Green fluorescent protein nanopolygons as monodisperse supramolecular assemblies of functional proteins with defined valency

Supramolecular protein assemblies offer novel nanoscale architectures with molecular precision and unparalleled functional diversity. A key challenge, however, is to create precise nano-assemblies of functional proteins with both defined structures and a controlled number of protein-building blocks. Here we report a series of supramolecular green fluorescent protein oligomers that are assembled in precise polygonal geometries and prepared in a monodisperse population. Green fluorescent protein is engineered to be self-assembled in cells into oligomeric assemblies that are natively separated in a single-protein resolution by surface charge manipulation, affording monodisperse protein (nano)polygons from dimer to decamer. Several functional proteins are multivalently displayed on the oligomers with controlled orientations. Spatial arrangements of protein oligomers and displayed functional proteins are directly visualized by a transmission electron microscope. By employing our functional protein assemblies, we provide experimental insight into multivalent protein–protein interactions and tools to manipulate receptor clustering on live cell surfaces.


Supplementary Figure 1. SDS-PAGE analysis of GFP oligomer variants with different linkers.
Oligomer mixtures were applied to a PAGE gel containing 0.1% SDS without boiling. The gel was analyzed by a fluorescent image analyzer with 470 nm-excitation and 530 nm-emission filters. Figure 2. Design of the charge variants of GFP monomer. a, Ribbon cartoon diagrams of the charge variants of GFP monomer. The GFP 11 strand is shown in blue. Substituted residues are shown in red, and mutations are indicated with arrows. b, Protein sequences of GFP monomer with net charges of -5, -7, -9 and -15. Blue, GFP 11; underlined, peptide linker; red, mutated residues. Figure 3. SEC analysis of GFP oligomer charge variants. GFP oligomers with net charges of -5 (a), -7 (b), -9 (c) and -15 (d) were analyzed using a superdex 200 column (10/300 GL).

Supplementary Figure 4. Partial purification of GFP oligomers by affinity chromatography.
GFP oligomers (net charge of -3) were eluted from a Ni-chelating column with elution solutions containing imidazole from 100 mM to 450 mM. Samples (without boiling) were analyzed by SDS-PAGE. The gel was analyzed by a fluorescent image analyzer with 470 nm-excitation and 530 nmemission filters. Figure 5. TEM images of partially purified GFP polygons with a net charge of -3 (wild type). a, Mixtures of GFP polygons (net charge -3, wild type) were partially purified by the electro-elution method and analyzed by SDS-PAGE (without boiling). b, TEM images of purified GFP polygons from fraction #1 to fraction #5. For fractions #4 and #5, polygonal GFP oligomers with more than 10 GFP monomers were also observed. GFP polygons (trimer, tetramer, and pentamer) were reacted with excess GFP 1-10, and resulting protein assemblies were analyzed in a native-PAGE gel.

Supplementary
Supplementary Figure 16. Representative TEM images of opened GFP oligomers from pentamer to decamer. Scale bars, 10 nm.

Supplementary Figure 17. Specific interaction of protein G-fused GFP polygons to surfacebound antibodies.
A SPR sensor chip surface was covered with mouse IgG1 (4000 RU), and GFP polygons (Polygon 3 mer & 7 mer) as well as protein G-fused GFP polygons (Protein G 3 mer & 7 mer) were applied at constant monomer concentration (10 μg ml -1 ). Binding curves were normalized by subtracting the reflective index changes upon sample injections.

Supplementary Figure 18. SPR responses upon multivalent protein G polygon binding to human Fc domain.
A SPR sensor chip surface was covered with recombinant human Fc domain (15000 RU), and protein G-fused polygons were applied at constant monomer concentration (5 μg ml -1 ). Binding curves were normalized by subtracting the reflective index changes upon sample injections. Here recombinant human Fc protein was employed, whereas human IgG mixtures (isolated directly from human serum) were used in Figure 5b in the main text. More homogeneous interaction between Fc domain and protein G was expected, and binding curves of Fc domain were more similar to the simulated curves (Supplementary Figure 19a) than those of human IgG. Figure 19. Simulated SPR binding curves of multivalent protein G polygons with surface-bound human IgG (a) and mouse IgG (b). a, 150 nM of protein G was applied to surface-bound human IgG. Association (180 sec) and dissociation (320 sec) phases were simulated with an association constant (k on ) 1 × 10 5 (M -1 s -1 ) and dissociation constants (k off ) 10 -2 , 10 -5 , and 10 -6 (s -1 ). b, 300 nM of protein G was applied to surface-bound mouse IgG1. Association (180 sec) and dissociation (320 sec) phases were simulated with an association constant (k on ) 0.5 × 10 5 (M -1 s -1 ) and dissociation constants (k off ) 10 -1 , 10 -3 , and 10 -5 (s -1 ). The units are relative response units (RU/RU max ). The simulation equation for association curves is [relative response units (RU/RU max ) = + /

Supplementary Figure 20. Confocal microscopy analysis of internalization of antibody-receptor clusters by protein G polygons with various valency. A549 cells were sequentially treated with
Cy5-Erbitux (10 μg ml -1 ) and protein G-fused GFP polygons (10 μg ml -1 ) with various valency or free protein G (10 μg ml -1 ). Erbitux alone or Erbitux-protein G bounded cells were incubated for 30 min at 37 °C, and receptor internalization was monitored by confocal microscopy. Cy5-Erbitux, red; protein G-polygon, green.

Supplementary Figure 21. Non-specific interactions of protein G polygons on cell surfaces.
A549 cells were treated with protein G polygons (10 μg ml -1 ) and incubated for 30 min at 37 °C. Nuclei were stained with DAPI. Protein G polygons (green) and DAPI (blue) were monitored using confocal microscopy.

Supplementary Figure 22. Flow cytometry analysis of antibody-mediated receptor internalization after 180 min.
Erbitux or Erbitux-protein G bounded cells were incubated for 180 min at 37 °C. Internalization of Cy5-Erbitux was quantified by flow cytometry, and relative fluorescence intensities are given in the cytometry profile data.

Supplementary Table 1. Protein sequences of GFP monomer variants with different linkers
Supplementary

MCherry, protein G and MBP fused GFP monomer
Protein fusions are shown in blue. Linkers are in underlined.