Enhanced signaling by the small guanosine triphosphatase Ras is common in T cell acute lymphoblastic leukemia/lymphoma (T-ALL, but the underlying mechanisms are unclear. Here, we identified the guanine nucleotide exchange factor RasGRP1 (Rasgrp1 in mice) as a Ras activator that contributes to leukemogenesis. We found increased RasGRP1 expression in many pediatric T-ALL patients, which we did not observe in rare early T cell precursor (ETP) T-ALL patients with KRAS and NRAS mutations, such as K-RasG12D. Leukemia screens in wild-type mice, but not in mice expressing the mutant K-RasG12D that encodes a constitutively active Ras, yielded frequent retroviral insertions that led to increased Rasgrp1 expression. Rasgrp1 and oncogenic K-RasG12D promoted T-ALL through distinct mechanisms. In K-RasG12D T-ALLs, we found that enhanced Ras activation did not lead to cell cycle arrest. In mouse T-ALL cells with increased Rasgrp1 expression, we found that Rasgrp1 contributed to a previously uncharacterized cytokine receptor–activated Ras pathway that stimulated the proliferation of T-ALL cells in vivo, which was accompanied by dynamic patterns of activation of effector kinases downstream of Ras in individual T-ALLs. Reduction of Rasgrp1 abundance reduced cytokine-stimulated Ras signaling and decreased the proliferation of T-ALL in vivo, suggesting that patients with this cancer should be screened for increased abundance of RasGRP1 to customize treatment.

The guanine nucleotide exchange factor (GEF) Vav1 is essential for transducing T cell antigen receptor (TCR) signals and therefore plays a critical role in the development and activation of T cells. It has been presumed that the GEF activity of Vav1 is important for its function; however, there has been no direct demonstration of this. Here, we generated mice expressing enzymatically inactive, but normally folded, Vav1 protein. Analysis of these mice showed that the GEF activity of Vav1 was necessary for the selection of thymocytes and for the optimal activation of T cells, including signal transduction to Rac1, Akt, and integrins. In contrast, the GEF activity of Vav1 was not required for TCR-induced calcium flux, activation of extracellular signal-regulated kinase (ERK) and protein kinase D1 (PKD1), and cell polarization. Thus, in T cells, the GEF activity of Vav1 is essential for some, but not all, of its functions.

Summary RasGRP proteins are activators of Ras and other related small GTPases by the virtue of functioning as guanine nucleotide exchange factors (GEFs). In vertebrates, four RasGRP family members have been described. RasGRP-1 through −4 share many structural domains but there are also subtle differences between each of the different family members. Whereas SOS RasGEFs are ubiquitously expressed, RasGRP proteins are expressed in distinct patterns, such as in different cells of the hematopoietic system and in the brain. Most studies have concentrated on the role of RasGRP proteins in the development and function of immune cell types because of the predominant RasGRP expression profiles in these cells and the immune phenotypes of mice deficient for Rasgrp genes. However, more recent studies demonstrate that RasGRPs also play an important role in tumorigenesis. Examples are skin- and hematological-cancers but also solid malignancies such as melanoma or prostate cancer. These novel studies bring up many new and unanswered questions related to the molecular mechanism of RasGRP-driven oncogenesis, such as new receptor systems that RasGRP appears to respond to as well as regulatory mechanism for RasGRP expression that appear to be perturbed in these cancers. Here we will review some of the known aspects of RasGRP biology in lymphocytes and will discuss the exciting new notion that RasGRP Ras exchange factors play a role in oncogenesis downstream of various growth factor receptors.

During T cell activation by antigen-presenting cells (APCs), the diverse spatiotemporal organization of components of T cell signaling pathways modulates the efficiency of activation. Here, we found that loss of the tyrosine kinase interleukin-2 (IL-2)–inducible T cell kinase (Itk) in mice altered the spatiotemporal distributions of 14 of 16 sensors of T cell signaling molecules in the region of the interface between the T cell and the APC, which reduced the segregation of signaling intermediates into distinct spatiotemporal patterns. Activation of the Rho family guanosine triphosphatase Cdc42 at the center of the cell-cell interface was impaired, although the total cellular amount of active Cdc42 remained intact. The defect in Cdc42 localization resulted in impaired actin accumulation at the T cell–APC interface in Itk-deficient T cells. Reconstitution of cells with active Cdc42 that was specifically directed to the center of the interface restored actin accumulation in Itk-deficient T cells. Itk also controlled the central localization of the guanine nucleotide exchange factor SLAT [Switch-associated protein 70 (SWAP-70)–like adaptor of T cells], which may contribute to the activation of Cdc42 at the center of the interface. Together, these data illustrate how control of the spatiotemporal organization of T cell signaling controls critical aspects of T cell function.

Ras GTPases are activated by RasGEFs and inactivated by RasGAPs, which stimulate the hydrolysis of RasGTP to inactive RasGDP. GTPase-impairing somatic mutations in RAS genes, such as KRASG12D, are among the most common oncogenic events in metastatic cancer. A different type of cancer Ras signal, driven by overexpression of the RasGEF RasGRP1 (Ras guanine nucleotide-releasing protein 1), was recently implicated in pediatric T-cell acute lymphoblastic leukemia (T-ALL) patients and murine models, in which RasGRP1 T-ALLs expand in response to treatment with interleukins (ILs) 2, 7 and 9. Here, we demonstrate that IL-2/7/9 stimulation activates Erk and Akt pathways downstream of Ras in RasGRP1 T-ALL but not in normal thymocytes. In normal lymphocytes, RasGRP1 is recruited to the membrane by diacylglycerol (DAG) in a phospholipase C-γ (PLCγ)-dependent manner. Surprisingly, we find that leukemic RasGRP1-triggered Ras-Akt signals do not depend on acute activation of PLCγ to generate DAG but rely on baseline DAG levels instead. In agreement, using three distinct assays that measure different aspects of the RasGTP/GDP cycle, we established that overexpression of RasGRP1 in T-ALLs results in a constitutively high GTP-loading rate of Ras, which is constantly counterbalanced by hydrolysis of RasGTP. KRASG12D T-ALLs do not show constitutive GTP loading of Ras. Thus, we reveal an entirely novel type of leukemogenic Ras signals that is based on a RasGRP1-driven increased in flux through the RasGTP/GDP cycle, which is mechanistically very different from KRASG12D signals. Our studies highlight the dynamic balance between RasGEF and RasGAP in these T-ALLs and put forth a new model in which IL-2/7/9 decrease RasGAP activity.