Verteporfin (VP) is an FDA-approved benzoporphyrin used in photodynamic therapy and is widely applied in preclinical studies as a purported light-independent inhibitor of the YAP-TEAD transcriptional interaction. VP is a potent photosensitizer that generates reactive oxygen species when illuminated. Observations of VP-associated protein aggregation, apparent YAP depletion, and in vitro cytotoxicity have raised the possibility that some reported effects reflect photochemical artifacts rather than true light-independent pharmacology. The present study tested whether ambient-light exposure during routine sample processing is a primary determinant of VP-associated high-molecular-weight complexes (HMWCs) and related immunoblot phenomena.
The authors treated MEL270, HEK293, and MCF-7 cells with vehicle, low-dose (LD) or high-dose (HD) VP under four systematically varied light/dark conditions that separated illumination during treatment from illumination during lysis and processing. Western blot analysis of p62, DIAP1, and ROCK1 showed that HMWC formation was determined principally by light exposure during sample processing. Samples that were treated in darkness but subsequently lysed and processed under ambient light displayed prominent HMWC bands for all three proteins, whereas matched samples processed entirely in darkness retained only the expected monomeric bands. Vehicle-treated samples processed under the same ambient-light workflow did not form HMWCs, indicating VP is required for the light-dependent cross-linking.
To determine whether intact cellular machinery is required, lysates from untreated cells were spiked with HD VP and incubated on ice. Lysates exposed to ambient light developed prominent HMWC bands for p62, DIAP1, and ROCK1 after 6 hours, whereas matched lysates incubated in near-darkness did not. In a light-exposure time course, cross-linked species were absent immediately after VP addition, first detectable at 1 hour, and increased through 6 hours. Vehicle-treated lysates carried through identical light workflows but without VP showed no HMWC formation. These results demonstrate that VP-induced cross-linking is a photochemical reaction that can occur post-lysis in the cell-free protein milieu and accumulates over hours of ambient-light exposure.
To probe reactive intermediates, cell homogenates were pre-incubated with the radical scavenger N-acetylcysteine (NAC) or the singlet oxygen quencher L-histidine prior to VP addition and ambient-light exposure. NAC markedly reduced p62 HMWC formation across all three cell lines, preserving monomeric p62 with only faint residual high-molecular-weight signal. L-histidine also reduced HMWC formation, though attenuation was less complete than with NAC. In both cases the monomeric p62 signal was maintained, indicating the scavengers limited cross-linking rather than promoting protein loss. These data are consistent with contributions from both Type I (radical-mediated) and Type II (singlet oxygen–mediated) photochemical pathways in VP-induced cross-linking.
Immunoblotting with an anti-methionine sulfoxide antibody revealed light-dependent methionine oxidation in all three cell lines. Under ambient light, immunoreactive bands in the 120–140-kDa range increased in intensity with VP dose and treatment duration; samples processed in near-darkness showed minimal methionine sulfoxide signal. The concordance between methionine oxidation and HMWC formation indicates that singlet oxygen generation occurred under these conditions. The authors note that methionine is one of several residues susceptible to singlet-oxygen attack (others include tryptophan, tyrosine, histidine, and cysteine) and that the specific residues forming the cross-links were not determined in this study.
Given VP's use as a YAP-TEAD inhibitor, the study examined YAP, phospho-YAP (Ser127), and TEF1/TEAD1. Under conditions where lysis and processing occurred in ambient light, both YAP and phospho-YAP displayed HMWC bands with reduced monomeric band intensity in a dose- and time-dependent manner across the three cell lines. When processing was performed in near-darkness, YAP and phospho-YAP monomeric bands were preserved and HMWCs were not detectable. By contrast, TEF1/TEAD1 showed substantially less HMWC formation; the monomeric TEF1 band remained largely stable across doses and time points, with only a discrete higher-molecular-weight species appearing at the highest dose and longest time under light. This differential susceptibility indicates protein-specific structural or chemical determinants of vulnerability to VP-mediated photo-cross-linking and explains how standard immunoblots can mimic selective disruption of the YAP-TEAD interaction.
Experiments that separated ambient illumination during the treatment phase from illumination during lysis and processing showed that HMWC formation was attenuated when lysis and processing occurred in near-darkness even if treatment had been under ambient light. This indicates that light exposure during the post-lysis sample-processing steps is the dominant determinant of VP-induced cross-linking, although treatment-phase illumination can contribute to some degree.
The source reports a single-experiment CCK-8 viability assay in which viability was lower under ambient light than in darkness at 24 hours for VP-treated cells, an observation consistent with the immunoblot findings. The authors explicitly state this viability observation requires independent replication. Details of the full viability dataset and statistical analysis are provided in the source repository, but the provided article text is truncated before complete viability results are presented in this summary.
These results support a light-artifact hypothesis: many VP-associated effects observed in cell-based assays—HMWC formation, apparent depletion of YAP on immunoblots, and cellular toxicity—can be driven substantially by ambient-light exposure during routine post-lysis handling rather than by an intrinsic light-independent pharmacological action of VP. The findings are especially relevant given ongoing clinical interest in VP as an anticancer agent and the parallel development of TEAD-targeted inhibitors. The data imply that preclinical studies using VP should rigorously control light exposure during lysis and processing and consider photochemical controls to distinguish true pharmacology from artifact.
All underlying data files, uncropped western blots, CCK-8 plate reader data, compiled datasets, and analysis code are available in the cited Zenodo repository. The article notes that some previously published experimental support for the light-artifact hypothesis was retracted and that the current study provides independent evidence. The provided source text is truncated before full presentation of the viability results; specific numerical viability outcomes, replication details, and some experimental parameters for the CCK-8 assay were not reported in the excerpted text. The authors also state that the single-experiment viability observation requires independent replication.