What we do in our laboratories

The following is a guide to ISH procedures as used in our laboratories. Our detailed protocols have been published elsewhere (Herbst and Niedobitek, 2001; Niedobitek and Herbst, 2001). Our protocols are fairly standardised and are equally applicable to sections from formalin-fixed, paraffin-embedded tissue blocks, frozen sections and cytological preparations. The only variable is the concentration of protease required (see below).

The detection of RNA transcripts in frozen sections may be more sensitive due to better RNA preservation (Frantz et al., 2001). However, this advantage is offset by the poorer morphology making identification of labelled cells more difficult. Thus, whenever possible, we prefer using paraffin sections even for the detection of low copy RNA species (Meru et al., 2002). Tissue sections are mounted on aminopropyltriethoxysilane (APES)-coated slides to prevent loss of sections during the lengthy hybridisation procedure. Preparation of sections necessary to make nucleic acids, DNA or RNA, accessible for ISH includes treatment with hydrochloric acid and a protease, usually pronase. The protease concentration required is lower for frozen sections and cytological specimens than for paraffin sections, but optimum conditions have to be established in each individual laboratory.

Occasionally, ISH is employed to detect viral DNA genomes in tissue sections [Figure 2.1(a)] (Herrmann et al., 2002; Niedobitek et al., 1989c; Reiss et al., 2002). For this purpose, we use plasmids carrying appropriate virus-specific inserts. Total plasmid DNA is labelled by nick translation using either 35S- or digoxigenin-dCTP and commercially available kits. In situations where detection of lytic virus replication is sufficient, we use digoxigenin-labelled probes. Hybridised probes are detected using standard immunohistochemical procedures. In experiments aimed at the detection of latent viral genomes, e.g. persistent Epstein-Barr virus (EBV) infection, the use of 35S-labelled probes is preferred because in our hands this methods offers higher sensitivity [Figure 2.1(a)].

RNA ISH generally is done using 35S-labelled riboprobes in our laboratories. This approach has proved useful in a variety of settings, e.g. for the detection of various cytokines [Figure 2.1(b)] (Beck et al., 2001; Herbst et al., 1996, 1997), recombination activating genes (Meru et al., 2001b, 2002), or collagens (Pohle et al., 1996). For transcripts expressed at high levels, non-radioactive probes may suffice (Pohle et al., 1996) but we recommend using radiolabelled probes when commencing ISH experiments for the detection of a new gene. Only when a distinct labelling pattern has been established and a signal is detected after only a few days of exposure, are attempts to use non-radioactive probes likely to be fruitful.

Figure 2.1 (a) DNA ISH using 35S-labelled probes reveals the presence of EBV DNA in tumour cells of a nasopharyngeal carcinoma (black grains). (b) RNA ISH with 35S-labelled riboprobes demonstrates expression of interleukin-10 in Hodgkin and Reed-Sternberg cells of a Hodgkin lymphoma (black grains). (c) Expression of the EBERs in Hodgkin and Reed-Sternberg cells of a Hodgkin lymphoma is visualised using digoxigenin-labelled RNA probes (red nuclear staining). (d) Double-labelling immunohistochemistry and ISH reveals expression of CD30 (red staining) in a proportion of EBV-positive cells (black grains) in infectious mononucleosis. A colour version of this figure appears in the colour plate section

Figure 2.1 (a) DNA ISH using 35S-labelled probes reveals the presence of EBV DNA in tumour cells of a nasopharyngeal carcinoma (black grains). (b) RNA ISH with 35S-labelled riboprobes demonstrates expression of interleukin-10 in Hodgkin and Reed-Sternberg cells of a Hodgkin lymphoma (black grains). (c) Expression of the EBERs in Hodgkin and Reed-Sternberg cells of a Hodgkin lymphoma is visualised using digoxigenin-labelled RNA probes (red nuclear staining). (d) Double-labelling immunohistochemistry and ISH reveals expression of CD30 (red staining) in a proportion of EBV-positive cells (black grains) in infectious mononucleosis. A colour version of this figure appears in the colour plate section

Exceptions to this rule are certain viral transcripts which are expressed at high copy numbers. A major focus of our laboratories is the study of EBV infection. Initially, we and others used DNA ISH for the detection of the virus (Anagnos-topoulos et al., 1989; Niedobitek et al., 1989c). This method has been replaced by ISH for the detection of the small EBV-encoded RNAs (EBERs). The EBERs are transcribed at high copy number and therefore can be detected using non-radioactive riboprobes or commercially available oligonucleotide probes. In our laboratories, detection of the EBERs is achieved using digoxigenin-labelled RNA probes in the setting of malignant tumours and reactive conditions [Figure 2.1(c)] (Beck et al., 2001; Niedobitek et al., 1992; Spieker et al., 2000). However, we feel that for the detection of latently infected B-cells, which may be very rare, radioactive ISH with 35S-labelled riboprobes is more robust [Figure 2.1(d)] (Meru et al., 2001a; Niedobitek et al., 1992).

As discussed, ISH can be used in double-labelling experiments combined with immunohistochemistry or applying more than one probe. We have employed this technique to determine the phenotype of EBV-infected cells by combining

EBER-specific ISH with 35S-labelled or digoxigenin probes with the immuno-histochemical detection of lineage-specific antigens [Figure 2.1(d)] (Niedobitek et al., 1992, 1997). In this scenario, immunohistochemistry is carried out first using RNase-free conditions followed by ISH. ISH can also be performed simultaneously applying different probes. This approach has been used for the detection of cytokine or light chain transcripts in EBV-infected cells (Beck et al., 2001; Herbst et al., 1992, 1997; Spieker et al., 2000). For such experiments, digoxigenin-labelled EBER-specific probes and 35S-labelled cytokine- or light chain-specific probes are simultaneously applied to sections. Following hybridisation and washing, the non-radioactive probe is detected first using standard immunohistochemical procedures followed by dipping of slides into a nuclear track emulsion. Finally, triple-labelling experiments have been carried out using immunohistochemistry for the detection of lineage-specific antigens followed by double ISH with digoxigenin-labelled EBER-specific probes and 35S-labelled light chain-specific probes (Spieker et al., 2000).

Was this article helpful?

0 0

Post a comment