Multiple Immunofluorescence Labeling of Formalin-Fixed Paraffin-Embedded Tissue

Protocol
Part of the Methods in Molecular Biology book series (MIMB, volume 724)

Abstract

Multiple immunofluorescent labeling of formalin-fixed paraffin-embedded (FFPE) tissue is not a routinely used method. At least in part, this is due to the perception that the innate autofluorescence of the FFPE material forbids the use of immunofluorescent labeling. As a result, immunohistochemical (immunoperoxidase) staining of FFPE material or cryosectioning methods is used instead. In this chapter, we describe a robust optimized method for high-resolution immunofluorescence labeling of FFPE tissue that involves the combination of antigen retrieval, indirect immunofluorescence, and confocal laser scanning microscopy. Once such samples have been prepared and imaged by confocal microscopy, they can be stored at −20°C for extensive periods (>250 days) and reexamined with minimal loss of quality. As a consequence, this method has the potential to open up the large archival sample collections to multiple immunofluorescent investigations.

Key words

Immunofluorescent labeling Formalin-fixed paraffin-embedded FFPE Confocal microscopy Antigen retrieval Pathology 

1 Introduction

Immunohistochemistry (IHC) is one of the pillars of modern diagnostic pathology and a fundamental research tool in both pathology and translational research laboratories. Currently, the most commonly used method is to stain the section with a primary antibody followed by a peroxidase-conjugated second layer antibody and development with a chromogenic substrate. While this method is robust and reliable, it is mainly limited to only revealing one protein at a time. There is, therefore, a need for a method that would allow multiple labeling of proteins in FFPE material. Recently, we described an optimized method for multiple immunofluorescent labeling of FFPE human tissue (1). Subsequently, we have extended this method for mouse tissue and in addition have tested out some variations to the original method. These additions to the method are highlighted in this chapter together with additional notes to help those using this method for the first time. We, and others, now routinely use this method for examining human tissue samples (2, 3) and experimental samples from in vivo studies (4, 5), and as a consequence we believe that this method will be of wide use to both pathologists and research scientists.

2 Materials

Unless otherwise stated, materials were stored at room temperature.

2.1 Tissue Fixation, Embedding, and Sectioning

  1. 1.

    10% Neutral-buffered formalin (Bios Europe, Lancashire, UK).

     
  2. 2.

    4% Paraformaldehyde (seeNote1). Prepared by heating the appropriate amount of paraformaldehyde in PBS on a stirring hotplate set at 150°C until the paraformaldehyde depolymerizes and goes into solution (2–3 h depending on volume being prepared). Once prepared, store the solution at room temperature.

     
  3. 3.

    Paraffin wax (Tissue-Tek; Sakura, Finland).

     
  4. 4.

    Tissue-Tek VIP automatic tissue processor (Sakura, Finland).

     
  5. 5.

    Superfrost plus glass slides (Cat No 631-0108, VWR International, Lutterworth, Leicestershire, UK).

     
  6. 6.

    Oxygen-free nitrogen gas (BOC gases, Guildford, Surrey, UK).

     

2.2 Antigen Retrieval

  1. 1.

    Xylene (Fisher Scientific, Loughborough, Leicestershire, UK).

     
  2. 2.

    Histoclear (Agar Scientific, Stanstead, Essex UK).

     
  3. 3.

    Ethanol (BDH, Poole, Dorset, UK).

     
  4. 4.

    Plastic slide rack and holder (Slide Staining System Easy Dip™ 720-0791VWR International Lutterworth Leicestershire, UK).

     
  5. 5.

    Target retrieval solution S1699 (Dako, Ely, Cambridgeshire, UK).

     

2.3 Immunofluorescent Labeling

  1. 1.

    ImmEdge pen (H-4000, Vector Laboratories, Peterborough, UK).

     
  2. 2.

    Phosphate-buffered saline (PBS, 10 mM phosphate buffer, 2.7 mM KCl, 137 mM NaCl, pH 7.4 Sigma-P4417-100TAB, Poole Dorset, UK).

     
  3. 3.

    Immunofluorescence buffer (IFF). PBS plus 1% BSA (A3059, Sigma, Poole, Dorset, UK) and 2% fetal calf serum (FCS). IFF is filtered through a 0.2 μm filter and stored in 20 ml aliquots at −20°C.

     
  4. 4.
    Primary antibodies (see Note 2). Details of the primary antibodies used in the figures are described in Figs. 1 and 2 and Table 1.
    Fig. 1.

    Expression of cytokeratin, α-smooth muscle actin and endosialin in normal human breast. 3 μm FFPE sections of human adult breast were stained for (a) DAPI (nuclear stain); (b) α-smooth muscle actin (αSMA); (c) endosialin; (d) wide spectrum cytokeratin; (e) merged image. αSMA is expressed by the myoepithelial cells and the pericytes closely associated with the endothelial cells in the vasculature. Luminal epithelial cells and non-pericyte stromal cells are αSMA negative. Endosialin expression is restricted to the stromal fibroblasts. The wide spectrum cytokeratin antibody detects the luminal cell cytokeratins and less efficiently the myoepithelial cell cytokeratins. Stromal cells are cytokeratin negative. Scale bar, 50 μm.

    Fig. 2.

    Expression of E-cadherin, muscle actin, and caveolin-1 in normal mouse small intestine. 3 μm FFPE sections of mouse adult intestine were stained for (a) DAPI (nuclear stain); (b) E-cadherin; (c) muscle actin (MAct); (d) caveolin-1 staining; (e) merged image. E-cadherin expression is restricted to the lateral membranes of the intestinal epithelial cells. Muscle actin and caveolin-1 are expressed by distinct stromal cell populations. Scale bar, 25 μm.

    Table 1

    Details of primary and secondary antibodies

    Antibody

    Species/Isotype

    Concentration

    Supplier

    Endosialin: B1/35.1

    Mouse IgG1

    4.0 μg/ml

    Isacke laboratory (6)

    α-Smooth muscle actin (αSMA)

    Mouse IgG2A

    0.88 μg/ml

    Sigma

    Cytokeratin: wide spectrum screening (WSS)

    Rabbit polyclonal

    1:500 dilution

    Dako

    Muscle actin (MAct)

    Mouse IgG1

    1.0 μg/ml

    Dako

    E-cadherin

    Mouse IgG2A

    0.25 μg/ml

    BD Biosciences

    Caveolin-1

    Rabbit polyclonal

    1.0 μg/ml

    Santa Cruz

    Alexa Fluor 555 anti-mouse IgG1

    Goat polyclonal

    2.0 μg/ml

    Invitrogen

    Alexa Fluor 488 anti-mouse IgG2A

    Goat polyclonal

    2.0 μg/ml

    Invitrogen

    Alexa Fluor 633 anti-rabbit Ig

    Goat polyclonal

    2.0 μg/ml

    Invitrogen

     
  5. 5.

    Alexa Fluor conjugated-secondary antibodies (Invitrogen, Paisley, UK) (see Note 3). Details of secondary antibodies used here are provided in Figs. 1 and 2 and Table 1.

     
  6. 6.

    DAPI (4′,6-diamidino-2-phenylindole; D21490, Invitrogen).

     
  7. 7.

    Vectashield (H-1000, Vector Laboratories).

     
  8. 8.

    Coverslips [22  ×  40 mm, 0.155–0.185 mm thickness (VWR International, Lutterworth Leicestershire, UK)].

     
  9. 9.

    Immersion oil (518F, Zeiss, Welwyn Garden City, Hertfordshire, UK).

     

3 Methods

3.1 Starting Material

Starting material is FFPE tissues. In our studies, these samples are typically (a) archival FFPE pathology specimens. Typically, these would have fixed overnight at room temperature in 10% neutral-buffered formalin before paraffin embedding or (b) human or rodent tissue samples that we have prepared. For this, small pieces of tissue 2–10 mm thick are fixed in 4% paraformaldehyde at room temperature with gentle tumbling for 1–16 h. Fixed material is then processed using a Tissue-Tek VIP automatic tissue processor with a standard 14 h protocol and embedded into paraffin wax.

3.2 Sectioning, Mounting onto Slides, and Slide Storage

  1. 1.

    Cut 3–4 μm sections from the embedded blocks and float onto a warm (42°C) water bath.

     
  2. 2.

    Pick sections from the water bath and place onto Superfrost plus slides.

     
  3. 3.

    Place slides into a vertical rack and dry overnight at 37°C in a fan-assisted cabinet.

     
  4. 4.

    The next day, label the slides and either use immediately or place into a container that is then purged with oxygen-free nitrogen gas. Store the slide containers at 4°C (seeNote4).

     
  5. 5.

    To carry out a labeling experiment, remove the required slides. Purge the container containing the remaining slides with oxygen-free nitrogen gas before replacing at 4°C.

     

3.3 Dewaxing and Antigen Retrieval

  1. 1.

    Load slides into glass or plastic slide racks.

     
  2. 2.

    Place the slide racks into a glass or plastic slide holder (seeNote5) and dewax by incubating for 2  ×  10 min in 100% xylene or Histoclear (seeNote6) with agitation about once a minute.

     
  3. 3.

    Rehydrate the dewaxed slides in ethanol as follows: 2  ×  10–20 s agitation in 100% ethanol, 1  ×  10–20 s agitation in 90% ethanol, 1  ×  10–20 s agitation in 70% ethanol, and 2  ×  10–20 s agitation in tap water.

     
  4. 4.

    Transfer the slide rack into a fresh slide holder containing 150 ml of Dako target retrieval solution (15 ml of stock into 135 ml H2O) that has been prewarmed in a 95°C water bath. Leave in the water bath for 30 min.

     
  5. 5.

    Remove the slide rack and holder from the water bath and leave on the bench at room temperature for 20 min.

     
  6. 6.

    Place slide rack under running cold tap water for 5 min.

     

3.4 Controls and Work-Up for Immuno­fluorescence Labeling

As for any multiple immunofluorescent labeling experiments, it is important to conduct the control and work-up experiments to ensure optimal immunolabeling.
  1. 1.

    Ascertain whether your primary antibody works on FFPE material (seeNote7). A good source of antibodies that will work in this method are ones that have already been worked up for IHC.

     
  2. 2.

    Titrate each primary antibody individually to ascertain the minimal working concentration. If the antibody has been titrated out for immunohistochemical staining of FFPE material, this will be a useful starting point for the immunofluorescence titration.

     
  3. 3.

    Wherever possible we combine the primary antibodies together for the first incubation and combine the fluorescent-conjugated secondary antibodies in the second incubation. However, it is essential to test that there is no cross reactivity of the antibodies (seestep 4 below). In addition, if the experiment includes a directly conjugated primary antibody, then these are incubated with the section after the indirect labeling has been completed.

     
  4. 4.

    Ensure that there is no antibody cross reactivity and minimal nonspecific labeling by undertaking the following controls (a) perform the staining reaction omitting each primary antibody one at a time but retaining all the fluorescent-conjugated secondary antibodies and (b) perform the staining reaction replacing the primary antibodies with the same concentration of isotype-matched Ig.

     

3.5 Immuno­fluorescence Labeling

  1. 1.

    Remove slides from slide rack, wipe around the section with a tissue to create an “island.” Carefully pipette 100–200 μl of PBS onto the section to prevent it from drying out.

     
  2. 2.

    Using an ImmEdge pen, draw a ring or rectangle around the “island.”

     
  3. 3.

    Shake the PBS off the slide and replace with 100–200 μl of IFF. Ensure that the whole island is wetted. If not, add an extra 200 μl of IFF and gently rock the slide until the whole island is wetted. Once the island has been wetted, shake off the IFF, and replace with 100–200 μl of IFF.

     
  4. 4.

    Place slides into a moist chamber at room temperature (seeNote8). Unless otherwise stated, all following incubations are at room temperature with gentle mixing on a rocking platform.

     
  5. 5.

    Incubate in 100–200 μl primary antibodies diluted in IFF for 60 min at room temperature or overnight at 4°C (seeNotes9 and 10).

     
  6. 6.

    Shake off the primary antibodies and replace with 100–200 μl PBS for 3  ×  5 min.

     
  7. 7.

    Shake of the PBS and incubate with 100–200 μl secondary antibodies diluted to 2 μg/ml in IFF for 60 min (seeNote10).

     
  8. 8.

    Wash 3  ×  5 min as described above using 100–200 μl PBS containing 1.43 μM DAPI (1:10,000 dilution of stock solution).

     
  9. 9.

    Rinse slides 1  × in PBS.

     
  10. 10.

    Shake off the PBS and mount the sections in Vectashield by adding 8–10 μl of Vectashield and lowering a coverslip onto the slide. Gently squeeze out the excess mountant and seal with clear nail varnish.

     
  11. 11.

    Once slides have been stained they can be examined immediately or stored at 4°C for up to 2 weeks. If it is envisaged that the slide will be examined a number of times over a long period, then storage at −20°C is highly recommended. We have reexamined slide that have been stored at −20°C for >250 days with minimal loss of quality (1).

     

3.6 Confocal Microscopy

To collect immunofluorescent images from FFPE sections is essential to use a confocal microscope. Details provided below are for the Leica SP2 confocal microscope that we have in our laboratory but this method should be adaptable for any confocal microscope that can collect images in sequential mode.
  1. 1.

    Visualize the stained slides using a suitable confocal microscope. In this study, a Leica SP2 confocal scanning microscope with the laser outputs controlled via the Acousto-Optical Tunable Filter (AOTF) and the four collection windows set using the Acousto-Optical Beam Splitter (AOBS) was used.

     
  2. 2.
    The settings used were as follows:
    • 403 nm Laser (25%) window 410–483 nm

    • 488 nm Laser (25%) window 493–538 nm

    • 543 nm Laser (100%) window 548–628 nm

    • 633 nm Laser (25%) window 638–700 nm.

     
  3. 3.

    We routinely collect images using a 20× dry lens (lens specification, HCPLAPOCS NA 0.7; Leica) at 1× zoom or with a 40× oil immersion lens (lens specification, HCXPLAPO NA 1.25; Leica) at 1× or 2× zoom. For the oil immersion lens, a small drop of immersion oil is placed in the center of the coverslip prior to imaging (seeNote11).

     
  4. 4.

    Collect confocal images using the microscope in sequential mode with a line average of 4 and a format of 1,024  ×  1,024 pixels.

     
  5. 5.

    The confocal microscope collects images in black and white and then assigns them a false color. Within the confocal software these colors can be reassigned (seeNote12).

     
  6. 6.

    In our laboratory, images are exported from the Leica confocal software into Adobe Photoshop CS2 v.9. In Figs. 1 and 2, no image manipulation was performed.

     

4 Notes

  1. 1.

    Although this fixative is usually referred to as paraformaldehyde, it is in fact formaldehyde prepared by depolymerizing paraformaldehyde.

     
  2. 2.

    Antibodies should be aliquoted and stored according to manufacturer’s instructions.

     
  3. 3.

    In our laboratory, we use almost exclusively Alexa Fluor-conjugated secondary antibodies but other commercially available conjugates are similar in performance. For example, FITC conjugates perform similar to Alex Fluor 488 conjugates. In our laboratory, we aliquot the secondary antibodies into 50 μl lots in 0.5 ml Eppendorf tubes, seal with parafilm and store at 4°C. A sensible precaution is to date the conjugates on arrival so that you know how long you have had them.

     
  4. 4.

    We have used cut sections that have been stored for 140 days in this way with no obvious loss of staining.

     
  5. 5.

    In our original description of this method (1), we used glass slide racks and holders. Subsequently, we have been using the plastic Easy Dip slide staining system that works extremely well.

     
  6. 6.

    In our original description of this method (1), we used xylene to dewax the slides. Subsequently, we have switched to using Histoclear. Both work equally well.

     
  7. 7.

    Antibodies which work by western blotting and immunofluorescence on paraformaldehyde-fixed material do not necessarily work on FFPE material. If testing out antibodies, for the first time, for their ability to work on FFPE material, we recommend identifying a suitable control tissue containing cell types that do and do not express the antigen. If such control tissue is not available, in our laboratory, we generate FFPE pellets of paired cell lines that do and do not express the antigen for antibody testing.

     
  8. 8.

    Moist chambers are dark or foil-covered plastic boxes with moist filter or chromatography papers in the base.

     
  9. 9.

    For staining overnight at 4°C the slides are not rocked. In our laboratory, we usually stain with the primary antibodies overnight at 4°C, but for some antibodies this incubation increases the level of nonspecific background staining. It is important to determine the optimal labeling conditions for each individual antibody.

     
  10. 10.

    As shown in this chapter, it is possible to stain with two monoclonal antibodies provided they are of different subclasses. Here we show examples of staining simultaneously with a mouse monoclonal IgG1 antibody, a mouse monoclonal IgG2A antibody, and a rabbit polyclonal antibody and detecting these with Alexa Fluor 555-conjugated anti-mouse IgG1, Alex Fluor 488-conjugated anti-mouse IgG2A, and Alexa Fluor 633-conjugated anti-rabbit Ig.

     
  11. 11.

    If slides are being stored at −20°C, any immersion oil remaining on the slide has to be carefully wiped away as it tends to creep over the slide during storage.

     
  12. 12.

    In double antibody labeling experiments, we normally depict the nuclear DAPI staining in blue. In triple antibody labeling experiments, we would normally depict the DAPI staining in gray. In cases where a series of experiments are being reported, the colors can be reassigned within the confocal software to maintain continuity.

     

Notes

Acknowledgements

This work was funded by Breakthrough Breast Cancer. We thank Jorge Reis-Filho, Kay Savage and Suzanne Parry for their help in developing this method.

References

  1. 1.
    Robertson, D., Savage, K., Reis-Filho, J. S. and Isacke, C. M. (2008) Multiple immunofluorescence labelling of formalin-fixed paraffin-embedded (FFPE) tissue. BMC Cell Biol. 9, 13.PubMedCrossRefGoogle Scholar
  2. 2.
    Simonavicius, N., Robertson, D., Bax, D. A., Jones, C., Huijbers, I. J. and Isacke, C. M. (2008) Endosialin (CD248) is a marker of tumor-associated pericytes in high-grade glioma. Mod Pathol. 21, 308–315.PubMedCrossRefGoogle Scholar
  3. 3.
    Pastrana, D. V., Tolstov, Y. L., Becker, J. C., Moore, P. S., Chang, Y. and Buck, C. B. (2009) Quantitation of human seroresponsiveness to Merkel cell polyomavirus. PLoS Pathog. 5, e1000578.PubMedCrossRefGoogle Scholar
  4. 4.
    Kendrick, H., Regan, J. L., Magnay, F. A., Grigoriadis, A., Mitsopoulos, C., Zvelebil, M., et al. (2008) Transcriptome analysis of mammary epithelial subpopulations identifies novel determinants of lineage commitment and cell fate. BMC Genomics 9, 591.PubMedCrossRefGoogle Scholar
  5. 5.
    Kosaka, N., Ogawa, M., Choyke, P. L., Karassina, N., Corona, C., McDougall, M., et al. (2009) In vivo stable tumor-specific painting in various colors using dehalogenase-based protein-tag fluorescent ligands. Bioconjug Chem. 20, 1367–1374.PubMedCrossRefGoogle Scholar
  6. 6.
    MacFadyen, J. R., Haworth, O., Roberston, D., Hardie, D., Webster, M. T., Morris, H. R., et al. (2005) Endosialin (TEM1, CD248) is a marker of stromal fibroblasts and is not selectively expressed on tumour endothelium. FEBS Lett. 579, 2569–2575.PubMedCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  1. 1.Breakthrough Centre, The Institute of Cancer ResearchRoyal Cancer HospitalLondonUK

Personalised recommendations