These experiments document the presence of enzymatic activities in extracts of commonly used cell lines which interfere with the determination of CAT activity. We suspect that the deacetylase activity is the most important, as the extract of the H4IIE C3 cells was capable of completely deacetylating the mono- and diacetylchloramphenicol formed during a 2-hr incubation of CAT with chloramphenicol and acetyl-CoA. The results of the inhibitor experiments are consistent with the presence of proteases which degrade CAT, or a serine carboxylesterase. The interference was also reduced by about half by EDTA; a metalloenzyme (either a protease or esterase) may therefore be involved. This interference appears to be a common phenomenon. We have surveyed 23 different cell types for the presence of the interfering activity and found it in 15. The interference was particularly prominent in several neuroendocrine and hepatoma cells. We took advantage of the effect of EDTA and the heat stability of CAT to eliminate the interference. Addition of 5 mM EDTA and a 10-min incubation of the sonicated cell suspension at 60 degrees prior to centrifugation abolished the interference in all cell lines tested. It is important to note that in order to reveal any CAT activity in some of the extracts (e.g., PC-12 or Hep3B), it was necessary to run the CAT assay for 2 hr. The control assays were therefore run almost to completion, and were well beyond the linear range of the assay. Therefore, the small differences which we observed between the heat-treated and control samples in some instances (e.g., rice, corn, or HeLa cells) will be dramatically amplified when the CAT assay is performed under conditions in which only a small percentage of the substrate is converted to product. After these studies had been performed, we found that others have also recommended heat treatment of the cell extract prior to CAT assay. We concur with this recommendation. We suggest that EDTA plus heat treatment of the cell extract should be incorporated into all CAT assay protocols, unless it has been previously determined that extracts of the cells used do not interfere. Furthermore, the heat treatment step should be used whenever the activity of promoter-CAT constructs is compared among different cell lines, as is often done to define tissue-specific expression.