MicroRNA-628-5p Facilitates Enterovirus 71 Infection by Suppressing TRAF3 Signaling

To date, EV71 infection still poses a great challenge to the health of infants and young children.¹ Much established evidence suggests that EV71 infection depends on a wide variety of host factors, including cell surface receptors for EV71 entry, innate immune response, miRNAs, and lncRNAs.^2,3 It is thought that diverse signaling pathways are required for EV71 to escape the host immune response.⁴ Innate immunity serves as the first line of defense against pathogens. Tumor necrosis factor receptor-associated factor (TRAF) proteins are essential components of signaling pathways activated by Toll-like receptor (TLR) or RIG-I-like receptor (RLR) family members. TRAF3 is widely expressed by many cell types, including all nucleated immune cells, in which it plays many roles in the regulation of immune functions.⁵ TRAF3 plays a significant role in regulating the functions of B and T lymphocytes. TRAF3 is a highly versatile regulator that positively controls type I interferon and cytokine production through interferon regulatory factors (IRFs) and nuclear factor-κB (NF-κB).⁶ A recent study indicated that EV71-induced ubiquitin-specific protease 19 (USP19) negatively regulates type I IFN signaling by targeting TRAF3.⁷

Currently, increasing attention has been paid to the roles of miRNAs in the EV71-host interaction. Upregulation of miR-493-3p⁸ and the miR-302 cluster⁹ suppresses the EV71-induced innate immune response via direct targeting of PTEN and karyopherin α2 (KPNA2), respectively. However, the functional roles of other miRNAs during EV71 infection remain largely unknown. In addition, the significance of these miRNAs needs to be addressed in clinical settings in EV71-infected patients.

miR-628-5p is involved in the initiation, proliferation, invasion, and metastasis of numerous types of human cancers^10,11 and the B cell receptor signaling pathway,¹² but the function of miR-628-5p during viral infection is poorly understood. A recent study indicated that the upregulation of miR-628-5p contributes to viral infection-associated disorder.¹³ In the present study, we showed by RNA-seq that the level of miR-628-5p was increased after EV71 infection. This result was confirmed by qRT-PCR at different time points after infection (Fig. 1a) and at different doses of EV71 (Fig. 1b). By performing computational analysis and luciferase reporter assays, we found that miR-628-5p could target TRAF3, as miR-628-5p mimics inhibited the luciferase activity of the wt TRAF3 3′UTR reporter (Fig. 1c). TRAF3 positively controls type I interferon and cytokine production through activation of IRF3 and NF-κB.⁶ We hypothesized that miR-628-5p upregulation helps EV71 evade innate immune responses by interfering with TRAF3 expression and its downstream signaling. To test this hypothesis, we performed Western blotting analysis of proteins in TRAF3-mediated signaling pathways. As shown in Fig. 1g, the expression of TRAF3 and two proteins downstream of TRAF3, NF-κB p65 and IRF3, was decreased in a dose (MOI)-dependent manner after EV71 infection. In addition, VP1 and cleaved Caspase-3 were increased in a dose (MOI)-dependent manner after EV71 infection. The mRNA expression of TRAF3 (Fig. 1d) and IFN-β transcription (Fig. 1e) were also decreased after overexpression of miR-628-5p. Interestingly, the overexpression of miR-628-5p could promote EV71 replication, as measured by qPCR and Western blotting (Fig. 1f, h), suggesting that EV71-induced miR-628-5p could target TRAF3 signaling to inhibit IFN-β transcription and promote EV71 replication.

Fig. 1.

a, b A total of 5 × 10⁵ RD cells were seeded into 6-well plates and then infected with EV71 at an MOI = 1 for the indicated doses or times. The levels of miR-628-5p in EV71-infected RD cells were detected by qRT-PCR. c Luciferase reporter assay of TRAF3 3′UTR. RD cells were cotransfected with the wild-type (wt) or indicated mutants of the TRAF3 3′UTR reporter plasmid and miR-628-5p mimics or NC mimics for 48 h. The cell lysate was prepared for luciferase assay. RD cells transfected with miR-628-5p mimics or NC were infected with EV71 (MOI = 1) for 24 h, and the transcription levels of TRAF3 (d) and IFN-β1 (e) and EV71 copy numbers (f) were measured by qRT-PCR. The expression levels of TRAF3, p65, IRF3, VP1, and cleaved Caspase-3 were measured by Western blotting (h). g Western blot analysis of TRAF3-mediated signaling after EV71 infection. RD cells were infected with EV71 at the indicated doses, and total protein was extracted. The expression levels of TRAF3, p65, IRF3, VP1, Caspase-3, and cleaved Caspase-3 were measured by Western blotting. i–k RD cells cotransfected with pcDNA3.1‐TRAF3 or pcDNA3.1 vector and miR-628-5p mimics or NC mimics were infected with EV71 (MOI = 1) for 24 h. The expression levels of TRAF3, p65, and IRF3 were measured by Western blotting (i), and the transcription levels of IFN-β1 (j) and EV71 replication (k) were measured by qRT-PCR. (l) RD cells (5 × 10⁵) were seeded into 6-well plates and then transfected with miR-628-5p mimics or NC mimics, (m) and miR-628-5p mimics+pcDNA3.1‐TRAF3 or NC mimics+pcDNA3.1‐TRAF3. After 24 h of transfection, RD cells were infected with EV71 (MOI = 1). Whole cell lysates and cytoplasmic and nuclear protein extracts were subjected to Western blot analysis. RD cells (5 × 10⁵) were seeded into 6-well plates and transfected with miR-628-5p inhibitor or NC inhibitor at a concentration of 250 nM for 24 h and then infected with EV71 (MOI = 1) for 24 h. The transcription levels of IFN-β1 (n) and EV71 replication (o) were tested by qRT-PCR. p, q Total RNA was extracted from the serum of healthy controls and EV71-infected patients by using the BIOG Free RNA Extraction Kit. The expression of miR-628-5p in the sera of healthy controls, mild cases, and severe cases (p) and in the sera of severe cases before and after treatment (q) was determined by qRT-PCR. Lamin B1 and β-actin were used as internal references for Western blotting, and GAPDH was used as an internal reference for qRT-PCR. All experiments were repeated at least three times. Bar graphs are presented as the mean ± SD, n = 3. *P < 0.05, **P < 0.01, ***P < 0.001

To determine whether miR-628-5p suppresses EV71-induced IFN-β transcription by directly targeting TRAF3, we cotransfected RD cells with either NC mimics or miR‐628-5p mimics with or without TRAF3. At 24 h post transfection, RD cells were infected with EV71 at an MOI of 1 for an additional 24 h. As shown in Fig. 1i, the expression levels of VP1 and activated Caspase-3 were clearly increased in RD cells transfected with pcDNA3.1‐TRAF3 in the presence of mimics, while p65 and IRF3 expression levels were decreased. Next, we assayed the transcription level of IFN-β1 mRNA (Fig. 1j) and EV71 copy numbers (Fig. 1k) by qRT-PCR. Overexpression of TRAF3 suppressed EV71 replication in RD cells and attenuated the effects of miR-628-5p overexpression on EV71 replication and IFN-β1 transcription. Taken together, our data suggest that miR-628-5p suppresses EV71-induced IFN-β transcription by directly targeting TRAF3.

To further investigate how miR-628-5p affects EV71-induced IFN-β transcription through p65 and IRF3, we transfected RD cells with miR-628-5p mimics or NC mimics (Fig. 1l) with or without pcDNA3.1‐TRAF3 (Fig. 1m). At 24 h after transfection, these cells were infected with EV71 for 24 h, and the expression levels of p65 and IRF3 in whole cell lysates and cytoplasmic and nuclear extracts were analyzed. As shown in Fig. 1l, m, mimic transfection clearly decreased TRAF3, p65, and IRF3 expression and increased VP1 expression in whole cell lysates. NF-κB p65 and IRF3 activation leads to their nuclear translocation. As shown in Fig. 1l, m, nuclear translocation of p65 and IRF3 was increased upon EV71 infection in a time-dependent manner, which could be suppressed by mimics. Moreover, TRAF3 overexpression significantly promoted p65 and IRF3 nuclear transport. Our data suggest that miR-628-5p suppressed TRAF3-mediated p65 and IRF3 nuclear translocation during EV71 infection, which further affected IFN-β transcription.

Multiple signaling pathways, miRNAs, and EV71 proteases have been reported to associate with innate immune evasion by EV71.⁴ However, there are still no effective drugs against EV71 infection. To further understand the effect of miR-628-5p on EV71 replication, we used a specific inhibitor to block the function of endogenous miR-628-5p. Transfection of the miR-628-5p inhibitor significantly blocked the downregulation of IFN-β1 mRNA compared with the NC transfection (Fig. 1n). EV71 replication was also decreased in RD cells in the presence of this inhibitor (Fig. 1o). Taken together, our data indicate that inhibiting the function of miR-628-5p could suppress EV71 replication. Finally, the expression of miR-628-5p was analyzed in serum samples from EV71-positive patients (mild cases: n = 52; severe cases: n = 42) and healthy individuals (n = 24). As shown in Fig. 1p, q, we found that the expression level of miR-628-5p in the sera of mild and severe cases was significantly higher than that in healthy individuals (Fig. 1p), and the expression of miR-628-5p was downregulated after treatment (Fig. 1q). Our study suggests that a high level of miR-628-5p may contribute to the development of EV71-infected HFMD. A number of serum exosomal or saliva-based miRNAs proposed as HFMD diagnostic markers were reported recently.^14,15 Our data are in agreement with published data showing that the serum level of miR-628-3p, which is in the same cluster as miR-628-5p, was significantly higher in patients with enterovirus infection than in the control group.¹⁵

In conclusion, EV71 infection induces miR-628-5p upregulation and further suppresses IFN-β transcription by inhibiting TRAF3 signaling. miR-628-5p also reduces the translocation of activated p65 and IRF3 from the cytosol into the nucleus and the subsequent transcription of IFN-β. Our findings showed a new strategy for EV71 to escape the host innate immune response. Our data also provide a new strategy to treat EV71 infection by using a miR-628-5p inhibitor.

Author contributions

G.C.D., Y.F.J., D.L. and S.Y.C. conceived and designed the research; D.L., C.Z., T.T.S., Y.D., R.H.D., and Y.L.G. performed the research; Y.F.J. and D.L. analyzed the data; and the manuscript was written by Y.F.J. and D.L. and revised by W.G.Z.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version of this article (10.1038/s41423-020-0453-4) contains supplementary material.