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

We report immunostaining with signal amplification by exchange reaction, which achieves highly multiplexed signal amplification via DNA-barcoded antibodies and orthogonal DNA concatemers generated by primer exchange reactions

Sinem K. Saka


Scholarcy highlights

  • For application of our in vitro extension and in situ assembly strategy on biological specimens, labeling specificity, resolution and amplification efficiency are the main considerations
  • To evaluate the specificity and preservation of morphology, we performed Immuno-SABER staining in cultured cells for microtubules as a test case for a densely arranged structural protein target
  • To validate its suitability for tissue imaging, we addressed three main considerations. Specific labeling of targets in commonly used sample preparations: We used SABER to label various protein targets in cell and tissue preparations, and reproduced the expected labeling patterns.. Efficient access to the target and quantifiable amplification without sacrificing resolution: SABER probes can penetrate into whole mount preparations and reach targets at 100-μm depth
  • Branched SABER generated ~50-fold amplification, and iterative SABER produced up to ~180-fold amplification. At these levels of amplification, SABER did not have a significant impact to the resolution for conventional microscopy and tissue imaging
  • We validated that the signal level can be further controlled by the concatemer length, and the concatemers do not hinder the access to antigens. Multiplexing: Through the use of orthogonal concatemers, we showed multiplexed amplification for 10 protein targets in the same sample, and simultaneous use of linear and branched SABER
  • Samples were incubated with DNA-conjugated primary antibodies diluted in incubation buffer overnight at 4°C, and washed with washing buffer for five times
  • Signal quantification—For quantification of signal amplification for CD8a labeling in tonsil samples in the Fig. 2 and Supplementary Fig. 3b,d, rectangular regions of interest covering 0.30–1.2 mm tile scans were selected after manual inspection to exclude areas with autofocusing errors or sectioning imperfections, and a customized CellProfiler routine was used first to remove masked autofluroescent structures identified in an independent channel, to calculate mean fluorescence signal/pixel for cell regions masked via thresholding of the CD8a signal

Need more features? Save interactive summary cards to your Scholarcy Library.