To the Editor

MDSCs are not a single defined cell population in myeloid cells, but a mixture of a large number of granulocytes, macrophages, and dendritic cells that are hindered in differentiation and maturation. The phenotypic identification of MDSCs is extremely complicated, MDSCs mainly co-express CD11b and Gr-1, which mainly including PMN-MDSCs (CD11b+ Ly6G+ Ly6Clow) and Mo-MDSCs (CD11b+ Ly6G− Ly6Chigh) in mice. They usually perform immunosuppressive function in different ways [1, 2]. Increasing evidences show that lncRNAs play an important role in the establishment of immune cell lineage and immune response because of its complexity in regulation, self-composition and structure [3]. However, the relationship between lncRNAs and MDSCs has not attracted widespread attention.

LncRNA Snhg6 is a novel lncRNA, which abnormally expresses in a variety of cancers [4,5,6]. By analyzing Arrarystar lncRNA microarray of Tu-MDSCs and SP-MDSCs (MDSCs derived from tumor tissue and spleen of mice with Lewis lung cancer xenograft, respectively), we finally chose lncRNA Snhg6, which is highly expressed in Tu-MDSCs, as the object of this study (Fig. 1a, Additional file 1: Fig. S1, Additional file 2: S2, Additional file 5: Table S1). To investigate the effects of lncRNA Snhg6 on MDSCs, we first transfected the specific siRNA (si-Snhg6) or overexpression lentivirus (Lv-Snhg6) of lncRNA Snhg6 in bone marrow cells and then induced MDSCs under the stimulation of GM-CSF and IL-6 (Additional file 3: Fig. S3, Additional file 6: Table S2.). The results revealed that the differentiation rate and absolute number of CD11b+ Gr-1+ MDSCs did not change significantly whether the expression of lncRNA Snhg6 was decreased or increased (Fig. 1b–g). Further studies showed that there was also no significant change in CD11b+ Ly6G+ Ly6Clow PMN-MDSCs, while the percentage of CD11b+ Ly6G− Ly6Chigh Mo-MDSCs was significantly reduced after lncRNA Snhg6–silencing (Fig. 1h, i). And overexpression of lncRNA Snhg6 increased the percentage of CD11b+ Ly6G− Ly6Chigh Mo-MDSCs (Fig. 1j, k). All of these indicated that lncRNA Snhg6 was involved in promoting the differentiation of Mo-MDSCs.

Fig. 1
figure 1

LncRNA Snhg6 promotes Mo-MDSCs but not affects the differentiation of PMN-MDSCs. a LncRNA Snhg6 expression in Tu-MDSCs compared with SP-MDSCs measured by qRT-PCR. b, c Typical flow cytometry of CD11b+ Gr-1+ MDSCs. d–g The percentage and absolute number of CD11b+ Gr-1+ MDSCs by statistical analyses. h, i Typical flow cytometry and the percentage of CD11b+ Ly6G+ Ly6Clow PMN-MDSCs and CD11b+ Ly6G− Ly6Chigh Mo-MDSCs after downregulating the expression of lncRNA Snhg6. j, k Typical flow cytometry and percentage of CD11b+ Ly6G+ Ly6Clow PMN-MDSCs and CD11b+ Ly6G− Ly6Chigh Mo-MDSCs after increasing the expression of lncRNA Snhg6. Each expression had three replicates, ns: no significance; * p < 0.05; ** p < 0.01

Full size image

The specific mechanism by which lncRNAs play a regulatory role is often determined by their subcellular location [7]. So next we detected the cellular distribution of lncRNA Snhg6 in MDSCs. RNA Fluorescence in situ Hybridization (RNA-FISH) revealed that lncRNA Snhg6 was located in both the cytoplasm and the nucleus (Fig. 2a). In addition, we also measured the expression of lncRNA Snhg6 in nuclear and cytoplasmic fractions of MDSCs by qRT-PCR. The results were consistent with RNA-FISH, which further verified that lncRNA Snhg6 was mainly located in the cytoplasm of MDSCs (Fig. 2b). Histone methyltransferase Enhancer of Zeste homolog 2 (EZH2) is a histone methyltransferase catalyzing the methylation of histone H3 at lysine 27. The latest research showed that an inhibitor of EZH2 activity—GSK343 could significantly promote the differentiation of hematopoietic stem cells (HPCs) into MDSCs in the presence of granulocyte–macrophage colony-stimulating factors (GM-CSF) and interleukin-6 (IL-6) in vitro [8]. In addition, the involvement of lncRNA Snhg6 in regulating EZH2 has also been reported [6, 9]. So we speculate that lncRNA Snhg6 may regulate the differentiation of MDSCs through EZH2. The following experiment proved that lncRNA Snhg6 could regulate the expression of EZH2 at the post-transcriptional rather than transcriptional level (Fig. 2c–f, Additional file 5: Table S1). Subsequently, the protein expression of EZH2 was detected at 0 h, 3 h, and 6 h, respectively, after adding cycloheximide (CHX). The results revealed that the stability of EZH2 protein significantly improved after downregulating lncRNA Snhg6 (Fig. 2g, h). Further Immunoprecipitation (IP) testing showed that the ubiquitination level of EZH2 was obviously reduced as lncRNA Snhg6 decreased (Fig. 2i). These suggest that lncRNA Snhg6 was likely to regulate the stability of EZH2 through protein-ubiquitination degradation pathway in the differentiation process of MDSCs. Of course, protein could be degraded either by the ubiquitin proteasome or through the lysosomal pathway after the protein is ubiquitinated [10]. The specific degradation mechanism of EZH2 in our study remains to be further study.

Fig. 2
figure 2

LncRNA Snhg6 reduces the stability of EZH2 protein. a The cellular localization of lncRNA Snhg6 was detected by Cy3-labeled lncRNA Snhg6 probe. Cy3-labeled 18S probe was used to indicate plasmid localization and Cy3-labeled U6 probe was used to indicate nuclear localization. DAPI was used to evaluate the cell nucleus. b Subcellular fractionation was isolated of MDSCs, and lncRNA Snhg6 localization was examined by qRT-PCR. 18S and U6 were used as cytoplasmic and nuclear indicators, respectively. c, d qRT-PCR were used to detect the expression of EZH2 at mRNA level. e, f Western blot were used to detect the expression of EZH2 at protein level. g Western blot were used to detect the expression of EZH2 with CHX (40 µg/ml) treated 0 h, 3 h and 6 h after transfecting si-Snhg6. h The statistical graph corresponding to the left. i RIP assays were used to investigate the ubiquitination of EZH2. Each expression had three replicates, ns: no significance; *p < 0.05; **p < 0.01

Full size image

The occurrence and development of tumors are inseparable from the tumor microenvironment with immunosuppressive characteristics, and the massive accumulation of immunosuppressive MDSCs in the tumor microenvironment is the main cause of tumor immune non-response. The previous experimental results in our laboratory confirmed that compared with SP-MDSCs, Tu-MDSCs had a stronger ability to inhibit CD4/CD8 T cells [11]. Therefore, we detected the inhibitory effect of MDSCs on CD4+ T proliferation and its immunosuppressive effector molecules arginase (Arg-1), nitric oxide (NO) and reactive oxygen species (ROS). All results showed that lncRNA Snhg6 did not participate in regulating the immunosuppressive function of MDSCs (Additional file 3: Fig. S3 and Additional file 4: Fig. S4, Additional file 5: Table S1).

In short, we found that lncRNA Snhg6 was involved in regulating the differentiation of MDSCs by reducing the protein stability of EZH2, but it did not affect the immunosuppressive function of MDSCs, which might provide a new perspective for the treatment of cancer.