Live Imaging to Quantify Cellular Radiosensitivity in Patient-Derived Tumor Organoids

Abstract

Radiation therapy (RT) is one of the mainstays of modern clinical cancer management. However, not all cancer types are equally sensitive to irradiation, often (but not always) because of differences in the ability of malignant cells to repair oxidative DNA damage as elicited by ionizing rays. Clonogenic assays have been employed for decades to assess the sensitivity of cultured cancer cells to ionizing irradiation, largely because irradiated cancer cells often die in a delayed manner that is difficult to quantify with short-term flow cytometry- or microscopy-assisted techniques. Unfortunately, clonogenic assays cannot be employed as such for more complex tumor models, such as patient-derived tumor organoids (PDTOs). Indeed, irradiating established PDTOs may not necessarily abrogate their growth as multicellular units, unless their stem-like compartment is completely eradicated. Moreover, irradiating PDTO-derived single-cell suspensions may not properly recapitulate the sensitivity of malignant cells to RT in the context of established PDTOs. Here, we detail an adaptation of conventional clonogenic assays that involves exposure of established PDTOs to ionizing radiation, followed by single-cell dissociation, replating in suitable culture conditions and live imaging. Non-irradiated (control) PDTO-derived stem-like cells reform growing PDTOs with a PDTO-specific efficiency, which is negatively influenced by irradiation in a dose-dependent manner. In these conditions, PDTO-forming efficiency and growth rate can be quantified as a measure of radiosensitivity on time-lapse images collected until control PDTOs achieve a predefined space occupancy.

Introduction

External beam radiation therapy (RT) is one of the mainstays of modern oncology, reflecting not only a pronounced anticancer activity associated with a well-defined spectrum of generally manageable side effects¹, but also an exceptionally widespread clinical availability (most cancer centers in developed countries are equipped with modern linear accelerators for external beam RT)². In line with this notion, RT is globally employed with success for both curative purposes, generally in the context of early-stage disease^3,4, and palliative applications, to contain symptoms (e.g., pain) from metastatic tumors⁵. That said, not all malignancies are equally sensitive to RT, reflecting a number of cancer cell-intrinsic and microenvironmental features^6,7,8,9. As a standalone example, various cancer-associated genetic and epigenetic alterations influencing DNA repair and redox homeostasis have been associated with increased or decreased intrinsic radiosensitivity, largely reflecting the ability of RT to kill cancer cells by causing direct and reactive oxygen species (ROS)-dependent damage to DNA and other macromolecules^10,11,12,13.

Over the past decades, clonogenic assays have been routinely employed to assess the radiosensitivity of cultured human and murine cancer cells^14,15,16. Indeed, while chemotherapeutics and other stressors often kill malignant cells in a fairly rapid manner, RT employed at clinically relevant doses most often elicit a delayed form of cell death that cannot be simply quantified with short-term flow cytometry- or microscopy-assisted techniques, such as measuring the cellular uptake of vital dyes (which selectively stain dead cells) 24-72 h after exposure to a cytotoxic stimulus¹⁷. Besides being straightforward and relying on rather inexpensive reagents, clonogenic assays have been particularly convenient as they allow for the implementation of the so-called "linear quadratic" model, which is a fairly simple mathematical tool for the quantitative analysis of RT dose response^18,19. Along similar lines, cultured cancer cells (including established cell lines as well as primary, patient-derived cells) have provided a convenient platform for testing various aspects of cancer biology, including (but not limited to) sensitivity to treatment in a rather straightforward but simplified manner, one major limitation being the absence of a structured tumor microenvironment (TME)^20,21. Indeed, two-dimensional cancer cell cultures are intrinsically unable to recapitulate cell-cell communications and interactions with the extracellular matrix and as they occur in cancer^22,23. Patient-derived tumor organoids (PDTOs) have emerged as tumor models that at least in part circumvent such limitation^24,25,26.

Unfortunately, conventional clonogenic assays cannot be employed to assess the radiosensitivity of PDTOs. On the one hand, irradiating established PDTOs may not necessarily compromise their growth as a multicellular unit, unless their stem-like compartment - which is responsible for the formation and maturation of PDTOs but exhibits increased resistance to DNA damage as compared to more differentiated PDTO-composing cells²⁷- is completely eradicated. On the other hand, irradiating PDTO-derived single-cell suspensions may not properly recapitulate the radiosensitivity of PDTO-composing (including stem-like) cells as exhibited in the context of formed PDTOs. Here, we detail a technique that harnesses live imaging of breast cancer PDTOs to monitor the PDTO-forming efficacy and growth rate of PDTO-derived cells previously exposed to ionizing irradiation compared to their unirradiated counterparts. With required variations largely reflecting the differential biology of individual PDTOs (e.g., culture media requirements, growth rate), we expect this protocol to be suitable for the study of radiosensitivity in a wide panel of PTDOs of both mammary and non-mammary origin.

Protocol

The reagents and equipment used in the study are listed in the Table of Materials.

1. Organoid culture

NOTE: TNBC#1 PDTOs were established in our lab based on tumor tissue surgically removed from a patient with triple-negative breast cancer (TNBC) who provided informed consent to participate in a biobanking protocol (IRB21-06023682). After validation by histology and RNA sequencing (RNAseq), TNBC#1 PDTOs are cultured in 66% matrigel drops (so-called 'domes') in enriched DMEM/F12 culture medium (Table 1) and passaged every 2-3 weeks at a 1:2-1:10 ratio²⁸.

1. Seed PDTOs at a 1:2-1:10 ratio about 2 weeks before irradiation in separate, non-adherent 6-well plates, one per each radiation dose tested (plus negative controls, which will be provided by PDTOs not exposed to ionizing radiation).
2. Add 3 mL of fresh enriched DMEM/F12 culture medium to the PDTOs every 3-4 days.
   NOTE: Seeding and culture conditions may require some adaptation to prevent PDTOs from acquiring a very recognizable, ring-like appearance that is indicative of core necrosis under the microscope and to ensure optimal viability at irradiation.

2. Radiation

1. PDTOs are ready for irradiation once they reach the average size (approximately 100 µm, as preliminary determined under bright field microscopy with a scale reference) and dome occupancy (approximately 80%). Passage the PDTOs at this stage to avoid overcrowding.
2. Radiate the PDTO-containing domes using the Small Animal Research Radition Platform (SARRP) irradiator (or an alternative irradiator for in vitro or in vivo applications) in open field mode (no collimator), delivering single doses of 2 Gy, 4 Gy, 6 Gy, 8 Gy, and 10 Gy. Non-irradiated PDTOs provide appropriate negative control conditions.
   NOTE: Dosimetry testing calibration should be performed prior to each experiment to ensure that Matrigel-enclosed PDTOs receive the planned radiation dose. These doses were chosen based on preliminary assessments of the radiosensitivity of TNBC#1 PDTOs and, hence, may require adjustments for other PDTOs.

3. Dissociation

1. Immediately after irradiation, wash each PDTO-containing matrigel dome with 2 mL of enriched DMEM/F12 culture medium and resuspend them in 1-3 mL of TrypL-E using a p1000 micropipette. Repeat about 30 times for each well.
2. Collect the cell suspension in a 50 mL centrifuge tube and incubate for ~5 min in a water bath at 37 °C.
3. Pellet the cells by centrifuging them at 800 x g for 5 min at room temperature.
4. Remove excess TrypL-E to leave ~1 mL in the tube and add 3x the original enzyme volume (see step 3.1) of enriched DMEM/F12 medium to wash the cells.
5. Filter the cell suspension using a 40 µm mesh filter. This step ensures that no cell clusters remain at plating and that each new PDTO originates from a single cell.
6. Count the cells in an automated cell counter upon staining with 10% trypan blue in PBS, and resuspend them to have 800 cells in 50 µL of enriched DMEM/F12 medium plus 66% matrigel at 4 °C (conditions for 1 dome).
7. Plate each dome in an individual well from a live imaging-compatible 48-well culture plate while avoiding the formation of bubbles. Technical triplicates for each radiation dose are recommended.
8. Place the plates in a cell culture incubator (37 °C, 5% CO₂) for 30 min to solidify the matrigel.
9. Add 0.5-1 mL of enriched DMEM/F12 medium, remove potential bubbles, and place plates in the live imaging system inside a cell culture incubator. Let the plate equilibrate to the set temperature for about 30 min before imaging.
   NOTE: Matrigel solidifies at temperatures >4 °C, implying that this temperature should be preserved until solidification is required.

4. Live imaging

1. On the Incucyte software, select Schedule to Aquire.
2. Click on the plus sign in the upper left corner of Launch Add Vessel.
3. To scan repeatedly or once, select Scan on Schedule, and then click on Next.
4. To create or restore vessel, select Create Vessel New, and then click on Next.
5. To select the scan Type, select Organoid, and then click on Next.
6. Scan settings: Choose QC < ' Phase + Brightfield Image Channels. '4x Objective' is automatically selected. Then click on Next.
7. Vessel Selection: Choose the appropriate software-validated plate according to the catalog number.
8. Vessel Location: Select the location of the plate on the instrument according to the layout, and then click on Next.
9. Scan Pattern: Select wells according to the plate layout, then click on Next.
10. Vessel Notebook: Name plate (Optional: name Cell Type and Passage). Create a plate Map by clicking on the plus sign in the upper right corner of the plate layout, 'Create plate map'.
11. In the Plate Map Editor, indicate cell type by clicking on the Create New Cell icon in the Cells section, apply to appropriate wells by clicking on the plus icon, and select the appropriate wells. Optional: The number of seeded cells and passages can be indicated. Repeat for each cell line.
12. Indicate the treatment by clicking on the Create New Cell icon in the Growth Conditions section, apply to appropriate wells by clicking on the plus icon, and select the appropriate wells. Repeat for each experimental condition. Next, click on Okay > Next.
13. Analysis Setup: 'Defer analysis until later' is automatically selected; click on Next.
14. Scan Schedule: Choose to scan twice a day, indefinitely. Click on Next. Monitor the darkness of control PDTOs for ideal endpoint, as they tend to collapse when overgrown.
15. Summary: Check the summary and click on Add to Schedule.
    NOTE: Scan properties can be visualized, and the schedule can be edited if needed by right-clicking on the scanning timeline.

5. Analysis

1. On the compatible software, select View. This can also be operated from a computer connected to the system via intranet.
2. Select the desired experiment by double-clicking on it.
3. Select the Launch Analysis icon.
4. Select the Create New Analysis Definition, and then click on Next.
5. Analysis Type: Select Organoid, and click on Next.
6. Image Channels: 'Phase + Brightfield' is selected by default, click 'Next'.
7. Select a set of representative images for analysis. Click on Next. It is recommended to choose early and late time points and different treatment conditions, e.g., control and different radiation doses.
8. Perform Analysis Definition for TNBC#1 following the steps below:
   1. On image: Select Preview Current or 'Preview All' to visualize the default mask of current or selected images. Repeat as analysis parameters are changed to update the mask.
   2. Segmentation: Set the Radius to 300, Sensitivity Background to 80, Edge Split to ON, and Edge Sensitivity to 70.
   3. Cleanup: Set Hole Fill to 5E+05 µm² and Adjust Size to 0 pixels.
   4. Filters: Set no min or max Area, min Eccentricity set to 0.19, and no max Eccentricity. Click on Next.
9. Scan times and wells: Select all desired scan time points and wells. Click on Next. Optional: Select Analyze Future Scans.
10. Save and Apply Analysis Definition: Enter the Definition Name, and click on Next.
11. Visualize the Summary, and click on Finish.
12. A pop-up window appears to inform the user that the analysis has entered the Analysis Queue. Click on Okay. Once the Analysis Queue window opens, the remaining tasks can be visualized.
13. If necessary, retrieve the completed analysis in View by selecting the blue arrow in the Analyses column, next to the Vessel Name, and right-clicking on the Analysis Definition Name. Next, either visualize the raw data on the software by clicking on Graph analysis Metrics, or export for further analysis by selecting Export Analysis Definition.
    NOTE: Raw data exportation can be customized in the Graph analysis Metrics window. Organoid Count, Total Area, Average Eccentricity, and Darkness can be separately exported, and data organized in the Select Grouping roll-down menu by clicking on None, Columns, Rows or plate Map Replicates. For TNBC#1, for raw data exportation, Graph Analysis Metrics, and for the Select Grouping roll-down option, the option None was chosen. These import parameters were chosen based on preliminary assessments with TNBC#1 PDTOs. They hence may require adjustments for other PDTOs, which is possible on the software based on the "Preview" features. We recommend adapting import parameters on (1) early images that mostly contain single cells or small PDTOs, and (2) images collected at late time points to ensure that large PDTOs are also acquired in both control and treated conditions.

Representative Results

TNBC#1 PDTOs were exposed to a single radiation dose of 0 (unirradiated controls), 2 Gy, 4 Gy, 6 Gy, 8 Gy, or 10 Gy on day 0. Immediately thereafter, PDTOs were dissociated to obtain a single-cell suspension for each experimental condition. PDTO-derived cells were next seeded in 48-well plates within 66% matrigel domes (50 µL each) deposited at the center of the wells, in 3 technical replicates per condition. Plates were placed in a live imaging system and imaged every 6 h using the organoid module 4X objective for a total of 30 days. Irradiation of TNBC#1 PDTOs with RT in a single dose of 2 Gy significantly delayed PDTO regrowth, while radiation doses ≥4 Gy (up to 10 Gy) completed prevented the regeneration of growing PDTOs (Figure 1).

Figure 1: Live imaging of patient-derived tumor organoids (PDTOs). Breast cancer TNBC#1 patient-derived tumor organoids (PDTOs) were left untreated or exposed to the indicated radiation dose, shortly followed by dissociation, replating, and imaging with a live cell imaging system. Representative images upon contrast adjustment (A) and quantitative assessments (B, mean ± SEM) are reported. Image scale: 1.19 x 1.16 mm (1.38 mm²). Scale bar: 400 µm; insets: 200 µm. ****p < 0.001 (two-way ANOVA, as compared to 0 Gy); ^####p < 0.001 (two-way ANOVA, as compared to 2 Gy). Please click here to view a larger version of this figure.

Component	Concentration
DMEM F/12	1X
GlutaMax	1X
Hepes	10 mM
PenStrep	100 U/mL
B27	1X
nAc	1.25 mM
Nicotinamide	10 mM
TGFbeta Receptor Inhibitor A83-01	0.5 μM
p38 MAP inhibitor p38i SB202190	1 μM
Noggin	10%
Rspondin Media	10%
FGF10	20 ng/mL
FGF7	5 ng/mL
NR (Heregulin)	5 nM
Primocin	100  μg/mL
Y-27632 (RhoKi)	5  μM
Epidermal Growth Factor hEGF	5 ng/mL

Table 1: PDTO culture medium composition.

Discussion

Here, we describe an adaptation of conventional clonogenic assays that harnesses breast cancer PDTOs and live imaging to quantify PDTO radiosensitivity based on (1) the persistence of PTDO-forming stem-like cells upon PDTO irradiation in vitro, and (2) the growth rate of the PDTOs these cells (may) generate. Critical steps of this protocol include (1) the establishment of PDTOs to a dome occupancy enabling good viability, (2) PDTO exposure to ionizing irradiation at different doses, including mock irradiated control conditions; (3) PDTO dissociation into single cells; (4) replating of PDTO-derived single cells; and (5) live imaging until a predefined dome occupancy in control conditions.

With minimal modifications, mostly linked to PDTO-forming efficiency and PDTO growth rate in control conditions (which are likely to vary considerably across different PDTOs), we expect this protocol to be adaptable to multiple commercially available live imaging platforms, as well as to numerous PDTOs, of both mammary and non-mammary derivation. On the one hand, baseline PDTO-forming efficiency upon single-cell dispersion is expected to influence the number of single cells to be reseeded to obtain new PDTOs at a density that enables not only image-based PDTO quantification but also the normal growth of untreated PDTOs. On the other hand, the PDTO growth rate is expected to affect the duration of the assay, as the merging of adjacent PDTOs should be avoided to preserve optimal analytical capacities. Thus, determining PDTO-forming efficiency and PDTO growth rate in control conditions for each individual PDTO is critical for the correct implementation of this protocol.

While this method is found to be fairly straightforward, it has some limitations. First, an excessively low number of PDTO-forming cells may impair the regeneration of normally growing PDTOs upon single-cell dissociation. Indeed, while routine PDTO passaging can be achieved by incomplete dissociation, quantitative assessments, as offered by this protocol, rely on the generation of a single-cell suspension (which can be visually confirmed during counting). Second, an excessively slow growth rate may considerably extend the duration of the assay and potentially result in an imprecise estimation of radiosensitivity. It is indeed conceivable that an extended culture time as required for slow-growing PDTOs to reach an endpoint-compatible size may offer extra chances for sublethally irradiated PDTO-forming cells to recover from stress and restore detectable (though slow-growing) PDTOs. Finally, dissociating PDTOs shortly after irradiation is expected to impose added stress on PDTO-forming cells, potentially increasing radiosensitivity as a result of altered cell-cell communications in the early phases of recovery. Delaying dissociation of 24-72 h may provide insights into the relative contribution of preserved cell-cell communications to the ability of PDTO-forming cells to recover from macromolecular damage as imposed by RT.

While live imaging platforms offer a convenient method to monitor PDTO formation and growth longitudinally, single endpoint images can also be used for the same objective. In this case, attention should be paid to ensure that adjacent PDTOs do not merge prior to imaging, and growth rate can be assessed as a function of PDTO size normalized to the duration of the assay.

In the era of personalized cancer medicine, assessing the sensitivity of individual tumors to treatment is of the utmost importance as it provides some hints for the design of individualized therapeutic strategies with improved efficacy²⁹. The protocol described herein fits into this endeavor by providing a convenient approach to estimating the cellular sensitivity of individual PDTOs to ionizing radiation (alone or combined with other therapeutic modalities). That said, whether measuring cancer cell radiosensitivity from PDTOs accurately predicts the response of individual patients with cancer to focal RT remains to be demonstrated. Correlating the radiosensitivity of distinct PDTOs to irradiation with the clinical response to RT of their respective donors will provide important insights into this possibility.

Materials

Name	Company	Catalog Number	Comments
40 µm mesh filter	Thomas Scientific	1164H35
B27	Invitrogen	17504-044
Cellometer Auto T4 Bright Field Cell Counter	Nexcelom
DMEM F/12	Corning	12634-010
Epidermal Growth Factor hEGF	Peprotech	AF-100-15
EVOS FL Digital Inverted Fluorescence Microscope	Thermo Fisher Scientific	12-563-460
FGF10	Peprotech	100-26
FGF7	Peprotech	100-19
GlutaMax	Invitrogen	35050061
Hepes	Invitrogen	15630-080
IncuCyte software 2021A	Sartorius		version: 2021A
Incucyte SX1	Sartorius		model SX1
Incucyte validated 48 well plate	Corning	3548
Matrigel	Discovery Labware	354230
nAc	Sigma Aldrich	A9165-5G
Nicotinamide	Sigma-Aldrich	N0636
Noggin			Purchased from the Englander Institute for Precision Medicine, Weill Cornell, NY, USA
Non-treated 6 well plate	Cellstar	657 185
NR (Heregulin)	Peprotech	100-03
p38 MAP inhibitor p38i SB202190	Sigma Aldrich	S7067
PBS	Corning	21-040-CV
PenStrep	Invitrogen	15140-122
Primocin	Invivogen	ant-pm-1
Rspondin Media			Purchased from the Englander Institute for Precision Medicine, Weill Cornell, NY, USA
Small Animal Radiation Research Platform (SARRP)	Xstrahl Ltd
TGFbeta Receptor Inhibitor A83-01	Tocris	2939
Trypan blue Stain (0.4%)	Gibco	15250-61
TrypLE	Gibco	112605-028
Y-27632 (RhoKi)	Selleck	S1049