Related Experiment Video
Updated: Jan 17, 2026

Modeling Breast Cancer in Human Breast Tissue using a Microphysiological System
Published on: April 23, 2021
Celastrol liposomes enhance anti-PD-L1 in triple-negative breast cancer by reprogramming immunosuppressive
Yujie Li1, Hongyan Zhang2, Jiale Wei1
1School of Pharmaceutical Sciences, Zhejiang Chinese Medical University, Hangzhou, Zhejiang, 310053, China.
Background:
Triple-negative breast cancer (TNBC) exhibits a low response rate to immune checkpoint inhibitors such as anti-PD-L1 antibodies, partly due to immune evasion mediated by membrane-bound PD-L1 (mPD-L1) on tumor cells and PD-L1⁺ derived extracellular vesicles (EVs). Cancer-associated adipocytes (CAAs) further exacerbate this immunosuppressive microenvironment by secreting free fatty acids and adipokines that promote mPD-L1 expression and PD-L1⁺ EVs release. Current strategies, such as high-dose antibodies or EVs secretion inhibitors, are limited by high toxicity, non-specificity, and neglect of CAAs' role, highlighting the need for safer and more effective combination therapies.
Purpose:
This study aims to develop a celastrol-loaded liposomal system (Cel/Lip) to improve the bioavailability and reduce the toxicity of celastrol (Cel), while evaluating its role in modulating PD-L1 expression and the tumor microenvironment.
Methods:
Cel/Lip was prepared using the ethanol injection method. A comprehensive evaluation of the combined therapeutic effects and mechanisms of Cel/Lip with anti-PD-L1 therapy was conducted through multiparametric assessments including flow cytometry, Western blot, and immunofluorescence microscopy. In vitro experiments analyzed the effects of Cel-Lip on mPD-L1 expression and PD-L1⁺ EVs secretion in 4T1 cells. In vivo studies using an orthotopic 4T1 tumor model assessed the impact of Cel-Lip on PD-L1 levels in tumor tissues and circulating EVs, CAAs infiltration, immunosuppressive cell populations, and response to anti-PD-L1 treatment.
Results:
We revealed that CAAs significantly elevated mPD-L1 expression in tumor cells and promoted PD-L1⁺ EVs production, consequently impairing CD8+T cell function and creating a major obstacle to immunotherapy efficacy. The prepared Cel/Lip exhibited a uniform particle size distribution (106.93 ± 0.33 nm). In vitro, this nanosystem simultaneously downregulated mPD-L1 expression and reduced PD-L1⁺ EVs release. In vivo, it significantly suppressed PD-L1 expression, inhibited CAAs-mediated lipid infiltration, and decreased immunosuppressive cell populations. Importantly, Cel/Lip demonstrated synergistic antitumor effects when combined with aPD-L1 therapy, offering a safe and highly effective treatment strategy with significant clinical potential for TNBC.
Conclusion:
Cel/Lip provides a safe and highly effective strategy for the treatment of TNBC. Cel/Lip reduces CAA-induced mPD-L1 upregulation on tumor cells, thereby remodeling the immunosuppressive tumor microenvironment (TME) by promoting cytotoxic T cell infiltration and activation. Furthermore, it suppresses the secretion of PD-L1⁺ EVs, thereby preventing systemic T cell exhaustion in circulation and enhancing the availability of aPD-L1 at target tumor sites. Together, these mechanisms overcome key limitations of aPD-L1 monotherapy and significantly amplify antitumor efficacy.

