Immune checkpoint blockade has reshaped modern oncology, but its success remains uneven. Drugs that disrupt the PD-1/PD-L1 pathway can produce long-lasting remissions in several cancers, yet many patients with solid tumors—including breast cancer—derive little or no benefit. The problem is especially perplexing when tumors carry abundant PD-L1, the molecule targeted by therapy. A study published in Science Bulletin now offers a possible explanation: PD-L1 may be doing far more than suppressing immune attack at the cancer-cell surface. In some tumors, it appears to enter the nucleus and reprogram gene activity from within.
The researchers found that nuclear PD-L1 was substantially more abundant in tumors from patients who failed to respond to immune checkpoint blockade than in tumors from patients who responded. Once inside the nucleus, PD-L1 acted as a transcriptional regulator, helping activate genes associated with immune suppression and resistance to anti-PD-1 treatment. This finding suggests that measuring PD-L1 only at the cell membrane may provide an incomplete picture of a tumor’s biology. A cancer cell could display high levels of the familiar checkpoint protein while simultaneously using a second, hidden form of PD-L1 to build a more hostile environment for immune cells.
The molecular trigger for this nuclear transformation was identified as DDB1, an adaptor protein within the CUL4A E3 ubiquitin ligase complex. The team reported that DDB1 promotes the attachment of K63-linked ubiquitin chains to PD-L1 at lysine 185, or K185. Ubiquitination is often described as a cellular tagging system, but ubiquitin chains can also change a protein’s location, interactions and activity. In this case, the modification appears to function as a dual-purpose switch: it helps PD-L1 reach the nucleus and enables the protein to operate once it arrives.
The first part of the mechanism involves another chemical modification at lysine 263. Acetylation at K263 normally restricts PD-L1 from entering the nucleus. DDB1-mediated ubiquitination at K185 counteracts this barrier, allowing PD-L1 to interact with the structural protein vimentin. The researchers propose that vimentin acts as a transport partner, helping shuttle the modified checkpoint protein from the cytoplasm into the nuclear compartment. This sequence of events gives PD-L1 access to DNA and separates its nuclear function from its conventional role at the plasma membrane.
The second part of the mechanism is even more unusual. The K63-linked ubiquitin chain was not merely a delivery signal; it was required for nuclear PD-L1 to recognize and bind particular genomic regions. The protein was found at promoters controlling immune checkpoint genes including CD276 and CD273, as well as genes involved in NF-κB signaling, such as TRAF1, BIRC3 and RELB. By influencing these transcriptional programs, nuclear PD-L1 could reinforce immune suppression and help cancer cells withstand attack after PD-1 inhibition.
Evidence for the importance of K185 came from a targeted mutation. When the researchers replaced lysine 185 with arginine, creating the K185R variant, PD-L1 could no longer receive the relevant ubiquitin modification. The mutant protein showed impaired nuclear accumulation and lost its ability to activate the identified gene network. Cut&Tag profiling and chromatin immunoprecipitation followed by quantitative PCR further showed that K185R PD-L1 no longer occupied the key promoter regions. These results link a single amino-acid site to both the localization and gene-regulating functions of the protein.
The discovery also points toward a potential way to disable this resistance pathway. The researchers tested thalidomide, an immunomodulatory drug approved for conditions including multiple myeloma and known to affect CRL4-related ubiquitin-ligase activity. In their experiments, thalidomide disrupted the interaction between DDB1 and PD-L1, reduced K63-linked ubiquitination and limited the protein’s movement into the nucleus. Rather than removing PD-L1 from the tumor altogether, the drug appeared to close the molecular route that allows the checkpoint protein to acquire its transcriptional role.
The most dramatic results came from experiments in mice bearing 4T1 breast tumors, a model known for aggressive growth and limited sensitivity to immunotherapy. Anti-PD-1 treatment alone or thalidomide alone produced comparatively modest effects, whereas the combination caused pronounced tumor shrinkage and complete regression in some animals. The treatment also altered the immune landscape inside the tumors. Flow-cytometry analyses showed increased infiltration by Granzyme B-positive cytotoxic T cells, which are capable of killing cancer cells, alongside a reduction in exhausted TIM3-positive, PD-1-positive T cells. Animals receiving the combination survived significantly longer than those treated with either agent alone.
The findings present PD-L1 as a protein with two distinct but connected identities: a membrane checkpoint that restrains T cells and a nuclear regulator that can promote an immunosuppressive transcriptional state. Blocking the first function with anti-PD-1 therapy may not be enough when the second remains active. Repurposing thalidomide to interfere with DDB1-dependent ubiquitination could therefore offer a strategy for attacking both layers of PD-L1 biology. The results remain preclinical, and the safety, dosing and effectiveness of this combination in people will require careful clinical testing. Even so, the study provides a compelling explanation for why PD-L1 abundance alone can fail as a predictor of immunotherapy response—and identifies a molecular vulnerability that may turn resistant tumors back into targets for immune attack.
Subject of Research: Nuclear PD-L1, DDB1-mediated K63-linked ubiquitination and resistance to anti-PD-1 immunotherapy in breast cancer.
Web References: https://doi.org/10.1016/j.scib.2026.07.077
References: Science Bulletin, DOI: 10.1016/j.scib.2026.07.077
Keywords: PD-L1, PD-1, immune checkpoint blockade, cancer immunotherapy, nuclear PD-L1, DDB1, ubiquitination, thalidomide, breast cancer, tumor microenvironment, NF-κB, T cells
