Lung cancer is projected to be the leading cause of cancer death in the United States; it has been estimated that lung cancer will be responsible for 350 deaths each day in 2022. Also, lung cancer will be the second most commonly affected cancer in both males and females in 2022. Besides, the 5-year relative survival of lung cancer patients is 22% in all races and stages (1). It has been reported that approximately 85% of lung cancer cases are grouped into NSCLCs; lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) are the most common subtypes of NSCLC (2). Considering the prevalence, associated burden, and poor survival of affected patients, there is a pressing need to identify novel therapeutic approaches to improve the prognosis of the patients.

Cancer immunotherapy is a novel therapeutic approach that leverages the host immune system to reject tumors (3). The immunosuppressive tumor microenvironment shields malignant cells from anti-tumoral immune responses and reprograms the residing immune cells in a way that they contribute to tumor growth and migration (4). Growing evidence has highlighted the significance of pathological expression of inhibitory immune checkpoint molecules in suppressing anti-tumoral immune responses (5). The PD-L1/PD-1 is one of the well-studied inhibitory immune checkpoint axes that can regulate immune responses. It has been reported that the increased expression of PD-L1 is associated with decreased overall survival and recurrence-free survival of NSCLC patients (6). Following the significant role of the PD-L1/PD-1 axis in facilitating tumor growth, various monoclonal antibodies targeting this axis have been developed, investigated, and approved for NSCLC patients (7). Besides the immunosuppressive role of tumor-intrinsic PD-L1, it is implicated in stimulating oncogenic signaling pathways. For instance, increased expression of PD-L1 in NSCLC cells increases tumor proliferation and colony numbers; however, PD-L1 depletion reserves these pro-tumoral effects in NSCLC cells (8). In cisplatin-treated NSCLC cells, PD-L1 knockdown decreases cell proliferation, migration, and invasion of NSCLC cells (6). It has been shown that the mitogen-activated protein kinase (MAPK) pathway stimulates PD-L1 expression in NSCLC cells, and inhibiting the MAPK pathway decreases PD-L1 protein and mRNA expression (9). Besides, the epidermal growth factor receptor (EGFR)-mediated pathway, as an oncogenic signaling pathway, upregulates PD-L1 expression in NSCLC cells (10). In colorectal cancer cells, PD-L1 knockdown has decreased the protein and mRNA expression of phosphatidylinositol-3-kinase and protein kinase B (PI3K) and protein kinase B (Akt), which are the essential factors of the PI3K/Akt oncogenic signaling pathway (11). Therefore, PD-L1 is considerably involved in oncogenesis and tumor development. Despite the promising results of administrating monoclonal antibodies against the PD-L1/PD-1 axis for NSCLC patients, these antibodies primarily target the PD-L1/PD-1 expression on various cells. Indeed, they are not designed to inhibit the mRNA or protein expression of oncogenic signaling pathway elements in malignant cells (12).

miRs are small non-coding RNAs that have considerable roles in the post-transcriptional regulation of gene expression (13). These non-coding RNAs substantially regulate various signaling pathways, and the downregulation of tumor-suppressive miRs and the upregulation of oncomiRs have been associated with tumor development (14). Recent studies have shown that ectopic expression of specific PD-L1-targeting miRs in cancerous cells, like colorectal cancer and triple-negative breast cancer cells, can suppress tumor-intrinsic PD-L1 expression, stimulate anti-tumoral immune responses, substantially inhibit oncogenic signaling pathways, and decrease tumor development (12, 15). Accumulating studies have identified PD-L1-inhibiting miRs that some of them can suppress NSCLC development (6, 16–18). Xie et al. have shown that the ectopic expression of miR-140, as a PD-L1-inhibiting miR, is associated with decreased expression of cyclin E and suppressed cell proliferation in NSCLC cells (19). Chen et al. have reported that miR-526b-3p increased expression downregulates the expression of c-Myc and MDR1 and enhances cisplatin-cytotoxicity in NSCLC cells (16). Song et al. have shown that increased expression of hsa-miR-138-5p decreases tumor proliferation and arrests the cell cycle of NSCLC cells via downregulating CCND3, Ki67, CDC20, and MCM2 (20). However, there is no study to comprehensively collate and investigate their role in NSCLC development.

In the current study, we aimed to thoroughly and systematically search the literature to identify the current knowledge of the significance of PD-L1-inhibiting miR ectopic expression in NSCLC development. We also leveraged the cancer genome atlas (TCGA) data to study their clinical significance in the affected patients. Besides searching the literature, we leveraged in silico data to predict the circular RNA (circRNA)/miR/PD-L1 axes in NSCLC, which might translate into introducing novel therapeutic approaches. Finally, we applied single-cell RNA sequencing to study the potential results of the ectopic expression of PD-L1-inhibiting miRs on the immune cells residing in the NSCLC microenvironment.

In 2022, lung cancer will be the second most commonly diagnosed and the most common cause of cancer-related death in men and women in the United States (1). NSCLC is responsible for the majority of lung cancer cases, and most patients are diagnosed when the disease is advanced (27). Although there have been notable advances in treating NSCLCs, disease progression and resistance development have been considerable burdens for affected patients (28). In this regard, a better understanding of NSCLC biology and its microenvironment can pave the way for developing novel treatments for affected patients.

As small non-coding RNAs, miRs have pivotal roles in the post-transcriptional regulation of genes. It is well-established that miRs can bind to the 3’ untranslated region (UTR) of target mRNAs and facilitate translation inhibition; therefore, they are involved in regulating various biological processes (29). Growing evidence has highlighted that specific miRs can pave the way for tumor development, and other miRs can inhibit tumor development and metastasis (12, 15). Tumor-suppressive miRs target the expression of oncogenes, while oncomiRs target the expression of tumor-suppressive genes. In this regard, the ectopic expression of tumor-suppressive miRs and suppressing the expression of oncomiRs can inhibit tumor development and migration (30). Since miRs can regulate a vast number of gene expressions, both directly and indirectly, they carry valuable therapeutic potential for cancer treatment (31).

Recently, circRNAs have emerged as a new class of non-coding of RNAs that considerably regulate gene expression (32). It has been reported that circRNAs are involved with miR sponges, protein/peptide translators, RNA-binding protein sponges, and regulating gene splicing and transcription (31). One of the most studied circRNA-mediated interactions is the cross-talk between circRNA and miRs; circRNAs have multiple sites for binding miRs that can “sponge” them and therefore regulate the expression of various genes (33). Thus, identifying the circRNA/miR/mRNA axes that have significant roles in cancer development/inhibition can be a milestone in cancer treatment.

Immunotherapy is one of the novel and promising treatments for a subset of cancer patients. Stimulating the host’s anti-tumoral immune responses is the cornerstone of immunotherapeutic approaches (3). Effective rejection of tumors can also eradicate cancerous cells and prevent tumor relapse; however, the immunosuppressive tumor microenvironment has been implicated in tumor development and shielding tumoral cells from anti-tumoral immune responses (34). Inhibitory immune checkpoint axes considerably attenuate anti-tumoral immune responses and contribute to tumor growth (3). The PD-1/PD-L1 axis is one of these inhibitory axes that can maintain this pathological immune tolerance against neoplastic cells (3). In this regard, multiple clinical studies have demonstrated the beneficial effect of administrating monoclonal antibodies that inhibit the PD-1/PD-L1 axes in NSCLC patients. Following the promising results, the FDA has approved pembrolizumab (anti-PD-1), nivolumab (anti-PD-1), atezolizumab (anti-PD-L1), and durvalumab (anti-PD-L1) for patients with NSCLC. Thus, targeting this inhibitory axis is pivotal in treating NSCLC (7).

Besides inhibiting anti-tumoral immune responses, recent findings have indicated that tumor-intrinsic inhibitory immune checkpoint molecules have oncogenic effects in various cancers, like triple-negative breast cancer, NSCLC, and melanoma (35–38). In TNBC cells, PD-L1 knockdown has substantially decreased cell viability, stemness, cell cycle progression, and migration. Also, tumor-intrinsic PD-L1 silencing has stimulated apoptosis of TNBC cells and decreased the CD25⁺ forkhead box P3 (Foxp3)⁺ Tregs (36). Qiu et al. have reported that PD-L1 overexpression has substantially increased cell proliferation and migration in glioblastoma cells via the mitogen-activated extracellular signal-regulated kinase (MEK)/extracellular signal-related kinase (Erk) pathway both in vitro and in vivo (39). Wei et al. have reported that increased expression of PD-L1 increases the self-renewal of colorectal cancer cells and PD-L1 knockdown decreases colorectal cancer stemness (40). Yu et al. have demonstrated that PD-L1 knockdown has substantially decreased cell viability and reduced tumor clonogenicity and migration in NSCLC cells (35). Hong et al. have shown that tumor-intrinsic PD-L1 silencing or PD-L1/PD-1 blockade can increase the proliferation and viability of CD8⁺ T-cells, upregulate the expression of IFN-γ and IL-4, and downregulate IL-10 expression in a co-culture system with NSCLC cells (41). Therefore, inhibiting tumor-intrinsic PD-L1 can stimulate anti-tumoral immune responses and suppress oncogenic signaling pathways, leading to suppressed tumor growth.

Considering the inhibitory effect of tumor-intrinsic PD-L1 on anti-tumoral immune responses, the established oncogenic effect of tumor-intrinsic PD-L1 on tumor development, and the multi-target and considerable effect of miRs on mRNAs expression, identifying PD-L1-inhibiting miRs and studying the overall effect of their ectopic expression on tumoral cells can pave the way for the development of novel therapeutic approaches. It has been reported that the ectopic expression of hsa-miR-497-5p, hsa-miR-424-5p, hsa-miR-34a-5p, hsa-miR-200c-3p, hsa-miR-138-5p, hsa-miR-383-5p, hsa-miR-195-5p, hsa-miR-570-3p, and hsa-miR-3609 can considerably inhibit tumor proliferation, migration, stemness, tumor-intrinsic PD-L1 expression in TNBC cells. Besides contributing to the transformation of the tumor microenvironment from immunosuppressive to pro-inflammatory, their ectopic expression has been associated with enhanced chemosensitivity in TNBC cells (15). Consistent with this, it has been shown that ectopic expression of hsa-miR-140-3p, hsa-miR-382-3p, hsa-miR-148a-3p, hsa-miR-93-5p, hsa-miR-200a-3p, hsa-miR-200c-3p, hsa-miR-138–5p, and hsa-miR-15b-5p can suppress PD-L1 expression and inhibit tumor proliferation and migration in colorectal cancer cells; these PD-L1-inhibiting miRs substantially transform the tumor microenvironment from immunosuppressive to pro-inflammatory and improve the chemosensitivity of colorectal cancer cells regardless of the microsatellite state (12). In the current study, we aimed to identify the effect of PD-L1-inhibiting miRs on NSCLC development and investigate the circRNA/miR/PD-L1 axes based on the literature and in silico data.

Zhu et al. have identified the FGD5-AS1/hsa-miR-142-5p/PD-L1 axis in NSCLC cells. They have reported that FGD5-AS1 inhibits hsa-miR-142-5p expression, and hsa-miR-142-5p ectopic expression downregulates PD-L1 expression and decreases the migration and invasion of cisplatin-resistant NSCLC cells (6). Yang et al. have identified the circCHST15/hsa-miR-194-5p/PD-L1 axis in NSCLC cells; they have shown that circCHST15 inhibits the expression of hsa-miR-194-5p and hsa-miR-194-5p ectopic expression decreases the proliferation and clonogenicity of NSCLC cells (18). Wang et al. have identified the circRNA-002178/hsa-miR-34a-5p/PD-L1 axis in NSCLC cells; circRNA-002178 inhibits the expression of hsa-miR-34a-5p, leading to the liberation of PD-L1 expression. Of interest, it has been reported that the exosomal plasmatic level of circRNA-002178 has a significant diagnostic value in LUAD patients (AUC = 0.9956, and P-value < 0.001) (42). Consistent with this, increased expression of hsa-miR-34a-5p mimic and hsa-miR-34a-5p inhibitor have decreased and increased PD-L1 expression in NSCLC cells (43, 44). It has been reported that hsa-miR-34a-5p overexpression inhibits the proliferation, migration, and invasion of LUAD cells. Also, hsa-miR-34a-5p overexpression stimulates apoptosis of LUAD cells (45). We have shown that hsa-miR-34a-5p is significantly enriched for regulating the PI3K/Akt pathway and cell cycle. Li et al. have highlighted the circRNA_0000284/hsa-miR-377-3p/PD-L1 axis in NSCLC cells, and the protumoral effect of circRNA_0000284 on cell proliferation, clonogenicity, migration, and invasion of NSCLC cells is mediated via the hsa-miR-377-3p/PD-L1 axis. They have shown that the expression level of circRNA_0000284 is substantially upregulated in NSCLC tissues compared to non-tumoral tissues, in higher stage NSCLC tissues compared to lower stage NSCLC tissues, in tissues with lymph node metastasis compared with the tissues without lymph node metastasis. Of interest, circRNA_0000284 expression level has been associated with the inferior overall survival of affected patients (46). Besides, it has been shown that hsa-miR-377-3p decreases the cell viability, migration, and invasion of NSCLC cells (47). We have shown that hsa-miR-377-3p is significantly enriched for regulating the cell cycle. Hong et al. have shed light on the circ-CPA4/hsa-let-7c-5p/PD-L1 axis in NSCLC cells; they have shown that circ-CPA4 targets hsa-let-7c-5p and liberate PD-L1 expression in NSCLC (41). Also, it has been reported that hsa-let-7c-5p suppresses the cell viability of NSCLC cells (48). Our results have indicated that increased expression of hsa-let-7c-5p is associated with improved disease-specific survival of LUSC patients. Also, we have demonstrated that hsa-let-7c-5p is significantly enriched for regulating the PI3K/Akt pathway. Wei et al. have highlighted the MALAT1/hsa-miR-200a-3p/PD-L1 axis in NSCLC cells; they have shown that MALAT1 sponges hsa-miR-200a-3p to liberate PD-L1 protein expression in NSCLC cells and MALAT1 overexpression increases clonogenicity, cell migration, proliferation, and invasion and decrease apoptosis in NSCLC cells. However, hsa-miR-200a-3p ectopic expression partially reverses the pro-tumoral effect of MALAT1 in NSCLC cells (49). Chen et al. have reported that hsa-miR-200a-3p suppresses tumor-intrinsic PD-L1 expression in NSCLC cells (50). Tan et al. have reported that ectopic expression of hsa-miR-200a-3p decreases the proliferation, invasion, and migration of NSCLC cells (51). Our results have shown that increased expression of hsa-miR-200a-3p is associated with enhanced disease-specific survival of LUAD patients. Besides, we have shown that hsa-miR-200a-3p is significantly enriched for regulating the cell cycle. Yang et al. have highlighted the CircCHST15/hsa-miR-155-5p/PD-L1 axis in NSCLC cells; circCHST15 suppresses hsa-miR-155-5p expression and liberates PD-L1 expression in NSCLC cells. Yang et al. have reported that hsa-miR-155-5p ectopic expression increases the proliferation and clonogenicity in H1395 and A549 cells (18). Consistent with this, Chen et al. have reported that hsa-miR-155-5p transfection decreases PD-L1 expression; however, Chen et al. have reported that hsa-miR-155-5p transfection suppresses the proliferation of A549 cells. They have shown that hsa-miR-155-5p increases the infiltration of IFN-γþ lymphocytes, CD4þ T lymphocytes, and CD8þ T lymphocytes, upregulates the expression of TNF-α and IFN-g, and downregulates IL-10 expression (52). In line with these, Huang et al. have demonstrated that hsa-miR-155-5p targets mRNA and protein PD-L1 expression in A549 and H1650 cells. Also, they have shown a negative correlation between hsa-miR-155-5p expression and PD-L1 expression in LUAD tissues (53).

It has been reported that hsa-miR-326 ectopic expression downregulates PD-L1 and B7-H3 expression in NSCLC cells. Besides, there have been upregulation in TNF-α, IFN-γ, and IL-2 expression and downregulation in IL-1β, IL-10, and TGF-β expression in the supernatant of hsa-miR-326 overexpressed cells. The overexpression of hsa-miR-326 has substantially decreased tumor growth, metastasis, and PD-L1 expression in animal models and increased the infiltration of CD8⁺ cells and TNF-α⁺ IFN-γ⁺ CD8⁺ T-cells population in tumor tissues in animal models of NSCLC (17). Chen et al. have demonstrated that miR-526b-3p can enhance cisplatin chemosensitivity, decrease the migration and proliferation of NSCLCs, and increase the population of CD8⁺ T-cells (16). Song et al. have reported that hsa-miR-138-5p inhibits NSCLC proliferation, decreases PD-L1 protein expression, and increases tumor-infiltrating mature DCs, CD4⁺ T cells, and CD8⁺ T-cells. Also, hsa-miR-138-5p downregulates PD-1 and PD-L1 expression in tumor-infiltrating DCs and increases the ability of DCs to induce cytotoxicity of CD8⁺ T-cells (20). It has been shown that hsa-miR-200c-3p downregulates PD-L1 mRNA and protein expression in NSCLC cells, and hsa-miR-200c-3p potentiates doxorubicin-induced cytotoxicity in NSCLC cells. Also, hsa-miR-200c-3p considerably enhances IFN-γ expression and the cytotoxic T-cell-mediated anti-tumoral effect against NSCLC cells (54). We have shown that hsa-miR-200c-3p is significantly enriched for regulating the PI3K/Akt pathway. Xie et al. have shown that miR-140 downregulates the protein and mRNA expression of PD-L1, arrests the cell cycle, and decreases the proliferation of NSCLC cells (19). Fujita et al. have demonstrated that the increased expression of hsa-miR-197-3p suppresses PD-L1 expression and enhances the chemosensitivity of NSCLC cells; hsa-miR-197-3p inhibition increases tumor growth and its migration and decreases the survival of affected animals (55). Cortez et al. have shown that hsa-miR-34c-5p and hsa-miR-34b suppress PD-L1 expression in NSCLC cells (43). It has been reported that hsa-miR-34c-5p increased expression decreases the migration and invasion of NSCLC cells (56). Our results have indicated that hsa-miR-34c-5p is significantly enriched for regulating the PI3K/Akt pathway and the cell cycle. It has been reported that hsa-miR-200b-3p and hsa-miR-200b-5p downregulate tumor-intrinsic PD-L1 expression in NSCLC cells (50, 57). The ectopic expression of hsa-miR-200b-3p has substantially decreased the migration and invasion of NSCLC cells (58). However, further studies are needed to clarify the overall effect of the hsa-miR-200b-5p ectopic expression on NSCLC cells. Despite the inhibitory role of hsa-miR-142-5p on tumor-intrinsic NSCLC cells, the overexpression of hsa-miR-142-5p increases tumor growth and decreases the apoptosis rate in NSCLC cells (59). In contrast, Jiang et al. have reported that hsa-miR-142-5p overexpression decreases cell proliferation, migration, and invasion and increases the apoptosis of NSCLC cells (60). Therefore, further studies are needed to evaluate the overall effect of the hsa-miR-142-5p ectopic expression on NSCLC development. Table 3 summarizes the tumor-suppressive PD-L1-inhibiting miRs in NSCLC.

System biology approaches have provided valuable insights into the expression profile of various human conditions (61, 62). Following systematically reviewing the literature, we aimed to leverage the data obtained from system biology to predict the potential circRNA/PD-L1-targeting miRs in NSCLC tissues. Our in silico results have highlighted the hsa_circ_0002476/hsa-miR-216a-5p/PD-L1 and hsa_circ_0063526/hsa-miR-1299/PD-L1 in NSCLC. Based on our in silico analyses, hsa-miR-216a-5p and hsa-miR-1299 can be significantly downregulated in NSCLCs and can suppress PD-L1 expression. Also, hsa_circ_0002476 and hsa_circ_0063526 can be significantly upregulated in NSCLC tissues, and hsa_circ_0002476 and hsa_circ_0063526 can inhibit hsa-miR-216a-5p and hsa-miR-1299 expression. Therefore, hsa_circ_0002476 and hsa_circ_0063526 might sponge hsa-miR-216a-5p and hsa-miR-1299, respectively, liberating PD-L1 expression in NSCLC. Consistent with our in silico investigation, Pang et al. have shown that hsa-miR-216a-5p is substantially decreased in NSCLC cells and tissues compared to normal cells and tissues, respectively. Also, the ectopic expression of hsa-miR-216a-5p has substantially decreased the proliferation, invasion, and migration and stimulated apoptosis in NSCLC cells (63). Recently, Chai et al. have demonstrated that miR-1299 expression has substantially decreased in NSCLC tissues compared to normal tissues (64). In line with these, Cao et al. have also reported that miR-1299 expression is substantially decreased in NSCLC cells and tissues compared to non-tumoral cells and tissues, respectively. Also, miR-1299 ectopic expression suppresses proliferation, migration, invasion, and epithelial-mesenchymal transition processes and stimulates apoptosis in NSCLC cells by regulating the PI3K/Akt pathway (65).

Single-cell RNA sequencing can be considered a milestone in understanding various cells residing in the tumor microenvironment and their expression profile (66). This novel approach enables us to dissect each cell of tumor tissue, thoroughly examine its expression profile, categorize them into groups, and investigate the expression of a specific gene in these cell groups (67). Following identifying the significance of ectopic expression of PD-L1-inhibiting miRs in NSCLC development and discussing the underlying mechanisms through systematic literature review and tumor bulk analysis, we aimed to answer the question of which cells residing in the tumor microenvironment of NSCLC might be liberated following ectopic expression of PD-L1-inhibiting miRs in NSCLC cells. For this aim, we analyzed 32,116 cells from 10 NSCLC tissues to identify which cells express PD-1 in the tumor microenvironment of those NSCLC tissues. Based on our analyses, tumor-infiltrated T-cells, CD8⁺ T-cells, CD4 Treg, DCs, and ILC2 might considerably express PD-1; therefore, ectopic expression of hsa-miR-194-5p, hsa-miR-326, hsa-miR-526b-3p, hsa-miR-34a-5p, hsa-miR-34c-5p, hsa-miR-138-5p, hsa-miR-377-3p, hsa-let-7c-5p, hsa-miR-200a-3p, hsa-miR-200b-3p, hsa-miR-200c-3p, and hsa-miR-197-3p might stimulate CD8⁺ T-cells and DCs-mediated anti-tumoral immune responses via inhibiting tumor-intrinsic PD-L1 expression in NSCLC.

The current study has several limitations that should be noted. First, we only included papers published in English before the mentioned date. The second limitation stems from the intrinsic limitation of single-cell RNA sequencing approaches, i.e., the count of sequenced cells in our single-cell RNA sequencing was limited compared to the count of sequenced cells in tumor bulk analysis. Nevertheless, this study has strengths as well. First, we systematically searched ten databases, i.e., Embase, Scopus, Web of Science, PubMed, Ovid, EBSCO, ProQuest, Cochrane library, African Index Medicus, and Pascal-Francis, to retrieve available papers published before the mentioned data. Second, we included in silico tumor bulk and single-cell RNA sequencing to enrich this study. Third, we predicted circRNA/miR/PD-L1 axes based on bioinformatic analyses and discussed their significance based on a thorough literature review of NSCLC development.

Overall, the ectopic expression of hsa-miR-194-5p, hsa-miR-326, hsa-miR-526b-3p, hsa-miR-34a-5p, hsa-miR-34c-5p, hsa-miR-138-5p, hsa-miR-377-3p, hsa-let-7c-5p, hsa-miR-200a-3p, hsa-miR-200b-3p, hsa-miR-200c-3p, and hsa-miR-197-3p, as PD-L1-inhibiting miR, can inhibit NSCLC development and stimulate anti-tumoral immune responses. Based on the single-cell RNA sequencing analyses, PD-1 might be considerably expressed on CD8⁺ T-cells and DCs; therefore, the ectopic expression of the abovementioned PD-L1-inhibiting miRs might stimulate the CD8⁺ T-cells and DCs-mediated anti-tumoral immune responses. Considering the multi-target nature of circRNA and miRs, identifying tumor-suppressive circRNA and miRs can provide a promising opportunity to overcome tumor growth, metastasis, and therapy resistance. Nevertheless, further studies are needed to identify, validate, and translate them into clinical practice.

Publicly available datasets were analyzed in this study. This data can be found here: geo databases, GSE158695, GSE63805, and GSE144945; and the TGCA database.

MS: developing the research question, performing the in silico and single-cell RNA sequencing analyses, searching the databases, selecting the studies from the initial systematic search, and writing the manuscript. FG and HH: selecting the studies from the initial systematic search. NA and NH: extracting the results from selected studies. AD and NSh: searching the literature. OB: reviewing and editing the manuscript. NSi and BB: reviewing and editing the manuscript and supervising the study. All authors contributed to the article and approved the submitted version.