TRIM21 belongs to the TRIM protein family and consists of a classic N-terminal RBCC motif and C-terminal PRY-SPRY domain. Among them, the RBCC motif can target protein to the proteasome via its E3 ubiquitin ligase activity and PRY-SPRY domain mediates immunoglobulin Fc fragment binding in a pincer-like interaction [32,33,34] (Figure 1a). To generate a new construct with both protein degradation activity and antibody-binding specificity, we fused the RBCC motif of TRIM21 to a nanobody that can bind to a targeted antigen. The corresponding protein was designated as TRIMbody, and the system was described as TRIMbody-Away. As shown in Figure 1b, TRIMbody is a multi-domain protein consisting of an N-terminal RING domain with E3 ubiquitin ligase activity, a B-box domain, a coiled-coil dimerization domain, and a C-terminal nanobody fragment which specifically recognizes intracellular proteins of interest.

To test whether TRIMbody could mediate degradation of the target protein, we chose EGFP protein as a proof-of-concept substrate. The previously reported nanobodies against EGFP, LaG16 and LaG2 were selected [35,36] and designed for bispecific anti-EGFP (αEGFP) nanobody. Then, an αEGFP TRIMbody was generated that comprised of an RBCC domain and bispecific αEGFP nanobody (Figure 1c). αEGFP TRIMbody could also be expressed as a soluble form in E. coli with the help of Lipoyl, as well as TRIM21. By analysis of SDS-PAGE, purified TRIM21 and αEGFP TRIMbody proteins exhibited major bands with molecular weights of 66 and 77 kDa, respectively (Figure 1d). Next, we examined the oligomeric state of the αEGFP TRIMbody by SEC analysis. The elute volume for αEGFP TRIMbody was determined at 10.3 mL, and 10.4 mL for purified TRIM21 protein, confirming that the αEGFP TRIMbody and TRIM21 have been correctly expressed (Figure 1e). Further, we measured the binding ability of αEGFP TRIMbody to EGFP by ELISA. Furthermore, a single-chain fragment of G12 antibody that recognizes HBsAg of HBV was used to generated αHBsAg TRIMbody and served as a negative control. We observed that the bispecific anti-EGFP nanobody (LaG16-LaG2) has a EC₅₀ value of 2.40 nM in EGFP binding, while the αEGFP TRIMbody showed more evident binding activity to EGFP, with an EC₅₀ value of 0.63 nM (Figure 1f), which could be due to dimerization of TRIM21 leading to enhanced binding avidity. Remarkably, there was no obvious EGFP binding activity for negative controls.

Based on the EGFP binding by αEGFP TRIMbody, we next investigated its target protein degradation function in EGFP-expressing cells. To obtain the stably expressing EGFP cell lines, we performed live cell FACS sorting of 293T cells infected with lentivirus to isolate EGFP-positive cells, and the sorted cells were kept growing and still showed that 99.9% of the population are EGFP-positive at least for five passages, suggesting that expression of EGFP in 293T cells is relatively stable (Figure 2a). Expressing vectors with αEGFP TRIMbody were transiently transfected into EGFP stably expressing 293T cells, and then EGFP fluorescence intensity of cells at the indicated time was determined and analyzed via flow cytometry analysis. Meanwhile, vector, RBCC only, TRIM21 only, αEGFP nanobody, and αHBsAg TRIMbody were used as the controls. We observed that transfection of αEGFP TRIMbody resulted in significant decrease in EGFP fluorescence at 24 and 48 h in a time-dependent manner, but the fluorescence has no significant decrease at 72 h compared with 48 h, which may result from dilution loss of transfected αEGFP TRIMbody plasmid due to transient transfection, leading to insufficient expression of αEGFP TRIMbody protein in cells, or compensatory supply of EGFP protein by constitutive expression of EGFP. In contrast, transfection of RBCC, TRIM21, and αEGFP nanobody showed no decrease of the fluorescence (Figure 2b), suggesting that the RBCC domain and αEGFP nanobody failed to trigger intracellular EGFP degradation. Moreover, we found that αHBsAg TRIMbody, which was confirmed to have no EGFP binding ability in ELISA (Figure 1f), also did not induce intracellular EGFP degradation in EGFP-expressing 293T cells (Figure 2b). The results showed that the combination of the RBCC domain and specific antibody are necessary for degradation of intracellular EGFP.

Additionally, to avoid instability and side effects of transient transfection, the inducible expression of αEGFP TRIMbody by Doxycycline (Dox) treatment in stably expressing EGFP cells was constructed using the Tet-On-3G system and further degradation of EGFP was measured by confocal laser scanning microscope or flow cytometry. This system was sensitive to Dox treatment and the optimal Dox concentration was 10 μg/mL. Fluorescence images of cells showed that EGFP was universal and stably expressed and mainly located in the cytoplasm. Notably, αEGFP TRIMbody was not expressed in cells under no Dox treatment, while addition of Dox caused αEGFP TRIMbody expression that co-localized with EGFP (Figure 2e), suggesting that αEGFP TRIMbody can be successfully induced and bound to EGFP via the specificity of the anti-EGFP nanobody. Following Dox addition, EGFP aggregated quickly in cytosol with a spotty pattern at 24 h. In contrast, no significant difference of EGFP was observed without Dox treatment (Figure 2c). Moreover, the quantitative EGFP fluorescence of cells after Dox induction was measured to assess the efficacy of induced αEGFP TRIMbody for EGFP degradation. Cells were treated with Dox and then fixed in 4% paraformaldehyde, and the EGFP fluorescence was analyzed via flow cytometry. The EGFP fluorescence was decreased by 25% at 24 h and had a 67% reduction after 48 and 72 h (Figure 2d). Next, we examined the EGFP and αEGFP TRIMbody protein level after Dox treatment by Western blot analysis and found that αEGFP TRIMbody proteins in cells were induced and kept stably expressed at 24, 48, and 72 h (Figure 2f,g). As for EGFP protein levels, no significant difference was observed in the absence of Dox. By contrast, substantial degradation of EGFP in Dox-treated groups was observed, with 40% reduction at 24 h and 60% decrease at 48 and 72 h after Dox addition (Figure 2g), indicating that intracellular EGFP degradation was subject to Dox induction and αEGFP TRIMbody expression.

Due to the fact that there was a significant decrease of EGFP protein level at 24 h induction by Dox, we expected that the αEGFP TRIMbody-mediated EGFP degradation process may happen earlier. Therefore, to further characterize the EGFP degradation dynamic pattern and visualize EGFP fluorescence puncta over time upon αEGFP TRIMbody expression, cells were exposed to Dox or vehicle for as long as 24 h and imaged every 6 h using a laser scanning confocal fluorescence microscope (Figure 3a). Meanwhile, we quantified the EGFP fluorescence intensity of cells by flow cytometry analysis (Figure 3b). The punctate EGFP was observed 6 h later by Dox induction and the EGFP fluorescence signal was significantly decreased after 12 h of Dox treatment, reaching maximal reduction of 30% of EGFP fluorescence intensity compared to control at 24 h (Figure 3a,b). Next, EGFP fluorescent puncta was examined from fluorescent images using ImageJ software and we observed it appear as early as 6 h in Dox-treated cells (Figure 3c). Additionally, the puncta area increased dramatically by at least 60-fold compared to controls at 12 and 24 h of Dox treatment (Figure 3d). We speculate that formation of EGFP puncta upon Dox treatment may be due to of EGFP in cytosol through αEGFP TRIMbody induction and recruitment.

Furthermore, we assessed the EGFP and αEGFP TRIMbody protein level by Western blot analysis and found that EGFP significantly reduced at 24 h of Dox treatment, but not for 6, 12, and 18 h of Dox induction (Figure 3e,f). Meanwhile, αEGFP TRIMbody was successfully induced at 6 h of Dox treatment, reaching to higher levels after adding Dox for 12 to 24 h. We noticed that the EGFP puncta appeared at 6 h of Dox treatment, accompanied with induction of αEGFP TRIMbody, and followed by the reduction of the EGFP fluorescence signal. Finally, the EGFP protein level was detected to decrease at 24 h of induction. These results implied that EGFP degradation happened with the formation of puncta, which may be through αEGFP TRIMbody-mediated intracellular EGFP aggregation, then followed by change of conformation of EGFP, leading to reduction of the fluorescence signal, finally destructed by the RBCC domain-involved degradation pathway. In addition, we performed live-cell imaging on cells and observed the dynamic process of degradation of intracellular EGFP during αEGFP TRIMbody induction.