Extracellular vesicles (EVs) are phospholipid bilayer vesicles secreted by cells that carry proteins and nucleic acids reflective of their cells of origin. Because EVs circulate in body fluids and remain relatively stable, they are attractive for biomarker discovery. However, EVs isolated from blood and other biofluids often lack tissue specificity, complicating efforts to trace their organ of origin and limiting diagnostic precision for diseases such as hepatocellular carcinoma (HCC).
To address this limitation, interest has shifted toward sampling the tissue interstitial fluid (TIF), the extracellular fluid that bathes cells in tissues. TIF contains EVs secreted locally and may preserve tissue-specific molecular signatures earlier and more directly than circulating biofluid EVs. TIF-derived EVs therefore represent a bridge between tissue biopsy and liquid biopsy and have been explored for diagnostic applications across multiple organs.
This study focused on optimizing the isolation of small EVs from hepatic TIF (TIF-sEVs) and profiling their non-coding RNA (ncRNA) cargo to identify candidate biomarkers specific to HCC.
The authors developed an optimized workflow for isolating TIF-sEVs from surgically obtained liver tissues. The protocol combined enzymatic digestion with differential and ultracentrifugation steps. Multiple digestion conditions and centrifugation repetitions were compared to determine a balance between EV yield and sample purity.
The optimal enzymatic conditions identified were digestion with Collagenase D (2 mg/mL) together with DNase I (40 U/mL) for 30 minutes. Compared with longer digestion (60 minutes) and no-digestion controls, the 30-minute condition produced EV preparations with fewer non-vesicular contaminants while preserving vesicle morphology. Additionally, repeating differential centrifugation steps reduced impurities relative to a single differential centrifugation.
Isolated particles were characterized by transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA), and Western blotting (WB). TEM showed the classic ‘cup-shaped’ EV morphology across tumor-derived (TEVs) and adjacent non-tumor-derived EVs (NEVs) under the tested conditions. The 0-minute and 30-minute digestion groups had cleaner TEM backgrounds, while the 60-minute group contained visible protein aggregates, lipoproteins, and other non-vesicular structures in addition to EVs.
NTA measurements indicated that the particle populations were predominantly below 200 nm, consistent with small EV size ranges. Western blots confirmed the presence of canonical EV marker proteins ALIX and CD63. The preparations also expressed the hepatocyte-specific membrane protein ASGPR, supporting the liver origin of the isolated vesicles.
Collectively, these characterizations indicated the optimized protocol produced TIF-sEVs that met expected morphological, size, and marker criteria.
The study selected 33 candidate ncRNAs for quantitative RT-PCR analysis: 11 long non-coding RNAs (lncRNAs) derived from TCGA-LIHC data and 22 microRNAs (miRNAs) from TCGA-LIHC and the GSE302990 dataset. qRT-PCR of TIF-sEV preparations demonstrated specific enrichment of multiple ncRNAs in TIF-sEVs compared with controls.
Examples of enriched lncRNAs included AL031985 and TMCC1-AS1, among a total of six lncRNAs reported as specifically enriched. Enriched miRNAs included miR-1224-5p and miR-483-5p, among eleven miRNAs identified as enriched in TIF-sEVs. The paper reports these enriched ncRNAs as candidate tissue-specific markers carried by hepatic TIF-sEVs.
To evaluate diagnostic potential, the authors integrated qRT-PCR results for all 33 candidate ncRNAs with clinical parameters using LASSO regression. The combined model selected five variables: serum ALB, PLT, DBIL, the lncRNA GAS5, and miR-194-5p. This composite signature achieved an area under the curve (AUC) of 0.960 for distinguishing HCC in the study dataset.
The report presents this signature as a proof-of-concept for combining TIF-sEV ncRNA content with routine clinical measures to improve HCC diagnostic performance.
The optimized enzymatic digestion (Collagenase D + DNase I for 30 minutes) plus repeated differential centrifugation yielded TIF-sEVs with expected morphology, size distribution, and marker expression while reducing non-vesicular contaminants. Detection of ASGPR alongside canonical EV markers supports hepatic origin, addressing one of the primary limitations of circulating EV biomarkers: tissue specificity.
Identification of multiple enriched ncRNAs in TIF-sEVs, and derivation of a high‑performing combined diagnostic signature (AUC 0.960) integrating ncRNA markers with routine clinical parameters, suggests that liver TIF-sEV profiling can provide tissue-tracing biomarker candidates for HCC. The authors position TIF-sEV analysis as an approach that connects tissue-level specificity with minimally invasive sampling strategies.
The study remains exploratory; the text reports assay development, candidate selection, and internal modeling, and it implies the need for further validation in larger, independent cohorts.
The paper describes an extraction workflow using surgical liver tissue followed by enzymatic digestion and sequential centrifugation culminating in ultracentrifugation to isolate small EVs. Characterization methods included TEM, NTA, and Western blotting for EV markers (ALIX, CD63) and the hepatocyte marker ASGPR. Candidate ncRNAs were selected from TCGA-LIHC and GSE302990 sources and profiled by qRT-PCR. LASSO regression was applied to integrate molecular and clinical variables.
Detailed experimental parameters, datasets, and supporting files are reported in the article and its Supporting Information.
This study establishes an optimized, reproducible protocol for isolating liver-derived TIF-sEVs using Collagenase D and DNase I digestion for 30 minutes with repeated differential centrifugation, and it demonstrates that TIF-sEVs carry enriched ncRNAs that may serve as tissue-specific biomarkers for HCC. A combined diagnostic signature incorporating ALB, PLT, DBIL, lncRNA GAS5, and miR-194-5p achieved an AUC of 0.960 in the reported analysis, highlighting the potential of integrating TIF-sEV molecular profiles with clinical measures. Further external validation is implied but not reported in the source.