Combined FISH and immunolabeling on paraffin-embedded tissue sections for the study of microchimerism.

Fluorescence in situ hybridization (FISH) has proven to be a very useful technique for the study of chimerism and microchimerism (2,4). Based on the detection of sex chromosomes, this technique allows for the identification and localization of donor and recipient cells in cases of sex-mismatched transplantation or cells from mothers and their male fetuses in studies of fetomaternal cell trafficking. For the further assessment of chimerism, the combination of FISH and techniques of protein detection allows for the determination of the phenotype of persistent chimeric cells. However, such combined techniques have proven to be difficult to apply to paraffin-embedded tissue specimens, which is the common preparation of tissue samples following biopsy, surgery, or autopsy. To date, studies of chimerism using simultaneous FISH and protein detection have focused on cytoplasmic proteins such as albumin and cytokeratin (1) or p53 (5). Methods for the identification of cells by sex chromosome constitution and the presence of antigens with a low level of expression or cellsurface molecules such as cluster of differentiation markers have been performed sequentially on a single tissue section or on thin serial sections. This approach requires a search for co-localized FISH and immunolabeling signals in two separate experiments, necessitating considerable microscope and image-capturing time to identify target cells. Artifacts and background can confound the visualization of DNA and protein signals in the same cell generated from two separate experiments. In addition, the localization of a cell by FISH in one tissue section and the determination of its phenotype by immunolabeling on a serial section either above or below the original section, which relies on the superimposition of two separate images to approximate colocalization, can be highly subjective. We previously reported a versatile method to perform FISH on a variety of paraffin-embedded tissues for studies of microchimerism (3). This protocol can be broken down into groups of processing steps, as shown in Table 1. We introduced the immunolabeling techniques at different points in the FISH procedure (i.e., between groups of FISH processing steps) to evaluate the quality of the FISH and immunostaining signals simultaneously. Immunofluorescence staining failed for all antigens when incorporated before the protease digestion (between FISH steps 4 and 5 or 8 and 9). After protease digestion, only cytokeratin could be stained (Figure 1E). Immunoperoxidase staining was successful for all antibodies (Figure 1, A–C), provided that the ethanol dehydration step (step 13 in FISH protocol) was removed. Though the peroxidase stain using aminoethylcarbazole is typically detected with a light microscope, it was also visible through single-bandpass red and dualbandpass filters with the fluorescent microscope. The peroxidase staining was more intense if it was performed after rather than before the acid and heat treatment in the FISH procedure (i.e., between FISH steps 8 and 9). The immunoperoxidase staining between steps 8 and 9 of the FISH procedure seems to be more widely applicable and generates a stronger stain; thus, it is the protocol we recommend to use. For both immunoperoxidase and immunofluorescence, no staining was obtained with isotypic control mouse IgG1 (Figure 1, D and F). All procedures were performed on human skin, liver, thyroid, and brain tissue. While immunoperoxidase staining Benchmarks