Uncovering the complexity of RAS signaling networks

by Matheny, Sharon A.

IMP is a newly described Ras effector that negatively regulates the ERK pathway, possibly at the level of KSR. Examination of the primary amino acid sequence reveals conserved motifs that suggest multiple modes of IMP regulation. One possible regulatory mechanism, Ras-stimulated autoubiquitination of IMP, was previously described. Here, we present data that suggests the existence of additional regulatory inputs. Introduction Mutations that result in a gain-of-function essentially reflect the loss of regulatory control over enzymatic activity. It is therefore just as important to understand how a protein is regulated as is its function. As a ubiquitin ligase, it was obvious to test IMP for self-modification; as discussed, we found that IMP does autoubiquitinate in the presence of oncogenic Ras (Ras12V). Analysis of the primary amino acid sequence indicated that IMP may be regulated on multiple levels in different ways. Two conserved motifs for posttranslational modification were identified: PXSP for ERK phosphorylation and ?KXE for sumoylation. Here, we show that IMP can be phosphorylated by ERK and that this correlates with IMP translocation to the nucleus. We also demonstrate that SUMO exerts negative effects on IMP inhibitory activity. Additionally, a predicted UBP- ZnF (SMART, Simple Modular Architecture Research Tool, http://smart.emblheidelberg.de) was examined for contribution to regulation of autoubiquitination. 43 44 Together, these data point to the existence of multiple regulatory inputs that control IMP activity. Results and Discussion As indicated in Chapter 2, IMP autoubiquitination may be induced upon Ras activation, resulting in downregulation of inhibition to allow signal transmission through the ERK pathway. To see if autoubiquitination could occur in response to a biological stimulus, we induced 293 cells with EGF in the presence or absence of a proteosome inhibitor (LLnL) to visualize accumulation of laddering due to poly-ubiquitination. As seen in Fig 3.1b, high molecular weight IMP species are present even without EGF in cells pretreated with LLnL, suggesting that IMP is continually ubiquitinated and degraded by the proteosome. Further, it confirms that the autoubiquitination of IMP induced by Ras12V in Chapter 2 is true to the biology of IMP in this pathway and not simply due to overexpression of an activator. The same bands accumulate in the absence of LLnL by 60 minutes of EGF exposure, indicating that the rate of IMP ubiquitination exceeds the rate of degradation, or that some poly-ubiquitinated IMP proteins are not degraded. The same experiment was performed in HeLa cells, in which similar laddering was observed (not shown). A puzzling phenomenon in HeLas had been noted in past experiments in which EGF stimulation caused a ‘spike’ in IMP protein levels. The band would appear around 5 or 10 minutes and disappear within two minutes. Interestingly, this is the same point at which MEK and ERK approach peak activation. Immunoblotting revealed that this was likely the result of epitope masking, as an IMP antibody to the N-terminus detected IMP in all 45 a