The interferon (IFN)-mediated protein kinase activity in extracts from mouse L-929 cells is manifested by the phosphorylation of an endogenous 67 kD molecular weight (mw) protein in the presence of double-stranded (ds) RNA. This protein kinase activity can also be assayed after partial purification on poly(I) X poly(C)-Sepharose under phosphatase-free conditions. By the use of this latter technique, here we investigated the distribution of the protein kinase activity in different cellular compartments. Most of the protein kinase activity is found in the post-ribosomal supernatant (S100) fraction, while a small portion of it is associated with the ribosomal salt wash (RSW: 0.5 M KCl eluate of ribosomal pellet) and nuclear fractions. These results are in contrast to several observations in the literature in which the protein kinase activity is thought to be associated with the ribosomal pellet. This controversy results from the conditions used for assay of the protein kinase activity. In fact, when the kinase is assayed in crude extracts supplemented with dsRNA, very little kinase activity is detectable in the S100 fraction compared to the RSW fraction. The S100 fraction contains a high level of phosphatase(s) activity which interferes with the protein kinase assay and might account for the misinterpretation observed in the literature. Some recent results have implicated a correlation between the dsRNA-dependent protein kinase responsible for the phosphorylation of the 67 kD protein and a polyamine-dependent protein kinase which phosphorylates a similar molecular weight protein, subunit of ornithine decarboxylase (Orn Dcase). Here, we show that Orn Dcase does not bind to poly(I) X poly(C)-Sepharose and polyamines do not substitute the requirement of dsRNA for the phosphorylation of the 67 kD protein.