Labeling Strategies Matter for Super-Resolution Microscopy: A Comparison between HaloTags and SNAP-tags

Abstract

Super-resolution microscopy requires that subcellular structures are labeled with bright and photostable fluorophores, especially for live-cell imaging. Organic fluorophores may help here as they can yield more photons-by orders of magnitude-than fluorescent proteins. To achieve molecular specificity with organic fluorophores in live cells, self-labeling proteins are often used, with HaloTags and SNAP-tags being the most common. However, how these two different tagging systems compare with each other is unclear, especially for stimulated emission depletion (STED) microscopy, which is limited to a small repertoire of fluorophores in living cells. Herein, we compare the two labeling approaches in confocal and STED imaging using various proteins and two model systems. Strikingly, we find that the fluorescent signal can be up to 9-fold higher with HaloTags than with SNAP-tags when using far-red rhodamine derivatives. This result demonstrates that the labeling strategy matters and can greatly influence the duration of super-resolution imaging.

Keywords: HaloTag; SNAP-tag; STED; fluorophores; live-cell imaging; microscopy; nanoscopy; self-labeling proteins; super-resolution microscopy.

Figure 1

Comparison of Golgi Labeling with HaloTag and SNAP-tag Fusion Proteins of Sialyltransferase (A) Top: Scheme of the labeling procedure. Bottom: Confocal images of live HeLa cells that have been treated as described in the scheme above. The white arrowheads indicate cells that express ManII-GFP and have been labeled with SiR-CA or SiR-BG. Scale bar: 20 μm. (B) Quantification of cells expressing ManII-GFP that are positive for SiR from three independent experiments (ST-Halo, 740 cells in total; ST-SNAP, 837 cells in total). (C) Fluorescence intensity distribution of HeLa cells that were incubated with SiR-CA or SiR-BG and that are transiently expressing ST-Halo-HA, ST-SNAP-HA, or no fusion protein. The number of cells (n) analyzed is indicated in the plot.

Figure 2

Investigation of Various Factors that Could Cause a Difference in Labeling Using SNAP-tags or HaloTags (A) Scheme of labeling procedures used in (B)–(D). (B) Plot showing the percentage of cells expressing ManII-GFP that have also been immunolabeled with an antibody against the HA tag in three independent experiments (ST-Halo-HA, 463 cells in total; ST-SNAP-HA, 489 cells in total). (C) Comparison of labeling efficiency of live and permeabilized cells using SNAP-tags and HaloTags from three independent experiments (ST-Halo-HA: live, 740 cells in total; fixed and permeabilized, 456 cells in total. ST-SNAP: live, 837 cells in total; fixed and permeabilized, 542 cells in total). (D) Average fluorescence intensity of immunostained cells as described in (A). (E) Intensity distribution in Drosophila egg chambers that are expressing Halo-SNAP-aPKC and have been labeled with SiR-CA or SiR-BG.

Figure 3

Comparison of HaloTag and SNAP-Tag Labeling with Various Fluorophores of Various Targets in HeLa Cells and Drosophila (A) Confocal images of HeLa cells expressing Halo and SNAP fusion proteins of sialyltransferase (ST), mannosidase II (ManII), outer membrane protein 25 (OMP25), and clathrin light chain (CLC) that have been labeled with the corresponding SiR substrates. Scale bar: 20 μm. (B) Labeling efficiency of different targets. The bar graph shows the number of ManII-GFP-expressing cells that were positive for labeling of a fusion protein with SiR from three independent experiments (total number of cells for ST: Halo, 740, SNAP, 837; ManII: Halo, 344, SNAP, 436; OMP25: Halo, 563, SNAP, 524; CLC: Halo, 630, SNAP, 460). Left bars, Halo; Right bars, SNAP. (C) Comparison of the ratio of the mean intensity of various SiR-labeled Halo and SNAP fusion proteins. The intensity distribution for each protein and number of cells analyzed are shown in Figure S4. (D) Comparison of the ratio of the mean intensity of Halo and SNAP fusion proteins labeled with TMR, JF549, SiR, and JF646 in HeLa cells and Drosophila egg chambers. The intensity distribution for each protein and number of cells analyzed are shown in Figures S6 and S7. Inset shows the dramatic difference in staining between JF646-CA and JF646-BG in egg chambers. Scale bar: 20 μm.

Figure 4

Comparison of Halo and SNAP Tagging in Live-Cell Super-Resolution Imaging (A) STED images of HeLa cells that are transiently expressing ManII-Halo or ManII-SNAP and that have been labeled with SiR-CA or SiR-BG, respectively (scale bar: 2 μm). The insets show the confocal image of the region highlighted with the green box. The vertical dark and light green lines indicate where the kymographs shown in the middle were taken (scale bar: 60 s). The plots show the average fluorescent signal as a function of position between the arrows shown in the confocal and STED images (dots, measured values; lines, fit). (B) Average initial intensity of STED movies of HeLa cells treated as described in (A) (n = 4 cells). (C) Average intensity over time of STED images of HeLa cells treated as described in (A) (n = 4 cells). (D) STED images of HeLa cells that are expressing Halo-CLC or SNAP-CLC and were labeled with SiR-CA or SiR-BG, respectively. The green boxes highlight clathrin-coated pits with a hollow center. Magnifications of the clathrin-coated pits highlighted with the dashed green boxes are shown in the upper right corner (scale bar: 1 μm). The plots show the average fluorescent signal as a function of position between the arrows shown in the STED images (dots, measured values; lines, fit). (E) STED images of Drosophila egg chambers expressing Halo-SNAP-aPKC that have been labeled with JF646-CA (top) or JF-646-BG (bottom). The insets show a confocal image of the area in the green dashed box. The values indicate the full-width half-maximum of line profiles taken between the arrowheads. Scale bar: 2 μm. (F) First frame of an STED video of Drosophila egg chambers that have been treated as described in (E). Scale bar: 1 μm. The green lines indicate where the kymographs shown next to it have been taken. Scale bar: 100 s. (G) Average initial intensity of STED movies of Drosophila egg chambers that have been treated as described in (E) (Halo, n = 4; SNAP, n = 6).