Long-term evolution of transposable elements

Transposable elements are often considered parasitic DNA sequences, able to invade the genome of their host thanks to their self-replicating ability. This colonization process has been extensively studied, both theoretically and experimentally, but their long-term coevolution with the genomes is still poorly understood. In this work, we aim to challenge previous population genetics models by considering features of transposable elements as quantitative, rather than discrete, variables. We also describe more realistic transposable element dynamics by accounting for the variability of the insertion effect, from deleterious to adaptive, as well as mutations leading to a loss of transposition activity and to nonautonomous copies. Individual-based simulations of the behavior of a transposable-element family over several thousand generations show different ways in which active or inactive copies can be maintained for a very long time. Results reveal an unexpected impact of genetic drift on the "junk DNA" content of the genome and strongly question the likelihood of the sustainable long-term stable transposition-selection equilibrium on which numerous previous works were based.     

Identification of functional domains of human erythrocyte spectrin.

Isolated human erythrocyte spectrin is a dimer of two unique polypeptide chains. The dimer (alpha beta) undergoes reversible salt- and temperature-dependent association to form (alpha beta)2 tetramers. Spectrin also binds with high affinity to a protein receptor on the cytoplasmic surface of erythrocyte membrane vesicles. By cleavage of spectrin at its cysteine residues with 2-nitro-5-thiocyanobenzoic acid, a 50,000-dalton peptide fragment has been isolated which inhibits the binding of spectrin to erythrocyte membrane vesicles. This peptide arises from a terminal region of the beta chain. An 80,000-dalton peptide generated by restricted trypsin digestion binds preferentially to dimeric spectrin. This peptide arises from a terminal portion of the alpha chain. Multiple peptides involved in noncovalent associations between the chains have also been identified. These associations indicate that the two subunits of spectrin are aligned parallel to one another and that the tetramer formation site and the high-affinity membrane binding site are in close proximity to one another.     

Identification and characterization of putative transposable DNA elements in solanaceous plants and Caenorhabditis elegans.

Several families of putative transposable elements (TrEs) in both solanaceous plants and Caenorhabditis elegans have been identified by screening the DNA data base for inverted repeated domains present in multiple copies in the genome. The elements are localized within intron and flanking regions of many genes. These elements consist of two inverted repeats flanking sequences ranging from 5 bp to > 500 bp. Identification of multiple elements in which sequence conservation includes both the flanking and internal regions implies that these TrEs are capable of duplicative transposition. Two of the elements were identified in promoter regions of the tomato (Lycoperiscon esculentum) polygalacturonase and potato (Solanum tuberosum) Win1 genes. The element in the polygalacturonase promoter spans a known regulatory region. In both cases, ancestral DNA sequences, which represent potential recombination target sequences prior to insertion of the elements, have been cloned from related species. The sequences of the inverted repeated domains in plants and C. elegans show a high degree of phylogenetic conservation. While frequency of the different elements is variable, some are present in very high copy number. A member of a single C. elegans TrE family is observed approximately once every 20 kb in the genome. The abundance of the described TrEs suggests utility in the genomic analysis of these and related organisms.     

Structural similarities between viroids and transposable genetic elements.

The primary structures of the tomato planta macho and tomato apical stunt viroids have been determined, and probable secondary structures are proposed. Both viroids can assume the rodlike conformation with extensive base-pairing characteristic of all known viroids. Sequence homologies between the two viroids (75%) and with members of the potato spindle tuber viroid group (73-83%) indicate that they both belong to this group. Comparative sequence analysis of all members of the group reveals striking similarities with the ends of transposable genetic elements. These similarities, the presence of inverted repeats often ending with the dinucleotides U-G and C-A, and flanking imperfect direct repeats suggest that viroids may have originated from transposable elements or retroviral proviruses by deletion of interior portions of the viral (or element) DNA.     

Inverted repeats of Tn5 are transposable elements.

Experiments presented here show that each of the 1.5-kilobase inverted repeats of the kanamycin-resistance transposon Tn5 is transposable; we designate them IS50-L (left) and IS50-R (right). By DNA sequence analyses, IS50 is 1533 base pairs (bp) long and generates 9-bp direct repeats of target sequences. The ends of IS50 comprise a hyphenated 8-of-9-bp inverted repeat and are not used with equal efficiency; the outside ends are more active than the inside ends, suggesting that a strong transposase recognition site at the outside ends, suggesting that a strong transposase recognition site at the outside end extends beyond the 8 bp common to both ends.     

Movement of yeast transposable elements by gene conversion.

We have constructed yeast strains in which Ty (transposon yeast) elements at the HIS4 locus are genetically marked with the yeast URA3 gene. By isolating and analyzing Ura- derivatives of these strains, we have detected a variety of Ty-mediated recombination events. In this paper, we describe events in which the DNA sequence of the Ty element at the HIS4 locus is replaced by the DNA sequence of a different Ty element. These replacements occur without alterations in the flanking DNA sequence and without chromosomal aberrations. We believe that these events result from gene conversion between the Ty element at HIS4 and a Ty element at a different site in the yeast genome. Gene conversion can occur between Ty elements that differ by large insertion and substitution mutations. These recombination events result not only in the movement of Ty sequences but also in alterations in expression of the adjacent HIS4 gene. Different Ty elements at the same site in the HIS4 regulatory region can result in His-, His+, and cold-sensitive His+ phenotypes. Several Ty elements render expression of the HIS4 gene subject to control by genes at the mating type locus.     

Evolution of P transposable elements: sequences of Drosophila nebulosa P elements.

P elements have been cloned and sequenced from Drosophila nebulosa. Their sequences have diverged less than 6% from P elements of Drosophila melanogaster. However D. nebulosa P elements have nucleotide changes that close all four open reading frames found in the D. melanogaster P element. Microinjection experiments show that D. nebulosa P elements cannot provide transposase function for D. melanogaster P elements, nor are D. nebulosa P elements mobilized by the transposase provided by a D. melanogaster P factor. Three D. nebulosa P elements appear to have integrated into the same position of a complex, centromeric repeated sequence. Comparison of nucleotide sequences suggests that D. nebulosa P elements have diverged upon different pathways from a common ancestor that was 99% homologous to the P elements of D. melanogaster.     

Human Tc1 and Tc2/Tc0 CD8 T-cell clones display distinct cell surface and functional phenotypes

It has recently become clear that distinct subsets of CD8 T cells, analogous to their CD4 counterparts, exist in rodents and humans. To examine functional differences between human CD8 T-cell subsets, we generated Tc1, Tc2, and Tc0 T-cell clones from the peripheral blood of healthy individuals. The majority of CD8 T-cell clones generated displayed a classic Tc1 phenotype, but 10% to 20% secreted interleukin (IL)-4 in addition to interferon-gamma (Tc0 phenotype). Generation of Tc2 clones was dependent on the use of anti-CD3 and anti-CD28 as the primary stimulus. The cytokine profiles of established clones remained susceptible to modification by the addition of IL-12 and IL-4. In addition, IL-12 enhanced and IL-4 inhibited the growth of Tc1 but not Tc2/0 CD8 T-cell clones. Significant functional differences were observed between the subsets. Tc2/0 clones expressed CD30 and CD40 ligand at a much higher level than Tc1 clones. Both Tc1 and Tc2/0 clones showed comparable cytotoxicity and produced similar levels of perforin and Fas L. However, Tc2 clones were much more resistant to activation-induced cell death and less susceptible to apoptosis by direct Fas ligation. Moreover, Tc1 and Tc2 clones had opposing effects on the development of CD4 effectors, promoting type 1 and type 2 responses, respectively. These data provide evidence for profound differences between human CD8 T-cell subsets that may be important in their functions as cytotoxic or immunoregulatory cells. (Blood. 2000;95:231-240)     

Tc4, a Caenorhabditis elegans transposable element with an unusual fold-back structure.

We have identified and characterized a family of transposable elements in the nematode Caenorhabditis elegans. The Tc4 transposable element family is present at about 20 copies per haploid genome in the C. elegans Bristol and Bergerac strains. Although Tc4 transposition events have not been observed in these wild-type strains, we have identified Tc4 transposition events in the mut-2 mutant strain TR679, in which the elements Tc1 and Tc3 also transpose at a higher frequency than in the wild type. We determined the sequence of one Tc4 element. This 1.6-kilobase element contains almost perfect inverted terminal repeats of 774 base pairs (bp) with a 57-bp unique internal sequence. Tc4 is a fold-back element, but its long inverted terminal repeats, unlike those of the fold-back elements of other organisms, do not consist of multiple short repeats. In the two cases studied, Tc4 insertion resulted in duplication of a TNA trinucleotide target site. The family of Tc4 elements differs from other C. elegans transposable element families in structure, degree of structural heterogeneity, and target-site specificity.     

Mapping of functional domains in adenovirus E1A proteins.

We have modified the E1A gene of human subgroup C adenovirus by introducing deletions in its coding sequence. Various truncated E1A proteins were expressed in Escherichia coli, purified, and microinjected via glass capillaries into Vero cells. We monitored their movement from the cell cytoplasm to the nucleus and their ability to induce expression of H5dl312, an adenovirus E1A deletion mutant. Our results show that the carboxyl terminus of E1A contains sequences essential for rapid and efficient nuclear localization. Essential information for efficient H5dl312 complementation is contained in an internal region, comprising sequences of both exons of the E1A gene. A first exon-encoded region, however, is sufficient to induce low levels of adenovirus gene expression. Information for nuclear localization and for H5dl312 complementation are therefore encoded by distinct domains of the E1A gene. In addition, we determined that the human c-myc product was unable to complement H5dl312.     

Full-length cDNA for rabbit tryptophan hydroxylase: functional domains and evolution of aromatic amino acid hydroxylases.

A full-length cDNA for tryptophan hydroxylase was cloned from rabbit pineal body by screening an expression library with antibody against rat phenylalanine hydroxylase, which crossreacts with rabbit tryptophan hydroxylase. Clones producing immunoreactive material contain sequences homologous to, yet distinct from, phenylalanine hydroxylase. The rabbit cDNA hybridizes to mRNA in pineal body and brainstem but not in liver. Comparison of the rabbit tryptophan hydroxylase sequence with the sequences of phenylalanine hydroxylase and tyrosine hydroxylase demonstrates that these three biopterin-dependent aromatic amino acid hydroxylases are highly homologous, reflecting a common evolutionary origin from a single primordial genetic locus. The pattern of sequence homology supports the hypothesis that the carboxyl-terminal two-thirds of the molecules constitute the enzymatic activity cores, and the amino-terminal thirds of the molecules constitute domains for substrate specificity.     

Identification of functional similarities between proteins using directed evolution

Protein sequences are often highly redundant and evolution can change them beyond recognition. It can therefore be difficult to identify proteins with functional or structural similarities by inspection of their sequences. Here we have used an experimental evolutionary approach to detect hidden similarities between the antisense RNA-binding protein Rop and other proteins. We created an envelope of functional Rop mutants by combinatorial mutagenesis, used the compilation of mutant sequences to search a database of protein structures, and thereby identified a segment of the enzyme valyl-tRNA-synthetase (ValRS). Further inspection revealed that the structures of the RNA-binding sites of both proteins are highly related, as indeed are the RNA ligands. From the known 3D structure of the ValRS in complex with tRNA, we were able to build a model of an RNA hairpin pair in complex with Rop that has proved to be consistent with the biochemical and NMR data for the interaction between Rop and RNA hairpins. We suggest that this approach (mutational envelope scanning), by generating sequence information de novo, can help uncover hidden similarities between proteins.     

Identification of a 95-kDa WEE1-like tyrosine kinase in HeLa cells.

Human WEE1 (WEE1Hu) was cloned on the basis of its ability to rescue wee1+ mutants in fission yeast [Igarashi, M., Nagata, A., Jinno, S., Suto, K. & Okayama, H. (1991) Nature (London) 353, 80-83]. Biochemical studies carried out in vitro with recombinant protein demonstrated that WEE1Hu encodes a tyrosine kinase of approximately 49 kDa that phosphorylates p34cdc2 on Tyr-15 [Parker, L. L. & Piwnica-Worms, H. (1992) Science 257, 1955-1957]. To study the regulation of WEE1Hu in human cells, two polyclonal antibodies to bacterially produced p49WEE1Hu were generated. In addition, a peptide antibody generated against amino acids 361-388 of p49WEE1Hu was also used. Unexpectantly, these antibodies recognized a protein with an apparent molecular mass of 95 kDa in HeLa cells, rather than one of 49 kDa. Immunoprecipitates of p95 phosphorylated p34cdc2 on Tyr-15, indicating that p95 is functionally related to p49WEEIHu, and mapping studies demonstrated that p95 is structurally related to p49WEE1Hu. In addition, the substrate specificity of p95 was more similar to that of fission yeast p107wee1 than to that of human p49WEE1. Finally, the kinase activity of p95 toward p34cdc2/cyclin B was severely impaired during mitosis. Taken together, these results indicate that the original WEE1Hu clone isolated in genetic screens encodes only the catalytic domain of human WEE1 and that the authentic human WEE1 protein has an apparent molecular mass of approximately 95 kDa.     

Recombination involving transposable elements: role of target molecule replication in Tn1 delta Ap-mediated replicon fusion.

Donor DNA molecules carrying Tn1 or Tn3 deletion mutants do not need to replicate in order to participate in replicon fusion recombination events during which the Tn1/Tn3 element is duplicated. We have assayed Tn1 delta Ap-mediated replicon fusion events involving plasmid R388 and the bacteriophage lambda-derived plasmid p lambda CM, and we find that the role of the recipient molecule is distinct. When p lambda CM carries Tn1 delta Ap, replicon fusion occurs in more than 1% of all cells assayed, whether or not p lambda CM::Tn1 delta Ap can replicate. In contrast, when R388 carries Tn1 delta Ap, replicon fusion occurs only when the p lambda CM target can replicate. Blocks to p lambda CM replication by prophage repressor or amber mutations of the O and P cistrons reduce replicon fusion so that it occurs in less than 1 out of 10(5) cells assayed.     

Diverse transposable elements are mobilized in hybrid dysgenesis in Drosophila virilis.

We describe a system of hybrid dysgenesis in Drosophila virilis in which at least four unrelated transposable elements are all mobilized following a dysgenic cross. The data are largely consistent with the superposition of at least three different systems of hybrid dysgenesis, each repressing a different transposable element, which break down following the hybrid cross, possibly because they share a common pathway in the host. The data are also consistent with a mechanism in which mobilization of a single element triggers that of others, perhaps through chromosome breakage. The mobilization of multiple, unrelated elements in hybrid dysgenesis is reminiscent of McClintock's evidence [McClintock, B. (1955) Brookhaven Symp. Biol. 8, 58-74] for simultaneous mobilization of different transposable elements in maize.     

Nucleotide sequence and evolution of ETn elements.

The ETn (for "early transposon") family of long repeated sequences in abundantly transcribed in early mouse embryos from retroviral-like long terminal repeats. Nucleotide sequencing of two elements does not reveal any long open reading frame nor significant homology to retroviral proteins. The genetic polymorphism, monitored by Southern blotting within and across mouse species, reflects a concerted mode of evolution for the ETn sequences.     

Morphological evolution through complex domains of fitness.

Computer simulated phenotypic walks through multi-dimensional fitness-landscapes indicate that (i) the number of phenotypes capable of reconciling conflicting morphological requirements increases in proportion to the number of manifold functional obligations an organism must perform to grow, survive, and reproduce, and (ii) walks over multi-task fitness-landscapes require fewer but larger phenotypic transformations than those through single-task landscapes. These results were determined by (i) simulating a "morphospace" containing 200,000 phenotypes reminiscent of early Paleozoic vascular sporophytes, (ii) evaluating the capacity of each morphology to perform each of three tasks (light interception, mechanical support, and reproduction) as well as the ability to reconcile the conflicting morphological requirements for the four combinatorial permutations of these tasks, (iii) simulating the walks obtaining all phenotypic maxima or optima within the seven "fitness-landscapes," and (iv) computing the mean morphological variation attending these walks. The results of these simulations, whose credibility is discussed in the context of early vascular land-plant evolution, suggest that both the number and the accessibility of phenotypic optima increase as the number of functional obligations contributing to total fitness increases (i.e., as the complexity of optimal phenotypes increases, the fitnesses of optima fall closer to the mean fitness of all the phenotypes under selection).