• 2018-07
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • CDK lacks DNA binding domains and nuclear


    CDK6 lacks DNA-binding domains and nuclear localization sequences, and therefore will probably be transported to the nucleus by a ‘piggy-back’ mechanism and needs to contact its specific sites indirectly through DNA-binding proteins. Besides NF-κB p65, a further candidate transcription factor is PAX4 (paired box 4), whose consensus DNA-binding motif is associated with a fraction of CDK6 peaks. However, CDK6 does not only associate with p65 because in BCR–ABL (breakpoint cluster region–Abelson oncogene homolog)-transformed B-acute lymphoid leukemia cells the kinase interacts also with STAT3 and the AP-1 transcription factor subunit JUN (Jun protooncogene) [39], as summarized in Figure 2B. The JUN protein is required for the chromatin recruitment of CDK6 to the vascular endothelial growth factor A (VEGFA) promoter and for expression of this pro-angiogenic factor. By contrast, binding of CDK6 to the p16INK4A promoter crucially depends on the presence of STAT3 [39], showing that CDK6 can utilize different transcription factor complexes to mediate its association with distinct chromatin regions. In agreement with the dynamic association of CDK6 with different transcription factor complexes, a previous study revealed the testosterone-inducible interaction of CDK6 with the androgen receptor, a nuclear hormone receptor that regulates transcription of the prostate-specific antigen (PSA) gene [40]. These larger transcriptional complexes contain further cofactors augmenting CDK6-driven transcription such as the p65 and nuclear hormone receptor-interacting protein TRIP6 41, 42. The exact composition and function of these nuclear complexes is unknown and requires more study. It is currently not clear whether the nuclear function of CDK6 in inflammatory gene expression depends on the presence of cyclins. ChIP experiments showed the co-occurrence of CDK6 and cyclin D2 at the VEGFA promoter and p16INK4A promoters. However, VEGFA mRNA transcription was independent from cyclin D [39]. CDK6 at p65-bound promoters is not associated with D-type cyclins [36]. By contrast, nuclear CDK4 and CDK6 binding with proteins regulating the po1 [e.g., p130 and E2F4 (E2F transcription factor 4)] are typically bound to cyclins [43]. These recent mechanistic studies help to explain the earlier finding that CDK6-deficient mice are protected from the lethal effect of LPS, a pathological process that is largely due to rapid release of inflammatory cytokines and chemokines [44]. Furthermore, unbiased screens identified the CDK6 locus to be among the regions conferring risk of rheumatoid arthritis [45].
    Kinase-Dependent and -Independent Functions of CDKs in Inflammation The contribution of CDK kinase activity to the stimulation of inflammatory gene expression does not reveal a coherent picture. There are cases where CDK kinase activity is clearly required, as seen by numerous studies showing the anti-inflammatory effects of CDK kinase inhibitors 46, 47, 48, 49. The relative contribution of CDK kinase function is less pronounced in the case of IL-1-triggered inflammatory gene expression. The CDK6 T177A mutant used by Handschick and coworkers [36] cannot undergo CAK-mediated activation but still leaves basal catalytic activity intact. This study also showed that CDK6 kinase activity is not augmented by IL-1, raising the possibility that basal CDK6 activity is sufficient to mediate some proinflammatory effects. Because the ATP-competitive inhibitor PD332991 (palbociclib) partially inhibits IL-1-induced NF-κB-dependent reporter gene activation [38], kinase-dependent and -independent functions of CDK6 cooperate to trigger IL-8 expression. A clearly kinase-independent function for CDK6 underlies its ability to trigger expression of p16INK4A and VEGF-A. This was revealed in experiments showing elevated levels of p16INK4A or VEGF-A in CDK6-deficient cells reconstituted with CDK6 mutants lacking any kinase activity and also in the presence of palbociclib [39]. In addition, CDK6 driven transcriptional activation of the PSA promoter is independent from its kinase activity [40]. These kinase-independent functions could be due to functional compensation of the inactivated kinase by other CDKs, a phenomenon that could also explain missing cell cycle phenotypes in most CDK knockout mice [2]. In addition, many protein kinases display phosphorylation-independent functions by functioning as adaptors or scaffold proteins 50, 51. This also explains why results from knockout systems often differ from those obtained with small-molecule inhibitors. Because the field of proinflammatory CDK function is still in its infancy, systematic studies will be necessary to reveal which of the numerous inflammatory input signals and effector pathways is amenable to kinase inhibitors. Figure 2C summarizes the current observations on chromatin functions of CDK6 and the differential requirements for kinase activity, cyclins, and nuclear cofactors.