Nanobody fusion enhances production of difficult-to-produce secretory proteins

Abstract

Secretory protein expression in mammalian cells is widely used in various fields, including biomedical research and biopharmaceutical production. However, achieving high-level expression of certain secretory proteins/peptides can be challenging. The naturally occurring N1 fragment of the prion protein is one of these difficult-to-produce secretory proteins, which hinders our understanding of its biological functions and limits its potential as a therapeutic molecule. To improve N1 production, we screened several well-folded protein domains and found that fusing N1 with a camelid nanobody (Nb) improved its translocation into the endoplasmic reticulum and significantly enhanced its secretion. Nb fusion does not alter the translocation mechanism, which remains dependent on the Sec61-Sec62-Sec63 complex. This approach also resulted in a significant increase in N1 production in the mouse brain using recombinant adeno-associated virus. Furthermore, fusing Nb to another unstructured protein, Shadoo (without glycosylphosphatidylinositol anchor), or a peptide hormone, somatostatin, also greatly increased their production, demonstrating the applicability of this approach to other proteins and peptides. The enhancement of N1 production is comparable or better than Fc fusion, and the effect is observed with all tested camelid Nb but not with a shark Nb and to a lesser extent with a human immunoglobulin heavy chain variable region. Importantly, the Nb in the fusion protein retained its antigen-binding capability, paving the way for the development of a dual-functional protein. Collectively, we present a novel strategy for enhancing the production of secretory proteins, which holds great promise in creating functional biological molecules for a wide range of applications.

Keywords: fusion protein; intracellular trafficking; intrinsically disordered protein; nanobody; prion; protein secretion.

Figure 1

Nb efficiently improves the production of unstructured N1 in cultured cells.A, schematic diagram of N1 fusion proteins. B, proteins in culture media were detected by Western blot analysis with 6D11 anti-PrP antibody. # Medium of N1-Nb expressing cells was diluted 375 times, whereas the other medium samples were undiluted. Asterisks indicate bands with expected molecular weight. C, fusion proteins in the soluble fraction of cell lysates were detected by Western blot analysis with 6D11 anti-PrP antibody. D, fusion proteins in the insoluble fraction of cell lysates were detected by Western blot analysis with 6D11 anti-PrP antibody. SP, signal peptide.

Figure 2

Nb enhances the translocation of N1 into endoplasmic reticulum.A, RT-PCR analysis of mRNA levels of N1^only and N1-Nb. Statistical analysis was performed with two-tailed unpaired t test (n = 12, p = 0.5442). B, immunofluorescence staining of N1^only, N1-Nb, and calnexin as indicated. The scale bar represents 5 μm. C, Pearson coefficients for colocalization of N1^only (n = 30) or N1-Nb (n = 9) with calnexin. Statistical analysis was performed using two-tailed unpaired t test (∗∗∗∗p = 0.000032). D, N1^only and N1-Nb, with or without the G92N and G34N point mutations, were detected in medium by immunoblot analysis using the 6D11 anti-PrP antibody. E, N1^only and N1-Nb in cell lysates, with or without the G92N point mutation, were treated with or without PNGase F as indicated. N1 and N1-Nb were detected by immunoblot analysis with the 6D11 anti-PrP antibody. F, N1-Nb and N1^only in the cell culture medium (top) and cell lysates (bottom) of wildtype, Sec61, Sec62, or Sec63 knockdown 293T cells were detected by immunoblot analysis using the 6D11 antibody. Bracket indicates unprocessed proteins, and red arrow indicates mature proteins. G, statistical analyses of N1-Nb and N1^only in the medium were conducted using one-tailed unpaired t test. The levels of N1-Nb and N1^only in the medium of Sec61, Sec62, or Sec63 knockdown 293T cells were compared with those of wildtype 293T cells (n ≥ 8, ∗∗∗∗p < 0.0001). H, statistical analyses of N1-Nb and N1^only in cell lysates were conducted using one-tailed unpaired t test. The levels of N1-Nb and N1^only in the cell lysates of Sec61, Sec62, or Sec63 knockdown 293T cells were compared with those of the wildtype 293T cells (n ≥ 7, left: ∗∗∗∗p < 0.0001, ∗∗p = 0.0038 and ^nsp = 0.071 and right: ∗∗p = 0.0033 and ^nsp > 0.05). I, the levels of N1-Nb in the medium and lysates of the indicated cells, with or without MG132 treatment, were determined by immunoblot analysis using the 6D11 antibody. Bracket indicates unprocessed proteins, and red arrow indicates mature proteins. J, transient transfected cells were treated with cycloheximide and collected at 0, 1, 4, and 8 h as indicated. N1^only and N1-Nb were detected by Western blot analysis with 6D11 antibody. β-actin was used as the loading control. Statistical analysis of intracellular levels of N1^only (n = 4) and N1-Nb (n = 6) (∗∗p = 0.0077 at 4 h and ∗∗p = 0.0089 at 8 h). K, media were collected at 0, 1, 4, and 8 h after cycloheximide treatment as indicated, and the presence of N1^only and N1-Nb was detected by Western blot analysis with 6D11 antibody. Statistical analysis of levels of N1^only (n = 4) and N1-Nb (n = 6) (∗p = 0.0191 at 4 h and ∗p = 0.0423 at 8 h). Statistical analyses of J and K were performed with two-way ANOVA followed by Bonferroni's multiple comparisons test.

Figure 3

Nb improves N1 production in mouse brain.A, schematic diagram illustrating intracerebroventricular injection of rAAV in newborn mice using BioRender. B, N1^only and N1-Nb in the total brain homogenates of wildtype (WT) and Prnp knockout (Prnp-KO) mice were detected by Western blot analysis with 6D11 anti-PrP antibody. Expected positions of N1, N1-Nb, and endogenous PrP were indicated. C, statistical analysis of N1^only and N1-Nb in Prnp KO and WT mouse brain as indicated (n = 3) (∗∗p = 0.00626 in KO mice and ∗∗p = 0.00145 in WT mice) through one-tailed unpaired t test. Total protein was used as the loading control. D, immunohistochemical staining of N1^only and N1-Nb in Prnp KO mice with 6D11 anti-PrP antibody. High-magnification images are shown in the bottom panel. The gamma values in all images are 0.8. Scale bars represent 1 mm in the top panel and 10 μm in the bottom panel.

Figure 4

Camelid Nb greatly improves N1 production.A, schematic diagram showing the N1 fused with Nb(GFP), Nb(ALFA), Nb(mCherry), or Fc. SP, signal peptide. B, fusion proteins in media, soluble and insoluble fractions of cell lysates were detected by Western blot analysis with 6D11 anti-PrP antibody. Bands with expected molecular weights in media were indicated by asterisks. Statistical analyses of fusion proteins in media (C) (∗∗∗p = 0.0002 and ∗∗p = 0.0033) and the soluble-to-insoluble ratios (D) (∗∗∗∗p < 0.0001, ∗∗∗p = 0.0005, ∗∗p = 0.0069, and ∗p = 0.0239) were carried out with one-way ANOVA (n = 5) followed by Tukey's multiple comparisons test, normalized by total protein. E, schematic diagram illustrating the fusion of N1 with vNAR and IGHV. F, fusion proteins in media, soluble and insoluble fractions of cell lysates were detected by Western blot analysis with 6D11 anti-PrP antibody. Bands with expected molecular weights in media were indicated by asterisks. Statistical analyses of fusion proteins in media (G) (∗∗∗∗p < 0.0001 and ∗∗p = 0.0064) and the soluble-to-insoluble ratios (H) (∗p = 0.0162) were carried out with one-way ANOVA (n = 3) followed by Tukey's multiple comparisons test, normalized by total protein. F, Flag tag; SP, signal peptide.

Figure 5

Nb enhances the production of Shadoo-ΔGPI and SST-28.A, schematic illustration of Shadoo-ΔGPI (Sha) and Shadoo-ΔGPI-Nb fusion protein (Sha-Nb). B, Sha and Sha-Nb in media, soluble and insoluble fractions of cell lysates were detected by Western blot analysis with V5 antibody. Asterisks indicate the expected position of Sha and Sha-Nb in media. C, secreted Sha-Nb proteins with or without deglycosylation with PNGase F as indicated were detected by Western blot analysis with V5 antibody. Statistical analyses of Sha and Sha-Nb in media (D) (∗p = 0.0275) and the soluble-to-insoluble ratios of Sha and Sha-Nb in cell lysates (E) (^nsp = 0.2222) were performed with two-tailed unpaired t test (n = 5), normalized by total protein. F, schematic depiction of SST-28^only and SST-28-Nb fusion protein. G, SST-28^only and SST-28-Nb in media, and soluble and insoluble fractions of cell lysates were detected by Western blot analysis with V5 antibody. The bottom panel is a longer-exposed blot to show SST-28^only. SP, signal peptide; V5, V5 tag.

Figure 6

Nb retains the antigen recognition capability in fusion protein. The N1-Nb fusion protein and N1^only were detected by immunofluorescent staining with the VHH antibody that recognizes Nb. Recombinant GFP was added to show that the Nb(GFP) in fusion was able to bind GFP. The scale bar represents 5 μm.