An RNA called mt-5'-tiRNA-Tyr was found at higher levels in colorectal cancer tissue and plasma, and its level distinguished colorectal cancer from comparison groups in an internally split plasma analysis. In the same study, changing the RNA in colorectal cancer cells was associated with changes in proliferation, the protein HARS2, mitochondrial protein production and succinate. The researchers describe mt-5'-tiRNA-Tyr as a candidate biomarker and a possible part of a biological pathway, while noting that the diagnostic result has not been tested in an independent external cohort.
From a broad screen to one candidate
The work combined exploratory PANDORA-seq, qRT-PCR validation, functional assays and RNA and protein mechanistic methods. The sequencing screen identified 1,631 small RNAs called tsRNAs, and 127 met the reported high-expression criteria: a fold change greater than 2 and a p-value below 0.05. The paper then examined mt-5'-tiRNA-Tyr across human specimens and laboratory models.
The human specimen component included 48 colorectal cancer and adjacent-tissue pairs. Plasma samples came from 104 colorectal cancer patients, 98 healthy donors, 50 intestinal-polyp patients and 32 postoperative colorectal cancer patients. The study used these groups for tissue comparisons and plasma discrimination.
Expression was higher in colorectal cancer tissue than adjacent tissue, and higher tissue expression was associated with poorer overall survival. The report also found significantly higher mt-5'-tiRNA-Tyr in colorectal cancer plasma. The study does not report exact expression fold changes or confidence intervals, so the direction of the association is clearer than its size.
A promising signal, with a narrow test
Using AUC, a summary score for how well a test separates groups, the standalone RNA scored 0.912 in the training data, with a 95% confidence interval of 0.858 to 0.966. At the reported cutoff of 3.452, sensitivity was 0.836 and specificity was 0.956. A combined panel scored 0.937, with a 95% confidence interval of 0.896 to 0.977, sensitivity of 0.849 and specificity of 0.912.
In internal validation, standalone mt-5'-tiRNA-Tyr had an AUC of 0.889, with a 95% confidence interval of 0.794 to 0.984, sensitivity of 0.710 and specificity of 0.933. The combined panel had an AUC of 0.941, with a 95% confidence interval of 0.887 to 0.995, sensitivity of 0.839 and specificity of 0.933. Reported AUCs for early-stage disease were 0.873 and 0.903, and for distinguishing colorectal cancer from intestinal polyps were 0.820 and 0.855. DeLong comparisons favored mt-5'-tiRNA-Tyr over most conventional markers, but the combined panel did not significantly outperform the RNA alone.
A proposed link to cell growth
To explore a possible mechanism, the researchers used RNA pull-down and RNA immunoprecipitation assays. The sense mt-5'-tiRNA-Tyr probe recovered more HARS2 than the antisense probe, and HARS2 immunoprecipitates were enriched for the RNA. These findings support an association between mt-5'-tiRNA-Tyr and HARS2. They did not quantify binding affinity or stoichiometry, and they did not establish direct displacement of mt-tRNA-His.
Further experiments linked the RNA to HARS2 succinylation, a chemical modification of the protein. Overexpression of mt-5'-tiRNA-Tyr was associated with higher HARS2 succinylation, while the HARS2-K91R variant showed lower succinylation and faster degradation than HARS2-WT, supporting K91 as the primary tested site. CPT1A overexpression was associated with higher HARS2 succinylation than KAT14 overexpression, and CPT1A knockdown with lower succinylation. The study identifies CPT1A as a candidate regulator, but its catalytic specificity and the spatial basis of its proximity to HARS2 were not established.
In cell experiments, mt-5'-tiRNA-Tyr mimics were associated with lower levels of mitochondria-encoded proteins and less newly made mitochondrial protein, a measure described as nascent mitochondrial translation. The mimics were also associated with altered HARS2-RNA binding and less mt-tRNA-His bound to HARS2. HARS2 overexpression and the K91R variant showed recovery of translation-related measures. Direct HARS2 aminoacylation activity and mt-tRNA-His charging were not measured, leaving that proposed step untested.
Metabolite findings fit the authors' proposed succinate feedback model. Inhibiting mt-5'-tiRNA-Tyr was associated with lower succinate and higher SDH activity, while overexpression showed the opposite pattern. Adding succinate was associated with increased HARS2 succinylation and with rescue of cell proliferation and xenograft growth. But succinate abundance and SDH activity do not establish the rate of flow through the relevant metabolic cycle or show that physiological succinate levels are sufficient for the proposed feedback relationship.
The cell and animal results were directionally consistent: inhibition coincided with lower colorectal cancer proliferation, overexpression with higher proliferation, and stable down-regulation with smaller xenograft growth and tumor weight. These were nonrandomized preclinical perturbation experiments, and exact effect sizes were not reported. They provide evidence from models, not direct evidence that the same effects occur in human patients.
What the study leaves open
The plasma findings need independent external validation. Biochemical work would also need to resolve whether mt-5'-tiRNA-Tyr directly displaces mt-tRNA-His under endogenous conditions, whether K91 succinylation changes HARS2's tRNA-charging activity, and whether endogenous succinate can reach levels sufficient to support the proposed feedback relationship. Until then, mt-5'-tiRNA-Tyr remains a candidate biomarker alongside a preclinical mechanistic model.
Paper data and sources
Original title: A Novel tRNA Half mt-5'-tiRNA-Tyr Promotes Colorectal Cancer Proliferation via Inducing HARS2 Succinylation.
Authors: Xinliang Gu, Danping Zhu, Xinwei Liu et al.
Journal/Repository: Advanced science (Weinheim, Baden-Wurttemberg, Germany)
Status: Peer-reviewed
First online: 2026-08-21
DOI: 10.1002/advs.77042
Original paper