/µm³) compared to live cell dominant regions (0.115 ± 0.050 ng/µm³).

In summary, low-coherence HT offers a real-time, label-free method for organoid imaging, enabling researchers to explore the detailed complexities of organoids without time-consuming preparation and labeling. Using low-coherence HT, we captured long-term, real-time observations of biological processes within organoids, such as cell apoptosis, migration, mitosis, and dynamics of subcellular organelles at a resolution between bright field microscopy and electron microscopy. These high-resolution 3D observations were difficult to achieve in real-time using conventional imaging methods, particularly in intact organoids in Matrigel media. This study presents advancements in two primary domains. Firstly, the necessity for sample preparation steps, including fixation and staining, is obviated while preserving the structural integrity of live organoids. Secondly, the quantitative analytical capabilities of low-coherence HT provide robust measures to assess organoid biology, including the RI values, making our imaging system a new tool for organoids in pharmacological screening applications.

Despite the advantages of low-coherence HT for organoid imaging, challenges arise when attempting to image thick organoids exceeding an axial size of approximately 200 µm. Current limitations in the working distance of the objective lens constrain the axial resolution, resulting in organoids positioned in the upper portion of the Matrigel dome yielding images of diminished quality. Furthermore, the reliance on organoid scattering signals can introduce sample-induced aberrations, further compromising image quality. However, the low-coherence HT offers considerable potential for imaging smaller organoids (less than 200 µm) during their nascent developmental stages—a significant advantage for early-stage drug response analysis using organoids.

In our current imaging system, we still need to address the challenge of low biomolecular specificity. A potential solution for overcoming this limitation is through the implementation of correlative imaging techniques, which can facilitate differentiation of individual subcellular structures. As recently demonstrated, deep-learning algorithms can accurately predict subcellular structures, with their performance attributed to the distinct features within the spatial distribution of RI associated with each structure^37–39, which can be applied to the present label-free time-lapse imaging of unlabeled live organoids. We expect that combining correlative imaging with the power of artificial intelligence could lead to improved biomolecular specificity in our imaging system. Furthermore, validating this approach using labeling agents that target specific molecules could offer additional confidence in its effectiveness.

The large size of volumetric organoid holotomograms, approximately 100 GB per organoid, and the associated processing time pose significant challenges in the current landscape. The emergence of high-throughput platforms equipped with advanced computing power will enable analyzing intricate cellular morphologies and dynamic changes within multiple organoids at high resolution. These technological advancements promise to revolutionize the field of regenerative medicine, offering a robust means of validating the quality of organoids destined for therapeutic applications. We believe this study provides evidence for the application of low-coherence HT as an organoid imaging modality. This could prompt a shift in drug screening paradigms, potentially leading to more personalized and efficient treatment strategies.

Acknowledgement

We thank all the members of the BMOL in KAIST, Center for Genome Engineering in IBS and Tomocube Inc. This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (RS-2023-00241278), NRF of Korea (2015R1A3A2066550, 2022M3H4A1A02074314), Institute of Information & Communications Technology Planning & Evaluation (IITP; 2021-0-00745) grant funded by the Korea government (MSIT), KAIST Institute of Technology Value Creation, Industry Liaison Center (G-CORE Project) grant funded by MSIT (N11230131) and Tomocube Inc. The authors thank Herve Hugonnet for technical discussion.

Author contributions

M.J.L. and J.H.L designed the study with input from Y.K.P. and B.K.K. Data acquisition was done by M.J.L. and J.H.L. J.H.L. performed data processing and M.J.L. performed data analysis. M.J.L. and J.H.L. wrote the manuscript, with feedback from G.K., H.J.K., S.M.L., J.M.H., B.K.K and Y.K.P.

Competing interests

M.J.L., J.H.L., H.J.K., S.M.L., and Y.K.P, have financial interests in Tomocube Inc., a company that commercializes holotomography and is one of the sponsors of the work.

Clevers, H. Modeling Development and Disease with Organoids. Cell 165, 1586-1597 (2016).
Fatehullah, A., Tan, S.H. & Barker, N. Organoids as an in vitro model of human development and disease. Nature Cell Biology 18, 246-254 (2016).
Kim, J., Koo, B.-K. & Knoblich, J.A. Human organoids: model systems for human biology and medicine. Nature Reviews Molecular Cell Biology 21, 571-584 (2020).
Sato, T., et al. Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Barrett's epithelium. Gastroenterology 141, 1762-1772 (2011).
Yui, S., et al. Functional engraftment of colon epithelium expanded in vitro from a single adult Lgr5+ stem cell. Nature medicine 18, 618-623 (2012).
Kim, M., et al. Patient-derived lung cancer organoids as in vitro cancer models for therapeutic screening. Nature communications 10, 3991 (2019).
Dye, B.R., et al. In vitro generation of human pluripotent stem cell derived lung organoids. elife 4, e05098 (2015).
Huch, M., et al. In vitro expansion of single Lgr5+ liver stem cells induced by Wnt-driven regeneration. Nature 494, 247-250 (2013).
Huch, M., et al. Long-term culture of genome-stable bipotent stem cells from adult human liver. Cell 160, 299-312 (2015).
Takebe, T., et al. Vascularized and functional human liver from an iPSC-derived organ bud transplant. Nature 499, 481-484 (2013).
Takasato, M., et al. Kidney organoids from human iPS cells contain multiple lineages and model human nephrogenesis. Nature 526, 564-568 (2015).
Homan, K.A., et al. Flow-enhanced vascularization and maturation of kidney organoids in vitro. Nature methods 16, 255-262 (2019).
Lancaster, M.A., et al. Cerebral organoids model human brain development and microcephaly. Nature 501, 373-379 (2013).
Quadrato, G., et al. Cell diversity and network dynamics in photosensitive human brain organoids. Nature 545, 48-53 (2017).
Fei, K., Zhang, J., Yuan, J. & Xiao, P. Present Application and Perspectives of Organoid Imaging Technology. Bioengineering 9, 121 (2022).
Keshara, R., Kim, Y.H. & Grapin-Botton, A. Organoid imaging: seeing development and function. Annual Review of Cell and Developmental Biology 38, 447-466 (2022).
Rios, A.C. & Clevers, H. Imaging organoids: a bright future ahead. Nature methods 15, 24-26 (2018).
Park, Y., Depeursinge, C. & Popescu, G. Quantitative phase imaging in biomedicine. Nat Photonics 12, 578-589 (2018).
Pirone, D., et al. Stain-free identification of cell nuclei using tomographic phase microscopy in flow cytometry. Nature photonics 16, 851-859 (2022).
Jin, D., Zhou, R., Yaqoob, Z. & So, P.T. Tomographic phase microscopy: principles and applications in bioimaging. JOSA B 34, B64-B77 (2017).
Wu, Y., et al. Intelligent frequency-shifted optofluidic time-stretch quantitative phase imaging. Opt Express 28, 519-532 (2020).
Liu, Y. & Uttam, S. Perspective on quantitative phase imaging to improve precision cancer medicine. Journal of Biomedical Optics 29, S22705-S22705 (2024).
Hugonnet, H., Lee, M. & Park, Y. Optimizing illumination in three-dimensional deconvolution microscopy for accurate refractive index tomography. Optics Express 29, 6293-6301 (2021).
Berg, S., et al. ilastik: interactive machine learning for (bio)image analysis. Nature Methods 16, 1226-1232 (2019).
Sato, T., et al. Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459, 262-265 (2009).
de Medeiros, G., et al. Multiscale light-sheet organoid imaging framework. Nature Communications 13, 4864 (2022).
Miura, S. & Suzuki, A. Generation of Mouse and Human Organoid-Forming Intestinal Progenitor Cells by Direct Lineage Reprogramming. Cell Stem Cell 21, 456-471.e455 (2017).
Dicarlo, M., et al. Quercetin Exposure Suppresses the Inflammatory Pathway in Intestinal Organoids from Winnie Mice. International Journal of Molecular Sciences 20, 5771 (2019).
Barer, R. Determination of dry mass, thickness, solid and water concentration in living cells. Nature 172, 1097-1098 (1953).
Popescu, G., et al. Optical imaging of cell mass and growth dynamics. American Journal of Physiology-Cell Physiology 295, C538-C544 (2008).
McKinley, K.L., et al. Cellular aspect ratio and cell division mechanics underlie the patterning of cell progeny in diverse mammalian epithelia. eLife 7, e36739 (2018).
Kaeffer, B., et al. Recovery of Exfoliated Cells From the Gastrointestinal Tract of Premature Infants: A New Tool to Perform “Noninvasive Biopsies?”. Pediatric Research 62, 564-569 (2007).
Ahlquist, D.A., Harrington, J.J., Burgart, L.J. & Roche, P.C. Morphometric analysis of the “mucocellular layer” overlying colorectal cancer and normal mucosa: Relevance to exfoliation and stool screening. Human Pathology 31, 51-57 (2000).
Loktionov, A. Cell exfoliation in the human colon: Myth, reality and implications for colorectal cancer screening. International Journal of Cancer 120, 2281-2289 (2007).
Grabinger, T., et al. Ex vivo culture of intestinal crypt organoids as a model system for assessing cell death induction in intestinal epithelial cells and enteropathy. Cell Death & Disease 5, e1228-e1228 (2014).
Pavillon, N., et al. Early Cell Death Detection with Digital Holographic Microscopy. PLoS One 7, e30912 (2012).
Jo, Y., et al. Label-free multiplexed microtomography of endogenous subcellular dynamics using generalizable deep learning. Nature Cell Biology 23, 1329-1337 (2021).
Bai, B., et al. Deep learning-enabled virtual histological staining of biological samples. Light: Science & Applications 12, 57 (2023).
Park, J., et al. Artificial intelligence-enabled quantitative phase imaging methods for life sciences. Nature Methods 20, 1645-1660 (2023).

There is a conflict of interest M.J.L., J.H.L., H.J.K., S.M.L., and Y.K.P, have financial interests in Tomocube Inc., a company that commercializes holotomography and is one of the sponsors of the work.

Download PDF

Journal Publication

published 01 Oct, 2024

Read the published version in Experimental & Molecular Medicine →

Editorial decision: revise
17 Jun, 2024
Review #2 received at journal
14 Jun, 2024
Review #1 received at journal
01 Jun, 2024
Reviewer #2 agreed at journal
13 May, 2024
Reviewer #1 agreed at journal
12 May, 2024
Reviewers invited by journal
12 May, 2024
Submission checks completed at journal
01 May, 2024
Editor assigned by journal
01 May, 2024
First submitted to journal
01 May, 2024

You are reading this latest preprint version

Long-Term Three-Dimensional High-Resolution Imaging of Live Unlabeled Small Intestinal Organoids Using Low-Coherence Holotomography

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials and Methods

Organoid culture

Imaging platform

Sample preparation for imaging

Data acquisition

Image processing

Feature extractions

Statistical analysis

Results

Low-coherence HT reveals distinct structures of native sIOs

Low-coherence HT captures rapid dynamic changes of sIOs

Observing drug responses in organoids with low-coherence HT

Quantification of viability using low-coherence HT reveals a decrease in the RI

Discussion

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1