Cell and mouse study links ELAVL1 to cervical cancer progression

A laboratory study combining cervical cancer cell experiments, molecular assays and a mouse xenograft model found that elevated ELAVL1 was associated with cervical cancer cell proliferation, migration and invasion. The researchers also identified a regulatory sequence linking ELAVL1 to activin A receptor-like type 1, or ACVRL1, and found that ACVRL1 expression was connected to caspase-14, or CASP14. Silencing ELAVL1 reduced ACVRL1 and CASP14 expression, while inhibiting either downstream factor partially reversed the malignant behaviors produced by ELAVL1 overexpression. These findings support a possible ELAVL1–ACVRL1–CASP14 regulatory pathway, although the experiments do not establish clinical effectiveness in patients.

ELAVL1, or Elav-like RBP 1, is an RNA-binding protein that recognizes AU-rich elements in target messenger RNAs and can influence RNA stability and translation. The study was motivated by evidence that ELAVL1 is frequently overexpressed in human malignancies and has previously been associated with aggressive features of cervical cancer. Its expression in cervical tumors has also been linked in earlier work to stabilization of messenger RNAs encoding proteins such as cyclin B1, TPD52 and HOXC6. The new experiments focused on whether ACVRL1 and CASP14 participate in a previously uncharacterized route through which ELAVL1 influences cervical cancer cells.

Expression analysis showed that ELAVL1 messenger RNA and protein levels were higher in the cervical cancer cell lines SiHa and HeLa than in the non-tumorigenic cervical epithelial cell line HCerEpiC. These comparisons used RT-qPCR and western blot analysis, with three technical replicates from three biological replicates for each cell line. ELAVL1 levels were also higher in cervical cancer tissues than in patient-matched adjacent non-tumor samples. The researchers additionally examined a public gene-expression dataset through TNMplot. In that independent analysis of 14 uterine tumor tissues and 14 paired adjacent normal tissues, ELAVL1 expression was significantly higher in tumor samples, with a reported p value of 0.00695. The experimental and public-dataset results therefore converged on ELAVL1 overexpression in cervical and uterine tumor tissue, although the clinical comparison was limited by its small sample size and did not establish patient outcomes.

Functional experiments in SiHa and HeLa cells indicated that ELAVL1 supports several malignant characteristics. Researchers reduced ELAVL1 expression with small interfering RNA and measured cell growth, migration, invasion and apoptosis. CCK-8 assays showed lower proliferation after knockdown, while wound-healing and Transwell assays showed reduced migration and invasion. For the wound-healing experiments, cells were pretreated with 2 μg/mL mitomycin C to reduce interference from proliferation, and migration was evaluated 48 hours after transfection. Cells with silenced ELAVL1 also had a higher proportion of apoptotic cells in flow-cytometry measurements and increased levels of cleaved caspase-3 and cleaved poly(ADP-ribose) polymerase, or PARP, proteins. In the reciprocal experiment, overexpressing ELAVL1 in SiHa cells increased proliferation, wound closure and invasion. Fluorescence imaging further associated ELAVL1 knockdown with reduced formation of F-actin-rich migratory and invasive protrusions. Together, these observations establish a functional association between ELAVL1 expression and aggressive cell behavior in the two cell-line models, without demonstrating that ELAVL1 alone initiates cervical cancer.

The investigators then used the ENCORI/starBase RNA-binding-protein prediction tool to search for a possible connection between ELAVL1 and ACVRL1. The analysis predicted an ELAVL1-binding site in the ACVRL1 messenger RNA 3′ untranslated region at chromosome 12 position 51922109. An RNA immunoprecipitation followed by quantitative PCR supported physical association between ELAVL1 protein and ACVRL1 RNA: ACVRL1 RNA was enriched in the ELAVL1 immunoprecipitate, while the immunoglobulin G control produced only faint nonspecific signal. In dual-luciferase reporter assays, ELAVL1 knockdown reduced activity from a reporter carrying the wild-type ACVRL1 3′ untranslated region. Altering the predicted binding sequence removed this response, suggesting that the candidate region participates in ELAVL1-dependent regulation. These experiments support a post-transcriptional interaction, although they do not establish the precise biochemical mechanism or whether ACVRL1 messenger RNA stabilization is the only consequence of the binding.

ACVRL1 and CASP14 were also investigated as potential interacting proteins. The study cited previous liquid chromatography–mass spectrometry screening that had identified CASP14 among 11 candidate ACVRL1-interacting proteins, then used AlphaFold3 to predict a possible structural interface between the two proteins. Immunofluorescence imaging showed that ACVRL1 and CASP14 colocalized in the nuclei of SiHa and HeLa cells, with CASP14 stained green and ACVRL1 stained red. These findings offer a structural and cellular basis for examining the pathway, but a predicted interface and spatial overlap alone do not demonstrate direct protein binding or a causal intracellular signaling sequence. The authors therefore tested the proposed regulatory relationships with gene-silencing experiments rather than relying on the imaging results alone.

When ELAVL1 was knocked down in SiHa and HeLa cells, both messenger RNA and protein levels of ACVRL1 and CASP14 decreased. Silencing ACVRL1 reduced CASP14 messenger RNA and protein levels without producing a significant change in ELAVL1 expression. In contrast, reducing CASP14 did not noticeably alter ELAVL1 or ACVRL1 expression. This ordering was consistent with an indirect model in which ELAVL1 acts upstream of ACVRL1, which in turn influences CASP14 abundance. The results do not, however, exclude feedback, intermediary regulators or other target RNAs, and knockdown experiments cannot by themselves establish direct causation for every step of the proposed pathway. The observations were measured in cell cultures, not in patients.

Co-transfection experiments provided functional evidence that ACVRL1 and CASP14 contribute to the phenotypes associated with ELAVL1. In SiHa cells overexpressing ELAVL1, simultaneous knockdown of ACVRL1 partially reversed the increases in proliferation, migration and invasion. Knocking down CASP14 produced similar partial reversals of all three behaviors. Additional flow-cytometry analysis showed that CASP14 depletion increased the proportion of apoptotic cells in SiHa and HeLa cultures relative to their negative controls. Because the downstream interventions only partially offset the ELAVL1 phenotype, the findings also suggest that ELAVL1 influences other targets or regulatory modules. The experiments nevertheless support a measurable contribution from the ACVRL1/CASP14 axis to the cellular effects of ELAVL1 in cervical cancer cells. The study used multiple independent biological replicates for the functional assays, with technical replicates used for selected molecular measurements.

The study extended its findings to an in vivo model using SiHa cells implanted subcutaneously in female nude mice. Ten mice were used for the xenograft experiment, with five assigned to the control condition and five to the ELAVL1-silencing condition. Once tumors had formed, researchers administered either an in vivo control or ELAVL1-targeting small interfering RNA at multiple tumor sites. RT-qPCR analysis showed that ELAVL1 messenger RNA was reduced in treated tumors. Tumor volume was significantly smaller in the ELAVL1-silencing group than in the control group, tumor weight was also significantly lower, and the messenger RNA level of the proliferation marker Ki-67 was reduced. The xenograft result indicates that local ELAVL1 silencing can suppress tumor growth in this particular mouse model. It does not demonstrate that a clinically deliverable ELAVL1 inhibitor would be safe or effective in people, and the subcutaneous xenograft reproduces only part of human cervical cancer biology.

Overall, the experiments support ELAVL1 as a cancer-promoting regulator in HPV-associated cervical cancer cell models and nominate ACVRL1 and CASP14 as components of a possible downstream pathway. SiHa cells are HPV16-positive, while HeLa cells are HPV18-positive, linking the cell models to the high-risk HPV types emphasized in the study. The research connects ELAVL1 binding to the ACVRL1 3′ untranslated region with changes in pathway expression, malignant cell behaviors and reduced tumor growth after knockdown. Important limitations include the small clinical sample, reliance on two cell lines and a mouse xenograft, possible influence from other ELAVL1 targets, and the absence of patient outcomes or clinical intervention data. The findings therefore identify a testable regulatory network and a potential target for further research, but they do not yet establish that manipulating this pathway is a safe or effective cervical cancer treatment in humans.

Subject of Research: Mechanistic study of ELAVL1 regulation and cervical cancer progression through the ACVRL1/CASP14 pathway.

Article Title: ELAVL1 overexpression potentiates cervical cancer malignancy via regulating the ACVRL1/CASP14 pathway

Article References: Liu, J., Wang, J., Gao, Y., Cui, X., Wang, W., Ma, H., Jiao, Y., Xu, J., Hao, J., Xu, Z., Li, L., & Hao, M. (2026). ELAVL1 overexpression potentiates cervical cancer malignancy via regulating the ACVRL1/CASP14 pathway. iScience, 29(11), Article 117956. https://doi.org/10.1016/j.isci.2026.117956

Image Credits: AI Generated

DOI: 10.1016/j.isci.2026.117956

Keywords: ELAVL1, cervical cancer, ACVRL1, CASP14, RNA-binding protein, HPV16, HPV18, cancer cell models, molecular regulation, xenograft tumors, overexpression, potentiates