Mouse intraductal modeling of primary ductal carcinoma in situ

Summary Mouse intraductal modeling enables efficient in vivo propagation of pre-invasive breast cancer lesions and provides a suitable micro-environment for creating patient-derived tumor xenograft models of estrogen-receptor-positive breast cancer. Here, we present a protocol for mouse intraductal modeling of primary ductal carcinoma in situ (DCIS). We describe steps for processing primary DCIS tissues and performing intraductal injections. We then detail procedures for processing intraductal lesions for 3D whole-mount imaging or serial transplantation using magnetic bead sorting. For complete details on the use and execution of this protocol, please refer to Hutten et al. (2023).1

In this step primary tissue is mechanically and enzymatically dissociated to a single cell solution suitable for intraductal injection.
1. Dissect resected sample with scalpels in a 6 cm petri dish in a few mL of Digestion solution.
CRITICAL: Take the potential presence of human pathogens into account and adhere to the corresponding biosafety protocols.
CRITICAL: To preserve tissue viability of the primary tissue, it should be processed as quickly as possible after surgical resection (1-5 h).Keep tissue at 4 C in the meantime.
Note: As a general rule of thumb the tissue pieces should be small enough to be taken up by a 25 mL pipette, but try to cut the tissue pieces as small as possible.
2. Transfer tumor pieces to a 50 mL Falcon tube and add 10 mL of digestion solution per 100 mg of tumor tissue.3. Incubate in shaker at 100 rpm at 37 C for 16 h overnight.
Note: Continuous movement of the tumor pieces in the digestion solution improves digestion and the final number of cells obtained.
Note: Coating of material with BSA minimizes loss of primary cells, which is specifically critical when dealing with small amounts of tissue.
10. Filter solution through a 70 mm cell strainer (2.5% BSA solution coated) into a 50 mL falcon tube (2.5% BSA solution coated), transfer into a 15 mL falcon tube (2.5% BSA solution coated ) and spin at 1500 3 g for 10 min.11.Aspirate supernatant and suspend cells in 100 mL of PBS (If a large pellet is present 1 mL of PBS can be used).
Note: keep the cells on ice from this point onwards.
12. Count cells: a.Take 10 mL of cell suspension and add 10 mL of trypan blue.b.Pipet 10 mL of this mix on a cell counting chamber and count cells with cell counter.c.Calculate the volume of cells needed to obtain 25,000 cells in 18 mL, which will be injected in each mammary gland.
Note: The single cell suspension will also contain immune-and stromal cells, which are included in the 25,000 cells.

Intraductal injections of the mammary gland
Timing: 20-25 min Timing: 10 min for setup Timing: 10-15 min per mouse This section describes the setup and steps for intraductal injections of the mammary gland.For intraductal injection the mouse needs to be immobilized and anesthetized.The nipple of the mouse should also be clearly visible, which is done by carefully shaving the area surrounding the nipple.
13. Prepare the setup in a laminar flow hood as shown in Figure 1.a. Cover the bottom of the cabinet with 3 sterile diapers.b.Install the camera, screen and small heating pad.c.Wipe everything that will be in contact with the mouse or you during surgery with disinfectant (e.g., VirkonS) before introduction into the biohazard cabinet or introduce them sterile: tissues, razor blades, 70% Ethanol, PBS, 15 mL Falcon tube, P20 pipet and tips, 1.5 mL Eppendorf tubes, anesthesia induction box, waste box, Evans blue (2 mg/mL), samples on ice, syringes, instruments (tweezers, Hamilton syringe) on a tissue right of the surgical area under the camera.d.Turn on heating pad (37 C).
CRITICAL: It is crucial to use the heating pad on 37 C to keep the mouse on the right body temperature.
e. Connect an anesthesia induction box / 5 mL syringe to the isoflurane station.
Note: 5 mL Syringes without plunger can be used as a head cone for the anesthesia for shaving and the injection.
14. Anesthetize the mouse with 2.5% isoflurane and 2 L per minute air flow until it falls asleep.15.Carefully shave around the nipples.16.Put the mouse in the isoflurane/5 mL syringe under the camera on a heating pad.17.Prepare droplets of 20 mL (2 mL Evans blue + 18 mL tumor cells) in the lid of an Eppendorf tube and fill the syringe.
Note: First clean the syringe with PBS.Also clean with PBS in between samples.
CRITICAL: While filling the syringe try to avoid air bubbles, as these can clog the mammary duct, resulting in failure of the injection.
18. Remove the cap of dead skin on the nipple with tweezers (not all nipples have this).19.Hold the nipple between the tweezers and inject the 20 mL cell suspension in the nipple (see Figure 2 and Methods video S1).
CRITICAL: It is not necessary to cut the tip of the nipple as described in other protocols.
CRITICAL: Check whether the injection was successful if the solution disperses in the mammary gland.A blue bulge directly at the place of injection indicates a failed injection (Figure 3).
Note: Depending on the experiment it is possible to inject multiple mammary glands.The preferred mammary glands are the 3 rd and 4 th glands on both sides.

Note:
The cell suspension should be gradually injected to avoid backflow of the solution.
20. Monitor outgrowth.For primary DCIS samples it takes 6-12 months on average for DCIS lesions to form, which can then be processed further.

Timing: 2 days
This section describes the protocol to produce 3D whole mounts of mammary glands to image the entire ductal tree and outgrown lesions.This technique can provide information on characteristics such as extent of growth, invasive potential, 3D morphology, growth location etc.
21. Sacrifice the animal and carefully dissect out the injected mammary glands.34.Mount the mammary gland between 2 coverslips using a whole-gland mounting device.
Note: For a detailed description of the whole-gland mounting device, see Hannezo & Scheele. 5.Image the tissue using a confocal microscope (see Figure 4).

Serial transplantation of MIND models
Timing: $20 h Timing: 3 h (for steps 37-58 after overnight digestion) This section describes the protocol for processing the outgrowth of the primary tissue in the previous section for serial transplantation.As MIND models often only show intraductal growth it is necessary to separate the human tumor cells from the mouse cells before reinjection.Here we describe a negative selection for unlabeled human cells using magnetic bead sorting.For tumors larger than 500 mm 3 , the following steps are not necessary and only steps 1 to 11 need to be repeated for serial transplantation.
36.Sacrifice the animal and carefully dissect out the injected mammary glands.37. Repeat step 1 to 11, only suspend cells in 300 mL of cold PBS with 0.5% BSA at step 11.38.Add 100 mL of human FcR Blocking Reagent for up to 10 8 total cells (20 mL of FcR per 60 mL buffer/ 10 7 cells).39.Label cells with biotinylated antibodies (Mouse Anti-MHC Class I and II) at a dilution of 1:100 (for 400 mL PBS with 0.5% BSA and FcR add 4 mL antibodies).Incubate for 20 min on ice.40.Wash cells by adding 5-10 mL of PBS with 0.5% BSA per 10 7 cells and centrifuge at 1500 3 g for 10 min.Aspirate supernatant completely then repeat wash.41.Resuspend pellet in 300 mL PBS with 0.5% BSA and add 100 mL of Anti biotin MicroBeads Ultra-Pure (20 mL of beads per 60 mL buffer).42.Mix well and incubate for 20 min in the refrigerator (2-8 C). 43.Wash cells by adding 5-10 mL of PBS with 0.5% BSA per 10 7 cells and centrifuge at 1500 3 g for 10 min.44.Resuspend up to 10 8 cells in 500 mL of PBS with 0.5% BSA.
Optional: Save 10 mL for pre-sort flow cytometry.45.Place LD column in the magnetic field of a suitable MACS Separator.46.Prepare LD column by rinsing with 2 mL of PBS with 0.5% BSA.52.Spin down the unlabeled cells from step 48 and 49 and resuspend in 110 mL PBS.
a. Use 10 mL for counting as described previously in step 12. b.Use another 10 mL for flow cytometry analysis.c.Store the rest of the cell suspension at 4 C for intraductal injection.
Flow cytometry analysis to detect human cells This section describes the flow cytometry analysis to determine the number of human tumor cells available for serial transplantation.
53. Label cells from step 52b with 1 mL of human specific EPCAM-e660 antibody (1:100), add 90 mL of cold FACS buffer and incubate for 20 min on ice covered with aluminum foil.
Optional: Also perform this and the following steps with the 10 mL saved for pre-sort flow cytometry from step 44.
54. Wash cells by adding 5 mL of cold FACS buffer and centrifuge at 1500 3 g for 5 min at 4 C. 55.Aspirate supernatant and suspend cells in 500 mL of cold FACS buffer and add 5 mL of DAPI (1:100).56.Perform flow cytometry analysis on these cells, the unstained cells from step 51 (control) and the pre-sort sample (optional).57.Calculate the number of human tumor cells in the sample by combining the percentage of EPCAM positive cells and the number of cells counted in step 52a.58.Calculate the volume of cells needed to obtain 25,000 human cells in 18 mL, which will be used for serial transplantation by intraductal injection in each mammary gland.
Note: At this point it is also possible to freeze down single-cell solution for later use.Only consider this when you have more than 500,000 cells.

EXPECTED OUTCOMES
Using this method, we were able to engraft primary DCIS lesions with a success rate of 88% at first passage, while a smaller percentage of 36% was successfully passaged serially as most DCIS lesions grow slow by nature (Figure 5).DCIS samples were injected into 1 to 9 mice depending on the sample size, with highly varying growth speeds between samples, ranging from palpable tumors within 6 months to microscopic lesions after 12 months.For more details see Hutten et al. 1 The DCIS-MIND models retain the features of the original tumor (e.g., growth pattern, ER, progesterone receptor and HER2 expression, as well as genomic aberrations) over multiple passages.As most DCIS-MIND models do not show palpable lesions, end-point analysis such as determining growth pattern and invasive progression is performed by histology and 3D whole mount imaging (Figure 6).Transplantable models can also be used to perform drug intervention studies, such as treatment of HER2-positive DCIS models with trastuzumab (Figure 7).Tumor free survival can be quantified for DCIS models with palpable tumors, otherwise it is possible to calculate extent of tumor growth at the end of treatment using H&E or whole mount analysis.

Protocol Potential solution
Provide estradiol in the drinking water of the mice a few days prior to intraductal injections.The estradiol results in enlarged nipples, which are easier to inject.

RESOURCE AVAILABILITY
Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Jos Jonkers, j.jonkers@nki.nl.

Materials availability
This study did not generate new unique reagents.

Figure 5 .
Figure 5. Engraftment rate of DCIS-MIND transplantation (A) percentage of successful engraftment of primary DCIS cells using MIND.(B) percentage of samples with successful serial transplantation.

Figure 6 .
Figure 6.DCIS MIND models retain features of the original tumors (A) H&E staining of DCIS samples from patients and corresponding DCIS-MIND models, showing the different matching growth patterns.(B) Maintenance of growth pattern over multiple passages.Black lines indicate concordance between passages, whereas orange lines indicate a discordance between passages.(C) Examples of immunohistochemistry for estrogen receptor (ER),progesterone receptor (PR), HER2 and Ki67 expression in MIND-DCIS lesions (Top row) versus matched primary DCIS lesions (Bottom row).(D) Maintenance of molecular subtype over multiple passages.Black lines indicate concordance between passages, whereas orange lines indicate a discordance between passages.E) H&E staining of DCIS-MIND models showing micro-invasion (left) or invasive growth (right).Blue arrows indicate micro-invasive cells.(F) Pie chart of the percentage of MIND-DCIS lesions with non-invasive growth, micro-invasion or invasive progression.(G) Maintenance of invasive or non-invasive growth over multiple passages.Black lines indicate concordance between passages, whereas orange lines indicate a discordance between passages.
Protocol P-buffer: mix 81 mL of 0.2 M Na 2 HPO 4 (28.39 g/L demineralized H 2 O, store at 20 C, sterile) with 19 mL of 0.2 M NaH 2 PO 4 (24 g/L demi H 2 O, store at 20 C, sterile) and add 100 mL of demineralized H 2 O. Adjust pH to 7.4 if necessary.
(Continued on next page)MATERIALS AND EQUIPMENTCRITICAL: ADDF+++ can be stored at 4 C for a maximum of 6 months CRITICAL: Digestion solution is prepared freshly immediately before usage Total N/A 10 mL STAR Protocols 4, 102526, September 15, 2023 CRITICAL: P-buffer can be stored at 4 C for a maximum of 6 months L-lysine in P-buffer: 5.848 g in 200 mL P-buffer.Adjust pH to 7.4 if necessary.CRITICAL: L-lysine in P-buffer can be stored at 4 C for a maximum of 6 months STEP-BY-STEP METHOD DETAILS Dissociation of primary DCIS tissue Timing: 16-18 h Timing: 5-10 min (for steps 1 and 2) Timing: Overnight $16 h (for step 3) Timing: 45-60 min (for steps 4-12)