Identification and characterization of nanobodies specific for the human ubiquitin-like ISG15 protein

Abstract

Interferon-induced ubiquitin (Ub)-like modifier Interferon Stimulated Gene 15 (ISG15) functions both intracellularly and as a secreted protein with cytokine-like properties. The ISG15 pathway is implicated in various diseases, including cancer and inflammatory disorders, but understanding its precise roles has been challenging because of limited availability of tools to study ISG15 biology. Here, we report the development of two novel nanobodies that target human ISG15, obtained through alpaca immunization and phage display. These nanobodies, VHH_ISG15-A and VHH_ISG15-B, exhibit nanomolar binding affinities and recognize distinct epitopes on ISG15's C- and N-terminal domains, respectively, as demonstrated by NMR and X-ray structural analyses. Both nanobodies enable the immunoprecipitation and proteomic identification of ISGylated substrates with minimal background contamination. VHH_ISG15-A is compatible with immunoblotting and recognizes unconjugated ISG15 under denaturing conditions. Functional assays showed that VHH_ISG15-A, but not VHH_ISG15-B, inhibits ubiquitin-specific peptidase 16-mediated deISGylation, likely by steric hindrance at the ISG15-binding interface. These results underscore the utility of VHH_ISG15-A and VHH_ISG15-B as tools to study ISG15 biology.

Keywords: ISG15; NMR; USP16; VHH; nanobody; protein structure.

Figure 1

Identification of human ISG15–specific nanobodies.A, scheme of alpaca immunization and ISG15 nanobody (VHH) selection by phage display. B, sequence alignment of the identified ISG15 nanobodies, with indication of the complementarity determining regions (CDRs). ISG15, Interferon Stimulated Gene 15.

Figure 2

Binding validation of ISG15-specific nanobodies.A and C, binding of nanobodies to hISG15-SBP-His as determined by size-exclusion chromatography (SEC). A, profiles of hISG15-SBP-His (green, peak A), VHH_ISG15-A (black, peak B), and a 1:2 M ratio mixture of hISG15-SBP-His:VHH_ISG15-A (blue, peaks C and D) are shown. B, profiles of hISG15-SBP-His (green, peak A), VHH_ISG15-B (orange, peak E), and a 1:2 M ratio mixture of hISG15-SBP-His:VHH_ISG15-B (pink, peaks F and G) are shown. C, profile of a 1:1:1 M ratio mixture of hISG15-SBP-His:VHH_ISG15-A:VHH_ISG15-B (red, peak H) is shown. SDS-PAGE gels of the 10 to 14 ml fractions corresponding to the VHH:ISG15–SBP complex separated by SEC are shown on the right panel. D, binding of VHHs to His-mISG15 as determined by SEC. Profiles of His-mISG15 (green), VHH (blue), and a 1:2 M ratio mixture of His-mISG15:VHH (red) are shown. E, representative sensorgrams from biolayer interferometry (BLI) analysis showing the binding interactions of ISG15 with VHH_ISG15-A (top panel) and VHH_ISG15-B (bottom panel), respectively. Experimental association and dissociation data (dots) are overlaid with globally fitted binding model curves (solid red lines). The data (n = 3) were used to calculate the equilibrium dissociation constant (K_D), association rate constant (K_on), and dissociation rate constant (K_off). ISG15, Interferon Stimulated Gene 15; SBP, streptavidin-binding peptide.

Figure 3

Binding characterization of ISG15-specific VHHs.A and B, estimated combined 1H/15N-chemical shift changes in ppm of 15N-labeled human ISG15 in the presence of excess VHH_ISG15-A (A) and VHH_ISG15-B (B). Individual ISG15 residues with chemical shift changes >0.1 ppm are indicated in purple, and those >0.2 ppm are indicated in red. C, crystal structure of VHH_ISG15-A in complex with human ISG15, resolved to 2.6 Å. Cartoon representation of ISG15 (blue) interacting with VHH_ISG15-A (tan) via the C-terminal domain. D, close-up view of the binding interface between ISG15 and VHH_ISG15-A. Key interface residues are shown in stick. E, structure of VHH_ISG15-B in complex with human ISG15. Cartoon representation of AlphaFold3 predicted complex of ISG15 (blue) and VHH_ISG15-B (green) bound via the N-terminal domain. F, close-up view of the predicted binding interface between ISG15 and VHH_ISG15-B. Key interface residues are shown in stick, H-bonds as dashed lines. G, SEC profiles depicting human ISG15 and VHH_ISG15-B binding. Left panel, ISG15 WT preincubated with VHHISG15-B WT at a 1:1 M ratio. Middle panel, ISG15 (V74D) preincubated with VHHISG15-B (F98D) at a 1:1 M ratio. Right panel, ISG15 (V74D) preincubated with VHHISG15-B WT at a 1:1 M ratio. ISG15, Interferon Stimulated Gene 15; SEC, size-exclusion chromatography.

Figure 4

Biochemical characterization of ISG15-specific VHHs.A, VHH_ISG15-A, but not VHH_ISG15-B, recognizes recombinant human ISG15 in immunoblot analysis. Recombinant hUbcH8-SBP-His (human), hISG15-SBP-His (human), and His-mISG15 (murine) proteins were separated by SDS-PAGE (left) and immunoblots (right) were performed with VHHs as the primary probes, followed by anti-HA monoclonal antibody to detect HA-tagged VHHs. Representative data of three (n = 3) independent experiments. B, VHH_ISG15-A, but not VHH_ISG15-B, detects cellular unconjugated ISG15 in HeLa cell lysates via immunoblot. HeLa-FLAG-ISG15:HeLa cell lysates exogenously expressing Tetra-FLAG–tagged ISG15. Representative data of three (n = 3) independent experiments. C, immunoprecipitation of ISGylated substrates using C-terminally biotin-tagged VHH_ISG15-A and VHH_ISG15-B. Biotin-tagged VHH_ISG15-A and VHH_ISG15-B were prebound to streptavidin beads, followed by incubation with 1 mg of IFN-β-stimulated HeLa cell lysates for immunoprecipitation. Immunoprecipitated proteins were eluted by boiling, separated by SDS-PAGE, and detected by immunoblotting with anti-human ISG15 antibody (left) and silver staining (right). Representative data of three (n = 2) independent experiments. D, heatmap displaying the intensities of the top 30 interactors of VHH_ISG15-A (VHH-A) and VHH_ISG15-B (VHH-B) and an ISG15 antibody (ISG15-Ab), identified by LC–MS/MS analysis in HAP1 USP18 KO cells treated with IFN-α (500 U/ml) for 48 h. This experiment was performed in technical duplicates, the enrichment background was corrected using a beads-only control, and the top 30 interactors were selected based on Manhattan distances. IFN, interferon; ISG15, Interferon Stimulated Gene 15; USP18, ubiquitin-specific peptidase 18.

Figure 5

VHH_ISG15-A, not VHH_ISG15-B, inhibits USP16-mediated deISGylation in vitro.A, surface/cartoon representation of ISG15 interacting with VHH_ISG15-A, VHH_ISG15-B (AF3 predicted), and USP16 (AF3 predicted). Top left, VHH_ISG15-A (tan) interacts with ISG15 (blue) via the C-terminal domain. Top right, VHH_ISG15-B (green) interacts with ISG15 via the C-terminal domain. Bottom left, complex of ISG15 with USP16 (brown) predicted with AF3 interacts with ISG15 at the N- and C-terminal domains. For clarity, residues 392 to 604 of USP16 are not shown here as AF3 predicts it to be disordered. Bottom right, overlay of the three complex structures, aligned on ISG15. Steric clash between VHH_ISG15-A and USP16 is clearly apparent. B, cleavage of recombinant human proISG15 proteins by recombinant human USP16 in the presence and absence of a twofold molar excess of VHHs over proISG15 at the indicated time points. Proteins were separated by SDS-PAGE and immunoblotted with anti-human ISG15 antibody. Representative data of three (n = 3) independent experiments. C, deconjugation of ISGylated substrates by recombinant human USP16 in the presence and absence of 10 μM VHHs. Proteins were separated by SDS-PAGE and immunoblotted with anti-human ISG15 antibody. Probing for β-actin served as a loading control. Representative data of two (n = 2) independent experiments. AF3, AlphaFold 3; ISG15, Interferon Stimulated Gene 15; USP16, ubiquitin-specific peptidase 16.