Comparison of indirect and direct in-situ polymerase chain reaction in cell preparations and tissue sections. Detection of viral DNA, gene rearrangements and chromosomal translocations.

Different approaches to the in-situ polymerase chain reaction (in-situ PCR) were compared in the detection and in-situ localization of chromosomal translocations (t14; 18) immunoglobulin gene rearrangements and viral DNA (cytomegalovirus, hepatitis B-virus) in cell suspensions, cytospins and tissue sections. Single and multiple primer pairs were compared in the amplification step of indirect in-situ PCR and long genomic probes or internal oligonucleotide probes in the subsequent in-situ hybridization (ISH). For direct in-situ PCR, in which amplification products were directly labeled with digoxigenin-11-dUTP during PCR and detected immunohistochemically, only single primer pairs were used for amplification. In-situ PCR yielded best results in the cell suspensions and worked less efficiently in cytospins or tissue sections. Quantification of the results obtained in artificial cell mixtures yielded only an approximate correlation between the number of expected and observed positive cells. The specificity of the results was greater with indirect in-situ PCR than direct in-situ PCR, where false positive results were frequent. Successful indirect in-situ PCR in tissue sections required the use of multiple primer pairs for amplification and genomic probes for detection by ISH. False positive results in direct in-situ PCR were caused by primer-independent, but DNA polymerase- and cycling-dependent incorporation of digoxigenin-labeled nucleotides into cellular DNA, possibly related to DNA repair and/or internal priming. Non-specific results were most marked in tissue sections and were much less frequent in cell suspensions. In-situ PCR includes a number of different techniques, which are not equally applicable to different starting materials. Accurate interpretation of the results requires vigorous controls.