S.H. Devoto, M. Mudryj, J. Pines, T. Hunter, J.R. Nevins, "A cyclin A-protein kinase complex possesses sequence-specific DNA binding activity: p33.MDSU/cdk2 is a component of the E2F-cyclin A complex," Cell, 68:167-76, 1992.
Joseph R. Nevins (Section of Genetics, Duke University Medical Center, Howard Hughes Medical Institute, Durham, N.C.): "A series of experiments performed in 1991 led to the realization that the cellular transcription factor E2F, previously studied as a component of transcription of an early adenovirus gene, was a target for the action of the retinoblastoma gene product. This finding coalesced studies directed at understanding the action of E1A as a transcriptional regula- tory protein, through the activation of E2F, together with studies directed at E1A's oncogenic role, by binding to proteins such as Rb. In short, it appeared that the binding of E1A to Rb was a consequence of the action of E1A to disrupt the E2F complex and activate E2F.
"At the same time that the E2F-Rb connection was made, it became clear that there was an additional E2F complex that contained the cell cycle regulatory protein cyclin A, another protein previously shown to be bound by adenovirus E1A. The E2F-cyclin A interaction was shown to accumulate in S phase of the cell cycle, coincident with the accumulation of the cyclin A protein. Although these findings were quite unexpected and striking, they were also perplexing to those in the cell cycle field, since previous studies had detailed the role of cyclin A as a cofactor for the cdc2 family of protein kinases.
"It was difficult to imagine a scenario in which cyclin A would interact with a kinase on the one hand and a transcription factor on the other. In this regard, the work of Steve Devoto, which demonstrated that the E2F-cyclin A complex also contained the cdk2 kinase as well as the Rb- related p107 protein--a result also described by Ed Harlow and colleagues at the same time (L.M. Cao, B. Faha, M. Dembski, et al., Nature, 355:176-9, 1992)--was important in placing the interaction in a more comfortable perspective. In short, this paper returned the cyclin A protein to its proper place--an association with the cdk2 kinase--but, in so doing, added an interesting twist: As a consequence of the presence of E2F, this kinase complex possessed sequence- specific DNA binding activity. I suspect that this aspect is responsible in part for the extensive citations.
"Although the ultimate significance of the association of a cyclin-dependent kinase with a transcription factor is still to be determined, these findings raise the interesting possibility that E2F might serve as a chaperone for the kinase complex, targeting the kinase to a site on DNA that may contain a substrate for the kinase. Whether this is significant for transcription or, alternatively, DNA replication, or even some other DNA-associated event, must await future studies."