Histochem Cell Biol (2014) 141:251–262 DOI 10.1007/s00418-013-1177-7
Original Paper
Stain–Decolorize–Stain (SDS): a new technique for multiple staining Jing Li · Yan Zhou · Jiang Gu
Accepted: 20 December 2013 / Published online: 5 January 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Multiple staining of more than one gene/antigen on a single tissue section is an indispensable tool in cell and tissue research. However, most of the available multiple staining techniques have limitations, and there has been no technique to simultaneously visualize and distinguish tissue antigens, nucleotide sequences and other chemical compounds on the same slide. Here, we present a practical and economic multiple stain technique, with which multiple cellular components including mRNA (with in situ hybridization), antigen epitope (with immunohistochemistry) and chemical molecules (with histochemistry) can be stained on a single tissue section to study their relationship. In addition, this technique also offers the possibility to evaluate morphology with an H&E staining on the same sections. We used the placenta, pancreas, breast ductal carcinoma, colon adenocarcinoma, cerebellum,
Jing Li and Yan Zhou contributed equally to this work. Electronic supplementary material The online version of this article (doi:10.1007/s00418-013-1177-7) contains supplementary material, which is available to authorized users. J. Li · J. Gu (*) Department of Pathology, Shantou University Medical College, Shantou, China e-mail: [email protected] Y. Zhou Novogene Bioinformatics Institute, Beijing, China Y. Zhou · J. Gu Department of Pathology, School of Basic Medical Sciences, Peking University, Beijing, China J. Gu Center of Translational Medicine, The Second Affiliated Hospital of Shantou University Medical College, Shantou, China
tonsil and heart tissue sections to evaluate the applicability of this new technique. The sensitivity and specificity of the technique have been tested, and an optimal protocol is recommended. Its applications in surgical pathology and research are discussed. This technique offers a novel tool to evaluate the relationship among multiple components at the same or adjacent locations to meet the needs of pathology diagnosis and research. Keywords Multiple staining · Microwave treatment · In situ hybridization · Immunohistochemistry · Masson stain
Introduction There is an increased demand for simultaneously visualization and analysis of cellular components including proteins, nucleotide sequences, carbohydrate, lipid and other molecules in the same field of visualization for diagnosis and research. Continuous efforts have been made to achieve this goal, and a number of techniques for multiple staining have been developed. These include the use of primary antibodies directly labeled with enzymes (Boorsma 1984), fluorophores (Valnes and Brandtzaeg 1982), colloidal gold (Gu et al. 1981), haptens (Wallace and Wofsy 1979; Van der Loos et al. 1989) or quantum dot (Riegler and Nann 2004), employing primary antibodies raised in different species (Campbell and Bhatnagar 1976) or of different immunoglobulin subclasses (Tidman et al. 1981), the tyramide signal amplification method (Shindler and Roth 1996; Hunyady et al. 1996) and sequential staining methods. In conjunction with the above, techniques to prevent crossreaction among visualization of different compounds have been developed. These include the antibody elution method (Nakane 1968; Tramu et al. 1978), the shielding method
13
252
(Sternberger and Joseph 1979) and employing monovalent F(ab) fragments for blocking excess binding sites from the primary antibodies (Carl et al. 1993; Negoescu et al. 1994; Brouns et al. 2002). All of these methods can achieve the goal of multiple staining to a certain extent, but each one has its limitations (Ino 2004; Tornehave et al. 2000). First, it is often cumbersome to obtain all the needed antibodies and reagents for these techniques. Second, the cross-reaction among the different staining techniques applied on one tissue section is often difficult to avoid. Third, the color reaction and color choices are limited, which makes the evaluation of different color labeled components difficult to distinguish. In 1995, Lan et al. reported a relatively simple and sensitive method in which as many as four antigens were sequentially stained on a single slide and more than one mouse monoclonal primary antibody were used. The key step in this method to avoid antibody cross-reactivity is a two 5-minute microwave treatments, which resulted in antibody denaturation (with moderate microwaving) or antibody elution (with repeated microwaving) (Tornehave et al. 2000). They successfully demonstrated four antigens by using four chromogens with different colors. However, as noted by Tornehave et al. (2000), this method makes it difficult to detect antigens that coexist in the same cellular compartment due to color mixing and shielding effects. In addition, it is challenging to select appropriate chromogens with good visual contrast. In 2009, Glass et al. reported a different approach for multiple IHC staining that enabled the simultaneous visualization of five antigens within a single section. This technique used alcohol-soluble peroxidase substrate, AEC and a previously established protocol to dissolve antibody–antigen complex (Tramu et al. 1978) that incubated sections in 0.15 M KMnO4/0.01 M H2SO4 solution for antibody stripping and photographing following each antigen colorization. Each color was then replaced by an artificial color, and the photographs are superimposed to form a single image. They successfully solved the problems of multiple color compatibility. However, KMnO4: H2SO4 is a strong oxidizing solution, which has been shown to destroy immunoreactivity if part of the antigenic epitope contains methionine residues or other oxidizable amino acid residues (Wang and Larsson 1985; Larsson et al. 1979a). The efforts of combining ISH and IHC on a single tissue section were also reported (Zaidi et al. 2000; Grin et al. 2013). However, the distribution of nucleotide sequences and antigens is difficult to observe on the same tissue section because of the shielding effect when they are located in the same structure or cell. Histochemical stains use a variety of dyes and techniques to stain particular tissues, structures or pathogens to assist pathologists with tissue-based diagnosis. Masson
13
Histochem Cell Biol (2014) 141:251–262
stain is one of the most widely used special stains in routine pathologic diagnosis that mainly identifies collagen and muscle that are not easily identifiable with H&E staining. However, there has been no method to combine histochemical stain and IHC or ISH on a single section. Here, we present a new method of multiple staining combining histochemical staining (Masson stain), ISH staining with NBT/BCIP as the substrate, and repeated staining of antigens with AEC as the substrate, and computer-assisted image processing. This technique can be used not only for the study of distributions of chemical characteristics, mRNA and multiple target antigens, but also for the detection of molecules with identical cellular localization on the same tissue section. In addition, this method also offers the convenience of viewing H&E staining of the same section after multiple stainings, which is difficult to achieve with the other multiple staining methods. This method thereby provides a mean to simultaneously evaluate distributions of genetic, immunohistochemical and histochemical information morphologically at the same location.
Materials and methods Specimens Tissue samples of human pancreas, heart and colon adenocarcinoma were collected from the Department of Pathology, Beijing University Health Science Center. The firsttrimester human placenta samples were collected from the First Affiliated Hospital of Shantou University Medical College. Human breast ductal carcinoma tissues were collected from the Cancer Hospital of Shantou University Medical College. Human tonsil and cerebellum samples were collected from the Department of Forensic Medicine, Shantou University Medical College. Informed consent was obtained. Freshly collected surgical tissue samples were dissected into small pieces and fixed in 4 % paraformaldehyde (PFA) for 4–6 h at room temperature. After washing with running water for 2 h, the tissue samples were dehydrated with a serial of ethanol and embedded in paraffin, and 4-μm-thick serial sections were cut. Masson staining Masson staining was performed according to the method of Kimura and McGinnis (1998) with modification. Briefly, after deparaffinization with xylene, rehydration with a serial of ethanol and counterstaining with hematoxylin, the tissue sections were dipped in 1 % hydrochloric acid (HCl) in 70 % alcohol for 1–5 s. After washing with distilled water, the sections were incubated with Ponceau-acid
Histochem Cell Biol (2014) 141:251–262
fuchsin for 10–15 min and washed with water again. The sections were immerged into 1 % phosphomolybdic acid for 5 min and then washed with 1 % acetic acid. Following dehydration with 95 and 100 % ethanols and xylene, the sections were mounted with neutral balata. The collagen fiber was stained green and the cell cytoplasm red. Immunohistochemistry Immunohistochemistry (IHC) was performed according to the method of Ye et al. (2007). Four-μm-thick paraffin sections were deparaffinized and immerged in 3 % hydrogen peroxide for 20 min to eliminate endogenous peroxidase activity. Antigen retrieval was performed by heating the tissue sections at 96 °C in 0.01 M citrate buffer (pH 6.0) for 15 min and then cooled to room temperature. Following incubation with 4 % BSA in 0.01 M PBS for 1 h, the sections were incubated with primary antibodies overnight at 4 °C. Primary antibodies are listed in Supplementary Material, Table S1. Following primary antibodies incubation and rinsing, the sections were incubated with a secondary antibody (Dako, Copenhagen, Denmark, reacting with both rabbit and mouse immunoglobulins) conjugated with peroxidase at 37 °C for 40 min. Following every step, the sections were rinsed with 0.01 M PBS 3 times for 5 min each. Positive signal was visualized with AEC (Dako, Copenhagen, Denmark), which gave a red staining signal. Slides were lightly counterstained with hematoxylin. In situ hybridization (ISH) In situ hybridization was performed according to the method of Chen and Gu (2007). Deparaffinized and rehydrated sections were incubated in 0.1 N HCl (diluted with RNase-free water) for 10 min. After washing with RNasefree water, the sections were heated to 95 °C in a microwave oven (2,450 Hz, 800 W) in 0.01 M citrate buffer (pH 6.0) for 15 min and then cooled to room temperature. Tissue sections were incubated with cRNA probes to human PLAP at 42 °C for 18–20 h (for human IGHM probe, the temperature was 50 °C). The sequences of primers are shown in Supplementary Material, Table S2. The probes were labeled with digoxigenin (Roche Diagnostics, Penzberg, Germany). Following hybridization, the section was washed in 5 × SSC at 50 °C and 2 × SSC plus 50 % formamide, 2 × SSC at 37 °C. After blocking with horse serum (1:100; Generay Biotech, Shanghai, China) at RT for 1 h, the slides were incubated with anti-digoxigenin antibody conjugated with alkaline phosphatase (dilution 1:500; Roche Diagnostics, Rotkreuz, Switzerland) at RT for 1 h. The reaction was visualized with nitroblue tetrazolium/5-bromo-4-choloro-3indolyl phosphate (NBT-BCIP; Promega, Madison, USA), which produced a dark purple precipitate.
253
Stain–Decolorize–Stain (SDS) method After visualization of the Masson staining, photomicrographs of the interested areas were taken with a light microscope. The slides were then soaked in xylene for 1 h to remove the coverslips and rehydrated in a serial graded ethanol from 100 to 70 %, which was prepared with DEPCtreated water. The red and green colors in Masson staining were completely dissolved in 80 and 70 % ethanol. After washing with DEPC-treated water, the slides were subjected to the ISH procedure. The hematoxylin stain was removed during the ISH microwave treatment in 0.01 M citrate buffer (pH 6.0) at 95 °C for 15 min. After taking photographs, the slides were immersed in distilled water at 50 °C for 3–5 min to remove the coverslips again and incubated with N,N-dimethylformamide (DMF) at 50 °C for 5–10 min in a shaking water bath to dissolve the NBT/BCIP dark purple precipitate. After PBS washing, the slides were treated with 3 % hydrogen peroxide and skipped microwave antigen retrieval and proceed with the first target antigen-specific antibody. After photomicrographing the interested area, slides were then soaked in 80 % alcohol for 10–30 min on a shaker to thoroughly dissolve the red AEC precipitate. The slides were then immersed in citrate buffer and subjected to another round of IHC staining using antibody for the second target antigen. This procedure began with a 2–3-min microwave (2,450 Hz, 800 W) treatment to heat the solution to 95 °C, followed with a 15-min heat (2 min off and 20 s on) to block antibody cross-reaction. Hydrogen peroxide step was used only in the first round of immunostaining. Peroxidase-labeled secondary antibody and AEC enzyme substrate identical to those used in the first round of IHC staining were used in the second and subsequent rounds of staining. Target antigens were all visualized with red AEC signal. After taking photomicrographs, the slides were again decolorized and used for IHC of subsequent target antigens as described above. After the last antigen decolorization, the same section was processed for H&E staining. Multiple staining of five target antigens and H&E staining on the same pancreatic tissue section Five antigens including insulin, glucagon, somatostatin, pancreatic polypeptide and cytokeratin were stained sequentially on a single pancreatic tissue section with this new technique to demonstrate the distributions of the five target antigens. The distributions of B cells, A cells, D cells, PP cells and ductal epithelium, of which these antigens are expressed adjacent to one another, were clearly demonstrated independently and sequentially on the same tissue section. After decolorizing the signal of the last
13
254
target antigen, the section was processed for H&E staining before image processing (described later). Multiple stain combining Masson staining, ISH and two target antigens on the same placental tissue section Masson staining was employed on a placental tissue section to stain the collagenous fibers in the villous stroma. After decoloration, the placental alkaline phosphatase (PLAP) mRNA was demonstrated with ISH using an antisense probe, which is a marker of the syncytiotrophoblast. Between the two procedures, the above-described decolorization method was used to dissolve the color of Masson staining. Following ISH, the NBT/BCIP precipitate was dissolved as described above. Two antibodies to E-cadherin and CD34, respectively, were sequentially applied on the same section with the above SDS method to demonstrate the distributions of cytotrophoblasts and endothelial cells in the chorion villi. After decolorizing the signal of the last antigen, the section was processed for H&E staining. Photographs were taken after each staining for subsequent image processing. In situ hybridization with another cRNA probe (targeted IGHM mRNA) followed with sequential immunostaining of three antigens (IgG, IgA and CD20) was also tested on the human tonsil tissue section. Antigens with similar cellular localization within the same cells In order to evaluate whether the SDS technique could be utilized to demonstrate several different antigens sharing a similar or identical cellular localization, 4 antigens on the same breast ductal carcinoma tissue section were tested. They included PCNA, E-cadherin, cytokeratin and Ki67 that were sequentially demonstrated on the same tissue section of a breast carcinoma with the SDS technique. Among them, PCNA and Ki67 are two nucleus antigens, E-cadherin a membranous antigen and cytokeratin an epithelial cytoplasmic antigen. Image processing To achieve an integrated picture that displays Masson staining, ISH and two antigens on the same placental tissue section, all the five antigens on one pancreatic tissue section or all the four antigens on one breast carcinoma tissue section at the same time, Adobe Photoshop (Adobe Systems, Mountain View, CA) was used for image processing. The Photoshop functions, “Extract” and “Replace Color”, were used for color replacement. The “Extract” function extracts a defined color, followed by replacing it with a different color with the “Replace Color”. For each set of photographs, one was chosen as a background. The other positive
13
Histochem Cell Biol (2014) 141:251–262
red signals were replaced with other distinct colors. These images were then superimposed onto the background photograph to form an integrated picture. To evaluate co-localization of two or more staining positivities, the “Blending Options” and “Overlay” functions of Photoshop were used to generate the new image where the overlapping of two colors resulting in a distinct third color, thereby clearly indicating collaboration. For the placental tissue section, pictures were taken after each staining to separately show the distributions of collagen, PLAP mRNA and the two antigens. The photograph of Masson staining was taken as a background, while the dark purple ISH positive signal and the two red IHC signals were artificially changed with Photoshop to cyan, blue and yellow, respectively, to assign each positive cell or structure a distinct color for easy recognition. The distinct colors representing different cells or structures were superimposed onto the background image to obtain an integrated photograph within which the distribution of all positive cells or structures and their relationships to one another were clearly illustrated. For the pancreatic tissue section, five pictures were obtained which separately showed the distribution of the five antigens. The image processing was the same to that described above. The photograph of cytokeratin staining was used as a background, and the red signals in the other four pictures were altered with Photoshop to yellow, purple, green and black, respectively, to make each antigen easily recognizable. The different photographs of different antigens from the same areas were then superimposed onto the background image, and the distributions of all antigens of interest were illustrated with different colors on the integrated picture. For breast ductal carcinoma tissue section, four photographs of the same area were taken. The photograph of cytokeratin staining was used as a background. The red positive signals of the other three images were altered with Photoshop to green, yellow and blue, respectively. As a new layer, the altered color of each photograph was superimposed onto the background image sequentially, and the function of “Overlay” was used to illustrate the color mixture of PCNA and Ki67 in the same nuclei. The distributions of all antigens in the same area were shown with different colors on the integrated photograph. Controls For each round of multiple IHC staining, the primary antibody was replaced with PBS as a negative control. For PLAP ISH staining, a cRNA sense probe was used as a negative control. We also controlled the specificity and sensitivity of this new technique. To control specificity, we omitted the
Histochem Cell Biol (2014) 141:251–262
primary antibody but used the secondary antibody or AEC alone, or both following the elution of the previous round of immunostaining. To control the sensitivity, we compared the staining intensities of the same antigen when it was stained first and last on the same tissue section.
Results Multiple staining of sequential Masson staining, ISH and IHC With this technique, the collagen fibers, PLAP mRNA and two antigens, E-cadherin and CD34, were sequentially demonstrated on the same placental tissue section (Fig. 1a–d). No cross-reaction between antibodies was observed. After photo image processing, the different cellular components were clearly identified in their own residence (Fig. 1e). The syncytiotrophoblasts (cyan) lay in the outer layer of the villi, while the cytotrophoblasts (dark blue) in the inner layer. Vessels with endothelial cells are shown in yellow. Around the vessels are collagen fibers shown in green. In addition, an H&E-stained photograph was also obtained of the same area for morphologic evaluation (Fig. 1f). The results of a combination of in situ hybridization with cRNA probe to IGHM mRNA and sequential immunostaining of three antigens on the same tonsil tissue section are shown in Supplementary Fig. S4.
255
on the breast cancer tissue section were identified by the membranous positive signals with the E-cadherin-specific antibody (Fig. 3b) or the cytoplasm positive signals with cytokeratin-specific antibody (Fig. 3c). No cross-reaction between antibodies was observed. After image processing, we could easily identify that in the carcinoma nest, some nuclei were positive for PCNA (green) only or Ki67 (blue) only (Fig. 3e), while some were positive for both (black, which was the overlapping of green and blue colors; arrowheads). The breast ductal carcinoma cell membranes were also clearly visible with E-cadherin staining (yellow) and the cytoplasm with cytokeratin in red. Different antigens with the same cellular localization were also tested with other tissues including human colon adenocarcinoma, heart and tonsil (Supplementary Fig. S3). The antigen-bound primary and secondary antibodies were all completely eluted from the tissue sections, and staining of a different antigen afterward was always successful without bringing up the positive signal of the previous antigen (Supplementary Fig. S1). When the same antigen was stained in the first round or the third round, there was no visible difference between the two positive signals (Supplementary Fig. S2). All negative and positive controls gave appropriate results, thereby establishing the specificity and sensitivity of this technique. The blue hematoxylin nuclear stain was completely removed with the microwave treatment.
Discussion Multiple staining of five antigens and H&E staining on a single tissue section Five antigens including insulin, glucagon, somatostatin, PP and cytokeratin were stained on a single pancreatic tissue section with this SDS technique (Fig. 2a–e). No cross-reaction was observed. Using superimposed photographs, the AEC red color for each antigen was replaced with a new color distinctly different from the color assigned to others. The distributions of these antigens and the cells in which they reside were clearly identified. The PP cells (black), A cells (green) and D cells (purple) are located at the periphery of the islet, while the B cells (yellow) occupy the central portion. Cytokeratin (red)-positive signal marked ductal epithelial cells within the exocrine tissue (Fig. 2f). An H&E staining was also obtained of the same area (Fig. 2g), offering clear morphologic information in addition to the immunohistochemical signals. Staining of antigens with the same cellular localization Numerous nuclei were positive for the antigens PCNA (Fig. 3a) and Ki67 (Fig. 3d). The ductal carcinoma cells
In this study, we developed protocols that can simultaneously demonstrate multiple cellular components including antigens, nucleotide sequences and chemical compounds on the same tissue section to discern their distribution and relationship. This technique combines IHC, ISH and histochemistry as well as H&E on a single tissue section, providing a powerful tool to meet the increasing demand for simultaneous visualization of multiple components at the same location on the same tissue section. About specificity and sensitivity A few critical points should be discussed for this new technique. One is the specificity, i.e., only the desired targets are specifically labeled without undesirable staining. The specificity was established with the following example. After positive staining of one antigen (MAP2 in neuron cytoplasm), the positive color was removed completely with 80 % ethanol followed by microwave treatment (remove the antibodies). The effectiveness of the above treatment was verified by the negative reaction when AEC and the secondary antibody were used
13
256
Histochem Cell Biol (2014) 141:251–262
Fig. 1 Multiple staining of sequential Masson staining, in situ hybridization and immunohistochemistry staining on a single placenta tissue section. a–f Are the same area of a placental tissue section. a–d Are sequentially staining demonstrating collagen fibers (green), PLAP mRNA (dark purple), E-cadherin (red) and CD34 (red), which, respectively, show the distributions of collagen fibers,
syncytiotrophoblasts, cytotrophoblasts and vascular endothelial cells on the same placental section. e Is a superimposed photomicrograph showing signals for collagen fibers in green, syncytiotrophoblasts in cyan, cytotrophoblasts in blue and vascular endothelial cells in yellow. f Shows a photograph of H&E staining in the same field after color elution. All scale bars 20 μm
alone or in combination. This was followed by another immunostaining (factor VIII for endothelial cells), which showed distinct positivity in endothelial cells. Next, the first antigen (MAP2) was stained again, and the positive MAP2 cytoplasmic signal was demonstrated again. The specificity and effectiveness of the color and antibody
elusion treatments were tested with many antibody and antigen combinations, and the specificity was repeatedly confirmed. In addition, the sensitivity of the test was not sacrificed with repeated elusion and staining steps as the subsequent positive signal had the same intensity to the initial staining of the same antigen. Therefore, the targets
13
Histochem Cell Biol (2014) 141:251–262
257
Fig. 2 Multiple staining of a single pancreas tissue section. a–g Are the same field of a pancreas tissue section. a–d and e Show the sequentially immunostaining results of pancreatic polypeptide, insulin, cytokeratin, somatostatin and glucagon on this pancreas tissue section (red signals), and no cross-reaction is identifiable. PP cells, B cells, ductal epithelium, D cells, A cells are marked, respectively,
by red coloration in a–d and e. f Is a superimposed photomicrograph showing signals for pancreatic polypeptide in black, insulin in yellow, cytokeratin in red, somatostatin in purple and glucagon in green color. g Shows a photomicrograph of the same field after H&E staining. All scale bars 20 μm
can be easily detected without losing reactivity during the relatively long procedure. The new technique is also practical. It can be carried out as a routine laboratory technique without demanding expensive equipment or challenging protocols.
About co‑localization Different antigens localized to the same cellular structure could be demonstrated with the SDS technique. Figure 3 and Supplementary Fig. S3 showed the staining of
13
258
Fig. 3 Staining of antigens in the same cellular localization. a–e Are the same field of a breast ductal carcinoma tissue section. Nuclei of proliferation cells including both cancer cells are positive for PCNA (a, red signals) and Ki67 (d, red signals) with certain overlapping that are clearly shown in black color (e, arrowheads). The breast ductal carcinoma cells are specifically identified by E-cadherin, which had a distinct membranous staining (b, red signals). The cyto-
13
Histochem Cell Biol (2014) 141:251–262
plasms of carcinoma cells show positive signals for cytokeratin (c, red signals). e Is an integrated photomicrograph showing signals for carcinoma cells’ membrane in yellow, cytoplasm in red, nucleuses solely positive for PCNA in green and for Ki67 in blue. The overlapping of PCNA and Ki67 is shown in black (arrowheads), which is the mixture of green and blue. All scale bars 20 μm
Histochem Cell Biol (2014) 141:251–262
differently paired antigens localized at the same cellular structure using this technique. Staining of the first antigen did not over shadow or affect the subsequent antigen stainings at the same location. We stained other antigens located in different cells or cellular structures in between the paired antigens to illustrate the specificity of the labeling (Fig. 3 and Supplementary Fig. S3) and obtain the desired results. About the heat treatment Since microwave heating was first introduced by Lan et al. (1995) as a simple step to block antibody cross-reactivity in sequential antigen stainings, its application has been expanded to combine with other techniques such as immunofluorescence (Tornehave et al. 2000) and spectral imaging (van der Loos 2008). We employed the heat treatment method together with chromogen dissolution for our new technique of multiple staining. With this technique, multiple staining can be carried out regardless of the species or immunoglobulin subclasses of the primary antibody. In addition, microwave treatment is an established method for in situ hybridization to expose mRNA to enhance detection sensitivity (Lan et al. 1996; Sperry et al. 1996). It has been reported that microwave treatment could not only enhance the mRNA signal to improve hybridization efficiency but also preserve tissue morphology (Lan et al. 1996; Sibony et al. 1995). As such, this method may find broad application in both diagnosis and research. Advantages of the SDS technique over other techniques The SDS technique described in this study has several advantages over previously established techniques (Tramu et al. 1978; Lan et al. 1995; Brouns et al. 2002; Sternberger and Joseph 1979). 1. Unlike the other multiple staining techniques, the various chromogens used for visualizing target antigens with different colors in this method can be replaced with AEC alone, saving both cost and effort in performing the protocol. 2. With no restriction on color selection and no need for choosing chromogens with good visual contrast, more target antigens may be stained. We demonstrated simultaneously staining of five antigens. Greater number could be achieved. 3. This SDS technique visualizes different molecules, nucleotide sequences and antigens sequentially, and the specific location of each target can be clearly demonstrated without worrying about color as the color deposit is removed by elution before proceeding to the next staining target. In addition, the shielding effects, i.e., the blocking of staining targets, often seen in mul-
259
tiple staining can be avoided. Previously, one molecule identification with color precipitation could shield others from subsequent detection. 4. This technique is uniquely suited to investigate collaboration of two or more antigens/nucleotide sequences. With the image processing software, the overlapping of two or more colors can be clearly identified and percentage calculated. 5. As the color for each cellular component is individually obtained and the final colors are artificially assigned, it would be possible to simultaneously display any combination of molecules of interest to exam their relationship on the same tissue section. 6. The same tissue section can be used for H&E staining after all the signals of multiple staining are dissolved, offering visualization of a clear morphology side-byside with visualization of nucleotide sequences, protein antigens and histochemical staining in the same visual field. This feature will be particularly beneficial for pathologic diagnosis by histopathologists who rely on H&E staining and morphology to make a judgment. Recommended protocols The sequence of the steps in SDS has been optimized, in which Masson staining should be performed first, and then ISH followed by IHC, as the Masson staining loses color easily in alcohol (70–80 % ethanol). If the section is treated with microwave first, the Masson staining would not work. ISH performed after Masson staining gave a slightly weaker positive signal than that of ISH alone, whereas there was no visible effect on IHC. Beraki et al. demonstrated successful ISH on a Giemsa prestained tissue section (Beraki et al. 2012). Some preferred antigen localization before ISH (Han et al. 1992), while others tended to carry out ISH for RNA or DNA before IHC (Rimsza et al. 1996; Beraki et al. 2012; Zaidi et al. 2000). It was reported that when ISH was performed after IHC, little or only very weak nucleotide signal could be detected with ISH (Reisenbichler et al. 2012). Our experience confirmed this claim. We performed each staining independently and compared the results with that obtained with the multiple staining procedure. It was found that no antigen/staining target was lost or gained due to the multiple staining technique. However, the multiple staining protocol does tend to weaken the staining intensity when the antigens were stained toward the end of the protocol. Therefore, we recommend that if quantitative analysis is desired, the operator of the procedure should evaluate the possible loss of reactivity by performing the staining of each target on serial section independently and compare the single staining results with that of the multiple staining side-by-side to
13
260
Histochem Cell Biol (2014) 141:251–262
Fig. 4 Principal steps of the SDS technique illustration. The optimal order of stainings is Masson stain first, and then in situ hybridization followed by immunohistochemistry with microphotographing and
color elution in between each separate staining. Microwave treatment is used for blocking antibody cross-reaction between each step of IHC. H&E staining can be performed by the end
make a determination of the quantitative differences and to control possible cross-reactivity among different steps. A step-by-step illustration of the SDS technique is shown in Fig. 4. In addition, probe concentration of ISH should be slightly higher than that for ISH alone, so is antibody concentration in the subsequent IHC. In addition to the order of technical steps, attention should be paid to the order of antibodies applied. Staining sequence of the antigens was rarely discussed when multiple staining techniques were introduced. We established a parameter for IHC as a reference to optimize the antibody incubation sequence. We named this parameter the “color developing duration (CDD) value.” It is defined as the time it takes from adding AEC solution to the moment that the target antigen becomes clearly visible. The CDD value for each antigen staining can be obtained easily in separate experiments. We found that the value differed with different target antigens, antibody manufacturers, and even diluents. The optimal staining sequence was designed by arranging target antigens from the highest CDD value to the lowest. That is, the faster the AEC signal of an antigen develops or the more easily an antigen could be stained, the later it should be employed in this SDS technique. In this way, multiple positive signals can be detected easily. We found no evidence of cross-reaction among the antibodies
or techniques in any of our experiments. The CDD value for any chromogen is a flexible time that is dependent on numerous factors including almost all steps and procedures of IHC staining process. For example, when doing heat-induced antigen retrieval (AR) pretreatment, various protocols of AR significantly influence the results of IHC including the “CDC” time. The operator should pay attention to repeating microwave treatment process (as well as other elute treatments) that may add some additional influence factors to not only the “CDD” value but also the intensity of IHC or ISH. This should be taken into consideration when quantitative evaluation of the results is the desired outcome. This technique is suitable for simultaneous demonstration of multiple antigens/targets to study their relationship but may not be ideal for quantitative evaluation. In conclusion, we have established a new method of multiple staining that combines histochemistry, in situ hybridization and immunohistochemistry. It is capable of detecting co-localization of multiple molecules at the same location. The specificity and sensitivity of this new technique were tested and validated. We also introduced a CDD value as an index to optimize the order of steps performed in this technique. It offers new possibility to comparatively analyze the distribution of different molecular components and their co-localization for diagnosis and research.
13
Histochem Cell Biol (2014) 141:251–262 Acknowledgments We thank Lu Yao, Na Niu, Zhuo Li, Yuxuan Liu, Yong Guo and Qi Cao from Department of Pathology of Peking University Health Science Center and Yiqun Geng, Tao Huang and Li Du from Department of Pathology of Shantou University Medical College for providing the tissue samples, comments and suggestions. Conflict of interest The authors declared no conflict of interest with respect to the research, authorship, and publication of this article. This work was supported by grants from the National Nature Science Foundation of China (No. 81030033, 30971150 to J.G.).
References Beraki E, Olsen TK, Sauer T (2012) Establishing a protocol for immunocytochemical staining and chromogenic in situ hybridization of Giemsa and Diff-Quick prestained cytological smears. Cytojournal 9:8 Boorsma DM (1984) Direct immunoenzyme double staining applicable for monoclonal antibodies. Histochemistry 80(2):103–106 Brouns I, Van Nassauw L, Van Genechten J, Majewski M, Scheuermann DW, Timmermans JP, Adriaensen D (2002) Triple immunofluorescence staining with antibodies raised in the same species to study the complex innervation pattern of intrapulmonary chemoreceptors. J Histochem Cytochem 50(4):575–582 Campbell GT, Bhatnagar AS (1976) Simultaneous visualization by light microscopy of two pituitary hormones in a single tissue section using a combination of indirect immunohistochemical methods. J Histochem Cytochem 24(2):448–452 Carl SAL, Gilleteferguson I, Ferguson DG (1993) An indirect immunofluorescence procedure for staining the same cryosection with 2 mouse monoclonal primary antibodies. J Histochem Cytochem 41(8):1273–1278 Chen Z, Gu J (2007) Immunoglobulin G expression in carcinomas and cancer cell lines. Faseb J 21(11):2931–2938 Glass G, Papin JA, Mandell JW (2009) SIMPLE: a sequential immunoperoxidase labeling and erasing method. J Histochem Cytochem 57(10):899–905 Grin A, Brezden-Masley C, Bauer S, Streutker CJ (2013) HER2 in situ hybridization in gastric and gastroesophageal adenocarcinoma: comparison of automated dual ISH to FISH. Appl Immunohistochem Mol Morphol Gu J, De Mey J, Moeremans M, Polak JM (1981) Sequential use of the PAP and immunogold staining method for the light microscopical double staining of the tissue antigens. Regul Pept 1(6):365–374 Han KH, Hollinger FB, Noonan CA, Yoffe B (1992) Simultaneous detection of HBV-specific antigens and DNA in paraffinembedded liver tissue by immunohistochemistry and in situ hybridization using a digoxigenin-labeled probe. J Virol Methods 37(1):89–97 Hunyady B, Krempels K, Harta G, Mezey E (1996) Immunohistochemical signal amplification by catalyzed reporter deposition and its application in double immunostaining. J Histochem Cytochem 44(12):1353–1362 Ino H (2004) Application of antigen retrieval by heating for double-label fluorescent immunohistochemistry with identical species-derived primary antibodies. J Histochem Cytochem 52(9):1209–1217 Kimura M, McGinnis MR (1998) Fontana-Masson–stained tissue from culture-proven mycoses. Arch Pathol Lab Med 122(12):1107–1111 Lan HY, Mu W, Nikolicpaterson DJ, Atkins RC (1995) A novel, simple, reliable, and sensitive method for multiple immunoenzyme
261 staining—use of microwave-oven heating to block antibody cross-reactivity and retrieve antigens. J Histochem Cytochem 43(1):97–102 Lan HY, Mu W, Ng YY, Nikolic-Paterson DJ, Atkins RC (1996) A simple, reliable, and sensitive method for nonradioactive in situ hybridization: use of microwave heating to improve hybridization efficiency and preserve tissue morphology. J Histochem Cytochem 44(3):281–287 Larsson LI, Childers S, Snyder SH (1979) Met- and leu-enkephalin immunoreactivity in separate neurons. Nature 282:407–410 Nakane PK (1968) Simultaneous localization of multiple tissue antigens using peroxidase-labeled antibody method—a study on pituitary glands of rat. J Histochem Cytochem 16(9):557–560 Negoescu A, Labatmoleur F, Lorimier P, Lamarcq L, Guillermet C, Chambaz E, Brambilla E (1994) F(Ab) secondary antibodies—a general-method for double immunolabeling with primary antisera from the same species—efficiency control by chemiluminescence. J Histochem Cytochem 42(3):433–437 Reisenbichler ES, Horton D, Rasco M, Andea A, Hameed O (2012) Evaluation of dual immunohistochemistry and chromogenic in situ hybridization for HER2 on a single section. Am J Clin Pathol 137(1):102–110 Riegler J, Nann T (2004) Application of luminescent nanocrystals as labels for biological molecules. Anal Bioanal Chem 379(7–8):913–919 Rimsza LM, Vela EE, Frutiger YM, Rangel CS, Solano M, Richter LC, Grogan TM, Bellamy WT (1996) Rapid automated combined in situ hybridization and immunohistochemistry for sensitive detection of cytomegalovirus in paraffin-embedded tissue biopsies. Am J Clin Pathol 106(4):544–548 Shindler KS, Roth KA (1996) Double immunofluorescent staining using two unconjugated primary antisera raised in the same species. J Histochem Cytochem 44(11):1331–1335 Sibony M, Commo F, Callard P, Gasc JM (1995) Enhancement of mRNA in situ hybridization signal by microwave heating. Lab Invest 73(4):586–591 Sperry A, Jin L, Lloyd RV (1996) Microwave treatment enhances detection of RNA and DNA by in situ hybridization. Diagn Mol Pathol 5(4):291–296 Sternberger LA, Joseph SA (1979) The unlabeled antibody method. Contrasting color staining of paired pituitary hormones without antibody removal. J Histochem Cytochem 27(11):1424–1429 Tidman N, Janossy G, Bodger M, Granger S, Kung PC, Goldstein G (1981) Delineation of human thymocyte differentiation pathways utilizing double-staining techniques with monoclonal antibodies. Clin Exp Immunol 45(3):457–467 Tornehave D, Hougaard DM, Larsson LI (2000) Microwaving for double indirect immunofluorescence with primary antibodies from the same species and for staining of mouse tissues with mouse monoclonal antibodies. Histochem Cell Biol 113(1):19–23 Tramu G, Pillez A, Leonardelli J (1978) An efficient method of antibody elution for the successive or simultaneous localization of two antigens by immunocytochemistry. J Histochem Cytochem 26(4):322–324 Valnes K, Brandtzaeg P (1982) Comparison of paired immunofluorescence and paired immunoenzyme staining methods based on primary antisera from the same species. J Histochem Cytochem 30(6):518–524 van der Loos CM (2008) Multiple immunoenzyme staining: methods and visualizations for the observation with spectral imaging. J Histochem Cytochem 56(4):313–328 Van der Loos CM, Das PK, Van den Oord JJ, Houthoff HJ (1989) Multiple immunoenzyme staining techniques. Use of fluoresceinated, biotinylated and unlabelled monoclonal antibodies. J Immunol Methods 117(1):45–52
13
262 Wallace EF, Wofsy L (1979) Hapten-sandwich labeling. IV. Improved procedures and non-cross-reacting hapten reagents for doublelabeling cell surface antigens. J Immunol Methods 25(3):283–289 Wang BL, Larsson LI (1985) Simultaneous demonstration of multiple antigens by indirect immunofluorescence or immunogold staining. Novel light and electron microscopical double and triple staining method employing primary antibodies from the same species. Histochemistry 83(1):47–56
13
Histochem Cell Biol (2014) 141:251–262 Ye J, Zhang B, Xu J, Chang Q, McNutt MA, Korteweg C, Gong E, Gu J (2007) Molecular pathology in the lungs of severe acute respiratory syndrome patients. Am J Pathol 170(2):538–545 Zaidi AU, Enomoto H, Milbrandt J, Roth KA (2000) Dual fluorescent in situ hybridization and immunohistochemical detection with tyramide signal amplification. J Histochem Cytochem 48(10):1369–1375
Original Paper
Stain–Decolorize–Stain (SDS): a new technique for multiple staining Jing Li · Yan Zhou · Jiang Gu
Accepted: 20 December 2013 / Published online: 5 January 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Multiple staining of more than one gene/antigen on a single tissue section is an indispensable tool in cell and tissue research. However, most of the available multiple staining techniques have limitations, and there has been no technique to simultaneously visualize and distinguish tissue antigens, nucleotide sequences and other chemical compounds on the same slide. Here, we present a practical and economic multiple stain technique, with which multiple cellular components including mRNA (with in situ hybridization), antigen epitope (with immunohistochemistry) and chemical molecules (with histochemistry) can be stained on a single tissue section to study their relationship. In addition, this technique also offers the possibility to evaluate morphology with an H&E staining on the same sections. We used the placenta, pancreas, breast ductal carcinoma, colon adenocarcinoma, cerebellum,
Jing Li and Yan Zhou contributed equally to this work. Electronic supplementary material The online version of this article (doi:10.1007/s00418-013-1177-7) contains supplementary material, which is available to authorized users. J. Li · J. Gu (*) Department of Pathology, Shantou University Medical College, Shantou, China e-mail: [email protected] Y. Zhou Novogene Bioinformatics Institute, Beijing, China Y. Zhou · J. Gu Department of Pathology, School of Basic Medical Sciences, Peking University, Beijing, China J. Gu Center of Translational Medicine, The Second Affiliated Hospital of Shantou University Medical College, Shantou, China
tonsil and heart tissue sections to evaluate the applicability of this new technique. The sensitivity and specificity of the technique have been tested, and an optimal protocol is recommended. Its applications in surgical pathology and research are discussed. This technique offers a novel tool to evaluate the relationship among multiple components at the same or adjacent locations to meet the needs of pathology diagnosis and research. Keywords Multiple staining · Microwave treatment · In situ hybridization · Immunohistochemistry · Masson stain
Introduction There is an increased demand for simultaneously visualization and analysis of cellular components including proteins, nucleotide sequences, carbohydrate, lipid and other molecules in the same field of visualization for diagnosis and research. Continuous efforts have been made to achieve this goal, and a number of techniques for multiple staining have been developed. These include the use of primary antibodies directly labeled with enzymes (Boorsma 1984), fluorophores (Valnes and Brandtzaeg 1982), colloidal gold (Gu et al. 1981), haptens (Wallace and Wofsy 1979; Van der Loos et al. 1989) or quantum dot (Riegler and Nann 2004), employing primary antibodies raised in different species (Campbell and Bhatnagar 1976) or of different immunoglobulin subclasses (Tidman et al. 1981), the tyramide signal amplification method (Shindler and Roth 1996; Hunyady et al. 1996) and sequential staining methods. In conjunction with the above, techniques to prevent crossreaction among visualization of different compounds have been developed. These include the antibody elution method (Nakane 1968; Tramu et al. 1978), the shielding method
13
252
(Sternberger and Joseph 1979) and employing monovalent F(ab) fragments for blocking excess binding sites from the primary antibodies (Carl et al. 1993; Negoescu et al. 1994; Brouns et al. 2002). All of these methods can achieve the goal of multiple staining to a certain extent, but each one has its limitations (Ino 2004; Tornehave et al. 2000). First, it is often cumbersome to obtain all the needed antibodies and reagents for these techniques. Second, the cross-reaction among the different staining techniques applied on one tissue section is often difficult to avoid. Third, the color reaction and color choices are limited, which makes the evaluation of different color labeled components difficult to distinguish. In 1995, Lan et al. reported a relatively simple and sensitive method in which as many as four antigens were sequentially stained on a single slide and more than one mouse monoclonal primary antibody were used. The key step in this method to avoid antibody cross-reactivity is a two 5-minute microwave treatments, which resulted in antibody denaturation (with moderate microwaving) or antibody elution (with repeated microwaving) (Tornehave et al. 2000). They successfully demonstrated four antigens by using four chromogens with different colors. However, as noted by Tornehave et al. (2000), this method makes it difficult to detect antigens that coexist in the same cellular compartment due to color mixing and shielding effects. In addition, it is challenging to select appropriate chromogens with good visual contrast. In 2009, Glass et al. reported a different approach for multiple IHC staining that enabled the simultaneous visualization of five antigens within a single section. This technique used alcohol-soluble peroxidase substrate, AEC and a previously established protocol to dissolve antibody–antigen complex (Tramu et al. 1978) that incubated sections in 0.15 M KMnO4/0.01 M H2SO4 solution for antibody stripping and photographing following each antigen colorization. Each color was then replaced by an artificial color, and the photographs are superimposed to form a single image. They successfully solved the problems of multiple color compatibility. However, KMnO4: H2SO4 is a strong oxidizing solution, which has been shown to destroy immunoreactivity if part of the antigenic epitope contains methionine residues or other oxidizable amino acid residues (Wang and Larsson 1985; Larsson et al. 1979a). The efforts of combining ISH and IHC on a single tissue section were also reported (Zaidi et al. 2000; Grin et al. 2013). However, the distribution of nucleotide sequences and antigens is difficult to observe on the same tissue section because of the shielding effect when they are located in the same structure or cell. Histochemical stains use a variety of dyes and techniques to stain particular tissues, structures or pathogens to assist pathologists with tissue-based diagnosis. Masson
13
Histochem Cell Biol (2014) 141:251–262
stain is one of the most widely used special stains in routine pathologic diagnosis that mainly identifies collagen and muscle that are not easily identifiable with H&E staining. However, there has been no method to combine histochemical stain and IHC or ISH on a single section. Here, we present a new method of multiple staining combining histochemical staining (Masson stain), ISH staining with NBT/BCIP as the substrate, and repeated staining of antigens with AEC as the substrate, and computer-assisted image processing. This technique can be used not only for the study of distributions of chemical characteristics, mRNA and multiple target antigens, but also for the detection of molecules with identical cellular localization on the same tissue section. In addition, this method also offers the convenience of viewing H&E staining of the same section after multiple stainings, which is difficult to achieve with the other multiple staining methods. This method thereby provides a mean to simultaneously evaluate distributions of genetic, immunohistochemical and histochemical information morphologically at the same location.
Materials and methods Specimens Tissue samples of human pancreas, heart and colon adenocarcinoma were collected from the Department of Pathology, Beijing University Health Science Center. The firsttrimester human placenta samples were collected from the First Affiliated Hospital of Shantou University Medical College. Human breast ductal carcinoma tissues were collected from the Cancer Hospital of Shantou University Medical College. Human tonsil and cerebellum samples were collected from the Department of Forensic Medicine, Shantou University Medical College. Informed consent was obtained. Freshly collected surgical tissue samples were dissected into small pieces and fixed in 4 % paraformaldehyde (PFA) for 4–6 h at room temperature. After washing with running water for 2 h, the tissue samples were dehydrated with a serial of ethanol and embedded in paraffin, and 4-μm-thick serial sections were cut. Masson staining Masson staining was performed according to the method of Kimura and McGinnis (1998) with modification. Briefly, after deparaffinization with xylene, rehydration with a serial of ethanol and counterstaining with hematoxylin, the tissue sections were dipped in 1 % hydrochloric acid (HCl) in 70 % alcohol for 1–5 s. After washing with distilled water, the sections were incubated with Ponceau-acid
Histochem Cell Biol (2014) 141:251–262
fuchsin for 10–15 min and washed with water again. The sections were immerged into 1 % phosphomolybdic acid for 5 min and then washed with 1 % acetic acid. Following dehydration with 95 and 100 % ethanols and xylene, the sections were mounted with neutral balata. The collagen fiber was stained green and the cell cytoplasm red. Immunohistochemistry Immunohistochemistry (IHC) was performed according to the method of Ye et al. (2007). Four-μm-thick paraffin sections were deparaffinized and immerged in 3 % hydrogen peroxide for 20 min to eliminate endogenous peroxidase activity. Antigen retrieval was performed by heating the tissue sections at 96 °C in 0.01 M citrate buffer (pH 6.0) for 15 min and then cooled to room temperature. Following incubation with 4 % BSA in 0.01 M PBS for 1 h, the sections were incubated with primary antibodies overnight at 4 °C. Primary antibodies are listed in Supplementary Material, Table S1. Following primary antibodies incubation and rinsing, the sections were incubated with a secondary antibody (Dako, Copenhagen, Denmark, reacting with both rabbit and mouse immunoglobulins) conjugated with peroxidase at 37 °C for 40 min. Following every step, the sections were rinsed with 0.01 M PBS 3 times for 5 min each. Positive signal was visualized with AEC (Dako, Copenhagen, Denmark), which gave a red staining signal. Slides were lightly counterstained with hematoxylin. In situ hybridization (ISH) In situ hybridization was performed according to the method of Chen and Gu (2007). Deparaffinized and rehydrated sections were incubated in 0.1 N HCl (diluted with RNase-free water) for 10 min. After washing with RNasefree water, the sections were heated to 95 °C in a microwave oven (2,450 Hz, 800 W) in 0.01 M citrate buffer (pH 6.0) for 15 min and then cooled to room temperature. Tissue sections were incubated with cRNA probes to human PLAP at 42 °C for 18–20 h (for human IGHM probe, the temperature was 50 °C). The sequences of primers are shown in Supplementary Material, Table S2. The probes were labeled with digoxigenin (Roche Diagnostics, Penzberg, Germany). Following hybridization, the section was washed in 5 × SSC at 50 °C and 2 × SSC plus 50 % formamide, 2 × SSC at 37 °C. After blocking with horse serum (1:100; Generay Biotech, Shanghai, China) at RT for 1 h, the slides were incubated with anti-digoxigenin antibody conjugated with alkaline phosphatase (dilution 1:500; Roche Diagnostics, Rotkreuz, Switzerland) at RT for 1 h. The reaction was visualized with nitroblue tetrazolium/5-bromo-4-choloro-3indolyl phosphate (NBT-BCIP; Promega, Madison, USA), which produced a dark purple precipitate.
253
Stain–Decolorize–Stain (SDS) method After visualization of the Masson staining, photomicrographs of the interested areas were taken with a light microscope. The slides were then soaked in xylene for 1 h to remove the coverslips and rehydrated in a serial graded ethanol from 100 to 70 %, which was prepared with DEPCtreated water. The red and green colors in Masson staining were completely dissolved in 80 and 70 % ethanol. After washing with DEPC-treated water, the slides were subjected to the ISH procedure. The hematoxylin stain was removed during the ISH microwave treatment in 0.01 M citrate buffer (pH 6.0) at 95 °C for 15 min. After taking photographs, the slides were immersed in distilled water at 50 °C for 3–5 min to remove the coverslips again and incubated with N,N-dimethylformamide (DMF) at 50 °C for 5–10 min in a shaking water bath to dissolve the NBT/BCIP dark purple precipitate. After PBS washing, the slides were treated with 3 % hydrogen peroxide and skipped microwave antigen retrieval and proceed with the first target antigen-specific antibody. After photomicrographing the interested area, slides were then soaked in 80 % alcohol for 10–30 min on a shaker to thoroughly dissolve the red AEC precipitate. The slides were then immersed in citrate buffer and subjected to another round of IHC staining using antibody for the second target antigen. This procedure began with a 2–3-min microwave (2,450 Hz, 800 W) treatment to heat the solution to 95 °C, followed with a 15-min heat (2 min off and 20 s on) to block antibody cross-reaction. Hydrogen peroxide step was used only in the first round of immunostaining. Peroxidase-labeled secondary antibody and AEC enzyme substrate identical to those used in the first round of IHC staining were used in the second and subsequent rounds of staining. Target antigens were all visualized with red AEC signal. After taking photomicrographs, the slides were again decolorized and used for IHC of subsequent target antigens as described above. After the last antigen decolorization, the same section was processed for H&E staining. Multiple staining of five target antigens and H&E staining on the same pancreatic tissue section Five antigens including insulin, glucagon, somatostatin, pancreatic polypeptide and cytokeratin were stained sequentially on a single pancreatic tissue section with this new technique to demonstrate the distributions of the five target antigens. The distributions of B cells, A cells, D cells, PP cells and ductal epithelium, of which these antigens are expressed adjacent to one another, were clearly demonstrated independently and sequentially on the same tissue section. After decolorizing the signal of the last
13
254
target antigen, the section was processed for H&E staining before image processing (described later). Multiple stain combining Masson staining, ISH and two target antigens on the same placental tissue section Masson staining was employed on a placental tissue section to stain the collagenous fibers in the villous stroma. After decoloration, the placental alkaline phosphatase (PLAP) mRNA was demonstrated with ISH using an antisense probe, which is a marker of the syncytiotrophoblast. Between the two procedures, the above-described decolorization method was used to dissolve the color of Masson staining. Following ISH, the NBT/BCIP precipitate was dissolved as described above. Two antibodies to E-cadherin and CD34, respectively, were sequentially applied on the same section with the above SDS method to demonstrate the distributions of cytotrophoblasts and endothelial cells in the chorion villi. After decolorizing the signal of the last antigen, the section was processed for H&E staining. Photographs were taken after each staining for subsequent image processing. In situ hybridization with another cRNA probe (targeted IGHM mRNA) followed with sequential immunostaining of three antigens (IgG, IgA and CD20) was also tested on the human tonsil tissue section. Antigens with similar cellular localization within the same cells In order to evaluate whether the SDS technique could be utilized to demonstrate several different antigens sharing a similar or identical cellular localization, 4 antigens on the same breast ductal carcinoma tissue section were tested. They included PCNA, E-cadherin, cytokeratin and Ki67 that were sequentially demonstrated on the same tissue section of a breast carcinoma with the SDS technique. Among them, PCNA and Ki67 are two nucleus antigens, E-cadherin a membranous antigen and cytokeratin an epithelial cytoplasmic antigen. Image processing To achieve an integrated picture that displays Masson staining, ISH and two antigens on the same placental tissue section, all the five antigens on one pancreatic tissue section or all the four antigens on one breast carcinoma tissue section at the same time, Adobe Photoshop (Adobe Systems, Mountain View, CA) was used for image processing. The Photoshop functions, “Extract” and “Replace Color”, were used for color replacement. The “Extract” function extracts a defined color, followed by replacing it with a different color with the “Replace Color”. For each set of photographs, one was chosen as a background. The other positive
13
Histochem Cell Biol (2014) 141:251–262
red signals were replaced with other distinct colors. These images were then superimposed onto the background photograph to form an integrated picture. To evaluate co-localization of two or more staining positivities, the “Blending Options” and “Overlay” functions of Photoshop were used to generate the new image where the overlapping of two colors resulting in a distinct third color, thereby clearly indicating collaboration. For the placental tissue section, pictures were taken after each staining to separately show the distributions of collagen, PLAP mRNA and the two antigens. The photograph of Masson staining was taken as a background, while the dark purple ISH positive signal and the two red IHC signals were artificially changed with Photoshop to cyan, blue and yellow, respectively, to assign each positive cell or structure a distinct color for easy recognition. The distinct colors representing different cells or structures were superimposed onto the background image to obtain an integrated photograph within which the distribution of all positive cells or structures and their relationships to one another were clearly illustrated. For the pancreatic tissue section, five pictures were obtained which separately showed the distribution of the five antigens. The image processing was the same to that described above. The photograph of cytokeratin staining was used as a background, and the red signals in the other four pictures were altered with Photoshop to yellow, purple, green and black, respectively, to make each antigen easily recognizable. The different photographs of different antigens from the same areas were then superimposed onto the background image, and the distributions of all antigens of interest were illustrated with different colors on the integrated picture. For breast ductal carcinoma tissue section, four photographs of the same area were taken. The photograph of cytokeratin staining was used as a background. The red positive signals of the other three images were altered with Photoshop to green, yellow and blue, respectively. As a new layer, the altered color of each photograph was superimposed onto the background image sequentially, and the function of “Overlay” was used to illustrate the color mixture of PCNA and Ki67 in the same nuclei. The distributions of all antigens in the same area were shown with different colors on the integrated photograph. Controls For each round of multiple IHC staining, the primary antibody was replaced with PBS as a negative control. For PLAP ISH staining, a cRNA sense probe was used as a negative control. We also controlled the specificity and sensitivity of this new technique. To control specificity, we omitted the
Histochem Cell Biol (2014) 141:251–262
primary antibody but used the secondary antibody or AEC alone, or both following the elution of the previous round of immunostaining. To control the sensitivity, we compared the staining intensities of the same antigen when it was stained first and last on the same tissue section.
Results Multiple staining of sequential Masson staining, ISH and IHC With this technique, the collagen fibers, PLAP mRNA and two antigens, E-cadherin and CD34, were sequentially demonstrated on the same placental tissue section (Fig. 1a–d). No cross-reaction between antibodies was observed. After photo image processing, the different cellular components were clearly identified in their own residence (Fig. 1e). The syncytiotrophoblasts (cyan) lay in the outer layer of the villi, while the cytotrophoblasts (dark blue) in the inner layer. Vessels with endothelial cells are shown in yellow. Around the vessels are collagen fibers shown in green. In addition, an H&E-stained photograph was also obtained of the same area for morphologic evaluation (Fig. 1f). The results of a combination of in situ hybridization with cRNA probe to IGHM mRNA and sequential immunostaining of three antigens on the same tonsil tissue section are shown in Supplementary Fig. S4.
255
on the breast cancer tissue section were identified by the membranous positive signals with the E-cadherin-specific antibody (Fig. 3b) or the cytoplasm positive signals with cytokeratin-specific antibody (Fig. 3c). No cross-reaction between antibodies was observed. After image processing, we could easily identify that in the carcinoma nest, some nuclei were positive for PCNA (green) only or Ki67 (blue) only (Fig. 3e), while some were positive for both (black, which was the overlapping of green and blue colors; arrowheads). The breast ductal carcinoma cell membranes were also clearly visible with E-cadherin staining (yellow) and the cytoplasm with cytokeratin in red. Different antigens with the same cellular localization were also tested with other tissues including human colon adenocarcinoma, heart and tonsil (Supplementary Fig. S3). The antigen-bound primary and secondary antibodies were all completely eluted from the tissue sections, and staining of a different antigen afterward was always successful without bringing up the positive signal of the previous antigen (Supplementary Fig. S1). When the same antigen was stained in the first round or the third round, there was no visible difference between the two positive signals (Supplementary Fig. S2). All negative and positive controls gave appropriate results, thereby establishing the specificity and sensitivity of this technique. The blue hematoxylin nuclear stain was completely removed with the microwave treatment.
Discussion Multiple staining of five antigens and H&E staining on a single tissue section Five antigens including insulin, glucagon, somatostatin, PP and cytokeratin were stained on a single pancreatic tissue section with this SDS technique (Fig. 2a–e). No cross-reaction was observed. Using superimposed photographs, the AEC red color for each antigen was replaced with a new color distinctly different from the color assigned to others. The distributions of these antigens and the cells in which they reside were clearly identified. The PP cells (black), A cells (green) and D cells (purple) are located at the periphery of the islet, while the B cells (yellow) occupy the central portion. Cytokeratin (red)-positive signal marked ductal epithelial cells within the exocrine tissue (Fig. 2f). An H&E staining was also obtained of the same area (Fig. 2g), offering clear morphologic information in addition to the immunohistochemical signals. Staining of antigens with the same cellular localization Numerous nuclei were positive for the antigens PCNA (Fig. 3a) and Ki67 (Fig. 3d). The ductal carcinoma cells
In this study, we developed protocols that can simultaneously demonstrate multiple cellular components including antigens, nucleotide sequences and chemical compounds on the same tissue section to discern their distribution and relationship. This technique combines IHC, ISH and histochemistry as well as H&E on a single tissue section, providing a powerful tool to meet the increasing demand for simultaneous visualization of multiple components at the same location on the same tissue section. About specificity and sensitivity A few critical points should be discussed for this new technique. One is the specificity, i.e., only the desired targets are specifically labeled without undesirable staining. The specificity was established with the following example. After positive staining of one antigen (MAP2 in neuron cytoplasm), the positive color was removed completely with 80 % ethanol followed by microwave treatment (remove the antibodies). The effectiveness of the above treatment was verified by the negative reaction when AEC and the secondary antibody were used
13
256
Histochem Cell Biol (2014) 141:251–262
Fig. 1 Multiple staining of sequential Masson staining, in situ hybridization and immunohistochemistry staining on a single placenta tissue section. a–f Are the same area of a placental tissue section. a–d Are sequentially staining demonstrating collagen fibers (green), PLAP mRNA (dark purple), E-cadherin (red) and CD34 (red), which, respectively, show the distributions of collagen fibers,
syncytiotrophoblasts, cytotrophoblasts and vascular endothelial cells on the same placental section. e Is a superimposed photomicrograph showing signals for collagen fibers in green, syncytiotrophoblasts in cyan, cytotrophoblasts in blue and vascular endothelial cells in yellow. f Shows a photograph of H&E staining in the same field after color elution. All scale bars 20 μm
alone or in combination. This was followed by another immunostaining (factor VIII for endothelial cells), which showed distinct positivity in endothelial cells. Next, the first antigen (MAP2) was stained again, and the positive MAP2 cytoplasmic signal was demonstrated again. The specificity and effectiveness of the color and antibody
elusion treatments were tested with many antibody and antigen combinations, and the specificity was repeatedly confirmed. In addition, the sensitivity of the test was not sacrificed with repeated elusion and staining steps as the subsequent positive signal had the same intensity to the initial staining of the same antigen. Therefore, the targets
13
Histochem Cell Biol (2014) 141:251–262
257
Fig. 2 Multiple staining of a single pancreas tissue section. a–g Are the same field of a pancreas tissue section. a–d and e Show the sequentially immunostaining results of pancreatic polypeptide, insulin, cytokeratin, somatostatin and glucagon on this pancreas tissue section (red signals), and no cross-reaction is identifiable. PP cells, B cells, ductal epithelium, D cells, A cells are marked, respectively,
by red coloration in a–d and e. f Is a superimposed photomicrograph showing signals for pancreatic polypeptide in black, insulin in yellow, cytokeratin in red, somatostatin in purple and glucagon in green color. g Shows a photomicrograph of the same field after H&E staining. All scale bars 20 μm
can be easily detected without losing reactivity during the relatively long procedure. The new technique is also practical. It can be carried out as a routine laboratory technique without demanding expensive equipment or challenging protocols.
About co‑localization Different antigens localized to the same cellular structure could be demonstrated with the SDS technique. Figure 3 and Supplementary Fig. S3 showed the staining of
13
258
Fig. 3 Staining of antigens in the same cellular localization. a–e Are the same field of a breast ductal carcinoma tissue section. Nuclei of proliferation cells including both cancer cells are positive for PCNA (a, red signals) and Ki67 (d, red signals) with certain overlapping that are clearly shown in black color (e, arrowheads). The breast ductal carcinoma cells are specifically identified by E-cadherin, which had a distinct membranous staining (b, red signals). The cyto-
13
Histochem Cell Biol (2014) 141:251–262
plasms of carcinoma cells show positive signals for cytokeratin (c, red signals). e Is an integrated photomicrograph showing signals for carcinoma cells’ membrane in yellow, cytoplasm in red, nucleuses solely positive for PCNA in green and for Ki67 in blue. The overlapping of PCNA and Ki67 is shown in black (arrowheads), which is the mixture of green and blue. All scale bars 20 μm
Histochem Cell Biol (2014) 141:251–262
differently paired antigens localized at the same cellular structure using this technique. Staining of the first antigen did not over shadow or affect the subsequent antigen stainings at the same location. We stained other antigens located in different cells or cellular structures in between the paired antigens to illustrate the specificity of the labeling (Fig. 3 and Supplementary Fig. S3) and obtain the desired results. About the heat treatment Since microwave heating was first introduced by Lan et al. (1995) as a simple step to block antibody cross-reactivity in sequential antigen stainings, its application has been expanded to combine with other techniques such as immunofluorescence (Tornehave et al. 2000) and spectral imaging (van der Loos 2008). We employed the heat treatment method together with chromogen dissolution for our new technique of multiple staining. With this technique, multiple staining can be carried out regardless of the species or immunoglobulin subclasses of the primary antibody. In addition, microwave treatment is an established method for in situ hybridization to expose mRNA to enhance detection sensitivity (Lan et al. 1996; Sperry et al. 1996). It has been reported that microwave treatment could not only enhance the mRNA signal to improve hybridization efficiency but also preserve tissue morphology (Lan et al. 1996; Sibony et al. 1995). As such, this method may find broad application in both diagnosis and research. Advantages of the SDS technique over other techniques The SDS technique described in this study has several advantages over previously established techniques (Tramu et al. 1978; Lan et al. 1995; Brouns et al. 2002; Sternberger and Joseph 1979). 1. Unlike the other multiple staining techniques, the various chromogens used for visualizing target antigens with different colors in this method can be replaced with AEC alone, saving both cost and effort in performing the protocol. 2. With no restriction on color selection and no need for choosing chromogens with good visual contrast, more target antigens may be stained. We demonstrated simultaneously staining of five antigens. Greater number could be achieved. 3. This SDS technique visualizes different molecules, nucleotide sequences and antigens sequentially, and the specific location of each target can be clearly demonstrated without worrying about color as the color deposit is removed by elution before proceeding to the next staining target. In addition, the shielding effects, i.e., the blocking of staining targets, often seen in mul-
259
tiple staining can be avoided. Previously, one molecule identification with color precipitation could shield others from subsequent detection. 4. This technique is uniquely suited to investigate collaboration of two or more antigens/nucleotide sequences. With the image processing software, the overlapping of two or more colors can be clearly identified and percentage calculated. 5. As the color for each cellular component is individually obtained and the final colors are artificially assigned, it would be possible to simultaneously display any combination of molecules of interest to exam their relationship on the same tissue section. 6. The same tissue section can be used for H&E staining after all the signals of multiple staining are dissolved, offering visualization of a clear morphology side-byside with visualization of nucleotide sequences, protein antigens and histochemical staining in the same visual field. This feature will be particularly beneficial for pathologic diagnosis by histopathologists who rely on H&E staining and morphology to make a judgment. Recommended protocols The sequence of the steps in SDS has been optimized, in which Masson staining should be performed first, and then ISH followed by IHC, as the Masson staining loses color easily in alcohol (70–80 % ethanol). If the section is treated with microwave first, the Masson staining would not work. ISH performed after Masson staining gave a slightly weaker positive signal than that of ISH alone, whereas there was no visible effect on IHC. Beraki et al. demonstrated successful ISH on a Giemsa prestained tissue section (Beraki et al. 2012). Some preferred antigen localization before ISH (Han et al. 1992), while others tended to carry out ISH for RNA or DNA before IHC (Rimsza et al. 1996; Beraki et al. 2012; Zaidi et al. 2000). It was reported that when ISH was performed after IHC, little or only very weak nucleotide signal could be detected with ISH (Reisenbichler et al. 2012). Our experience confirmed this claim. We performed each staining independently and compared the results with that obtained with the multiple staining procedure. It was found that no antigen/staining target was lost or gained due to the multiple staining technique. However, the multiple staining protocol does tend to weaken the staining intensity when the antigens were stained toward the end of the protocol. Therefore, we recommend that if quantitative analysis is desired, the operator of the procedure should evaluate the possible loss of reactivity by performing the staining of each target on serial section independently and compare the single staining results with that of the multiple staining side-by-side to
13
260
Histochem Cell Biol (2014) 141:251–262
Fig. 4 Principal steps of the SDS technique illustration. The optimal order of stainings is Masson stain first, and then in situ hybridization followed by immunohistochemistry with microphotographing and
color elution in between each separate staining. Microwave treatment is used for blocking antibody cross-reaction between each step of IHC. H&E staining can be performed by the end
make a determination of the quantitative differences and to control possible cross-reactivity among different steps. A step-by-step illustration of the SDS technique is shown in Fig. 4. In addition, probe concentration of ISH should be slightly higher than that for ISH alone, so is antibody concentration in the subsequent IHC. In addition to the order of technical steps, attention should be paid to the order of antibodies applied. Staining sequence of the antigens was rarely discussed when multiple staining techniques were introduced. We established a parameter for IHC as a reference to optimize the antibody incubation sequence. We named this parameter the “color developing duration (CDD) value.” It is defined as the time it takes from adding AEC solution to the moment that the target antigen becomes clearly visible. The CDD value for each antigen staining can be obtained easily in separate experiments. We found that the value differed with different target antigens, antibody manufacturers, and even diluents. The optimal staining sequence was designed by arranging target antigens from the highest CDD value to the lowest. That is, the faster the AEC signal of an antigen develops or the more easily an antigen could be stained, the later it should be employed in this SDS technique. In this way, multiple positive signals can be detected easily. We found no evidence of cross-reaction among the antibodies
or techniques in any of our experiments. The CDD value for any chromogen is a flexible time that is dependent on numerous factors including almost all steps and procedures of IHC staining process. For example, when doing heat-induced antigen retrieval (AR) pretreatment, various protocols of AR significantly influence the results of IHC including the “CDC” time. The operator should pay attention to repeating microwave treatment process (as well as other elute treatments) that may add some additional influence factors to not only the “CDD” value but also the intensity of IHC or ISH. This should be taken into consideration when quantitative evaluation of the results is the desired outcome. This technique is suitable for simultaneous demonstration of multiple antigens/targets to study their relationship but may not be ideal for quantitative evaluation. In conclusion, we have established a new method of multiple staining that combines histochemistry, in situ hybridization and immunohistochemistry. It is capable of detecting co-localization of multiple molecules at the same location. The specificity and sensitivity of this new technique were tested and validated. We also introduced a CDD value as an index to optimize the order of steps performed in this technique. It offers new possibility to comparatively analyze the distribution of different molecular components and their co-localization for diagnosis and research.
13
Histochem Cell Biol (2014) 141:251–262 Acknowledgments We thank Lu Yao, Na Niu, Zhuo Li, Yuxuan Liu, Yong Guo and Qi Cao from Department of Pathology of Peking University Health Science Center and Yiqun Geng, Tao Huang and Li Du from Department of Pathology of Shantou University Medical College for providing the tissue samples, comments and suggestions. Conflict of interest The authors declared no conflict of interest with respect to the research, authorship, and publication of this article. This work was supported by grants from the National Nature Science Foundation of China (No. 81030033, 30971150 to J.G.).
References Beraki E, Olsen TK, Sauer T (2012) Establishing a protocol for immunocytochemical staining and chromogenic in situ hybridization of Giemsa and Diff-Quick prestained cytological smears. Cytojournal 9:8 Boorsma DM (1984) Direct immunoenzyme double staining applicable for monoclonal antibodies. Histochemistry 80(2):103–106 Brouns I, Van Nassauw L, Van Genechten J, Majewski M, Scheuermann DW, Timmermans JP, Adriaensen D (2002) Triple immunofluorescence staining with antibodies raised in the same species to study the complex innervation pattern of intrapulmonary chemoreceptors. J Histochem Cytochem 50(4):575–582 Campbell GT, Bhatnagar AS (1976) Simultaneous visualization by light microscopy of two pituitary hormones in a single tissue section using a combination of indirect immunohistochemical methods. J Histochem Cytochem 24(2):448–452 Carl SAL, Gilleteferguson I, Ferguson DG (1993) An indirect immunofluorescence procedure for staining the same cryosection with 2 mouse monoclonal primary antibodies. J Histochem Cytochem 41(8):1273–1278 Chen Z, Gu J (2007) Immunoglobulin G expression in carcinomas and cancer cell lines. Faseb J 21(11):2931–2938 Glass G, Papin JA, Mandell JW (2009) SIMPLE: a sequential immunoperoxidase labeling and erasing method. J Histochem Cytochem 57(10):899–905 Grin A, Brezden-Masley C, Bauer S, Streutker CJ (2013) HER2 in situ hybridization in gastric and gastroesophageal adenocarcinoma: comparison of automated dual ISH to FISH. Appl Immunohistochem Mol Morphol Gu J, De Mey J, Moeremans M, Polak JM (1981) Sequential use of the PAP and immunogold staining method for the light microscopical double staining of the tissue antigens. Regul Pept 1(6):365–374 Han KH, Hollinger FB, Noonan CA, Yoffe B (1992) Simultaneous detection of HBV-specific antigens and DNA in paraffinembedded liver tissue by immunohistochemistry and in situ hybridization using a digoxigenin-labeled probe. J Virol Methods 37(1):89–97 Hunyady B, Krempels K, Harta G, Mezey E (1996) Immunohistochemical signal amplification by catalyzed reporter deposition and its application in double immunostaining. J Histochem Cytochem 44(12):1353–1362 Ino H (2004) Application of antigen retrieval by heating for double-label fluorescent immunohistochemistry with identical species-derived primary antibodies. J Histochem Cytochem 52(9):1209–1217 Kimura M, McGinnis MR (1998) Fontana-Masson–stained tissue from culture-proven mycoses. Arch Pathol Lab Med 122(12):1107–1111 Lan HY, Mu W, Nikolicpaterson DJ, Atkins RC (1995) A novel, simple, reliable, and sensitive method for multiple immunoenzyme
261 staining—use of microwave-oven heating to block antibody cross-reactivity and retrieve antigens. J Histochem Cytochem 43(1):97–102 Lan HY, Mu W, Ng YY, Nikolic-Paterson DJ, Atkins RC (1996) A simple, reliable, and sensitive method for nonradioactive in situ hybridization: use of microwave heating to improve hybridization efficiency and preserve tissue morphology. J Histochem Cytochem 44(3):281–287 Larsson LI, Childers S, Snyder SH (1979) Met- and leu-enkephalin immunoreactivity in separate neurons. Nature 282:407–410 Nakane PK (1968) Simultaneous localization of multiple tissue antigens using peroxidase-labeled antibody method—a study on pituitary glands of rat. J Histochem Cytochem 16(9):557–560 Negoescu A, Labatmoleur F, Lorimier P, Lamarcq L, Guillermet C, Chambaz E, Brambilla E (1994) F(Ab) secondary antibodies—a general-method for double immunolabeling with primary antisera from the same species—efficiency control by chemiluminescence. J Histochem Cytochem 42(3):433–437 Reisenbichler ES, Horton D, Rasco M, Andea A, Hameed O (2012) Evaluation of dual immunohistochemistry and chromogenic in situ hybridization for HER2 on a single section. Am J Clin Pathol 137(1):102–110 Riegler J, Nann T (2004) Application of luminescent nanocrystals as labels for biological molecules. Anal Bioanal Chem 379(7–8):913–919 Rimsza LM, Vela EE, Frutiger YM, Rangel CS, Solano M, Richter LC, Grogan TM, Bellamy WT (1996) Rapid automated combined in situ hybridization and immunohistochemistry for sensitive detection of cytomegalovirus in paraffin-embedded tissue biopsies. Am J Clin Pathol 106(4):544–548 Shindler KS, Roth KA (1996) Double immunofluorescent staining using two unconjugated primary antisera raised in the same species. J Histochem Cytochem 44(11):1331–1335 Sibony M, Commo F, Callard P, Gasc JM (1995) Enhancement of mRNA in situ hybridization signal by microwave heating. Lab Invest 73(4):586–591 Sperry A, Jin L, Lloyd RV (1996) Microwave treatment enhances detection of RNA and DNA by in situ hybridization. Diagn Mol Pathol 5(4):291–296 Sternberger LA, Joseph SA (1979) The unlabeled antibody method. Contrasting color staining of paired pituitary hormones without antibody removal. J Histochem Cytochem 27(11):1424–1429 Tidman N, Janossy G, Bodger M, Granger S, Kung PC, Goldstein G (1981) Delineation of human thymocyte differentiation pathways utilizing double-staining techniques with monoclonal antibodies. Clin Exp Immunol 45(3):457–467 Tornehave D, Hougaard DM, Larsson LI (2000) Microwaving for double indirect immunofluorescence with primary antibodies from the same species and for staining of mouse tissues with mouse monoclonal antibodies. Histochem Cell Biol 113(1):19–23 Tramu G, Pillez A, Leonardelli J (1978) An efficient method of antibody elution for the successive or simultaneous localization of two antigens by immunocytochemistry. J Histochem Cytochem 26(4):322–324 Valnes K, Brandtzaeg P (1982) Comparison of paired immunofluorescence and paired immunoenzyme staining methods based on primary antisera from the same species. J Histochem Cytochem 30(6):518–524 van der Loos CM (2008) Multiple immunoenzyme staining: methods and visualizations for the observation with spectral imaging. J Histochem Cytochem 56(4):313–328 Van der Loos CM, Das PK, Van den Oord JJ, Houthoff HJ (1989) Multiple immunoenzyme staining techniques. Use of fluoresceinated, biotinylated and unlabelled monoclonal antibodies. J Immunol Methods 117(1):45–52
13
262 Wallace EF, Wofsy L (1979) Hapten-sandwich labeling. IV. Improved procedures and non-cross-reacting hapten reagents for doublelabeling cell surface antigens. J Immunol Methods 25(3):283–289 Wang BL, Larsson LI (1985) Simultaneous demonstration of multiple antigens by indirect immunofluorescence or immunogold staining. Novel light and electron microscopical double and triple staining method employing primary antibodies from the same species. Histochemistry 83(1):47–56
13
Histochem Cell Biol (2014) 141:251–262 Ye J, Zhang B, Xu J, Chang Q, McNutt MA, Korteweg C, Gong E, Gu J (2007) Molecular pathology in the lungs of severe acute respiratory syndrome patients. Am J Pathol 170(2):538–545 Zaidi AU, Enomoto H, Milbrandt J, Roth KA (2000) Dual fluorescent in situ hybridization and immunohistochemical detection with tyramide signal amplification. J Histochem Cytochem 48(10):1369–1375
