Attempts have been made in recent years to increase the sensitivity of nucleic acid detection in situ. Basically two different strategies have been employed: amplification of target and amplification of signal.
Amplification of target, RNA or DNA, by in situ PCR, either with direct incorporation of labelled nucleotides or followed by detection using labelled oligonucleotide probes, has been reviewed elsewhere (Hawkins and Dodd, 2000; Long, 1998; Nuovo, 2001). Alternatively, Hofler et al. (1995) described a method for the in situ detection of measles virus RNA employing reverse transcriptase with primers including a T7 RNA polymerase site to generate a cDNA intermediate followed by T7 RNA polymerase-directed generation of RNA transcripts in situ and detection using 35S-labelled oligonucleotide probes. A related technique, primed in situ labelling (PRINS), is based on the hybridisation of oligonucleotide probes followed by extension using Taq polymerase in the presence of labelled nucleotides (Coullin et al., 2002; Wilkens et al., 1997).
Signal amplification has been another focus of research and these approaches will be discussed in more detail. When using radioactive probes, using more than one labelled nucleotide, e.g. 35S-ATP in addition to 35S-UTP, in the transcription reaction may improve sensitivity (Franco et al., 2001; Ky and Shughrue, 2002; Sedlaczek et al., 2001), although others have not found this to be beneficial (Poulsom et al., 1998). Increasing the photographic development time may also lead to a more intense signal. However, this also increases the background and a better signal-to-noise ratio is achieved by increasing the exposure time (Franco et al., 2001).
With regards to non-radioactive probes, application of multiple layers of detection systems (e.g. mouse anti-digoxigenin antibody followed by biotiny-lated horse anti-mouse reagent and an enzyme-labelled avidin-biotin complex) may improve the signal intensity (Niedobitek et al., 1989b; Speel et al., 1998). However, such systems are also prone to an increased background (Speel et al., 1998). The most significant and practically useful advance in this field has been the description of tyramide signal amplification (TSA) systems. Initially developed for membrane immunoassays (Bobrow et al., 1989, 1991), this method has been successfully applied to immunohistochemistry and ISH (Adams, 1992; Aigner et al., 1999; Herbst et al., 1998; Niedobitek et al., 1997; Speel et al., 1998; Zaidi et al., 2000; Zehbe et al., 1997). TSA is based on the horseradish peroxidase-catalysed deposition of hapten-labelled tyramide molecules at the site of antibody or probe binding. In the case of ISH, peroxidase usually is brought to the site of hybridisation by employing a hapten-labelled probe, followed by binding of a peroxidase-labelled hapten-specific antibody (Speel et al., 1998). Since any of the reagents employed for probe detection may contribute to background labelling, use of probes directly labelled with peroxidase followed by TSA has been advocated (van de Corput et al., 1998). However, such probes are not widely available yet. Use of biotin-labelled probes and biotinylated tyramide can result in excessive background staining due to the presence of endogenous biotin in many tissues (Niedobitek and Herbst, 1991; Niedobitek et al., 1989a; Speel et al., 1998). This can be avoided by using tyramides linked to other markers e.g. digoxigenin, fluorochromes, or dinitrophenol (DNP) (Kolquist et al., 1998; Lewis et al., 2001; Schmidt et al., 1997; Speel et al., 1998; Zaidi et al., 2000). Application of TSA systems in ISH results in an increased sensitivity with an estimated amplification factor of 5-25 (Schmidt et al., 1997; Speel et al., 1998). Accordingly, probe concentration may be reduced significantly (Yang et al., 1999). Using TSA, detection of rare RNA transcripts by ISH has been reported, for example including those encoding for the reverse transcriptase component of human telomerase (Kolquist et al., 1998), insulin (Speel et al., 1998), and collagens (Aigner et al., 1999). In the latter study, two rounds of TSA using biotinylated tyramide have been employed (Aigner et al., 1999). At the DNA level, detection of single copies of human papillomavirus (HPV) genomes by ISH has been reported with biotin-labelled DNA probes using TSA followed by streptavidin nanogold and autometallography with silver acetate (Zehbe et al., 1997). A commercially available detection system using a secondary antibody linked to a dextran polymer coupled to numerous enzyme molecules has been reported to achieve almost the same sensitivity as TSA detection reagents in HPV-specific ISH (Wiedorn et al., 2001). Another method of potential use for ISH has been designated rolling circle amplification (RCA). This method is based on the observation that circularised oligonucleotides can support a replication reaction analogous to the replication of certain viral genomes. Application of this technique to RNA ISH has been reported by Zhou et al. (2001), who have used digoxigenin-labelled probes specific for |-actin. Bound probe was detected using an anti-digoxigenin antibody covalently linked to an RCA primer. Alternatively, RCA can be supported by primers which, in addition to a target-specific sequence, contain an RCA primer sequence (Lizardi et al., 1998; Zhong et al., 2001). A circular oligonucleotide is then hybridised to the RCA primer, and RCA is initiated by DNA polymerase generating a large DNA molecule which remains attached to the site of hybridisation (Baner et al., 1998; Zhong et al., 2001; Zhou et al., 2001). This is detected either by direct incorporation
WHAT WE DO IN OUR LABORATORIES
of labelled nucleotides or by hybridisation to a labelled probe (Lizardi et al., 1998; Zhong et al., 2001; Zhou et al., 2001). The potential usefulness of this approach for increasing the sensitivity of ISH requires further investigation.
Was this article helpful?