The t-CyCIF Process

In t-CyCIF, a form of multi-round multiplex tissue immunofluorescence, ~5 µm thick sections are cut from FFPE blocks, the standard in most histopathology services, followed by dewaxing and antigen retrieval in the usual manner; to reduce auto-fluorescence a cycle of “pre-staining” is performed. Subsequent t-CyCIF cycles each involve four steps (Figure 1): (i) immuno-staining with antibodies against protein antigens (three antigens per cycle in the implementation described here) (ii) staining with a DNA dye (commonly Hoechst 33342) to mark nuclei and facilitate image registration across cycles (iii) four-channel imaging at low and high magnifications (iv) fluorophore bleaching followed by a wash step and then another round of immuno-staining. The signal-to-noise ratio often increases with cycle number. When no more t-CyCIF cycles are to be performed, the specimen is stained with H&E to enable conventional histopathology review. Individual image panels are stitched together and registered across cycles followed by image processing and segmentation to identify cells and other structures.

Figure 1: Assembling a high-plex t-CyCIF image using an iterative process

Imaging

Imaging of t-CyCIF specimens can be performed on a variety of fluorescent microscopes each of which represent a different tradeoff between data acquisition time, image resolution and sensitivity (Figure 2). Specimens several square centimeters in area are often examined at a resolution of ~1 µm on slide scanners, but high resolution image is obtained using confocal or structured illumination microscopes.

Figure 2: t-CyCIF of a metastatic melanoma at different resolutions

Antibodies for tissue-based CyCIF

In the first cycle of t-CyCIF is possible to use indirect immunofluorescence and secondary antibodies. In all other cycles antibodies are directly conjugated to fluorophores, typically Alexa 488, 555 or 647. As an alternative to chemical coupling we have tested the Zenon™ antibody labelling method from ThermoFisher in which isotype-specific Fab fragments pre-labelled with fluorophores are bound to primary antibodies to create immune complexes; the immune complexes are then incubated with tissue samples. This method is effective with some but not all primary antibodies.

To date, we have tested commercial antibodies against ~200 proteins for their compatibility with t-CyCIF. These antibodies include lineage makers, cytoskeletal proteins, cell cycle regulators, the phosphorylated forms of signaling proteins and kinases, transcription factors, markers of cell state including quiescence, senescence, apoptosis, stress, etc. (see Table 1). Currently we rely exclusively on commercial antibodies that have previously been validated using immuno-histochemistry (IHC) or conventional immunofluorescence. We compare staining by t-CyCIF and what has previously been reported for IHC staining (Figure 3) We also compare directly antibodies against the same antigen by using different antibodies in different channels; this enables pixel-level comparison of the same cells (Figure 4).

Figure 3: Anti-PD1 staining in two successive sections of human tonsil by t-CyCIF on the left and IHC in the middle; DNA stained in blue. Right panel shows fraction of positive cells for several antibodies by the t-CyCIF and IHC.

Figure 4: Correlation of anti-PD1 staining by four different antibodies scored on a pixel-by-pixel basis as determined from a single section of human tonsil. Antibody 2 performs poorly in this comparison.

Efforts to date do not constitute a sufficient level of testing or validation for clinical studies and patterns of staining described in this site or in our publications should therefore be considered illustrative of the t-CyCIF approach rather than definitive descriptions. We are currently assembling an OMERO database of matched t-CyCIF and IHC images across multiple tissues and knockdown cell lines to further advance antibody validation. This date will be available near the end of 2018.

Image processing and data analysis

Image processing and data analysis are demanding in the case of high-plex tissue images; we use software tools developed by others supplemented by a growing number of specialized methods (code can be found at our GitHub repository. Once cells are segmented and turned into intensity information, tools such as t-SNE can be used in much the same way as with mass cytometry and other high-dimensional data (Figure 5).

Figure 5: t-CyCIF of human small intestine with analysis

Human subjects disclaimer

Human specimens were retrieved from the archives of the Brigham and Women’s Hospital under a discarded/excess tissue protocol as detailed in Institutional Review Board (IRB) protocol IRB17-1688 (2018) for research deemed to “involve no more than minimal risk to the subjects.”

Funding

This work was made possible by NIH grants P50-GM107618, U54-HL127365, U2C-CA233262, U2C-CA233280, U54-CA225088, the Ludwig Center at Harvard, and the Ludwig Cancer Research Foundation.