Phosphorylation of small GTPases has been

Phosphorylation of small GTPases has been also observed to affect binding affinity for the GDP/GTP cycle regulators notably GDP dissociation inhibitor (GDI) [10]. Indeed the EGF or cAMP-dependent phosphorylation of Cdc42 is associated with enhanced Cdc42–GDI interaction [8], [11]. RhoA inhibition by PKA or cGK phosphorylation has been shown to occur also through enhanced RhoGDI interaction rather than direct perturbation of guanine nucleotide exchange factor (GEF) or GTPase activating protein (GAP) status [7], [8], [12]. Lastly, phosphorylation of the small GTPases could change the affinity for or discriminate their partners. For example phosphorylation of Rem and Gem, GTPases of the RGK family, is critical for their association with 14-3-3 proteins [13], [14]. An alternative pathway was proposed for regulating RhoA signalling whereby GTP-bound RhoA, when phosphorylated by PKA or cGK, could be separated from its putative effector(s) independently of its GTP/GDP cycling [7]. That was illustrated nicely in the recent work of Nusser et al. which has shown that serine phosphorylation of RhoA differentially affects binding to its effectors, PKN/PRK, ROCK and rhotekin [15]. RhoB belongs to the highly related Rho GTPase family (including Cdc42, Rac, RhoA and RhoC) best characterized for their effects on the chlorpromazine hydrochloride cytoskeleton [16]. While its close relatives, RhoA and RhoC, promote oncogenesis and invasion, RhoB rather plays a negative regulator role in oncogenesis [17], [18], [19], [20], [21]. Moreover we [20] and others [22], [23] have shown that RhoB protein expression levels are inversely related to malignancy of various tumors. Furthermore RhoB is essential for UVB-induced cell survival [24] and inhibits radiation-induced mitotic cell death [25]. Additionally endosomal RhoB has been implicated in the regulation of the endocytic trafficking of different proteins such as EGF-R [26], [27], [28] as well as in the outward movement of Src to plasma membrane [29] or CXCR2 sorting decision [30]. Unlike most small GTPases, RhoB is short lived and is an immediate-early response gene induced by growth factors [24], [31], [32] and genotoxic stress [24], [33], [34]. RhoB is a highly controlled protein, regulated at many steps of its biosynthetic pathway including transcriptional and post-transcriptional activity [24], [35], [36], [37], [38] and RNA [24], [39], [40] or protein stability [41]. Moreover post-translational modifications are very important in RhoB activity such as prenylation [42] or palmitoylation [43].
Materials and methods
Results
Discussion We provide both in vitro and in vivo evidence for phosphorylation of the small GTPase RhoB on serine 185 by an isoform of CK1, and for a critical role of this post-translational modification in its cellular activities. The in silico analysis by NetPhos2.0 software predicted that CK1 could phosphorylate RhoB but on Thr77, Ser88 or Thr163. Nevertheless the mass spectrometry and nano-LC-MS/MS analysis of RhoB phosphorylated in vitro by CK1 allowed the identification of a RhoB monophosphorylated isoform and indicated that RhoB is phosphorylated by CK1 on serine 185. The radical reduction of the phosphorylation by CK1 of GST-RhoBS185A strengthened this observation. Unexpectedly the sequence around Ser185 does not conform to the canonical consensus for CK1 substrates; which stipulates either phosphoamino-acids or acidic residues in position n-3 from the serine [56], as for the RhoB potential sites proposed by NetPhos2.0 (Fig. 1). The sequence downstream of Ser185 is also distinct to the sequence recognized by CK1 in NF-AT4 or β-catenin, including an SLS motif followed downstream by a cluster of acidic residues [57]. Thus the phosphorylation of a substrate by CK1 appears to be not strictly dependent on a consensus sequence, as it was shown that it may depend on the tertiary structure of the substrate [56]. In contrast NetPhos2.0 analysis predicted that Ser185 could be a substrate of the kinases CamKII and PKA. The in vitro kinase reactions with purified kinases showed that CamKII was able to phosphorylate GST-RhoB (data not shown) while on the contrary PKA did not produce any detectable phosphorylated GST-RhoB (Fig. 2). Thus whereas RhoA and RhoB are highly homologous proteins they appeared to be specific substrates of distinct kinases, such as CK1 and PKA. It is noteworthy that the phosphorylable amino acids are located in the hyper variable region of the Rho proteins that could explain this discrepancy. For instance the RhoA Ser188, target of PKA, follows the basic amino acids of the C-terminal region not present in RhoB, which instead possess palmitoylated cysteine.