Although multiple phosphorylation sites are often clustered in substrates, the mechanism of phosphorylation within clusters has not been systematically investigated. Intriguingly, in addition to acidic residues, protein kinase CK2 can use phosphoserine residues as consensus determinants suggesting that CK2 may act in concert with other kinases. We used a peptide array approach to outline optimal consensus sequences for hierarchical phosphorylation by CK2, both in the context of processive, multisite phosphorylation, and in concert with a priming proline-directed kinase. Results suggest that hierarchical phosphorylation involving CK2 requires precise positioning of either multiple phosphodeterminant residues or specific combinations of canonical determinants and phosphodeterminants, and can be as enzymatically favorable as canonical CK2 phosphorylation. Over 1600 human proteins contain at least one CK2 hierarchical consensus motif, and ~20% of these motifs contain at least one reported in vivo phosphorylation site. These motifs occur non-randomly in the human proteome, with significant enrichment in proteins controlling specific cellular processes. Taken together, our results provide strong in vitro evidence that hierarchical phosphorylation may contribute to the regulation of crucial biological processes. In addition, the results suggest a mechanism by which CK2, a constitutively active kinase, can be a regulatory participant in cellular processes.

CK2 is a constitutively active protein kinase with key regulatory roles in many cellular signaling events which has been implicated in several human diseases. To investigate its roles in biological events and potential as a therapeutic target, several potent CK2 inhibitors have been developed including TBB and its derivatives that have been employed in many studies. Despite the utility of these compounds, a precise understanding of their mode of action within cells remains incomplete. In fact, cells are typically treated with inhibitor concentrations (>5 μM) that are orders of magnitude higher than their in vitro inhibitory constants (<0.05 μM). Accordingly, we hypothesized that CK2 inhibitors could have off-target effects that are not recognized when inhibitors are profiled using panels of recombinant protein kinases. To address this issue, we combined structural modeling with inhibitor-affinity purification and proteomics to test the specificity of derivatives of TBB using whole cell lysates of HeLa cells. While these studies confirmed that CK2 does bind to the immobilized inhibitor, several other abundant ATP/GTP-binding proteins were also identified and confirmed. These results suggest that highly abundant nucleotide binding proteins may limit the bioavailability of the free inhibitor and interactions with CK2 in the cellular environment. This article is part of a Special Issue entitled: From protein structures to clinical applications.