As a linker protein of cytoskeleton and nucleoskeleton, SYNE3/nesprin-3 has been shown to play a vital role in 3D cell migration  and tumor cell movement  in previous studies. However, these researches too limited to reveal SYNE3 functions in tumor. Therefore, in our study, we hoped to provide some basic knowledge of SYNE3 and analyze its roles in cancer.
Based on our analysis, SYNE3 was conserved among mammals and its family members, suggesting some vital roles of SYNE3 to keep through evolution. Moreover, SYNE3 shared more similarities with SYNE1 and SYNE2 than SYNE4 in its structure and may in its functions as well.
We investigated SYNE3 expression in various tissues. In normal human tissues, SYNE3 was strongly detected in breast and lung, corresponding with the highest mRNA level in these two tissues. Moreover, the expression difference between normal and tumor tissues was most obvious in breast and lung as well in both IHC and GEPIA. However, conflicts also existed, for instance, kidney was of low mRNA but moderate to high SYNE3 staining. For one hand, our tissues were just randomly selected cases, so they cannot represent general conditions like data from UCSC or GEPIA. For another hand, the mRNA level was not always consistent with protein level, though there was no related study on SYNE3 expression. In addition, though SYNE3 had been reported to exist widely, its expression was not always high enough to be detected.
In our prognostic analysis, we found higher SYNE3 expression level associated with longer OS in patients with KIRC, LUAD, CESC or HNSC except LGG with negative correlation. In most cases, the expression of SYNE3 was also downregulated in tumor. Recent study showed that knockdown of SYNE3 led to DNA damage, genome organization loss, and transcriptional changes , and these events were closely related to tumorigenesis for this mechanism had been reported in SYNE1. Therefore, SYNE3 participated in some anti-tumor events and there might be various mechanisms to adjust SYNE3 expression.
CeRNA is a conception firstly proposed in 2011. In ceRNA network, non-coding RNA, likely lncRNA or circRNA, can competitively bind with miRNA, thereby weaken the repression from miRNA to mRNA. We identified hsa-miR-330-3p and hsa-miR-149-5p as key regulatory miRNAs of SYNE3. MiR-330-3p was a recognized cancer-promoting factor. It was reported that miR-330-3p can target the gene of program cell death 4 (PDCD4) to promote tumor . MiR-330-3p is also related with metastasis and invasion of lung cancer, and glutamate receptor 3 (GRIA3)  and human manganese superoxide dismutase 2 (hSOD2)  are its two target genes. Researches found miR-330-3p is correlated with bad prognosis in breast cancer as well , in which miR-330-3p can target Collagen and Calcium Binding EGF Domains 1 (CCBE1) to promote cancer metastasis . Higher miR-330-3p led to lower OS in renal cell carcinoma (RCC), though miR-330-3p also presented some tumor-suppressing functions in RCC , suggesting that further studies were needed to explain this paradox. However, the function of miR-149-5p depends on specific tumor types. MiR-149-5p can help to disable the invasion and proliferation of medullary thyroid carcinoma cells  and suppress the development of non-small cell lung cancer  and hepatocellular carcinoma [25, 26]. Combined with our survival analysis, tumor-promoting functions of miRNAs displayed more obvious in KIRC, CESC, LUAD and HNSC. While for LGG, there might be other mechanisms related to its unique progressive features or micro-environment, causing a contrary result.
In our ceRNA network, NEAT1 and MIR503HG are tumor suppressors targeting both miR-330-3p and miR-149-5p. NEAT1 can be mediated by Tumor protein P53 (P53), a critical tumor suppressor, to suppress transformation and cancer initiation . MIR503HG can target heterogeneous nuclear ribonucleoprotein A2/B1 (HNRNPA2B1) and NF-κB signaling pathway to inhibit migration of hepatocellular carcinoma . Meanwhile, only a few of the rest lncRNAs have been reported, including both tumor suppressors and tumor enhancers. These results showed double regulations in this ceRNA network of SYNE3, and tumor-suppresing lncRNAs seem to role more dominantly.
As for PPI network of SYNE3, it was reasonable for SYNE3 to interact with Sad1 And UNC84 Domain Containing (SUN) family, a set of cytoskeleton linker proteins and the other SYNE members. The function of SYNE3 is correlated with these binding partners, which was supported by GO and KEGG analysis. Based on our analysis, SYNE3 is correlated with some tumorous processes. P53 and 3 interacting genes of SYNE3 were involved in P53 signaling pathway, including cyclin-dependent kinase 1 (CDK1), Caspase 8 (CASP8) and cyclin B1 (CCNB1). CDK1 can adjust the cell cycle , CASP8 is referred as an apoptosis regulator , and CCNB1 is vital for mitosis and proliferation . Moreover, apoptosis, a pathway closely to P53, also enriched 5 SYNE3 interacting genes, involving CASP8 again, LMNA, CASP6, LMNB1 and LMNB2. Among them, downregulation of CASP6 induced suppression of the apoptosis of chronic myeloid leukemia cells . LMNB1 and LMNB2 are coding genes for lamin B1 and lamin B2 respectively, which are nucleoskeleton components associated with cancer and aging . Besides, 9 interacting genes of SYNE3 code ribosomal proteins, indicating a possible association with transcription. Based on these, SYNE3 may suppress cancer by promoting tumor apoptosis, which can be realized through adjustment of cell cycle, recruitment of immune cells and alteration of nucleoskeleton and ribosome.
We finally focused on SYNE3 and immune property of tumor microenvironment. SYNE3 showed good correlation with tumor immune infiltration, especially in LUAD, as its OS was most relevant with its tumor microenvironment, especially with B cell and DC cell. It has already been reported that B-cell infiltration correlated with good prognosis in LUAD . Dendritic cell is responsible for antigen presentation to T- and B-cells and activation natural killer (NK) cells and can activate anti-tumor immunity . Therefore, regulating the infiltration of B cell and DC cell can be an effective pathway for SYNE3 to influence the survival of patients with LUAD. GSEA analysis also showed that higher SYNE3 expression relating to pathways enhancing B cell and DC cell. Furthermore, SYNE3 also significantly correlated with markers of B cell and DC cell. Therefore, we supposed that higher expression of SYNE3 helped with the production and collection of B cell and DC cell, so that SYNE3 altered tumor microenvironment and presented significant and reasonable prognostic value in LUAD.