- Email: [email protected]
Simultaneous detection of two independent antigens by double staining with two mouse monoclonal antibodies
Journal of Virological Methods 73 (1998) 89 – 97
Simultaneous detection of two independent antigens by double staining with two mouse monoclonal antibodies Norihiro Teramoto a,b,*, Laszlo Szekely a, Katja Pokrovskaja a, Li Fu Hu a, Tadashi Yoshino b, Tadaatsu Akagi b, George Klein a b
a Microbiology and Tumor Biology Center, Karolinska Institute, S171 77, Stockholm, Sweden Department of Pathology, Okayama Uni6ersity Medical School, 2 -5 -1 Shikata-cho, Okayama, Japan
Received 15 October 1997; received in revised form 2 March 1998; accepted 2 March 1998
Abstract Simultaneous detection of two antigens by immunostaining usually requires primary antibodies from two different species or a hapten modification of one of the antibodies if they are from the same species. A novel double staining method is described for immunodetection of two independent antigens using two mouse monoclonal antibodies. The principle of the method is the following: The first antigen is detected by a monoclonal antibody that is diluted so extensively that it cannot be recognized with conventional detection systems. A highly sensitive biotin – tyramide amplification system is used to identify this antibody. The second antigen is stained with a monoclonal antibody by dilution and detected by conventional immunostaining. The method was tested for both alkaline-phosphatase staining on paraffin sections and immunofluorescence staining on cultured cells in cytospin preparation. The absence of cross-reaction in the former system was demonstrated by the mutually exclusive detection of T- and B-cells in human lymph nodes or T-cells and carcinoma cells in nasopharyngeal carcinoma biopsies. Similarly, the EBV encoded EBNA2 and ZEBRA proteins showed a mutually exclusive staining by immunofluorescence on B95-8 cells. The method could be used to demonstrate co-expression of two independent antigens in the same cells, such as PCNA and keratin in carcinoma cells in paraffin sections and for EBNA2 and LMP1 EBV proteins in immunofluorescence preparations of B95-8 cells. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Double staining method; Independent antigens; Mouse monoclonal antibodies
Abbre6iations: AEC, 3-amino-9-ethyl carbazole; APAAP, alkaline-phosphatase anti-alkaline phosphatase; BCIP/NBT, 5-bromo-4chloro-3-indolyl phosphate/nitro-blue tetrazolium; CSA, catalyzed signal amplification; DAB, 3’3-diaminobenzidine; DABCO, 1,4-diazabicyclo (2.2.2) octane; EPOS, enhanced polymer one-step staining; FITC, fluorescein isothiocyanate; Mab, monoclonal antibody; NPC, nasopharyngeal carcinoma; PCNA, proliferating cell nuclear antigen; TRITC, tetra-methylrhodamine isothiocyanate isomer R. * Corresponding author. Tel.: + 46 8 7286746; fax: + 46 8 330498; e-mail: [email protected] 0166-0934/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S0166-0934(98)00048-2
90
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
1. Introduction Immunostaining is used widely for the detection and measurement of protein expression. Double staining with two antibodies, derived from two different animal species, that are directed against two different proteins is quite straightforward (Mason et al., 1983; Key et al., 1993). No reliable method exists, however, that would permit the simultaneous use of two different unlabeled antibodies from the same species due to the unavoidable cross-reactions between the secondary antibodies. These cross-reactions occur at two different levels: 1. the first antibody is recognized by the conjugated antibody reagent targeted against the second antibody 2. the conjugated reagent that is already bound to the first antibody captures molecules when the second antibody is applied Earlier, it was suggested that it is possible to carry out double staining with two independent, unlabeled mouse monoclonal antibodies (Mab-s) if the first Mab is detected with peroxidase conjugated secondary antibody that catalyzes 3’3-diaminobenzidine (DAB) deposition. The deposited DAB was supposed to mask the first Mab and the peroxidase conjugated secondary antibody and prevent the interaction with the antibodies in the subsequent staining steps (Valnes and Brandtzaeg, 1982, 1984; Chen et al., 1987). This method was not free of cross-reactions, however. Moreover the extensive deposition of DAB limited the spatial resolution (Valnes and Brandtzaeg, 1984). The regular method of double staining with two Mabs involves direct labeling of one antibody with small haptens such as fluorescein isothiocyanate (FITC), biotin, gold particles or enzymes such as alkaline phosphatese or horseradish peroxidase (Riesenberg et al., 1993; Shibuya et al., 1993). These labels can either directly visualize the modified Mab or be detected by a second layer of antibody directed against the conjugated material. As an alternative, the primary antibody can be labeled with a large complex of enzymes sitting on a polymer backbone in the so called enhanced polymer one-step staining (EPOS) method (Pastore et al., 1995). Labeling of the primary anti-
body is tedious and expensive, however and the chemical modification can destroy the antigen binding site. Recently, a new staining enhancement system was introduced, based on peroxidase-mediated deposition of biotin–tyramide around the primary antibody (Bobrow et al., 1989; Merz et al., 1995). This staining system, designated as CSA (catalyzed signal amplification), is now available commercially from Dakopatts (Glostrup, Denmark). The optimal concentration of primary antibodies in the CSA method is more than 50 times lower than in other detection systems (Merz et al., 1995). A novel double staining protocol is described now using two unlabeled mouse Mabs. The method is based on the difference in sensitivity between the CSA method and the conventional detection systems. The first mouse monoclonal antibody is applied in a dilution that does not allow direct visualization by the second detection system. This first Mab is visualized by the CSA method. In the subsequent steps the second unlabeled mouse monoclonal antibody is applied and detected by a conventional enzyme or fluorochrome conjugated antibody. We used several mouse monoclonal antibodies to assess the validity of the method, including mouse monoclonal antibodies against lymphoid markers, a proliferation marker and different viral proteins. Three EBV-encoded proteins were studied, EBNA2, LMP1 and ZEBRA. EBNA2 and LMP1 are proteins associated with latency that contribute to the EBV mediated growth transformation of B-lymphocytes. They are expressed in EBV-immortalized lymphoblastoid cell lines (LCL-s), but not in type I Burkitt lymphoma cells (Rowe et al., 1992). ZEBRA protein (encoded by the open reading frame BZLF-1) is associated with the viral lytic cycle. It initiates a cascade of lytic phase protein expression (Miller, 1990). In EBV-transformed LCL-s only very few cells express the ZEBRA protein. In the present paper, we demonstrate that the double staining method has a wide range of application, including staining of paraffin sections and
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
91
Table 1 Antibodies used in this study Antibody to
CD3 (D) CD45RO* CD79a (D) PCNA** (D) Cytokeratin*** (D) EBNA2* LMP1 (D) ZEBRA (D) APAAP, mouse monoclonal (D) TRITC-conjugated rabbit anti-mouse Ig (D) Peroxidase-conjugated sheep anti-mouse Ig (A) FITC-conjugated streptavidin (P)
Clone
UCHT1 UCHL1 JCB117 PE10 MNF116 PE2 CS1-4 BZ-1 AP7/6/7 — — —
Dilution of antibodies 1sta
2ndb
1:5000 1:200 1:2500 1:5000 — 1:40 — — — — — —
— 1:2 1:25 — 1:50 1:1 1:30 1:20 1:50 1:30 1:1000 1:50
Antigen retrival procedure
Autoclaving in H2O — Autoclaving in H2O Autoclaving in H2O Trypsinization — — — — — — —
All antibodies are mouse monoclonal unless otherwise stated. D, Dakopatts; A, Amersham; P, Pharmingen. a Dilution of monoclonal antibodies for the 1st antigen detected by the CSA method. b Dilution of antibodies for the 2nd antigen or as secondary antibodies. * Supernatant of cultured hybridoma cell lines; ** proliferating cell nuclear antigen; *** MNF116 reacts with a wide spectrum of keratin Nos. 5, 6, 8, and 17.
immunofluorescent detection of proteins on single cell level.
2. Materials and methods
2.1. Specimens Sections (3–5 mm) were cut from paraffin blocks that were prepared by routine procedures. Biopsy specimens included 3 nasopharyngeal carcinomas (NPC), 3 lung cancers, 4 reactive lymph nodes, and 1 tonsil. The following cell lines were used for double immunofluorescence staining: an EBV producing marmoset lymphoblastoid cell line, B95-8; an in vitro EBV transformed human lymphoblastoid line, SIB1; an EBV positive, type I Burkitt lymphoma line, Rael; EBV negative mouse myeloma line, SP2/0. All cells were cultured at 37°C in 5% CO2 in RPMI 1640 medium containing 10% fetal calf serum. Before staining, the human cells were mixed in 4 – 6:1 proportion with the internal staining control SP2/0 cells. The mouse and human cells can be distinguished read-
ily by their characteristic chromatin distribution using Hoechst staining. Cytospin smears were prepared by centrifuging the cells on glass slides for 2 min at 900 rpm. The cells were fixed immediately, without drying, in acetone for 20 min. B95-8 and SIB1 are known to express EBNA2 and LMP1. About 2–4% of the B95-8 cells express ZEBRA. Rael is non-producer cell line. It expresses only EBNA1 as the sole EBV-encoded protein (Masucci et al., 1989).
2.2. Double staining procedures The sections were deparaffinized and rehydrated in xylene and a descending alcohol series. If necessary, antigen retrieval procedures were used (Goddard et al., 1991; Ryong et al., 1991). The combination of the retrieval procedures and mouse monoclonal antibodies are listed in Table 1. In order to determine the optimal conditions for the first antibody, mouse monoclonal antibodies which were serially diluted were visualized by alkaline-phosphatase anti-alkaline phosphatase (APAAP) after 15 min incubation at room temperature (Cordell et al., 1984). Dilution that gave
92
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
no detectable signal with APAAP was used for CSA detection. For double staining, the rehydrated paraffin sections were soaked in 5% H2O2 for 5 min, followed by incubation with blocking solution supplied in the CSA kit (Dakopatts, Glostrup, Denmark) for at least 5 min at room temperature. Subsequently the slides were incubated with the highly diluted first mouse monoclonal antibody (listed in Table 1) for no more than 16 min. The sections were processed according to the supplier’s instructions for the CSA kit, and color development was carried out using DAB or 3-amino-9-ethyl carbazole (AEC). The sections were incubated with the second mouse monoclonal antibody at the concentration indicated in Table 1 at 4°C overnight. The second Mab was visualized using the APAAP method with a 5bromo-4-chloro-3-indolyl phosphate/nitro-blue (BCIP/NBT) substrate. The sections were counterstained with methyl green and mounted with Tris-buffered glycerol. For double immunofluorescence staining, the cell smears were incubated with 1.5% blocking solution (Boehringer Manheim, Manheim, Germany) for at least 20 min. The blocking was followed by incubation with the highly diluted first mouse monoclonal antibody (listed in Table 1) for no more than 16 min. The cells were incubated with peroxidase-conjugated sheep antimouse antibody (1:1000, Amersham, Buckingham, UK) for 15 min and subsequently for 15 min with the amplification solution of the CSA kit to generate the biotinylated tyramid deposits. The second mouse monoclonal antibodies were used at concentrations indicated in Table 1 at 4°C, overnight. The biotin – tyramide deposits corresponding to the first Mab and the second Mab itself were detected simultaneously by a mixture of FITC-conjugated streptavidin and tetramethylrhodamine isothiocyanate isomer R (TRITC)-conjugated rabbit anti-mouse immunoglobulin, respectively. The mixture also contained 40 ng/ml Hoechst 33258 to visualize DNA. The slides were washed regularly three times with PBS between the different staining steps in both staining protocols. The stained cells were mounted with 70% glycerol, 2.5% 1,4-diazabicyclo
(2.2.2) octane (DABCO) (Sigma), 50 mM Tris– HCl, pH 8.5. The microscopy, image capturing with a cold CCD camera and subsequent image processing was carried out as previously described (Szekely et al., 1995). In the staining controls, color development for the first mouse monoclonal antibodies by DAB, AEC or FITC-conjugated streptavidin or incubation with the second mouse monoclonal antibodies was omitted. Cross-reactions were assessed by detecting the bound first mouse monoclonal antibody with the second detection system.
3. Results
3.1. Double immunochemical staining on paraffin sections The proper dilution of the first Mab is crucial for the successful double staining using CSA as the first staining step. It should be under the detectability by the conventional immunostaining procedure that is used to detect the second Mab. Fig. 1 shows the results of the antibody titration (1:50, 500, 5000) for the detection of proliferating cell nuclear antigen (PCNA) in the germinal center of a human lymph node using the APAAP procedure. After 15 min incubation, PCNA was detected at 1:50 and 1:500 but not at 1:5000 dilution using the APAAP procedure (Fig. 1a–c), whereas CSA clearly visualized PCNA even at 1:5000 dilution (Fig. 2e and g) indicating that 1:5000 dilution is suitable for double staining. It was found that most mouse monoclonal antibodies were below the detection level of APAAP but above the sensitivity of CSA, when they were diluted 50–100 times more than their usual working concentration. The mutually exclusive double staining pattern on paraffin embedded material was demonstrated by staining CD45RO-positive (primed) T-cells and CD79a-positive B-cells in human reactive lymph nodes and tonsil sections (Fig. 2a). A similar mutually exclusive staining pattern was obtained when CD45RO-positive T-cells (Fig. 2c) or CD79a-positive B-cells (data not shown) were stained in combination with anti-cytokeratin
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
93
Fig. 1. Sensitivity titration of the first monoclonal antibody to determine the dilution that is no longer detected by the conventional detection system reserved for the second monoclonal antibody. Different dilutions of anti-PCNA antibody (PE10) were incubated with a lymph node section at room temperature for 15 min and visualized by the APAAP method. PCNA was strongly stained in the germinal center of a lymph node at 1:50 (a), faintly stained at 1:500 (b), and was not detectable at 1:5000 dilution (c).
(MN116) antibody that visualized epithelial cells in tonsil or three different nasopharyngeal carcinoma sections. The staining pattern was clear and well defined even if the antigens were adjacent to each other indicating that biotin – tyramide and subsequent DAB deposition did not block second antibody reaction (Fig. 2a and c). The CSA/APAAP double staining was able to resolve antigens that were present in the same cell, even on paraffin sections, as long as one antigen was nuclear and the other was associated with the cytoplasm. For example PCNA nuclear staining could be detected in germinal center B-cells that were visualized by CD79a antibody (Fig. 2e). PCNA was also detected in the nuclei of lung carcinoma cells, identified by anti-cytokeratin antibody (Fig. 2g). No staining was seen in the control sections on which DAB or AEC color development and the incubation with the second mouse monoclonal antibody were omitted. This shows that the first mouse monoclonal antibodies was not visualized by the APAAP method (Fig. 2b, d, f and h).
3.2. Double immunofluorescence staining on single cell le6el EBV encoded viral antigens with known expression pattern were used to explore the feasibility of CSA based double staining for fluorescence detection. EBNA2, one of the latency associated nuclear antigens, was visualized in the nuclei of B95-8 and SIB1 cells by FITC-conjugated strep-
tavidin that detected the biotinylated tyramide deposits that formed at the site of the antigen (Fig. 3a and e). The fine intranuclear resolution of the immunofluorescence, characteristic for EBNA-2 by conventional fluorescence staining, was lost, however. Approximately 2% of the B958 cells expressed the lytic cycle associated EBV protein, ZEBRA. Double immunofluorescence staining where EBNA-2 was detected by PE-2 monoclonal antibody followed by CSA and FITC-conjugated streptavidin reaction (Fig. 3a), and the ZEBRA protein by Mab BZ-1 followed by conventional staining with TRITC-conjugated anti-mouse serum (Fig. 3b) showed that the ZEBRA positive cells were negative for EBNA2 (Fig. 3c). No cross-reaction was detected between the two staining systems as demonstrated by capturing nonoverlapping red and green monochromatic signals from the same cells using a cold CCD camera. 30–50% Of the EBNA-positive B95-8 and SIB cells were also positive for the membrane associated viral protein LMP-1. LMP-1 was localized in membrane associated patches on the cell surface and scattered foci in the cytoplasm. Double staining of EBNA-2 and LMP-1 with unlabeled mouse Mab-s of PE-2 and CS1-4, respectively, was used to carry out CSA based double fluorescence staining within the same cell. EBNA-2 is a known positive regulator of LMP-1 expression. Double immunostaining of the two proteins showed that the LMP1-positive cells expressed relatively high amounts of EBNA2.
94
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
Fig. 2
Fig. 3
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
The following controls were included in these experiments: the Burkitt lymphoma cell line Rael, known to express EBNA-1 exclusively, was consistently negative for PE-2, BZ-1 and CS1-4 (Fig. 3d). All preparations used for double staining contained 15– 25% EBV negative Sp2/0 mouse myeloma cells. These cells were easily recognized by their pericentromeric heterochromatin granules that gave them a characteristic appearance on Hoechst 33258 DNA staining (Fig. 3, nuclei labeled with an ). Mab anti-CD3 antibody that did not react with any of the examined B-cell lines was used as a negative control, included as the first reagent in the control stainings (Fig. 3h).
4. Discussion The possibility was demonstrated to use, simultaneously, two mouse monoclonal antibodies, without any chemical modification, for double immunostaining without cross-reaction. The method is based on the selective sensitivity of the secondary detection systems. It is therefore very important to establish the optimal concentration of the first antibody that allows detection by the CSA method, but not by the conventional detec-
95
tion system. This needs to be determined for each antigen–antibody pair. Since the local concentration of the antigen may vary widely for different samples and since a high concentration of a given antigen may capture relatively high amounts of the first Mab, even from a very diluted solution, it is essential to run the titration for the non-CSA based detection system on parallel samples and to demonstrate the absence of any reactivity at the selected working dilution. Although the method presented only used two unlabeled mouse Mabs, it is likely that it could be applied to other unlabeled monoclonal or polyclonal antibody pairs from the same species. Combination with additional directly labeled antibodies from the same species or unlabeled antibodies from a different species might be used for triple or even higher order probing. It was shown that the method can be used for two color immunofluorescence with two different mouse monoclonal antibodies and that the visualization of the two antigens does not interfere with each other. It was found that two different antigens can be distinguished in paraffin sections, even if they are expressed in the same cell. The resolution of the CSA method is not sufficient for fine intracellular localization, due to the dissemination of the catalyzed tyramides. This dissemination might be reduced by an increase of
Fig. 2. Double staining with two mouse monoclonal antibodies on paraffin sections. The first antibodies were visualized with DAB (brown) or AEC (brownish red) using the CSA method. The second antibodies were visualized with BCIP/NBT (dark blue) using the APAAP method. (a) T-cells (anti-CD45RO, CSA method, brown) and B-cells (anti-CD79a, APAAP method, blue) show a mutually exclusive localization in a lymph node (magnification × 150). Note that adjacent cells were stained in different colors, indicating that the DAB deposition did not block the reaction. (c) Double staining of infiltrating T-lymphocytes (anti-CD45RO, CSA method, brown) and cancer cells (anti-cytokeratin, APAAP method, blue) in nasopharyngeal carcinoma tissue ( × 200). (e) PCNA (CSA method, brown) in the nuclei of germinal center B-cells (anti-CD79a, APAAP method, blue) ( × 100). (g) PCNA (CSA method, brown) in the nuclei of lung cancer cells (anti-cytokeratin, APAAP method, blue) ( × 200). (b, d, f and h) Control stainings to demonstrate the absence of cross-reaction between the first mouse monoclonal antibodies and the APAAP method. Fig. 3. Double immunofluorescence staining on single cells. Mutually exclusive expression of EBNA-2 (a) (Mab PE-2, CSA method, FITC, green) and ZEBRA protein (b) (Mab BZ-1, conventional indirect staining, TRITC, red) in B95-8 cells nuclei are counterstained with Hoechst 33258 (blue). Tri-color overlap (c). Note the absence of cross-reaction from the EBNA-2 positive cells by the TRITC-conjugated anti-mouse serum targeted against the anti-ZEBRA. Double negative staining control Rael cells (d). Detection of two independent antigens in the same cells. B95-8 cells were double stained for EBNA-2 (e) (Mab PE-2, CSA method, FITC, green) and LMP-1 (f) (Mab CS1-4, conventional indirect staining, TRITC, red). TRITC-conjugated rabbit anti-mouse antibody did not show any reaction with the nuclei of EBNA-2 positive cells. Tri-color overlap (g). Non-reactive antibody control anti-CD3 on B95-8 cells (Mab UCHT1, CSA method, FITC, green) shows no reaction. The cells indicated by an are Sp2/0 mouse myeloma cells that serve as internal negative controls.
96
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
viscosity during the deposition of tyramide, however (Van Gijlswijk et al., 1996). Using this novel method, it was shown that ZEBRA-positive B95-8 cells that have entered lytic cycle do not express EBNA2. This is in agreement with a previous paper by Sinclair et al. (1992) who showed that ZEBRA represses the activity of the Cp promoter. To our knowledge, this is the first demonstration of the single cell level of EBNA2 down-regulation in EBV-carrying lymphoid cells that have switched on ZEBRA. The LMP1 promoter is known to be activated by EBNA2 expression (Fahraeus et al., 1990). Our staining results are in line with these findings. The double staining method described above may have wide application since it can be used both for paraffin sections and cell smears. It does not require specialized techniques or apparatus.
Acknowledgements We thank Mie Inoko for her assistance in preparing the materials. This work was supported by grants from the Swedish Cancer Society (Cancerfonden), the Swedish Research Council, and by a matching grant from the Concern Foundation, LA, and the Cancer Research Institute, NY.
References Bobrow, M., Harris, T., Shaughnessy, K., Litt, G., 1989. Catalyzed reporter deposition, a novel method of signal amplification. Application to immunoassays. J. Immunol. Methods 125, 279 – 285. Chen, K., Demetris, A.J., Vanthiel, D.H., Whiteside, T.L., 1987. Double immunoenzyme staining method for analysis of tissue and blood lymphocyte subsets with monoclonal antibodies. Lab. Invest. 56, 114–119. Cordell, J., Fallini, B., Erber, W., et al., 1984. Immunoenzymatic labeling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes). J. Histochem. Cytochem. 32, 219 – 229. Fahraeus, R., Jansson, A., Ricksten, A., Sjo¨o¨blom, A., Rymo, L., 1990. Epstein-Barr virus-emcoded nuclear antigen 2
.
activates the viral latent membrane protein promoter by modulating the activity of a negative regulatory element. Proc. Natl. Acad. Sci. 87, 7390 – 7394. Goddard, M.J., Wilson, B., Grant, J.W., 1991. Comparison of commercially available cytokeratin antibodies in normal and neoplastic adult epithelial and non-epithelial tissues. J. Clin. Pathol. 44, 660 – 663. Key, G., Petersen, J.L., Becker, M.H., et al., 1993. New antiserum against Ki-67 antigen suitable for double immunostaining of paraffin wax sections. J. Clin. Pathol. 46, 1080 – 1084. Mason, D.Y., Abdulaziz, Z., Falini, B., Stein, H., 1983. Single and double immunoenzymatic techniques for labeling tissue sections with monoclonal antibodies. Ann. New York Acad. Sci. 420, 127 – 133. Masucci, M.G., Contreras-Salazar, B., Ragnar, E., et al., 1989. 5-Azacytidine up regulates the expression of Epstein-Barr virus nuclear antigen 2 (EBNA-2) through EBNA-6 and latent membrane protein in the Burkitt’s lymphoma line Rael. J. Virol. 63, 3135 – 3141. Merz, H., Malisius, R., Mannweiler, S., et al., 1995. ImmunoMax. A maximized immunohistochemical method for the retrieval and enhancement of hidden antigens. Lab. Invest. 73, 149 – 156. Miller, G., 1990. The switch between latency and replication of Epstein-Barr virus. J. Infect. Dis. 161, 833 – 844. Pastore, J., Clampett, C., Miller, J., Porter, K., Miller, D., 1995. A rapid immunoenzyme double labeling technique using EPOS reagents. J. Histotechnol. 18, 35 – 39. Riesenberg, R., Oberneder, R., Kriegmair, M., et al., 1993. Immunocytochemical double staining of cytokeratin and prostate specific antigen in individual prostatic tumour cells. Histochem. J. 99, 61 – 66. Rowe, M., Lear, A.L., Croom, C.D., Davies, A.H., Rickinson, A.B., 1992. Three pathways of Epstein-Barr virus gene activation from EBNA1-positive latency in B lymphocytes. J. Virol. 66, 122 – 131. Ryong, W.S., Iwaki, T., Kitamoto, T., Tateishi, J., 1991. Hydrated autoclave pretreatment enhances TAU immunoreactivity in formalin-fixed normal and Alzheimer’s disease brain tissues. Lab. Invest. 64, 693 – 702. Shibuya, M., Ito, S., Davis, R.L., Hoshino, T., 1993. Immunohistochemical double staining with immunogold-silver and alkaline phosphatase to identify nuclear markers of cellular proliferation. Biotech. Histochem. 68, 17 – 19. Sinclair, A.J., Brimmell, M., Farrell, P.J., 1992. Reciprocal antagonism of steroid hormones and BZLF1 in switch between Epstein-Barr virus latent and productive cycle gene expression. J. Virol. 66, 70 – 77. Szekely, L., Jiang, W.Q., Pokrovskaja, K., Wiman, K.G., Klein, G., Ringertz, N., 1995. Reversible nucleolar translocation of Epstein-Barr virus-encoded EBNA-5 and hsp70 proteins after exposure to heat shock or cell density congestion. J. Gen. Virol. 76, 2423 – 2432. 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, 518 – 524.
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97 Valnes, K., Brandtzaeg, P., 1984. Paired indirect immunoenzyme staining with primary antibodies from the same species. Application of horseradish peroxidase and alkaline phosphatase as sequential labels. Histochem. J. 16, 477– 487.
97
Van Gijlswijk, R.P.M., Wiegant, J., Raap, A.K., Tanke, H.J., 1996. Improved localization of fluorescent tyramides for fluorescence in situ hybridization using dextran sulfate and polyvinyl alcohohol. J. Histochem. Cytochem. 44, 389 – 392.
.
Simultaneous detection of two independent antigens by double staining with two mouse monoclonal antibodies Norihiro Teramoto a,b,*, Laszlo Szekely a, Katja Pokrovskaja a, Li Fu Hu a, Tadashi Yoshino b, Tadaatsu Akagi b, George Klein a b
a Microbiology and Tumor Biology Center, Karolinska Institute, S171 77, Stockholm, Sweden Department of Pathology, Okayama Uni6ersity Medical School, 2 -5 -1 Shikata-cho, Okayama, Japan
Received 15 October 1997; received in revised form 2 March 1998; accepted 2 March 1998
Abstract Simultaneous detection of two antigens by immunostaining usually requires primary antibodies from two different species or a hapten modification of one of the antibodies if they are from the same species. A novel double staining method is described for immunodetection of two independent antigens using two mouse monoclonal antibodies. The principle of the method is the following: The first antigen is detected by a monoclonal antibody that is diluted so extensively that it cannot be recognized with conventional detection systems. A highly sensitive biotin – tyramide amplification system is used to identify this antibody. The second antigen is stained with a monoclonal antibody by dilution and detected by conventional immunostaining. The method was tested for both alkaline-phosphatase staining on paraffin sections and immunofluorescence staining on cultured cells in cytospin preparation. The absence of cross-reaction in the former system was demonstrated by the mutually exclusive detection of T- and B-cells in human lymph nodes or T-cells and carcinoma cells in nasopharyngeal carcinoma biopsies. Similarly, the EBV encoded EBNA2 and ZEBRA proteins showed a mutually exclusive staining by immunofluorescence on B95-8 cells. The method could be used to demonstrate co-expression of two independent antigens in the same cells, such as PCNA and keratin in carcinoma cells in paraffin sections and for EBNA2 and LMP1 EBV proteins in immunofluorescence preparations of B95-8 cells. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Double staining method; Independent antigens; Mouse monoclonal antibodies
Abbre6iations: AEC, 3-amino-9-ethyl carbazole; APAAP, alkaline-phosphatase anti-alkaline phosphatase; BCIP/NBT, 5-bromo-4chloro-3-indolyl phosphate/nitro-blue tetrazolium; CSA, catalyzed signal amplification; DAB, 3’3-diaminobenzidine; DABCO, 1,4-diazabicyclo (2.2.2) octane; EPOS, enhanced polymer one-step staining; FITC, fluorescein isothiocyanate; Mab, monoclonal antibody; NPC, nasopharyngeal carcinoma; PCNA, proliferating cell nuclear antigen; TRITC, tetra-methylrhodamine isothiocyanate isomer R. * Corresponding author. Tel.: + 46 8 7286746; fax: + 46 8 330498; e-mail: [email protected] 0166-0934/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S0166-0934(98)00048-2
90
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
1. Introduction Immunostaining is used widely for the detection and measurement of protein expression. Double staining with two antibodies, derived from two different animal species, that are directed against two different proteins is quite straightforward (Mason et al., 1983; Key et al., 1993). No reliable method exists, however, that would permit the simultaneous use of two different unlabeled antibodies from the same species due to the unavoidable cross-reactions between the secondary antibodies. These cross-reactions occur at two different levels: 1. the first antibody is recognized by the conjugated antibody reagent targeted against the second antibody 2. the conjugated reagent that is already bound to the first antibody captures molecules when the second antibody is applied Earlier, it was suggested that it is possible to carry out double staining with two independent, unlabeled mouse monoclonal antibodies (Mab-s) if the first Mab is detected with peroxidase conjugated secondary antibody that catalyzes 3’3-diaminobenzidine (DAB) deposition. The deposited DAB was supposed to mask the first Mab and the peroxidase conjugated secondary antibody and prevent the interaction with the antibodies in the subsequent staining steps (Valnes and Brandtzaeg, 1982, 1984; Chen et al., 1987). This method was not free of cross-reactions, however. Moreover the extensive deposition of DAB limited the spatial resolution (Valnes and Brandtzaeg, 1984). The regular method of double staining with two Mabs involves direct labeling of one antibody with small haptens such as fluorescein isothiocyanate (FITC), biotin, gold particles or enzymes such as alkaline phosphatese or horseradish peroxidase (Riesenberg et al., 1993; Shibuya et al., 1993). These labels can either directly visualize the modified Mab or be detected by a second layer of antibody directed against the conjugated material. As an alternative, the primary antibody can be labeled with a large complex of enzymes sitting on a polymer backbone in the so called enhanced polymer one-step staining (EPOS) method (Pastore et al., 1995). Labeling of the primary anti-
body is tedious and expensive, however and the chemical modification can destroy the antigen binding site. Recently, a new staining enhancement system was introduced, based on peroxidase-mediated deposition of biotin–tyramide around the primary antibody (Bobrow et al., 1989; Merz et al., 1995). This staining system, designated as CSA (catalyzed signal amplification), is now available commercially from Dakopatts (Glostrup, Denmark). The optimal concentration of primary antibodies in the CSA method is more than 50 times lower than in other detection systems (Merz et al., 1995). A novel double staining protocol is described now using two unlabeled mouse Mabs. The method is based on the difference in sensitivity between the CSA method and the conventional detection systems. The first mouse monoclonal antibody is applied in a dilution that does not allow direct visualization by the second detection system. This first Mab is visualized by the CSA method. In the subsequent steps the second unlabeled mouse monoclonal antibody is applied and detected by a conventional enzyme or fluorochrome conjugated antibody. We used several mouse monoclonal antibodies to assess the validity of the method, including mouse monoclonal antibodies against lymphoid markers, a proliferation marker and different viral proteins. Three EBV-encoded proteins were studied, EBNA2, LMP1 and ZEBRA. EBNA2 and LMP1 are proteins associated with latency that contribute to the EBV mediated growth transformation of B-lymphocytes. They are expressed in EBV-immortalized lymphoblastoid cell lines (LCL-s), but not in type I Burkitt lymphoma cells (Rowe et al., 1992). ZEBRA protein (encoded by the open reading frame BZLF-1) is associated with the viral lytic cycle. It initiates a cascade of lytic phase protein expression (Miller, 1990). In EBV-transformed LCL-s only very few cells express the ZEBRA protein. In the present paper, we demonstrate that the double staining method has a wide range of application, including staining of paraffin sections and
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
91
Table 1 Antibodies used in this study Antibody to
CD3 (D) CD45RO* CD79a (D) PCNA** (D) Cytokeratin*** (D) EBNA2* LMP1 (D) ZEBRA (D) APAAP, mouse monoclonal (D) TRITC-conjugated rabbit anti-mouse Ig (D) Peroxidase-conjugated sheep anti-mouse Ig (A) FITC-conjugated streptavidin (P)
Clone
UCHT1 UCHL1 JCB117 PE10 MNF116 PE2 CS1-4 BZ-1 AP7/6/7 — — —
Dilution of antibodies 1sta
2ndb
1:5000 1:200 1:2500 1:5000 — 1:40 — — — — — —
— 1:2 1:25 — 1:50 1:1 1:30 1:20 1:50 1:30 1:1000 1:50
Antigen retrival procedure
Autoclaving in H2O — Autoclaving in H2O Autoclaving in H2O Trypsinization — — — — — — —
All antibodies are mouse monoclonal unless otherwise stated. D, Dakopatts; A, Amersham; P, Pharmingen. a Dilution of monoclonal antibodies for the 1st antigen detected by the CSA method. b Dilution of antibodies for the 2nd antigen or as secondary antibodies. * Supernatant of cultured hybridoma cell lines; ** proliferating cell nuclear antigen; *** MNF116 reacts with a wide spectrum of keratin Nos. 5, 6, 8, and 17.
immunofluorescent detection of proteins on single cell level.
2. Materials and methods
2.1. Specimens Sections (3–5 mm) were cut from paraffin blocks that were prepared by routine procedures. Biopsy specimens included 3 nasopharyngeal carcinomas (NPC), 3 lung cancers, 4 reactive lymph nodes, and 1 tonsil. The following cell lines were used for double immunofluorescence staining: an EBV producing marmoset lymphoblastoid cell line, B95-8; an in vitro EBV transformed human lymphoblastoid line, SIB1; an EBV positive, type I Burkitt lymphoma line, Rael; EBV negative mouse myeloma line, SP2/0. All cells were cultured at 37°C in 5% CO2 in RPMI 1640 medium containing 10% fetal calf serum. Before staining, the human cells were mixed in 4 – 6:1 proportion with the internal staining control SP2/0 cells. The mouse and human cells can be distinguished read-
ily by their characteristic chromatin distribution using Hoechst staining. Cytospin smears were prepared by centrifuging the cells on glass slides for 2 min at 900 rpm. The cells were fixed immediately, without drying, in acetone for 20 min. B95-8 and SIB1 are known to express EBNA2 and LMP1. About 2–4% of the B95-8 cells express ZEBRA. Rael is non-producer cell line. It expresses only EBNA1 as the sole EBV-encoded protein (Masucci et al., 1989).
2.2. Double staining procedures The sections were deparaffinized and rehydrated in xylene and a descending alcohol series. If necessary, antigen retrieval procedures were used (Goddard et al., 1991; Ryong et al., 1991). The combination of the retrieval procedures and mouse monoclonal antibodies are listed in Table 1. In order to determine the optimal conditions for the first antibody, mouse monoclonal antibodies which were serially diluted were visualized by alkaline-phosphatase anti-alkaline phosphatase (APAAP) after 15 min incubation at room temperature (Cordell et al., 1984). Dilution that gave
92
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
no detectable signal with APAAP was used for CSA detection. For double staining, the rehydrated paraffin sections were soaked in 5% H2O2 for 5 min, followed by incubation with blocking solution supplied in the CSA kit (Dakopatts, Glostrup, Denmark) for at least 5 min at room temperature. Subsequently the slides were incubated with the highly diluted first mouse monoclonal antibody (listed in Table 1) for no more than 16 min. The sections were processed according to the supplier’s instructions for the CSA kit, and color development was carried out using DAB or 3-amino-9-ethyl carbazole (AEC). The sections were incubated with the second mouse monoclonal antibody at the concentration indicated in Table 1 at 4°C overnight. The second Mab was visualized using the APAAP method with a 5bromo-4-chloro-3-indolyl phosphate/nitro-blue (BCIP/NBT) substrate. The sections were counterstained with methyl green and mounted with Tris-buffered glycerol. For double immunofluorescence staining, the cell smears were incubated with 1.5% blocking solution (Boehringer Manheim, Manheim, Germany) for at least 20 min. The blocking was followed by incubation with the highly diluted first mouse monoclonal antibody (listed in Table 1) for no more than 16 min. The cells were incubated with peroxidase-conjugated sheep antimouse antibody (1:1000, Amersham, Buckingham, UK) for 15 min and subsequently for 15 min with the amplification solution of the CSA kit to generate the biotinylated tyramid deposits. The second mouse monoclonal antibodies were used at concentrations indicated in Table 1 at 4°C, overnight. The biotin – tyramide deposits corresponding to the first Mab and the second Mab itself were detected simultaneously by a mixture of FITC-conjugated streptavidin and tetramethylrhodamine isothiocyanate isomer R (TRITC)-conjugated rabbit anti-mouse immunoglobulin, respectively. The mixture also contained 40 ng/ml Hoechst 33258 to visualize DNA. The slides were washed regularly three times with PBS between the different staining steps in both staining protocols. The stained cells were mounted with 70% glycerol, 2.5% 1,4-diazabicyclo
(2.2.2) octane (DABCO) (Sigma), 50 mM Tris– HCl, pH 8.5. The microscopy, image capturing with a cold CCD camera and subsequent image processing was carried out as previously described (Szekely et al., 1995). In the staining controls, color development for the first mouse monoclonal antibodies by DAB, AEC or FITC-conjugated streptavidin or incubation with the second mouse monoclonal antibodies was omitted. Cross-reactions were assessed by detecting the bound first mouse monoclonal antibody with the second detection system.
3. Results
3.1. Double immunochemical staining on paraffin sections The proper dilution of the first Mab is crucial for the successful double staining using CSA as the first staining step. It should be under the detectability by the conventional immunostaining procedure that is used to detect the second Mab. Fig. 1 shows the results of the antibody titration (1:50, 500, 5000) for the detection of proliferating cell nuclear antigen (PCNA) in the germinal center of a human lymph node using the APAAP procedure. After 15 min incubation, PCNA was detected at 1:50 and 1:500 but not at 1:5000 dilution using the APAAP procedure (Fig. 1a–c), whereas CSA clearly visualized PCNA even at 1:5000 dilution (Fig. 2e and g) indicating that 1:5000 dilution is suitable for double staining. It was found that most mouse monoclonal antibodies were below the detection level of APAAP but above the sensitivity of CSA, when they were diluted 50–100 times more than their usual working concentration. The mutually exclusive double staining pattern on paraffin embedded material was demonstrated by staining CD45RO-positive (primed) T-cells and CD79a-positive B-cells in human reactive lymph nodes and tonsil sections (Fig. 2a). A similar mutually exclusive staining pattern was obtained when CD45RO-positive T-cells (Fig. 2c) or CD79a-positive B-cells (data not shown) were stained in combination with anti-cytokeratin
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
93
Fig. 1. Sensitivity titration of the first monoclonal antibody to determine the dilution that is no longer detected by the conventional detection system reserved for the second monoclonal antibody. Different dilutions of anti-PCNA antibody (PE10) were incubated with a lymph node section at room temperature for 15 min and visualized by the APAAP method. PCNA was strongly stained in the germinal center of a lymph node at 1:50 (a), faintly stained at 1:500 (b), and was not detectable at 1:5000 dilution (c).
(MN116) antibody that visualized epithelial cells in tonsil or three different nasopharyngeal carcinoma sections. The staining pattern was clear and well defined even if the antigens were adjacent to each other indicating that biotin – tyramide and subsequent DAB deposition did not block second antibody reaction (Fig. 2a and c). The CSA/APAAP double staining was able to resolve antigens that were present in the same cell, even on paraffin sections, as long as one antigen was nuclear and the other was associated with the cytoplasm. For example PCNA nuclear staining could be detected in germinal center B-cells that were visualized by CD79a antibody (Fig. 2e). PCNA was also detected in the nuclei of lung carcinoma cells, identified by anti-cytokeratin antibody (Fig. 2g). No staining was seen in the control sections on which DAB or AEC color development and the incubation with the second mouse monoclonal antibody were omitted. This shows that the first mouse monoclonal antibodies was not visualized by the APAAP method (Fig. 2b, d, f and h).
3.2. Double immunofluorescence staining on single cell le6el EBV encoded viral antigens with known expression pattern were used to explore the feasibility of CSA based double staining for fluorescence detection. EBNA2, one of the latency associated nuclear antigens, was visualized in the nuclei of B95-8 and SIB1 cells by FITC-conjugated strep-
tavidin that detected the biotinylated tyramide deposits that formed at the site of the antigen (Fig. 3a and e). The fine intranuclear resolution of the immunofluorescence, characteristic for EBNA-2 by conventional fluorescence staining, was lost, however. Approximately 2% of the B958 cells expressed the lytic cycle associated EBV protein, ZEBRA. Double immunofluorescence staining where EBNA-2 was detected by PE-2 monoclonal antibody followed by CSA and FITC-conjugated streptavidin reaction (Fig. 3a), and the ZEBRA protein by Mab BZ-1 followed by conventional staining with TRITC-conjugated anti-mouse serum (Fig. 3b) showed that the ZEBRA positive cells were negative for EBNA2 (Fig. 3c). No cross-reaction was detected between the two staining systems as demonstrated by capturing nonoverlapping red and green monochromatic signals from the same cells using a cold CCD camera. 30–50% Of the EBNA-positive B95-8 and SIB cells were also positive for the membrane associated viral protein LMP-1. LMP-1 was localized in membrane associated patches on the cell surface and scattered foci in the cytoplasm. Double staining of EBNA-2 and LMP-1 with unlabeled mouse Mab-s of PE-2 and CS1-4, respectively, was used to carry out CSA based double fluorescence staining within the same cell. EBNA-2 is a known positive regulator of LMP-1 expression. Double immunostaining of the two proteins showed that the LMP1-positive cells expressed relatively high amounts of EBNA2.
94
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
Fig. 2
Fig. 3
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
The following controls were included in these experiments: the Burkitt lymphoma cell line Rael, known to express EBNA-1 exclusively, was consistently negative for PE-2, BZ-1 and CS1-4 (Fig. 3d). All preparations used for double staining contained 15– 25% EBV negative Sp2/0 mouse myeloma cells. These cells were easily recognized by their pericentromeric heterochromatin granules that gave them a characteristic appearance on Hoechst 33258 DNA staining (Fig. 3, nuclei labeled with an ). Mab anti-CD3 antibody that did not react with any of the examined B-cell lines was used as a negative control, included as the first reagent in the control stainings (Fig. 3h).
4. Discussion The possibility was demonstrated to use, simultaneously, two mouse monoclonal antibodies, without any chemical modification, for double immunostaining without cross-reaction. The method is based on the selective sensitivity of the secondary detection systems. It is therefore very important to establish the optimal concentration of the first antibody that allows detection by the CSA method, but not by the conventional detec-
95
tion system. This needs to be determined for each antigen–antibody pair. Since the local concentration of the antigen may vary widely for different samples and since a high concentration of a given antigen may capture relatively high amounts of the first Mab, even from a very diluted solution, it is essential to run the titration for the non-CSA based detection system on parallel samples and to demonstrate the absence of any reactivity at the selected working dilution. Although the method presented only used two unlabeled mouse Mabs, it is likely that it could be applied to other unlabeled monoclonal or polyclonal antibody pairs from the same species. Combination with additional directly labeled antibodies from the same species or unlabeled antibodies from a different species might be used for triple or even higher order probing. It was shown that the method can be used for two color immunofluorescence with two different mouse monoclonal antibodies and that the visualization of the two antigens does not interfere with each other. It was found that two different antigens can be distinguished in paraffin sections, even if they are expressed in the same cell. The resolution of the CSA method is not sufficient for fine intracellular localization, due to the dissemination of the catalyzed tyramides. This dissemination might be reduced by an increase of
Fig. 2. Double staining with two mouse monoclonal antibodies on paraffin sections. The first antibodies were visualized with DAB (brown) or AEC (brownish red) using the CSA method. The second antibodies were visualized with BCIP/NBT (dark blue) using the APAAP method. (a) T-cells (anti-CD45RO, CSA method, brown) and B-cells (anti-CD79a, APAAP method, blue) show a mutually exclusive localization in a lymph node (magnification × 150). Note that adjacent cells were stained in different colors, indicating that the DAB deposition did not block the reaction. (c) Double staining of infiltrating T-lymphocytes (anti-CD45RO, CSA method, brown) and cancer cells (anti-cytokeratin, APAAP method, blue) in nasopharyngeal carcinoma tissue ( × 200). (e) PCNA (CSA method, brown) in the nuclei of germinal center B-cells (anti-CD79a, APAAP method, blue) ( × 100). (g) PCNA (CSA method, brown) in the nuclei of lung cancer cells (anti-cytokeratin, APAAP method, blue) ( × 200). (b, d, f and h) Control stainings to demonstrate the absence of cross-reaction between the first mouse monoclonal antibodies and the APAAP method. Fig. 3. Double immunofluorescence staining on single cells. Mutually exclusive expression of EBNA-2 (a) (Mab PE-2, CSA method, FITC, green) and ZEBRA protein (b) (Mab BZ-1, conventional indirect staining, TRITC, red) in B95-8 cells nuclei are counterstained with Hoechst 33258 (blue). Tri-color overlap (c). Note the absence of cross-reaction from the EBNA-2 positive cells by the TRITC-conjugated anti-mouse serum targeted against the anti-ZEBRA. Double negative staining control Rael cells (d). Detection of two independent antigens in the same cells. B95-8 cells were double stained for EBNA-2 (e) (Mab PE-2, CSA method, FITC, green) and LMP-1 (f) (Mab CS1-4, conventional indirect staining, TRITC, red). TRITC-conjugated rabbit anti-mouse antibody did not show any reaction with the nuclei of EBNA-2 positive cells. Tri-color overlap (g). Non-reactive antibody control anti-CD3 on B95-8 cells (Mab UCHT1, CSA method, FITC, green) shows no reaction. The cells indicated by an are Sp2/0 mouse myeloma cells that serve as internal negative controls.
96
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97
viscosity during the deposition of tyramide, however (Van Gijlswijk et al., 1996). Using this novel method, it was shown that ZEBRA-positive B95-8 cells that have entered lytic cycle do not express EBNA2. This is in agreement with a previous paper by Sinclair et al. (1992) who showed that ZEBRA represses the activity of the Cp promoter. To our knowledge, this is the first demonstration of the single cell level of EBNA2 down-regulation in EBV-carrying lymphoid cells that have switched on ZEBRA. The LMP1 promoter is known to be activated by EBNA2 expression (Fahraeus et al., 1990). Our staining results are in line with these findings. The double staining method described above may have wide application since it can be used both for paraffin sections and cell smears. It does not require specialized techniques or apparatus.
Acknowledgements We thank Mie Inoko for her assistance in preparing the materials. This work was supported by grants from the Swedish Cancer Society (Cancerfonden), the Swedish Research Council, and by a matching grant from the Concern Foundation, LA, and the Cancer Research Institute, NY.
References Bobrow, M., Harris, T., Shaughnessy, K., Litt, G., 1989. Catalyzed reporter deposition, a novel method of signal amplification. Application to immunoassays. J. Immunol. Methods 125, 279 – 285. Chen, K., Demetris, A.J., Vanthiel, D.H., Whiteside, T.L., 1987. Double immunoenzyme staining method for analysis of tissue and blood lymphocyte subsets with monoclonal antibodies. Lab. Invest. 56, 114–119. Cordell, J., Fallini, B., Erber, W., et al., 1984. Immunoenzymatic labeling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes). J. Histochem. Cytochem. 32, 219 – 229. Fahraeus, R., Jansson, A., Ricksten, A., Sjo¨o¨blom, A., Rymo, L., 1990. Epstein-Barr virus-emcoded nuclear antigen 2
.
activates the viral latent membrane protein promoter by modulating the activity of a negative regulatory element. Proc. Natl. Acad. Sci. 87, 7390 – 7394. Goddard, M.J., Wilson, B., Grant, J.W., 1991. Comparison of commercially available cytokeratin antibodies in normal and neoplastic adult epithelial and non-epithelial tissues. J. Clin. Pathol. 44, 660 – 663. Key, G., Petersen, J.L., Becker, M.H., et al., 1993. New antiserum against Ki-67 antigen suitable for double immunostaining of paraffin wax sections. J. Clin. Pathol. 46, 1080 – 1084. Mason, D.Y., Abdulaziz, Z., Falini, B., Stein, H., 1983. Single and double immunoenzymatic techniques for labeling tissue sections with monoclonal antibodies. Ann. New York Acad. Sci. 420, 127 – 133. Masucci, M.G., Contreras-Salazar, B., Ragnar, E., et al., 1989. 5-Azacytidine up regulates the expression of Epstein-Barr virus nuclear antigen 2 (EBNA-2) through EBNA-6 and latent membrane protein in the Burkitt’s lymphoma line Rael. J. Virol. 63, 3135 – 3141. Merz, H., Malisius, R., Mannweiler, S., et al., 1995. ImmunoMax. A maximized immunohistochemical method for the retrieval and enhancement of hidden antigens. Lab. Invest. 73, 149 – 156. Miller, G., 1990. The switch between latency and replication of Epstein-Barr virus. J. Infect. Dis. 161, 833 – 844. Pastore, J., Clampett, C., Miller, J., Porter, K., Miller, D., 1995. A rapid immunoenzyme double labeling technique using EPOS reagents. J. Histotechnol. 18, 35 – 39. Riesenberg, R., Oberneder, R., Kriegmair, M., et al., 1993. Immunocytochemical double staining of cytokeratin and prostate specific antigen in individual prostatic tumour cells. Histochem. J. 99, 61 – 66. Rowe, M., Lear, A.L., Croom, C.D., Davies, A.H., Rickinson, A.B., 1992. Three pathways of Epstein-Barr virus gene activation from EBNA1-positive latency in B lymphocytes. J. Virol. 66, 122 – 131. Ryong, W.S., Iwaki, T., Kitamoto, T., Tateishi, J., 1991. Hydrated autoclave pretreatment enhances TAU immunoreactivity in formalin-fixed normal and Alzheimer’s disease brain tissues. Lab. Invest. 64, 693 – 702. Shibuya, M., Ito, S., Davis, R.L., Hoshino, T., 1993. Immunohistochemical double staining with immunogold-silver and alkaline phosphatase to identify nuclear markers of cellular proliferation. Biotech. Histochem. 68, 17 – 19. Sinclair, A.J., Brimmell, M., Farrell, P.J., 1992. Reciprocal antagonism of steroid hormones and BZLF1 in switch between Epstein-Barr virus latent and productive cycle gene expression. J. Virol. 66, 70 – 77. Szekely, L., Jiang, W.Q., Pokrovskaja, K., Wiman, K.G., Klein, G., Ringertz, N., 1995. Reversible nucleolar translocation of Epstein-Barr virus-encoded EBNA-5 and hsp70 proteins after exposure to heat shock or cell density congestion. J. Gen. Virol. 76, 2423 – 2432. 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, 518 – 524.
N. Teramoto et al. / Journal of Virological Methods 73 (1998) 89–97 Valnes, K., Brandtzaeg, P., 1984. Paired indirect immunoenzyme staining with primary antibodies from the same species. Application of horseradish peroxidase and alkaline phosphatase as sequential labels. Histochem. J. 16, 477– 487.
97
Van Gijlswijk, R.P.M., Wiegant, J., Raap, A.K., Tanke, H.J., 1996. Improved localization of fluorescent tyramides for fluorescence in situ hybridization using dextran sulfate and polyvinyl alcohohol. J. Histochem. Cytochem. 44, 389 – 392.
.