**3. Kit**

The mouse dominant spotting mutation at the *W* locus was first described in the early 1900s (Durham, 1908). Mutations at this locus were studied extensively not only because they produced visible coat color changes, but also because mutant mice showed defects in hematopoiesis, mast cell development and gametogenesis (Russell, 1979). However, it was not until 1988 that the gene product at the *W* locus was found to encode for the cellular homolog of the *kit* oncogene which had been molecularly identified a few years earlier (Besmer et al., 1986; Chabot et al., 1988; Geissler et al., 1988).

Kit is a type III RTK that shares structural similarities with platelet-derived growth factor receptors (PDGFRs) and , Flt3 (also known as Flk-2, discussed below) and CSF1R. They

Regulation of Tyrosine Kinase Signaling by Cbl in Hematopoietic Stem Cells 121

Flt3, another member of the type III RTKs, was originally identified by two separate groups through homology screening for TKs (Matthews et al., 1991; Rosnet et al., 1991). Its expression is detected in placenta, gonads, brain and hematopoietic cells, but its role outside of the hematopoietic system is not clear at present. Ligand for Flt3 (Flt3 ligand; FL) was identified a few years later and its transcript is expressed in wide range of both fetal and

Roles of Flt3 in HSCs appear to vary among species and also dependent upon developmental stages. The most primitive self-renewing HSCs with long-term reconstituting potential (LT-HSCs) are not found within Flt3+ LSK cells in adult mouse bone marrow while the same biological activity was detected in both Flt3+ and Flt3- populations in fetal liver (Adolfsson et al., 2001; Christensen & Weissman, 2001). Notably, human HSCs with multi-lineage reconstituting activity are Flt3+ (Sitnicka et al., 2003). Flt3 deficient mice are viable and fertile, but show defects in B lymphocyte progenitors and dendritic cell generation (Mackarehtschian et al., 1995). The role of the FL-Flt3 axis in HSC maintenance and expansion remains controversial. Mackarehtschian et al. originally reported that Flt3 deficient bone marrow cells showed defects in lymphoid and myeloid reconstitution upon transplanting into myeloablated hosts (Mackarehtschian et al., 1995), while a more recent report by Buza-Vidas et al. concluded that Flt3 and FL were dispensable for maintenance and posttransplantation expansion of mouse HSCs (Buza-Vidas et al., 2009). Partly based on the phenotypes of Flt3 and FL deficient mice, models were proposed that Flt3 might function in the lineage restriction process from HSCs to lymphoid progenitors (Luc et al., 2007). However, in human, activating mutations in the *FLT3* gene, either in the form of internal tandem duplication (ITD) mutation in the juxtamembrane domain or point mutations in the kinase domain, are more frequently associated with myeloid malignancies rather than with lymphoid malignancies (Stirewalt & Radich, 2003). Clearly, the roles of Flt3

in the normal and pathological hematopoiesis need to be further delineated.

Flt3 interaction may be mediated through this adaptor protein.

Cbl becomes tyrosine phosphorylated upon Flt3 engagement (Lavagna-Sévenier et al., 1998). It has been shown to physically interact with Flt3, and overexpression of an E3 ligasedefective mutant Cbl inhibited FL-induced Flt3 ubiquitylation and internalization, indicating involvement of Cbl in Flt3 signaling regulation (Sargin et al., 2007). Mice expressing a RF mutant Cbl from its endogenous locus are hypersensitive to FL stimulation (Rathinam et al., 2010), and we confirmed a similar phenotype in mouse bone marrow cells deficient in both Cbl and Cbl-b (Naramura, manuscript in preparation). Furthermore, deletion of FL blocks leukemia development in Cbl RING finger mutant mice (Rathinam et

Nevertheless, the mode of interaction between Cbl and Flt3 has not been clarified. Direct binding between Cbl and Flt3 has not been demonstrated. The sequences surrounding tyrosine 589 partially conform to the canonical Cbl(TKB) recognition sequence, and this region shares a very high homology to the sequences surrounding tyrosine 568 (a candidate Cbl binding site) in Kit. Notably, this is also the region frequently affected by ITD mutations. Alternatively, or in addition to the direct binding, because Flt3 is known to interact with Grb2 (Dosil et al., 1993; Zhang et al., 1999), a Cbl-binding adaptor protein, Cbl-

adult tissues (Lyman et al., 1993; Hannum et al., 1994).

**4. Flt3** 

al., 2010).

are characterized by an extracellular domain with five immunoglobulin-like domains, a single transmembrane domain and an intracellular tyrosine kinase domain that is split into two by an intervening sequence.

HSCs are functionally defined as rare cells with the capacity to self-renew and give rise to all cell types of the hematopoietic lineage, including erythrocytes, granulocytes, monocytes, megakaryocytes and lymphocytes. No single marker specific for HSCs is known today. However, it is widely accepted that, in mice, most HSCs reside in a population of cells that express Kit and another cell surface protein Sca-1 and lack the expression of committed lineage markers (Ikuta & Weissman, 1992). Thus, Kit expression is intimately tied to HSCs.

The ligand for Kit is called stem cell factor (SCF) and encoded by the Steel (*Sl*) locus. The phenotypes of the *Sl* mutant mice are in most cases similar to those of *W* mutant mice, affecting hematopoiesis, mast cell development, fertility and coat colors (Galli et al., 1993). Collectively, these observations firmly established the essential roles of the SCF-Kit axis in these biological processes.

In the mouse embryo, hematopoietic cells are found in the blood islands in the yolk sac starting around embryonic day 7. Subsequently, at day 10-11 of gestation, HSCs migrate to the fetal liver and then to the spleen and the bone marrow, the primary hematopoietic organs in adult. The hematopoietic defect in *W* mice is detected throughout the course of development. Syngeneic transplantation experiments demonstrated that the defect exerted by *W* mutations was intrinsic to hematopoietic cells. The hematopoietic microenvironment in these animals are not affected and able to support hematopoiesis of normal donorderived cells (Russell, 1979).

Kit activity is regulated at various levels. Ligand-receptor engagement of Kit initiates receptor dimerization and subsequent activation of its TK activity. Extensive biochemical studies have mapped intracellular phosphorylated tyrosine residues and their interacting proteins. These include Src family TKs, phosphatases such as SHP1 and SHP2, phospholipase C, p85 subunit of phosphoinositide-3 kinase, (p85(PI3K)) and adaptor proteins such as Grb2 and APS (Lennartsson et al., 2005).

Cbl becomes phosphorylated when Kit-expressing cells are stimulated with SCF (Wisniewski et al., 1996). Earlier studies suggested that Cbl interacted with Kit indirectly through Grb2 (Brizzi et al., 1996), CrkL and p85(PI3K) (Sattler et al., 1997), and APS (Wollberg et al., 2003). More recent data suggest that Cbl binds to Kit directly at tyrosine 568, which is located in the juxtamembrane domain, and tyrosine 936, which is located in the carboxyterminal tail, ubiquitinylate Kit and target them for degradation (Masson et al., 2006). Both Cbl and Cbl-b function similarly towards Kit (Zeng et al., 2005). Hematopoietic cells deficient in Cbl functions are hypersensitive to stimulations through Kit (Naramura et al., 2010; Rathinam et al., 2010). These data all together strongly support that Kit may be one of the physiological targets of Cbl proteins in the HSC compartment.

Structurally, sequences surrounding tyrosine 568 partially conform to the canonical Cbl(TKB) recognition sequence while those around tyrosine 936 do not. Further analyses into the mechanisms of binding between Cbl and Kit may reveal novel molecular interactions that remained unknown so far.
