Supplementary Figure 6: Sequence validation for Exchange-SABER. | Nature Biotechnology

Supplementary Figure 6: Sequence validation for Exchange-SABER.

From: Immuno-SABER enables highly multiplexed and amplified protein imaging in tissues

Supplementary Figure 6

(a) 32 SABER sequences were extended to ~650 bases in vitro and examined by gel shift assay visualized with SybrGold on 6% denaturing PAGE gels. Qualitatively, 18 of the 32 sequences (such as #29) displayed a broader distribution with a ladder of shorter products visible albeit these bands being much dimmer than the predominant concatemer band. Although being extended, one sequence (#51) had a more even distribution in the upper length regime without a clear predominant band. Based on these distributions, particular applications may favor a subset of the sequence library, i.e. high efficiency primers (for example primers 27, 28, 30, 31, 37, 38, 41, 44, 47, 49, 50, 54) may be more favorably utilized for more quantitative experiments or for lower abundance proteins. (b-d) In situ performance and crosstalk analysis. BS-C-1 cells were stained with bridge DNA-conjugated antibodies targeting α-Tubulin on a 96-well plate. Concatemers extended from each primer were hybridized to the bridges creating an array of wells labeled with primer sequences p.25-p.56. For each concatemer two sets of wells were prepared: (i) cognate group to be incubated with the corresponding imager strands, and (ii) crosstalk group where we added mixtures of imagers except the cognate imager strand. For each primer (e.g. p.25), both cognate and crosstalk wells were prepared by either applying the corresponding Alexa647-imager (e.g. i.25*) or all the imagers except the cognate one (e.g. -i.26* to i.56*). Images were captured in 16-bit (0-65,535). Representative images are shown for Primer 25 (p.25) (c) and Primer 27 (p.27) concatemers (d). Crosstalk images are displayed with two different intensity scales to render the crosstalk signal visible. The fluorescence signals were quantified and plotted in the log scale and displayed as a heatmap. Consistent with the in vitro gel shift assay in panel a, sequences that had lower extension efficiency (particularly primers 29, 32, and 51) tended to yield less fluorescence signal compared to sequences that that extend with higher efficiency. Non-negligible crosstalk signal was only detected for Primer 27 concatemer (red box). (e) Crosstalk analysis of primer sequence p.27. BS-C-1 cells were fixed and stained with DNA-conjugated antibodies targeting alpha-Tubulin. Concatemers extended from primer sequence p.27 were hybridized to the antibodies, followed by addition of non-cognate imager strands. We first grouped every five imager strands and determined that p.27 had crosstalk with imager 44-49. We then tested individual imager strands from imager 44-49 and determined the strand responsible for crosstalk as imager strand 48 (i.48*), which is excluded from the library for further multiplexed imaging in presence of p.27.

Back to article page