Long-term three-dimensional high-resolution imaging of live unlabeled small intestinal organoids using low-coherence holotomography

The prevailing challenges in live unlabeled high-resolution imaging of native organoids stem from technical issues like complex sample handling and optical scattering in three-dimensional architectures. 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 successfully captured high-resolution morphological intricacies and dynamic events within mouse small intestinal organoids at a subcellular resolution. Furthermore, this method provides a unique advantage in differentiating between viable and non-viable organoids, thereby expanding its potential applications in organoid-based research.

Initial steps involve 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 samples 11 .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.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, ilastik 12 (Extended Data Fig. 1).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.
Through optical sectioning of reconstructed HT images, we elucidate the complex internal architecture of sIOs (Fig. 1b).Remarkably, we were able to identify goblet, Paneth, and enteroendocrine cells, each exhibiting typical morphologies consistent with previously reported electron microscopic studies 13 .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.
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. 1c and Movie 1 online).These observations correlate well with established mitotic features from prior studies 14 .Moreover, we documented epithelial cell exfoliation (Fig 1d) and subsequent cellular apoptosis and chromatin condensation (Movie 2 online).While previous studies were limited to observing intestinal cell exfoliation 15,16 , 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 enterocytes 17 .
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. 1e).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 this period, revealing an exponential growth trend (Fig. 1f).Variability in organoid dimensions was noted, attributed to the shedding of apoptotic cells; these fluctuations were quantified as alterations in the absolute growth rate (Fig. 1g).Importantly, a temporal increase in growth rate was observed as the experiment progressed.For each timeframe snapshot, the z-position was adjusted to the best focal plane for the targeted cells.Scale bar = 20 µm.e, Sequential images illustrating sIO development.For each time point, the upper row presents the focal plane, while the depth-colorcoded projections are displayed in the lower row.Scale bar = 50 µm.f-g, A graphical illustration of sIO's size (f) and its respective absolute growth rate (g) over time.
To evaluate cell death 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 (Extended Data Fig. 2).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 2a).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. 2b).The observed reduction in organoid volume can likely be attributed to cellular shrinkage and the exfoliation and dissociation of dead cells, as previously reported 18 .Quantitative tracking of organoid size further supported these observations.While the volume of the vehicle-treated sIOs increased, that of the cisplatin-treated sIOs decreased (Fig. 2c).Using the linearly proportional relationship between drymass density and the refractive index (RI) 19,20 , we calculated the protein density in sIOs.The vehicle-treated group maintained a consistent protein density, whereas a drastic decrease was observed in the cisplatin-treated group after 12 hours (Fig 2c).By computing the product of organoid size and protein density, we determined the overall protein mass.Here, an increase was noted in the vehicle-treated group, while a decrease was evident in the cisplatin-treated group.
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. 2d).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 2e).The cisplatin-treated group showed lower protein density (mean ± std; 0.232 ± 0.103 ng/μm 3 ), while vehicle-treated group showed higher (0.277 ± 0.085 ng/μm 3 ).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 (Extended Data Fig 3).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 hypothesis 21 .To validate our hypothesis, we stained sIO with Hoechst and Calcein-AM and take their fluorescent images with lowcoherence HT (Fig 2f).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 region and dead cell dominant region.Then we compared protein density of co-stained region (live cells dominant) and Hoechst-only stained region (dead cells dominant).The histogram of protein density revealed different distribution between co-stained region and Hoechst-only region (Fig 2g).The dead cell dominant region showed lower protein density (0.220 ± 0.054 ng/μm 3 ) compared to live cell dominant region (mean ± std; 0.317 ± 0.067 ng/μm 3 ).In the merged image, the magenta and orange boxes indicate live cell-dominant and dead cell-dominant regions, respectively.Scale bar = 50 µm.g, Histogram illustrating the protein density distribution within the boxed regions of (f).Each color corresponds to the respective color of the boxes.
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 lowcoherence HT, we captured long-term, real-time observations of biological processes within organoids, such as cell apoptosis, migration, mitosis, and dynamics of subcellular organelles within resolutions 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 various measures to assess organoid biology, including the RI values, making our imaging system a new tool for organoids in pharmacological screening applications.

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-Lcysteine (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 view 11 .Lowcoherence 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 coverslipbottomed 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.

Fig 1 :
Fig 1: High-resolution holotomograms of live mouse sIOs via low-coherence HT | a, Overview of the low-

Fig 2 :
Fig 2: In-depth quantitative analysis of sIOs leveraging low-coherence HT | a, Diagram illustrating holotomogram-based viability assessment of drug-treated sIOs.Red arrowheads highlight the time points at which images were captured.b, Time-lapse images of sIOs following drug treatment.Each colored box is magnified, with the color bar indicating depth of samples.Scale bar = 20 µm.c, Longitudinal quantitative assessment of sIOs.Every parameter is calibrated against the initial measurement; the dotted line signifies the baseline value.d, Illustration of the subvolume sectioning strategy alongside representative images.Scale bar = 20 µm.e, scattered plot of protein density between vehicle-treated sIO and cisplatin-treated sIO. each dot represents a subvolume of