Gerard J. Nuovo

Director, MGN Medical Research Laboratories, Setauket, New York 11733, USA

Received: 07/25/96; Accepted: 10/02/96; On-line 11/01/96

1. ABSTRACT

RT in situ PCR allows for the routine and rapid detection of low copy viral and human RNAs. Success with RT in situ PCR is best accomplished with formalin fixed, paraffin embedded material, which allows the study of archival material. The key variable for RT in situ PCR is protease digestion. The optimal digestion time, which is determined by testing a variety of protease digestion times, is defined by an intense signal in the nuclei of most cells irrespective of the primers used, and a loss of this signal with overnight digestion in DNase. This permits the target specific direct incorporation of the labeled nucleotide into the amplified cDNA. A lack of signal with the negative control (DNase, no RT) and an intense nuclear signal in most cells with the positive control (no DNase) is prerequisite for success with RT in situ PCR. The localization of the signal (cytoplasmic for human mRNAs and restricted to certain cell types) is another important indicator of successful RT in situ PCR. The one step rTth system allows for the reproducible amplification and detection of low copy RNA targets within a few hours. Matrix metalloprotease (MMPs) and their inhibitors (TIMPs) in cervical cancer are used as a model system for RT in situ PCR. Analysis of MMP and TIMP expression in cervical cancer demonstrates the following: 1) the signal localizes to the cytoplasm of invasive cancer cells and the surrounding stromal cells; 2) no signal is evident in the adjacent carcinoma in situ cells (non invasive component) or the normal epithelium.

Cervical cancers of poor prognosis showed a marked increase in the percentage of cells expressing MMP versus TIMP as compared to microinvasive cervical cancer, which has an excellent prognosis.