[Frontiers in Bioscience 2, c15-29, September 1, 1997]

	[Frontiers in Bioscience 2, c15-29, September 1, 1997 Reprints PubMed CAVEAT LECTOR
	IN SITU PCR. OVERVIEW OF PROCEDURES AND APPLICATIONS Carlos A. Muro-Cacho, M.D., Ph.D. Department of Pathology, H. Lee Moffitt Cancer Center and Research Institute and University of South Florida College of Medicine, Tampa, Florida, USA Received 1/10/97 Accepted 7/15/97 6. CONTROLS In In situ PCR protocols, it is imperative to control the numerous steps involved, not only to interpret a given signal in the proper context of sensitivity and specificity, but also to identify and correct some of the many potential pitfalls that are likely to occur (appendix 6). 6.1 Causes of false negativity. Poor thermal conduction by the slide or the block, uneven convection, adsorption of reagents to the glass, presence of DNA polymerase inhibitors, evaporation, excessive washing, and leakage of reagents, can all account for false negativity or inconsistency in the results (43-48). Control for some of these steps should be done during the optimization of the technique. Since most In situ PCR protocols are designed to detect nucleic acids within cells placed on slides, the starting material may be present in low amounts (43-45) and uncontrolled fixation of archival material may have further reduced the nucleic acid content, compromising the sensitivity of the procedure. DNA extracted from paraffin blocks is often shorter than 400 bp and mRNA fragments, in paraffin blocks, are usually smaller than 200 bp (1-10, 43-45). Although, some authors have been able to obtain full length DNA from paraffin-embedded material (2, 12-14), other investigators have succeed with In situ PCR only when using multiple primers designed to produce short PCR products and relatively long DNA probes or cocktails of oligonucleotide probes (8, 43-45). To ensure that a negative result is not due to poor penetration of reagents and accessibility of the target, it is advisable to include a control sample in which a sequence, such as actin, is amplified. 6.2 Causes of False Positivity. Specificity can be tested by recovering the product from the slide and analyzing it by gel electrophoresis. In indirect methods, the ISH step introduces an extra level of specificity, determined by the probe. Furthermore, another slide hybridized with an irrelevant probe should give no signal. "Nested" PCR can increase the specificity of PCR by using a second set of primers to amplify sequences within the first amplicon (46-48). Furthermore, testing, in the absence of PCR components (enzyme, primers or probe), should be done to make sure that the detection system does not render a non-specific signal. In experiments with virally-infected cells, it is advisable to mix infected and non-infected cells, at different ratios, and to correlate the results of the test with the expected sensitivity. 6.3 DNA-repair. Non-specific incorporation of labeled nucleotides may occur as a consequence of "repair" of DNA breaks by DNA polymerase (44). Excessive protease digestion (removal of histones), insufficient DNAse treatment (DNase may break DNA into oligonucleotides), poor fixation, suboptimal processing of the sample, apoptosis, or previous irradiation of the specimen can all induce DNA breaks. The slide, in which the DNA polymerase is omitted, however, should not show this artifact. This type of false positivity is extremely difficult to eradicate but it may be partially reduced by using, exonuclease-free, DNA polymerase or by performing several amplification cycles in the absence of the labeled nucleotide (44, 45). 6.4 Mispriming. The frequent amplification of "non-desired" sequences seems to be dependent upon factors such as specificity of primers, pH and ion concentration in the PCR mixture, and annealing temperature (1-8). Primers should be designed to detect short template sequences avoiding homology to non-desired sequences and between themselves. Since misprimimg occurs at lower melting temperatures, withholding the Taq polymerase, until 55^o C is reached reduces non-specific amplification without compromising the specific primer-target annealing. This "hot-start" procedure has been successfully applied to In situ PCR (46-48). An alternative method has been developed which takes advantage of the affinity interaction of Taq and Anti-Taq mAb (46). This interaction initially blocks the activity of the enzyme but the complex dissociates, at higher temperatures, freeing the enzyme at the appropriate time for specific amplification. In the hands of some researchers, this method has provided results similar to hot-start technique with less manipulation of the sample (40-42). Other "cold-start" procedures, using E. coli single-strand-DNA-binding protein SSB, or T4 gene protein, at ambient temperatures, have been devised (1, 5, 8, 40, 42, 48, 50-53). 6.5 Endogenous priming. This artifact, thought to occur due to priming of endogenous DNA and cDNA fragments, is difficult to eradicate and therefore the use of specific primers is imperative (1, 8). 6.6 Diffusion of the PCR products. Excessive protease digestion (loss of cellular boundaries) and excessive number of amplification cycles may contribute to the diffusion of the amplified product (8). If a product diffuses out of its original location, it may be preferentially amplified, producing a signal in a different cellular compartment or even in an extracellular location with the possibility of "labeling" adjacent negative cells (8, 12, 21). Proper processing of the sample, use of aldehyde-based fixatives, low number of cycles, generation of longer amplicons or more "bulky" products (21, 41), or the use of direct methods (incorporation of digoxigenin) may all reduce this type of false positivity (1, 4, 12, 48).

[Frontiers in Bioscience 2, c15-29, September 1, 1997
Reprints
PubMed
CAVEAT LECTOR

IN SITU PCR. OVERVIEW OF PROCEDURES AND APPLICATIONS

Carlos A. Muro-Cacho, M.D., Ph.D.

Department of Pathology, H. Lee Moffitt Cancer Center and Research Institute and University of South Florida College of Medicine, Tampa, Florida, USA

Received 1/10/97 Accepted 7/15/97

6. CONTROLS

In In situ PCR protocols, it is imperative to control the numerous steps involved, not only to interpret a given signal in the proper context of sensitivity and specificity, but also to identify and correct some of the many potential pitfalls that are likely to occur (appendix 6).

6.1 Causes of false negativity.

Poor thermal conduction by the slide or the block, uneven convection, adsorption of reagents to the glass, presence of DNA polymerase inhibitors, evaporation, excessive washing, and leakage of reagents, can all account for false negativity or inconsistency in the results (43-48). Control for some of these steps should be done during the optimization of the technique. Since most In situ PCR protocols are designed to detect nucleic acids within cells placed on slides, the starting material may be present in low amounts (43-45) and uncontrolled fixation of archival material may have further reduced the nucleic acid content, compromising the sensitivity of the procedure. DNA extracted from paraffin blocks is often shorter than 400 bp and mRNA fragments, in paraffin blocks, are usually smaller than 200 bp (1-10, 43-45). Although, some authors have been able to obtain full length DNA from paraffin-embedded material (2, 12-14), other investigators have succeed with In situ PCR only when using multiple primers designed to produce short PCR products and relatively long DNA probes or cocktails of oligonucleotide probes (8, 43-45). To ensure that a negative result is not due to poor penetration of reagents and accessibility of the target, it is advisable to include a control sample in which a sequence, such as actin, is amplified.

6.2 Causes of False Positivity.

Specificity can be tested by recovering the product from the slide and analyzing it by gel electrophoresis. In indirect methods, the ISH step introduces an extra level of specificity, determined by the probe. Furthermore, another slide hybridized with an irrelevant probe should give no signal. "Nested" PCR can increase the specificity of PCR by using a second set of primers to amplify sequences within the first amplicon (46-48). Furthermore, testing, in the absence of PCR components (enzyme, primers or probe), should be done to make sure that the detection system does not render a non-specific signal. In experiments with virally-infected cells, it is advisable to mix infected and non-infected cells, at different ratios, and to correlate the results of the test with the expected sensitivity.

6.3 DNA-repair.

Non-specific incorporation of labeled nucleotides may occur as a consequence of "repair" of DNA breaks by DNA polymerase (44). Excessive protease digestion (removal of histones), insufficient DNAse treatment (DNase may break DNA into oligonucleotides), poor fixation, suboptimal processing of the sample, apoptosis, or previous irradiation of the specimen can all induce DNA breaks. The slide, in which the DNA polymerase is omitted, however, should not show this artifact. This type of false positivity is extremely difficult to eradicate but it may be partially reduced by using, exonuclease-free, DNA polymerase or by performing several amplification cycles in the absence of the labeled nucleotide (44, 45).

6.4 Mispriming.

The frequent amplification of "non-desired" sequences seems to be dependent upon factors such as specificity of primers, pH and ion concentration in the PCR mixture, and annealing temperature (1-8). Primers should be designed to detect short template sequences avoiding homology to non-desired sequences and between themselves. Since misprimimg occurs at lower melting temperatures, withholding the Taq polymerase, until 55^o C is reached reduces non-specific amplification without compromising the specific primer-target annealing. This "hot-start" procedure has been successfully applied to In situ PCR (46-48). An alternative method has been developed which takes advantage of the affinity interaction of Taq and Anti-Taq mAb (46). This interaction initially blocks the activity of the enzyme but the complex dissociates, at higher temperatures, freeing the enzyme at the appropriate time for specific amplification. In the hands of some researchers, this method has provided results similar to hot-start technique with less manipulation of the sample (40-42). Other "cold-start" procedures, using E. coli single-strand-DNA-binding protein SSB, or T4 gene protein, at ambient temperatures, have been devised (1, 5, 8, 40, 42, 48, 50-53).

6.5 Endogenous priming.

This artifact, thought to occur due to priming of endogenous DNA and cDNA fragments, is difficult to eradicate and therefore the use of specific primers is imperative (1, 8).

6.6 Diffusion of the PCR products.

Excessive protease digestion (loss of cellular boundaries) and excessive number of amplification cycles may contribute to the diffusion of the amplified product (8). If a product diffuses out of its original location, it may be preferentially amplified, producing a signal in a different cellular compartment or even in an extracellular location with the possibility of "labeling" adjacent negative cells (8, 12, 21). Proper processing of the sample, use of aldehyde-based fixatives, low number of cycles, generation of longer amplicons or more "bulky" products (21, 41), or the use of direct methods (incorporation of digoxigenin) may all reduce this type of false positivity (1, 4, 12, 48).