No abstract

In an attempt to define the molecular basis of the functional diversity of K+ channels, we have isolated overlapping rat brain cDNAs that encode a neuronal delayed rectifier K+ channel, K,4, that is structurally related to the Drosophila Shaw protein. Unlike previously characterized mammalian K+ channel genes, which each contain a single protein-coding exon, K,4 arises from alternative exon usage at a locus that also encodes another mammalian Shaw homolog, NGK2. Thus, the enormous diversity of K+ channels in mammals can be generated not just through gene duplication and divergence but also through alternative splicing of RNA.The electrical excitability of cells of the nervous system is regulated, in part, by voltage-sensitive K+ channels (1). The structure of neuronal K+ channels that have been cloned has been conserved from Drosophila to humans and consists of proteins containing six hydrophobic, putative transmembrane domains, one of which, S4, has been implicated in sensing changes in the membrane potential (2-4). Although they are structurally similar, these K+ channels can be divided into four distinct classes by virtue of their sequence homology to channels encoded at the Drosophila Shaker, Shab, Shal, and Shaw loci (5-11). In Drosophila, diversity of the K+ channels within a given class arises, in part, from differential splicing of RNA encoding the channels (5-11). In contrast, mammalian K+ channel genes that have been described to date have no introns in the coding region (12)(13)(14) and the diversity of these channels appears to have resulted from gene duplication and divergence. We describe here, however, the cloning and expression of a cDNA encoding a mammalian Shaw-type K+ channel, Kv4, which arises from alternative splicing of a transcript from a gene that also encodes NGK2 (15), another mammalian channel related to Shaw. 11 MATERIALS AND METHODS K,4 cDNA Clones. An oligo(dT)-primed cDNA library, constructed from mRNA isolated from the brains of 2-weekold rats, was screened using a probe prepared from clone K41 (12), encoding part of a Shaker-type channel, labeled to a specific activity of >1 x 109 by random hexamer-primed labeling (17). The filters were hybridized at 60'C in 5 x SSPE (lx SSPE = 0.18 M NaCl/10 mM sodium phosphate, pH 7.4/1 mM EDTA), 5x Denhardt's solution, 0.5% SDS, and 1% glycerol with 200 ,ug of salmon sperm DNA per ml and probe at 1 x 106 cpm/ml and were washed to a final stringency of 0.2x SSPE/0.1% SDS at 220C. A cDNA (SA) containing a 2.6-kilobase (kb) insert was isolated and subcloned into pBS+ for sequencing. To obtain the 3' end of the coding region, another oligo(dT)-primed cDNA library, constructed from mRNA isolated from the brains of 14-day-old rats, was screened using an oligonucleotide probe (0123) that included bases 1260-1359 from 5A (Fig. 1). The probe was labeled to a specific activity of >1 x 108 cpm/pmol by filling in two overlapping oligonucleotides with the Klenow fragment of DNA polymerase I and all four [a-32P]dNTPs. Filters were hybridized at 4...

We describe a novel and general way of generating high affinity peptide (HAP) binders to receptor tyrosine kinases (RTKs), using a multi-step process comprising phage-display selection, identification of peptide pairs suitable for hetero-dimerization (non-competitive and synergistic) and chemical synthesis of heterodimers. Using this strategy, we generated HAPs with K(D)s below 1 nM for VEGF receptor-2 (VEGFR-2) and c-Met. VEGFR-2 HAPs bound significantly better (6- to 500-fold) than either of the individual peptides that were used for heterodimer synthesis. Most significantly, HAPs were much better (150- to 800-fold) competitors than monomers of the natural ligand (VEGF) in various competitive binding and functional assays. In addition, we also found the binding of HAPs to be less sensitive to serum than their component peptides. We believe that this method may be applied to any protein for generating high affinity peptide (HAP) binders.

No abstract