Denmotoxin, a three-finger toxin from the colubrid snake Boiga dendrophila (Mangrove Catsnake) with bird-specific activity

Boiga dendrophila (mangrove catsnake) is a colubrid snake that lives in Southeast Asian lowland rainforests and mangrove swamps and that preys primarily on birds. We have isolated, purified, and sequenced a novel toxin from its venom, which we named denmotoxin. It is a monomeric polypeptide of 77 amino acid residues with five disulfide bridges. In organ bath experiments, it displayed potent postsynaptic neuromuscular activity and irreversibly inhibited indirectly stimulated twitches in chick biventer cervicis nerve-muscle preparations. In contrast, it induced much smaller and readily reversible inhibition of electrically induced twitches in mouse hemidiaphragm nerve-muscle preparations. More precisely, the chick muscle alpha(1)betagammadelta-nicotinic acetylcholine receptor was 100-fold more susceptible compared with the mouse receptor. These data indicate that denmotoxin has a bird-specific postsynaptic activity. We chemically synthesized denmotoxin, crystallized it, and solved its crystal structure at 1.9 A by the molecular replacement method. The toxin structure adopts a non-conventional three-finger fold with an additional (fifth) disulfide bond in the first loop and seven additional residues at its N terminus, which is blocked by a pyroglutamic acid residue. This is the first crystal structure of a three-finger toxin from colubrid snake venom and the first fully characterized bird-specific toxin. Denmotoxin illustrates the relationship between toxin specificity and the primary prey type that constitutes the snake's diet.     

Organization of the human hepatocyte growth factor-encoding gene.

Human genomic phage libraries were screened for the human hepatocyte growth factor (HGF)-encoding gene (HGF) using a cDNA encoding the human protein as a probe. Characterization of the clones revealed that this gene is composed of 18 exons interrupted by 17 introns spanning approx. 70 kb. The first exon contains the 5'-untranslated region and the signal peptide. The next ten exons encode the alpha-chain which contains four kringle structures. Each kringle domain is encoded by two exons as observed in other kringle-containing proteins. The twelfth exon contains the short spacer region between the alpha- and beta-chains and the remaining six exons comprise the beta-chain. The beta-chain is structurally similar to the catalytic domains of serine proteases; amino acid substitutions in the active site were found. The organization of the HGF gene is highly homologous to those of the serine proteases involved in blood coagulation and fibrinolysis, especially with that of plasminogen. This suggests that the human HGF gene is evolutionally related to these genes.     

Cloning and characterization of the first amphibian bradykinin gene

More than ten bradykinin-related peptides and their cDNAs have been identified from amphibians, but their genes are unknown. In present study, four cDNAs encoding one, two, four and six copies of bradykinin-related peptides were cloned from the frog (Odorrana grahami) skin cDNA library, respectively. Three bradykinin-related peptides (bradykinin, Thr6-bradykinin, Leu5Thr6-bradykinin) were deduced from these four cDNA sequences. Based on the cDNA sequence, the gene sequence encoding an amphibian bradykinin-related peptide from O. grahami was determined. It is composed of 7481 base pairs including two exons and two introns. The first exon codes signal peptide and the second exon codes acidic spacer peptide and Thr6-bradykinin. The promoter region of the bradykinin gene contains several putative recognition sites for nuclear factors, such as SRY, GATA-1, LYF-1, DeltaE, CDXA, NKX-2.5, MIF1 and S8. The current work may facilitate to understand the regulation and possible functions of amphibian skin bradykinin-related peptides.     

Genomic organization of four novel nondisulfide-bridged peptides from scorpion Mesobuthus martensii Karsch: gaining insight into evolutionary mechanism

At least 25 nondisulfide-bridged peptides (NDBPs) have been identified and characterized from scorpions. However, the genomic organization of the genes that encode these peptides have not been reported yet. BmKa1, BmKa2 and BmKb1 are three novel genes that code for NDBPs identified by our group from Mesobuthus martensii Karsch. Based on their cDNA sequences, the genomic DNA sequences encoding these peptides were obtained using the PCR method. Sequence analysis showed that three distinct genomic structural patterns are used to encode these three peptides. The BmKa1 gene is not interrupted by any introns. However, the BmKa2 gene is composed of two exons, interrupted by a 67 bp intron that is located in the DNA region encoding the mature peptide. Two genomic homologues of the BmKb1 cDNA sequence, named BmKb1′ and BmKb2, respectively, were obtained. The BmKb1′ gene contains one intron of 593 bp, inserted into the DNA region that encodes the signal peptide. Similarly, the BmKb2 gene also contains an intron that interrupts the exon that encodes the NDBP signal peptide. The amino acid sequences deduced for BmKb2 and BmKb1′ differ only at one position. The data suggest that the genomic organizational pattern of NDBPs displays more divergence than that exhibited by the genes that encode disulfide-bridged peptides from scorpions.     

Structurally conserved alpha-neurotoxin genes encode functionally diverse proteins in the venom of Naja sputatrix

The structure and organization of the genes encoding the long-chain neurotoxins and four other isoforms of weak neurotoxins in the venom of Naja sputatrix are reported. The genes contained three exons interrupted by two introns, a structure similar to other members of the three-finger toxin family. The proteins encoded by these genes, however, show varied affinity towards nicotinic acetylcholine receptors. Phylogenetic analysis of these genes showed that the weak neurotoxin gene is confined to a distinct group. We also observe that specific mutations of the gene provide the diversity in function in these toxins while maintaining a common structural scaffold. This forms the first report where the molecular basis of evolution of postsynaptic neurotoxins from an ancestral gene can be demonstrated using the same species of snake.     

Structure and evolution of the bovine prothrombin gene

The cloned bovine prothrombin gene has been characterized by partial DNA sequence analysis, including the 5' and 3' flanking sequences and all the intron-exon junctions. The gene is approximately 15.4 x 10(3) base-pairs in length and comprises 14 exons interrupted by 13 introns. The exons coding for the prepro-leader peptide and the gamma-carboxyglutamic acid-containing region are similar in organization to the corresponding exons in the factor IX and protein C genes. This region has probably evolved as a result of recent gene duplication and exon shuffling events. The exons coding for the kringles and the serine protease region of the prothrombin gene are different in organization from the homologous regions in other genes, suggesting that introns have been inserted into these regions after the initial gene duplication events.     

Partial structure of the gene for chicken cartilage proteoglycan core protein

A genomic DNA fragment (gCORE-1), encoding a portion of the cartilage proteoglycan core protein, has been isolated from a phage library using cDNA as a probe. The genomic insert is about 17 kilobase pairs; two BamHI fragments of the insert (1.3 and 4.8 kilobase pairs) contain most of the hybridizable sequences found in the cDNA. Sequence analysis of these fragments shows that they contain a total of five exons that encompass 216 amino acid residues, all of which are identical to those of the corresponding cDNA sequence. Three of the exons, which are adjacent to one another, are very similar to the corresponding exons in the gene of a rat hepatic lectin as well as to an exon in the gene of human pulmonary surfactant-associated protein. There is a strong degree of conservation of amino acid sequences encoded in the three genes, although there is no similarity between their introns. The sizes of the five exons in gCORE-1, except for one (which is indeterminate because only a partial cDNA sequence is available), are less than 184 base pairs, whereas the sizes of the introns range from 218 to greater than 2629 base pairs. Four of the introns interrupt an exon codon at either their donor or acceptor sites, between the first and second nucleotides. Only one intron does not split a codon. Intron and exon boundary sites are in agreement with known consensus sequences for introns. The dispersed distribution and relatively small size of the exons, if representative of the entire gene, suggest that the complete gene which codes for the core protein may be quite sizable.     

Structure of the human hepatic triglyceride lipase gene

The structure of the human hepatic triglyceride lipase gene was determined from multiple cosmid clones. All the exons, exon-intron junctions, and 845 bp of the 5' and 254 bp of the 3' flanking DNA were sequenced. Comparison of the exon sequences to three previously published cDNA sequences revealed differences in the sequence of the codons for residues 133, 193, 202, and 234 that may represent sequence polymorphisms. By primer extension, hepatic lipase mRNA initiates at an adenine 77 bases upstream of the translation initiation site. The hepatic lipase gene spans over 60 kb containing 9 exons and 8 introns, the latter being all located within the region encoding the mature protein. The exons are all of average size (118-234 bp). Exon 1 encodes the signal peptide, exon 4, a region that binds to the lipoprotein substrate, and exon 5, an evolutionarily highly conserved region of potential catalytic function, and exons 6 and 9 encode sequences rich in basic amino acids thought to be important in anchoring the enzyme to the endothelial surface by interacting with acidic domains of the surface glycosaminoglycans. The human lipoprotein lipase gene has been recently reported to have an identical exon-intron organization containing the analogous structural domains [Deeb & Peng (1989) Biochemistry 28, 4131-4135]. Our observations strongly support the common evolutionary origin of these two lipolytic enzymes.     

The genes for the frog skin peptides GLa, xenopsin, levitide and caerulein contain a homologous export exon encoding a signal sequence and part of an amphiphilic peptide

From genomic libraries of Xenopus laevis, parts of the genes coding for the precursors of the skin peptides GLa (peptide with amino-terminal glycine and carboxy-terminal leucinamide), xenopsin and levitide have been isolated and sequenced. The gene for prepropeptide GLa comprises four exons, separated by relatively small introns. The gene for preproxenopsin is composed of five exons, of which all but the last one have been analyzed. This is a large gene encompassing at least 25,000 base pairs. In addition, two exons of the gene for preprolevitide have been isolated. A comparison of these genes reveals the presence of a homologous exon. This exon contains 161 bp, starts one base pair prior to the initiation codon and encodes a signal peptide and part of a pro region with processing sites. In addition, the two genes for preprocaerulein analyzed previously [Vlasak et al. (1987) Eur. J. Biochem. 169, 53-58] also contain a similar exon. This demonstrates the existence of a homologous export exon in genes encoding the precursors of different skin peptides.     

Exons of the human pancreatic polypeptide gene define functional domains of the precursor

Pancreatic polypeptide is a 36-amino acid peptide which inhibits pancreatic exocrine function. We have previously determined from the nucleotide sequence of a cDNA that pancreatic polypeptide is derived from a 95-amino acid precursor, prepropancreatic polypeptide. Pulse-chase studies have suggested that the precursor is cleaved to produce three peptides: pancreatic polypeptide, an icosapeptide, and a smaller peptide. In the present study, we have used the cloned cDNA as a hybridization probe to isolate the pancreatic polypeptide gene from a human bacteriophage genomic library. The nucleotide sequence of 2.8 kilobases of DNA representing the entire human pancreatic polypeptide gene was determined. The gene contains four exons and three introns. Exon 1 encodes the 5'-untranslated region of the mRNA, exon 2 encodes the signal sequence and the sequence of pancreatic polypeptide, exon 3 encodes the icosapeptide, and exon 4 encodes a carboxyl-terminal heptapeptide and the 3'-untranslated region of the mRNA. By Southern blot analysis, the gene detected in a pancreatic polypeptide-producing islet cell tumor was indistinguishable from that in normal human leukocytes. The structure of the human pancreatic polypeptide gene is consistent with the hypothesis that prepropancreatic polypeptide generates three distinct peptides, each encoded by a separate exon. Increased expression of pancreatic polypeptide in the islet cell tumor does not appear to be correlated with major alterations in pancreatic polypeptide gene structure.     

Evolution of collagen IV genes from a 54-base pair exon: A role for introns in gene evolution

The exon structure of the collagen IV gene provides a striking example for collagen evolution and the role of introns in gene evolution. Collagen IV, a major component of basement membranes, differs from the fibrillar collagens in that it contains numerous interruptions in the triple helical Gly-X-Y repeat domain. We have characterized all 47 exons in the mouse alpha 2(IV) collagen gene and find two 36-, two 45-, and one 54-bp exons as well as one 99- and three 108-bp exons encoding the Gly-X-Y repeat sequence. All these exons sizes are also found in the fibrillar collagen genes. Strikingly, of the 24 interruption sequences present in the alpha 2-chain of mouse collagen IV, 11 are encoded at the exon/intron borders of the gene, part of one interruption sequence is encoded by an exon of its own, and the remaining interruptions are encoded within the body of exons. In such "fusion exons" the Gly-X-Y encoding domain is also derived from 36-, 45-, or 54-bp sequence elements. These data support the idea that collagen IV genes evolved from a primordial 54-bp coding unit. We furthermore interpret these data to suggest that the interruption sequences in collagen IV may have evolved from introns, presumably by inactivation of splice site signals, following which intronic sequences could have been recruited into exons. We speculated that this mechanism could provide a role for introns in gene evolution in general.     

Exon structure of a mannose-binding protein gene reflects its evolutionary relationship to the asialoglycoprotein receptor and nonfibrillar collagens

Cloned cDNAs encoding mannose-binding proteins isolated from rat liver have been used to isolate one of the genes encoding this group of proteins. This gene, which encodes the minor form of binding protein (designated MBP-A), has been characterized by sequence analysis. The protein-coding portion of the mRNA for the MBP-A is encoded by four exons separated by three introns. The NH2-terminal, collagen-like portion of the protein is encoded by the first two exons. These exons resemble the exons found in the genes for nonfibrillar collagens in that the intron which divides them is inserted between the first two bases of a glycine codon and the exons do not have the 54- or 108-base pair lengths characteristic of fibrillar collagen genes. The carbohydrate-binding portion of MBP-A is encoded by the remaining two exons. This portion of the protein is homologous to the carbohydrate-recognition domain of the hepatic asialoglycoprotein receptor, which is encoded by four exons. It appears that the three COOH-terminal exons of the asialoglycoprotein receptor gene have been fused into a single exon in the MBP-A gene. The organization of the MBP-A gene is very similar to the arrangement of the gene encoding the highly homologous pulmonary surfactant apoprotein, although one of the intron positions is shifted by a single amino acid. The 3' end of a mannose-binding protein pseudogene has also been characterized.     

Organization of the murine Ig-related lambda 5 gene transcribed selectively in pre-B lymphocytes.

lambda 5 is an immunoglobulin lambda light chain-related gene which is selectively transcribed in murine pre-B lymphocytes to yield a 1.2 kb poly(A)+ mRNA. Comparison of the nucleotide sequence of a 1 kb cDNA clone with the sequence of a genomic clone isolated from 70Z/3 murine pre-B lymphoma cells shows lambda 5 is composed of three exons spanning a 3.75 kb DNA segment. Conserved splice signal sequences at all exon/intron boundaries and the presence of a long open reading frame indicate that a functional mRNA molecule can be made. Exon I contains a cap-site and a potential ATG start codon as well as sequences encoding a signal peptide. This gene could encode a lambda 5 protein of 209 amino acids which has, however, not yet been identified. The 3' portion of exon II and all of exon III shows strong sequence homologies to J lambda L and C lambda L exons. Homology to the lambda L chain genes is lost in the 5' portion of exon II and throughout exon I. In exon I short homologies to leader sequences and to VH framework 1 sequences are seen.     

Isolation and sequence of a rat chymotrypsin B gene

A cDNA clone encoding part of chymotrypsin B was isolated from a cDNA library prepared from rat pancreatic mRNA and used as a probe to isolate the chymotrypsin B gene. The nucleotide sequence of this gene is presented. The 4709-base pair transcribed portion of the isolated gene was inferred from the cDNA and gene sequence, and the 5' border was determined by primer extension on pancreatic polyadenylated RNA. The coding portion of the gene is interrupted by six introns. The active site residues His 57, Asp 102, and Ser 195 are encoded by separate exons. Moreover, two regions of the enzyme which form the substrate-binding pocket are also encoded by separate exons. Thus, the substrate specificity and catalytic activity of the enzyme are produced by joining several exons encoding protein segments that are intrinsically catalytically inactive. The number and location of the intron/exon junctions of the chymotrypsin gene as compared to those of other serine protease genes, as well as the location of the genes on separate chromosomes, suggest that the duplication that resulted in the formation of the chymotrypsin gene was an ancient evolutionary event.