First printed in R&D Systems' 1997 Catalog.
fms-like tyrosine kinase-3 (Flt-3 or STK-1) is a newly discovered member of the third class of receptor-type tyrosine kinases (RTKs).1, 2 To date, there are at least seven RTK classes. Summarizing relevant classes, molecules in class III possess five immunoglobulin-like (Ig-like) extracellular domains, while class IV molecules express three Ig-like extracellular domains, and class V molecules possess seven Ig-like extracellular domains.2 Within these three classes (III, IV, & V), four Flt (based on fms homology) and two Flk (fetal liver kinase) gene products have been identified, although the molecules identified as Flt-3 and Flk-2 are the human and mouse homologues of the same factor.
Figure 1. Flt-3 & bone marrow B cell lymphopoiesis |
The Flt designation (Flt-1) was originally applied to a transmembrane glycoprotein that contained both a cytoplasmic tyrosine kinase domain and an extracellular cysteine organization reminiscent of the fms family of receptors.3 Functionally, Flt-1 is known to bind both VEGF and Placental Growth Factor, and has been renamed VEGF receptor 1 (VEGFR1). The designation Flt-2 is no longer used as this receptor has been shown to bind FGF acidic, FGF basic and FGF-4, and has subsequently been renamed FGFR1.4, 5 Flt-3 is unique and binds to its own ligand, Flt-3L or FL. Flt-4 has recently been shown to bind to the newly discovered VEGF-C, and has thus been renamed VEGFR3.5-10
Two mouse Flk receptors have been isolated from mouse fetal liver hematopoietic stem cells.11, 12 Studies have shown the first Flk molecule to be a second receptor for VEGF and this is now known as VEGFR2. The second receptor, Flk-2, is the mouse equivalent of human Flt-3, and now shares the Flt-3/Flk-2 designation.12-14 For recent reviews of the Flt-3 receptor and its ligand see references 15-18.
Human Flt-3 is a 160 kDa, type I transmembrane glycoprotein originally cloned from a CD34+ stem cell library. It is 993 amino acid (aa) residues in length, and contains a 25 aa residue signal sequence, a 516 aa residue extracellular segment, a 21 aa residue transmembrane span, and a 431 aa residue cytoplasmic domain.14, 19 Within the cytoplasmic domain, there is a 75 aa residue interruption of the tyrosine kinase elements that constitute the receptor's intrinsic tyrosine kinase activity.14, 19, 20 Human Flt-3 shows 85% aa sequence identity overall to mouse Flt-3, although the mouse molecule is seven amino acids longer.19, 21 The predicted molecular weight of Flt-3 is approximately 110 kDa, indicating considerable glycosylation.15, 19 Flt-3 shares its class III RTK designation with stem cell factor receptor (KIT),22 M-CSF receptor (c-fms/FMS)23, 24 and the type A and type B receptors for PDGF (PDGFRA and PDGFRB).25, 26
Within the Flt-3/KIT/FMS group, these three receptors are all believed to play a role in hematopoietic progenitor cell development. Overall, KIT, FMS and Flt-3 receptors show only 9% identity in their extracellular segments, but 57% identity in their kinase domains.14 Although the receptor binding site(s) for Flt-3-ligand are unknown, the first three N-terminal Ig-like domains on KIT and FMS have been identified as crucial for cytokine binding, suggesting an equivalent situation for Flt-3.27, 28 If Flt-3 is analogous to KIT in terms of ligand-receptor interactions, the fourth Ig-like Flt-3 domain will be found to be crucial in Flt-3 dimerization and signal transduction.29 This possibility is supported by the recent observation that a variant isoform of mouse Flt-3, devoid of the fifth Ig-like domain, is unaffected with respect to Flt-3-ligand binding and tyrosine kinase activation.30
Flt-3 is currently found on CD34+ fetal liver and bone marrow stem cells,31 pre-B cells,14, 32 pro-B cells,32 immature thymocytes,3 monocytes,32 and AML (acute myeloid leukemia) plus B cell precursor ALL (acute lymphoblastic leukemia) blast cells.18, 33-35 Although Flt-3 was cloned from a mouse placenta cDNA library, it is unclear which placental cells are responsible for its expression.21 Cell lines known to express Flt-3 include 697 and ALL-1 pre-B ALL cells, EOL-1 and EOL-3 AML-BC cells, and MOLM-1 CML cells.35 While Flt-3 is commonly associated with CD34+ stem cells,19, 36 it is not clear if all CD34+ cells express Flt-3. For instance, it has been suggested that in the CD34+ stem cell population Flt-3 expression is concentrated on CD117(c-kit)+ stem cells rather than CD117low stem cells.34
Flt-3 ligand (FL/Flt-3L) has been cloned and found to be a type I transmembrane glycoprotein of approximately 30 kDa.7, 37 Although it is known to exist naturally as a non-disulfide-linked homodimer, multiple forms of the ligand are known, including a soluble form.7, 38 Generally, the extracellular portion is approximately 160 aa residues in length and the cytoplasmic segment is approximately 20-30 aa residues in length.7, 37, 38 Mouse and human FL are about 70% identical at the amino acid level, and both show cross-species reactivity.7, 37-39 The Kd for FL binding to Flt-3 is reported to be approximately 0.7 nM.10 In their extracellular domains, Flt-3 ligand, stem cell factor (SCF), and FMS-ligand (M-CSF) show a four-helix bundle structure and demonstrate four conserved cysteines.37, 38 Cells known to express FL include human and mouse T cell lines37, 38 a thymic stromal cell line,7 and bone marrow fibroblasts.40
Functionally, hematopoietic growth factors have been described as belonging to one of three groups. The first or multilineage group, consisting of IL-3 and GM-CSF, acts on early CFUs (CFU-GEMM, CFU-GM) and BFUs (BFU-E, BFU-MK). The second or unilineage group, consisting of Epo, G-CSF, IL-5, M-CSF and Tpo, acts on later hematopoietic progenitors (i.e., CFU-E, CFU-Mk, and CFU-Eo). The third or "potentiating" group, consisting of IL-6, IL-11, LIF, FGF basic, SCF and FL, acts principally to potentiate the activities of other hematopoietic factors.41 Within the third group, SCF and FL both show marked activity on hematopoietic stem cells and thus could be considered special circumstance/stem cell growth factors. Some of the myelopoietic potentiating effects attributed to FL include: 1) an expansion of CD34+ CD38- cell number in conjunction with SCF and IL-3;42 2) an increase in HPP-CFC and CFU-GM number when incubated with IL-3, IL-6 and Epo;43 and 3) in the presence of GM-CSF, the formation of CFU-GM from CD34-Lin- cells.44 Individual and direct myelopoietic effects of FL include an increase in CFU-GM, CFU-GEMM and HPP-CFC survival43, 45 and a preferential induction of macrophages in differentiating CD34+ cultures.46 FL alone has minimal or no effects on erythroid and megakaryocyte progenitors.44, 46, 47
The FL/Flt-3 system also plays an important role in lymphopoiesis. In particular, FL has been shown in vitro to synergize with IL-7 in inducing the proliferation of both B cell progenitors (B220low CD43+CD24-)48 and later stage pro-B cells (B220+CD43+CD24+).49 This activity has been confirmed in mice made deficient for Flt-3. In these mice hematopoietic populations are essentially normal but marked deficiencies of early B cell progenitors are found in the bone marrow. Pro-B (or B220+CD43+) cells occur in numbers less than one-half those of controls, while pre-B (or B220+CD43-) cells show numbers less than 75% of normal controls.50 This has led to the suggestion that FL, perhaps expressed constitutively by bone marrow fibroblasts, is a normal regulator of B cell lymphopoiesis, while cytokines produced by activated lymphocytes (such as IL-3) synergize with FL in times of stress to accelerate B cell development.51
Finally, at least two cytokines have now been identified that downmodulate FL activity. TNF-alpha and TGF-beta have both been shown to block FL-induced or potentiated hematopoietic activity, with TGF-beta in particular demonstrating an ability to decrease Flt-3 protein levels and reverse the FL-induced decrease in the time that hematopoietic progenitors spend in the G1 phase of the cell cycle.52-54