A peptide tag-specific nanobody enables high-quality labeling for dSTORM imaging

Abstract

Dense fluorophore labeling without compromising the biological target is crucial for genuine super-resolution microscopy. Here we introduce a broadly applicable labeling strategy for fixed and living cells utilizing a short peptide tag-specific nanobody (BC2-tag/bivBC2-Nb). BC2-tagging of ectopically introduced or endogenous proteins does not interfere with the examined structures and bivBC2-Nb staining results in a close-grained fluorophore labeling with minimal linkage errors. This allowed us to perform high-quality dSTORM imaging of various targets in mammalian and yeast cells. We expect that this versatile strategy will render many more demanding cellular targets amenable to dSTORM imaging.

Fig. 1

Comparison and characterization of BC2-nanobody (BC2-Nb) formats for wide-field and dSTORM imaging. a Schematic illustration of the BC2-Nb dye-conjugation strategies. Monovalent and bivalent BC2-Nbs were either conjugated with Alexa Fluor 647 (AF647) via N-hydroxysuccinimide (NHS) ester (left panel) or linked to AF647 by enzymatic sortase coupling (right panel). Wide-field imaging of chemically fixed HeLa cells expressing mCherry-vimentin_BC2T (mCherry-VIM_BC2T) stained with modified BC2-Nbs. Monovalent versions of the BC2-Nbs (NHS- and sortase-coupled) are depicted on the left panel, corresponding bivBC2-Nbs are displayed on the right side. Stainings with NHS-conjugated nanobodies are shown in two different image contrasts, the upper half in the same brightness and contrast as the sortase-coupled nanobodies; in the lower half with an adjusted contrast. Scale bars, 25 µm. b Representative dSTORM images of chemically fixed HeLa cells expressing vimentin_BC2T, stained with the monomeric NHS-conjugated BC2-Nb_{AF647 (NHS)} (left) and the sortase-coupled bivBC2-Nb_{AF647 (sort)} (right). Scale bars, images 5 µm, insets 1 µm. Image reconstruction details are given in Methods section. c Assessment of staining quality in wide-field fluorescence imaging. Labeling of the different nanobody formats was quantified by calculating the ratio of the signal intensity of mCherry-VIM_BC2T expressing cells to non-transfected cells (background), (BC2-Nb_{AF647 (NHS)}: n = 115; bivBC2-Nb_{AF647 (NHS)}: n = 134; BC2-Nb_{AF647 (sort)}: n = 150; bivBC2-Nb_{AF647 (sort)}: n = 195) (Methods section, Supplementary Fig. 1). d Assessment of bivBC2-Nb_AF647 staining of endogenous β-catenin. Bar chart summarizes measured nanobody per µm² values for untransfected chemically fixed HeLa cells stained with GFP-Nb_AF647 or bivBC2-Nb_AF647 in comparison to chemically fixed HeLa cells transiently expressing _BC2TLC3B stained with bivBC2-Nb_AF647, errors given as standard deviation (S.D.), N = 3 cells for each condition (Methods section, Supplementary Fig. 2). e Quantification of completeness of labeling for the bivBC2-Nb and SNAP-tag labeling systems using FtnA-oligomers of 24 subunits. Bar chart summarizes median values of FtnA-24mer fluorescence intensities as percentage of theoretical maxima (Methods section, Supplementary Fig. 3)

Fig. 2

Super-resolution imaging and analysis of differently labeled vimentin constructs. a Schematic illustration of labeling strategies used for comparative SRM imaging of native or ectopically expressed vimentin. b Representative PALM/dSTORM images of chemically fixed Hela cells expressing the corresponding constructs outlined in a or native vimentin (left panel). Insets show magnifications of representative vimentin filaments of varying thickness (1—thick, 2—medium, 3—thin, peripheral). Scale bars, 5 µm in main images, 1 µm in insets. c Filament widths as histograms (left) with a bin size of 75 nm (x-axis) plotted against relative fraction (y-axis). Full data are represented underneath the histograms as box + scatter plots with the same x-axis. The box marks the three quartiles and the whiskers mark 95% of all the data. The average lengthwise fluorophore coverage was calculated for each bin and plotted (right) as mean filament width (black line) and standard deviation (colored area) against relative fraction covered by fluorophores (y-axis). Width and lengthwise fluorophore coverage were analyzed for a total of 676 (bivVB6-Nb_AF647), 295 (PAmCherry), 724 (GFP-Nb_AF647), and 620 (bivBC2-Nb_AF647) filaments, N = 5 cells for each condition, cells, and selected filaments are shown in Supplementary Fig. 4. Image reconstruction details are given in Methods section

Fig. 3

Super-resolution imaging of transiently expressed BC2-tagged proteins in chemically fixed cells. Representative dSTORM images of a a HeLa cell expressing _BC2Tlamin, b U2OS cell expressing tubulin_BC2T (filament width statistics in Supplementary Fig. 9a), c HeLa cell expressing _BC2Tactin (coverage statistics in Supplementary Fig. 9b), d HeLa cells expressing _BC2TLC3B either left untreated or treated with rapamycin. Bar charts represent the degree of clustering, given as a relative fraction of cluster points versus noise points, errors given as standard deviation (S.D.). Histograms represent cluster diameters as determined by DBSCAN analysis with a bin size of 100 nm (x-axis) plotted against relative fraction (y-axis). Full data are represented underneath the histograms as box + scatter plots with the same x-axis. The box marks the three quartiles and the whiskers mark 95% of all the data. Total number of clusters n = 342 in non-treated cells, n = 405 in treated cells, N = 3 cells for untreated cells and N = 4 cells for rapamycin-treated cells (Supplementary Fig. 11), and e HeLa cells expressing _BC2TGFP-GPI. All cells were stained with bivBC2-Nb_AF647 (Methods section). Scale bars, images 5 µm, insets 1 µm. Crossed out rectangles mark the position of fiducial markers used for drift correction. Image reconstruction details are given in Methods section

Fig. 4

Visualization of endogenously expressed BC2-tagged actin labeled with bivBC2-Nb_AF647. a Wide-field images of chemically fixed wild-type A549 and HeLa (-wt; left panel), as well as chemically fixed A549-_BC2TACTB and HeLa-_BC2TACTB cells (right panel). Cells were either left untreated (0 h) or stimulated for 48 h with TGFβ (5 ng ml⁻¹) followed by staining with phalloidin_AF555 and bivBC2-Nb_AF647. Scale bars, 25 µm. b dSTORM image of a representative HeLa-_BC2TACTB cell. Scale bars, image 5 µm, insets 1 µm. Image reconstruction details are given in Methods section. Imaging sequence taken from raw data acquisition can be found in Supplementary Movie 5, assessment of AF647 photophysics under dSTORM imaging conditions can be found in Supplementary Fig. 15

Fig. 5

Super-resolution imaging and single-particle tracking in live HeLa cells. a dSTORM image of live _BC2TGFP-GPI expressing HeLa cells stained with bivBC2-Nb_AF647 and insets in gray scale overlaid with single-particle trajectories of immobile (diffusion coefficient below 0.02 µm² s⁻¹, left) and mobile (diffusion coefficient above 0.02 µm² s⁻¹, right). Color scale of diffusion coefficients is given under insets. Scale bars 10 µm in images, 2 µm in insets. Supplementary Movies 1–4 show the recorded live _BC2TGFP-GPI dynamics of the whole cell and the corresponding insets. b Live-cell dSTORM images of two HeLa-_BC2TACTB cells stained with bivBC2-Nb_ATTO655. Wide-field fluorescence images in upper left corners. Scale bars in 10 µm, 2 µm in insets. Image reconstruction details are given in Methods section. Imaging sequence taken from raw data acquisition can be found in Supplementary Movie 6, assessment of ATTO655 photophysics under live cell dSTORM imaging conditions can be found in Supplementary Fig. 15