Warning! You won't be able to use the quotation basket until you enable cookies in your Web browser.
Warning! Your Web browser is no longer supported. Please upgrade to a modern browser.

Detection of Multiple Cell Molecules and mRNA

Double immunoenzymatic staining is a widely used technique to directly analyse the expression of multiple molecules in a single tissue section. One of the aims of our work using the BondTM system was to explore the capability of the automated device to perform multiple labeling.

We were able to show that multiple staining can be easily performed with Bond, it requires a considerably reduced amount of practical time (compared to the manual protocol) and is highly reproducible in terms of high quality and background-free staining. We further explored the possibility to demonstrate that single or double detection of mRNA by in situ hybridization can be combined with immunoenzymatic labeling. Automated labeling could also be performed not only on formalin-fixed paraffin-embedded (FFPE) tissues but also on peripheral blood and bone marrow smears, opening the possibility of using this procedure in the analysis of hematological/cytology samples.

Figures

Figure 1
Examples of automated triple immunostaining including: human tonsil, human bone marrow, white-cell enriched blood smear, lymphocyte predominant Hodgkin lymphoma, breast carcinoma and classical Hodgkin lymphoma

A. Automated triple immunostaining of normal human tonsil. CD20-positive B cells were revealed first by the immunoperoxidase procedure (brown), and immuno-alkaline phosphatase staining was then performed for CD34 (purple) and Ki67 (blue). The inset shows at higher magnification Ki67-positive (blue)/CD20-positive (brown) double positive proliferating B cells as well as CD20-positive only B cells, the vascular endothelium is labeled with CD34 (purple). Note the absence of non-specific background labeling in the section. EDTA-based (pH8.8) antigen retrieval solution, Leica Microsystems.

B. Automated triple immunostaining of a normal human bone marrow trephine shows nuclear expression of GATA-1 (brown) in some erythroid cells which express Glycophorin A (red). CD34 (blue) labels vascular endothelium and some mononuclear cells.

C. A white cell-enriched blood smear is triple immunostained for B cells (CD20, brown), T cells (CD3, blue), and monocytes (CD68, red). Fixed in Acetone: Methanol (1:1) 5min. No antigen retrieval.

D. Automated double immunolabeling of a tissue array core of a lymphocyte predominant Hodgkin lymphoma (LPHL) case. The neoplastic cells express CD20 (brown) and are surrounded by CD3-positive (blue) T cells. EDTA-based (pH8.8) antigen retrieval solution, Leica Microsystems.

E. Automated triple immunostaining of a tissue array core of a breast carcinoma labeled for CD3 (brown), FOXP3 (blue, nuclear), and cytokeratins 5, 6,18 (red). Note the cells dual labeled for CD3 and FOXP3 represent regulatory T cells.

F. Combined immunostaining and in situ hybridization (ISH). A case of classical Hodgkin lymphoma shows EB virus (brown) transcripts in the nuclei of a large CD30-positive (red) neoplastic cell surrounded by CD20-single positive (blue) B cells. Enzyme antigen retrieval solution, 10 min, Leica Microsystems.

Introduction

Immunoenzymatic techniques are widely employed approaches to study in routine diagnosis and research the expression of proteins in tissues and cell samples. These procedures are commonly performed using horseradish peroxidase as the enzyme label, since it can be easily visualized using the diaminobenzidine (DAB) substrate that leaves an insoluble reaction-product marking its location. However, in the 1970s an alternative enzyme label, alkaline phosphatase, began to be utilised by a number of laboratories, particularly when the APAAP procedure was developed.1 The localisation of alkaline phosphatase can be revealed with a number of substrates which also yield insoluble reaction-products of different colors. The combination of the two protocols led to the development of double immunoenzymatic techniques, allowing the study of the localisation of two different cell markers in tissue sections by revealing them in two colours, one in brown (for the peroxidase) and the other in red or blue (for the alkaline phosphatase).2-4 The mainstay of the double immunoenzymatic technique at present is undoubtedly in histopathological research; it increases the amount of information that can be gained from a single slide and also has some beneficial practical aspects such as saving storage space, material and reagents. However, there are also some drawbacks to be mentioned, for example, interpretation of the staining results can be difficult if both immunoenzymatic reactions are not of comparable high quality (i.e. unspecific staining/background should be excluded) and it can require a long time since the two immunoenzymatic procedures are usually performed in sequence.

The aim of this study was to develop a multiple staining approach by using the Bond-maxTM immunostainer (Leica Microsystems) in order to improve the quality of the staining, to speed up the performance and ultimately to demonstrate that the procedure can be easily applied in the research and routine laboratory settings.

Materials and Methods

Samples

Tissues

A variety of formalin-fixed paraffin-embedded (FFPE) normal human tissue blocks (including bone marrow and tonsil) as well as a series of classical Hodgkin lymphomas (cHL) were obtained from the authors’ institution. Blocks of human neoplasms were kindly provided by a number of collaborators (a 1.0 mm tissue-array containing different types of lymphoma was a gift of Dr. G. Roncador, Madrid, Spain and a breast carcinoma tissue-array was provided by Prof. M-L. Hansmann, Frankfurt am Main, Germany).

Hematological Specimens

Blood samples were obtained from a healthy donor and mononuclear cell preparations were performed as previously described.5

Single Immunoenzymatic Staining

Staining was carried out on the Leica Bond-max automated immunostainer following the manufacturer’s protocols. Primary antibodies were diluted to optimal concentration using the Bond Primary Antibody Diluent, for each protocol adhesive labels were printed and applied to the slide before putting on the Bond Universal Covertiles and loading into the immunostainer. Washing steps between each reagent were performed using 1x Bond Wash Solution.

Double Immunoenzymatic Staining

The technique was performed using two polymer detection systems from Leica Microsystems: a) the Bond Polymer Refine – DS9800, (a peroxidase-based detection reagent) and b) the Bond Polymer AP Red – DS9305, (an alkaline phosphatase detection reagent). Table 1 details the protocol applied. The material used included FFPE tissue sections and hematological samples, all steps up to the primary antibody were omitted when this latter material was investigated. To note is that when the second staining procedure is performed new adhesive labels are required, since it is not yet possible to combine the two protocols as the program of the immunostainer stands at present.* The alkaline phosphatase reaction was visualised with either the Fast Red substrate included in the Bond Polymer AP Red detection system or with a different commercially available Fast Blue substrate.6 Hematoxylin counterstaining was performed only if the Fast Red substrate had been used.

Triple Immunoenzymatic Staining

Triple immunoenzymatic labelling in a single section was performed by a further cycle of immuno-alkaline phosphatase (by using the alternative substrate i.e. Fast Red if Fast Blue had been used for the previous reaction and vice versa) being carried out after the first two antibodies were detected (Table 1, Part 2). No hematoxylin counterstain was applied.

 “ImmunoISH”: Combined ISH and Immunoenzymatic Labeling Procedure

In situ hybridization for cellular mRNA (including probes for EBER, kappa and lambda immunoglobulin light chain) was also performed on FFPE tissue sections using the Bond immunostainer according to the manufacturer’s protocol. In Table 2 (Parts 1 and 2) the full protocol is reported for single and dual label ISH. When ISH was combined with multiple immunoenzymatic labeling, the ISH technique was performed first. A one- or eventually two-step immunoenzymatic staining was performed using the same approach as detailed above for multiple staining.

Antibodies and mRNA Probes

The antibodies used in this study include: CD3 (mouse monoclonal, clone LN10, PA0553, Leica Microsystems); CD20 (mouse monoclonal, clone L26, M0755, Dako, Ely, UK); CD34 (mouse monoclonal, clone QBEnd 10, M7165, Dako, Ely, UK); CD68 (mouse monoclonal, clone PG-M1, Prof. B. Falini, Perugia, Italy); 7 Cytokeratins 5, 6, 18 (mouse monoclonal, clone LP34, M0717, Dako, Ely, UK); FOXP3 (mouse monoclonal, clone 236A/E7, MCA2376, AbD Serotec, Oxford, UK); Gata1 (contact corresponding author for details); Glycophorin A (mouse monoclonal, clone JC159, LRF Immunodiagnostic Unit, Oxford, UK); Ki-67 (mouse monoclonal, clone MIB-1, M7240, Dako, Ely, UK); EBER probe (PB0589, Leica Microsystems).

This study was further expanded and details of the additional antibodies and mRNA probes used are reported elsewhere.6

Step

Temp

Time

Part 1

 

 

Bake

60 °C

30 min

Dewax

72 °C

3 min

Alcohol

Room

x 3

Heat Pretreatment

100 °C

20 min

Primary Antibody

Room

15 min

Peroxide Block

Room

5 min

Post Primary (Refine)

Room

8 min

Polymer (Refine)

Room

8min

DAB (Refine)

Room

10 min

Deionized Water

Room

x 3

Part 2

 

 

Primary Antibody

Room

30 min

Post Primary (AP)

Room

20 min

Polymer (AP)

Room

30 min

Fast Red (AP) (or Fast Blue)

Room

20 min (or manual)

Hematoxylin (Optional)

Room

5 min

Deionized Water

Room

x 3

Table 1. Double Staining Protocol. Washes were performed after each step with 1x Bond Wash Solution except where replaced by alcohol or deionised water as shown (*Editor: Note that it is possible with the 3.5 version of Bond software (and later versions) to employ double staining protocols on a single slide without reapplying labels.).
Table 1. Double Staining Protocol. Washes were performed after each step with 1x Bond Wash Solution except where replaced by alcohol or deionised water as shown (*Editor: Note that it is possible with the 3.5 version of Bond software

Step

Temp

Time

Part 1

 

Bake

60 °C

30 min

Dewax

72 °C

3 min

Alcohol

Room

x 3

Enzyme Pretreatment

Room

10 min

Probe Hybridisation

37 °C

1-2 hours

Peroxide Block

Room

5 min

Anti-Fluorescein

Room

15 min

Post Primary (Refine)

Room

8 min

Polymer (Refine)

Room

8 min

DAB (Refine)

Room

10 min

Deionised Water

Room

x 3

Part 2 (option A)

 

Probe Hybridization

37 °C

1-2 hours

Anti-Fluorescein

Room

15 min

Post Primary (AP)

Room

20 min

Polymer (AP)

Room

30 min

Fast Blue

Room

Manual

Part 2 (option B)

 

Primary Antibody

Room

30 min

Post Primary (AP)

Room

20 min

Polymer (AP)

Room

30 min

Fast Blue

Room

Manual

Part 3

 

Primary Antibody

Room

30 min

Post Primary (AP)

Room

20 min

Polymer (AP)

Room

30 min

Fast Red (AP)

Room

20 min

Table 2. ImmunoISH Protocol. Washes were performed after each step with 1x Bond Wash Solution except where replaced by alcohol or deionised water as shown (*Editor: Note that it is possible with the 3.5 version of Bond software (and later versions) to employ double staining protocols on a single slide without reapplying labels.).
Table 2. ImmunoISH Protocol. Washes were performed after each step with 1x Bond Wash Solution except where replaced by alcohol or deionised water as shown.

Results

Double Immunoenzymatic Labeling

The established protocol suitable to detect pairs of antigens in tissue sections by using the automated immunostainer revealed that independent of the type of tissue and antibody used, the staining was clear, reproducible and in absence of non-specific background staining. Examples of double immunostaining for markers present in different cell types are shown in Figure 1.

Triple Immunoenzymatic Labeling

Detection of three molecules in a single tissue section was made possible by the addition of a second immuno-alkaline phosphatase procedure to detect the third molecule (Figure 1). Blue and red chromogen-substrates were used for the two sequential immuno-alkaline phosphatase reactions. We preferred to develop the Fast Red substrate last, since we noted that development of the substrate reaction for the third marker commonly caused the existing alkaline phosphatase reaction product to change color (e.g. the blue product deepened towards a purple color – see Figure 1).

Classical Hodgkin Disease and In Situ Hybridization

The approach of automated multiple labeling was also extended to detect the mRNA of Epstein Barr Virus (EBV, EBER) by using the ISH technique in combination with one or two step immuno-enzymatic labeling (“immunoISH”). For this purpose cases of classical Hodgkin lymphoma were used to detect EBV transcript in the Hodgkin and Reed-Sternberg cells that were labeled for CD30 whereas the bystander B cells were immunostained for CD20 (Figure 1).

Multiple Staining in Hematological Samples

Hematological samples comprising peripheral blood and bone marrow smears were also subjected to multiple labeling. The results showed that the material is suitable for this type of approach (as illustrated in Figure 1) and also open new possibilities for the employment of fine needle aspirates and other biological fluids such as CSF.

Discussion

In this study we were able to demonstrate that the overall quality for multiple antigen detection in tissue sections can be effectively improved by using an automated immunostainer. When multiple immunolabeling is performed one of the recurrent problems is the presence of background which causes difficulties in the interpretation of results. This is therefore one of the major concerns of employing this technique in routine diagnostic laboratories particularly if a manual protocol is used. In the present study we used the Bond automated immunostainer and established a multiple staining protocol that minimized the occurrence of background and facilitated the interpretation of multiple antigen expression in a single tissue section.

One of the factors that we think may have influenced the absence of background in presence of a strong positive staining is the nature of the polymer and the optimized washing procedures that the Bond uses. Another important aspect of the Bond system used in this study was the feasibility to combine usage of different detection systems (i.e. alkaline phosphatase and peroxidase) to produce two labeling reaction products that were clearly distinct from each other, but also compatible with hematoxylin counterstain that was omitted only when a Fast Blue substrate was used (Figure 1).

It was possible therefore to clearly detect two markers present in either distinct cell populations or different cellular compartments within the same cell. We also noticed that for molecules present at the same sub-cellular location, a mixed staining reaction product intermediate between the colors of the two chromogens used was observed. In this regard, it is important to keep in mind, as previous immunoenzymatic labelling studies reported,8 that the insoluble immunoperoxidase reaction product tends to “mask” the area where its reaction product is deposited altering the visualization of the second reaction product. Therefore, with judicious selection of antibody concentration it may be possible to achieve satisfactory dual labeling for molecules localized in close proximity considering that the degree of masking depends on factors such as the density of the peroxidase reaction product.3, 9 The clarity and reproducibility of the staining obtainable using the automated approach described herein would clearly be of value in these circumstances.

Although labeling of two markers at the same site by a double immunoenzymatic technique should normally be avoided (double immunofluorescent labeling is generally the preferred method of choice10, 11), it is not generally appreciated that this procedure offers a very effective means of detecting cells that carry only one of a pair of markers.

Having confirmed the reliability of automated double staining, we went on to develop a method to perform a triple labeling. A range of tissues (either in form of whole or tissue-array sections) and hematological samples (including blood and bone marrow smears as well as bone marrow trephines that undergo a decalcification process) originating also from different institutes and are therefore subject to slightly different processing procedures, were used and the results obtained were of the same quality independent of whether a double or a triple labeling was performed (Figure 1). There is little evidence in literature of either two-12, 13 or three-color8, 9, 14 immunoenzymatic labeling of human tissue as an easily reproducible procedure; those studies that have been published have not had the benefit of being performed on automated equipment.

Similar to the multiple immunoenzymatic labeling, we also demonstrated that by using the Bond, it is possible to perform ISH for two species of cellular mRNA detecting the reaction products in two contrasting enzymatic substrates. To the best of our knowledge this is the first study reporting on dual-color enzymatic ISH analysis of routinely processed human tissue samples. Furthermore, we have shown that it is possible to combine ISH labeling (for one or two mRNA species) with single or double immunostaining (Figure 1). Although immunohistochemistry and ISH have been combined in many studies to identify two markers, it is commonly performed in the reverse order from that used in the present paper (i.e. immunostaining typically precedes ISH)15 also we have not identified previous reports of triple “immunoISH” labeling of either variety on paraffin samples.

Our findings show that double and triple immunostaining can be applied not only to FFPE tissue sections but can also be extended to blood (Figure 1) and bone marrow smears (data not shown). In diagnostic laboratories it is a fairly common occurrence to receive samples with limited cell numbers (e.g. fine needle aspirate, CSF sample), which do not easily lend themselves for phenotypic analysis by the standard technique of flow cytometry. We suggest that in these circumstances automated double or triple labeling represents a simple and quick means of identifying cell populations.

Conclusion

This study proved that by using an automated immunostainer it was possible to develop a protocol capable of producing high quality and reproducible staining for multiple markers in a wide range of tissues and hematological samples. The flexibility and simplicity of the technique also allows performance of more sophisticated assays combining double immunostaining with ISH.

The automated procedure described herein is now extensively used in the author’s laboratory given its reproducibility and the saving in operator time. At present, the principal use of these procedures will continue to be found in the research environment but we envisage that for the simultaneous detection of two (or three) different markers in a diagnostic context (e.g. to reveal a pair of markers whose co-expression is unique to neoplastic cells), double immunoenzymatic labeling will become mandatory and we suggest that this approach will then likely to be the procedure of choice.

References

  1. Cordell JL, Falini B, Erber WN, Ghosh AK, Abdulaziz Z, MacDonald S, et al. Immunoenzymatic labeling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes). J Histochem Cytochem 1984;32(2):219-29.
  2. Mason DY, Stein H, Naiem M, Abdulaziz Z. Immunohistological analysis of human lymphoid tissue by double immunoenzymatic labelling. J Cancer Res Clin Oncol 1981;101(1):13-22.
  3. Valnes K, Brandtzaeg P. Comparison of paired immunofluorescence and paired immunoenzyme staining methods based on primary antisera from the same species. J Histochem Cytochem 1982;30(6):518-24.
  4. Falini B, Abdulaziz Z, Gerdes J, Canino S, Ciani C, Cordell JL, et al. Description of a sequential staining procedure for double immunoenzymatic staining of pairs of antigens using monoclonal antibodies. J Immunol Methods 1986;93(2):265-73.
  5. Marafioti T, Paterson JC, Ballabio E, Reichard KK, Tedoldi S, Hollowood K, et al. Novel markers of normal and neoplastic human plasmacytoid dendritic cells. Blood 2008:Epub ahead of print.
  6. Paterson J, Ballabio E, Mattsson G, Turner S, Mason D, Marafioti T. Labeling of multiple cell markers and mRNA using automated apparatus. Appl Immunohistochem Mol Morphol 2008:In press.
  7. Falini B, Flenghi L, Pileri S, Gambacorta M, Bigerna B, Durkop H, et al. PG-M1: a new monoclonal antibody directed against a fixative-resistant epitope on the macrophage-restricted form of the CD68 molecule. Am J Pathol 1993;142(5):1359-72.
  8. Van der Loos CM, Das PK, Houthoff HJ. An immunoenzyme triple-staining method using both polyclonal and monoclonal antibodies from the same species. Application of combined direct, indirect, and avidin-biotin complex (ABC) technique. J Histochem Cytochem 1987;35(11):1199-204.
  9. Van der Loos CM, Das PK, Van den Oord JJ, Houthoff HJ. Multiple immunoenzyme staining techniques. Use of fluoresceinated, biotinylated and unlabelled monoclonal antibodies. J Immunol Methods 1989;117(1):45-52.
  10. Mason DY, Micklem K, Jones M. Double immunofluorescence labelling of routinely processed paraffin sections. J Pathol 2000;191(4):452-61.
  11. Gellrich S, Ventura R, Jones M, Tan SY, Mason DY. Immunofluorescent and FISH analysis of skin biopsies. Am J Dermatopathol 2004;26(3):242-7.
  12. Short BG, Zimmerman DM, Schwartz LW. Automated double labeling of proliferation and apoptosis in glutathione S-transferase-positive hepatocytes in rats. J Histochem Cytochem 1997;45(9):1299-305.
  13. Herawi M, Epstein JI. Immunohistochemical antibody cocktail staining (p63/HMWCK/AMACR) of ductal adenocarcinoma and Gleason pattern 4 cribriform and noncribriform acinar adenocarcinomas of the prostate. Am J Surg Pathol 2007;31(6):889-94.
  14. Chauhan SC, Vinayek N, Maher DM, Bell MC, Dunham KA, Koch MD, et al. Combined staining of TAG-72, MUC1, and CA125 improves labeling sensitivity in ovarian cancer: antigens for multi-targeted antibody-guided therapy. J Histochem Cytochem 2007;55(8):867-75.
  15. Pineau I, Barrette B, Vallieres N, Lacroix S. A novel method for multiple labeling combining in situ hybridization with immunofluorescence. J Histochem Cytochem 2006;54(11):1303-13.