A receptor-Fc based SFV replicon Trim-Away platform for targeted viral protein degradation

Abstract

Background

Developing effective antiviral strategies is urgently needed during global viral pandemics. Traditional approaches, including small-molecule inhibitors, neutralizing antibodies, and RNA interference (RNAi), often face challenges such as drug resistance, limited specificity, and inefficient delivery. These limitations highlight the pressing need for innovative strategies focused on the targeted degradation of viral proteins.

Methods

We developed an optimized Trim-Away system employing a receptor-Fc fusion protein strategy. This system integrates the E3 ubiquitin ligase TRIM21 with engineered receptor-Fc proteins to ensure highly specific recognition and intracellular degradation. A key innovation is the use of the Semliki Forest virus (SFV) self-amplifying replicon (pSFV). This platform enables sustained and robust expression of the Trim-Away components. Furthermore, this plasmid-based delivery eliminates the need for protein purification, thereby streamlining the process and improving delivery efficiency.

Results

The system effectively degrades diverse viral targets. Specifically, it successfully degraded the spike proteins of both wild-type SARS-CoV-2 and its Omicron variant. It also targeted adeno-associated virus (AAV) capsid proteins. In vivo assays further confirmed that the self-amplifying replicon markedly reduces AAV-encoded luciferase expression. These data demonstrate that the system maintains high potency even at low dosages.

Conclusions

Our findings demonstrate that the pSFV-driven Trim-Away system is a powerful tool for viral protein degradation. The receptor-Fc strategy provides a significant advantage against rapidly mutating viruses. This study establishes a versatile and adaptable platform for future antiviral intervention.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12985-026-03235-w.

Background

COVID-19 and other viral pandemics have highlighted the severe impact of viral infections on global health [1]. While current antiviral strategies include vaccines, drugs, and public health measures, they face significant limitations [2]. Therefore, there is a critical need to innovate antiviral approaches to better protect public health.

Targeting viral structural proteins for degradation represents a promising antiviral strategy [3]. Directly inducing protein degradation offers rapid action, flexibility, reduced genomic interference, and the ability to address challenging targets, making it an attractive approach in both research and therapy [4]. This strategy aims to deplete essential viral components, thereby curtailing the virus’s ability to replicate and infect host cells [5]. So far, several methods have been employed to achieve protein degradation, such as the ubiquitin-proteasome system (UPS) [6, 7], the autophagy-lysosome pathway [7, 8], and small-molecule drugs [9, 10], among others. Among these, proteolysis-targeting chimeras (PROTACs), bifunctional molecules that recruit E3 ubiquitin ligases to viral proteins for UPS-mediated degradation, have shown potential against viruses like HPV [11] and influenza [12] but suffer from insufficient specificity against rapidly mutating strains and vulnerability to viral resistance [13–15].

Trim-Away is an advanced strategy that depends on the Ubiquitin-Proteasome System (UPS) [16]. It utilizes TRIM21, a member of the E3 ligase family [17], to target and degrade specific proteins rapidly and efficiently. TRIM21 binds to the Fc region of antibodies attached to target proteins, marking the antibody-antigen complex for rapid degradation by the proteasome [18]. Compared to PROTACs, which use small molecule linkers to recruit E3 ubiquitin ligases, Trim-Away employs antibodies as connectors, providing greater specificity and flexibility [19, 20]. This biological approach reduces off-target effects and other side effects commonly associated with small molecule linkers in PROTACs. However, Trim-Away’s reliance on high-affinity antibodies presents critical limitations, particularly when targeting mutant viral proteins [16]. Mutations can reduce antibody binding affinity, necessitating the continuous development of high-affinity antibodies, which is time-consuming and resource-intensive [21]. Furthermore, the efficiency of delivering antibodies into the cytoplasm of target cells remains to be improved [22].

To overcome these drawbacks, we modified the Trim-Away system by replacing the antibody’s Fab fragment with host receptors that naturally bind viral proteins, retaining the Fc region to ensure TRIM21 recruitment. This receptor-based design leverages inherent viral protein-receptor binding affinity, eliminating the need for antibody screening and enhancing capture of even highly infectious mutant viral strains [23].

In experimental validation, the Trim-Away system effectively targeted adeno-associated virus (AAV) capsid proteins via an AAVR-Fc fusion, utilizing AAVR as the native cell surface receptor for viral recognition. This targeted degradation profoundly depleted intracellular viral capsid proteins, thereby severely restricting the overall production of functional recombinant AAV particles. Consistent in vivo results showed reduced AAV-delivered luciferase activity following pSFV-Trim-Away-mediated interference. The pSFV platform is a plasmid-based self-amplifying replicon system previously constructed in our laboratory. Notably, high and sustained Trim-Away expression via the self-replicating pSFV plasmid enabled efficient degradation of SARS-CoV-2 spike protein across multiple variants at low doses.

Collectively, our modified Trim-Away strategy preserves the core advantages of the original system while addressing its key limitations, providing a robust and adaptable antiviral platform to combat evolving viral threats.

Results

Efficient degradation of Fc fusion proteins by modified Trim-Away system

TRIM21 is an E3 ubiquitin ligase that binds antibodies and mediates their degradation via the proteasome. However, variability in antibody binding restricts its efficacy, especially for rapidly mutating targets such as viral proteins. To address this limitation, we engineered a receptor-Fc fusion protein, where the Fc domain is specifically recognized by TRIM21. Upon binding of TRIM21 to the Fc portion of the fusion protein, TRIM21 catalyzes the conjugation of ubiquitin chains to the target receptor, marking it for proteasomal degradation (Fig. 1A).

Fig. 1.

Design and functional verification of the modified Trim-Away system. (A) Degradation schematic diagram of the receptor-Fc fusion modified Trim-Away. (B) 293T cells co-transfected with pcDNA3.1-TRIM21 and pcDNA3.1-eGFP (Control) or pcDNA3.1-eGFP-Fc (Trim-Away) were photographed by fluorescence microscope at 24 and 48 h post transfection. The percentage of eGFP fluorescent area was quantified. Data are presented as mean ± SD from three independent experiments (n = 3). Statistical significance was determined by unpaired t-test. ** P < 0.01, *** P < 0.001; ns, not significant

To validate the ability of the Trim-Away system to degrade Fc fusion proteins, we fused enhanced green fluorescent protein (eGFP) to the N-terminus of the Fc domain. Degradation of eGFP-Fc was assessed in HEK-293T cells via fluorescence microscopy at 24 and 48 h post-co-transfection with pcDNA3.1-TRIM21 and pcDNA3.1-eGFP-Fc (Trim-Away group). Compared to the control group (co-transfection with pcDNA3.1-TRIM21 and pcDNA3.1-eGFP), the Trim-Away-treated group exhibited a significant reduction in eGFP fluorescence intensity at both 24 and 48 h post-transfection. The time-dependent decrease in fluorescence intensity in the treated group confirmed efficient degradation of the eGFP-Fc fusion protein, thereby verifying the feasibility and efficacy of the engineered Trim-Away system (Fig. 1B). These results demonstrate that engineered Fc fusion proteins can be efficiently degraded by the Trim-Away system, supporting the potential of receptor-Fc fusion strategies for application in targeted protein degradation.

Specific degradation of AAV receptor and capsid proteins by Trim-Away

To confirm the specificity and efficacy of the Trim-Away system, we first evaluated its capacity to degrade the Fc-fused extracellular domain of adeno-associated virus receptor (AAVR-Fc). HEK-293T cells were co-transfected with a TRIM21-encoding plasmid and either an AAVR-Fc-encoding plasmid or an unfused AAVR-encoding plasmid. Co-expression of TRIM21 resulted in a significant reduction in AAVR-Fc protein levels, whereas it exerted no significant effect on the unfused AAVR (Fig. 2A&B). This result confirms the Fc-dependent specificity of TRIM21-mediated degradation.

Fig. 2.

Specific degradation of AAV receptor and capsid proteins. (A, B) Western blot (A) and relative quantitative analysis (B) evaluating TRIM21-mediated degradation of AAVR-Fc or AAVR. (C-E) Western blot (C) and corresponding quantifications of receptor (D) and AAV capsid proteins (VPs) (E) under different co-transfection conditions (Group 1: pAAV-R2C9 encoding VPs alone; Group 2: +AAVR; Group 3: +AAVR-Fc; Group 4: +AAVR+TRIM21; Group 5: +AAVR-Fc+TRIM21). (F, G) Western blot (F) and quantification (G) of intracellular VPs following AAV transduction in cells pre-transfected with TRIM21 and either AAVR (Control) or AAVR-Fc (Trim-Away). Data are mean ± SD (n = 3). Unpaired t-test: ** P < 0.01, *** P < 0.001; ns, not significant

Next, we investigated whether this system could mediate the degradation of AAV viral capsid proteins (VPs). Plasmids encoding VPs were co-transfected with various combinations of AAVR, AAVR-Fc, and TRIM21. The protein levels of VP1, VP2, and VP3 were markedly reduced exclusively in the presence of both AAVR-Fc and TRIM21 (Fig. 2C-E, Group 5). The omission of either TRIM21 (Groups 2 and 3) or the Fc domain (Group 4) resulted in the failure of target degradation, maintaining the expression of VPs at baseline levels. This confirms the strict dual dependency of the degradation mechanism (Fig. 2C-E).

To further confirm the efficacy of the system against intact viral particles, we introduced the Trim-Away system into cells following AAV transduction. We then performed western blot analysis to detect VP protein expression, and the results showed that intracellular VP protein levels in the Trim-Away treated group were also significantly reduced compared to the control group (Fig. 2F&G).

Taken together, these data demonstrate that the AAVR-Fc-guided Trim-Away system can specifically recognize and degrade the target fusion protein. It can also effectively degrade AAV capsid proteins, both when co-expressed with the system and when derived from intact viral particles.

Trim-Away restricts AAV yield via capsid degradation

Having confirmed that the Trim-Away system efficiently degrades AAV capsid proteins, we next investigated the direct impact of this degradation on new recombinant AAV production and viral transduction. Recombinant AAV carrying the eGFP gene (rAAV-eGFP) was produced in HEK-293T cells via a triple-plasmid co-transfection system comprising pAAV-eGFP, pAAV-helper, and pAAV-R2C9. During rAAV production, cells in the experimental group were co-transfected with Trim-Away system plasmids (encoding TRIM21 and AAVR-Fc), while cells in the control group were co-transfected with plasmids encoding TRIM21 and AAVR. This control system cannot target capsid proteins due to the absence of an Fc tag.

Subsequently, rAAV particles were harvested from cell lysates and analyzed. Results showed that rAAV particles produced in the presence of the functional Trim-Away system contained significantly lower levels of VPs compared to those from the control group (Fig. 3A), suggesting that the intracellular degradation of capsid proteins markedly reduces the pool of structural protein available for packaging. To evaluate how this affects the yield of functional virus, fresh HEK-293T cells were transduced with rAAV produced under both experimental conditions. At 48 and 72 h post-transduction, fluorescence microscopy revealed a significant reduction in the number and fluorescence intensity of eGFP-positive cells transduced with rAAV from the Trim-Away experimental group (Fig. 3B).

Fig. 3.

Trim-Away Impairs the Production of Recombinant AAV Particles. (A) Western blot analysis of VP levels in recombinant AAV produced via triple-plasmid transfection (pAAV-eGFP, pAAV-helper, and pAAV-R2C9) in cells co-transfected with plasmids encoding TRIM21 and AAVR-Fc versus TRIM21 and AAVR. (B) Representative fluorescence microscopy images and corresponding quantitative analysis of cells transduced with recombinant AAV. (C) Cells expressing eGFP following transduction with recombinant AAV were counted by flow cytometry. Control: cells co-transfected with plasmids encoding TRIM21 and AAVR, Trim-Away: cells co-transfected with plasmids encoding TRIM21 and AAVR-Fc. Data are presented as mean ± SD from three independent experiments (n = 3). Statistical significance was determined by unpaired t-test. * P < 0.05, *** P < 0.001; ns, not significant

Furthermore, we determined the percentage of eGFP-positive cells via flow cytometry. The results showed that this percentage decreased significantly from 50.7% in the control group to 22.8% in the experimental group, indicating a marked reduction in viral transduction. (Fig. 3C). Collectively, these findings demonstrate that the Trim-Away system effectively degrades essential viral building blocks, leading to a profound reduction in the overall yield of functional recombinant AAV particles.

Trim-Away mediated inhibition of AAV transduction in vivo

To further verify the antiviral efficacy of the Trim-Away system in a complex physiological environment, we established a mouse model using recombinant AAV vectors encoding a luciferase reporter gene. This model allowed for dynamic monitoring of viral transduction efficiency via living bioluminescence imaging. To achieve continuous target degradation in vivo, a self-amplifying replicon system based on the Semliki Forest virus (pSFV) was employed. Unlike conventional non-replicating vectors, this replicon enables sustained and high-level expression [24] of Trim-Away components to more efficiently replenish the TRIM21 and receptor-Fc molecules consumed during viral co-degradation, thereby enhancing the interference efficacy of the Trim-Away system against AAV transduction. Experimental mice were randomly assigned to a negative control (NC) group or a Trim-Away treatment group. Bioluminescence imaging was performed serially on days 3, 5, 7, 9, 11, and 13 post-AAV injection (PAI). In the NC group, robust bioluminescence was detectable as early as day 3, with high signal intensity and wide distribution indicating efficient viral entry and transgene expression. This robust luminescent signal was sustained throughout the 13-day observation period.

In contrast, mice receiving Trim-Away treatment exhibited markedly suppressed luminescence. Signals in this group were predominantly at background levels (visualized as blue), indicating that the transduction was effectively blocked (Fig. 4A). Quantitative analysis confirmed these visual observations. When total flux was tracked over time, the NC group maintained consistently high values (fluctuating around 10⁸ p/s). In comparison, the Trim-Away group exhibited values approximately two orders of magnitude lower, typically ranging from 10⁶ to 10⁷ p/s (Fig. 4B). A similar trend was observed for average radiance measurements (Fig. S1A). The NC group maintained average radiance levels above 10⁵ p/s/cm²/sr, whereas the Trim-Away group exhibited minimal activity mostly below 10⁴ p/s/cm²/sr. Importantly, statistical analysis revealed that these reductions were highly significant at all examined time points (P < 0.001). The suppression of the luciferase signal in the treated group was not transient but remained stable throughout the entire experiment. The slight increase in signal intensity observed in the treatment group after day 9 PAI may be attributed to the exhaustion of the Trim-Away system. These in vivo findings are consistent with our in vitro data, providing compelling evidence that the Trim-Away system significantly impairs AAV transduction in living organisms.

Fig. 4.

In vivo inhibition of AAV transduction via a self-amplifying replicon-based Trim-Away system. (A) Representative bioluminescence images of mice at indicated days post-AAV injection (PAI) with luciferase-encoding AAV9. Prior to the AAV challenge, mice were intramuscularly administered PBS (NC group) or a plasmid mixture of pSFV-TRIM21 and pSFV-AAVR-Fc (Trim-Away group). (B) Quantitative analysis of whole-body Total Flux over a 13-day period. (C) Body weight changes tracked over 21 days PAI. Data are mean ± SD (n = 3). *** P < 0.001 compared to the NC group

Systemic safety and potential cytotoxicity were evaluated following the administration of the engineered degradation components. Body weights of mice were recorded continuously over a 21-day period PAI, displaying stable growth trends comparable to the control group (Fig. 4C). Major organs including the heart, liver, spleen, lungs, and kidneys were harvested for hematoxylin and eosin staining. There are no signs of pathological necrosis or inflammatory infiltration (Fig. S1B). The construct design lacks secretory signal peptides, which is expected to confine the decoy receptors to the cytosol. This spatial restriction aims to circumvent unintended binding with native extracellular ligands, contributing to the safety profile of this targeted protein degradation platform.

Trim-Away exhibits effective degradation to SARS-CoV-2 spike protein in co-transfected 293T cells

To evaluate the potential of the Trim-Away system as a therapeutic strategy against SARS-CoV-2, we first confirmed the specificity of TRIM21-mediated degradation. HEK-293T cells were co-transfected with a TRIM21-encoding plasmid and either a plasmid encoding ACE2-Fc or untagged ACE2. As shown in Fig. 5A, co-expression of TRIM21 led to a substantial reduction in ACE2-Fc protein levels. In contrast, no significant decrease in protein abundance was observed for untagged ACE2 (Fig. 5B). This confirms that the degradation process is strictly Fc-dependent, thereby sparing endogenous ACE2 from unintended targeting. Next, we constructed a self-amplifying plasmid system for expressing Trim-Away components and tested its ability to degrade the viral antigen. Western blot analysis showed that in cells expressing the complete self-amplifying system, the levels of both full-length SARS-CoV-2 spike protein and its S1 subunit were markedly lower compared to control cells (Fig. 5C), confirming the efficacy of the self-amplifying replicon-based expression strategy.

Fig. 5.

Specific SFV replicon-based Trim-Away degradation of SARS-CoV-2 Spike. (A-B) TRIM21 degrades Fc-tagged ACE2 but not untagged ACE2. (C) Validation of Spike degradation via the SFV replicon-based system. (D) Effective degradation of wild-type and Omicron Spike variants. (E) Cells were co-transfected with 2 µg of Spike-encoding plasmid and decreasing amounts of pSFV or pcDNA3.1(+) Trim-Away plasmids (2 µg to 0.25 µg in total for TRIM21 and ACE2-Fc). Spike and TRIM21 expression is shown in the immunoblot. TRIM21 levels were used to indicate the system input. Relative Spike expression was normalized to β-actin. Data are presented as mean ± SD from three independent experiments (n = 3). Statistical significance was determined by one-way ANOVA followed by Dunnett’s multiple comparisons test (*** P < 0.001). Sp: SFV replicon vector

Given the rapid evolution of the virus, we further investigated whether the system retains efficacy against the Omicron variant. Results demonstrated that the Trim-Away system degraded the Omicron spike protein with efficiency comparable to that of the wild-type spike protein (Fig. 5D). This indicates that by targeting the conserved ACE2-Spike interaction, our strategy confers broad-spectrum activity against emerging SARS-CoV-2 variants.

To evaluate the dose dependency of the self-amplifying replicon, a parallel comparative assay was performed using the conventional pcDNA3.1(+) expression vector as a control (Fig. 5E). Plasmids encoding the Trim-Away components based on either the pSFV or pcDNA3.1(+) system were transfected at progressively decreasing dosages from 2 µg to 0.25 µg. Effective target degradation was observed across all tested dosages in both experimental groups. The degradation efficiency was comparable between the two systems at the 2 µg dosage. As the transfection amount of the Trim-Away plasmids decreased, the disparity in degradation efficacy between the pcDNA3.1(+) system and the pSFV system was observed to widen gradually. Even at the lowest tested dosage, the pSFV replicon still exhibited excellent degradation efficacy.

Discussion

Current antiviral strategies, ranging from small-molecule inhibitors to neutralizing antibodies and RNA interference, often face bottlenecks such as off-target effects and high development costs [25]. Targeted protein degradation offers a distinct mechanism to overcome these limitations. In this study, we established a modified Trim-Away system based on self-amplifying replicon technology. The platform demonstrates robust efficacy in degrading diverse viral proteins, intercepting the viral life cycle at the packaging and transduction stages.

The fundamental mechanism of Trim-Away relies on the specific recruitment of the cellular degradation machinery [16]. TRIM21 functions as a cytosolic antibody receptor that binds the Fc domain with high affinity, activating its E3 ubiquitin ligase activity [26]. This triggers the proteasomal destruction of the bound target. Our initial validation using eGFP-Fc fusion proteins confirmed that this machinery functions efficiently within the intracellular environment. Unlike extracellular neutralization, this approach physically eliminates the target protein. The successful degradation of eGFP-Fc provided the molecular basis for extending this strategy to pathogenic viral components. The apparent slower degradation kinetics in our plasmid-based system reflect a dynamic balance between continuous target protein expression and targeted degradation, differing from the rapid clearance of preexisting proteins observed with direct intracellular delivery [16]. However, the plasmid approach provides practical advantages in preparation cost and broader application scalability.

Building on this mechanism, we engineered an AAVR-Fc fusion to target adeno-associated virus (AAV) capsid proteins. This design bypasses the need for antigen-specific antibodies. By replacing the variable region of an antibody with the natural receptor AAVR [27], we ensured high-affinity binding to the viral capsid. Because AAV capsid proteins VP1, VP2, and VP3 share the conserved VP3 structural core recognized by AAVR [28, 29], the receptor decoy effectively mediates their simultaneous degradation. Our results showed that this receptor-decoy strategy degraded synthesized viral proteins and intercepted incoming virus particles at the post-entry stage. Without secretory signal sequences, the physical contact and subsequent degradation complex formation are expected to occur primarily in the cytosol [30]. Upon binding to the Fc domain of the decoy receptor, the E3 ubiquitin ligase activity of TRIM21 is activated. This leads to the rapid ubiquitination of the target complex and its subsequent proteasomal destruction. Consequently, the efficient cytosolic clearance of these structural components limits the availability of essential substrates for the viral packaging process, leading to a substantial reduction in the total output of recombinant AAV. The slightly reduced degradation efficiency of receptor-Fc in the presence of bulky viral targets likely reflects steric hindrance and proteasomal kinetic barriers. However, this process remains mechanically coupled to the forced unfolding and efficient co-degradation of the attached viral proteins via TRIM21 [30, 31].

To address the delivery challenges associated with protein-based therapies [32], we employed a pSFV self-amplifying replicon vector. This vector drives high-level, sustained expression of the Trim-Away components. In vivo validation in a mouse model confirmed the potency of this approach. Against a rigorous systemic challenge of 1 × 10¹¹ GC AAV9, it is speculated that the widespread bioluminescence and sustained expression kinetics observed in the mock-treated mice are likely attributable to the well-documented properties of the AAV9 vector, including systemic leakage and long-term episomal stability [33–36]. In contrast, the pSFV-delivered Trim-Away system achieved a 100-fold reduction in viral transduction. High statistical significance (P < 0.001) persisted throughout the 13-day observation period. The sustained suppression of luciferase expression in the mouse model underscores the advantage of the replicon system over transient gene or protein delivery, which often suffers from rapid metabolic turnover. A slight signal rebound was observed in the treatment group after day 9. This phenomenon is tied to the inherently slow and asynchronous uncoating process of AAV [35]. The pSFV-driven Trim-Away system acts transiently, exerting robust initial viral clearance before its components naturally diminish. As the intracellular degradation pressure wanes, a minor fraction of surviving AAV particles eventually complete uncoating. Transcription from these persistent residual genomes accounts for the late-stage cumulative bioluminescence.

Expanding the scope of this platform, we demonstrated its versatility and potency against SARS-CoV-2. Using an ACE2-Fc decoy, the system effectively degraded the spike protein of both the wild-type virus and the Omicron variant. Receptor-based targeting offers a critical advantage. The conservation of the receptor-binding domain ensures that the virus has limited capacity to escape through mutation [37]. Notably, this degradation was achieved with extremely low transfection dosages. The self-amplifying nature of the pSFV vector allowed for significant protein reduction even at minimal plasmid inputs. This high potency suggests that the catalytic amplification of the plasmid replicon can compensate for low delivery efficiency, a common hurdle in gene therapy.

However, this study has several limitations. The cytosolic location of TRIM21 and the receptor complex limit their access to proteins like Spike that transit rapidly through the secretory pathway, meaning degradation must occur during a narrow window of translocation, a phase where nascent viral proteins remain accessible to cytosolic ubiquitination machineries [38, 39]. A potential improvement involves anchoring TRIM21 to the cytosolic face of the ER membrane [40, 41]. This strategy would preserve its redox-sensitive RING domain while maximizing spatial proximity to membrane-bound targets [42, 43]. In addition, while our results with AAV and spike proteins are promising, we utilized surrogate markers and viral components rather than live, replicating viruses. AAV is not a human pathogen. It was used strictly as a safe surrogate model to establish the proof-of-concept for the targeted degradation mechanism. The system must be tested against live, highly pathogenic viruses to fully confirm its clinical and practical value. Additionally, the exact in vivo biodistribution of the pSFV and AAV vectors was not directly quantified. Mapping the vector dissemination in future studies would help clarify the precise extent of AAV systemic leakage and the tissue-level co-localization dynamics of the two vectors.

Although the pSFV plasmid provides a robust proof-of-concept, its inherent transient expression shapes its specific clinical utility. This short-term dynamic limits general long-term prophylaxis but aligns effectively with acute clinical scenarios. The system can provide pre-exposure prophylaxis for high-risk populations facing imminent viral exposure, establishing a temporary but robust defensive barrier. Additionally, early post-exposure administration could acutely intercept invading capsids before widespread replication occurs. To realize these clinical applications, transitioning the architecture into a self-amplifying RNA lipid nanoparticle (SaRNA-LNP) platform represents a feasible translational direction [44]. SaRNA-LNP systems circumvent the safety concerns of DNA vectors and facilitate precisely modulated tissue tropism. Specific targeted delivery can be achieved by engineering the LNPs with targeting ligands or optimizing their lipid composition [45]. When combined with localized administration routes, such as aerosolized inhalation [46], this approach restricts therapeutic expression strictly to the primary infection site, mitigating the risk of systemic toxicity and establishing a versatile interference platform against emerging viral threats.

Materials and methods

Plasmid cloning and construction

The sequence of TRIM21 was derived from the amino acid sequence of the human TRIM21 protein (NCBI protein accession number NP_003132), with a Flag tag appended to its N-terminus, and was codon-optimized for expression in human cells. This construct was cloned into the pcDNA3.1(+) vector and the self-amplifying vector pSFV via homologous recombination. The pSFV self-amplifying replicon vector was constructed in our laboratory. The sequences encoding the non-structural proteins (RNA-dependent RNA polymerase) of the Semliki Forest virus (SFV, Accession P08411) along with other essential viral elements were inserted into the pVAX1 plasmid backbone maintained in the laboratory. The nucleotide sequence that encodes the extracellular domain of AAVR (NCBI protein accession number NP_079150) was derived from the KIAA0319L gene. This specific segment was subsequently codon-optimized in order to ensure efficient expression in human cells. The AAVR-Fc fusion protein was constructed by overlap extension PCR (OE-PCR), which linked the AAVR sequence to the Fc fragment of IgG (NCBI protein accession number 4CDH_A). The extracellular domain of ACE2 (NCBI protein accession number NP_001358344) sequence, encoding the human cell surface receptor that binds to the SARS-CoV-2 spike protein, was similarly codon-optimized. The Fc-tagged SARS-CoV-2 spike protein receptor (ACE2-Fc) was generated by ligating the optimized ACE2 sequence to an Fc fragment using OE-PCR. These constructs were subsequently cloned into pcDNA3.1(+) and pSFV vectors through homologous recombination with a His tag. The coding sequences for the AAV capsid proteins (NCBI protein accession number NP_049542) and the full-length wild-type and Omicron variant spike proteins (NCBI protein accession numbers QIH45093 and UHS85575) were all codon-optimized for human cells and cloned into the pcDNA3.1(+) vector via homologous recombination. All required gene sequences were synthesized by Sangon Biotech. The AAV-packaging plasmids (pAAV-eGFP, pAAV-helper, and pAAV-R2C9) were purchased from Tianjin Wuyuan Biology and maintained in our laboratory.

Cell lines and transfection

Human HEK-293T cells (CRL-3216, ATCC) and African green monkey kidney Vero E6 cells (CRL-1586, ATCC) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Gibco), 100 µg/mL streptomycin, and 100 IU/mL penicillin (Gibco, Grand Island, NY, USA). For transfection, cells were prepared by seeding them in 6-well plates 24 h before transfection to reach 70–80% confluency. To maintain uniform transfection efficiency, the total mass of transfected DNA was kept constant across all groups by supplementing with an empty vector. HEK-293T cells were transfected using poly-ethyleneimine (PEI) at a weight ratio of PEI to plasmid DNA of 2:1. PEI-DNA complexes were incubated at room temperature for 15 min before being added to the cells. The culture medium was replaced with fresh DMEM 4–6 h post-transfection. Vero E6 cells were transfected using Lipofectamine Nucleic Acid Transfection Reagent (Yeasen, 40802E S03). The transfection procedure was similar to that for PEI-mediated transfection of HEK-293T cells. The transfection reagent-DNA complexes were prepared, added to the cells, and followed by medium replacement 4–6 h later. Cells were harvested 48 h post-transfection for lysis. For protein extraction, cells were first washed with cold phosphate-buffered saline (PBS) and then lysed in RIPA buffer on ice for 30 min. Cell lysates were centrifuged at 12,000 rpm for 10 min at 4 °C, and the supernatant containing total proteins was collected for subsequent analysis.

Western blot

Protein samples were mixed with 5× reducing loading buffer prior to analysis. Western blot (WB) analysis was performed following standard protocols with minor modifications. Briefly, equal amounts of protein were separated by SDS-PAGE and transferred onto polyvinylidene difluoride (PVDF) membranes. Membranes were blocked with 5% non-fat milk in Tris-buffered saline with Tween 20 (TBST) for 1 h at room temperature, followed by incubation with primary antibodies overnight at 4 °C.

Primary antibodies used were as follows: anti-Flag (ProteinTech, Cat. No. 20543-1-AP), 6×His-HRP-conjugated (Cat No. HRP-66005), β-actin-HRP conjugate (Pro-teinTech, Cat. No. HRP-60008), anti-SARS-CoV-2 Spike (Cell Signaling Technology, Cat. No. 99423 S), anti-human IgG Fc (Cell Signaling Technology, Cat. No. 32935 S), anti-GFP (ProteinTech, Cat. No. 50430-2-AP), and anti-AAV VP1/VP2/VP3 (Progen, Cat. No. 61084). After three washes with TBST, membranes were incubated with the appropriate secondary antibody [Goat Anti-Rabbit IgG(H + L)-HRP; ProteinTech, Cat. No. SA00001-2] for 1 h at room temperature.

For detection, membranes were washed again and visualized using an enhanced chemiluminescence (ECL) substrate, followed by imaging with a chemiluminescent imaging system. Band intensities were quantified using ImageJ software to verify protein expression; relative expression was calculated as the ratio of target protein to internal control and normalized to the control group.

Detection of VPs in AAV transduced cells

Endotoxin-free plasmids (pcDNA3.1-TRIM21, pcDNA3.1-AAVR-Fc, and pcDNA3.1-AAVR) were prepared for co-transfection. HEK-293T cells were seeded in 6-well plates at 50% confluency. A mixture containing 1.5 µg each of pcDNA3.1-TRIM21 and either pcDNA3.1-AAVR-Fc or pcDNA3.1-AAVR was co-transfected into HEK-293T cells via the PEI transfection method.

At 4–6 h post-transfection, 1 µL of AAV viral stock was added to 2 mL of complete medium supplemented with 2 µL of 1000× polybrene. The mixture was thoroughly mixed and used to replace the culture medium in each well to initiate viral transduction. At 24 h post-transduction, the medium was replaced with fresh complete medium to maintain optimal cell growth and transduction conditions. Cells were further incubated for 48 h, then harvested and lysed in RIPA buffer. Protein extracts were subjected to WB analysis to detect viral capsid proteins (VPs).

The expression level of VPs determined by WB was used to assess the efficiency of TRIM21-mediated degradation, providing insights into the Trim-Away degradation mechanism.

AAV production and interference

To assemble AAV, HEK-293T cells were co-transfected with pAAV-eGFP, pAAV-helper, and pAAV-R2C9 at a molar ratio of 1:2:1. For the control group, cells were co-transfected with pcDNA3.1-TRIM21 and pcDNA3.1-AAVR; for the experimental group, cells were co-transfected with pcDNA3.1-TRIM21 and pcDNA3.1-AAVR-Fc. All transfections were performed using the PEI method, with cells seeded to 50% confluency in 6-well plates prior to transfection.

At 48 h post-transfection, cells from both groups were harvested for protein extraction. Cell lysates were subjected to WB analysis to detect VPs, which served as indicators of AAV production yield and the efficacy of Trim-Away-mediated degradation. Primary antibodies specific to AAV VPs were used, followed by appropriate HRP-conjugated secondary antibodies. Relative VP levels between the control and experimental groups were analyzed to evaluate TRIM21-mediated degradation.

Recombinant AAV transduction and detection

HEK-293T cells were transduced with recombinant AAV either pre-treated with the Trim-Away system or left untreated. Producer cells from both the treatment and control groups underwent identical experimental procedures. Reagent volumes were maintained strictly equal across all steps. To accurately assess the differential yield of functional virions, equal volumes of the resulting recombinant AAV lysates were used to transduce fresh HEK-293T cells.

At 48 h and 72 h post-transduction, cells were examined using a fluorescence microscope to observe and capture eGFP fluorescence, enabling visual confirmation of successful transduction and eGFP reporter gene expression. For the quantitative analysis of fluorescence microscopy images, the percentage of eGFP fluorescent area and the integrated density were determined using ImageJ software. A consistent threshold was applied to segment the fluorescent signal from the background prior to measurement. Additionally, flow cytometry was employed to determine the proportion of eGFP-positive cells.

Flow cytometry steps included: washing cells with cold PBS, trypsinization, resuspension in PBS, and filtration through a 40 μm strainer to obtain a single-cell suspension. Analysis was performed using a flow cytometer equipped with a 488 nm laser for eGFP detection. Data were collected and analyzed to determine the percentage of eGFP-positive cells, indicating recombinant AAV transduction efficiency.

Additionally, WB analysis was conducted to detect and quantify eGFP expression levels in transduced cells. Cells were lysed to extract total protein for WB, and eGFP expression levels were compared between the Trim-Away-treated and control groups. Flow cytometric data were analyzed using FlowJo software.

In vivo delivery and bioluminescence imaging

To evaluate viral interference in a physiological context, male BALB/c mice were used. On day 0, Trim-Away components were delivered by complexing 2.5 µg each of self-amplifying plasmids (pSFV-AAVR-Fc and pSFV-TRIM21) with TransPi™ In Vivo Lung Delivery Reagent (Polygene Bio, Cat. No. 1000IVL), following the manufacturer’s instructions for efficient plasmid delivery. This reagent exhibits high lung-targeting efficiency via tail vein injection, but mediates localized in situ delivery when administered intramuscularly. The formulated plasmid complexes were administered via localized intramuscular injection into the hind limb muscles.

The control group received an equivalent volume of Dulbecco’s Phosphate-Buffered Saline (DPBS). At 24 h post-delivery, the recombinant AAV9 vectors (1 × 10¹¹ genome copies per mouse) were injected into the same site. This localized injection strategy ensured the spatial co-localization of the Trim-Away components and the AAV9 particles.

Bioluminescence imaging was initiated on day 3 post-AAV injection to monitor viral transduction kinetics. Prior to each imaging session, mice were anesthetized with isoflurane and maintained under anesthesia throughout the procedure. The luciferase substrate, D-luciferin potassium salt (Yeasen Biotech, Cat. No. 40902ES03), was prepared and administered to mice via intraperitoneal injection.

Following a 10-minute incubation to allow systemic substrate distribution, bioluminescent signals were captured using the PerkinElmer Lumina XRMS Series III imaging system. Acquisition parameters (including exposure time and binning) were optimized to ensure data accuracy without signal saturation. All images were processed using Living Image software (Version 4.4). Quantitative analysis was performed by defining consistent regions of interest (ROI) encompassing the entire body of each mouse, and data were expressed as total flux (p/s) and average radiance (p/s/cm²/sr).

Conclusions

Our study establishes the self-amplifying replicon-based Trim-Away system as a modular and potent antiviral platform. By combining the specificity of receptor-Fc fusions with the sustained expression of self-amplifying replicon, we achieved effective degradation of viral targets in both in vitro and in vivo models. This transient degradation system can be applied to pre-exposure prophylaxis and acute post-exposure interventions. Future investigations should focus on the long-term safety profile of sustained TRIM21 expression and the potential application of this technology to other high-risk pathogens where traditional therapeutics have failed.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AAV

Adeno-associated virus

AAVR

Adeno-associated virus receptor

ACE2

Angiotensin-converting enzyme 2

ECD

Extracellular domain

GC

Genome copies

PROTACs

Proteolysis-targeting chimeras

RBD

Receptor-binding domain

RNAi

RNA interference

rAAV

Recombinant adeno-associated virus

SARS-CoV-2

Severe acute respiratory syndrome coronavirus 2

SFV

Semliki Forest virus

TRIM21

Tripartite motif-containing protein 21

UPS

Ubiquitin-proteasome system

VPs

Viral capsid proteins

Author contributions

Conceptualization, Shoutao Zhang; methodology, Yiming Han, Zhen Tian and Shoutao Zhang; software, Zhen Tian, Xiaodi Wang and Pengbo Wang; validation, Qisheng Dong; formal analysis, Qisheng Dong, Xiaodi Wang and Pengbo Wang; investigation, Yiming Han, Qisheng Dong, Zhen Tian, Xiaodi Wang and Pengbo Wang; resources, Shoutao Zhang and Quanyong Liu; data curation, Yiming Han; writing-original draft preparation, Yiming Han and Qisheng Dong; writing-review and editing, Yiming Han, Quanyong Liu, Yangxue Liu, Qi Guo and Shoutao Zhang; visualization, Yiming Han and Quanyong Liu; supervision, Shoutao Zhang; project administration, Yiming Han, Yangxue Liu, Qi Guo and Shoutao Zhang; funding acquisition, Shoutao Zhang.

Funding

This research was supported by the Natural Science Foundation of Henan Province of China (Grant No.232300421117 and No.232102311149), Science and Technology Research and Development Joint Fund Project of Henan Province (Grant No.245101610002 and No.245101610009) and “Open Competition” Project for Key Technology Research in Major Industries of Henan Province (Grant No. 251000310100).

Data availability

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.

Declarations

Ethics Approval

All animal procedures were performed in accordance with national guidelines and approved by the Ethics Committee of Zhengzhou University (ethics approval number: ZZUIRB2025-103).

Informed consent

Not applicable.

Competing interests

The authors declare no competing interests.