In Situ Detection of Metabolically Active Cells in Hepatocellular Carcinoma Tissue Using MTT-Based Cryosection Imaging

Summary

Here, we present a simple protocol to detect the metabolically active cells in hepatocellular carcinoma tissue using MTT-based cryosection imaging. This method may also be used to detect active cells in situ in other tissues or organs.

Abstract

Emerging evidence, including prior studies, highlights a subpopulation of cells within hepatocellular carcinoma (HCC) tissues that exhibit superior metabolic activity and stress resistance. These cells are key drivers of tumor progression and therapy resistance. However, existing methods for in situ viable cell detection often compromise tissue integrity, alter cellular viability, or are technically demanding. This study presents a novel, simple, non-destructive approach for in situ detection of metabolically active cells in HCC tissues through mitochondrial-dependent reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to insoluble formazan crystals. By optimizing tissue fragment size (1 × 1 × 0.2 cm³), culture conditions (20% FBS, 1 mg/mL MTT, and 3 h incubation with intermittent swirling), and cryosectioning parameters (20 µm thickness, 4% PFA fixation, and DAPI counterstaining), spatial mapping of viable HCC cells was achieved within 5 h. This cost-effective protocol requires no specialized equipment and maintains tissue architecture, enabling spatial identification of high-viability and stress-resistant cell clusters in surface regions. Limitations include progressive viability loss beyond 6 h and crystal displacement during sectioning, which can be mitigated through FBS supplementation and controlled freezing. This method provides a practical platform for single-cell isolation of therapy-resistant subpopulations, advancing HCC microenvironment research.

Introduction

Hepatocellular carcinoma (HCC), the third leading cause of cancer-related deaths globally, is characterized by high heterogeneity and recurrence rates¹. This spatial and temporal heterogeneity makes certain cell populations within tumor tissues more active than others in metabolism, proliferation, progression, stress tolerance, and drug resistance. The in situ detection of viable cells within HCC tissues holds profound clinical and biological significance, as these cells encompass proliferating subclones, therapy-resistant populations, and metastasis-initiating cells. Such viable cells could serve as critical drivers of tumor recurrence, immune evasion, and therapeutic failure².

However, current spatial profiling technologies (e.g., spatial transcriptomics) require specialized instrumentation and are associated with high costs, while traditional bulk analyses often overlook the spatial heterogeneity of viable cells and their dynamic interactions with stromal components³. For instance, residual viable HCC cells post-resection exhibit distinct metabolic profiles that cannot be faithfully recapitulated in dissociated cell systems⁴. Thus, advancing spatially resolved detection technologies is imperative to unravel the functional crosstalk between viable tumor cells and their microenvironment, ultimately informing precision therapeutic strategies. Although conventional methods such as flow cytometry or immunohistochemistry (IHC) have been used by our group to detect the viable cells in HCC, they require tissue dissociation or fixation, thereby losing spatial context and hindering the analysis of cell-cell interactions^5,6. Thus, developing non-destructive, high-resolution technologies for in situ detection of viable cells is critical for elucidating HCC resistance mechanisms and guiding precision therapies.

To rapidly detect and display the viable cells in HCC tissues in situ without destructive treatment, a simple 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)-based tissue culture method was developed. Briefly, the fresh thin HCC tissues (approximately 1 × 1 × 0.2 cm³) were first cultured in DMEM medium for 3 h containing 20% fetal bovine serum (FBS) and MTT (1 mg/mL). Then, after 4% paraformaldehyde (PFA) fixation and DAPI-containing mounting, OCT-embedded frozen tissue slides (20 µm thickness) were subjected to microscopic observation under bright or UV light. The viable cells will display purple, purple-red, or purple-black coloration. Using this protocol, we successfully detected the viable cells in HCC tissues within 5 h. Meanwhile, this method preserves 3D microarchitecture, is compatible with downstream multi-omics analysis, and is simple, as well as time- and cost-friendly. However, this method exhibits certain limitations. The multi-step processes of washing, sectioning, and culturing may induce structural deformation and apoptosis in tissues, especially in high metabolic samples, such as HCC tissue in this study. Additionally, increased thickness of cryosection slices compromises the clarity and flatness of microscopic observations.

Protocol

All experiments were conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the First Affiliated Hospital of Quanzhou (No. FJMU IACUC2021257). HCC tissues were surgically removed and collected at Quanzhou First Hospital, affiliated to Fujian Medical University, with informed consent from the patients. This experiment requires standard protective measures, including wearing a laboratory coat, a disposable mask, and gloves.

1. Tissue preparation

1. Place a sterile 9 cm Petri dish on ice. Add 30 mL of ice-cold Phosphate Buffer Saline (PBS).
2. Immediately transfer the surgically resected human HCC tissue into the PBS.
3. Gently swirl the dish for 5 s to remove blood contaminants. Aspirate PBS using a sterile pipette.
   NOTE: Repeat three times.
4. Transfer the tissue to the lid of the Petri dish.
   NOTE: The tissues easily adhere to the unwetted Petri dish lid for better handling subsequently.
5. Trim the tissue into approximately 1 × 1 × 0.2 cm³ fragments using a sterile surgical blade (No. 11).
   NOTE: The length and width of the fragments could be changed, but the depth should be less than 0.2 cm.
6. Return fragments to a fresh 9 cm dish filled with 30 mL of ice-cold PBS.
7. Gently swirl the dish for 5 s to remove blood contaminants. Aspirate PBS using a sterile pipette.
   NOTE: Repeat step 1.7 three times. One tissue fragment boiled in PBS (95 °C, 10 min) was set as a negative control. Perform all steps under sterile conditions.

2. Tissue culture

1. Add 1.8 mL of complete culture medium DMEM (prewarmed to 37 °C) containing glucose (final concentration = 4.5 g/L) and L-glutamine (final concentration = 2 mM) into each well of a 6-well plate. Then, add 0.6 mL FBS (final concentration = 20% v/v), and 0.6 mL MTT stock (final concentration = 1 mg/mL) into each well of a 6-well plate.
2. Transfer the tissue fragments into wells using sterile forceps.
3. Place the 6-well plate in the incubator. Incubate for 3 h at 37 °C with 5% CO₂.
   NOTE: Gently swirl the plate for 3 s every 30 min. The incubation could be terminated once the surface of the tissue fragments turns purple or purple-black.

3. Termination and fixation

1. Aspirate the MTT medium. Wash tissue fragments with ice-cold saline.
   NOTE: Repeat five times.
2. Add 2 mL of 4% neutral-buffered formalin (NBF) into each well of a 6-well plate.
3. Place the plate in the refrigerator at 4 °C for 1 h.

4. Cryosection

1. Blot the fragments gently on clean absorbent paper.
2. Place tissue in a cryomold filled with OCT compound.
   NOTE: Ensure the tissue is fully covered with OCT compound.
3. Freeze the tissue in a low-temperature refrigerator at -80 °C for 20 min.
4. Mount the frozen block on a cryostat chuck at -20 °C.
5. Cut 100 µm sections at -20 °C until the tissue surface is exposed.
   NOTE: The tissue surface displays a purple-like coloration.
6. Collect 20 µm slices using adhesive-coated slides.
   NOTE: The tissue slices should present in a shade of purple, purple-red, or purple-black.
7. Fix the slices in 4% paraformaldehyde (PFA) at room temperature for 2 min.
8. Gently rinse the slides with PBS for 1 min.
   NOTE: Repeat three times.

5. Mounting

1. Apply 1 drop of DAPI-containing (final concentration: 1 µg/mL) mounting medium..
2. Cover with a coverslip.
3. Seal the edges with nail polish.
   NOTE: Slides with sealed edges can be stored at 4 °C in the dark if microscopy is not performed immediately.

6. Microscopy

1. Visualize purple, purple-red, or purple-dark formazan deposits (viable cells) at 20× magnification under brightfield using brightfield microscopy.
2. Detect blue DAPI (nuclei) under UV illumination using fluorescence microscopy.
3. Merge brightfield and fluorescence channels for spatial correlation.
   NOTE: The software used here is NIS Element F (version 4.00.06).

Results

This study successfully detected metabolically active cells in HCC tissue without disrupting tissue architecture. Firstly, the brightfield images (Figure 1A) showed that the tissue surface exhibited dense cellular coverage and multilayered stacking of cells across the slide, confirming the preserved architectural integrity of the HCC tissue throughout culture and processing. Notably, clusters of cells within superficial regions displayed distinct purple-red formazan deposits, indicative of mitochondrial activity in viable cell populations. Secondly, DAPI nuclear counterstaining can easily identify the location and outline of cells in fluorescence images (Figure 1B). These images also exhibited uniformly intact nuclei across the section, confirming cellular structural preservation achieved by our protocol. Thirdly, after merging the brightfield images (Figure 1A) and their corresponding fluorescence images (Figure 1B), we can more easily and accurately identify and distinguish the viable cells by the precise colocalization of formazan-positive cells (purple-red) with DAPI-positive nuclei (blue), enabling in situ mapping of viable cells within the tissue (Figure 1C).

Unlike the surface-proximal cryosections, the brightfield images (Figure 2A) of the cryosections derived from deeper tissue regions showed no detectable formazan signals, indicating an absence of viable cells. However, the fluorescence images (Figure 2B) of these regions still showed nuclear integrity, ruling out procedural artifacts in cryosectioning. Meanwhile, merged images (Figure 2C) further validated the spatial consistency between non-viable cells (brightfield) and intact nuclei (DAPI), highlighting a gradient of metabolic activity from the tissue surface to its core.

Figure 1: Microscopy of surface-proximal cryosections. (A) Brightfield images; (B) Fluorescence images; (C) Merged images. Scale bar = 50 µm. Please click here to view a larger version of this figure.

Figure 2: Microscopy of deep-tissue cryosections. (A) Brightfield images; (B) Fluorescence Images; (C) Merged images. Scale bar = 50 µm. Please click here to view a larger version of this figure.

Discussion

Current methodologies for in situ viable cell detection still face multifaceted limitations. IHC and immunofluorescence (IF), which are widely used to identify proliferation markers or apoptosis-related proteins, suffer from disruption of membrane integrity induced by formaldehyde fixation⁷. Moreover, antibody penetration in thick tissue sections is suboptimal⁸. Fluorescent viability probes such as Calcein-AM, though effective in labeling esterase-active live cells ex vivo, exhibit reduced specificity in hypoxic HCC regions due to altered enzymatic activity and nonspecific diffusion into necrotic areas⁹. Laser capture microdissection (LCM) enables precise isolation of viable cell clusters but operates at low throughput and sacrifices spatial context, limiting its utility in studying cell-cell communication dynamics¹⁰. Emerging technologies like multiphoton microscopy offer deep-tissue imaging but require high costs and induce phototoxicity during prolonged imaging, which can artificially perturb cell viability¹¹. Nanoparticle-based probes designed to target HCC-specific surface markers enhance signal specificity but face challenges such as nonspecific uptake by Kupffer cells and potential induction of oxidative stress, confounding viability assessments¹².

This study developed a simple approach for in situ detection of viable cells in HCC tissues by leveraging the metabolic reduction of MTT to insoluble formazan crystals within live cells. Unlike conventional MTT-based cell viability detection protocols that require formazan solubilization, our method preserves crystal deposition in situ. This method enables direct visualization of viable cells without disrupting tissue architecture, while also providing operational simplicity and cost-effectiveness. Moreover, this is a dynamic viability assessment during ex vivo tissue culture, a critical feature for identifying stress-resistant cell subpopulations under near-physiological conditions. This protocol is particularly suited for rapid viability screening in resource-limited settings or intraoperative margin assessment.

Some limitations also emerged in this experiment. The first is the progressive loss of viability. It was found that extended culture durations >6 h) led to gradual cell death, likely due to hypoxia and nutrient depletion. To address this, the culture conditions were optimized by increasing the FBS concentration to 20% to enhance nutrient supply, and by limiting incubation to 3 h to balance signal intensity and viability preservation. The second limitation is the restricted penetration of MTT. Microscopy results indicated that formazan signals were predominantly localized to superficial tissue regions (depth <100 µm), reflecting poor MTT diffusion into dense HCC matrices. This limitation could be addressed by sectioning tissues into 1 × 1 × 0.2 cm³ fragments, increasing the MTT concentration to 1 mg/mL, and swirling the plate for 3 s every 30 min. The third issue is crystal loss during processing, specifically formazan leaching during sectioning and washing, which generates false negatives. This was minimized by increasing cryosection thickness to 20 µm and fixing slides with 4% PFA prior to washing. The fourth challenge involves signal specificity. It is easy for colors from endogenous pigments to overlap, complicating formazan identification. To enhance specificity, DAPI nuclear counterstaining was incorporated in the mounting medium to distinguish viable cells (formazan+ nuclei+) from debris. Negative controls were also established using heat-inactivated tissues to validate signal thresholds.

In summary, spatially restricted clusters of viable, stress-resistant cells were identified within HCC tissues, a population hypothesized to drive chemoresistance and recurrence¹³. This protocol's capacity to map metabolically active cells in situ paves the way for single-cell isolation and subsequent multi-omics profiling, potentially uncovering novel therapeutic targets.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
-80 °C freezer (upright)	Thermo Fisher Scientific	TSX60086V
Clean bench (vertical laminar flow)	AIRTECH	SW-CJ-1FD
CO2 incubator (165 L)	Thermo Fisher Scientific	Heracell VIOS 160i (e.g., 51030400 / 51033559)
Coverslip (24×50 mm, #1.5)	CITOGLAS (Citotest)	0341-3610
Culture petri dish (90 mm)	NEST	752001 (example)
DAPI-containing mounting medium	Beyotime	P0131 (5 mL/25 mL)
DMEM (high glucose, w/ pyruvate)	Invitrogen (Gibco)	11995-065
FBS (Qualified, US origin)	Invitrogen (Gibco)	A3160501
Fluorescent microscope	Nikon	ECLIPSE Ti2 series
Freezing microtome (cryostat)	Leica	CM1950
Microscope slides (plain, 25×75 mm)	CITOGLAS (Citotest)	0317-0001
MTT reagent (powder)	Beyotime	ST316
Nail polish (clear)	Beyotime	—
NBF (Neutral Buffered Formalin)	Beyotime	—
OCT compound	SAKURA (Sakura Finetek)	4583
Optical microscope (upright)	Nikon	ECLIPSE Ci series (e.g., Ci-L)
PBS (500 mL)	Beyotime	C0221A
PFA, 4% in PBS (fixative)	Beyotime	P0099
Pipette tip (1 mL, low-binding)	KIRGEN	KG1313-L
Pipette tip (200 µL)	Axygen (Corning)	T-200-Y (bulk) / T-200-Y-R-S (racked, sterile)
Pipettor (single-channel)	Thermo Fisher Scientific	Finnpipette F2 (range-specific cat#)
Scalpel	HYSTIC	model varies (e.g., supplier ref. 100022208326)
Tweezer	HYSTIC	model varies