Claudin-1 Distinguishes Stress-Induced Premature Senescence in Human Vascular Smooth Muscle Cells

In a study published in GeroScience, researchers identified claudin-1 as a selective marker capable of distinguishing stress-induced premature senescence from replicative senescence in human vascular smooth muscle cells. Cardiovascular diseases remain the leading global cause of mortality, with disorders like hypertension, myocardial infarction, and stroke frequently originating from atherosclerosis, an inflammatory arterial disease marked by vessel narrowing from atherosclerotic plaque accumulation. Arterial plaques are largely composed of vascular smooth muscle cells that display hallmarks of cellular senescence, an irreversible cessation of cell division. This arrest arises through two main routes: replicative senescence, caused by telomere erosion and exhausted proliferative capacity from repeated divisions, or stress-induced premature senescence, triggered by external insults such as oxidative stress or chemotherapeutic agents. Although both senescence forms share established markers, including heightened senescence-associated beta-galactosidase activity, increased p53, p21, and p16, and reduced lamin B1 and Ki67 expression, distinguishing between them in vascular tissue has remained an unresolved challenge.

To identify unique molecular signatures of stress-induced premature senescence, the authors established controlled in vitro culture models using human vascular smooth muscle cells from young male donors. Premature senescence was induced by treating young cells with one hundred nanomolar doxorubicin, a DNA-damaging agent, or seven point five micromolar curcumin, which induces senescence independently of direct DNA damage. Replicative senescence was established by serial passaging until cells lost division potential, extending up to passage twenty-two. Senescence induction was validated against young controls: control populations exhibited bromodeoxyuridine incorporation above seventy percent and senescence-associated beta-galactosidase activity below ten percent, whereas senescent cultures displayed under five percent bromodeoxyuridine-positive cells and over eighty percent beta-galactosidase positivity. Transcriptomic profiling using Affymetrix HuGene arrays identified CLDN1, encoding claudin-1, as the most substantially upregulated transcript in prematurely senescent cells, showing tenfold elevation in doxorubicin-treated cells and fourfold increase in curcumin-treated cells, while remaining unchanged in replicative senescence. Crucially, CLDN1 was the only member among twenty-four claudin genes displaying this selective induction.

To confirm the microarray results, researchers performed quantitative real-time PCR using the Pfaffl method, normalizing target expression against reference genes GAPDH, EIF2B, and DAD1. Quantitative amplifications verified robust CLDN1 transcript upregulation in doxorubicin-treated and curcumin-treated vascular smooth muscle cells. However, qPCR assays also revealed that cells undergoing replicative senescence exhibited an approximate twofold increase in CLDN1 mRNA relative to young controls. To evaluate whether increased mRNA translated into protein, the authors conducted Western blot analyses normalized to GAPDH. Strikingly, claudin-1 protein expression was elevated exclusively in cells undergoing stress-induced premature senescence. In replicatively senescent cells, despite the twofold mRNA increase, claudin-1 protein remained undetectable or at negligible baseline levels matching control cells. This divergence demonstrated an uncoupling between transcriptional activity and protein translation, indicating that while promoter activation occurs across both senescence pathways, specific post-transcriptional repressive mechanisms, potentially mediated by regulatory microRNAs, actively suppress claudin-1 protein accumulation during replicative senescence.

To investigate the epigenetic mechanisms governing CLDN1 gene expression, the research team analyzed promoter histone modifications using chromatin immunoprecipitation coupled with sequencing. The ChIP-seq experiments examined two distinct histone marks: trimethylated lysine 4 of histone H3 (H3K4me3), an epigenetic mark of active transcription and open promoter chromatin, and trimethylated lysine 9 of histone H3 (H3K9me3), indicative of condensed, transcriptionally silent heterochromatin. In proliferating young control cells, H3K4me3 formed a narrow peak of 1,646 base pairs directly at the transcription start site, displaying the highest signal intensity. In contrast, prematurely senescent cells treated with curcumin or doxorubicin and replicatively senescent cells all exhibited broadened, lower H3K4me3 peaks measuring 3,078 base pairs, 2,295 base pairs, and 2,354 base pairs, respectively. Concurrently, H3K9me3 immunoprecipitation showed no significant enrichment across the gene under any experimental condition. These profiles revealed that an enlarged, permissive H3K4me3 domain promotes chromatin accessibility at the CLDN1 promoter in both senescence states, confirming that open chromatin drives transcription while downstream regulatory mechanisms govern protein translation.

Beyond total protein abundance, the spatial distribution of claudin-1 within the cell established a notable divergence between senescence modalities. Claudins canonically operate as integral transmembrane components of tight junctions alongside occludins, junctional adhesion molecules, and zonula occludens-1, sealing intercellular spaces and regulating paracellular permeability across epithelial and endothelial barriers. Vascular smooth muscle cells, however, do not assemble tight junctions under physiological conditions. Using immunofluorescence microscopy and Western blot analyses of biochemically isolated cytoplasmic and nuclear fractions, validated for purity using GAPDH and HDAC1, researchers mapped the intracellular localization of claudin-1. In young control and replicatively senescent vascular smooth muscle cells, weak residual claudin-1 localized strictly within the cytoplasm or was undetectable. In stark contrast, prematurely senescent cells displayed pronounced claudin-1 accumulation predominantly concentrated within the cell nucleus. Parallel evaluations across human umbilical vein endothelial cells, immortalized keratinocytes, and epithelial cancer cell lines confirmed classical junctional membrane localization, underscoring that nuclear claudin-1 accumulation represents an atypical, cell-type-specific phenotype restricted to prematurely senescent vascular smooth muscle cells.

To confirm that claudin-1 upregulation represents a true hallmark of premature senescence rather than a secondary artifact of high cell density or reversible division arrest, researchers evaluated additional experimental conditions. Proliferating young vascular smooth muscle cells were maintained in serum-free medium for seven days to induce quiescence, a reversible arrest where cells resume division upon serum reintroduction. In parallel, young control and replicatively senescent cells were cultured at confluences exceeding ninety percent, corresponding to densities above nine thousand cells per square centimeter. Immunoblotting demonstrated that quiescent cells maintained detectable lamin B1 expression, confirming the absence of permanent senescence, while anillin, an essential cytokinesis protein and marker of dividing cells, was absent in all non-proliferating states. Crucially, claudin-1 protein expression remained undetectable or minimal in quiescent cells and confluent cultures. Elevated claudin-1 protein was detected exclusively in prematurely senescent populations, including cells treated with one hundred and fifty micromolar hydrogen peroxide to induce oxidative stress, demonstrating that claudin-1 accumulation occurs independently of cell confluence and reversible growth arrest.

In parallel with claudin-1, the investigators characterized anillin (ANLN) as an informative negative biomarker of cell proliferation across both senescence pathways. Anillin is a conserved scaffolding protein critical for cytokinesis, interacting with cytoskeletal actin to facilitate the assembly and contraction of the cleavage furrow during cell division. Microarray profiling and quantitative PCR demonstrated that ANLN messenger RNA was profoundly downregulated across all senescent conditions, including doxorubicin, curcumin, and replicative senescence. Western blot analyses corroborated this transcriptional decline, showing that anillin protein was expressed in young dividing control cells and disappeared in non-proliferating cells across premature senescence, replicative senescence, and serum-starved quiescence. While immunofluorescence staining revealed nuclear anillin localization in approximately fifty percent of proliferating control cells, senescent cells displayed significantly diminished nuclear fluorescence. Notably, high-resolution confocal microscopy revealed discrete anillin foci within the cell nuclei of a specific subset of prematurely senescent cells. These nuclear anillin foci were absent in young controls and replicatively senescent cells, providing an additional structural feature accompanying stress-induced vascular aging.

To address whether claudin-1 elevation constitutes an immediate acute stress response or a sustained adaptation characteristic of established senescence, the researchers performed temporal kinetic analyses following senescence induction. Vascular smooth muscle cells were evaluated at twenty-four and forty-eight hours post-treatment with curcumin or doxorubicin, using heme oxygenase-1 (HO-1) as a reference marker of immediate acute cellular stress. In curcumin-treated cells, HO-1 protein expression increased rapidly within twenty-four hours before undergoing marked downregulation as permanent senescence became established. In contrast, claudin-1 protein was undetectable at twenty-four hours and began accumulating only at forty-eight hours post-treatment. This delayed kinetics confirmed that claudin-1 does not function as a first-response stress defense, but rather accumulates as cells progress into stable senescence. The authors also investigated upstream signaling pathways, evaluating sirtuin 1, a negative regulator of claudin-1 that universally declines during both replicative and premature senescence. Because sirtuin 1 decreases in all senescence models, and neither tumor necrosis factor-alpha nor interleukin-1-beta exhibited transcriptional or protein upregulation, distinct regulatory mechanisms govern selective claudin-1 protein accumulation.

Extending their in vitro findings to clinical vascular pathology, the investigators analyzed primary vascular smooth muscle cells isolated ex vivo from carotid atherosclerotic plaques harvested from six patients undergoing endarterectomy surgery. Explanted plaque-derived cells were cultured under uniform conditions and analyzed by Western blotting alongside in vitro control, prematurely senescent, and replicatively senescent cells. Immunoblots revealed that cells from all six patient donors lacked anillin protein, confirming that plaque smooth muscle cells were non-proliferating. However, claudin-1 expression exhibited striking donor-dependent variability across the patient samples. Explanted cells from two donors demonstrated robust claudin-1 protein bands alongside absent anillin, indicating a predominant population of prematurely senescent cells. In two other donors, weak claudin-1 bands suggested a heterogeneous cellular mixture containing both prematurely and replicatively senescent populations. In the remaining two donors, claudin-1 protein was completely undetectable, matching the profile of replicatively senescent cells. These ex vivo findings illustrate the biological heterogeneity of human atherosclerotic plaques, suggesting that individual lesions experience different burdens of environmental stress and replicative exhaustion.

While presenting claudin-1 as a promising candidate marker for identifying stress-induced premature senescence in vascular smooth muscle cells, the authors acknowledged notable methodological limitations and highlighted avenues for future inquiry. A primary experimental constraint was the scarce number of primary cells obtainable from human carotid plaque explants, which precluded simultaneous testing of a comprehensive senescence marker panel across all individual donor samples. Consequently, the authors emphasized that their ex vivo conclusions reflect group-level observations rather than an assumption that claudin-1 is uniformly elevated across all senescent plaque cells. Additionally, the authors noted that while doxorubicin, curcumin, and hydrogen peroxide all effectively triggered claudin-1 elevation, external transcriptomic datasets indicate that ionizing radiation does not upregulate claudin-1 in vascular smooth muscle cells, indicating that premature senescence markers may depend on specific inducing stimuli. Furthermore, parallel analyses conducted on human fibroblasts yielded inconclusive replicative senescence profiles, underscoring that claudin-1 induction is cell-type specific. Future investigations will seek to clarify the direct molecular mechanisms driving claudin-1 upregulation and evaluate whether claudin-1 modulation influences vascular cell migration, atherosclerotic plaque stability, or vascular disease progression.

Subject of Research: Identification of claudin-1 as a differential marker of stress-induced premature senescence in vascular smooth muscle cells

Article Title: Claudin-1 as a potential marker of stress-induced premature senescence in vascular smooth muscle cells

Article References: Gadecka, A., Janiszewska, D., Dudkowska, M., Koblowska, M., Buko, T., Iwanicka-Nowicka, R., Kossowska, H., Bojakowski, K., Mosieniak, G., & Bielak-Zmijewska, A. (2026). Claudin-1 as a potential marker of stress-induced premature senescence in vascular smooth muscle cells. GeroScience. https://doi.org/10.1007/s11357-026-02634-9

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02634-9

Keywords: vascular smooth muscle cells, cellular senescence, stress-induced premature senescence, replicative senescence, claudin-1, atherosclerosis, anillin, epigenetics, potential, marker, stress-induced, premature