Quantitative measurements of the changes in protein thiols in cervical intraepithelial neoplasia and in carcinoma of the human uterine cervix provide evidence for the existence of a biochemical field effect.

Quantitative micromethods have been used for measuring reactive protein thiols (PSHr), total reactive protein sulfur (TRPS), total protein thiols (PSHt), and protein disulfides (PDS) in fixed frozen sections of human uterine cervix. PSHr and TRPS were stained using 2,2'-dihydroxy-6,6'-dinaphthyl disulfide; PSHt and PDS were stained using mercurochrome methods. Microspectrophotometric measurements were made on the stained sections using a microdensitometer with associated data processing; the results obtained for areas of epithelium and stroma were converted to absorbance values per micron 2. Samples of uterine cervix that were diagnosed as containing cervical intraepithelial neoplasia (CIN) I-III or carcinoma were examined and compared with samples of normal uterine cervix. Measurements were made not only on identified lesions but also on apparently normal tissue obtained from the same cervix. Epithelial/stroma ratios (E/S) were calculated for PSHr, TRPS, PSHt and PSHt + PDS; in addition, the double ratios of PSHr/TRPS and PSHt/PSHt + PDS were also calculated for E/S. The mean E/S values for PSHr and PSHt were significantly different for all types of lesion compared with control samples. The E/S ratios for apparently normal tissue obtained from cervices with CIN or carcinoma were also significantly different compared with corresponding control values, indicating a field effect. There was a considerable degree of overlap between individual values in the control groups versus those obtained with each type of lesion. The corresponding mean E/S values for TRPS and for PSHt + PDS in the samples containing lesions were not significantly different from control means except for the group containing CaCx. However, the mean values for the double ratios (PSHr/TRPS and PSHt/PSHt + PDS) were significantly different in the groups containing lesions compared with the controls. Moreover, apparently normal tissue obtained from cervices containing CIN or carcinoma had different mean values compared with the controls, confirming the existence of a field effect. The degree of overlap of individual values in the lesion groups compared with the control values was much less with double ratio values than previously noted for single ratio values. In consequence, the double ratio measurements clearly discriminated CIN I + II and CIN-III from controls. Our data show that CIN is associated with marked changes in tissue protein thiols and disulfides and that these differences extend to neighboring apparently normal tissue indicative of a field effect.