John G. Flanagan

The kit ligand: A cell surface molecule altered in steel mutant fibroblasts

The c-kit proto-oncogene, the gene at the mouse W developmental locus, is one of a substantial group of genes that appear to encode cell surface receptors but for which the ligands are unknown. We have characterized the kit ligand by a generally applicable approach: the receptor extracellular domain was genetically fused to placental alkaline phosphatase, producing a soluble receptor affinity reagent with an enzyme tag that could be easily and sensitively traced. This fusion protein, APtag-KIT, was used to demonstrate a specific binding interaction (KD = 3 x 10(-8) M) with a ligand on 3T3 fibroblast lines. In situ staining showed labeling over the whole surface of the 3T3 cells, but not extending to adjacent nonexpressing cells. These findings provide direct molecular evidence that the kit ligand can exist as a cell surface protein. Binding was not detected on 3T3 fibroblasts carrying the steel (Sl) mutation, confirming the biological significance of the binding activity and demonstrating that mutations at the Sl locus affect the expression or structure of the kit ligand.

Complementary gradients in expression and binding of ELF-1 and Mek4 in development of the topographic retinotectal projection map

Topographic maps with a defined spatial ordering of neuronal connections are a key feature of brain organization. Such maps are believed to develop in response to complementary position-specific labels in presynaptic and postsynaptic fields. However, the complementary labeling molecules are not known. In the well-studied visual map of retinal axons projecting to the tectum, the labels are hypothesized to be in gradients, without needing large numbers of cell-specific molecules. We recently cloned ELF-1 as a ligand for Eph family receptors. Here, RNA hybridization shows matching expression gradients for ELF-1 in the tectum and its receptor Mek4 in the retina. Binding activity detected with alkaline phosphatase fusions of ELF-1 and Mek4 also reveals gradients and provides direct evidence for molecular complementarity of gradients in reciprocal fields. ELF-1 and Mek4 may therefore play roles in retinotectal development and have properties predicted of topographic mapping labels.

Transmembrane form of the kit ligand growth factor is determined by alternative splicing and is missing in the SId mutant

The ligand (KL) for the c-kit receptor is a growth factor encoded at the mouse steel (Sl) locus. KL exists in both cell surface and soluble forms, though little is known of the regulation and functional significance of these forms. We show here that tissue-specific alternative splicing gives two types of KL mRNA. Both encode a transmembrane domain, but in transfected cells one produced the soluble form of KL at relatively high levels, whereas the other preferentially gave the cell surface form. Cell surface KL not only stimulated proliferation, but also mediated cell-cell adhesion. The SId allele, which impairs development of hematopoietic cells, melanocytes, and germ cells, has a deletion in the KL gene removing the transmembrane and intracellular domains. Expression of a corresponding cDNA gave a soluble protein that stimulated cellular proliferation but was not associated with the cell surface. These results provide evidence that cell surface KL has a critical role in the intact organism.

Heparin is required for cell-free binding of basic fibroblast growth factor to a soluble receptor and for mitogenesis in whole cells.

Heparin is required for the binding of basic fibroblast growth factor (bFGF) to high-affinity receptors on cells deficient in cell surface heparan sulfate proteoglycan. So that this heparin requirement could be evaluated in the absence of other cell surface molecules, we designed a simple assay based on a genetically engineered soluble form of murine FGF receptor 1 (mFR1) tagged with placental alkaline phosphatase. Using this assay, we showed that FGF-receptor binding has an absolute requirement for heparin. By using a cytokine-dependent lymphoid cell line engineered to express mFR1, we also showed that FGF-induced mitogenic activity is heparin dependent. Furthermore, we tested a series of small heparin oligosaccharides of defined lengths for their abilities to support bFGF-receptor binding and biologic activity. We found that a heparin oligosaccharide with as few as eight sugar residues is sufficient to support these activities. We also demonstrated that heparin facilitates FGF dimerization, a property that may be important for receptor activation.

Genetic Analysis of Ephrin-A2 and Ephrin-A5 Shows Their Requirement in Multiple Aspects of Retinocollicular Mapping

Ephrin-A2 and -A5 are thought to be anteroposterior mapping labels for the retinotectal/retinocollicular projection. Here, gene disruptions of both these ephrins are characterized. Focal retinal labeling reveals moderate map abnormalities when either gene is disrupted. Double heterozygotes also have a phenotype, showing an influence of absolute levels. In vitro assays indicate these ephrins are required for repellent activity in the target and also normal responsiveness in the retina. In double homozygotes, anteroposterior order is almost though not completely lost. Temporal or nasal retinal labelings reveal quantitatively similar but opposite shifts, with multiple terminations scattered widely over the target. These results indicate an axon competition mechanism for mapping, with a critical role for ephrins as anteroposterior topographic labels. Dorsoventral topography is also impaired, showing these ephrins are required in mapping both axes.

PTPσ Is a Receptor for Chondroitin Sulfate Proteoglycan, an Inhibitor of Neural Regeneration

Toward Neuronal Regeneration Neurons in the central nervous system that are severed or crushed do not regenerate well. Part of the problem derives from the glial scars left behind after such damage. The scar tissue contains sulfated proteoglycans that seem to inhibit axon regeneration. Shen et al. (p. 592, published online 15 October) have now identified a protein tyrosine phosphatase (PTP) in mouse neuronal membranes that functions as a receptor for the proteoglycans. Neurons that lacked this particular PTP showed improved regeneration. Regeneration remained incomplete, presumably due to other inhibitory factors in the way of complete axon regeneration. Mouse neurons that lack a receptor for inhibitory proteoglycans show improved regeneration. Chondroitin sulfate proteoglycans (CSPGs) present a barrier to axon regeneration. However, no specific receptor for the inhibitory effect of CSPGs has been identified. We showed that a transmembrane protein tyrosine phosphatase, PTPσ, binds with high affinity to neural CSPGs. Binding involves the chondroitin sulfate chains and a specific site on the first immunoglobulin-like domain of PTPσ. In culture, PTPσ–/– neurons show reduced inhibition by CSPG. A PTPσ fusion protein probe can detect cognate ligands that are up-regulated specifically at neural lesion sites. After spinal cord injury, PTPσ gene disruption enhanced the ability of axons to penetrate regions containing CSPG. These results indicate that PTPσ can act as a receptor for CSPGs and may provide new therapeutic approaches to neural regeneration.

Topographically Specific Effects of ELF-1 on Retinal Axon Guidance In Vitro and Retinal Axon Mapping In Vivo

Topographic maps, which maintain the spatial order of neurons in the order of their axonal connections, are found throughout the nervous system. In the visual retinotectal projection, ELF-1, a ligand in the tectum, and its receptors in the retina show complementary gradients in expression and binding, indicating they may be positional labels for map development. Here we show that ELF-1 acts as a repellent axon guidance factor in vitro. In vivo, when the tectal ELF-1 pattern is modified by retroviral overexpression, retinal axons avoid ectopic ELF-1 patches and map to abnormally anterior positions. All these effects were seen on axons from temporal but not nasal retina, indicating that ELF-1 could determine nasal versus temporal retinotectal specificity, and providing a direct demonstration of a cell recognition molecule with topographically specific effects on neural map development.

Ephrin-B Reverse Signaling Is Mediated by a Novel PDZ-RGS Protein and Selectively Inhibits G Protein–Coupled Chemoattraction

Transmembrane B ephrins and their Eph receptors signal bidirectionally. However, neither the cell biological effects nor signal transduction mechanisms of the reverse signal are well understood. We describe a cytoplasmic protein, PDZ-RGS3, which binds B ephrins through a PDZ domain, and has a regulator of heterotrimeric G protein signaling (RGS) domain. PDZ-RGS3 can mediate signaling from the ephrin-B cytoplasmic tail. SDF-1, a chemokine with a G protein-coupled receptor, or BDNF, act as chemoattractants for cerebellar granule cells, with SDF-1 action being selectively inhibited by soluble EphB receptor. This study reveals a pathway that links reverse signaling to cellular guidance, uncovers a novel mode of control for G proteins, and demonstrates a mechanism for selective regulation of responsiveness to neuronal guidance cues.

Semaphorin 5A Is a Bifunctional Axon Guidance Cue Regulated by Heparan and Chondroitin Sulfate Proteoglycans

The response of neuronal growth cones to axon guidance cues depends on the developmental context in which these cues are encountered. We show here that the transmembrane protein semaphorin 5A (Sema5A) is a bifunctional guidance cue exerting both attractive and inhibitory effects on developing axons of the fasciculus retroflexus, a diencephalon fiber tract associated with limbic function. The thrombospondin repeats of Sema5A physically interact with the glycosaminoglycan portion of both chondroitin sulfate proteoglycans (CSPGs) and heparan sulfate proteoglycans (HSPGs). CSPGs function as precisely localized extrinsic cues that convert Sema5A from an attractive to an inhibitory guidance cue. Therefore, glycosaminoglycan bound guidance cues provide a molecular mechanism for CSPG-mediated inhibition of axonal extension. Further, axonal HSPGs are required for Sema5A-mediated attraction, suggesting that HSPGs are components of functional Sema5A receptors. Thus, neuronal responses to Sema5A are proteoglycan dependent and interpreted according to the biological context in which this membrane bound guidance cue is presented.

Heterodimers of Placenta Growth Factor/Vascular Endothelial Growth Factor ENDOTHELIAL ACTIVITY, TUMOR CELL EXPRESSION, AND HIGH AFFINITY BINDING TO Flk-1/KDR

Here we show that the Escherichia coli expressed monomers of placenta growth factor (PLGF) and vascular endothelial growth factor (VEGF) can be re-folded in vitro to form PLGF/VEGF heterodimers. The purified recombinant PLGF/VEGF heterodimers and VEGF homodimers have potent mitogenic and chemotactic effects on endothelial cells. However, PLGF/VEGF heterodimers display 20-50-fold less mitogenic activity than VEGF homodimers. In contrast, PLGF homodimers have little or no effect in these in vitro assays. We also demonstrate the presence of natural PLGF/VEGF heterodimers in the conditioned media of various human tumor cell lines. While PLGF/VEGF heterodimers bind with high affinity to a soluble Flk-1/KDR receptor, PLGF homodimers fail to bind to this receptor. Cross-linking of I-ligands to human umbilical vein endothelial cells reveals that PLGF/VEGF heterodimers and VEGF homodimers, but not PLGF homodimers, form complexes with membrane receptors. VEGF homodimers and PLGF/VEGF heterodimers stimulate tyrosine phosphorylation of a 220-kDa protein, the expected size for the KDR receptor in human umbilical vein endothelial cells, whereas PLGF homodimers are unable to induce tyrosine phosphorylation of this protein. These data indicate that PLGF may modulate VEGF-induced angiogenesis by the formation of PLGF/VEGF heterodimers in cells producing both factors.