Generation of 3D ex vivo mouse- and patient-derived glioma explant slice model for integration of confocal time-lapse imaging and spatial analysis

Summary

Development of spatial-integrative pre-clinical models is needed for glioblastoma, which are heterogenous tumors with poor prognosis. Here, we present an optimized protocol to generate three-dimensional ex vivo explant slice glioma model from orthotopic tumors, genetically engineered mouse models, and fresh patient-derived specimens. We describe a step-by-step workflow for tissue acquisition, dissection, and sectioning of 300-μm tumor slices maintaining cell viability. The explant slice model allows the integration of confocal time-lapse imaging with spatial analysis for studying migration, invasion, and tumor microenvironment, making it a valuable platform for testing effective treatment modalities.

For complete details on the use and execution of this protocol, please refer to Comba et al. (2022).¹

Graphical abstract

Highlights

• •Detailed protocol of 3D ex vivo explant slices from mouse- and patient-derived gliomas
• •Description of tumor tissue sectioning and culture under physiological conditions
• •Integrated procedure to analyze tumor migration and the role of new therapeutic targets
• •This explant model allows histopathological, spatial TME, and treatment efficacy studies

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Development of spatial-integrative pre-clinical models is needed for glioblastoma, which are heterogenous tumors with poor prognosis. Here, we present an optimized protocol to generate three-dimensional ex vivo explant slice glioma model from orthotopic tumors, genetically engineered mouse models, and fresh patient-derived specimens. We describe a step-by-step workflow for tissue acquisition, dissection, and sectioning of 300-μm tumor slices maintaining cell viability. The explant slice model allows the integration of confocal time-lapse imaging with spatial analysis for studying migration, invasion, and tumor microenvironment, making it a valuable platform for testing effective treatment modalities.

Before you begin

The protocol below describes the specific steps to perform ex-vivo analysis in explant slices from mouse and patient-derived glioma models. This is an integrated method that enables the use of explant slices for time-lapse confocal imaging to evaluate tumor cell migration within the tumor core and at the tumor border. Moreover, a detailed mathematical method was used to evaluate single cellular dynamics and cellular pair-wise correlations from time lapse imaging. The preclinical mouse models, genetic or orthotopic, are a powerful strategy to evaluate the role of new targets in glioma migration. The GFP expression in tumor cells growing in TdTomato expressing parenchyma allows the evaluation of the tumor border and normal tissue invasion. Furthermore, the staining of patient-derived glioma explants with Calcein AM facilitates the analysis of cell viability and evaluation of patterns of migration using time-lapse confocal imaging. This explant slice model enables post-imaging spatial analysis from the same tissue including H&E histo-pathological and morphological tests, TME studies using immuno-histochemistry and multiplex analysis to evaluate stromal cells, immune cells, and tumor extracellular matrix composition and dysregulation. Additionally, explants are suitable to evaluate the response to standard of care treatment alone or in combination with novel anti-tumor therapies.

Institutional permissions

All animal work must adhere to the relevant jurisdictions be they institutional, local, and/or federal. All work described herein was performed under the approval of the Institutional Animal Care and Use Committee (IACUC) at the University of Michigan. For this protocol, patient informed consent was obtained prior to tissue resection as required by the Institutional Review Board–approved protocols at the University of Michigan (HUM00057130 and HUM00024610).

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
CD3ε (D7A6E™) XP® mAb (1:400 dilution)	Cell Signaling	Cat# 85061S
CD45 (Intracellular Domain) (D9M8I) XP® mAb IgG (1:1000 dilution)	Cell Signaling	Cat# 13917S
Anti-KI67 pAb (1:400 dilution)	Abcam	Cat# ab15580
Recombinant Anti-Iba1 antibody [EPR16588] mAb (1:2000 dilution)	Abcam	Cat# ab178846
Anti-Nestin pAb (1:800 dilution)	Novus Biologicals	Cat# NB100-1604
Anti-SOX2 pAb (1:200 dilution)	Abcam	Cat# ab97959
Anti-Glial Fibrillary Acidic Protein Antibody pAb (1:500 dilution)	Sigma Aldrich	Cat# AB5541
GFP Antibody pAb (1:500 dilution)	Rockland Immunochemicals	Cat# 600-101-215
Cleaved Caspase-3 (Asp175) pAb (1:400 dilution)	Cell Signaling	Cat# 9661S
Anti-CD68 antibody pAb (1:1000 dilution)	Abcam	Cat# ab125212
Donkey anti-Mouse IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 647 (1:1000 dilution)	Invitrogen	Cat# A31571
Donkey anti-Rabbit IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 555 (1:1000 dilution)	Invitrogen	Cat# A31572
Alexa Fluor® 647 AffiniPure Donkey Anti-Chicken IgY (IgG) (H+L) (1:1000 dilution)	Jackson Immunoresearch	Cat# 703-605-155
Donkey anti-Goat IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor™ Plus 488 (1:1000 dilution)	Invitrogen	Cat# A32814TR
Alexa Fluor® 488 AffiniPure Donkey Anti-Chicken IgY (IgG) (H+L) (1:1000 dilution)	Jackson Immunoresearch	Cat# 703-545-155
	Jackson Immunoresearch	Cat# 703-585-155
Biological samples
Human high-grade glioma (HGG) samples	University of Michigan Hospital	N/A
Chemicals, peptides, and recombinant proteins
Hoechst 33342, Trihydrochloride, Trihydrate	Invitrogen	Cat# H3570
Calcein, AM	Invitrogen	Cat# C1430
D-Luciferin potassium salt	MediLumine	Cat# 222PS
In vivo jetPEI®	Polyplus Transfection	Cat# 201-50G
200 proof ethanol	Decon Laboratories	Cat# 2701
Paraformaldehyde	Sigma-Aldrich	Cat# P6148-1KG
Sodium hydroxide pellets	Fisher Scientific	Cat# S320-1
DPBS supplemented with magnesium and calcium	Gibco	Cat# 14040-133
DPBS without calcium or magnesium anhydrous	Gibco	Cat# 2376618
DMEM high-glucose without phenol red	Gibco	Cat# 21063-029
DMEM F-12	Gibco	Cat# 11330-032
Neurobasal™ medium	Gibco	Cat# 21103-049
HBSS	Gibco	Cat# 14175-095
DPBS without calcium or magnesium	Gibco	Cat# 14190-144
FBS	Biowest	Cat# S1620
Insulin, human recombinant	MilliporeSigma	Cat# 91077C
2-Mercaptoethanol 1000×	Gibco	Cat# 21985-023
50× B-27™ Supplement	Gibco	Cat# 17504-044
100× N-2 Supplement	Gibco	Cat# 17502-048
MEM NEAA 100×	Gibco	Cat# 11140-050
GlutaMAX™ 100×	Gibco	Cat# 35050-061
EGF	Peprotech	Cat# AF-100-15-1MG
FGF	Peprotech	Cat# 100-18B-1MG
Bovine serum albumin (BSA)	Sigma-Aldrich	Cat# A2153-100G
100× Antibiotic-Antimycotic (Anti-Anti)	Gibco	Cat# 15240-062
Normocin™	Invivogen	Cat# ANT-NR-1
StemPro™ Accutase™ Cell Dissociation Reagent	Gibco	Cat# A1110501
Laminin	Gibco	Cat# 23017015
Matrigel®	Corning	Cat# 354230
Agarose (molecular grade, low melting point)	Fisher Bioreagents	Cat# BP165-25
Compressed gas 95% oxygen and 5% carbon dioxide	Cryogenic Gases	Cat# CM-2-1
Compressed gas 100% oxygen	Cryogenic Gases	Cat# OXYE-OP-50
Hydrochloric acid 36.5%–38.0%	Fisher Chemical	Cat# A144500
Ketamine / Ketaset™ (120 mg/kg body weight)	Zoetis; MWI Veterinary Supply	Cat# 000680
Dexmedetomidine / Dexdmonitor™ (0.5 mg/kg body weight)	Zoetis	N/A
Atipamezole / Antisedan™ (1.0 mg/kg body weight)	Zoetis; MWI Veterinary Supply,	Cat# 32800
Isoflurane / Fluriso™ (1%–5%)	VetOne	Cat# 501017
Carprofen / Rimadyl™ (5 mg/kg body weight)	Zoetis	N/A
Buprenorphine (0.1 mg/kg body weight)	Par Pharmaceutical; MWI Veterinary Supply	Cat# 060969
Experimental models: Cell lines
NPA glioma neurospheres (RTK/RAS/PI3K overactivation, knockdown of p53-GFP and ATRX-GFP)	Nunez et al., 2019; Calinescu et al., 2015, Comba et al., 2020; Comba et al., 2022.	N/A
NPD glioma neurospheres (RTK/RAS/PI3K overactivation, knockdown of p53-GFP, and overexpression of PDGFβ-GFP ligand)	Comba et al., 2020; Comba et al., 2022	N/A
Experimental models: Organisms/strains
C57BL6 6–8-week-old males and females’ mice	Taconic Biosciences	Cat# B6NTac
ROSAmT/mG B6.129(Cg)-Gt(ROSA)26Sortm4(ACTB-tdTomato,-EGFP)Luo/J- 6–8 week old male and females’ mice	Jackson Laboratory	Cat# 007676
Recombinant DNA
pT/CAGGS-NRASV12 (NRASG12V Overexpression)	Addgene	Cat# 20205
pT2/shp53/GFP4 (p53 Knockdown)	Addgene	Cat# 20208
pT2/shAtrx-GFP4 (Atrx Knockdown)	Addgene	Cat# 124259
pT2-shp53-PDGFb-GFP4 (p53 Knockdown, PDGFβ ligand overexpression)	N/A	N/A
Software and algorithms
Image J	N/A	https://imagej.net/software/fiji/downloads
TrackMate plugin on Image J	N/A	https://imagej.net/software/fiji/downloads
Julia	N/A	https://julialang.org/downloads/
RStudio	N/A	https://support--rstudio-com.netlify.app/products/rstudio/download/
Other
Forceps	Fine Surgical Tools	Cat# 11000-12
Small surgical scissors	Fine Surgical Tools	Cat# 14060-09
Large surgical scissors	Fine Surgical Tools	Cat# 14002-13
Scalpel handle	Stoelting	Cat# 52171P
Carbon steel #15 scalpel blade	Exelint International	Cat# 29532
Rongeours	Fine Science Tools	Cat# 16021-14
Gas anesthesia system	Caliper Life Sciences	Model No. XGI-8
Anesthesia vaporizer	Midmark	Model No. Matryx™ VIP 3000
Model 701 10 μL syringe	Hamilton Company	Cat# 7635-01
30-gauge 1.25-inch point style needle	Hamilton Company	Cat# 7803-07
33-gauge 1.5-inch point style needle	Hamilton Company	Cat# 7803-05
Digital lab standard rat stereotaxic	Stoelting	Cat# 51900
Mouse and neonate adapter	Stoelting	Cat# 51625
Quintessential stereotaxic injector	Stoelting	Cat# 53311
150W Dual gooseneck illuminator	Stoelting	Cat# 59259
Digital just for mouse stereotaxic instrument	Stoelting	Cat# 51730D
Rodent warmer control box X1	Stoelting	Cat# 53800
Ideal micro-drill	Cellpoint Scientific	Cat# 67-1204
Stemi 2000 stereomicroscope	Zeiss	Cat# 455052-0000-000
Fiber-Lite® high intensity LED illuminator	Dolan-Jenner	Cat# MILEDUSB1
Tissue culture laminar flow hood	The Baker Company, Sterigard III Advance	Model SG403
Vibratome	Leica	Model VT1000 s
Mini perforated spoon	Moria	Cat# 10370-19
Single-sided razorblade	American Line	Cat# 66-0362
Double-sided razorblade	Personna	Cat# 74-0002
Hot water bath	Labcare America	Cat# 51221048
10 mL serological pipet	GenClone	Cat# 12-104
PIPETMAN Classic P1000 micropipette	Gilson	Cat# F123602
PIPETMAN Classic P200 micropipette	Gilson	Cat# F123601
PIPETMAN Classic P10 micropipette	Gilson	Cat# F144802
0.45 μm syringe filter	GVS	Cat# FJ25BSCPS0004AL01
30 mL syringe	BD	Cat# 309651
6-well plate	GenClone	Cat# 25-105
Millicell cell culture insert, 30 mm, hydrophilic PTFE, 0.4 μm	Millicell	Cat# PICM0RG50
50 mL conical tubes	CellPro	Cat# CW5603
10 cm Petri dish	Olympus Plastics	Cat# 25-202
Cyanoacrylate adhesive (Krazy™ Glue)	N/A	N/A
Peel away truncated 12 × 12 mm embedding molds	Electron Microscopy Sciences	Cat# 70181
Bucket of ice	N/A	N/A
CO2 water-jacketed cell culture incubator	Forma Scientific	Model 3110 Series II
S1 Pipet filler	Thermo Scientific	Cat# 9521
1000 μL filter tip	Globe Scientific Inc.	Cat# 150835
200 μL barrier tip	Olympus Plastics	Cat# 24-412
10 μL reach barrier tip	Olympus Plastics	Cat# 24-401
IVIS® Spectrum	PerkinElmer	Cat# 124262
Stereomicroscope	Olympus	Model No. SZX16
Light engine	Lumencor	Model No. Sola SM 5-LCR-SB
Nunc glass base dish, 27 mm	Thermo Scientific	Cat# 150682
LSM 880 confocal laser scanning microscope with Airyscan. The scan-head is attached to a Zeiss Axio Observer Z1 inverted microscope with motorized, programmable x,y,z stage, additional piezo z-stage, joystick and touch panel docking control station. A definite focus (Version 2) attachment corrects for focus drifts during time lapse experiments. Equipped with temperature and CO2 control chambers.	Zeiss, Germany	Model No. 880
Disposable base molds	Fisher Scientific	Cat# 22363554
Swingsette™ tissue processing/embedding cassettes	Simport	Cat# M515-2
Tissue Path™ MACROSETTE™ processing/embedding cassettes	Fisherbrand	Cat# 15182706
Rectangular foam biopsy pads	Fisherbrand	Cat# 22038221
Microspatula	Fisherbrand	Cat# 21-401-10
Fully enclosed tissue processor	Leica, Germany	Model No. ASP 300S
Embedding system cooling module HistoCore Arcadia C and Automatic paraffin dispenser HistoCore Arcadia H	Leica Biosystems, Germany	Model Nos. Arcadia C and Arcadia H
Instant sealing sterilization pouch	Fisherbrand	Cat# 0181255
Milli-Q® water purification system	Millipore	Cat# Z00QSV001
Timer	N/A	N/A
Thermometer	N/A	N/A

Materials and equipment

Primary Mouse Culture Medium

Reagent	Final concentration	Amount
DMEM F12	74%	37 mL
FBS	25%	12.5 mL
100× Anti-Anti Solution	1%	500 μL
Total		50 mL

Store at 4°C for up to two weeks.

Secondary Mouse Culture Medium

Reagent	Final concentration	Amount
DMEM F12	95.6%	478 mL
50× B-27™ supplement	1%	5 mL
100× N-2 supplement	2%	10 mL
Normocin™	0.2%	1 mL
100× Anti-Anti	1%	5 mL
1000× EGF	60 ng/mL	500 μL
1000× FGF	60 ng/mL	500 μL
Total		500 mL

Store at 4°C for up to two weeks.

Human Culture Medium

Reagent	Final concentration	Amount
Neurobasal medium	47.44%	237.188 mL
DMEM-F12	47.44%	237.188 mL
50× B-27™ supplement without vitamin A	1%	5 mL
100× N-2 supplement	2%	10 mL
100× GlutaMAX™	1%	5 mL
100× Anti-Anti	1%	5 mL
1000× 2-mercaptoethanol	0.1%	500 μL
Insulin	2.5 μg/mL	125 μL
Total		500 mL

Store at 4°C for up to two weeks.

4% Paraformaldehyde

Reagent	Final concentration	Amount
Milli-Q® water	N/A	1 L
Paraformaldehyde	40 g/mL	80 g
Sodium hydroxide	N/A	1 pellet
2× DPBS	50%	1 L
1 N Hydrochloric acid	N/A	Depends on how much is needed to reach a pH of 7.4 for any given batch.

Note: Combine the Milli-Q® water, paraformaldehyde, and sodium hydroxide on a heated magnetic stirrer in a chemical fume hood. Once dissolved, filter the solution into the 2× DPBS and adjust the pH to 7.4 dropwise with the hydrochloric acid.

Store at 4°C for up to one month.

CRITICAL: Paraformaldehyde causes eye, respiratory, and dermatologic damage. Always manipulate it in a chemical safety cabinet or laminar flow hood while wearing proper PPE such as gloves and eye protection. If the temperature reaches 70°C, paraformaldehyde combusts, therefore the temperature must be closely monitored.

• •30 mg/mL Luciferin solution: dissolve 1 g D-Luciferin potassium salt in 33.33 mL DPBS without calcium or magnesium.

Store at −20°C as 1 mL aliquots for up to 6 months.

• •Milli-Q® water: Use the Milli-Q® water purification system according to manufacturer recommendations (https://www.emdmillipore.com/US/en/product/Milli-Q-Direct-Water-Purification-System, MM_NF-C85358).
• •1 N Hydrochloric acid: combine 8.3 mL of hydrochloric acid (36.5%–38.0%) with 91.7 mL of Milli-Q® water.

Store in a proper chemical safety cabinet for up to one month.

• •2× DPBS: dissolve 38.2 g anhydrous DPBS in 1000 mL of Mili-Q® water.

Store at 20°C–25°C for up to one month.

• •70% ethanol: combine 700 mL 200 proof ethanol and 300 mL Milli-Q® water.

Store at 20°C–25°C for up to one month.

• •100× laminin coating solution stock: dissolve 1 mg of laminin in 1 mL DPBS supplemented with Ca²⁺ and Mg²⁺.

Store at −80°C for up to one year.

• •1% (v/v) laminin coating solution: combine 1 mL of DPBS supplemented with Ca2+ and Mg2+ and 10 μL of 100× laminin for a final laminin concentration of 10 μg/mL.
• •Matrigel® solution: combine 1 part Matrigel® to 2 parts human culture medium.

Store at −80°C as 100 μL aliquots for up to one year.

• •Sectioning medium: add 5 mL of 100× Anti-Anti solution to 500 mL of DMEM high glucose without phenol red.

Store at 4°C for up to two weeks.

• •4% (m/v) agarose: add 4 g molecular grade, low melting point agarose to 100 mL HBSS. Dissolve by swirling and heating in the microwave until light boiling is observed and the solution is transparent.

Store at 4°C for up to two weeks.

CRITICAL: If using a container with a cap or lid, make sure to loosen it so the container does not build up pressure while heating to prevent container breakage and injury.

• •1000× FGF/FGF stock solutions: Add 53 mg of bovine serum albumin (BSA) to 53 mL of DPBS to form a 0.01% BSA solution. Mix vigorously by vortexing and filter through a 0.45 μm filter on a 30 mL syringe. Dissolve 1 mg EGF or FGF in the 1 mL of filtered 0.01% BSA. Aliquot into 600 μL Eppendorf tubes.

Store at −20°C for up to 6 months.

• •2% (m/v) agarose: add 2 g molecular grade, low melting point agarose to 100 mL HBSS. Dissolve by swirling and heating in the microwave until light boiling is observed and the solution is transparent.

CRITICAL: If using a container with a cap or lid, make sure to loosen it so the container does not build up pressure while heating to prevent container breakage and injury.

• •1 mM Calcein AM stock solution: dissolve 1 mg Calcein in 1 mL DMSO.

Store at −20°C as 10 μL aliquots for up to 6 months.

• •2 μM Calcein solution: add 10 μL of the 1 mM Calcein stock solution to 5 mL of secondary mouse culture medium.
• •Hoechst 33342: dilute stock Hoechst 33342 1:4000 in secondary mouse culture medium.
• •Autoclaving: the day before procuring the samples, place all autoclavable equipment in a new instant sealing sterilization pouch and place the sealed pouch in an autoclave to be sterilized.
• •Laminar flow hood setup: the night before procuring the samples, place the autoclaved Crosstex bag full of equipment along with all non-autoclavable in the laminar flow hood, thoroughly spray everything with 70% (v/v) ethanol, and expose to UV light overnight.
• •Cell culture incubator: ensure that the cell culture incubator is at 37°C, is humidified and has an atmosphere of 21% O₂ and 5% CO₂.
• •Water bath. Bring the water bath to 40°C.
• •Confocal microscope: 1 h before beginning time-lapse imaging, turn on the microscope, lasers, electronic stage, humidified stage top incubator to 37°C with 5% CO₂, and Zen Black.

Step-by-step method details

Tumor generation in genetically engineered mouse model (GEMM) of glioma

Timing: 2 h (30 min for plasmids mix and 1.5 h for a cage of 5 mice)

The Sleeping Beauty (SB) transposon system was used to generate a GEMM of glioma by inducing genetic modifications into the neural precursor cells of the subventricular zone in neonatal pups. This method was used as previously described.^1,2,3,4,5

• 1.
  Anesthetize one-day-old C57BL/6 mouse pups via hypothermia for 2 min.

  • a.
    Place them in a stereotaxic apparatus to orthotopically inject the plasmid combination encoding the desired genetic modifications into the subventricular zone to induce tumorigenesis.
  • b.
    Delivery of plasmids is achieved with in vivo DNA transfection agent jetPEI®.

Note: The genetic alterations include: RTK/RAS/PI3K overactivation (NRAS-G12V), p53 knockdown, and ATRX knockdown for the NPA model, and RTK/RAS/PI3K overactivation (NRAS-G12V), p53 knockdown, and PDGFβ ligand overexpression for NPD model. The chosen genetic models include the use of the following plasmids: (1) pT2C-LucPGK-SB100×, (2) pT2-NRAS-G12V; (3) pT2-shP53-GFP4; (4) pT2-shAtrx-GFP4; (5) pT2-shP53-Pdgfβ-GFP4.

• 2.To evaluate the evolution of the tumor in vivo, each week, use IVIS® Spectrum imaging system to measure bioluminescence.

Note: Tumors generated with the SB system can be initially detected 30–40 days post injection, for NPA model, and 20–30 days post injection, for NPD model.

• 3.Before an animal’s tumor burden reaches the humane endpoint, as defined by the IACUC, or when obtaining an IVIS signal of 10⁶ to 10⁷ photon/seconds, the animal should be euthanized and its brain removed.

CRITICAL: Tumor burden is assessed with behavioral monitoring (hunched posture, scruffy fur, lack of ambulation, and/or periocular porphyrin staining). For time lapse scanning imaging, GEMM tumors should reach the moribund stage. Consistent weekly imaging using IVIS is crucial.

CRITICAL: Explants obtained from mice in the latest stage will exhibit tumor necrosis and will not display cell migration and proliferation.

• 4.For brain acquisition proceed to step (15).
• 5.Use an epifluorescent stereomicroscope to confirm the presence of a tumor before proceeding to the next step.

Note: Tumors generated with the SB system can be dissected using an inflorescent stereomicroscope and dissociated to generate glioma neurospheres which are maintained through culture and used for further orthotopic implantation.

Tumor initiation via orthotopic implantation of glioma neurospheres

Timing: 30 min per animal implanted

Intracranial orthotopic implantation of glioma neurospheres obtained from the above mentioned GEMM was performed as previously described.^1,5

• 6.Dissociate the cultured neurospheres with StemPro Accutase and resuspend them in DMEM F-12 without FBS or supplements, to a final concentration of 50,000 cells/μL.
• 7.Use 6–8 –week-old C57BL/6 mice or ROSAmT/mG mice for orthotopic intracranial stereotactic implantation.

Note: ROSAmT/mG mice express TdTomato protein in all cells. This characteristic enables the visualization of GFP+ tumor cells migrating into the normal brain parenchyma.

• 8.After anesthetizing the mice with a 100 μL intraperitoneal injection of ketamine at 120 mg/kg and dexmedetomidine at 0.5 mg/kg, give them preemptive analgesia by a 100 μL subcutaneous injection of carprofen at 5 mg/kg and shave their heads before securing them in a stereotaxic apparatus.
• 9.Retract the scalp and drill a burr hole 1 mm anterior and 1.5 mm lateral relative to bregma.
• 10.Inject 50,000 NPA or NPD cells suspended in 1 μL of DMEM F-12 without FBS or supplements into a pocket spanning 3 mm–3.5 mm deep in the striatum.
• 11.Use bone wax and sutures to seal the skull and scalp, respectively, before reversing the anesthesia with a 100 μL intraperitoneal injection of atipamezole at 1.0 mg/kg and additional analgesia with a 100 μL subcutaneous injection of buprenorphine at 0.1 mg/kg.
• 12.Observe the mice until they are bright, alert, and responsive before returning them to normal housing. Usual recovery time is 30 min.
• 13.Before animal tumor burden reaches the humane endpoint, as defined by the IACUC, or when obtaining an IVIS signal of 10⁶ to 10⁷ photon/second, the animal should be to be euthanized and its brain removed.

CRITICAL: Tumor burden is assessed with behavioral monitoring (hunched posture, scruffy fur, lack of ambulation, and/or periocular porphyrin staining). For time lapse-imaging, mice should nearly reach the moribund stage. Consistent weekly imaging using IVIS is crucial. Explants from mice euthanized too late will exhibit tumors with necrosis and will not display cell migration and proliferation.

Preparation prior to sample acquisition

Timing: 1 h

Preparation of cell culture inserts, media for tissue collection and agarose for tissue embedding.

• 14.Human: At least 1 h before procuring the patient-derived tumor samples, pipet 100 μL of the Matrigel® solution (1 volume of Matrigel® mixed with 2 volumes of cell culture media) onto cell culture inserts and incubate for 1 h. After that, add 1 mL of Secondary mouse culture medium to avoid desiccation of the Matrigel®.

Note: To use the patient-derived tumor explants as a high-throughput platform for drug screening, tissue can be placed on 24- to 96-well-plate inserts.

• 15.Mouse: At least 1 h before procuring the mouse samples, pipet 1,000 μL of the laminin coating solution onto cell culture inserts in a six well plate and incubate for 1 h.
• 16.Melt the 4% (m/v) agarose in the microwave until it has begun to boil and becomes a transparent liquid, approximately 30 s.

CRITICAL: If using a container with a cap or lid, make sure to loosen it so the container does not build up pressure while heating to prevent container breakage and injury.

CRITICAL: The lid cannot be removed, only loosened, to maintain the sterility achieved by boiling the agarose.

• 17.Place the melted 4% (m/v) agarose in the 40°C water bath to prevent solidification.
• 18.Pipet 50 mL of human culture medium into a 50 mL conical tube and warm it to 37°C.
• 19.In the laminar flow hood, place the container of sectioning medium in the ice bucket. Using plastic tubing and a sterile 10 mL pipette, oxygenate the medium with 95% O₂ and 5% CO₂ gas for at least 20 min.
• 20.Prepare 30 mL of the cold, oxygenated human culture medium into a 50 mL conical tube for sample collection. Place tube in an ice bucket until collection.

Dissection and acquisition of mouse brain

Timing: 20 min per mouse

The following steps describe how to perform the dissection and acquisition of a mouse brain post tumor intracranial orthotopic surgery or a GEMM.

• 21.As indicated previously, select a mouse post intracranial orthotopic surgical implantation or a GEMM of glioma mouse with an IVIS signal between 10⁶ to 10⁷ photon/sec.
• 22.
  Put the mouse under anesthesia via inhalation using Fluriso™.

  • a.
    Place the mouse in the gas anesthesia system anesthesia chamber and turn the 100% oxygen tank on to between 345 and 448 kPa.
  • b.
    Adjust the airflow to the anesthesia chamber to 1.5 L per minute.
  • c.
    Set the anesthesia vaporizer to 0.25%.
  • d.
    Wait 1–2 min for the mouse to lose consciousness.
  • e.
    Turn off the oxygen, airflow, and vaporizer before removing the mouse.
• 23.
  Mouse brain resection.

  • a.
    Place the mouse on a sterile surface in the biological safety cabinet.
  • b.
    Spray the head and neck of the mouse with 70% ethanol to wet the fur.
  • c.
    Euthanize the mouse via decapitation with the large surgical scissors.

    CRITICAL: Minimize time between euthanasia and vibrosectioning to prevent cell death. Place the head in a 10 cm Petri dish, covered, and transport it on ice to the laminar flow hood.

  • d.
    With the large surgical scissors, open the scalp, moving rostral to caudal.
  • e.
    Dissect the brain using rongeurs.

    CRITICAL: Only squeeze until separation of the frontal bones is observed upon breaking the interfrontal suture. Squeezing too much will damage the brain.

  • f.
    With the rongeurs, remove any muscle and cervical vertebrae from the caudal side of the head. Begin removing the bones of the skull from rostral to caudal.
  • g.
    Once the dorsal surface of the brain is fully exposed, remove it from the skull by cutting the cranial nerves on the ventral side with the small surgical scissors and gently dislodging the olfactory bulbs with forceps.

Human high grade glioma sample acquisition

Timing: 30 min of transportation between operating room and laboratory

Patient-derived brain tumor samples obtained at the operation room.

• 24.Patient-derived glioma tumor samples are acquired at the moment of the surgery by specialized neurosurgeons.
• 25.Avoid necrotic regions of the sample.
• 26.Immediately place the samples in cold, oxygenated human culture medium on ice for transportation to the laboratory for processing; limiting the time spent at 20°C–25°C increases tissue viability.

CRITICAL: Limiting the time between tissue resection and culturing increases model viability.

Embedding the mouse or human tumor specimen in 4% agarose

Timing: 30 min

Mouse brain tumor tissue or patient-derived tumor samples are embedded in agarose to generate a semi rigid block to facilitate sectioning.

• 27.Place a 10 cm Petri dish on ice and add cold, oxygenated sectioning medium to submerge the sample. Transfer the sample to the 10 cm Petri dish.
• 28.Transfer the 4% (m/v) agarose to the laminar flow hood and pipette enough into a truncated 12 × 12 mm embedding mold to fill it, approximately 10 mL, avoiding bubbles, and leaving enough room to submerge the sample without causing the agarose to overflow.

CRITICAL: With the thermometer, ensure the agarose is less than or equal to 37°C before embedding to prevent damaging the tissue sample.

• 29.
  With the forceps, submerge the sample in the agarose. Hold it in place with a 10 μL pipet tip until the agarose has cooled and become viscous enough to hold the tissue in place but has not yet solidified so the pipet tip can be removed.

  • a.
    Mouse.

    • i.
      For coronal sectioning of a mouse brain, the rostral-caudal axis should be perpendicular to the blade on the vibratome and the specimen dish.

      Note: In the 12 × 12 mm truncated mold, it must therefore be oriented vertically with the olfactory bulbs facing down.

  • b.
    Human.

    • i.
      For sectioning patient-derived tumor tissue, orient the sample so that the area of each vibrosection is maximized by ensuring the longest axis is parallel to the blade on the vibratome.
• 30.Place the truncated 12 × 12 mm embedding mold on ice until that agarose is fully solidified, about 5 min.
• 31.Cut the corners of the truncated 12 × 12 mm embedding mold with the single sided razorblade and remove the agarose block.
• 32.Glue the agarose block to the specimen disc with cyanoacrylate adhesive and allow it to set for 5–8 min on ice. To make coronal sections of a mouse brain, the olfactory bulbs must be face up with the ventral side of the brain towards the blade.

CRITICAL: if the tissue fell out of the desired orientation during agarose solidification, trim the bottom of the agarose block at an angle with a single sided razor blade to account for the tissue misalignment. The agarose block, depending on how much agarose was added to the truncated 12 × 12 mm embedding mold, may need to be trimmed regardless of tissue orientation so the blade fixture on the vibratome can advance.

• 33.Place a 10 cm Petri dish on ice in the laminar flow hood and fill it with cold, oxygenated sectioning medium. This is where freshly taken sections will be placed.

CRITICAL: No more than 90 min should elapse between tissue resection and culturing to limit excessive cell death.

Sectioning with vibratome

Timing: 30–60 min depending on tissue type

Mouse brain tumors and patient-derived samples are sectioned using an automatic vibratome.

• 34.Fasten the buffer tray to the vibratome and fill the cooling bath outside of the buffer tray with ice.
• 35.Clean the double-sided razorblade first with acetone and then with 70% (v/v) ethanol. Carefully break the razorblade in half and clamp half of the blade to the knife holder. Fasten the knife holder to the vibratome.

CRITICAL: If sectioning multiple samples, change the razorblade between them.

• 36.Fill the buffer tray with enough cold, oxygenated sectioning medium to submerge the agarose block. Run surgical tubing into the laminar flow hood from the tank of 95% O₂ and 5% CO₂ gas and oxygenate the cold sectioning medium in the buffer tray using a 1000 μL pipet tip on the end of the surgical tubing.
• 37.Fasten the specimen disc with the attached agarose block to the buffer tray.
• 38.Set the section thickness to 250–300 μm on the vibratome.
• 39.Adjust the start and stop points of blade advancement according to the position of agarose block. Set the knife feed speed and knife sectioning frequency. Utilize continuous, automatic blade advancement.
• 40.Begin sectioning, transferring the tissue sections to the 10 cm Petri dish on ice containing the sectioning medium using the mini perforated spoon. Separate the agarose from the tissue with forceps.
• 41.Repeat steps 38–39 until all the desired tissue is sectioned. For mouse tissue, continue. For patient-derived tumor tissue, jump to step 49 under ‘Staining of patient-derived tumor samples.’
• 42.For mouse tissue, use an epifluorescent stereomicroscope to select only the vibrosections that contain GFP+ tumor. Use the microscope to determine which sections bear tumor and mark them by annotating the lid of the 10 cm Petri dish with a marker.
• 43.Remove the six well plate containing the cell culture inserts and the 50 mL conical tube containing the primary mouse culture medium from the cell culture incubator and place them in the laminar flow hood.
• 44.Pipette off the laminin solution and wash the laminin-coated cell culture inserts once with DPBS supplemented with magnesium and calcium then twice with the warmed primary mouse culture medium.
• 45.
  Transfer the explants to the laminin-coated cell culture inserts.

  • a.
    Pipet enough primary mouse culture medium into the insert to barely cover the explants.
  • b.
    Pipet 2 mL of medium into the bottom of the well outside of the cell culture insert.

CRITICAL: adding too much medium will cause the explants to float, not adhere to the laminin coating, and make time-lapse imaging impossible. Ensure that tissue is covered by medium.

• 46.Place the six well plate containing the vibrosections plated on the laminin-coated cell culture inserts back in the cell culture incubator at 37°C.
• 47.After 6–18 h, transfer the cell culture insert to be imaged to a glass bottom dish using sterile forceps in a laminar flow hood and change the media to secondary mouse culture medium. Add 2 mL per dish and keep it in the incubator for 20 min to stabilize.
• 48.Transfer the cell culture inserts to the confocal microscope for time-lapse imaging.

Staining of patient-derived tumor samples for tumor viability and visualization analysis

Timing: 1.5 h

Patient-derived brain tumor samples are stained with Hoechst and Calcein for tumor viability analysis and fluorescence microscope migration studies.

• 49.Start the staining of the explants with Hoechst 33342 and Calcein AM in a six well plate without the inserts.
• 50.Incubate the explants for 1.5 h in Hoechst (1:4000) and 2 μM Calcein then wash with warm human culture medium three times.
• 51.Wash Matrigel®-coated cell culture inserts with warm human culture medium three times.
• 52.In the laminar flow hood, transfer the explants to the Matrigel®-coated cell culture inserts and pipet enough secondary mouse culture medium to barely cover the explants. Pipet 2 mL of medium into the bottom of the well outside of the cell culture insert.

CRITICAL: adding too much media will cause the explants to float, not adhere to the Matrigel® coating, and make time-lapse imaging impossible. Conversely, too little media will lead to tissue death.

Time-lapse single photon laser scanning confocal imaging of tumor cell migration

Timing: 45 min (The time for setting the microscope could vary depending on the tissue; bigger tissue size and better quality requires more imaging sites. Time lapse confocal imaging can be performed from 24 to 72 h)

Explants slices are incubated in a humidified, temperature and CO2 controlled incubation chamber included as part of a laser scanning confocal microscope for time lapse imaging.

• 53.This methodology is detailed for an inverted Zeiss LSM880 laser scanning confocal microscope with AiryScan (Carl Zeiss, Jena, Germany). Another similar confocal microscope could be used.
• 54.Start the microscope and set the temperature of the humidified incubation chamber to 37°C. Turn on the CO₂ and set it to 5%. Let temperature and CO₂ levels to equilibrate for 20 min.

CRITICAL: Utilize an extra container with sterile water placed into the big incubator chamber to maintain the humidity and prevent excessive media evaporation (especially important for longer imaging experiments).

• 55.Transfer the glass bottom dish with the tissue sections into the microscope incubator chamber. Allow media to equilibrate at 37°C and 5% CO₂ for 15 min.
• 56.Secure the glass bottom plate to avoid movement of the explant section during imaging.
• 57.After sections were maintained in the incubator for 15 min, you can start the microscope settings for time lapse confocal imaging.
• 58.Select the objective to use. For confocal imaging, explants are placed into a 27 mm Nunc glass base dish and on top of an insert.

Note: Under these conditions it is recommended to use an air 10×/0.3 EC Plan-Neofluar with a working distance of 5.2 mm or air 20×/0.4 LD Plan-Neofluar with a 7.9 mm working distance objective or similar.

• 59.Turn on the Argon 488 nm and the DPSS 561 nm excitation lasers, set the laser power for each laser equally for all positions.

CRITICAL: To avoid damage in the tissue and photobleaching, set laser power to the minimum level possible.

• 60.For GFP+ tumors and Tdtomato+ parenchyma, perform simultaneous excitation at 488 nm and 561 nm using scan mode: frame; frame size 1024 × 1024 or higher and averaging mode: line.
• 61.Use ocular observation and the locate function to select a position (x,y,z) of interest. Save each position. Different positions and different slice can be imaged at the same time.
• 62.To address the analysis of the explant in 3D, set the Z stack positions. Select the first (bottom) and the last (top) positions on focus for the explant. To do 2D image scanning, select an intermediate position between the extreme positions of the explants. This will avoid imaging the explant on the top or the bottom surface of the slices and will validate the depth of imaging.
• 63.Use the Time Series function to set the time intervals. Usually, >300 cycles at an interval of 10 min.
• 64.Set number of tile scan per location. Using a 10× objective, do not use tile scan but use 2 × 2 tile scan for a 20× objective.
• 65.Select Focus strategy to Autofocus Mode “Fluorescence” for every time and every position.
• 66.Use AutoSave imagining or Streaming function.
• 67.Start running the time lapse scanning imaging. Imaging can be performed for between 24 and 72 h.

Treatment response using glioma explants slices

Timing: 24–72 h (time will vary depending on the treatment)

Explant slices could be cultured in different plate size or multiwell plates for treatment analysis.

• 68.Explants can be used to study the response to drug or standard of care treatment.
• 69.Place explant on top of inserts in different wells as required depending on the experimental conditions and number of repetitions.

Note: For high-throughput screening of different compounds or treatments, the use of 96 well plates or higher is recommended.

• 70.Apply media with the corresponding treatment into each well.

Note: Explants under treatment can be used for the analysis of cell viability, apoptosis or even cell migration and invasion using confocal imaging.

• 71.After treatment, remove media and fix the tissue in 4% paraformaldehyde (PFA) for 24–48 h.

Tissue fixation

Timing: 24 h (tissue fixation could be performed between 24 and 48 h)

After the tissue has been used for time lapse confocal imaging or for drug screening and treatment response, explants can be fixed, cryopreserved and cryosectioned or paraffin embedded and micro-sectioned for further experimental approaches such as histopathological analysis using H&E staining, immunohistochemistry, immunofluorescence, spatial transcriptomic (RNA-Scope) and other desirable spatial and molecular methods (See Figure 3, 4, and 5).

• 72.After completion of image acquisition, remove the sample dish from the microscope platform.
• 73.Remove the secondary mouse culture medium from the cell culture inserts and dish and fill them with 4% PFA.

CRITICAL: PFA and its vapors are hazardous, always utilize proper personal protective equipment and perform actions under a chemical hood to avoid exposure of the eyes and skin.

• 74.Secure the lid of the 10 cm Petri dish with parafilm and leave the tissue at 4°C for 24–48 h.
• 75.In a chemical hood, use a pipette to remove as much of the 4% PFA as possible, discarding it properly. Wash the tissue sections with DPBS without Ca²⁺ or Mg²⁺ two times.

Pause point: If you cannot proceed to tissue embedding, you can leave the slices in DPBS at 4°C for 2–4 days.

• 76.Continue to either Tissue Processing for Cryosections or Agarose Embedding for Tissue Processing in Paraffin.

Figure 3.

Tissue processing for histopathological and tumor microenvironment spatial analysis

(A) Explants can be fixed after imagining or treatment evaluation is done. Media is removed and replaced with 4% paraformaldehyde (PFA) and, placed at 4°C for 48 h. Explant are pre embedded in 2% agarose, to be further embedded in paraffin.

(B) Following tissue fixation, tissue must be processed for dehydration (ethanol series), cleared (xylene), then embedded with paraffin.

(C) Tissue microsections of 5 μm are used for histopathological analysis (Hematoxylin and Eosin staining) and/or immunohistochemistry/immunofluorescence. Image shows representative 5 μm H&E sections of mouse NPA glioma explant slice and human glioblastoma explants showing tissue morphology after 4 days of culture. Scale bars: 50 μm.

Figure 4.

Experimental control of the tumor viability and analysis of the microenvironment landscape in glioma mouse explant models

(A) Immunofluorescence of explants sections that were never subjected to imaging (Control) compared with sections of explants imaged for 72 h. (A) Analysis of tumor viability. Green fluorescent protein (GFP) is labeling tumor cells. Left: KI67, marker of active cell proliferation (Red); Middle: CASP3, cleaved caspase 3, apoptosis marker (Red); Left: Survivin, anti-apoptotic marker (Red). Top: control, bottom: post-imaging sections. Scale bar: 50 μm.

(B) Quantification of double labeled positive cells, as a percentage of the total cells in the field. Per sample 5 random fields were quantified. Biological replicate (n = 3). Left: Tumor cells (GFP+) and KI67 + cells. Middle: Tumor cells (GFP+) and KI67 + cells. Left: Tumor cells (GFP+) and Survivin+ cells. Error bars represent ±SEM. Unpaired two-sided t-test. Ns = no significant.

(C) Analysis of tumor microenvironment and tumor invasive border. Nuclei are label with DAPI (blue). GFP (green) is labeling tumor cells. Left: Tumor cell density, GFP positive cells; Middle: Macrophages/microglia, ionized calcium-binding adapter molecule 1 (IBA1) positive cells (Red) and GFP+ tumor cells; Left: Reactive astrocytes, glial fibrillary acidic protein (GFAP) positive cells (pink) and GFP+ tumor cells (green). White dotted lines show the tumor invasive border. Top: control, bottom: 72 h post-imaging sections. Scale bar: 50 μm.

(D) Quantification of positive cells, as a percentage of the total cells per field. Per sample 5 random fields were quantified. Biological replicates (n = 3). Left: Tumor cells (GFP+); Middle: Left: Macrophages/microglia (IBA+); Astrocytes (GFAP+). Error bars represent ±SEM. Unpaired two-sided t-test. ns = no significant.

Figure 5.

Experimental approaches: analysis of the extracellular matrix components and response to treatment

(A) Hematoxylin and eosin (H&E) stained explants sections. Control, non-treated and collagenase treated (15 U/mL, for 48 h).

(B) Immunofluorescence of explants sections. Collagen 1a1 (COL1A1) green, Nuclei are label with DAPI (blue).

(C) Quantification of collagen areas (green immunofluorescent areas). Per sample 10 random fields were quantified (n = 3). Error bars represent ±SEM. Unpaired two-sided t-test. ∗∗p < 0.0042.

Tissue processing for cryosections

Timing: 24 h (tissue is incubated overnight)

Tissue could be embedded in OCT media to generate cryosections. Cryosections will maintain the staining with dyes such as Hoechst and Calcein AM for further analysis.

• 77.After PFA fixation, transfer tissue to 30% sucrose (m/v) solution overnight at 4°C.
• 78.Prepare a plastic container filled with cold isopentane/2-methylbutane and place into a metal container filled with liquid nitrogen.
• 79.With a micropipette, spread a thin layer of OCT media in the bottom of a disposable plastic base mold.
• 80.Using a flat micro-spatula remove the brain from the 30% sucrose solution and transfer an explant from the culture insert onto the OCT coating. Fill the plastic mold with OCT assuring that the tissue slices are submerged.
• 81.Using forceps, quickly place the cryomold with OCT and the explants into the isopentane/2-methylbutane.
• 82.Upon solidification of the OCT, remove the cryomold with the explants in dry ice. Store it at −80°C.

Agarose embedding for tissue processing in paraffin

Timing: 30 min

Explant slices are embedded in agarose to facilitate tissue processing and paraffin embedding.

• 83.Melt the 2% (m/v) agarose by swirling and heating in the microwave until light boiling is observed and the solution is transparent.

CRITICAL: If using a container with a cap or lid, make sure to loosen it so the container does not build up pressure while heating to prevent container breakage and injury. Place the agarose in a 40°C water bath to maintain its liquid state.

• 84.With a micropipette, spread 500 μL of 2% agarose into a thin, even coat in the bottom of a disposable plastic base mold.
• 85.With a flat micro-spatula, transfer an explant from a cell culture insert onto the agarose bed. Fill the plastic mold the rest of the way, fully submerging the tissue slices.
• 86.Allow the agarose to solidify at 20°C–25°C. Carefully remove the agarose block with a microspatula.
• 87.Place the agarose block in a labeled tissue MACROSETTE™, sandwiched between two rectangular foam biopsy pads.
• 88.Repeat steps 81–84 for all vibrosections to be embedded in paraffin. Multiple sections can be placed in the sabe mold.
• 89.Store the MACROSETTES™ in 70% ethanol for up to 24 h before embedding in paraffin. If it will be longer than 24 h, store in DPBS without Ca²⁺ or Mg²⁺ before transferring to 70% ethanol.

Tissue processing and paraffin embedding

Timing: 14 h (program runs overnight)

Explant slices are processed in an alcohol series and subsequently embedded in paraffin for further analysis in tissue sections.

Note: Alternative tissue processing procedures shorter than the one utilized in this protocol can be performed. The following steps were performed in an automated Leica ASP 300S however other devices can be used or the process can be done by hand. Following tissue fixation, the tissue must be dehydrated (usually with ethanol), cleared (usually with xylene), then infiltrated with paraffin.

• 90.Immerse MACROSETTES™ in 70% ethanol at 20°C–22°C under a vacuum for 1 h, let drain for 30 s.
• 91.Immerse MACROSETTES™ in 80% ethanol at 20°C–22°C under a vacuum for 1 h, let drain for 1 min.
• 92.Immerse MACROSETTES™ in 95% ethanol at 20°C–22°C under a vacuum for 1 h, let drain for 1 min.
• 93.Immerse MACROSETTES™ in 95% ethanol at 20°C–22°C under a vacuum for 1 h, let drain for 1 min.
• 94.Immerse MACROSETTES™ in 100% ethanol at 20°C–22°C under a vacuum for 1 h, let drain for 1 min.
• 95.Immerse MACROSETTES™ in 100% ethanol at 20°C–22°C under a vacuum for 1 h, let drain for 1 min.
• 96.Immerse MACROSETTES™ in 100% ethanol at 20°C–22°C under a vacuum for 1 h, let drain for 2 min.
• 97.Immerse MACROSETTES™ in xylene at 20°C–22°C under a vacuum for 1 h, let drain for 1 min.
• 98.Immerse MACROSETTES™ in xylene at 20°C–22°C under a vacuum for 1 h, let drain for 1 min.
• 99.Immerse MACROSETTES™ in xylene at 20°C–22°C under a vacuum for 1 h, let drain for 2 min.
• 100.Immerse MACROSETTES™ in paraffin at 58°C under a vacuum for 1 h, let drain for 2 min.
• 101.Immerse MACROSETTES™ in paraffin at 58°C under a vacuum for 1 h, let drain for 2 min.
• 102.Immerse MACROSETTES™ in paraffin at 58°C under a vacuum for 1 h, let drain for 2 min.
• 103.Place the MACROSETTES™ in the paraffin bath, tapping them with forceps to allow all bubbles to escape, preventing the MACROSETTES™ from floating.
• 104.Ensure the cold plate is turned on.
• 105.Wait 10–15 min, or until the agarose blocks within the MACROSETTES™ have become fully transparent.
• 106.Remove the agarose block from the MACROSETTE™ and, with a single-sided razor blade, trim excess agarose away. Take care to go slowly as the agarose can fracture during trimming.
• 107.Place the trimmed agarose block in a heated metal mold and transfer the mold to the cold plate. Press down on the agarose block to adhere it to the metal mold. Apply enough pressure to remove any air bubbles trapped between the agarose block and the metal mold however do not fracture the agarose.
• 108.Fill the metal mold with liquid paraffin, place the corresponding microcassette on top of the mold, and add enough additional liquid paraffin to fill the microcassette.
• 109.Gently transfer the metal mold and microcassette to the cold plate and allow the paraffin to solidify.
• 110.Once the paraffin is cool to the touch, carefully remove the paraffin block and microcassette.
• 111.The explants are now ready to be sectioned on a microtome at 5 μm, mounted on slides, and stained as needed for further experiments including H&E, immunohistochemistry, immunofluorescence, and multiplex assays such as RNAScope.

Post-imaging statistical analysis of tumor cell migration

Timing: 1.5 h per movie

To determine the movement of cells at various locations of the tumor and analyze patterns of migration, perform a localized statistical analysis. Select localized areas based on the organization of cells in clusters migrating together with a similar distribution.

Experimental dataset (time-lapse imaging)

The raw dataset consists of time-lapse videos acquired with a laser scanning confocal microscope (LSM 880, Axio Observer, Zeiss, Germany) using Zen (Blue edition) version 2.5 at the time interval of Δt = 10 min per frame using 10× objective covering approx. 850.19 × 850.19 μm (size in pixels: 1024 × 1024). Statistical analysis involves the TrackMate plugin, Image J Software, Julia, RStudio, and scripts that are available on GitHub and Zenodo (https://doi.org/10.5281/zenodo.7587235). Step-by-step information to analyze the glioma dynamics is provided below:

• 112.
  Tracking using TrackMate.

  • a.
    To track the migration of cells, use the software Fiji (https://imagej.net/software/fiji/) with the plugin TrackMate (https://imagej.net/plugins/trackmate).
  • b.
    The start panel will launch, showing information about the image dimensions.
  • c.
    To capture cells to track click “Next”, a new panel will pop up to select cell size which is called “blob”. Select 20 μm and a threshold of 1 together with the DoG method (Difference of Gaussian detectors).
  • d.
    When the progress bar has reached the end, click “Next.”
  • e.
    A tracking panel opens, we use the “Simple LAP” tracker and click “Next.”
  • f.
    A “Track filter” panel opens. Remove tracks according to their properties and select at least 10 tracks for each spot and remove the ones that are below 10 tracks, click “Next.”
  • g.
    For each time-lapse video, several trajectories of cells are developed in the analysis panel. Click the “Analysis” tab and three windows pop up with “Spots”, “Tracks Statistics” and “Link in Track Statistics.” Use the two CSV files named ”Spots” and “Tracks Statistics” for further analysis.
• 113.
  Step B: Trajectories graph using spots.csv and tracks.csv files employing Julia scripts.

  • a.
    As a result of TrackMate analysis, we obtain two CSV files.
  • b.
    As an example, and illustration, one can find these files on GitHub (https://github.com/smotsch/analysis_glioma) in the folder data tracking step1_data_trackMate/NPA_stich_3.
  • c.
    The positions (X, Y) of the cells (see the file ∗spots.csv) along with the trajectories (∗tracks.csv) are also provided there.
  • d.
    On running the script, individual cell trajectories graph of the time-lapse datasets is generated as shown in Figure 2.
• 114.
  Step C: Filtering.

  • a.
    To reduce this erratic behavior, we need to filter the trajectories.
  • b.
    From the data obtained with TrackMate, we then smooth trajectories which allows for estimating the velocity of each cell at any given time.
  • c.
    Use a Gaussian kernel as a filter (with standard deviation σ = 2 and a stencil of 9 points).
  • d.
    Run the script in Julia to perform this step, the script is in the folder src at GitHub (https://github.com/smotsch/analysis_glioma/tree/main/src) naming as step2_filter_data.jl.
  • e.
    On running the script, individual cell trajectories graph of the time-lapse datasets will be filtered.
  • f.
    This filtering step also estimates the positions Xi(t) and velocities Vi(t) of the cells i.
  • g.
    From the velocity Vi(t), we also deduce the velocity direction θi(t). We can now apply several statistics to investigate cell behavior. A path obtained with ImageJ (dashed line) and the filtered path (red) obtained after filtering should be applied. The smooth path allows estimating the velocity of the cell at each time step.
  • h.
    All the trajectories are saved in jld2 file in the folder data_tracking at GitHub (https://github.com/smotsch/analysis_glioma/tree/main/data_tracking) naming as step2_data_filtered.
• 115.
  Step D: Localized Statistics: velocity or correlation distribution.

  Note: Each cell is characterized by a position xi ∈ R2 and a velocity vi ∈ R2. The velocity vector v can be split into magnitude; speed (scalar) c = |v| and angle direction θ such that v = c(cos θ, sin θ). To investigate the distribution of velocity angle θ, speed c and velocity vector v, three statistics will be performed as shown in Figure 2F.

  • a.
    Distribution of velocity angle gives an indication of the overall direction of the cells.
  • b.
    Overall, statistics analysis in specific zones is assigned manually on the basis of the organization/pattern in glioma explants once the movie is complete.
  • c.
    The coordinates of each zone are determined by using Fiji coordinate function. Fiji will display the coordinates of the cursor position x, and y position in μm.
  • d.
    Coordinates allow us to perform the statistics in specific zones and the files of x and y coordinates save in a JSON file format called “coordinates.json” (the representative template file is provided at GitHub: https://github.com/smotsch/analysis_glioma/tree/main/data_tracking/step3_data_zone/NPA_stich_3).
  • e.
    “Coordinate.json” will be used in Julia to assign different zones: script to run and generate zone is available as “step3_zone_create_df.jl” at GitHub (https://github.com/smotsch/analysis_glioma/blob/main/src/step3_zone_create_df.jl).
  • f.
    Execution of script will generate data frames containing the analysis of each zone. Repeat run twice to execute velocity or correlation analysis using Julia script provided at GitHub (https://github.com/smotsch/analysis_glioma/blob/main/src/step3_zone_create_df.jl).
  • g.
    Line 13 of the script needs to change for velocity and correlation as shown below:

    velocity_or_correlation = “velocity” # “velocity” or “correlation”

    velocity_or_correlation = “correlation” # “velocity” or “correlation”
  • h.
    The final graph of the velocity or correlation analysis can be generated using Julia scripts provided at GitHub (https://github.com/smotsch/analysis_glioma/tree/main/src).
  • i.
    “step3_zone_velocity_plot.jl.”
  • j.
    “step3_zone_correlation_plot.jl.”
  • k.
    Results pertaining to angle velocity, speed distribution and velocity heatmap of each zone can be easily accessed as a pdf in df velocity folder while correlation, nematic correlation, and relative position can be accessed in df_correlation folder [The illustration of the datasets result is represented on GitHub (https://github.com/smotsch/analysis_glioma/tree/main/data_tracking/step3_data_zone/NPA_stich_3).
• 116.
  Step E: Likelihood Analysis using RStudio scripts to classify collective patterns of migration.

  • a.
    To classify a pattern of migration or invasion in each zone (flock, stream or swarm), utilize the collection of angles θi inside a specific zone and assess three distributions, which are (wrapped) Gaussian for a flock, a symmetric Gaussian for a stream and the constant function for a swarm (Figures 2H and 2I).
  • b.
    Akaike weight distribution analysis matches the best distribution of θi.
  • c.
    To determine the likelihood analysis and specify whether the zone is flock, stream or swarm, run the R script provided on GitHub (https://github.com/smotsch/analysis_glioma/blob/main/src/step4_zone_testing_distribution.r).
  • d.
    The zone distribution graph will be generated in R Studio (https://www.r-project.org/) along with the Akaike weight results.
  • e.
    AW close to 1 will be considered as the highest likelihood and disregards the other ones that are not close to AW = 1.

Figure 2.

Confocal time-lapse imaging of mouse glioma explant slices: Study of tumor migration and invasion

(A) Inserts holding 300 μm explants slices are placed in Zeiss LSM880 laser scanning confocal microscope incubator chamber and maintained at 37°C and 5% of CO₂.

(B and C) Confocal microscope focal plane can cover 100–150 μm. A high-resolution z-stack is obtained and the plane for time-lapse imaging is selected at the middle of the z-stack to avoid imaging at the bottom of the explant.

(D) Representative images of time-lapse confocal imaging of NPA GFP+ neurospheres orthotopically implanted in a TdTomato mouse. Scale bar: 100 μm (Corresponding with Methods video S1).

(E) Tracking analysis of individual GFP+ tumor cell paths using the TrackMate plugin from Image-J.

(F and G) Statistical analysis developed to study glioma cells dynamic. (F) Three Individual statistics were explored to visualize the distribution of velocity: distribution of velocity angle (θ), speed distribution (c), and the velocity vector (v) distribution. (G) Illustration of the method used to determine relative position and pairwise correlation. Relative Position: Frequency for each zone of the relative position between two cells. In the frame of a reference of one cell, we estimate the probability to have another cell xj nearby. Pair-wise correlation: correlation with nearby neighbors is estimated using the velocity directions ωi and ωj and the relative position xi-xj.

(H and I) Likelihood Analysis to classify collective patterns of migration. (H) Angle Velocity graph represent the three different patterns of migration defined by a collection of angles θ_i : Streams (parallel opposite directions), Flock (same directions) and Swarm (multiple directions). (I) Angle velocity was used to obtain the histogram with the three distributions and calculate the likelihood that corresponds Gaussian for a flock, a symmetrize Gaussian for a stream and the constant function for a swarm.

Expected outcomes

Mouse glioma tumors are generated by inducing genetic modifications in subependymal stem cells, or by orthotopic implantation of neurospheres glioma cells. Using GEMM of glioma tumors can be detected using IVIS imaging after 30–40 days of plasmid injection. Tumors are expected to grow, and tumor size can be analyzed every week until obtaining a maximum signal of 10⁶–10⁷ photon/sec (GA). In our experiments, the NPD genetic model of glioma has a median survival of 74 days and NPA tumors have a median survival of 68 days as described previous work.¹ However, different genetic models could present a different time point to reach the end stage. Tumors generated by orthotopic implantation will be used when tumors reach an IVIS signal of 10⁶–10⁷ photon/sec. Tumor growth and expected moribund stage will vary depending on the individual tumor model used; for NPA intracranial implanted neurospheres used in our model, animals are expected to be euthanized after approximately 19–20 days (GA). After approx. 3 weeks we expect to euthanize mice and detect GFP+ tumors using epifluorescent stereomicroscopy; at this time, tumor size is expected to be between 3 to 6 mm of diameter.

Patient-derived tumor specimens are collected as soon as they are taken from the surgery site and maintained in cold oxygenated human culture medium for transportation. Under these conditions of transportation and maintenance, we expect to obtain high tumor viability after several days of culture (GA).

After tumor sample acquisition (mouse and/or human), tissue is processed in no more than 90 min. The process involves, embedding in 4% agarose at a temperature not above 37°C, then tissue block is maintained on ice and sectioned in cold and oxygenated conditions. We anticipated to get high tissue viability and morphology quality if tissue is processed following our protocol guidelines (Figures 1A–1I). For mouse tissue, explant slices showing GFP+ expression will be selected for culture (Figures 1J and 1K). Patient-derived tumor tissue will be stained with Calcein AM, a permeable dye, and Hoechst 33342 to evaluate tumor tissue viability and migration analysis (Figure 1L).

Figure 1.

Step by step of the protocol: embedding and sectioning fresh glioma tissue

(A) In the laminar flow hood, media should be kept on ice. Set up the knife holder to the vibratome.

(B) Vibratome’s buffer holder is surrounded with ice, to maintain the low temperature while cutting. Buffer holder is filled with cold sectioning media, with constant oxygen supply.

(C) Dissected brain, or biopsy, is placed in an embedding mold filled with 4% agarose.

(D) The mold is place on ice until it solidifies.

(E) Agarose block is cut out of the mold and glued to the specimen disc.

(F) Specimen disc is placed on ice until glue is dry.

(G) Specimen disc with the attached agarose block is placed and fasten in the buffer tray.

(H and I) 250–300 μm coronal are collected with the mini perforated spoon, and transferred to media-containing plates.

(J) Selected sections are placed in laminin or Matrigel coated inserts.

(K) Explants fluorescence is corroborated using. Murine models exhibit a red fluorescent (TdTomato+) parenchyma and green fluorescent (GFP+) tumor.

(L) Human tissue is stained with Hoescht 3334 and Calcein AM cell permanent dye (green fluorescent) to analyze tumor viability.

After 6–24 h of culture slices are placed in an incubator chamber for time-lapse confocal imaging. In our model we used an inverted Zeiss LSM880 laser scanning confocal microscope with an incubation chamber that enable tissue viability over more than 72 h (Methods video S1). You can expect longer incubation periods if the media is replaced every two days. To validate the imaging depth, we performed Z-stack acquisitions (Figure 2). Using time-lapse confocal imaging of glioma explants ex vivo, it is possible to analyze cell migration patterns within the tumor core and at the tumor border in explant slices generated from mouse glioma models in TdTomato mice (Figure 2D). Patient-derived glioma explants stained with a combination of the cell permeable Calcein AM dye (cytoplasm) and Hoechst 33342 (nuclei) will enable the visualization of tumor migration by time-lapse confocal imaging (Figure 1L). We predict that other tumor models could also be adapted to be used for time lapse confocal imaging using this protocol. We found that movies performed every 10 min for at least 100–150 cycles are suitable for statistical analysis of single and collective cellular migration and invasion. Cells are individually tracked using the TrackMate plugging (Figure 2E). Since individual paths are usually erratic, the trajectories are filtered using the Gaussian Kernel method for a more precise analysis of cell velocity. Different parameters such as cell positions x_i(t), cell velocities v_i(t), and cell velocity direction θ_i (t) are obtained. Using these parameters, it is possible to apply several statistics to study cell behavior (Figures 2F–2I). We proposed the use of three statistics to understand the distribution of cell velocity: (1) velocity angle [θ]: this parameter gives an indication of the average direction of all cells); (2) speed of the cells [c] and (3) velocity vector [v = (vx, vy)] using a heatmap plot to observe the distribution (Figure 2F). Further, to study the collective migration of cells we proposed to use pairwise analysis. From this analysis is it possible to obtain the (1) Relative Position of cells: this statistic indicates the density of neighboring cells, using the reference of the position of any cell (xi), the probability to have another cell (x_j) close is calculated. (2) Correlation orientation or velocity correlation indicates the correlation: ω_i ·ω_j, depending on the position of a nearby cell x_j (Figure 2G). Moreover, using the velocity angle (cell orientations) it is possible to predict specific patterns of collective migration such as Streams (cells moving in opposite directions), Flocks (cells moving with the same direction) or as Swarms (moving without any preferred direction). Given the dataset we used a likelihood estimation to determine which distribution is more likely (Figures 2H and 2I). Altogether these statistical migration analyses can be used to understand the role of different target genes, specific mutations or drug treatment on glioma invasion and progression.

Following movie acquisition or drug treatment, sections are fixed in 4% PFA for 24–48 h and processed for paraffin embedding or cryopreservation for further analysis (Figures 3A and 3B). Fixation of tissue enables the analysis of tumor cell morphology and histopathological features following time lapse imagining or treatment. Figure 3C shows the preservation of the tumor histological structure in mouse and human specimens after more than 4 days of tissue culture. Tumor tissue viability and the cell microenvironment could be preserved for longer periods if media is replaced periodically every two or three days (Figures 4A–4D).

To check for possible tissue damage because of the exposition of explants to time-lapse confocal imaging we recommend performing post-imaging controls. We observed that tumor cells’ viability and apoptosis were not significantly different from tissue exposed to 72 h of imaging compared to tissue control maintained in an incubation chamber at 37°C and 5% CO₂ for the same period (Figures 4A–4D). Moreover, the evaluation of TME showed that GFP+ tumor cells, IBA1+ macrophages/microglia and GFAP+ astrocytes are maintained after imaging of explant slices for 72 h (Figures 4C and 4D). Besides, imaging, and post-analysis of explant slices allowed the analysis of the tumor border and invasion (Figure 4C).

On the other hand, as shown in Figure 5, fixed explant slices can be stained with H&E for evaluation of histology and ECM composition in response to treatment by IHC of collagen or other ECM components (Figures 5A–5C).

In summary, this protocol provides an integrated approach using mouse and patient-derived explant slices to evaluate cell proliferation, cell migration and invasion, TME and ECM composition and response to treatment. This model represents a versatile model and a potential pipeline to analyze tumor responses to standard of care and clinical trials treatments and evaluate high-throughput drug screening using patient-derived human explants slices.

Limitations

There are a few limitations to this technique and laboratory facilities required. One limitation is that the tumor section may not be able to be imaged for more than 3–5 days because the media needs to be changed to retain cell viability. Additionally, cells may become photobleached after time lapse confocal imaging for several days and imaging quality will decline over time. The use of the minimum laser power is recommended. Depth of imaging within the explant slice using confocal imaging is also limited to around 100–150 μm, depending on the microscope utilized. For high throughput drug screening, plates with 96 or higher number of wells with inserts should be used.

To perform this protocol, animal husbandry under institutional regulations is required for mouse models and caring for mice with tumors. For different radiation treatments, an irradiation core will be utilized. Human samples will be acquired from physicians in partnering hospitals. For post-analysis, a tissue embedding core may be necessary.

Troubleshooting

Problem 1

Tissue dislodges from the agarose block during sectioning. Agarose and tissue may have a very different density, and embedding was not appropriated (Related to steps 29–30).

Potential solution

• •Use different concentrations (m/v) of agarose depending on the tumor tissue. Re-embed the tissue.
• •Use always brand-new blades.

Problem 2

Mouse tumor explant slices do not show positive GFP tumor. A possible cause is that tumor was too small at the time of euthanasia (Related to step 21).

Potential solution

• •Perform the optimization of tumor growth in your animal model. If possible, check animals via IVIS imaging before tissue acquisition. Do not use tumor with less than 10⁶ photons/sec signal.

Problem 3

Explants slices are floating and not adhered to the insert (Related to steps 45–47).

Potential solution

• •Add media in the well below the insert and only a thin layer of media on the top of the explant.
• •Ensure to perform the coating at least 1 h before placing the explant on the inserts.

Problem 4

Affected and impaired cell viability or migration (Related to steps 53–67).

Potential solution

• •Avoid using agarose at more than 37°C.
• •Acquire the tissue directly from OR and transport the tissue immediately in clod oxygenated media.
• •Keep media cold and oxygenated during all sectioning process.
• •Organize the materials before start and perform the procedure within 90 min.
• •Necrotic tissue will be inviable for migration analysis.

Problem 5

During embedding of fixed tissue, explants float up when filling the mold with agarose (Related to STEPS 82–85).

Potential solution

• •Place the explant on the thin layer of agarose quickly before it solidified. Then with a micropipette, place a couple drops of liquid agarose on the explants and allow it to solidify before filling the plastic mold the rest of the way.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact [Pedro Lowenstein] (pedrol@umich.edu).

Materials availability

All materials associated with this protocol are in the paper. Cells, plasmids, and other reagents developed in this study could be made available upon request to pedrol@umich.edu.

Data and code availability

The protocol includes a code for glioma cells dynamic analysis. The analysis can be performed using the Julia Programing Language. Link for this project Script and their dependencies can be found at public GitHub repository: https://github.com/smotsch/analysis_glioma and Zenodo: https://doi.org/10.5281/zenodo.7587235.