Facts

  • The expression of circRNAs is frequently dysregulated in human cancers.

  • CircRNAs play different roles during tumorigenesis and cancer progression.

  • CircRNAs regulate T cells, NK cells, and macrophages to reshape the tumor microenvironment.

  • CircRNA regulation of the tumor microenvironment provides potential therapeutic opportunities for cancer treatment.

Questions

  • Why circRNAs have multiple functions in the same or different human cancers. What are the underlining molecular determinants of this specificity?

  • Does dose-dependent targeting of circRNAs work in mouse models, at least in three-dimensional tumor organoid models?

  • Is circRNA targeting applicable in clinical trials?

  • Can we design prophylactic or therapeutic anti-cancer approaches based on genetic of polymorphisms of circRNAs?

Introduction

Circular RNAs (circRNA) is a closed circular molecule that is resistant to exonucleases, and is thus stable and widespread in animals and plants. CircRNA was discovered in 1976 when the Sanger team studied virus-like RNAs [1]. In 1991, Nigro et al. [2] accidentally discovered a normal novel RNA product. Due to its low expression and the limitations of detection technology, circRNA was originally considered to be an aberrant product of RNA splicing. Recently, with advances in high-throughput sequencing technology, increasing numbers of circRNAs have been characterized and their roles and mechanisms have become active areas of investigation [3,4,5].

The immune system maintains homeostasis through immunomodulation, surveillance, and the prevention of pathogen invasion. The immune response coordinates a variety of immune cells and has antiviral, antibacterial, and antitumor functions. With rapid developments in oncology, immunology, molecular biology, and related disciplines, immunotherapies such as immune checkpoint inhibitors, tumor vaccines, and adoptive cell therapy have revolutionized cancer treatment. However, therapeutic responses, especially those of solid tumors, have been unsatisfactory in clinical trials and clinical applications. Recent studies have demonstrated that circRNAs are involved in cancer development [6,7,8] and immune responses [9,10,11,12]. In this review, we discuss the roles of circRNAs in the regulation of immune cells, immune-related molecules, and tumor immunity. We anticipate that this summary of current knowledge will facilitate the development of strategies to target circRNAs in the immune microenvironments of human cancers.

Biogenesis and function of circRNAs

CircRNA is a class of non-coding RNA generated from precursor messenger RNA (mRNA). Most circRNAs originate from exons in gene coding regions; others originate from 3′–untranslated regions (UTRs), 5'-UTRs, introns, intergenic regions, and antisense RNA [13, 14]. CircRNAs can be divided into four categories based on their sequence origin: (1) exonic circular RNAs (EciRNAs) derived from exons of the parent gene; (2) lasso-type or circular intronic RNAs (ciRNAs) derived from introns; (3) exonic–intronic circular RNAs (EIciRNAs) derived from both exons and introns; and (4) other circRNAs, including those derived from antisense strand transcripts (antisense circRNAs) and those derived from intergenic sequences or other unannotated genomic sequences (intergenic circRNAs) [15]. About 80% of circRNAs are EciRNAs localized mainly to the cytoplasm, whereas ciRNAs and EIciRNAs are often localized to the nucleus. CircRNAs are relatively evolutionarily conserved in different species. Jeck et al. [16] used the genome-wide RNase R enrichment method to detect >25,000 circRNAs in fibroblasts. Wang et al. [17] observed circRNA expression in fungi, plants, and prokaryotes, reflecting a high degree of conservation and widespread distribution among species. The expression of the same circRNA varies greatly under diseased and non-diseased conditions, among tissues, and during different time periods. The half-life of circRNAs exceeds that of the associated linear mRNA, as the covalent closed-loop structure lacks 5′ and 3′ends, which makes circRNAs more resistant to the exonuclease RNase R [18].

CircRNAs have four main biological functions (Fig. 1). (1) As they contain a large number of micro-RNA (miRNA) binding sites, they serve as molecular sponges and compete for miRNA binding to target mRNAs, thereby upregulating the expression of target genes [19,38,39,40]. Tumor cell-derived circRNAs have recently been reported to play a vital and direct role in tumor immune escape (Table 1). Mechanically, circRNAs enhance the interaction between the immunosuppressive molecule programmed death receptor 1 (PD-1) and its ligand (PD-L1) by upregulating PD-1 expression in T cells, suppressing T-cell activation and cytokine secretion. Exosomes derived from different tumor cells deliver various circRNAs to T cells to inhibit their killing ability via PD-1 upregulation. Those derived from ovarian cancer cells were found to deliver circ-0001068 into T cells, increasing PD-1 expression via miR-28-5p sponging and thereby causing T-cell exhaustion [41]. In lung adenocarcinoma, circRNA-002178 was found to enter CD8+ T cells via exosomes and upregulate PD-1 expression by absorbing miR-34a [42]. circRNA can also upregulate the expression of the immune checkpoint molecules PD-L1 and CD73 on tumor cell surfaces via miRNA sponging, which helps tumor cells to escape recognition and death by T cells. Multiple studies have shown that circRNAs regulate PD-L1 expression via the circRNA–miRNA–mRNA axis, for instance, the circRNA of vimentin, CDR1-AS, hsa_circ_0003288, hsa_circ_0000190, hsa_circ_0046523, circ-CPA4, hsa-circRNA-002178, circ_0000284, circ_001678, circ-HSP90A, and circIGF2BP3 (Table 1) [43,44,114,115].

TAMs form the most abundant immune cell population in the TME. CircRNAs in tumor cells can regulate macrophage polarization through multiple pathways in the TME (Table 3, Fig. 4): (1) they induce crosstalk between tumor cells and macrophages (Fig. 4A), (2) they promote chemokine secretion from tumor cells (Fig. 4B), (3) tumor cell-derived circRNAs in exosomes enter macrophages to play a regulatory role (Fig. 4C), and (4) they promote tumor-cell expression of cytokines such as IL-4 and PD-L1 (Fig. 4D). In-depth investigation of the mechanisms underlying these roles and preclinical studies are urgently needed.

Table 3 circRNAs regulate .macrophages.
Full size table
Fig. 4: Circular RNAs (circRNAs) regulate macrophage polarization to promote tumor progression.
figure 4

A CircRNAs mediate crosstalk between tumor cells and macrophages, induce M2 macrophage polarization, and impair T cell function, resulting in the formation of an immunosuppressive tumor microenvironment (TME). B CircRNAs upregulate chemokine expression in tumor cells and induce M2 macrophage polarization. C Tumor cells release exosomes containing circRNAs into macrophages to enhance M2 macrophage polarization. D CircRNAs promote the secretion of inflammatory factors and immunosuppressive molecules in tumor cells to recruit and induce M2 macrophage polarization and disable T cells.

Full size image

CircRNAs regulate neutrophils, myeloid-derived suppressor cells, and cancer-associated fibroblast

The neutrophils are also an important component in the TME, participating in different stages of tumor development and progression such as tumorigenesis, proliferation and metastasis. Neutrophils could play dual roles as a pro-tumor(N2) or tumor suppressor (N1) in the tumor microenvironment due to heterogeneous phenotypes and functional diversity. Recently, mounting evidence show that circular RNA affects tumor development by regulating the function of neutrophils. In bladder cancer, circDHTKD1 recruited and activated neutrophils by inducing CXCL5 expression, and then neutrophils participated in lymphangiogenesis by secreting VEGF-C, facilitating lymphatic metastasis of bladder cancer cells [116]. But in CRC, circPACRGL mainly promoted differentiation of N1 to N2 neutrophils by sponging miR-142-3p/miR-506-3p, N2 neutrophils increased the expression of transforming growth factor-β1 (TGF-β1), which promoted CRC cell proliferation, migration and invasion [117]. Although the underlying mechanism is not very clear, but the diversity and plasticity of neutrophils maybe act as a potential and promising immunotherapy target in clinical treatment.

Myeloid-derived suppressor cells (MDSC) are also another key player in TME. In addition to the immunosuppressive effect, MDSC can also exert tumor-promoting effects by promoting angiogenesis, invasion and metastasis. More details about non-coding RNAs including circRNAs modulate MDSCs in TME have been summarized elsewhere [118].

Cancer-associated fibroblasts (CAFs), also named as tumor-associated fibroblast, are a key factor in tumor microenvironment. It plays important role in tumor growth and metastasis due to diverse functions, such as interactions with cancer cells and crosstalk with infiltrating leukocytes and so on. In pancreatic cancer, Hu et al found that circFARP1 upregulated the expression and secretion of LIF via CAV1 in CAFs to induce chemoresistance [51,52,53, 57]. Stromal cells, such as cancer-associated fibroblasts, endothelial cells, and pericytes, are important TME components, and much more research is warranted to explore their potential regulation. Although TME reprogramming is considered to be a potentially effective strategy for tumor eradication and the improvement of tumor immunotherapy efficacy, there is still a long way to go before we can conquer cancer. For example, does dose targeting of circRNAs work in mouse models, at least in three-dimensional tumor organoid models? Is circRNA targeting applicable in clinical trials? Thorough investigations of circRNAs using animal models would help to accelerate the translation of basic research into clinical practice. We believe that an improved understanding of circRNA functions and mechanisms related to tumorigenesis and immunotherapy would certainly contribute to the development of new therapeutic strategies for cancer.