The programmed death-1 (PD-1) and the PD ligand 1 (PD-L1) interaction represents a key immune checkpoint within the tumor microenvironment (TME), and PD-1 blockade has led to exciting therapeutic advances in clinical oncology. Although IFN-γ–dependent PD-L1 induction on tumor cells was initially thought to mediate the suppression on effector cells, recent studies have shown that PD-L1 is also expressed at high level on tumor-associated macrophages (TAMs) in certain types of tumors. However, the precise role of PD-L1 expression on TAMs in suppressing antitumor immunity within the TME remains to be defined. Using a myeloid-specific Pdl1-knockout mouse model, here we showed definitive evidence that PD-L1 expression on TAMs is critical for suppression of intratumor CD8+ T cell function. We further demonstrated that tumor-derived Sonic hedgehog (Shh) drives PD-L1 expression in TAMs to suppress tumor-infiltrating CD8+ T cell function, leading to tumor progression. Mechanistically, Shh-dependent upregulation of PD-L1 in TAMs is mediated by signal transducer and activator of transcription 3, a cascade that has not been previously reported to our knowledge. Last, single-cell RNA sequencing analysis of human hepatocellular carcinoma revealed that PD-L1 is mainly expressed on M2 TAMs, supporting the clinical relevance of our findings. Collectively, our data revealed an essential role for Shh-dependent PD-L1 upregulation in TAMs in suppressing antitumor immunity within the TME, which could lead to the development of new immunotherapeutic strategies for treating cancer.
Amy J. Petty, Rui Dai, Rosa Lapalombella, Robert A. Baiocchi, Don M. Benson, Zihai Li, Xiaopei Huang, Yiping Yang
Usage data is cumulative from August 2025 through August 2026.
| Usage | JCI | PMC |
|---|---|---|
| Text version | 2,578 | 247 |
| 285 | 69 | |
| Figure | 1,070 | 20 |
| Citation downloads | 318 | 0 |
| Totals | 4,251 | 336 |
| Total Views | 4,587 | |
Usage information is collected from two different sources: this site (JCI) and Pubmed Central (PMC). JCI information (compiled daily) shows human readership based on methods we employ to screen out robotic usage. PMC information (aggregated monthly) is also similarly screened of robotic usage.
Various methods are used to distinguish robotic usage. For example, Google automatically scans articles to add to its search index and identifies itself as robotic; other services might not clearly identify themselves as robotic, or they are new or unknown as robotic. Because this activity can be misinterpreted as human readership, data may be re-processed periodically to reflect an improved understanding of robotic activity. Because of these factors, readers should consider usage information illustrative but subject to change.