CDKs mainly consist of an N-terminal and a C-terminal lobe with distinct functions. The N-terminal lobe (N-lobe) contains a glycine-rich inhibitory loop and a so-called C helix that binds cyclins. When cyclin is bound, two alpha helices change position to enable ATP binding, in addition to the movement of the activation loop. In addition to activating the CDKs, cyclins can directly bind the substrate or localize the CDK to a subcellular area where the substrate is found. For example,
cyclin B1 and
B2 can localize CDK1 to the nucleus and the Golgi, respectively, through a localization sequence outside the CDK-binding region. The identity of the CDK-activating kinase (CAK) that carries out this phosphorylation varies among different model organisms. The timing of this phosphorylation also varies; in
mammalian cells, the activating phosphorylation occurs after cyclin binding, while in yeast cells, it occurs before cyclin binding. CAK activity is not regulated by known cell cycle pathways, and it is the cyclin binding that is the limiting step for CDK activation.
Non-cyclin activators Several proteins other than cyclins can activate CDKs. For instance, certain
viruses encode proteins with
sequence homology to cyclins. One much-studied example is
K-cyclin (or v-cyclin) from
Kaposi sarcoma herpes virus, which activates CDK6. The vCyclin-CDK6 complex promotes an accelerated transition from G1 to S phase. This leads to the removal of inhibition on Cyclin E–CDK2's enzymatic activity and promotes transformation and tumorigenesis. During neuronal differentiation, CDK5 is activated by the
p35 and
p39 proteins. This activation is important in growth of the
dendritic spine and in
synapse formation. The
RINGO/Speedy proteins can also activate CDKs (primarily CDK1 and 2), despite lacking homology to cyclins. These proteins alter the substrate specificity of the CDKs in addition to regulating activity. ==Medical significance==