Protocol for isolating patient-derived ascites cells and extracellular vesicles from gastric cancer peritoneal metastases

Summary

Peritoneal carcinomatosis (PC) cells and extracellular vesicles (EVs) from gastric cancer ascites are valuable for studying tumor-stroma interactions. Here, we present a protocol for isolating PC cells and EVs from patient ascites. We describe steps for PC cell culture, iodixanol-based EV purification, electron microscopy, nanoparticle tracking analysis, flow cytometry, and proteomic profiling. This protocol also includes TD-139 loading into exosomes for functional assays to evaluate their role in modulating the tumor microenvironment.

For complete details on the use and execution of this protocol, please refer to Fan et al.¹

Graphical abstract

Highlights

• •Isolation and culture of patient-derived PC cells from malignant ascites
• •Density gradient ultracentrifugation for extracellular vesicle purification
• •Extracellular vesicle characterization by TEM, NTA, and flow cytometry
• •Proteomic profiling of extracellular vesicles using timsTOF HTmass spectrometry (Bruker)

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

Peritoneal carcinomatosis (PC) cells and extracellular vesicles (EVs) from gastric cancer ascites are valuable for studying tumor-stroma interactions. Here, we present a protocol for isolating PC cells and EVs from patient ascites. We describe steps for PC cell culture, iodixanol-based EV purification, electron microscopy, nanoparticle tracking analysis, flow cytometry, and proteomic profiling. This protocol also includes TD-139 loading into exosomes for functional assays to evaluate their role in modulating the tumor microenvironment.

Before you begin

This protocol is designed primarily for isolating peritoneal carcinomatosis (PC) cells and extracellular vesicles (EVs) from malignant ascites of gastric adenocarcinoma patients for subsequent functional and molecular analyses. We have successfully applied this protocol to ascites samples from human gastric adenocarcinoma. Additionally, the protocol can be adapted to other malignancies characterized by abundant malignant ascites, such as high-grade serous ovarian carcinoma.

Institutional permissions

Obtain approval from the relevant institutional ethics review board before beginning the collection of patient samples.

CRITICAL: All human specimen collections must adhere strictly to applicable institutional and national ethical guidelines and regulations.

Identify and recruit patients diagnosed with PC requiring therapeutic paracentesis, without additional selection criteria.

Obtain written, informed consent from all participating patients before sample collection.

Collect ascitic fluid samples from patients according to approved clinical protocols.

Note: At The University of Texas MD Anderson Cancer Center, patient-derived ascites specimens were collected under Institutional Review Board-approved protocol (LAB01-543). All experiments comply with institutional guidelines and the Declaration of Helsinki (1964 and later amendments). Researchers at other institutions should obtain equivalent approvals and permissions from their respective ethical committees prior to initiating experiments.

Key resources table

REAGENT or RESOURCE	RESOURCE	IDENTIFIER
Antibodies
CD9 antibody (1:100)	BD Biosciences	Cat# 558749, RRID: AB_397103
CD63 antibody (1:100)	BD Biosciences	Cat# 564221, RRID: AB_2738678
CD81 antibody (1:100)	BD Biosciences	Cat# 559519, RRID: AB_397260
Chemicals, peptides, and recombinant proteins
TD-139	SelleckChem	Cat# S0471
Critical commercial assays
OptiPrep iodixanol 60% (w/v), 1.320 g/mL	Sigma-Aldrich	Cat# D1556
Glutaraldehyde solution	Electron Microscopy Sciences	Cat# 16220
Formvar/carbon-coated copper grids	Electron Microscopy Sciences	Cat# FCF200-Cu
Poly-L-lysine solution	Sigma-Aldrich	Cat# P4707
Uranyl acetate	Electron Microscopy Sciences	Cat# 22400
Experimental models: Cell lines
Human: AGS cell line	ATCC	N/A
Murine: KP-Luc2 cell line	Fan et al.2	NA
Software and algorithms
ImageJ	N/A	https://imagej.nih.gov/ij/
Other
Steriflip Vacuum filter unit, 0.22 μm	MilliporeSigma	Cat# SCGP00525
RPMI-1640	Gibco	Cat# 11875093
Dulbecco’s modified Eagle’s medium (DMEM)	Gibco	Cat# 11965092
Fetal bovine serum	Gibco	Cat# 26140079
Penicillin-Streptomycin	Gibco	Cat# 15140122
Phosphate-buffered saline	Gibco	Cat# 10010023
70 μm Cell strainer	Corning	Cat# 352350
T25 Flask	Corning	Cat# 353108
T225 Flask	Corning	Cat# 431082
Polycarbonate ultracentrifuge tubes	Beckman Coulter	Cat# 355618
0.2 μm Filters	MilliporeSigma	Cat# SLGP033RS

Materials and equipment

Reagent	Final concentration	Amount
OptiPrep (iodixanol)	1.14 g/mL	38.6 mL
PBS	N/A	11.4 mL
Total	N/A	50 mL

Step-by-step method details

Isolation of patient-derived PC cells from ascites

Timing: 1–3 h for initial processing; 1–2 weeks for culture

This step outlines how to isolate and culture adherent tumor cells from patient-derived ascites. The resulting epithelial cell cultures can be used for downstream applications including exosome production and functional assays.

• 1.Collect ascitic fluid (∼500–1000 mL) via therapeutic paracentesis.
• 2.Immediately transport the fluid on ice to the lab and process within 2 h.
• 3.Transfer fluid to 50 mL conical tubes and centrifuge at 500 × g for 10 min at 4°C.
• 4.Discard the supernatant and resuspend the pellet in 5–10 mL PBS, adjusted the volume based on the pellet size.
• 5.Filter through a 70 μm strainer to remove clumps and debris.
• 6.Centrifuge again at 500 × g for 5 min.
• 7.Resuspend the cell pellet in RPMI-1640 with 10% FBS and 1× Pen/Strep.
• 8.Plate into T25 flasks or 6-well plates and incubate at 37°C, 5% CO₂.
• 9.Change media every 2–3 days to remove non-adherent cells.
• 10.Passage when adherent cells reach 70–80% confluence.

Note: All procedures involving cell isolation and processing should be performed under sterile conditions in a biosafety cabinet.

Isolation of ascites-derived EVs by density gradient flotation

Timing: ∼6 h

This step describes the isolation of EVs, including exosomes, from malignant ascites using iodixanol-based density gradient ultracentrifugation. The resulting EV fraction is suitable for downstream characterization by TEM, NTA, and flow cytometry, as well as proteomic and functional assays. This method yields highly enriched EVs with reduced contamination from protein aggregates and non-vesicular particles, enabling high-confidence analyses of EV cargo and biological activity.

The full workflow is illustrated in Figure 1.

• 11.Centrifuge 10–50 ml ascitic fluid at 2000 × g for 20 min at 4°C to remove cells.
• 12.Transfer supernatant to a new tube and centrifuge at 16,500 × g for 30 min at 4°C.

Note: This step is to remove large extracellular vesicles (LEVs, >∼200 nm diameter) and non-vesicular microparticles.³

• 13.Filter the supernatant using a pre-wetted 0.22 μm vacuum tube top filter that has been pre-wetted with sterile PBS.
• 14.Add 1.3 g of OptiPrep 1.320 per g of filtered ascites sample and mix to final sample density of 1.17 g/mL.
• 15.Load densified sample into the bottom of polycarbonate ultracentrifuge tube.

Note: Fill volume should be ∼80% of tube volume capacity.

• 16.
  Overlay densified ascites with 2–4 mL of 1.14 g/mL iodixanol/PBS gradient solution.

  • a.
    Prepare the gradient solution by mixing 0.977 g of PBS per g of OptiPrep 1.320.
  • b.
    Dispense iodixanol/PBS solution over ascites layer using syringe fitted with 18 Ga needle. Take care to minimize mixing of the two layers.
• 17.Overlay 1.14 g/mL layer with 0.5 mL of PBS, again using syringe and avoiding excessive mixing.
• 18.Ultracentrifuge at 65,000 × g for >4 h (or overnight) at 8°C.
• 19.
  Collect exosomes/small EVs from the top of the gradient using a 1–3 mL syringe fitted with a 20–21 Ga needle.

  • a.
    Harvesting PBS layer along with 90% of the 1.14 g/mL iodixanol/PBS layer volume.
  • b.
    Avoiding the 1.17 g/mL sample layer (Figure 2).
• 20.Store exosomes at −80°C.

CRITICAL: Exosome samples can typically be stored at −80°C for up to 1 year. However, for optimal preservation of biological functionality, usage within 3–6 months is recommended. Avoid repeated freeze-thaw cycles, as this can significantly affect exosome integrity and biological activity. Aliquot samples if frequent usage is anticipated.

Figure 1.

A schematic overview of the experimental workflow for isolating exosomes from cell-free ascitic fluid

The process involves sequential centrifugation steps to remove cells and debris, followed by ultracentrifugation at 100,000 × g to enrich extracellular vesicles. Further purification is performed using iodixanol density gradient ultracentrifugation to isolate exosome-enriched fractions based on buoyant density (δ = 1.14–1.17 g/mL).

Figure 2.

Isolation of exosome-enriched fractions using iodixanol density gradient ultracentrifugation

Representative image showing the formation of exosome-enriched layers following iodixanol-based density gradient ultracentrifugation. The exosome-containing fraction was collected for downstream validation and mass spectrometry–based proteomic profiling.

Exosome production from GAC cell line

Timing: 2–3 days

This step outlines the procedure for generating exosome-rich conditioned medium from GAC cell lines, including AGS and KP-Luc2. Following serum-free culture, exosomes are harvested from the conditioned medium by differential centrifugation and ultracentrifugation. The resulting exosomes are suitable for downstream applications such as proteomic analysis, functional assays, and in vivo delivery studies.

The full workflow is illustrated in Figure 3.

• 21.Culture human AGS cells or murine KP-Luc2 cells until they reach ∼70–80% confluency.
• 22.Wash the cells twice with PBS to remove residual serum and debris.
• 23.Replace the medium with serum-free DMEM without antibiotics.
• 24.Incubate the cells for 48 h at 37°C with 5% CO₂.
• 25.Collect the conditioned medium and centrifuge at 800 × g for 5 min at 4°C to remove cells.
• 26.Transfer the supernatant and centrifuge again at 2,000 × g for 10 min at 4°C to eliminate cell debris.
• 27.Filter the resulting supernatant through a 0.2 μm filter to remove larger vesicles and apoptotic bodies.

Note: Vesicles and cellular debris are larger than 200 nm.

• 28.
  Ultracentrifuge the filtered medium at 100,000 × g for 3 h at 4°C.

  • a.
    The exosome pellet will typically form as a translucent or slightly visible pellet at the bottom of the ultracentrifuge tube.
  • b.
    Carefully remove the supernatant without disturbing this pellet.
• 29.Discard the supernatant and resuspend the exosome-containing pellet in PBS.

Note: Gently pipette PBS along the side of the tube and carefully resuspend the pellet to avoid excessive shear forces that could damage the exosomes.

• 30.Perform a second ultracentrifugation at 100,000 × g for 3 h at 4°C to further purify the exosomes.
• 31.Resuspend the final exosome pellet in 100 μL PBS and stored at −80°C for downstream analyses, for functional assays, or for in vivo studies.

CRITICAL: Exosome samples can typically be stored at −80°C for up to 1 year. However, for optimal preservation of biological functionality, usage within 3–6 months is recommended. Avoid repeated freeze-thaw cycles, as this can significantly affect exosome integrity and biological activity. Aliquot samples if frequent usage is anticipated.

Figure 3.

Workflow for exosome isolation from AGS gastric cancer cells

(A) AGS gastric cancer cells were cultured in serum-free medium for 48 h to eliminate exogenous vesicle contamination. Conditioned media were then collected and filtered through a 0.22 μm membrane.

(B) The filtered supernatant was transferred to ultracentrifugation tubes for downstream processing.

(C) Exosomes were pelleted by ultracentrifugation and resuspended in PBS.

(D) The resuspended exosomes were incubated overnight at 4°C on a rotator to facilitate uniform dispersion and recovery.

Nanoparticle tracking analysis

Timing: 15–20 min/sample

This step describes the use of NTA to quantify and characterize exosomes based on their size and concentration. Exosome samples are diluted to optimal particle density and analyzed using the NanoSight LM10 system. This approach enables reproducible measurement of size distribution and particle concentration for quality control and downstream experimental standardization.

• 32.
  Dilute exosome samples 1:1000 in sterile PBS to achieve an estimated particle concentration of approximately 1×10⁹ particles/mL.

  • a.
    Typically, the concentration of exosomes harvested from conditioned medium of a T225 flask ranges from approximately 1×10¹¹ to 1×10¹² particles/mL, thus requiring dilution to the target concentration.
  • b.
    Ensure the sample is free of aggregates or debris by gently vortexing and filtering through a 0.22 μm syringe filter if needed.
• 33.Load 1 mL of the diluted exosome solution into the sample chamber of the NanoSight LM10.
• 34.Set the temperature to 25°C and the syringe pump speed to 20 for consistent flow.
• 35.Allow the system to equilibrate for 90 seconds before image acquisition.
• 36.Capture three independent 30-second video recordings per sample under identical settings to ensure reproducibility.
• 37.Perform particle size distribution and concentration analysis using NTA software.
• 38.Adjust camera level and detection threshold to maintain optimal particle visibility.

Note: Typically between 30–80 particles per frame.

• 39.Analyze each video independently and report the mean particle size (mode and mean), concentration (particles/mL), and standard deviation (Figure 4).
• 40.Clean the chamber thoroughly between samples with PBS and 70% ethanol to avoid cross-contamination.

Figure 4.

NTA of exosomes derived from patient ascites and AGS gastric cancer cells

Representative NTA profiles showing the size distribution and particle concentration of exosomes isolated from (left) patient-derived ascites and (right) AGS cell-conditioned media. Both populations exhibit a typical exosome size range of approximately 30–150 nm.

Transmission electron microscopy

Timing: 4–5 h

This step outlines the transmission TEM workflow for ultrastructural characterization of exosomes. Fixed vesicles are applied to poly-L-lysine–coated copper grids, stained with uranyl acetate, and visualized under a transmission electron microscope. This method enables high-resolution assessment of exosome morphology, membrane integrity, and size distribution.

• 41.
  Fixation:

  • a.
    Resuspend isolated exosomes in 2.5% glutaraldehyde (in 0.1 M phosphate buffer, pH 7.4).
  • b.
    Incubate at 4°C for 30 min.

Note: This step stabilizes membrane structures.

• 42.Pellet recovery (optional): Centrifuge at 100,000 × g for 1 h at 4°C to remove excess fixative. Gently resuspend the exosome pellet in sterile PBS.
• 43.
  Grid preparation:

  • a.
    Use formvar/carbon-coated copper grids (200 mesh).
  • b.
    Glow discharge grids for 30–60 seconds using a plasma cleaner to make the surface hydrophilic, which improves vesicle adhesion.
• 44.Prepare a control grid using PBS without exosomes to monitor for staining artifacts and contaminants.
• 45.Incubate grids with 0.01% poly-L-lysine for 5 min at room temperature to promote exosome binding.
• 46.Blot off excess solution with filter paper.
• 47.Apply 5–10 μL of fixed exosome suspension to each grid and incubate at room temperature for 1 h in a humidified chamber to prevent evaporation.
• 48.Gently wash the grids five times with PBS, each for 3 min, to remove unbound particles.
• 49.
  Staining:

  • a.
    Stain grids with 1% aqueous uranyl acetate (freshly prepared and filtered) for 1 min.
  • b.
    Blot off excess stain and air-dry the grids completely in a clean, dust-free area for at least 15 min.
• 50.Examine grids under a JEOL JEM-1010 transmission electron microscope at 80 kV.
• 51.Acquire images at ×30,000–80,000 magnification (Figure 5).
• 52.Use multiple grids and fields to ensure representative data.
• 53.Image analysis (optional): Use ImageJ or other analysis software to quantify vesicle diameter and confirm size distribution.

Figure 5.

TEM images of exosomes from ascites and AGS gastric cancer cells

Representative TEM micrographs depicting the morphology of exosomes isolated from (left) patient-derived ascites and (right) AGS gastric cancer cell-conditioned media. Both exosome populations exhibit the characteristic cup-shaped or spherical morphology. Scale bars: 100 nm.

Flow cytometry of bead-bound exosomes

Timing: ∼2 days

This step describes the flow cytometry–based characterization of exosomes using aldehyde/sulfate latex bead coupling and fluorescent antibody staining. Exosomes are immobilized on beads, blocked, and labeled with fluorophore-conjugated antibodies targeting surface markers such as CD9, CD63, and CD81. Flow cytometric analysis enables quantitative assessment of exosome surface marker expression and population purity.

• 54.
  Day 1 – Bead Coupling (Exosome-bead binding).

  • a.
    Mix approximately 5 × 10⁹ purified exosomes with 10 μL of aldehyde/sulfate latex beads in 500 μL PBS.
  • b.
    Rotate gently at room temperature for 15 min to allow initial binding.
  • c.
    Add an additional 400 μL PBS to the mixture to prevent aggregation.
  • d.
    Continue rotating overnight at 4°C on a nutating mixer or rotator.
• 55.
  Day 2 – Blocking and Antibody Staining.

  • a.
    Quenching unreacted sites:

    • i.
      Add 200 μL of 1 M glycine to quench residual aldehyde groups.
    • ii.
      Rotate gently at room temperature for 1 h.
    • iii.
      Centrifuge at 3000 × g for 5 min and discard supernatant.
  • b.
    Blocking:

    • i.
      Resuspend the bead pellet in 100 μL of 10% BSA in PBS.
    • ii.
      Incubate at room temperature for 1 h to block nonspecific binding sites.
    • iii.
      Wash once with PBS containing 2% BSA.
  • c.
    Primary antibody incubation:

    • i.
      Prepare a staining cocktail in 50 μL of 2% BSA/PBS, including fluorophore-conjugated monoclonal antibodies (Figure 6).
    • ii.
      Resuspend the beads in the staining mix and incubate at 4°C for 1 h in the dark.
    • iii.
      Gently vortex every 15–20 min to prevent settling.
  • d.
    Washing:

    • i.
      Wash beads twice with 500 μL PBS + 2% BSA, centrifuging at 3000 × g for 5 min each time.
    • ii.
      Resuspend in 50 μL of 2% BSA/PBS.
    • iii.
      Secondary antibody staining (if needed):
    • iv.
      Incubate with fluorophore-conjugated secondary antibody in 50 μL of 2% BSA for 30 min at 4°C, protected from light.
    • v.
      Wash three times with 2% BSA/PBS as above.
    • vi.
      Final resuspension and analysis:
    • vii.
      Resuspend stained beads in 300 μL PBS for flow cytometry acquisition.
    • viii.
      Analyze samples on a Gallios 561 flow cytometer (Beckman Coulter) or equivalent.
    • ix.
      Use log scale settings for all fluorescence channels.
    • x.
      Set FSC/SSC gates to exclude debris and doublets and apply appropriate single-stain controls and FMO controls for compensation.
  • e.
    Data analysis:

    • i.
      Perform analysis using FlowJo v10.
    • ii.
      Quantify the percentage of bead-bound events positive for each exosome marker (CD9, CD63, CD81) and mean fluorescence intensity (MFI) for relative expression comparisons.

Note: It is recommended to always include isotype controls or beads-only controls to establish background fluorescence. Avoid harsh vortexing, as latex beads can be fragile after coupling. Use freshly prepared exosomes whenever possible and minimize freeze-thaw cycles prior to staining. Stained samples should be stored at 4°C for no more than 24 h and must be protected from light to maintain signal integrity.

Figure 6.

Workflow for flow cytometric detection of exosomes

(A) Schematic illustration of the exosome flow cytometry. Exosomes are incubated with antibodies recognizing surface markers CD9, CD63, and CD81.

(B) Exosome–antibody mixtures are incubated at room temperature with gentle rotation to facilitate binding prior to flow cytometric analysis.

(C) Flow cytometry characterization of bead-coupled EVs isolated from ascites.

Proteomic profiling of EVs by Evosep-timsTOF HT mass spectrometry

Timing: 2–3 days

This step describes the proteomic profiling of exosome-derived proteins using Evosep-timsTOF HT mass spectrometry in data-independent acquisition mode. Following detergent-based lysis, protein reduction, alkylation, and tryptic digestion, peptides are analyzed using high-throughput LC-MS/MS. This workflow enables comprehensive and high-confidence identification of exosomal protein cargo for downstream functional and biomarker analyses.

• 56.
  Lyse exosomes/sEVs by sonicating them in 0.1 M Tris-HCl buffer (pH 8.0) containing 0.2% DDM.

  • a.
    Proceed with one-pot reduction and alkylation by adding 2× concentrated 20 mM TCEP and 80 mM iodoacetamide.
  • b.
    Incubate the reaction mixture in the dark at 95°C for 5 min.
• 57.Digest proteins overnight at 37°C using trypsin at a 1:100 (enzyme: protein) ratio.
• 58.Quantify digested peptide concentration using the tryptophan fluorescence assay.
• 59.
  Load 500 ng of peptides onto EvoTips.

  • a.
    Desalt the peptides and inject them using the Evosep chromatography system with a PepSep C18 column (15 cm × 150 μm, 1.5 μm particle size).
  • b.
    Perform LC-MS/MS acquisition in high-resolution diaPASEF mode using a Bruker timsTOF HT mass spectrometer and a 15 SPD gradient.
• 60.
  Process raw DIA files using Spectronaut v19.7.250203.62635 with the 2024 UniProt human protein database.

  • a.
    Set methionine oxidation (+15.994915) as a variable modification.
  • b.
    Set cysteine carbamidomethylation (+57.021464) as a fixed modification.
  • c.
    Filter identified features at a 1% false discovery rate.

Generation of exoTD-139

Timing: ∼1 day

exoTD-139 refers to exosomes loaded with TD-139, a specific inhibitor targeting Galectin-3. The purpose of generating exoTD-139 is to utilize exosomes as carriers for targeted delivery of TD-139 to recipient cells or tissues, thereby enhancing therapeutic efficacy and specificity compared to free drug administration.

• 61.
  Isolation of Exosomes from AGS or KP-Luc2 Cells.

  • a.
    Culture AGS and KP-Luc2 cells in T-175 flasks to 70–80% confluency.
  • b.
    Replace the growth medium with serum-free, antibiotic-free DMEM and incubate for 48 h.
  • c.
    Collect the conditioned media and perform differential centrifugation:
  • d.
    Centrifuge 800 × g for 5 min at 4°C.
  • e.
    Centrifuge 2,000 × g for 10 min at 4°C.
  • f.
    Filter the supernatant through a 0.22 μm filter.
  • g.
    Ultracentrifuge at 100,000 × g for 3 h at 4°C.
  • h.
    Wash the exosome pellet with PBS and repeat ultracentrifugation.
  • i.
    Resuspend purified exosomes in sterile PBS and quantify particle concentration using NTA.
  • j.
    Adjust to 1 × 10⁹ particles/mL before proceeding.
• 62.
  TD-139 Loading into exosomes.

  • a.
    Mix ∼1 × 10⁹ exosomes with 300 μg of TD-139 in a total volume of 500 μL PBS.
  • b.
    Incubate the mixture at room temperature for 24 h with gentle rotation (e.g., nutator) to facilitate passive loading.
• 63.
  Removal of Free (Unbound) TD-139.

  • a.
    Transfer the exoTD-139 mixture to Amicon Ultra centrifugal filters (30 kDa cutoff).
  • b.
    Centrifuge at 4,000 × g for 10 min at 4°C.
  • c.
    Wash the retained exosome fraction twice with sterile PBS by adding 500 μL PBS and repeating centrifugation.
• 64.
  Recovery and Storage.

  • a.
    Recover the purified exoTD-139 fraction by inverting the Amicon filter into a clean collection tube and spinning briefly at 1,000 × g at 4°C.
  • b.
    Quantify particle concentration using NTA if needed.
  • c.
    Use freshly prepared exoTD-139 immediately for downstream assays (e.g., cell treatment, uptake assay, in vivo injection).
  • d.
    If necessary, store at 4°C short-term (within 24 h) or at −80°C for longer-term storage; avoid repeated freeze–thaw cycles.

Note: Ensure that the TD-139 stock is sterile-filtered prior to use. Although passive loading is simple, it may result in low encapsulation efficiency; electroporation or sonication can be considered for improved loading with appropriate optimization. To quantify TD-139 incorporation, perform UV-Vis spectroscopy or LC-MS analysis after washing. Include a vehicle control using DMSO-treated exosomes in all functional assays to account for potential solvent effects.

Expected outcomes

Primary patient-derived epithelial tumor cells isolated from malignant ascites can be successfully cultured, with adherent colonies typically emerging within 7 to 10 days.^2,4 This model provides a clinically relevant platform for studying peritoneal metastasis in cancers such as diffuse-type gastric carcinoma and high-grade serous ovarian cancer, where ascitic fluid is often abundant and tumor cell-rich.^5,6,7 Once cultured, these cells serve as a valuable source for isolating EVs, which range from 30 to 150 nm in diameter, as confirmed by NTA and TEM. EVs derived from ascites or cell culture supernatants harbor molecular signatures reflective of their cell of origin, making them potent mediators and biomarkers of disease. The single-step density gradient flotation approach described here facilitates efficient and rapid isolation of highly enriched circulating exosomes /small extracellular vesicles from background non-vesicular contaminants including soluble proteins and immune complexes via a single four-hour ultracentrifugation step. For deep-discovery molecular cargo profiling, this method is preferred as it enables separation of bona fide vesicle subpopulations from extracellular non-lipid containing extracellular particles such as protein-protein and protein-nucleic acid complexes. A limitation of this approach is that some lower density lipoprotein subclasses present in biofluids may co-fractionate along with the exosome/sEV yields as is common with many EV isolation methods. The method presented here balances purity and yield for the purpose of untargeted molecular profiling. Advanced proteomic profiling using LC-HDMSE yields high-confidence EV cargo characterization, enabling insight into tumor progression and intercellular communication. Combined imaging and biophysical techniques such as TEM and NTA ensure rigorous assessment of vesicle morphology, size distribution, and particle concentration. Furthermore, flow cytometry using bead-coupled EVs confirms the presence of canonical surface markers including CD9, CD63, and CD81, validating the identity of the isolated vesicles. Functionally, exoTD-139, exosomes loaded with the galectin-3 inhibitor TD-139, demonstrate robust performance in both in vitro functional assays and in vivo delivery models, supporting their potential use in targeted modulation of the tumor microenvironment.

Limitations

Density gradient ultracentrifugation effectively separates EVs based on their density, producing samples with fewer contaminants like proteins and non-vesicular particles compared to simpler methods. This higher purity makes subsequent experiments, such as proteomic analysis or functional assays, more reliable. However, density gradient ultracentrifugation is slower, requires special equipment, and involves more technical handling than other isolation methods, like size-exclusion chromatography or polymer-based precipitation, which might yield more EVs. In summary, although density gradient ultracentrifugation takes longer and might yield fewer EVs, its main strength is providing cleaner EV samples, essential for accurate downstream analysis.

The protocol has several potential limitations that may affect reproducibility and downstream applications. The yield and purity of patient-derived tumor cells are highly dependent on both the tumor cell content and the volume of ascitic fluid obtained, which can vary significantly across patients. During early culture, there is a risk of contamination with fibroblasts or mesothelial cells, which may compromise the epithelial tumor cell population. The efficiency of exosome recovery and purity is also sensitive to ultracentrifugation parameters and gradient consistency, particularly when using iodixanol-based separation. Finally, loading of TD-139 into exosomes using a passive diffusion method may result in variable drug encapsulation efficiency, potentially affecting the biological potency of exoTD-139 preparations.

Troubleshooting

Problem 1

Low cell yield from ascites (related to Step 1).

Culturing epithelial tumor cells from ascitic fluid often results in a low cell yield due to limited tumor cell content, particularly in samples rich in inflammatory or stromal cells. This variability can hinder downstream analysis and experimental reproducibility. To ensure optimal results, consider the following strategies.

Potential solution

• •Ascitic fluid volume: Increasing the volume of ascites collected per patient can improve tumor cell yield. Alternatively, pooling ascites samples from patients with similar clinical backgrounds may also enhance cell recovery.
• •Rapid processing: Promptly process the collected ascites within 2 h to maintain cell viability and minimize cell loss.
• •Optimized culture conditions: Maintain consistent culture conditions, such as temperature, media composition, and gas environment (37°C, 5% CO₂), to support the survival and proliferation of tumor cells.

Problem 2

Variability in ascitic fluid composition across patient samples (related to Step 1).

Variations in ascitic fluid composition, influenced by clinical factors such as prior chemotherapy or infection, can lead to inconsistent experimental outcomes.

Potential solution

• •Documentation: Maintain detailed records of clinical background and sample characteristics for each patient-derived sample.
• •Biological replicates: Include multiple biological replicates to account for variability and enhance experimental robustness.
• •Validation: Validate findings across multiple patient samples to confirm reproducibility and reliability of experimental results.

Problem 3

Fibroblast overgrowth in primary cultures (related to Step 1).

Fibroblasts from stromal contamination can rapidly overtake slower-growing epithelial cancer cells in primary cultures derived from ascites. This can compromise the purity and utility of the epithelial tumor cell populations.

Potential solution

• •Cell sorting: Employ fluorescence-activated cell sorting or magnetic-activated cell sorting (MACS) with epithelial-specific markers to enrich tumor cell populations and reduce stromal contamination before culturing.
• •Selective adherence: Carefully control plating times and culture conditions to preferentially support epithelial cell attachment and minimize fibroblast growth.

Problem 4

Poor EV recovery (related to Steps 2 and 3).

Low EV yields are commonly due to issues with density gradient ultracentrifugation, particularly during iodixanol layering.

Potential solution

• •Gradient preparation: Ensure careful and accurate layering of the iodixanol gradients to maintain clear separation of EV fractions.
• •Controlled conditions: Strictly regulate ultracentrifugation parameters (speed, time, temperature) to optimize EV recovery.
• •Quality reagents: Regularly verify gradient integrity and utilize freshly prepared reagents to enhance yield and consistency.

Problem 5

Weak EV staining in downstream analyses (related to Steps 2 and 3).

Weak or inconsistent EV staining in methods such as flow cytometry can result from suboptimal antibody performance or low target expression.

Potential solution

• •Antibody optimization: Conduct titration experiments to determine optimal antibody concentrations and incubation durations.
• •Validated reagents: Use antibodies specifically validated for EV surface markers to ensure reliable detection.
• •Controls: Include both positive and negative staining controls to clearly distinguish specific signals from background noise.

Problem 6

Lack of functional effect with exoTD-139 (related to Step 8).

Failure to observe biological responses using exosome-loaded TD-139 may indicate issues with loading efficiency or drug integrity.

Potential solution

• •Drug verification: Confirm the biological activity and stability of TD-139 prior to loading onto exosomes.
• •Enhanced loading: Extend incubation periods between exosomes and TD-139 or consider alternative loading methods (e.g., electroporation or sonication).
• •Analytical confirmation: Utilize analytical techniques such as nanoparticle tracking analysis, high-performance liquid chromatography (HPLC), or mass spectrometry to quantitatively confirm loading efficiency.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to the lead contact, Jaffer A. Ajani (jajani@mdanderson.org).

Technical contact

Technical questions on executing this protocol should be directed to and will be answered by the technical contacts, Yibo Fan (yfan3@mdanderson.org) and Jody V. Vykoukal (jvykouka@mdanderson.org).

Materials availability

This study did not generate new unique reagents.

Data and code availability

No custom code was used in this protocol.

Acknowledgments

We are thankful to Tracy Stafford (Scientific Publications Services, Research Medical Library, The University of Texas MD Anderson Cancer Center) for the editorial assistance. This research was funded by Public Health Service Grant DF56338 (S.S.); Department of Defense grants CA160433, CA170906, CA160445, CA200990, and CA210439 (S.S. and J.A.A.); National Institutes of Health grants CA129906, CA138671, and CA172741 (J.A.A.); and K99 CA286745 to G.Z. The research was also supported in part by the Caporella family, the Park family, the Dallas family, the Dio family, the Frankel family, the Kushner family, the Kohn family, the Smith family, an anonymous donor, the Richard L. Duchossois Memorial Foundation, the Shipman family, the McNeil family, the Stupid Strong Foundation (Dallas, TX), and the Gastric Cancer Foundation (San Francisco, CA) (J.A.A.), as well as The University of Texas MD Anderson Cancer Center multidisciplinary grant programs.

Author contributions

Conceptualization, J.A.A.; methodology, Y.F., S.S., M.P.P., G.Z., J.V.V., H.K., K.Y., J.J., G.A.C., R.E.W., L.W., and S.H.; investigation, Y.F., M.P.P., and J.V.V.; visualization, Y.F., S.S., M.P.P., J.V.V., and J.A.A.; funding acquisition, S.S. and J.A.A.; project administration, S.S. and J.A.A.; supervision, J.A.A.; writing – original draft, Y.F. and J.A.A.; writing – review and editing, Y.F., J.V.V., H.K., and J.A.A.

Declaration of interests

The authors declare no competing interests.