Double and triple immunocytochemical labelling at the light microscope level in histopathology

Adding value to liver (and allograft) biopsy evaluation using a combination of multiplex quantum dot immunostaining, high-resolution whole-slide digital imaging, and automated image analysis

Various technologies including nucleic acid, protein, and metabolic array analyses of blood, liver tissue, and bile are emerging as powerful tools in the study of hepatic pathophysiology. The entire lexicon of liver disease, however, has been written using classical hematoxylin-eosin staining and light microscopic examination. The authors' goal is to develop new tools to enhance histopathologic examination of liver tissue that would enrich the information gained from liver biopsy analysis, enable quantitative analysis, and bridge the gap between various "-omics" tools and interpretation of routine liver biopsy results. This article describes the progress achieved during the past 2 years in developing multiplex quantum dot (nanoparticle) staining and combining it with high-resolution whole-slide imaging using a slide scanner equipped with filters to capture 9 distinct fluorescent signals for multiple antigens. The authors first focused on precise characterization of leukocyte subsets, but soon realized that the data generated were beyond the practical limits that could be properly evaluated, analyzed, and interpreted visually by a pathologist. Therefore, the authors collaborated with the open source FARSIGHT image analysis project (http://www.farsight-toolkit.org). FARSIGHT's goal is to develop and disseminate the next-generation toolkit of automated image analysis methods to enable quantification of molecular biomarkers on a cell-by-cell basis from multiparameter images. The resulting data can be used for histocytometric studies of the complex and dynamic tissue microenvironments that are of biomedical interest. The authors envisage that these tools will eventually be incorporated into the routine practice of surgical pathology and precipitate a revolution in the specialty.

Multiple antigen immunostaining procedures

Detection of multiple antigens in the same tissue section can be done by combining a range of immunohisto/cytochemical techniques based either on light microscopic chromogenic precipitates or fluorochrome labeling. Light microscopic techniques preferred for this purpose use combinations of immunogold silver staining (black precipitate), immunoperoxidase, immunoalkaline phosphatase and immunogalactosidase methods using chromogens of different colors. Fluorochrome labels favored for these combinations include AMCA (blue), FITC (green), rhodamine (orange-red) and Cy5 (far red), their matching synthetic members from the Alexa series, or quantum dots. Antibodies directly labeled or those from noncross-reacting animal species (e.g., mouse, rabbit, goat, guinea pig etc.) can be applied simultaneously. When the antigens of interest are in separate cells or cell compartments (e.g., in cell membrane, cytoplasm or nucleus), and only cross-reacting antibodies are available, there have also been ways of avoiding unwanted cross-talk. These include the exploitation of the shielding effect of chromogens; inactivation of immuno-sequences of the first staining by using either acidic elution, formaldehyde fixation or microwave heating; combining unlabeled and hapten-labeled antibodies; or using labeled monovalent F(ab) secondary antibodies. In this chapter we briefly discuss the principle of multiple antigen immunolabeling and provide useful protocols for its performance.

Double immunostaining with p63 and high-molecular-weight cytokeratins distinguishes borderline papillary lesions of the breast

Papillary breast lesions remain a source of diagnostic confusion because the full range of epithelial proliferations may arise within, or secondarily involve, papilloma. The expression of p63 and high-molecular-weight cytokeratins (HMWCK) was studied simultaneously in 33 papillary lesions including intraductal papilloma (IP, n = 10), atypical papilloma (AP, n = 8) and intraductal papillary carcinoma (IPC, n = 15) by double immunostaining. The myoepithelial cell nuclei were stained dark brown whereas the cytoplasms of usual ductal hyperplasia (UDH) and myoepithelium were stained purple. The myoepithelial layer was recognized as a dark brown dotted line at the epithelial stromal junction in all IP (10/10), most AP (7/8) and some IPC (7/15), suggesting that the retained myoepithelial layer in the papillary processes does not necessarily guarantee benignity. However, the malignant epithelial cells in AP and IPC were typically recognized as monotonous populations unstained with either chromogen. These monotonous cells contrasted with the proliferating cells of UDH in papilloma, which had intense purple cytoplasm in a mosaic-like fashion. The present data suggest that the double immunostaining with the two popular antibodies p63 and HMWCK is a useful tool for reproducible classification of papillary breast lesions.

How we process trephine biopsy specimens: epoxy resin embedded bone marrow biopsies

Improved cytomorphology of semithin resin sections over paraffin wax embedded sections may be important in diagnostic haematopathology. However, resin embedding can make immunohistochemical antigen detection or DNA isolation for clonal gene rearrangement assays difficult. This review describes the processing of bone marrow biopsies using buffered formaldehyde based fixation and epoxy resin embedding, with or without EDTA decalcification. Traditional semithin resin sections are completely rehydrated after etching in home made sodium methoxide solution. Resin elimination allows high resolution staining of tissue components with common histological stains. Efficient antigen retrieval and the Envision-HRP system permit the immunohistological detection of many antigens of diagnostic relevance, with retention of high quality cytomorphology. Furthermore, DNA can be extracted for clonality analysis. The technique can be completed within a similar time period to that of paraffin wax processing with only approximately 30% increase in cost. This technique has been used for diagnosis in over 4000 bone marrow biopsies over the past 14 years. By meeting traditional and contemporary demands on the haematopathologist, it offers a powerful alternative to paraffin wax processing for diagnosis and research.

E-, N- and P-cadherin, and alpha-, beta- and gamma-catenin protein expression in normal, hyperplastic and carcinomatous human prostate

The expression of E-, N- and P-cadherin, alpha-, beta- and gamma-catenin, and actin was studied by immunohistochemistry, ELISA, and Western blot analysis in normal prostates, and in the prostates of men with benign prostatic hyperplasia and men with prostatic carcinoma, in order to evaluate their possible role in the pathogenesis of these diseases. Present results reveal that the immunophenotype of hyperplastic prostates differs from those of both normal and carcinomatous prostates in the intracellular distribution (observed by immunohistochemistry) and the intensity (measured by ELISA) of immunoreactions to cadherins, catenins, and actin. Hyperplastic prostates differ form normal prostates in the weaker immunoreaction to the three cadherin types, the two catenins, and actin, as well as in the intracellular distribution of P-cadherin, beta- and gamma-catenin, and actin. Differences between benign prostatic hyperplasia and prostatic carcinoma are less marked because hyperplastic prostates differ from carcinomatous prostates only in the weaker immunoreactions to P-cadherin, and alpha-catenin. The most remarkable findings in this study were: (1) alpha-catenin production was elevated in prostatic carcinoma in comparison with benign prostatic hyperplasia and normal prostate; and (2) P-cadherin expression in benign prostatic hyperplasia is reduced with regard to those of normal and carcinomatous prostates. It may be concluded that a decreased immunoreaction to cadherins, catenins, and actin, as well as changes in the intracellular distribution of actin in prostatic cells are not necessarily suggestive of malignancy, because these alterations are also present in BPH, and thus, the loss of cadherin-catenin-mediated adhesion alone is not sufficient to establish an invasive phenotype.

Sialoadhesin-Positive Host Macrophages Play an Essential Role in Graft-Versus-Leukemia Reactivity in Mice

We recently established an effective immune T-cell–mediated graft-versus-leukemia (GVL) murine model system in which complete tumor remissions were achievable even in advanced metastasized cancer. We now describe that this T-cell–mediated therapy is dependent on host macrophages expressing the lymphocyte adhesion molecule sialoadhesin (Sn). Depletion of Kupffer cells in tumor-bearing mice during adoptive immunotherapy (ADI) or the treatment of these animals with anti-Sn monoclonal antibodies led to complete or partial inhibition of the immune T-cell–mediated therapeutic effect. Furthermore, Sn+ host macrophages in livers formed clusters during ADI with donor CD8 T cells. To test for a possible antigen presentation function of these macrophages, we used as an in vitro model the antigen β-galactosidase for which a dominant major histocompatibility complex (MHC) class I Ld-restricted peptide epitope is known to be recognized by specific CD8 cytotoxic T lymphocytes (CTL). We demonstrate that purified Sn+ macrophages can process exogenous β-galactosidase and stimulate MHC class I peptide-restricted CTL responses. Thus, Sn+ macrophages, which are significantly increased in the liver after ADI, may process tumor-derived proteins via the MHC class I pathway as well as via the MHC class II pathway, as shown previously, and present respective peptide epitopes to CD8 as well as to CD4 immune T cells, respectively. The synergistic interactions observed before between immune CD4 and CD8 T cells during ADI could thus occur in the observed clusters with Sn+ host macrophages.

Sialoadhesin-Positive Host Macrophages Play an Essential Role in Graft-Versus-Leukemia Reactivity in Mice

We recently established an effective immune T-cell–mediated graft-versus-leukemia (GVL) murine model system in which complete tumor remissions were achievable even in advanced metastasized cancer. We now describe that this T-cell–mediated therapy is dependent on host macrophages expressing the lymphocyte adhesion molecule sialoadhesin (Sn). Depletion of Kupffer cells in tumor-bearing mice during adoptive immunotherapy (ADI) or the treatment of these animals with anti-Sn monoclonal antibodies led to complete or partial inhibition of the immune T-cell–mediated therapeutic effect. Furthermore, Sn+ host macrophages in livers formed clusters during ADI with donor CD8 T cells. To test for a possible antigen presentation function of these macrophages, we used as an in vitro model the antigen β-galactosidase for which a dominant major histocompatibility complex (MHC) class I Ld-restricted peptide epitope is known to be recognized by specific CD8 cytotoxic T lymphocytes (CTL). We demonstrate that purified Sn+ macrophages can process exogenous β-galactosidase and stimulate MHC class I peptide-restricted CTL responses. Thus, Sn+ macrophages, which are significantly increased in the liver after ADI, may process tumor-derived proteins via the MHC class I pathway as well as via the MHC class II pathway, as shown previously, and present respective peptide epitopes to CD8 as well as to CD4 immune T cells, respectively. The synergistic interactions observed before between immune CD4 and CD8 T cells during ADI could thus occur in the observed clusters with Sn+ host macrophages.

Differences Between Graft-Versus-Leukemia and Graft-Versus-Host Reactivity. I. Interaction of Donor Immune T Cells With Tumor and/or Host Cells

Graft-versus-leukemia (GVL) and Graft-versus-host (GVH) reactions were compared after systemic transfer of allogeneic antitumor immune T lymphocytes from B10.D2 (H-2d; MIsb) into DBA/2 (H-2d; MIsa) mice. Before immune cell transfer, recipient DBA/2 mice were sublethally irradiated with 5 Gy to prevent host-versus-graft reactivity. Recipients were either bearing syngeneic metastatic ESb lymphomas (GVL system) or were normal, non–tumor-bearing mice (GVH system). We previously reported that this adoptive immunotherapy protocol (ADI) had pronounced GVL activity and led to immune rejection of even advanced metastasized cancer. In this study, monoclonal antibodies were used for immunohistochemical analysis of native frozen tissue sections from either spleen or liver to distinguish donor from host cells, to differentiate between CD4 and CD8 T lymphocytes, and to stain sialoadhesin-positive macrophages at different time points after cell transfer. The kinetics of donor cell infiltration in spleen and liver differed in that the lymphoid organ was infiltrated earlier (days 1 to 5 after transfer) than the nonlymphoid organ (days 5 to 20). After reaching a peak, donor cell infiltration decreased gradually and was not detectable in the spleen after day 20 and in the liver after day 30. The organ-infiltrating donor immune cells were mostly T lymphocytes and stained positive for CD4 or CD8 T-cell markers. A remarkable GVL-associated observation was made with regard to a subset of macrophages bearing the adhesion molecule sialoadhesin (SER+ macrophages). In the livers of tumor-bearing mice, their numbers increased between days 1 and 12 after ADI by a factor greater than 30. Double-staining for donor cell marker and SER showed that the sialoadhesin-expressing macrophages were of host origin. The SER+ host macrophages from GVL livers were isolated by enzyme perfusion and rosetting 12 days after ADI, when they reached peak values of about 60 cells per liver lobule, and were tested, without further antigen addition, for their capacity to stimulate an antitumor CD8 T-cell response. The results of this immunologic analysis suggest that these cells in the liver function as scavengers of the destroyed metastases and as antigen-processing and -presenting cells for antitumor immune T cells.

Dual staining of lymphocyte membrane antigens with colloidal gold and biotinylated horseradish peroxidase

Double immunoenzymatic labelling procedures for the localization of antigens on cells in tissue sections using horseradish peroxidase (HRP) and alkaline phosphatase have been described previously, but mainly for detecting antigens on different cells. With this type of staining when two antigens are present on the same cell, an optimal colour combination that shows a high contrast between the basic colour of each enzyme substrate product is difficult to achieve and the interpretation of their mixed colour intermediate is subjective. We present a method for the simultaneous demonstration of two antigens on the same cell. The method can be used to label either single cells in suspension, or cells in paraffin fixed tissue, using a combination of a particulate label, colloidal immunogold-silver, and an enzymatic label HRP-DAB. The method is easy to perform and utilises commercially available staining kits.

Ultrastructural characteristics of glial fibrillary acidic protein expression in epoxy resin-embedded human brain tumors

Thirteen surgically removed, epoxy resin (Durcupan ACM or Epon 812)-embedded human brain tumors were examined for glial fibrillary acidic protein (GFAP) content in semithin and ultrathin sections with the immunogold-silver staining method. Mild aldehyde fixation and the hydrophobic resin embedding did not interfere with the antigenicity, since silver intensification of the immunogold marker provided excellent visualization of the reaction on both light microscopic and ultrastructural levels. The GFAP reaction was usually localized on the glial intermediate filament bundles, usually correlating well with the amount of filaments. The unstained filamental regions of two ependymomas might correspond to the vimentin expression revealed by double labeling in semithin sections. Occasional GFAP immunopositivity without filamental appearance was observed in one of the oligodendrogliomas, as patchy electron-dense cytoplasmic corpuscules, in Rosenthal fibers and in some mainly necrobiotic tumor cells, reflecting a possible connection with glial filaments.

Protein gene product 9.5 and ubiquitin immunoreactivities in rat epididymis epithelium

A quantitative immunohistochemical study was performed of the distribution of protein gene product 9.5 (PGP, a soluble protein localized in neurons and neuroendocrine cells as well as in some non-nervous cells) and ubiquitin along the rat epididymis. In the ductuli efferents, PGP immunoreaction was observed in the whole cytoplasm of some columnar cells; a smaller number of columnar cells showed ubiquitin immunoreactivity with limited apical and basal cytoplasmic localization. In the proximal caput epididymidis, the whole cytoplasm of all columnar cells showed PGP immunoreactivity, ubiquitin immunostaining was negative in this region. In the middle and distal caput epididymidis and the distal cauda, the apical cytoplasm of some columnar cells and the whole cytoplasm of some basal cells showed immunoreactivity to PGP. In these regions, immunoreactivity to ubiquitin was positive in the supranuclear cytoplasm of some columnar cells but not in the basal cells. No immunoreactivity to PGP or ubiquitin was detected in the corpus epididymis and the proximal cauda. Double immunostaining revealed that all the epididymal ubiquitin immunoreactive cells were also PGP immunoreactive, whereas most PGP immunoreactive cells did not immunoreact to ubiquitin. In ubiquitin-PGP immunoreactive cells, the site of the PGP immunoreaction differed from that of the ubiquitin immunoreaction. PGP-ubiquitin immunoreactive cells also seemed to be immunoreactive to anti-AE1/AE3 keratin antibodies. The spermatozoal heads were immunoreactive to PGP antibodies in the epididymal regions from proximal caput to distal cauda but not in the ductuli efferents. The findings suggest that non-ubiquitinated PGP immunoreactive proteins are secreted in the epididymis, mainly in the proximal caput, and attach to spermatozoa.

Practical suggestions for successful immunoenzyme double-staining experiments

Many methodologies exist to perform an immunoenzyme double staining. Hence, the practical problem arises as to which of these methods is optimal for one's own experimental design. A process of selection is described which is derived from our own practical experience. First, a general strategy is outlined for the handling of tissue sections to be used for multiple staining methods. Secondly, the selection of an appropriate immunoenzyme double-staining concept is made using a flow chart. Thereafter we give criteria for the definitive selection of an immunoenzyme double-staining protocol based on the characteristics of the tissue or cell type under study. Particular attention is given to the selection of appropriate detection systems, applying enzymes or gold particles, and good contrasting colour combinations. The problems of visualizing co-localization using immunoenzyme double staining are dealt with, and suggestions are made to adapt the method, if necessary, in order to optimize it.

Immunohistochemical identification of tubular segments in percutaneous renal biopsies

To identify the renal cortical tubular segments involved in tubulo-interstitial disease in formalin-fixed, paraffin-embedded percutaneous kidney biopsies, we developed multiple immunolabeling protocols using segment-specific tubular markers. The present study of biopsies from patients with minimal change or thin basement membrane nephropathy provides a baseline for interpretation of histopathology. Proximal tubules were stained either by the PAS reaction or by the biotinylated Phaseolus vulgaris erythroagglutinin (PHA-E)-streptavidin-gold-silver system (brush borders black). The anti-Tamm-Horsfall (THP) antibody-immunoperoxidase (aminoethylcarbazole, AEC-IPO), and anti-epidermal cytokeratins (ECK) antibodies-immunoalkaline-Fast Blue BB methods marked the distal straight tubules and the cortical collecting system red-brown and blue, respectively. When these immunolabelings were combined, the coapplication of AEC-PO-labeled peanut agglutinin (PNA) or anti-epithelial membrane antigen antibody-AEC-IPO technique (both are markers for distal nephron) visualized the apical membranes of distal convoluted tubules. In the protocol PHA-E + PNA + THP + ECK, the tubular basement membranes were outlined by the anti-laminin antibody-AEC-IPO staining, carried out simultaneously. The protocol PNA + THP + ECK + PAS was found to be quite appropriate multiple immunolabeling method for the tubules, and is recommended for use as a tool in the study of tubulo-interstitial diseases.