Roose LabHome.html
Department of Anatomy

Antigen receptor - Ras signals and T cell function

LAB PAPERSLab_Papers.html

Roose Laboratory | University of California, San Francisco | Department of Anatomy | 513 Parnassus Avenue HSW 1330 | San Francisco CA 94143-0452

Our laboratory is interested in genetic lesions that lead to abnormal Ras signal transduction events and development of diseases such as autoimmunity and cancer. Active Ras communicates to many downstream targets and affects many cell biological responses. The potency of active Ras is illustrated by the fact that activating mutations in the Ras genes are among the most common genetic lesions leading to cancer. Our goal is to understand the mechanism of activation of the small GTPase Ras by Ras exchange factors, comparing normal physiology with pathology.

There are four main directions in the lab:

  1. 1. Mechanistic understanding of Ras activation by Ras exchange factors.

2. Altered Ras-kinase signaling in autoimmune T cells.

  1. 3. Interfering with oncogenic Ras signals in leukemia.

4. Ras exchange factors in epithelial cells and carcinoma.

Below are several vignettes of our research.

We previously established that lymphocytes express two types of Ras activators or RasGEFs (Ras guanine nucleotide exchange factors), Rasgrp and Sos. Both types of RasGEFs are activated downstream of antigen receptors, namely the TCR or BCR, to exchange the GDP on Ras for GTP. Thus, Rasgrp and Sos seemingly fulfill redundant functions. However, mouse models pointed to a non-redundant role for Rasgrp1 in thymocyte development and our cell line studies revealed that Rasgrp1 dominates Ras activation with Sos depending on Rasgrp1. With in silico and in vitro approaches we uncovered that Rasgrp signals to Ras in a graded, or analog, manner, while Sos does to in an exponential, or digital way. We predicted that these distinct Ras activation patterns play an important role to regulate T cell function. We are currently testing these hypotheses with synergistic in silico, in vitro, and in vivo  approaches.

Meet our collaborators
Click !Collaborators.html

The activity of the RasGEF Sos with its diverse protein domains is regulated in many different ways. These include but are not limited to membrane recruitment by the adapter Grb2 and interactions with lipids in the plasma membrane. We focus on the allosteric pocket in Sos. The crystal structure of Sos with Ras revealed that Sos contains a second binding pocket for Ras, in addition to the traditional GEF pocket where Ras binds and its GDP is removed. This second pocket enables a positive feedback loop: the product RasGTP binds back to Sos to induce an allosteric change to robustly enhance Sos’ GEF activity. In our in silico and in vitro models the allosteric pocket in Sos is critical for digital Ras activation. These studies also explained the dependence of Sos on Rasgrp1: Rasgrp1 generates the initial RasGTP required to efficiently engage Sos’ positive feedback loop. We are continuing these directions to uncover the biological role of Sos’ allosteric pocket.

The biological role of allosteric activation of SOS

Much less is known about the different domains or the regulation of the RasGEF Rasgrp1. Its C1 domain mediates membrane recruitment by binding to the second messenger DAG (diacylglycerol), bringing it in proximity to Ras. We and others have established that phosphorylation of Rasgrp1 on threonine 184 (T184 positioned in the REM domain) modestly enhances its RasGEF activity. The amino acid sequence of Rasgrp1 also predicts a pair of EF hands, structures that often bind calcium to subsequently undergo conformational changes. In B cells these EF hands assist in membrane recruitment, but their role in T cells is not understood. We are utilizing collaborative biophysical, computational, genetic, and biochemical approaches to understand the molecular mechanisms of Rasgrp1 regulation.

Molecular mechanisms of Rasgrp1 regulation

Our computational models predicted that RasGTP levels exponentially increase when a threshold in signals over Sos is surpassed (so that the allosteric feedback mechanism is efficiently engaged). Modeling increasing signals over Rasgrp in the third dimension demonstrates that this threshold moves leftwards so that less signals over SOS are required to induce the jump to high RasGTP. In the extreme condition of high Rasgrp1 signals the computational model predicts spontaneous Ras signals. Hyperactive, oncogenic Ras signals are frequently found in different cancer types. We are currently exploring how deregulation of Rasgrp1 may result in cancer, such as leukemia and colorectal cancer.

Deregulated Rasgrp1 and cancer

We and others uncovered that lymphocytes demonstrate significant levels of tonic or constitutive signals in the resting state. While constitutive BCR signals clearly have a survival function, the importance of constitutive TCR signals is less well understood. We have published how constitutive signals can either repress gene expression or maintain expression levels of other gene sets. We are exploring the biochemical connection between tonic TCR signals and the gene regulation in the nucleus and are investigating mouse models to define the biological roles of these signals.

Tonic signaling to control gene expression in T cells

We collaborate with many different groups and benefit from these collaborations by tapping into computational models, biophysical approaches, structural insights, and patient samples. We combined these collaborative efforts with our own expertise in the biochemistry of Ras signal transduction, cell line approaches, and various mouse models.