Nucleotide sequence and organization of the human S-protein gene: repeating peptide motifs in the "pexin" family and a model for their evolution

The S-protein/vitronectin gene was isolated from a human genomic DNA library, and its sequence of about 5.3 kilobases including the adjacent 5' and 3' flanking regions was established. Alignment of the genomic DNA nucleotide sequence and the cDNA sequence indicated that the gene consisted of eight exons and seven introns. The intron positions in the S-protein gene and their phase type were compared to those in the hemopexin gene which shares amino acid sequence homologies with transin and the S-protein. Three introns have been found at equivalent positions; two other introns are very close to these positions and are interpreted as cases of intron sliding. Introns 3-7 occur at a conserved glycine residue within repeating peptide segments, whereas introns 1 and 2 are at the boundaries of the Somatomedin B domain of S-protein. The analysis of the exon structure in relation to repeating peptide motifs within the S-protein strongly suggests that it contains only seven repeats, one less than the hemopexin molecule. A very similar repeat pattern like that in hemopexin is shown to be present also in two other related proteins, transin and interstitial collagenase. An evolutionary model for the generation of the repeat pattern in the S-protein and the other members of this novel "pexin" gene family is proposed, and the sequence modifications for some of the repeats during divergent evolution are discussed in relation to known unique functional properties of hemopexin and S-protein.     

A rapidly evolving region in the immunoglobulin heavy chain loci of rat and mouse: postulated role of (dC-dA)n.(dG-dT)n sequences

The nucleotide sequences of the introns that are located between the C4 exon and the first membrane exon of mouse and rat immunoglobulin epsilon-chain genes have been determined. The rat intron sequence was found to contain four separate clusters of repetitive sequences all of which consisted of (dC-dA)n.(dG-dT)n dinucleotide repeats. A comparison between this chromosomal region in mouse and rat revealed four deletions or duplications, three of which have occurred inside or at the borders of the CA clusters. Rearrangements have occurred inside or at the borders of all four repeats after the evolutionary separation of mouse and rat. The sequence comparison reveals in addition a duplication, connected to the CA repeats, which has occurred early in evolution, before the evolutionary divergence of mouse and rat. These findings suggest that (dC-dA)n.(dG-dT)n sequences are potential targets for recombination events.     

Nucleotide sequence of the gene for the b subunit of human factor XIII

Factor XIII (Mr 320,000) is a blood coagulation factor that stabilizes and strengthens the fibrin clot. It circulates in blood as a tetramer composed of two a subunits (Mr 75,000 each) and two b subunits (Mr 80,000 each). The b subunit consists of 641 amino acids and includes 10 tandem repeats of 60 amino acids known as GP-I structures, short consensus repeats (SCR), or sushi domains. In the present study, the human gene for the b subunit has been isolated from three different genomic libraries prepared in lambda phage. Fifteen independent phage with inserts coding for the entire gene were isolated and characterized by restriction mapping, Southern blotting, and DNA sequencing. The gene was found to be 28 kilobases in length and consisted of 12 exons (I-XII) separated by 11 intervening sequences. The leader sequence was encoded by exon I, while the carbonyl-terminal region of the protein was encoded by exon XII. Exons II-XI each coded for a single sushi domain, suggesting that the gene evolved through exon shuffling and duplication. The 12 exons in the gene ranged in size from 64 to 222 base pairs, while the introns ranged in size from 87 to 9970 nucleotides and made up 92% of the gene. The introns contained four Alu repetitive sequences, one each in introns A, E, I, and J. A fifth Alu repeat was present in the flanking 3' end of the gene. Two partial KpnI repeats were also found in the introns, including one in intron I and one in intron J. The KpnI repeat in intron J was 89% homologous to a sequence of approximately 2200 nucleotides flanking the gene coding for human beta globin and approximately 3800 nucleotides from the L1 insertion present in the gene for human factor VIII. Intron H also contained an "O" family repeat, while two potential regions for Z-DNA were identified within introns G and J. One nucleotide change was found in the coding region of the gene when its sequence was compared to that of the cDNA. This difference, however, did not result in a change in the amino acid sequence of the protein.     

Genomic DNA sequence of the cystic fibrosis transmembrane conductance regulator (CFTR) gene.

The gene responsible for cystic fibrosis, the most common severe autosomal recessive disorder, is located on the long arm of human chromosome 7, region q31-q32. The gene has recently been identified and shown to be approximately 250 kb in size. To understand the structure and to provide the basis for a systematic analysis of the disease-causing mutations in the gene, genomic DNA clones spanning different regions of the previously reported cDNA were isolated and used to determine the coding regions and sequences of intron/exon boundaries. A total of 22,708 bp of sequence, accounting for approximately 10% of the entire gene, was obtained. Alignment of the genomic DNA sequence with the cDNA sequence showed perfect colinearity between the two and a total of 27 exons, each flanked by consensus splice signals. A number of repetitive elements, including the Alu and Kpn families and simple repeats, such as (GT)17, (GATT)7, and (TA)14, were detected in close vicinity of some of the intron/exon boundaries. At least three of the simple repeats were found to be polymorphic in the population. Although an internal amino acid sequence homology could be detected between the two halves of the predicted polypeptide, especially in the regions of the two putative nucleotide-binding folds (NBF1 and NBF2), the lack of alignment of the nucleotide sequence as well as the different positions of the exon/intron boundaries does not seem to support the hypothesis of a recent gene duplication event. To facilitate detection of mutations by direct sequence analysis of genomic DNA, 28 sets of oligonucleotide primers were designed and tested for their ability to amplify individual exons and the immediately flanking sequences in the introns.     

Duplication of seven exons in LDL receptor gene caused by Alu-Alu recombination in a subject with familial hypercholesterolemia

A defective LDL receptor gene in a child with familial hypercholesterolemia produces a receptor precursor that is 50,000 daltons larger than normal (apparent Mr 170,000 vs. 120,000). The elongated protein resulted from a 14 kilobase duplication that encompasses exons 2 through 8. The duplication arose from an unequal crossing-over between homologous repetitive elements (Alu sequences) in intron 1 and intron 8. The mutant receptor has 18 contiguous cysteine-rich repeat sequences instead of the normal nine. Seven of these duplicated repeats are derived from the ligand-binding domain, and two repeats are part of the epidermal growth factor precursor homology region. The elongated receptor undergoes normal carbohydrate processing, its apparent molecular weight increases to 210,000, and the receptor reaches the cell surface where it binds reduced amounts of LDL but undergoes efficient internalization and recycling. The current findings support an evolutionary model in which homologous recombination between repetitive elements in introns leads to exon duplication during evolution of proteins.     

Comparative analysis of the structural organization of two closely related vitellogenin genes in X. laevis

The structural organization of the two closely related vitellogenin genes A1 and A2 has been determined and compared by electron microscopy. In both genes the mRNA-coding sequence of 6 kb is interrupted 33 times, leading to a total gene length of 21 kb for gene A1 and 16 kb for gene A2. Thus both genes have a mean exon length of 0.175 kb, while the mean intron length is 0.45 kb in gene A1 and 0.31 kb in gene A2. Because the introns interrupt the structural sequence at homologous positions in genes A1 and A2, we suggest that these two genes are the products of a duplication of an ancestral gene which had an intron-exon arrangement similar to that of the extant genes. Since the duplication event, the sequence and length of the analogous introns have changed rapidly, whereas homologous exons have diverged to an extent of only 5% of their sequences. The results suggest different mechanisms of evolution for exons and introns. While the exons evolved primarily by point mutations, such mutations, as well as deletion, insertion and duplication events, were important in the evolution of the introns.     

The role of introns in repeat protein gene formation

Genes composed of tandem repetitive sequence motifs are abundant in nature and are enriched in eukaryotes. To investigate repeat protein gene formation mechanisms, we have conducted a large-scale analysis of their introns and exons. We find that a wide variety of repeat motifs exhibit a striking conservation of intron position and phase, and are composed of exons that encode one or two complete repeats. These results suggest a simple model of repeat protein gene formation from local duplications. This model is corroborated by amino acid sequence similarity patterns among neighboring repeats from various repeat protein genes. The distribution of one- and two-repeat exons indicates that intron-facilitated repeat motif duplication, in which the start and end points of duplication are located in consecutive intronic regions, significantly exceeds intron-independent duplication. These results suggest that introns have contributed to the greater abundance of repeat protein genes in eukaryotic versus prokaryotic organisms, a conclusion that is supported by taxonomic analysis.     

Genomic organization of the human multidrug resistance (MDR1) gene and origin of P-glycoproteins

The MDR1 gene, responsible for multidrug resistance in human cells, encodes a broad specificity efflux pump (P-glycoprotein). P-glycoprotein consists of two similar halves, each half including a hydrophobic transmembrane region and a nucleotide-binding domain. On the basis of sequence homology between the N-terminal and C-terminal halves of P-glycoprotein, we have previously suggested that this gene arose by duplication of a primordial gene. We have now determined the complete intron/exon structure of the MDR1 gene by direct sequencing of cosmid clones and enzymatic amplification of genomic DNA segments. The MDR1 gene includes 28 introns, 26 of which interrupt the protein-coding sequence. Although both halves of the protein-coding sequence are composed of approximately the same number of exons, only two intron pairs, both within the nucleotide-binding domains, are located at conserved positions in the two halves of the protein. The other introns occur at different locations in the two halves of the protein and in most cases interrupt the coding sequence at different positions relative to the open reading frame. These results suggest that the P-glycoprotein arose by fusion of genes for two related but independently evolved proteins rather than by internal duplication.     

Evolution of EF-hand calcium-modulated proteins. IV. Exon shuffling did not determine the domain compositions of EF-hand Proteins

Summary In the previous three reports in this series we demonstrated that the EF-hand family of proteins evolved by a complex pattern of gene duplication, transposition, and splicing. The dendrograms based on exon sequences are nearly identical to those based on protein sequences for troponin C, the essential light chain myosin, the regulatory light chain, and calpain. This validates both the computational methods and the dendrograms for these subfamilies. The proposal of congruence for calmodulin, troponin, C, essential light chain, and regulatory light chain was confirmed. There are, however, significant differences in the calmodulin dendrograms computed from DNA and from protein sequences. In this study we find that introns are distributed throughout the EF-hand domain and the interdomain regions. Further, dendrograms based on intron type and distribution bear little resemblance to those based on protein or on DNA sequences. We conclude that introns are inserted, and probably deleted, with relatively high frequency. Further, in the EF-hand family exons do not correspond to structural domains and exon shuffling played little if any role in the evolution of this widely distributed homolog family. Calmodulin has had a turbulent evolution. Its dendrograms based on protein sequence, exon sequence, 3′-tail sequence, intron sequences, and intron positions all show significant differences.     

A new repetitive sequence uniquely present in the decay-accelerating factor genes

 The decay-accelerating factor (DAF, CD55) protects cells from autologous complement attack on self cell membranes. We have previously reported that the seventh exon encoding the serine/threonine-rich(S/T)-abc region of the guinea pig DAF gene is composed of five homologous repeats of about 51 base pairs, and that differential usage of these repeats produces the various lengths observed in the S/T region of guinea pig DAF. In this study, we found that the seventh intron of the guinea pig DAF gene was wholly composed of 18 tandem repeats homologous to the repeating unit of the S/T-abc exon. This type of repetitive structure, although the number of repeats was variable, was also found in the corresponding exons and introns of all DAF genes of other species so far tested including human and seven other primates and mouse, in which alternative splicing in this region has not been found. This suggested that generation of the repetitive sequences spanning the exon and intron regions had occurred before the diversification of these species. In addition, all the intron sequences of the tested DAF genes had no stop codon when they were presumably translated in the same reading frame as the seventh and eighth exons, except for that of one of two duplicated mouse DAF genes. These findings and significant interspecies identities of the intron sequence suggest that the intron sequence conceivably could be translated in some tissues and/or in some stages of development although to date we have not yet succeeded in detecting mRNA for this region. Received: 7 July 1997 / Revised: 11 August 1997     

Nucleolar introns from Physarum flavicomum contain insertion elements that may explain how mobile group I introns gained their open reading frames.

Comparison of two group I intron sequences in the nucleolar genome of the myxomycete Physarum flavicomum to their homologs in the closely related Physarum polycephalum revealed insertion-like elements. One of the insertion-like elements consists of two repetitive sequence motifs of 11 and 101 bp in five and three copies, respectively. The smaller motif, which flanks the larger, resembles a target duplication and indicates a relationship to transposons or retroelements. The insertion-like elements are found in the peripheral loops of the RNA structure; the positions occupied by the ORFs of mobile nucleolar group I introns. The P. flavicomum introns are 1184 and 637 bp in size, located in the large subunit ribosomal RNA gene, and can be folded into group I intron structures at the RNA level. However, the intron 2s from both P. flavicomum and P. polycephalum contain an unusual core region that lacks the P8 segment. None of the introns are able to self-splice in vitro. Southern analysis of different isolates indicates that the introns are not optional in myxomycetes.     

Conservation of the positions of metazoan introns from sponges to humans

Sponges (phylum Porifera) are the phylogenetic oldest Metazoa still extant. They can be considered as reference animals (Urmetazoa) for the understanding of the evolutionary processes resulting in the creation of Metazoa in general and also for the metazoan gene organization in particular. In the marine sponge Suberites domuncula, genes encoding p38 and JNK kinases contain nine and twelve introns, respectively. Eight introns in both genes share the same positions and the identical phases. One p38 intron slipped for six bases and the JNK gene has three more introns. However, the sequences of the introns are not conserved and the introns in JNK gene are generally much longer. Introns interrupt most of the conserved kinase subdomains I-XI and are found in all three phases (0, 1 and 2). We analyzed in details p38 and JNK genes from human, Caenorhabditis elegans and Drosophila melanogaster and found in most genes introns at the positions identical to those in sponge genes. The exceptions are two p38 genes from D. melanogaster that have lost all introns in the coding sequence. The positions of 11 introns in each of four human p38 genes are fully conserved and ten introns occupy identical positions as the introns in sponge p38 or JNK genes. The same is true for nine, out of ten introns in the human JNK-1 gene. The introns in human p38 and JNK genes are on average more than ten times longer than corresponding introns in sponges. It was proposed that yeast HOG1-like kinases (from i.e. Saccharomyces cerevisiae and Emericella nidulans) and metazoan p38 and JNK kinases are orthologues. p38 and JNK genes were created after the split from fungi by the duplication and diversification of the HOG1-like progenitor gene. Our results further support the common origin of p38 and JNK genes and speak in favor of a very early time of duplication. The ancestral gene contained at least ten introns, which are still present at the very conserved positions in p38 and JNK genes of extant animals. Four of these introns are present at the same positions in the HOG-like gene in the fungus E. nidulans. The others probably entered the ancestral gene after the split of fungi, but before the duplication of the gene and before the creation of the common, urmetazoan progenitor of all multicellular animals. A second gene coding for an immune molecule is described, the allograft inflammatory factor, which likewise showed a highly conserved exon/intron structure in S. domuncula and in human. These data show that the intron/exon borders are highly conserved in genes from sponges to human.     

Molecular structure and genetic stability of human hypoxanthine phosphoribosyltransferase (HPRT) gene duplications.

We have determined the genetic stability of three independent intragenic human HPRT gene duplications and the structure of each duplication at the nucleotide sequence level. Two of the duplications were isolated as spontaneous mutations from the HL60 human myeloid leukemia cell line, while the third was originally identified in a Lesch-Nyhan patient. All three duplications are genetically unstable and have a reversion rate approximately 100-fold higher than the rate of duplication formation. The molecular structures of these duplications are similar, with direct duplication of HPRT exons 2 and 3 and of 6.8 kb (HL60 duplications) or 13.7 kb (Lesch-Nyhan duplication) of surrounding HPRT sequence. Nucleotide sequence analyses of duplication junctions revealed that the HL60-derived duplications were generated by unequal homologous recombination between clusters of Alu repeats contained in HPRT introns 1 and 3, while the Lesch-Nyhan duplication was generated by the nonhomologous insertion of duplicated HPRT DNA into HPRT intron 1. These results suggest that duplication substrates of different lengths can be generated from the human HPRT exon 2-3 region and can undergo either homologous or nonhomologous recombination with the HPRT locus to form gene duplications.     

Structure, polymorphism, and novel repeated DNA elements revealed by a complete sequence of the human alpha-fetoprotein gene

The human alpha-fetoprotein gene spans 19,489 base pairs from the putative "Cap" site to the polyadenylation site. It is composed of 15 exons separated by 14 introns, which are symmetrically placed within the three domains of alpha-fetoprotein. In the 5' region, a putative TATAAA box is at position -21, and a variant sequence, CCAAC, of the common CAT box is at -65. Enhancer core sequences GTGGTTTAAAG are found in introns 3 and 4, and several copies of glucocorticoid response sequences AGATACAGTA are found on the template strand of the gene. There are six polymorphic sites within 4690 base pairs of contiguous DNA derived from two allelic alpha-fetoprotein genes. This amounts to a measured polymorphic frequency of 0.13%, or 6.4 X 10(-4)/site, which is about 5-10 times lower than values estimated from studies on polymorphic restriction sites in other regions of the human genome. There are four types of repetitive sequence elements in the introns and flanking regions of the human alpha-fetoprotein gene. At least one of these is apparently a novel structure (designated Xba) and is found as a pair of direct repeats, with one copy in intron 7 and the other in intron 8. It is conceivable that within the last 2 million years the copy in intron 8 gave rise to the repeat in intron 7. Their present location on both sides of exon 8 gives these sequences a potential for disrupting the functional integrity of the gene in the event of an unequal crossover between them. There are three Alu elements, one of which is in intron 4; the others are located in the 3' flanking region. A solitary Kpn repeat is found in intron 3. The Xba and Kpn repeats were only detected by complete sequencing of the introns. Neither X, Xba, nor Kpn elements are present in the related human albumin gene, whereas Alu's are present in different positions. From phylogenetic evidence, it appears that Alu elements were inserted into the alpha-fetoprotein gene at some time postdating the mammalian radiation 85 million years ago.     

Complete nucleotide and encoded amino acid sequence of a mammalian myosin heavy chain gene. Evidence against intron-dependent evolution of the rod

The complete nucleotide sequence and exon/intron structure of the rat embryonic skeletal muscle myosin heavy chain (MHC) gene has been determined. This gene comprises 24 X 10(3) bases of DNA and is split into 41 exons. The exons encode a 6035 nucleotide (nt) long mRNA consisting of 90 nt of 5' untranslated, 5820 nt of protein coding and 125 nt of 3' untranslated sequence. The rat embryonic MHC polypeptide is encoded by exons 3 to 41 and contains 1939 amino acid residues with a calculated Mr of 223,900. Its amino acid sequence displays the structural features typical for all sarcomeric MHCs, i.e. an amino-terminal "globular" head region and a carboxy-terminal alpha-helical rod portion that shows the characteristics of a coiled coil with a superimposed 28-residue repeat pattern interrupted at only four positions by "skip" residues. The complex structure of the rat embryonic MHC gene and the conservation of intron locations in this and other MHC genes are indicative of a highly split ancestral sarcomeric MHC gene. Introns in the rat embryonic gene interrupt the coding sequence at the boundaries separating the proteolytic subfragments of the head, but not at the head/rod junction or between the 28-residue repeats present within the rod. Therefore, there is little evidence for exon shuffling and intron-dependent evolution by gene duplication as a mechanism for the generation of the ancestral MHC gene. Rather, intron insertion into a previously non-split ancestral MHC rod gene consisting of multiple tandemly arranged 28-residue-encoding repeats, or convergent evolution of an originally non-repetitive ancestral MHC rod gene must account for the observed structure of the rod-encoding portion of present-day MHC genes.     

Essential role of duplications of short motif sequences in the genomic evolution of Bombyx mori

Summary The Bombyx fibroin gene has a discrete mosaic structure of various repetitive sequences, which may have evolved through various repeating arrangements. Detailed sequence analysis of the fibroin gene containing coding and noncoding regions revealed that the whole sequence could be arranged as an array of short repetitive sequences. A portion of the intron of the fibroin gene is one of interspersed repetitive elements. We cloned a 1.5-kb DNA fragment of the Bombyx genome that contains interspersed elements homologous to the intron sequence. Sequence comparison between the intron and the 1.5-kb fragment shows that partial duplication has frequently occurred in evolutionary progress, and the resultant repetitive blocks of short motif sequences are abundant in the genome. These facts suggest that tandem duplication of the short motif sequence is an important rearrangement in genomic evolution of the fibroin gene. Offprint requests to: S. Ichimura     

A glimpse of a putative pre-intron phase of eukaryotic evolution

Comparison of the exon-intron structures of ancient eukaryotic paralogs reveals the absence of conserved intron positions in these genes. This is in contrast to the conservation of intron positions in orthologous genes from even the most evolutionarily distant eukaryotes and in more recent paralogs. The lack of conserved intron positions in ancient paralogs probably reflects the origination of these genes during the earliest phase of eukaryotic evolution, which was characterized by concomitant invasion of genes by group II self-splicing elements (which were to become introns in the future) and extensive duplication of genes.     

Phylogenetic distribution of intron positions in alpha-amylase genes of bilateria suggests numerous gains and losses

Most eukaryotes have at least some genes interrupted by introns. While it is well accepted that introns were already present at moderate density in the last eukaryote common ancestor, the conspicuous diversity of intron density among genomes suggests a complex evolutionary history, with marked differences between phyla. The question of the rates of intron gains and loss in the course of evolution and factors influencing them remains controversial. We have investigated a single gene family, alpha-amylase, in 55 species covering a variety of animal phyla. Comparison of intron positions across phyla suggests a complex history, with a likely ancestral intronless gene undergoing frequent intron loss and gain, leading to extant intron/exon structures that are highly variable, even among species from the same phylum. Because introns are known to play no regulatory role in this gene and there is no alternative splicing, the structural differences may be interpreted more easily: intron positions, sizes, losses or gains may be more likely related to factors linked to splicing mechanisms and requirements, and to recognition of introns and exons, or to more extrinsic factors, such as life cycle and population size. We have shown that intron losses outnumbered gains in recent periods, but that "resets" of intron positions occurred at the origin of several phyla, including vertebrates. Rates of gain and loss appear to be positively correlated. No phase preference was found. We also found evidence for parallel gains and for intron sliding. Presence of introns at given positions was correlated to a strong protosplice consensus sequence AG/G, which was much weaker in the absence of intron. In contrast, recent intron insertions were not associated with a specific sequence. In animal Amy genes, population size and generation time seem to have played only minor roles in shaping gene structures.     

Structure of the human CRFB4 gene: Comparison with its IFNAR neighbor

The cytokine receptor family consists of a growing number of structurally and evolutionarily related transmembrane receptors. CRFB4 and IFNAR are two of the most similar members of this family. They are encoded by two neighboring genes on both human chromosome 21 and murine chromosome 16. The sequence of the human CRFB4 gene was determined from the first exon to the last intron. The nature of the repetitive sequences present in the introns was analyzed and compared with those present in the human IFNAR gene. This analysis leads to considerations of the antiquity of the duplication that gave rise to both genes from a common ancestor. A pseudogene for USF has been identified in the IFNAR gene and a new definition for the repetitive sequence MER37 is proposed. The polymorphism associated with two CA repeats present in the CRFB4 gene is described.The nucleotide sequence reported in this paper has been deposited to GenBank with accession numbers U08988 and U12021 Correspondence to: G. Lutfalla