Research note: the putative nuclear localization signal of duck circovirus type 1 cap protein is critical for VLP assembly

Abstract

Duck circovirus type 1 (DuCV-1) threatens the duck industry globally, with no approved vaccines available. This study aimed to establish a prokaryotic expression system for DuCV-1 virus-like particle (VLP) and clarify the role of the putative nuclear localization signal (NLS) in capsid (Cap) protein function and assembly. Codon-optimized full-length Cap and putative NLS-deleted dCap (residues 1–36) were expressed in Escherichia coli. Deletion of the putative NLS reduced Cap expression by 50%, solubility from 88% to 32.6%, and completely abolished VLP assembly. Full-length Cap self-assembled into 13–17 nm icosahedral VLP, structurally consistent with native virions. Homology modeling revealed the Cap monomer adopts a conserved jelly-roll fold, with the putative NLS mediating critical inter-subunit interactions at the 2-fold axis of the icosahedral capsid. Collectively, the putative NLS is an essential multifunctional motif for Cap protein folding, solubility, and VLP assembly, providing a theoretical and experimental basis for DuCV VLP-based vaccine development.

Introduction

Duck circovirus (DuCV) is a major pathogen threatening the global duck farming industry, causing growth retardation, immunosuppression, and secondary infections that lead to significant economic losses(Lei et al., 2024; Wang, et al., 2022). Among three DuCV genotypes, DuCV-1 is the most prevalent, widely distributed in East Asia (30–50% infection rate), and exhibits higher Cap protein conservation than DuCV-2/3, making it a preferred target for vaccine development(Li, et al., 2025).

Currently, no commercial DuCV vaccines are available. Inactivated vaccines are costly with limited efficacy, while infectious clone-based vaccines raise biosafety concerns (Li, et al., 2015; Zhaolong, et al., 2020). Virus-like particles (VLP) are ideal vaccine candidates due to their safety and strong immunogenicity (Mohsen and Bachmann, 2022; Nooraei, et al., 2021), but the structural basis of DuCV VLP assembly and the role of Cap’s putative nuclear localization signal (putative NLS) remain unclear. This study focused on exploring the putative NLS’s effects on Cap protein expression, solubility, and VLP assembly, and deciphering the underlying structural mechanism. It establishes a foundation for DuCV VLP vaccine development.

Materials and methods

Construction and expression of recombinant proteins

DuCV-1 Cap sequence was retrieved from NCBI GenBank (MN928800). The putative NLS was predicted using the SignalP-6.0 online tool (https://services.healthtech.dtu.dk/services/SignalP-6.0/), identifying a conserved motif ("RRAYRGRRKRRGLRRRFRRRRLRIARPRRR") corresponding to amino acid residues 7–36. Codon-optimized Cap and putative NLS-deleted dCap were cloned into pET30a+ and transformed into E. coli BL21 (DE3). Expression was induced with 0.5 mM IPTG at 37°C for 6 hours, followed by ultrasonication and fractionation into soluble and insoluble fractions.

Purification of cap and preparation of VLP

Soluble proteins were purified via Ni Sepharose affinity chromatography (GE Healthcare, IL, USA) with equilibration (20 mM sodium dihydrogen phosphate, 300 mM NaCl, 20 mM imidazole, pH 8.0), wash (50 mM imidazole), and elution (500 mM imidazole, pH 6.0) buffers.

For VLP assembly, purified Cap protein (1 mg/mL) was dialyzed to remove imidazole and promote self-assembly. The protein solution was placed in a dialysis bag with a 3.5 kDa molecular weight cut-off (Solarbio, Beijing, China) and dialyzed against assembly buffer (20 mM sodium dihydrogen phosphate, 300 mM sodium chloride, pH 6.0) at 4°C with constant stirring at 100 rpm for 24 hours. The assembly buffer was replaced three times during dialysis to ensure complete removal of imidazole. After dialysis, the sample was centrifuged at 12000 × g for 10 minutes at 4°C to remove aggregates, and the supernatant containing VLP was collected for structural and morphological analysis.

SDS-PAGE, western blot, and TEM analysis

Proteins were separated by 15% SDS-PAGE gel, transferred to PVDF membranes, and probed with anti-His monoclonal antibody (1:5000) and HRP-conjugated secondary antibody (1:10000). Bands were visualized with Luminata Forte substrate and quantified via Image J. VLP were observed by transmission electron microscopy (TEM, JEM-2100f, JEOL, Tokyo, Japan) after negative staining with 2% uranyl acetate.

Structural modeling and epitope prediction

Homology modeling of the DuCV-1 Cap monomer and VLP was performed using the SWISS-MODEL web server (https://swissmodel.expasy.org/), with the crystal structure of bat circovirus VLP (PDB ID: 6rpo) as the reference template. The model with the highest Quaternary Structure Quality Estimate (QSQE) and Global Model Quality Estimate (GMQE) scores was selected for further analysis. The three-dimensional (3D) structure was visualized and analyzed using PyMOL v3.11 software (Schrödinger, LLC, USA), including secondary structure identification, subunit interaction analysis, and surface epitope mapping.

Results and discussion

Putative NLS is essential for cap protein expression and solubility

Expression and solubility assays showed that deletion of the putative NLS had a profound impact on Cap protein production (Fig. 1A-1C). Under identical induction conditions, the total expression level of dCap was approximately 50% lower than that of Cap. Moreover, the soluble fraction of Cap accounted for 88% of total expression, while only 32.6% of dCap was soluble. This indicates that the putative NLS plays a crucial role in promoting correct protein folding and reducing inclusion body formation. This finding differs from observations in PCV2, where NLS deletion enhances Cap expression, and it highlights species-specific regulatory mechanisms of DuCV-1 Cap expression among circoviruses (Mo, et al., 2019).

Fig. 1.

Expression, purification, and in vitro self-assembly of DuCV Cap and dCap proteins.

(A) Schematic of recombinant protein constructs. (B) Protein purification. Lane M: Protein marker; NC: Negative control; T: Total protein; S: Soluble fraction; E1–E8: Elution fractions. (C) Solubility analysis. Upper panel: Quantitative analysis of band intensity using Image J software; lower panel: Corresponding Western blot results. (D) TEM image of Cap self-assembled into VLP (pH 6.0). (E) TEM image of dCap (failed to assemble into VLP and formed irregular aggregates only under pH 3.0. Scale bar = 50 nm.

Purification efficiency also differed between the two proteins: dCap achieved a 14-fold enrichment after nickel affinity chromatography, while Cap only showed a 1.45-fold enrichment. This discrepancy is likely due to the highly positive charge of the putative NLS region (containing 19 arginines and 1 lysine), which mediates non-specific electrostatic interactions with the nickel matrix, reducing target protein elution efficiency. These results suggest that while the putative NLS is essential for Cap protein function, its positive charge may pose challenges for purification. They provide insights for optimizing purification strategies in future large-scale production.

Putative NLS mediates VLP assembly via inter-subunit interactions

VLP assembly was assessed by TEM after dialysis across pH 3.0–9.0. The full-length Cap protein self-assembled into uniform spherical icosahedral VLP with a diameter of 13–17 nm, consistent with the size of naturally occurring DuCV virions (15–16 nm) reported previously (Fig. 1D)(Hattermann, et al., 2003). In contrast, under the same assembly conditions, the dCap protein failed to form regular VLP and only aggregated into irregular multimers (Fig. 1E). This indicates that the putative NLS is an indispensable structural motif for DuCV-1 VLP assembly, whereas the NLS of PCV2 is not required for VLP formation (Mo, et al., 2019).

Homology modeling analysis revealed the structural mechanism underlying NLS-mediated VLP assembly. The Cap monomer adopts a conserved jelly-roll fold (1 α-helix, 8 β-sheets, 7 loops) (Fig. 2A), and 60 monomers assemble into a typical icosahedral capsid (Fig. 2B), which is consistent with the assembly mode of other members of the Circoviridae family (Khayat, et al., 2011). The surface structure of the VLP (Fig. 2C) exhibits functionally relevant features. Loop CD forms a prominent "spike-like" protrusion likely mediating host cell receptor binding and Loop GH forms a "channel-like" structure potentially involved in genome release (Fig. 2D)(Khayat, et al., 2011). The positive charge of the putative NLS faces the interior of the capsid (Fig. 2H), which may facilitate viral genome packaging through electrostatic interactions with the negatively charged phosphate backbone of viral DNA.

Fig. 2.

Structural simulation and analysis of DuCV Cap and capsid.

(A) Ribbon diagram of the DuCV Cap monomer; secondary structures (loops, β-strands, and α-helix) are color-coded. (B) Schematic diagram of the icosahedral capsid showing the 2-fold, 3-fold, and 5-fold symmetry axes. (C) Surface structure of the icosahedral capsid. (D) Cross-sectional view of the capsid. (E) Cross-sectional view showing the inner surface of the capsid. (F) The 2-fold symmetry axis harboring the putative NLS, with two subunits symmetrically distributed on the capsid. (G) Outer surface morphology of the two adjacent subunits flanking the 2-fold symmetry axis on the capsid. (H) Interaction sites between the two adjacent subunits flanking the 2-fold symmetry axis. Owing to the highly dynamic nature of the NLS structure that precludes accurate prediction, all structures illustrated herein exclude the NLS region.

Further analysis showed that the putative NLS region is localized near the 2-fold symmetry axis of the capsid, mediating critical inter-subunit interactions (Figs. 2F, 2G). This mediating role is achieved through two coordinated molecular mechanisms involving the NLS and its adjacent regions. Specifically, tyrosine 42 (TYR42) and tyrosine 114 (TYR114)—residues adjacent to the putative NLS—form a conserved hydrophobic core between adjacent subunits (Fig. 2H), and this hydrophobic interaction is highly conserved in DuCV-1. The core of this interaction relies on hydrophobic forces, and notably, in DuCV-2/3 where TYR42 is mutated to phenylalanine (PHE), the key hydrophobic interaction is retained. This is because both tyrosine and phenylalanine are aromatic amino acids with conserved benzene ring structures, which are essential for maintaining the hydrophobic core. This hydrophobic interaction serves as a structural anchor for subunit dimerization, while the positively charged amino acid residues within the putative NLS itself form additional electrostatic interactions and hydrogen bonds with residues from neighboring subunits. Together, these complementary interactions (hydrophobic from NLS-adjacent residues and electrostatic/hydrogen bonds from NLS-inherent residues) synergistically stabilize the capsid structure. Deletion of the putative NLS not only eliminates the electrostatic and hydrogen bond interactions mediated by its positively charged residues but may also perturb the spatial conformation of the adjacent TYR42 and TYR114, thereby compromising the formation of the conserved hydrophobic core. Collectively, these disruptions of inter-subunit interactions lead to the loss of VLP assembly ability. These structural insights provide a basis for understanding the DuCV assembly mechanism, but these speculations require further experimental validation.

Implications for DuCV vaccine development

This study establishes a robust, cost-effective, and scalable E. coli-based DuCV-1 VLP expression system, offering distinct advantages over existing platforms: it avoids incomplete inactivation risks and high costs of PBMC-derived inactivated vaccines, while being non-replicative and inherently safe compared to infectious clone-based vaccines (Zhaolong, et al., 2020).

The putative NLS, identified as critical for Cap protein folding, solubility, and VLP assembly, is indispensable for recombinant constructs. However, its positive charge may hinder purification. Future work could explore alternative strategies (e.g., ion-exchange chromatography) or soluble tags to improve yields without compromising NLS function.

In conclusion, the insights into the putative NLS’s pivotal role and the established E. coli system lay a solid foundation for developing safe, effective, and industrially applicable DuCV vaccines. Future research should focus on verifying VLP immunogenicity and protective efficacy in ducks.

CRediT authorship contribution statement

Xinnuo Lei: Writing – original draft, Visualization, Validation, Software, Methodology, Formal analysis, Data curation, Conceptualization. Peiqing Wei: Visualization, Validation, Investigation, Formal analysis, Data curation. Dongliang Wang: Visualization, Validation, Software, Resources, Investigation. Yifan Jiang: Visualization, Validation, Formal analysis, Data curation. Wanting Yu: Visualization, Validation, Methodology. Anping Wang: Validation, Supervision, Resources, Funding acquisition. Zhi Wu: Writing – review & editing, Supervision. Shanyuan Zhu: Writing – review & editing, Supervision, Project administration, Funding acquisition.

Disclosures

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.