**4. Discussion**

As the most prominent component of GTCs, EGCG displayed broad and distinguished anti-viral activity compared to other GTC members [36]. Here, we showed that EGCG exhibited an inhibitory effect on the PCV2 infection at the concentration of 100 μM, without causing significant toxicity to the host cells. EGCG acted directly on PCV2 virions rather than any cellular factors with or without the stimulation of virions, and EGCG inhibited virion attachment to host cells by competitively interacting with capsid with cell surface receptor, but it did not affect other post-binding stages.

Heparan sulfate is a linear, synthetic, acidic polysaccharide belonging to the glycosaminoglycan (GAG) family, which is widely distributed on the surface and in extracellular matrix of various mammalian cells [37]. HS was attached to the protein core and consisted of repeating 1 to 4 linked disaccharide units, one of which is anα-d-glucosamine residue and the other is an anionic acid. HS chains acquired negative charge though epimerization and terminal sulfation processes, thereby facilitating HS in establishing electrostatic interaction with positively charged motifs and residues [38], including signaling pathway proteins, growth factors, plasma proteins, immune modulators and viral structural proteins [39]. In general, these protein ligands possessed distinct HS binding region, which acted as a complementary structure for heparin-protein interactions [40]. Based on the known ligands' structural details, the molecular modeling of protein-glycosaminoglycan revealed that HS binding regions formed a hydrophilic pocket wrapped around and folded over heparin oligosaccharides. The key factors to produce such hydrophilic pocket were the clusters contributed by basic amino acids, especially arginines and lysines [41,42].

The EGCG interacted with PCV2 capsid. Due to the considerable affinity, this interaction could competitively impair the binding of capsid to HS, thereby inhibiting the attachment of virions to the host cell surface. Meanwhile, the EC could not interact with the virus capsid, and hence did not exert an antiviral effect. Structurally, EGCG possessed numerous hydroxyl radicals attached on an extra benzene ring which is not the case for EC (Figure 1a), and the result of molecular docking showed that these additional free radicals were involved in forming hydrogen bonds with ASP70 and ARG-73 of capsid. Indeed, the mutagenesis assay demonstrated that aspartic acids (ASP70 and ASP78) and arginines (ARG51 and ARG73) were essential for EGCG- capsid interaction. It is also worth noticing that the arginine residues of GAG ligand were also important for the formation of a center with a high positive charge density, which could electrostatically interact with the acidic groups of HS; this was supported by the result of the replacement of ARG51 and ARG73 which remarkably attenuated the affinity of capsid to HS. Based on these findings, we conclude that the interaction of EGCG with PCV2 capsid might primarily depend on the formation of hydrogen bonds by specific positively charged amino acids in capsid, such as ARG51 and ARG73, while newly formed links might thus interfere with the original electrostatic interaction between capsid and HS, resulting in the impairment of virus attachment to the cell surface.

In previous studies, EGCG showed similar antiviral activity against various viruses, EGCG inhibited the infection of PRRSV with an EC50 of 48.2–63.09 μM in vitro [43], and EGCG exhibited an antiviral effect against JEV with an IC50 of 4.9–20 μM in vitro [44]. In our study, the EC50 of EGCG against PCV2 in vitro was calculated as 37.79 ± 1.64 μM. These comparable findings suggested that the antiviral activity of EGCG in vitro did not reach the efficacy of traditional antiviral drugs, such as ribavirin. It is worth noting that although this level of potency of EGCG at this stage is not of practical value in the treatment of clinical virus infections, the identification of key amino acids of viral proteins that bind EGCG is valuable for the development of structure-based anti-virus drugs, such as the EGCG-based synthetic EGCG palmitate, whose EC50 against PRRSV is decreased to 5.86–12.69 μM, which is nearly 5 times lower than for the original EGCG [45]. On the other hand, more research data focusing on in vivo antiviral effects of EGCG and detailed structural analysis of EGCG-protein interaction should be studied in future research to facilitate a more comprehensive assessment of the antiviral potential of EGCG.

**Author Contributions:** Conceptualization, J.L., J.G. and J.Z; methodology, J.L.; software, J.L.; validation, D.S., S.W.; formal analysis, J.L. and D.S.; investigation, D.S.; resources, Y.D.; data curation, J.L. and J.G.; writing—original draft preparation, J.L.; writing—review and editing, J.G. and J.Z.; visualization, J.L. and D.S.; supervision, J.G. and J.Z.; project administration, J.Z.; funding acquisition, J.Z. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Key Research & Development Program of China (Grant No. 2017YFD0500101 & 2015BAD12B01) and the National Natural Science Foundation of China (Grant No. 31402198).

**Acknowledgments:** We thank the National Key Laboratory of Natural Medicines for their support of MST instruments and technology, and also the instrumental and technical assistance on heparin column chromatography from technicians of general electric (GE) healthcare life sciences.

**Conflicts of Interest:** The authors declare no competing interests. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
