A CDK-activating kinase network is required in cell cycle control and transcription in fission yeast

Cyclin-dependent kinases (CDKs) involved in cell cycle control require activation by phosphorylation, but CDK-activating kinase (CAK) has diverged between metazoans and budding yeast [1, 2]. Fission yeast has two CAKs: the essential Mcs6 complex [3-6], homologous to the metazoan CDK7 complex implicated in cell cycle control and transcription [7-9]; and Csk1 [5, 6, 10], a nonessential ortholog of budding yeast Cak1 [2, 11]. Both can activate the major CDK, Cdc2 [6], but Csk1 can also activate Mcs6 [5, 6], so it was unclear whether the pathway is a linear cascade or a network [12]. Here, we show that a mutation, mcs6-13, which selectively abrogates CDK activation, blocks both G1/S and G2/M transitions, but only when csk1(+) is absent. In contrast, gradual depletion or rapid inactivation of Mcs6 in csk1(+) cells causes cell separation defects or growth arrest, respectively, accompanied by decreased phosphorylation of RNA polymerase 11 (RNAP 11), but not of Cdc2. Finally, neither cell cycle arrest nor CAK failure is recapitulated by a second mutation in mcs6-13 that prevents Mcs6 activation by Csk1, indicating that Csk1 activates Cdc2 directly in vivo. Thus, Mcs6 acts in concert with Csk1 to activate Cdc2 and independently to support transcription and facilitate cell separation. Csk1 likewise has multiple physiologic targets, including Mcs6 and Cdc2.