UltraPlex Hapten-Based Multiplexed Fluorescent Immunohistochemistry

Abstract

The UltraPlex method for multiplexed two-dimensional fluorescent immunohistochemistry is described, in which hapten tags conjugated to primary antibodies facilitate multiplexed imaging of four or more antigens per tissue section at once. Anti-hapten secondary antibodies labeled with fluorophores provide amplified signal for detection, which is accomplished using a standard fluorescent microscope or digital slide scanner. The protocol is rapid, straightforward, and utilizes conventionally prepared tissue samples. The resulting staining is highly sensitive and specific, enabling high resolution imaging of multiple cellular subtypes within tissue samples. Tumor cells and tumor-infiltrating lymphocytes are presented as examples. Multiple 4-plex stained tissue samples can be digitally overlaid to create 8-plex (or more) high content images, enabling visualization of distribution of complex cellular subtypes across tissues.

1. INTRODUCTION

In recent decades, there has been tremendous progress in understanding the role of inflammation in disease and treating illness by targeting the immune system rather than the disease process itself. While blood biomarkers can give a window into inflammatory disease, it is often important to examine the affected tissue directly. This has led to increasing demand for tools to characterize the tissue immune microenvironment, particularly in oncology, where novel antibody and cellular immunotherapies that drive an anti--tumor immune response are making dramatic impacts.

The inflammatory infiltrate in tumors is traditionally examined by microscopy analysis of tumor infiltrating lymphocytes (TILs) in thin sections cut from formalin--fixed, paraffin--embedded (FFPE) tissue blocks and stained with hematoxylin and eosin (H&E). While TILs can be generally recognized by their characteristic H&E staining pattern, this basic method fails to differentiate among lymphocyte subsets, leaving the nature of the immune infiltrate a mystery. Chromogenic immunohistochemistry (IHC) improves upon H&E staining by using a primary antibody (1° Ab) to detect a specific protein marker, allowing straightforward identification of (e.g.) the CD8+ cytotoxic lymphocytes among the TILs. However, the practical limit for chromogenic IHC is two antigens on a single sample, which is insufficient to assess many immune cell phenotypes.

Given this technology gap, there has been growing interest in multiplexed detection as a strategy to examine tissue to better define the cell types and their activation states within the immune microenvironment (reviewed by [1, 2]). Rather than relying on chromogens, for multiplexed immunofluorescence (MxIF), each target is labeled with a unique fluorophore which can be detected at a distinct wavelength using a standard fluorescent microscope. Much like the multiplexed fluorescent detection used for flow cytometry, a simple solution is to use simultaneous staining with fluorescently labeled 1° Abs for direct MxIF. However, in order to detect many relevant antigens, the amplification provided by fluorescent secondary antibodies (2° Abs) is required. A long--standing problem in this indirect immunofluorescence approach is that 2° Abs detect their cognate 1° Abs based on unique features of the constant region linked to species of origin (mouse, rabbit, etc.) and/or isotype. Typically, the different 1° Abs in a MxIF panel must come from different species. However, many of the best antibodies to detect immune markers are rabbit monoclonals, obviating simultaneous MxIF detection with 2° Abs. Simultaneous MxIF is also limited by the number of channels available for detection. Most widefield and confocal fluorescence research microscopes offer four to seven fluorescent channels, which can potentially be extended by spectral unmixing [3, 4] or other strategies, but imaging whole slides is often challenging. However, whole slide scanners are available from various companies which support fluorescent scanning of four or more fluorescence channels. To solve the various problems commonly encountered with conventional approaches, UltraPlex MxIF technology provides signal amplification by indirect immunofluorescence detection while remaining compatible with 1° Abs derived from any species and a wide range of existing imaging platforms. To assemble an UltraPlex MxIF panel, a set of four 1° Abs are covalently coupled to peptide--like hapten tags. These barcoded tags are then recognized by four fluorescently labeled, high affinity (K_D < 10⁻⁻¹⁰ M) anti--hapten monoclonal 2° Abs. The hapten--coding strategy permits both the 1° Abs and 2° Abs to be applied as “cocktails” without risk of cross reactivity (Figure 1A). Along with high specificity, because each 1° Ab is conjugated to multiple haptens and each anti--hapten 2° Ab is conjugated to multiple fluorophore molecules, UltraPlex offers high sensitivity, with detection equivalent to that of conventional anti--species 2° Abs and superior to that of directly conjugated fluorescent 1° Abs (Figure 1B).

Figure 1.

UltraPlex MxIF technology. (A) Workflow schematic in which a cocktail of hapten-labeled primaries is incubated on the tissue followed by a cocktail of fluorophore-labeled anti-hapten antibodies. (B) Images of luminal B “triple positive” breast cancer tissue stained with (left) a cocktail of directly labeled fluorescent 1° Abs detecting HER2, ER and PR; (center) rabbit anti-HER2 1° Ab followed by fluorophore-labeled anti-rabbit 2° Ab; (right) an UltraPlex MxIF staining panel consisting of hapten-labeled HER2, ER and PR 1° Abs followed by fluorophore-labeled anti-hapten 2° Abs. In the left panel, use of directly labeled antibodies renders the ER and PR signals too dim to be seen. In the center panel, only a single antigen (Her2) can be probed as all 1° Abs are of rabbit origin. Only the UltraPlex approach can be used to multiplex all three antigens while clearly revealing each antigen on the tissue.

In contrast to the limited use of anti-species 2° Abs in MxIF applications, UltraPlex MxIF is limited only by the number of available hapten/anti--hapten pairs. However, MxIF of four antigens with four fluorescent colors on each slide (plus a nuclear counterstain) is well matched to the capabilities of commercial fluorescent microscopes or digital slide scanners. Thus, we have constructed panels of four 1° Abs that recognize related antigens such as TIL subsets or tumor cell markers. The UltraPlex MxIF experiment starts with cutting and processing tissue sections exactly as for conventional IHC. Either frozen or FFPE tissues are acceptable. Then, a “cocktail” of four hapten--barcoded 1° Abs is applied to each section, incubated for 1 h, and washed. This is followed by applying a cocktail of four fluorescent anti--hapten 2° Abs, incubation for 1 h, and washing. Each slide is then coverslipped and imaged by fluorescence microscopy or digital slide scanning. Optionally, UltraPlex hapten--modified microspheres can be used for instrument calibration and image quantitation. Images can be analyzed using standard biological image processing freeware (ImageJ Fiji, [5-7]). Alternatively, commercially available digital pathology software packages can be used to facilitate quantitative and spatial analysis of features of the immune microenvironment revealed by UltraPlex MxIF staining. UltraPlex panels are able to simultaneously assess a number of parameters of potential prognostic and therapeutic significance in disease (Figure 2). UltraPlex staining of normal tonsil tissue (Figure 2A) with CD4, CD8, Ki--67 and CD31 antibodies shows T cells residing primarily in the interfollicular areas (CD4 and CD8), proliferating cells in the germinal centers (Ki--67) and well organized endothelial cells lining the blood vessels (CD31) within this immune tissue. Use of this same panel to stain an aggressive form of breast cancer, triple negative breast cancer (TNBC, Figure 2B), shows infiltration of T cells (CD4 and CD8), significant cellular proliferation (Ki--67), and a high degree of angiogenesis (CD31). The latter is shown by the numerous small spheroid budding structures stained by CD31 within the tumor tissue.

Figure 2.

UltraPlex 4-plex staining panel utilized on two types of tissue. CD4 (green), CD8 (yellow), Ki--67 (purple), and CD31 (red) 4-plex images of (A) tonsil tissue and (B) triple negative breast cancer tissue (TNBC). CD4 and CD8 positive T cell subtypes are visible in both images, as are proliferating Ki67 positive cells, either in the germinal centers of tonsil tissues or throughout the tumor tissue. Note CD31 staining of fully formed vesicles in the tonsil, compared to the punctate CD31 staining on TNBC indicating widespread angiogenesis.

The identification of cells co--expressing two or three markers further demonstrates the capabilities of the UltraPlex mxIF technology (Figure 3). Here, melanoma tissue was stained with a 4--plex panel detecting CD4, CD8, FoxP3, and T--bet leading to identification of eight phenotypes including CD4+ FoxP3+, CD4+ T--bet+, CD8+ T--bet+ and CD4+ FoxP3+ T--bet+ co--expressors. Each of these phenotypes represents an immune cell type with a unique biological function.

Figure 3:

UltraPlex single cell phenotyping. 4-plex staining of immune cell markers CD4 (purple), CD8 (green), FoxP3 (yellow), and T-bet (magenta) on melanoma tissue (left) allows identification of CD4+ FoxP3+, CD4+ T-bet+, CD8+ T-bet+, and CD4+ FoxP3+ T-bet+ co-expressing cells (right). Each of these phenotypes represents a unique immune cell function.

As a simple approach to allow rapid high multiplexing, multiple 4--plex panels can be applied to serial FFPE tissue sections. The images from two or more sections are digitally aligned and the data combined into a composite image, yielding virtual images with 8 or more antigens. As an example (Figure 4), a digital 8--plex image of tumor tissue with immune infiltrate was produced by overlaying serial sections stained with two 4-plex MxIF panels: a tumor cell panel including Her2, ER, PR, and Ki67; and a TIL panel including CD3, CD4, CD8, and CD20. The 4-plex images are used to identify cellular subsets, and the 8-plex image allows visualization of cell subsets across tissue.

Figure 4:

Digital 8-plex image of tumor tissue obtained by overlaying images of 4-plex stained tissue sections. 4-plex staining of serially cut tissue sections reveals tumor cell antigens (left, 4-plex A) including Her2 (red), ER (blue), PR (green), and Ki67 (magenta) or TIL antigens (right, 4-plex B) including CD3 (cyan), CD4 (light green), CD8 (orange), and CD20 (light magenta). The 4-plex images are then digitally overlaid to create a digital 8-plex (center), allowing visualization of both tumor cells and TIL subsets across the tumor tissue.

2. MATERIALS

The UltraPlex staining procedure requires standard molecular biology laboratory materials, including adjustable volume pipettes, volumetric glassware, balance, pH meter, mixing apparatus, purified water, and a ventilated chemical fume hood. Utilize appropriate personal protective equipment as needed during staining procedure. Dispose of all used materials in accordance with laboratory regulations.

2.1. UltraPlex Fluorescent Immunohistochemistry

1. Tissue slides: Tissue may be prepared as either formalin fixed paraffin embedded (FFPE) sections (see Note 1) or frozen sections (see Note 2). Prepared tissue slides can be mounted with either 1 or 2 tissue sections of up to 1 cm diameter each. Alternatively, tissue microarrays may be used. Tissue sections should be cut to 3 -- 5 μm thickness and evenly spaced across slide surface. Preferably, tissue will be selected from serially cut sections, enabling digital image overlay. All tissue should be mounted on positively charged slides for enhanced adherence, preventing sample loss during staining process. Glass slides with ‘frosted’ labeling ends are preferred for sample identification, marked by the researcher using a solvent--resistant pencil.

2. Glass slide staining containers: with slide holder inserts, solvent--resistant, volume capacity ≥ 200 mL.

3. Deparaffinization reagent: xylene, ACS grade, ≥ 98%. Xylene is flammable and hazardous by skin contact or vapor inhalation; see Note 3 for safety considerations. Alternatively, if using autostainer, use deparaffinization reagent provided by autostainer manufacturer.

4. Ethanol: histology grade, ≥ 95%. Ethanol is flammable; use appropriate storage and working conditions.

5. Purified water: Distilled or deionized water (dH2O). For some steps, tap water may be acceptable (see Note 4).

6. Heat induced epitope retrieval (HIER) buffer: Select UltraPlex kits will include HIER buffer 10X concentrate. For most applications, 1X HIER buffer consists of 10 mM sodium citrate, pH 6.0. Add 2.94 g sodium citrate tribasic dihydrate and 0.5 mL Tween--20 to 800 mL of dH2O and use a magnetic stirrer to mix until dissolved. Adjust pH to 6.0. Add dH2O to bring volume to 1000 mL. HIER buffer can be stored for up to 3 months at 4 °C, but pH should be checked prior to each experiment if stored solution is used. 10X HIER buffer concentrate may also be prepared if desired, and then diluted to 1X and pH checked just prior to experiment. If using autostainer, use HIER buffer provided by autostainer manufacturer. If researchers provide 1° Abs, appropriate HIER buffer should be empirically determined for optimal epitope detection of the provided 1° Abs.

7. Heat--resistant plastic “Coplin” style slide jars, buffer volume capacity ≥ 50 mL.

8. Pressure cooker or histology steamer capable of maintaining a constant temperature of ≥ 120 °C with interior capacity sufficient to hold at least two of the Coplin jars described above.

9. 10% neutral buffered formalin: necessary only if using frozen unfixed tissue. Formalin is toxic; use appropriate storage and working conditions.

10. Phosphate buffered saline (PBS) buffer 1X: 137 mM sodium chloride (NaCl), 2.7 mM potassium chloride (KCl), 10 mM sodium phosphate dibasic (Na₂HPO₄), 1.8 mM potassium phosphate monobasic (KH₂PO₄), pH 7.4. Many premixed commercially available PBS formulations are suitable. If 10X PBS concentrate is obtained, it should be diluted to 1X using dH2O prior to use.

11. Wash buffer: Select UltraPlex kits will include 10X wash buffer concentrate. Generally, a suitable 1X wash buffer can be prepared as 0.2% Tween--20 in PBS pH 7.4. For 1 L of 1X wash buffer, add 2 mL Tween--20 to 998 mL PBS. Wash buffer should be prepared fresh and not stored. Alternatively, if using autostainer, use wash buffer provided by autostainer manufacturer.

12. Hydrophobic "PAP" slide pen.

13. Blocking buffer: Provided with UltraPlex kit at working concentration. Consists of 3% normal rabbit serum and 0.1% Triton--X--100 in PBS.

14. Primary antibodies (1° Abs): UltraTag--conjugated 1° Abs, provided with UltraPlex kit. Each of the UltraTag--1° Abs is provided as a 100X solution. Dilute 100--fold into antibody diluent buffer just prior to use in a volume suitable for applying 150 μL per tissue section. Alternatively, researcher--provided 1° Abs may be used instead of Cell IDx 1° Abs, but these must be first conjugated to UltraTag barcodes; for more information, see Note 5.

15. Secondary antibodies (2° Abs). Anti--UltraTag rabbit monoclonal antibodies conjugated to spectrally distinct 'CL' series fluorophores, provided with UltraPlex kit. Each 2° Ab staining solution is provided as a 100X concentrate. Dilute 100--fold into antibody diluent buffer just prior to use in a volume suitable for applying 150 μL per tissue section. User--provided 2° Abs are not compatible with the UltraPlex system.

16. Calibration microspheres: UltraTag barcoded, uniformly sized 6 μm particles suitable for microscopy imaging, provided in solution for labeling with anti--UltraTag fluorescent 2° Abs alongside tissue staining experiment. Calibration microspheres are an optional UltraPlex kit component and may also be purchased separately as needed.

17. Antibody diluent buffer: Provided with UltraPlex kits at working concentration. Consists of 1% bovine serum albumin and 0.2% Tween--20 in PBS.

18. Slide staining tray with water reservoir for humidification and light--blocking lid.

19. Mounting medium: Select UltraPlex kits will include mounting medium. In general, an aqueous antifade mounting media including DAPI nuclear counterstain is suitable. It is critical to use an aqueous mounting medium without solvents or phenylenediamine, which can quench UltraPlex fluorophores. For example, VectraShield and PolyMount mounting media are incompatible with UltraPlex assay fluorophores.

20. Cover glass: 24 x 50 mm. Other lengths of cover glass may be used; however, the cover glass must be able to completely cover the tissue specimen. A cover glass thickness of 0.13 – 0.19 mm (designated #1 or #1.5) is recommended for use with many standard microscope objective lenses.

21. Clear nail polish for permanently sealing edges of coverslip, if desired.

22. Slide storage boxes.

2.2. UltraPlex Slide Imaging

UltraPlex stained tissue slides may be imaged using a conventional widefield fluorescent microscope or a digital fluorescent slide scanner. For comments on use of a confocal fluorescence microscope, see Note 6.

For imaging UltraPlex slides using a widefield fluorescent microscope, the following system components are recommended:

1. Inverted widefield fluorescence microscope with mechanical X--Y stage, ocular lenses for specimen viewing, multiple objective turret, brightfield illumination, port for external fluorescence excitation lamp, and port for digital camera mounting.

2. Light source capable of fluorophore excitation within a range of approximately 300--800 nm. Mercury, metal halide, or LED light sources may provide acceptable excitation.

3. Fluorescence filter sets for 4--plex assay (Table 1):

Table 1.

Commonly used fluorescence filter sets compatible with the CL fluorophores used in the UltraPlex MxIF assay. Wavelength units shown are in nanometers (nm). These and comparable filter sets are commercially available from several companies.

Fluorophore	Fluor excitation	Fluor emission	Filter Set Name	Excitation Filter	Bandpass Filter	Emission Filter
CL490	491	515	EGFP/FITC	470/40	495	525/50
CL550	550	565	Cy3/TRITC	545/25	565	605/70
CL650	655	676	Cy5	620/60	660	700/75
CL750	759	780	Cy7	710/75	760	810/90
DAPI	358	461	DAPI	350/50	400	460/50

Additional filter sets and fluorophores will be required if performing 5-- or 6--channel custom UltraPlex assays, or if any filter set listed above is not available to the researcher. (see Note 7).

• 4.Objective lens(es) with 4x--100x magnification as desired. If available, 'Apochromat' objectives are preferred for high compatibility with multiple channel fluorescence and excellent optical flatness.
• 5.High resolution monochrome digital camera. Cooled CCD or CMOS cameras have both been used with success.
• 6.Computer workstation running image acquisition software.

For imaging UltraPlex slides using a digital slide scanner, components should consist minimally of the materias below. For component specifications of a commercial digital slide scanner validated in the laboratory by Cell IDx for scanning UltraPlex slides, see Note 8.

• 7.Automated slide loading compartment with ≥ 8 slide capacity.
• 8.Objective lenses of 1.25x--40x.
• 9.Light source capable of fluorophore excitation within a range of approximately 300--800 nm.
• 10.Emission filter sets compatible with CL flurophore excitation/emission spectra.
• 11.High resolution camera with multi--megapixel sensors for a large field of view and a wide dynamic range.
• 12.Computer workstation running image acquisition software capable of opening and viewing microscopy images generated by digital slide scanner. Software provided by scanner manufacturer is generally preferable.

The following components are required for digital image overlay and analysis:

• 13.Technical software for image processing and analysis. ImageJ--Fiji (NIH) freeware is recommended, as described in Section 3.5.
• 14.Large capacity (≥ 1 terabyte) external storage drive for storing and transporting digital image files (optional, but recommended).

3. METHODS

3.1. UltraPlex Fluorescent Immunohistochemistry Manual Staining Procedure

This procedure begins with several steps for preparation of FFPE tissue for staining. If frozen tissue is preferred, omit Steps 1--5 below and instead perform the brief steps described in Note 9. Then, complete the procedure below, starting at Step 6. This procedure is written for staining of 10 slides with 1 tissue section per slide. Each tissue section is treated with 150 μL per reagent unless otherwise indicated. Researchers may scale volumes and quantities accordingly depending on size of tissue sections and number of slides per assay. Control tissue slides should be included with all experiments (see Note 10).

1. Prepare 100 mL of HIER buffer and verify pH = 6.0. Add 50 mL of prepared HIER buffer to each of 2 heat--resistant plastic Coplin jars. Loosely affix the jar lids. Place containers into pressure cooker or histology steamer apparatus. Preheat apparatus to ≥ 120 °C. This step serves to preheat the HIER buffer, improving heat--induced epitope retrieval.

2. In a chemical fume hood, fill a glass slide staining container with 200 mL xylene. Deparaffinize tissue by immersing slides 2 X 5 min in xylene, using fresh xylene for the second immersion step. Transfer used xylene into a labeled and sealed solvent waste container for disposal.

3. Fill glass slide staining container with 200 mL of 95% ethanol. Immerse slides 2 min. Perform subsequent 2 min immersions in 200 mL each of descending ethanol concentrations at 90% (2 min), 70% (2 min), and 50% ethanol (2 min). Use dH2O to dilute 100% ethanol to desired final percentage.

4. Fill a clean glass slide staining container with dH2O. Immerse slides for 2 X 5 min, using fresh dH2O for the second immersion step.

5. Temporarily remove plastic Coplin jars from heated antigen retrieval apparatus using heat--resistant gloves. Evenly distribute slides into preheated plastic Coplin jars containing HIER buffer and return containers to heated apparatus. Steam for 15 min. Power off the apparatus and cool slides within for 10 min. Release pressure from apparatus, if using a pressure cooker. Using heat resistant gloves, remove slide containers and place on benchtop.

6. Transfer slides to a staining container filled with 200 mL wash buffer and rest for 20 min on benchtop.

7. Fill humidified staining tray with dH2O to level recommended by tray manufacturer.

8. Remove slides from wash buffer. Remove excess slide moisture by blotting the edge of glass with absorbant lens paper (e.g. “Kimwipes”) or lint--free paper towels. Use caution to avoid contact with tissues.

9. Transfer slides to elevated supports within humified staining tray, at least 1 cm above the level of the water in tray basin.

10. Allow slides to briefly air dry (1 min). Excessive moisture on slide glass may interfere with application of hydrophobic pen liquid.

11. Encircle each tissue section with hydrophobic pen. Allow to dry for 1 min.

12. Pipet 150 μL of blocking buffer per tissue section. Cover humidified tray. Block slides 20 min.

13. Prepare the 1° Ab staining solution. Into 1.5 mL antibody diluent buffer, pipet 15 μL of each 100X 1° Ab concentrate. Mix by pipetting and place tube on ice.

14. Remove blocking buffer by tilting each slide to a 90° angle on a paper towel. Place slides flat on elevated supports in humidity tray.

15. Pipet 150 μL of the diluted 1° Ab staining solution onto each tissue section. Place lid on tray. Incubate for 60 min.

16. Fill a clean glass slide staining container with 200 mL of wash buffer. Load slides into container insert, slowly immerse in container, and incubate for 3 min. Repeat 3 times, adding fresh wash buffer each time.

17. Prepare the 2° Ab staining solution. Into 1.5 mL antibody diluent buffer, pipet 15 μL of each 100X 2° Ab concentrate. Mix by pipetting and place tube on ice.

18. After Step 16 is concluded, tip off excess wash buffer onto paper towels. Place slides flat in staining tray.

19. Pipet 150 μL of the diluted 2° Ab staining solution onto each tissue section. Place lid on tray. Incubate for 60 min.

20. Wash slides as described in Step 16.

21. Add 200 mL of dH2O to a clean glass slide staining container. Incubate slides 3 min.

22. Align slides in staining tray and apply 2 drops of Fluoroshield^™ plus DAPI to each slide, ensuring tissue is covered by media. Incubate for 5 min in the staining tray with the light--blocking lid in place.

23. Apply cover glass to each slide, avoiding the introduction of air bubbles to the specimen under the glass.

24. Allow mounting media to harden by letting slides sit for at least 1 hour at room temperature before imaging. Use light--blocking tray lid or foil to protect from ambient light. Additionally, clear nail polish may also be applied to the borders of the cover glass for immobilization.

25. When slides are dry, load into slide box for transport to imaging and long--term storage.

26. Proceed to imaging.

3.2. UltraPlex Fluorescent Immunohistochemistry Automated Staining Procedure

This section of the protocol summarizes the use of UltraPlex reagents within an established autostaining method. This methods employs FFPE tissue and a Leica Bond Rx automated research stainer. Other autostainers have been reported to work with UltraPlex reagents, such as the Ventana Discovery.

Manufacturer protocol dewaxing and epitope retrieval functions are run on the Bond Rx instrument, followed by a custom program for blocking and antibody incubation optimized for UltraPlex staining. For more detailed information on routine use of the Leica Bond Rx, please see manufacturer user manual.

1. Load required premixed solutions into appropriate reservoirs in Bond Rx instrument, including Leica Bond Dewax Solution(s), Bond Epitope Retrieval Solution 1/2, and Bond Wash Solution. If these solutions already exist in the instrument reservoirs, verify solutions are fresh and that volume levels are sufficient to run desired number of UltraPlex slides.

2. Add dH2O to appropriate reservoir.

3. Add blocking buffer to appropriate reservoir.

4. Prepare 1° Ab and 2° Ab UltraPlex antibody solutions diluted to 1X in desired amount of antibody diluent buffer. Add 300 μL to total preparation volume to account for dead volume in solution reservoirs of autostainer.

5. Transfer 1° Ab and 2° Ab solutions to appropriate instrument reservoirs.

6. Load FFPE slides into Bond Rx instrument.

7. Run standard 'Dewax' function using program [*D].

8. Run standard 'HIER' function using program [ER2*H2(20)].

9. Run UltraPlex IF function using custom program as shown in Table 2 below:

10. At the conclusion of the Bond Rx autostaining protocol, remove slides from instrument. Proceed to tissue mounting procedure at Section 3.1, Step 21.

Table 2.

Custom program for automated UltraPlex MxIF slide staining using Leica Bond Rx autostainer.

Step	Reagent	Step Type	Time (Min:Sec)	Temp	Dispense Vol (μL)
1	dH2O	Reagent Incubation	0:00	Ambient	150
2	Bond Wash	Wash Step	0:10	Ambient	150
3	Bond Wash	Wash Step	0:10	Ambient	150
4	Blocking Soln	Reagent Incubation	20:00	Ambient	150
5	1' Ab	Reagent Incubation	60:00	Ambient	150
6	Bond Wash	Wash Incubation	3:00	Ambient	150
7	Bond Wash	Wash Step	3:00	Ambient	150
8	Bond Wash	Wash Step	3:00	Ambient	150
9	2' Ab	Reagent Incubation	60:00	Ambient	150
10	Bond Wash	Wash Step	3:00	Ambient	150
11	Bond Wash	Wash Step	3:00	Ambient	150
12	Bond Wash	Wash Step	3:00	Ambient	150

3.3. Preparation of UltraTag Calibration Microspheres

It is most efficient to perform this brief procedure concurrently with 2° Ab labeling of tissues as described in Section 3.1, Step 19.

1. Prepare 100 μL of each anti--UltraTag 2° Ab by adding 1 μL of 100X concentrate to 99 μL of antibody dilution buffer.

2. Briefly vortex each tube of UltraTag microspheres.

3. Add 10 μL of each UltraTag microsphere suspension to corresponding anti--UltraTag 2° Ab solution. Mix by pipetting.

4. Incubate 60 min at ambient temperature with gentle mixing.

5. Centrifuge tubes for 5 minutes at 2000 x g at ambient temperature.

6. Carefully remove supernatant.

7. Wash microspheres by adding 1 mL of PBS and pipetting to mix.

8. Centrifuge as described in Step 5 above.

9. Resuspend in 100 μL PBS.

10. Pipet 100 μL of microsphere suspension onto standard glass slides. Use a separate slide for each microsphere suspension.

11. Apply coverglass.

12. Remove excess moisture from slide surface.

13. Seal coverglass edges with clear nail polish.

14. Allow slides to dry in the dark alongside UltraPlex stained tissue slides until imaging session.

3.4. UltraPlex Slide Imaging Procedure

Whether utilizing a conventional widefield fluorescent microscope or digital slide scanner, the workflow is similar for imaging UltraPlex stained slides. Order the slides such that the first in the group are the calibration microspheres, followed by the positive tissue antigen controls, the unstained and the 2° Ab--only tissue controls, and lastly the experimental tissue slides.

1. Power on fluorescence excitation lamp, microscope components, digital camera, and computer workstation.

2. Open image acquisition software.

3. Clean residual mounting media from slides, if necessary, using lens paper and a small amount of water or mild glass cleaner.

4. Place slides into autoloader of digital slide scanner, if applicable, according to manufacturer’s instructions.

5. Position first slide on microscope stage with coverglass facing objective lens.

6. Rotate desired magnification objective into place under slide (typically 10--40x).

7. Move the DAPI filter set into place.

8. Using the camera software, focus the calibration microspheres.

9. Adjust the DAPI signal exposure time such that the majority (> 90%) of microspheres are visible, but not exhibiting signal overexposure. Record the optimal signal exposure time in milliseconds.

10. Move the CL490 filter set into place.

11. Adjust the CL490 signal exposure time such that the majority (> 90%) of microspheres are visible, but not exhibiting signal overexposure (Figure 5). Record the optimal signal exposure time in milliseconds.

12. Proceed to the next calibration slides for CL550, CL650, and CL750, repeating Steps 10-11 for each fluorescent channel. This will calibrate the microscope for tissue imaging. Then, proceed to tissue slides.

13. Move the first antigen positive tissue slide into place on the microscope stage.

14. Using the camera software, observe the DAPI signal at low magnification to focus the tissue.

15. Proceed to image all slides using the calibrated exposure times determined in steps 9--12.

16. If using a standard microscope, capture at least 3 regions of interest (ROI) per tissue section. If using digital scanner, capture at least 80% of tissue area. Images taken in each fluorescent channel must be recorded for each ROI or tissue section. If performing serial section alignment as described in Section 3.4, imaged tissue ROIs must be aligned using distinct features of tissue morphology known to be present across serially cut sections, such as ducts, crypts, or other macrostructures.

17. Save each acquired image as a separate, 16--bit grayscale TIFF file, unless otherwise mandated by the acquisition software. JPEG or otherwise compressed files are not recommended due to loss of important pixel data by compression algorithms.

18. Keep a record of the image file names while saving the TIFF files. It is crucial to be able to properly match image files with their corresponding tissues after the imaging session is complete. The fluorescent channel of each image should also be noted.

19. After saving the TIFF files to a hard drive location, copying the TIFF files to an external drive or server is recommended, as the file set may be quite large (gigabytes) and use a significant amount of hard drive space. The external storage location also allows data portability.

Figure 5.

UltraTag calibration microspheres visualized at 20X magnification using a widefield fluorescent microscope, here shown labeled with anti-UltraTag CL490 2° Ab. Microspheres allow calibration of instrument settings and subsequent evaluation of tissue antigen expression levels across fluorescent channels.

3.5. UltraPlex Digital Image Overlay Procedure

The following procedure describes how to prepare a digital 8--plex image using two 4--plex images as input. The 4--plex images should be derived from serially cut tissue sections to ensure biological relevance of signal colocalization. The digital image overlay procedure can be scaled up as needed, i.e. to 12--plex (or higher). The workflow described below is for ImageJ--Fiji 2019 and should be updated by the user as needed for compliance with future versions of the software.

1. Open ImageJ--Fiji software on computer workstation.

2. Perform digital alignment of tissue images:

   1. Install .jar files for plugins "StackReg" and "TurboReg" [8, 9] in the ImageJ--Fiji plugins folder.
   2. Open TIFF files in ImageJ--Fiji.
   3. Combine images into a stack using the menu commands Image → Stacks → Images To Stack.
   4. Convert stack to an 8--bit file. (Image → Type → 8--bit)
   5. Run plugin "StackReg". Follow plugin menu prompts to perform alignment. Adjust parameters to achieve optimal alignment across series of images.
   6. Save image stack as a new TIFF file for record keeping purposes. The TIFF stack is then used in several applications in the following steps.
3. Utilize UltraPlex 4--plex images to identify cell types of interest:

   1. Convert the TIFF stack prepared above to individual images using Image → Stacks → Stack to Images.
   2. Select 4 images from one ROI with staining of cell marker panel of interest (for example, tumor cell markers or lymphocyte subset markers).
   3. Create a pseudocolored 4--plex composite image using the selected images using Image → Color → Merge Channels and completing the menu prompts. The resulting image will enable identification and quantitation of cell types displayed in the 4--plex image (Figure 4, left and right panels).
   4. Repeat steps a--c for each ROI and 4--plex panel desired.
   5. Save each composite image as a new TIFF for analysis and record keeping.
   6. Perform quantitation on 4--plex images. For example, total counts of cell subsets, distance between types of immune cells and tumor cells, or percent of cells exhibiting a specific marker. Many detailed procedures for biological image quantitation using ImageJ--Fiji can be found in [5-7] and elsewhere.
4. Overlay UltraPlex 4--plex images to create 8-- or 12--plex (or more) images:

   1. Open 4--plex composite TIFFs desired for overlaying purposes.
   2. Select the first 4--plex composite window.
   3. Overlay the second 4--plex composite window using Image → Overlay → Add Image… and completing the menu prompts to create the overlaid image. The resulting image should appear comparable to Figure 4, center panel.
   4. Repeat step c to overlay additional 4--plex composites as desired.
   5. Flatten finalized composite image using Image → Overlay → Flatten.
   6. Save each flattened image as a TIFF for further analysis (if necessary) and record keeping purposes.

4. NOTES

1. FFPE tissue slides should be prepared from FFPE tissue blocks according to conventional procedure [10]. Tissue sections should be cut to a thickness of 5 μm and affixed to positively charged slides with a 'frosted' labeling area (e.g. FisherBrand "SuperFrost Plus"). Following standard FFPE slide preparation, FFPE tissue slides should be oven--baked at 60 °C for up to 2 hours to ensure adherence of tissue sections to slides. FFPE slides may be stored at ambient temperature. Prior to staining, inspect slides with brightfield microscopy to assess tissue embedding integrity. Tissue should appear uniformly embedded within paraffin and affixed to the slide surface. Tissue without this appearance may be subject to partial or complete dissociation of the tissue from the slides, and therefore, an abrogation of the staining process. If tissue slides pass visual inspection, proceed to staining.

2. Frozen tissue sections are generally cut from a frozen tissue block embedded in O.C.T. cryoprotectant [10]. It is necessary for sections and tissue blocks to be stored at −80°C to prevent ice crystal formation. Once tissue is thawed, it must be used immediately to prevent cellular autolysis. If tissue has been previously fixed, either by immersion or perfusion, HIER may be desired, as determined by the researcher. If frozen tissue is unfixed, a short fixation step is recommended (e.g. in 10% neutral buffered formalin as described in Note 9). A fixation step helps to ensure epitope recognition by the 1° Abs used in the UltraPlex assay, which are typically optimized for detection of fixed epitopes in FFPE tissue.

3. Xylene is hazardous by inhalation or skin contact. Always work with xylene in a ventilated chemical fume hood wearing protective gloves and eyewear. Xylene is also highly flammable; keep away from heat sources. Store in flammable--safe cabinet. For disposal, place in a labeled and sealed hazardous waste container. Do not mix with other waste chemicals. Never dispose of xylene in laboratory sinks.

4. While purified dH2O is generally preferred for all molecular biology procedures, tap water is often used for routine histology wash steps in research and clinical labs. If necessary, a cold tap water bath can be created in a clean laboratory sink for slides immobilized in an immersion rack. The tap water stream should not run directly over the slides, so that it does not disturb the tissue adhering to the slides. If desired, a tap water bath can be used after the ethanol rehydration step (Section 3.1, Step 3), and/or just prior to mounting slides (Section 3.1, Step 21), in place of the noted dH2O steps. Tap water may also be used to fill the humidified staining tray. However, for buffer preparation, tap water is not recommended and dH2O should be used.

5. An UltraTag barcode conjugation kit is available from Cell IDx that enables rapid (< 3 hr) conjugation of UltraTags to user--provided 1° Abs. The kit is resin spin--column based and requires only basic laboratory equipment (pipets, microcentrifuge).

In brief, the materials for this procedure include:

1. Desalting and buffer exchange spin columns

2. Conjugation buffers

3. UltraTag barcodes in solution

4. User--provided 1'Abs, ≥ 50 μg at 1.0 ± 0.1 mg/mL

The conjugation procedure briefly involves:

1. 1° Ab preparation for conjugation (1 hr)

2. Conjugation reaction (1 hr)

3. Conjugate purification (10 min)

A detailed protocol and kit purchasing information are available at http://www.cellidx.com.

• 6.Confocal microscopes have not yet been extensively validated with the UltraTag multiplex technology. Widefield microscopy has been initially used as it enables more sensitive detection of fluorescent signals (many confocal microscopes compromise signal brightness in order to resolve a single focal plane). However, a confocal microscope with similar system components to the widefield microscope described in Section 2.2 would likely be effective. Because signal intensity is an issue, confocal microscopes are equipped with high powered excitation lasers rather than a typical broad spectrum excitation lamp used for widefield microscopy. If using excitation lasers for UltraPlex imaging, the researcher should identify the optimal excitation laser for each fluor used in the UltraPlex assay. For example, for CL490, a 488 nm excitation laser would be appropriate, while for CL650, a 633 or 640 nm laser would likely be effective.
• 7.The procedure described herein utilizes a 4--plex fluorescent panel consisting of CL490, CL550, CL650, and CL750. Using these CL fluors with the appropriate filter sets shown in Table 1 results in excellent fluorescent channel specificity, with extremely low (if any) spectral crosstalk across fluorescent channel. This feature obviates the need for complex spectral unmixing procedures to correct spectral crosstalk. If a higher multiplexed assay is desired, a 6--plex panel has been validated by Cell IDx using narrow bandwidth excitation and emission filters (see Table 3). If narrow bandwidth filter sets are not available to the researcher, to enable the 6--plex a spectral unmixing procedure must be conducted to correct for spectral crosstalk, for example as described in [3].
• 8.
  Component specifications for the Leica Aperio Versa digital slide scanner:

  1. Objectives: 1.25x HC PL Fluotar 0.04 NA; 5x HC PL Fluotar 0.15 NA; 10x HC PL Fluotar 0.32 NA; 20x HC PL APO 0.80 NA; 40x – HC PL APO 0.95 NA.
  2. Light source: X--Cite series 120PCQ high--pressure metal halide arc lamp.
  3. Fluorescence filter sets (Table 3):
  4. Camera: Zyla 5.5 sCMOS (Andor Technology)
  5. Imaging software: Aperio ImageScope (Leica)
• 9.
  This note briefly summarizes steps specific to the preparation of unfixed frozen tissue sections for the UltraPlex assay. Perform the steps below instead of Steps 1--5 in Section 3.1, and then continue the Section 3.1 protocol at Step 6.

  1. Remove slides from freezer. Thaw 10 min on benchtop. Transfer to a staining container filled with 200 mL wash buffer for 5 min.
  2. In a chemical fume hood, fill a staining container with 200 mL of 10% neutral buffered formalin. Immerse slides for 10 min. Discard used formalin in a labeled waste container.
  3. Proceed to Step 6 of Section 3.1.

  If previously fixed frozen tissue is to be used, the researcher should determine if an HIER procedure is necessary to render fixed epitopes accessible to 1° Abs. If HIER is desired, follow the HIER procedure in Section 3.1 (Steps 1 and 5) and then continue through the protocol from Step 6 onward. If HIER is not desired, simply start the Section 3.1 staining procedure at Step 6.

• 10.
  It is recommended to include in each experiment the following control tissue sections:

  1. Tissue antigen positive controls for each antigen, one for each fluorescent channel.
  2. Unstained control to evaluate tissue autofluorescence.
  3. 2° Ab only control to evaluate possible nonspecific binding of 2° Ab (nonspecific binding of anti--UltraTag 2° Abs is usually minimal on most tissues).

Table 3.

Filter sets used for a 6--plex UltraPlex MxIF assay using a commercially available fluorescent digital slide scanner (Leica Aperio Versa). Wavelength units shown are in nanometers (nm). These and comparable filter sets are commercially available from several companies.

Fluorophore	Fluor excitation	Fluor emission	Excitation Filter	Bandpass Filter	Emission Filter
CL490	491	515	485/25	505	525/30
CL550	550	565	546/22	565	590/33
CL650	655	676	635/30	660	680/30
CL700	707	730	695/30	720	740/35
CL750	759	780	750/20	770	800/50
DAPI	358	461	350/50	400	460/50