Spy&Go purification of SpyTag-proteins using pseudo-SpyCatcher to access an oligomerization toolbox

Abstract

Peptide tags are a key resource, introducing minimal change while enabling a consistent process to purify diverse proteins. However, peptide tags often provide minimal benefit post-purification. We previously designed SpyTag, forming an irreversible bond with its protein partner SpyCatcher. SpyTag provides an easy route to anchor, bridge or multimerize proteins. Here we establish Spy&Go, enabling protein purification using SpyTag. Through rational engineering we generated SpyDock, which captures SpyTag-fusions and allows efficient elution. Spy&Go enabled sensitive purification of SpyTag-fusions from Escherichia coli, giving superior purity than His-tag/nickel-nitrilotriacetic acid. Spy&Go allowed purification of mammalian-expressed, N-terminal, C-terminal or internal SpyTag. As an oligomerization toolbox, we established a panel of SpyCatcher-linked coiled coils, so SpyTag-fusions can be dimerized, trimerized, tetramerized, pentamerized, hexamerized or heptamerized. Assembling oligomers for Death Receptor 5 stimulation, we probed multivalency effects on cancer cell death. Spy&Go, combined with simple oligomerization, should have broad application for exploring multivalency in signaling.

Fig. 1

Overview of Spy&Go. SpyTag/SpyCatcher interact irreversibly by spontaneous isopeptide bond formation, promoted by E77. SpyTag purification requires the generation of reversible interaction with SpyDock in 3 steps: blocking reaction, enhancing efficiency of SpyTag binding and precise immobilization on resin. Spy&Go then enables purification of SpyTag-fusions, setting the stage for modular oligomerization or multimerization

Fig. 2

Design of SpyDock. a All small residues except D blocked SpyCatcher reaction. E77 was mutated to the indicated residue and incubated with SpyTag-MBP for 24 h, before boiling in SDS and analysis by SDS-PAGE with Coomassie staining. b Additional mutations in SpyCatcher enhanced Spy&Go purification. The original SpyCatcher, the evolved SpyCatcher002 or SpyCatcher2.1 bearing the E77A mutation were compared for capture and elution of SpyTag-MBP (SDS-PAGE with Coomassie staining). Further E77X mutations were similarly explored on SpyCatcher2.1. c Positions of mutations screened for SpyDock acceleration, based on Protein Data Bank 4MLI; CnaB2 triad represented in spheres, anchoring sites in yellow, and SpyCatcher2.1 accelerating mutations in green. d S49C resin anchoring allowed efficient SpyDock purification. SpyCatcher2.1 E77A S49C was coupled to SulfoLink resin and tested for SpyTag-MBP capture and release (SDS-PAGE with Coomassie staining)

Fig. 3

Affinity of SpyDock for its targets. ITC binding isotherms for SpyDock binding at 25 °C to a SpyTag-MBP or b SpyTag002-MBP. Error bars represent the uncertainty in fit to the binding curve using a 1:1 binding model

Fig. 4

Spy&Go from bacterial expression. a SpyTag-MBP was purified from E. coli clarified lysate by Spy&Go. Protein: input SpyTag-MBP protein. T: total pooled elutions. Purity of T was determined by densitometry (right); gray represents background lane intensity (mean ± 1 s.d., n = 3). b Ni-NTA purification of SpyTag-MBP from the same lysate via its His-tag. c Spy&Go purification of scPvuII-SpyTag (SpyTag at an internal loop, shown schematically) from bacterial lysate. d Spy&Go purification of the nanobody αDR5-SpyTag (C-terminal fusion) from bacterial lysate. All fractions were analyzed by SDS-PAGE with Coomassie staining

Fig. 5

Spy&Go from mammalian expression. a HEK293T cells were transfected with the extracellular region of EpCAM fused to SpyTag and a His-tag (EpCAM-SpyTag). EpCAM-SpyTag was purified from the clarified cell supernatant using Spy&Go. Fractions were analyzed by SDS-PAGE with Coomassie staining. T: total pooled elutions. Resin: resin post-elution. b Ni-NTA purification of EpCAM-SpyTag as in a. c Spy&Go purification of CyRPA-SpyTag from Expi293HEK cells as in a

Fig. 6

Construction of oligomeric toolbox. a Spy Oligomerization Toolbox from establishing a panel of coiled coils with valency from 2 to 7 for fusion to SpyCatcher002. The coiled coils are presented with each color representing one chain and the N-terminus colored as a blue ball. The C-terminus of SpyCatcher002, where it will be linked to the coiled coil, is colored as a red ball. b DLS to show assembly of each SpyCatcher002-oligomer, with hydrodynamic radius (R_h) (mean ± 1 s.d., n = 10) for each assembly. c SEC-MALS to show assembly of each SpyCatcher002-oligomer. The peak shows the normalized absorbance units (AU) at 280 nm of the SpyCatcher002-oligomer species from SEC. The horizontal line shows the distribution of molar mass (g/mol) of the species in the peak from MALS. Expected and observed molecular weight (M_r) is shown alongside, with error bars representing the uncertainty in fit to the molar mass curve

Fig. 7

Oligomer panel tested for cancer cell killing. a Cartoon depicting depletion of free SpyTag-ligand using Spy&Go resin. b Coupling of SpyTag-fusion to coiled coil series. αDR5-SpyTag was incubated with each SpyCatcher002-coiled coil and analyzed by SDS-PAGE with Coomassie staining. In Recaptured lanes, excess αDR5-SpyTag was removed from the coiled coil conjugate by an additional passage through Spy&Go resin. c Dose-response curve of MDA-MB-231 cell viability when treated with different concentrations of αDR5 conjugated to each SpyCatcher002-oligomer platforms. The line at 50% cell viability shows the cut-off for EC₅₀ calculation. The x-axis is normalized to the concentration of αDR5 monomer. Error bars represent mean ±1 s.d., n = 2. d MDA-MB-231 viability upon incubation as in c with the building blocks of αDR5 alone or coiled coils alone. Error bars represent mean ± 1 s.d., n = 3