Functional characterization of human NDR kinases : novel regulatory mechanisms and a path towards direct substrates

Protein kinases are important regulators of signal transduction pathways in both unicellular and multicellular organisms. They play critical roles in biological processes such as cell growth, division, differentiation, adhesion, motility and cell death. Given their central role in mediating cellular function and physiological responses, perturbation of protein kinase signaling can cause many diseases, including cancer and diabetes. Out of the 518 protein kinases encoded by the human genome, approximately 60 belong to the AGC group of Ser/Thr protein kinases, including the NDR kinase family. Members of this family are highly conserved from yeast to men and regulate important processes such as mitotic exit, cell polarity, neuronal and epithelial morphology, growth, proliferation and apoptosis. Despite the fact that NDR kinase family members regulate important cellular processes, direct downstream targets have only been identified recently. Furthermore, the regulation of NDR kinase signaling by upstream kinases of the Ste20-like family or co-activator proteins of the MOB family is also remarkably conserved. The human genome encodes for four NDR kinases: NDR1, NDR2, LATS1 and LATS2. Whereas the molecular mechanisms of NDR kinase regulation have mostly been worked out using human NDR1/2 kinases, biological functions have just started to emerge. Human NDR kinases were implicated in regulating centrosome duplication, mitotic chromosome alignment and apoptosis signaling. Additionally, the human MOB family consists of six distinct members (hMOB1A, -1B, -2, -3A, -3B and -3C), with hMOB1A/B the best studied due to their tumor suppressive functions through regulation of NDR/LATS kinases. The roles of the other MOB proteins are not as well defined.
We investigated the role of human MOB proteins in NDR/LATS kinase regulation. We found that three hMOB proteins did not bind to or activate all human NDR kinases and that hMOB2 was an NDR-specific binder. Furthermore, we describe competitive binding of hMOB1A/B and hMOB2 towards the NTR of human NDR1/2. Interestingly, in contrast to hMOB1A/B, hMOB2 is bound to unphosphorylated NDR1/2. Moreover, RNAi-mediated depletion of hMOB2 protein resulted in increased NDR activity. Consistent with these findings, hMOB2 overexpression impaired not only okadaic acid-induced activation of NDR but also the functional roles of NDR in death receptor-induced apoptosis and centrosome duplication. In summary, our data indicate that hMOB2 is a negative regulator of human NDR1/2 kinases.
Additionally, we established a basis for the discovery of additional human NDR kinase substrates. We employed the chemical genetic method developed by Shokat and colleagues to create analog-sensitive variants of NDR1/2 kinases. Subsequently, we have tried to identify direct targets of analog-sensitive NDR1(M166G) by performing in vitro kinase assays on cell lysates and immunocomplexes in the presence of a radiolabeled ATP analog and observed a specific and reproducible pattern of labeled bands in reactions containing NDR1(M166G) immunocomplexes. Our data together with the recent identification of the first in vivo substrate of human NDR1/2 kinases, p21, should stimulate further efforts to dissect the downstream signaling of mammalian NDR kinases.