Simple blocking system for use with maleimide-labeled nucleic acid probes.

Here, we describe our investigation of the preparation and use of chemically labeled DNA probes for fluorescence in situ hybridization (FISH) experiments. Compared to enzymatic labeling of DNA, the chemical labeling process is expected to be significantly more robust, cheaper and easier to scale up (1,2,8). Running a major probe isolation/preparation facility in the western US, we have been providing collaborators with relatively large amounts of probe DNA, labeled with either biotin or digoxigenin. We became progressively interested in preparing fluorochrome-labeled DNA probes, because they eliminate many of the time-consuming detection steps (3,9,11,12) and often result in levels of background signal much lower than those obtained with biotinylated probes (13). Furthermore, we recently developed hybridization protocols for formalinfixed, paraffin-embedded tissue sections that completely eliminate the need to remove unbound probe or even to wash the tissue after hybridization. At the same time, hybridization signals were of such high intensity as to allow the use of low magnification (i.e., low numerical aperture [N.A.]) microscope lenses. So far, the major application of this technique has been the identification of the species of cells in heterologous transplantations such as human tissue transplanted into mice (4,7,10, 12–14), but it appears that the same methodology will be suitable to detect cells from any species in heterologous transplantations. Probes used in previous experiments (4,10,14) were kindly provided by an industrial sponsor applying a proprietary chemical labeling technique (1). These probes were used to develop the published hybridization protocols (10, 14) and created a demand for chemically labeled DNA probes. The fact that there were no commercial sources of fluorochrome-labeled COT-1 DNA probes led to a severe probe shortage and stimulated our developments. Most chemical DNA-labeling protocols published thus far have been either laborintensive, proprietary or they yielded probes of generally low quality. A recently introduced DNA labeling kit (FastTAG; Vector Laboratories, Burlingame, CA, USA) attracted our attention because it allows DNA modification by either heat or UV light. While different haptens such as biotin and dinitrophenyl (DNP) were available for DNA labeling with the kit, we were primarily interested in labeling our DNA samples with fluorescein for use in FISH. The FastTAG kit is based on covalent binding of a proprietary heator photo-activated crosslinker to the DNA. The crosslinker carries an internal disulfide bond that is subsequently reduced. Fluorescein maleimide, or any other sulfhydryl (SH) reactive molecule, can then be coupled to the thiol group (5). In the standard protocol, most of the uncoupled fluorescein maleimide is removed from the DNA probe preparation by ethanol-precipitation. Before use in FISH, the supplier of the FastTAG kit recommends purification of the DNA probes by gel filtration. Using this kit for labeling probe DNA turned out to be relatively simple. Here, we used human COT-1 DNA (Life Technologies, Gaithersburg, MD, USA) because we wanted to apply the probe for the labeling of human cells in transplantation tissues (6,10,14). However, we expect that the procedure will work equally well with any other highly repeated or genomic DNA samples. Approximately 1 μg of DNA was labeled with the FastTAG kit following the manufacturer’s instructions. We used the 100-W Hg lamp of a Metalloplan microscope (Leitz, Wetzlar, Germany) equipped with a filter set for UV-excitation (Leitz) to induce photocrosslinking. The microscope objective was removed, and the sample was placed in an ice-water bath on the microscope stage and exposed for