Instability of repetitive DNA sequences: The role of replication in multiple mechanisms

Rearrangements between tandem sequence homologies of various lengths are a major source of genomic change and can be deleterious to the organism. These rearrangements can result in either deletion or duplication of genetic material flanked by direct sequence repeats. Molecular genetic analysis of repetitive sequence instability in Escherichia coli has provided several clues to the underlying mechanisms of these rearrangements. We present evidence for three mechanisms of RecA-independent sequence rearrangements: simple replication slippage, sister-chromosome exchange-associated slippage, and single-strand annealing. We discuss the constraints of these mechanisms and contrast their properties with RecA-dependent homologous recombination. Replication plays a critical role in the two slipped misalignment mechanisms, and difficulties in replication appear to trigger rearrangements via all these mechanisms.

The architecture of the human Rad54–DNA complex provides evidence for protein translocation along DNA

Proper maintenance and duplication of the genome require accurate recombination between homologous DNA molecules. In eukaryotic cells, the Rad51 protein mediates pairing between homologous DNA molecules. This reaction is assisted by the Rad54 protein. To gain insight into how Rad54 functions, we studied the interaction of the human Rad54 (hRad54) protein with double-stranded DNA. We have recently shown that binding of hRad54 to DNA induces a change in DNA topology. To determine whether this change was caused by a protein-constrained change in twist, a protein-constrained change in writhe, or the introduction of unconstrained plectonemic supercoils, we investigated the hRad54–DNA complex by scanning force microscopy. The architecture of the observed complexes suggests that movement of the hRad54 protein complex along the DNA helix generates unconstrained plectonemic supercoils. We discuss how hRad54-induced superhelical stress in the target DNA may function to facilitate homologous DNA pairing by the hRad51 protein directly. In addition, the induction of supercoiling by hRad54 could stimulate recombination indirectly by displacing histones and/or other proteins packaging the DNA into chromatin. This function of DNA translocating motors might be of general importance in chromatin metabolism.

Two highly homologous ribonuclease genes expressed in mouse eosinophils identify a larger subgroup of the mammalian ribonuclease superfamily.

Two putative ribonucleases have been isolated from the secondary granules of mouse eosinophils. Degenerate oligonucleotide primers inferred from peptide sequence data were used in reverse transcriptase-PCR reactions of bone marrow-derived cDNA. The resulting PCR product was used to screen a C57BL/6J bone marrow cDNA library, and comparisons of representative clones showed that these genes and encoded proteins are highly homologous (96% identity at the nucleotide level; 92/94% identical/similar at the amino acid level). The mouse proteins are only weakly homologous (approximately 50% amino acid identity) with the human eosinophil-associated ribonucleases (i.e., eosinophil-derived neurotoxin and eosinophil cationic protein) and show no sequence bias toward either human protein. Phylogenetic analyses established that the human and mouse loci shared an ancestral gene, but that independent duplication events have occurred since the divergence of primates and rodents. The duplication event generating the mouse genes was estimated to have occurred < 5 x 10(6) years ago (versus 30 to 40 x 10(6) years ago in primates). The identification of independent duplication events in two extant mammalian orders suggests a selective advantage to having multiple eosinophil granule ribonucleases. Southern blot analyses in the mouse demonstrated the existence of three additional highly homologous genes (i.e., five genes total) as well as several more divergent family members. The potential significance of this observation is the implication of a larger gene subfamily in primates (i.e., humans).

Evolution of alcohol dehydrogenase genes in the palm and grass families.

The alcohol dehydrogenase (Adh; alcohol:NAD+ oxidoreductase, EC 1.1.1.1) gene family has two or three loci in a broad array of angiosperm species. The relative stability in the number of Adh loci led Gottlieb [Gottlieb, L. D. (1982) Science 216, 373-380] to propose that the Adh gene family arose from an ancient gene duplication. In this study, the isolation of three loci from the California fan palm (Washingtonia robusta) is reported. The three loci from palm are highly diverged. One palm Adh gene, referred to here as adhB, has been completely sequenced, including 950 nucleotides of the upstream regulatory region. For the second locus, adhA, 81% of the exon sequence is complete. Both show the same basic structure as grass Adh genes in terms of intron number and intron location. The third locus, adhC, for which only a small amount of sequence is available (12% of exon sequence) appears to be more highly diverged. Comparison of the Adh gene families from palms and grasses shows that the adh1 and adh2 genes of grasses, and the adhA and adhB genes of palms, arose by duplication following the divergence of the two families. This finding suggests that the multiple Adh loci in different monocot lineages are not the result of a single ancestral duplication but, rather, of multiple duplication events.

The evolution of the vertebrate Dlx gene family.

The vertebrate Dlx gene family consists of homeobox-containing transcription factors distributed in pairs on the same chromosomes as the Hox genes. To investigate the evolutionary history of Dlx genes, we have cloned five new zebrafish family members and have provided additional sequence information for two mouse genes. Phylogenetic analyses of Dlx gene sequences considered in the context of their chromosomal arrangements suggest that an initial tandem duplication produced a linked pair of Dlx genes after the divergence of chordates and arthropods but prior to the divergence of tunicates and vertebrates. This pair of Dlx genes was then duplicated in the chromosomal events that led to the four clusters of Hox genes characteristic of bony fish and tetrapods. It is possible that a pair of Dlx genes linked to the Hoxc cluster has been lost from mammals. We were unable to distinguish between independent duplication and retention of the ancestral state of bony vertebrates to explain the presence of a greater number of Dlx genes in zebrafish than mammals. Determination of the linkage relationship of these additional zebrafish Dlx genes to Hox clusters should help resolve this issue.

Identification of interdomain sequences promoting the intronless evolution of a bacterial protein family.

In the evolution of eukaryotic genes, introns are believed to have played a major role in increasing the probability of favorable duplication events, chance recombinations, and exon shuffling resulting in functional hybrid proteins. As a rule, prokaryotic genes lack introns, and the examples of prokaryotic introns described do not seem to have contributed to gene evolution by exon shuffling. Still, certain protein families in modern bacteria evolve rapidly by recombination of genes, duplication of functional domains, and as shown for protein PAB of the anaerobic bacterial species Peptostreptococcus magnus, by the shuffling of an albumin-binding protein module from group C and G streptococci. Characterization of a protein PAB-related gene in a P. magnus strain with less albumin-binding activity revealed that the shuffled module was missing. Based on this fact and observations made when comparing gene sequences of this family of bacterial surface proteins interacting with albumin and/or immunoglobulin, a model is presented that can explain how this rapid intronless evolution takes place. A new kind of genetic element is introduced: the recer sequence promoting interdomain, in frame recombination and acting as a structure-less flexibility-promoting spacer in the corresponding protein. The data presented also suggest that antibiotics could represent the selective pressure behind the shuffling of protein modules in P. magnus, a member of the indigenous bacterial flora.

The root of the universal tree and the origin of eukaryotes based on elongation factor phylogeny.

The genes for the protein synthesis elongation factors Tu (EF-Tu) and G (EF-G) are the products of an ancient gene duplication, which appears to predate the divergence of all extant organismal lineages. Thus, it should be possible to root a universal phylogeny based on either protein using the second protein as an outgroup. This approach was originally taken independently with two separate gene duplication pairs, (i) the regulatory and catalytic subunits of the proton ATPases and (ii) the protein synthesis elongation factors EF-Tu and EF-G. Questions about the orthology of the ATPase genes have obscured the former results, and the elongation factor data have been criticized for inadequate taxonomic representation and alignment errors. We have expanded the latter analysis using a broad representation of taxa from all three domains of life. All phylogenetic methods used strongly place the root of the universal tree between two highly distinct groups, the archaeons/eukaryotes and the eubacteria. We also find that a combined data set of EF-Tu and EF-G sequences favors placement of the eukaryotes within the Archaea, as the sister group to the Crenarchaeota. This relationship is supported by bootstrap values of 60-89% with various distance and maximum likelihood methods, while unweighted parsimony gives 58% support for archaeal monophyly.

Naturally occurring variation in copia expression is due to both element (cis) and host (trans) regulatory variation.

Significant differences in levels of copia [Drosophila long terminal repeat (LTR) retrotransposon] expression exist among six species representing the Drosophila melanogaster species complex (D. melanogaster, Drosophila mauritiana, Drosophila simulans, Drosophila sechellia, Drosophila yakuba, and Drosophila erecta) and a more distantly related species (Drosophila willistoni). These differences in expression are correlated with major size variation mapping to putative regulatory regions of the copia 5' LTR and adjacent untranslated leader region (ULR). Sequence analysis indicates that these size variants were derived from a series of regional duplication events. The ability of the copia LTR-ULR size variants to drive expression of a bacterial chloramphenicol acetyltransferase reporter gene was tested in each of the seven species. The results indicate that both element-encoded (cis) and host-genome-encoded (trans) genetic differences are responsible for the variability in copia expression within and between Drosophila species. This finding indicates that models purporting to explain the dynamics and distribution of retrotransposons in natural populations must consider the potential impact of both element-encoded and host-genome-encoded regulatory variation to be valid. We propose that interelement selection among retrotransposons may provide a molecular drive mechanism for the evolution of eukaryotic enhancers which can be subsequently distributed throughout the genome by retrotransposition.

Evolutionary changes in the expression pattern of a developmentally essential gene in three Drosophila species.

The hypothesis that morphological evolution may largely result from changes in gene regulation rather than gene structure has been difficult to test. Morphological differences among insects are often apparent in the cuticle structures produced. The dopa decarboxylase (Ddc) and alpha-methyldopa hypersensitive (amd) genes arose from an ancient gene duplication. In Drosophila, they have evolved nonoverlapping functions, including the production of distinct types of cuticle, and for Ddc, the production of the neurotransmitters, dopamine and serotonin. The amd gene is particularly active in the production of specialized flexible cuticles in the developing embryo. We have compared the pattern of amd expression in three Drosophila species. Several regions of expression conserved in all three species but, surprisingly, a unique domain of expression is found in Drosophila simulans that does occur in the closely related (2-5 million years) Drosophila melanogaster or in the more remote (40-50 million years) Drosophila virilis. The "sudden" appearance of a completely new and robust domain of expression provides a glimpse of evolutionary variation resulting from changes in regulation of structural gene expression.

Circular RNAs from transcripts of the rat cytochrome P450 2C24 gene: correlation with exon skipping.

The cytochrome P450 2C24 gene is characterized by the capability to generate, in rat kidney, a transcript containing exons 2 and 4 spliced at correct sites but having the donor site of exon 4 directly joined to the acceptor site of exon 2 (exon scrambling). By reverse transcriptase-PCR analysis, it is now shown that the only exons present in the scrambled transcript are exons 2, 3, and 4 and that this molecule lacks a poly(A)+ tail. Furthermore, the use of PCR primers in both orientations of either exon 2 or exon 4 revealed that the orders of the exons in the scrambled transcript are 2-3-4-2 and 4-2-3-4, respectively. These results, combined with the observation that P450 2C24 is a single-copy gene, with no duplication of the exon 2 to exon 4 segment, suggest that the scrambled transcript has properties consistent with that of a circular molecule. In line with this is the observation of an increased resistance of the transcript to phosphodiesterase I, a 3'-exonuclease. Moreover, an alternatively processed cytochrome P450 2C24 mRNA, lacking the three scrambled exons and having exon 1 directly joined to exon 5, has been identified in kidney and liver, tissues that express the scrambled transcript. This complete identity of the exons that are absent in the alternatively processed mRNA but present in the scrambled transcript is interpreted as indicative of the possibility that exon scrambling and exon skipping might be interrelated phenomena. It is therefore proposed that alternative pre-mRNA processing has the potential to generate not only mRNAs lacking one or more exons but also circular RNA molecules.

A new hemoglobin gene from soybean: a role for hemoglobin in all plants.

We have isolated a new hemoglobin gene from soybean. It is expressed in cotyledons, stems of seedlings, roots, young leaves, and in some cells in the nodules that are associated with the nitrogen-fixing Bradyrhizobium symbiont. This contrasts with the expression of the leghemoglobins, which are active only in the infected cells of the nodules. The deduced protein sequence of the new gene shows only 58% similarity to one of the soybean leghemoglobins, but 85-87% similarity to hemoglobins from the nonlegumes Parasponia, Casuarina, and barley. The pattern of expression and the gene sequence indicate that this new gene is a nonsymbiotic legume hemoglobin. The finding of this gene in legumes and similar genes in other species strengthens our previous suggestion that genomes of all plants contain hemoglobin genes. The specialized leghemoglobin gene family may have arisen from a preexisting nonsymbiotic hemoglobin by gene duplication.

Stimulation of intrachromosomal homologous recombination in human cells by electroporation with site-specific endonucleases.

In somatic mammalian cells, homologous recombination is a rare event. To study the effects of chromosomal breaks on frequency of homologous recombination, site-specific endonucleases were introduced into human cells by electroporation. Cell lines with a partial duplication within the HPRT (hypoxanthine phosphoribosyltransferase) gene were created through gene targeting. Homologous intrachromosomal recombination between the repeated regions of the gene can reconstruct a functioning, wild-type gene. Treatment of these cells with the restriction endonuclease Xba I, which has a recognition site within the repeated region of HPRT homology, increased the frequency or homologous recombination bv more than 10-fold. Recombination frequency was similarly increased by treatment with the rare-cutting yeast endonuclease PI-Sce I when a cleavage site was placed within the repeated region of HPRT. In contrast, four restriction enzymes that cut at positions either outside of the repeated regions or between them produced no change in recombination frequency. The results suggest that homologous recombination between intrachromosomal repeats can be specifically initiated by a double-strand break occurring within regions of homology, consistent with the predictions of a model.

Unexpected beta2-microglobulin sequence diversity in individual rainbow trout.

For mammals beta2-microglobulin (beta2m), the light chain of major histocompatibility complex (MHC) class I molecules, is invariant (or highly conserved) and is encoded by a single gene unlinked to the MHC. We find that beta2m of a salmonid fish, the rainbow trout (Oncorhynchus mykiss), does not conform to the mammalian paradigm. Ten of 12 randomly selected beta2m cDNA clones from an individual fish have different nucleotide sequences. A complex restriction fragment length polymorphism pattern is observed with rainbow trout, suggesting multiple beta2m genes in the genome, in excess of the two genes expected from the ancestral salmonid tetraploidy. Additional duplication and diversification of the beta2m genes might have occurred subsequently. Variation in the beta2m cDNA sequences is mainly at sites that do not perturb the structure of the mature beta2m protein, showing that the observed diversity of the trout beta2m genes is not primarily a result of pathogen selection.

DNA rearrangement mediated by inverted repeats.

Inverted repeats of DNA are widespread in the genomes of eukaryotes and prokaryotes and can mediate genome rearrangement. We studied rearrangement mediated by plasmid-borne inverted repeats in Escherichia coli. We show that inverted repeats can mediate an efficient and recA-independent recombination event. Surprisingly, the product of this recombination is not that of simple inversion between the inverted repeats, but almost exclusively an unusual head-to-head dimer with complex DNA rearrangement. Moreover, this recombination is dramatically reduced by increasing the distance separating the repeats. These results can be readily explained by a model involving reciprocal switching of the leading and lagging strands of DNA replication within the inverted repeats, which leads to the formation of a Holliday junction. Reciprocal strand switching during DNA replication might be a common mechanism for genome rearrangement associated with inverted duplication.

The chromosomal distribution of mouse odorant receptor genes.

Odorant receptors (ORs) on nasal olfactory sensory neurons are encoded by a large multigene family. Each member of the family is expressed in a small percentage of neurons that are confined to one of several spatial zones in the nose but are randomly distributed throughout that zone. This pattern of expression suggests that when the sensory neuron selects which OR gene to express it may be confined to a particular zonal gene set of several hundred OR genes but select from among the members of that set via a stochastic mechanism. Both locus-dependent and locus-independent models of OR gene choice have been proposed. To investigate the feasibility of these models, we determined the chromosomal locations of 21 OR genes expressed in four different spatial zones. We found that OR genes are clustered within multiple loci that are broadly distributed in the genome. These loci lie within paralogous chromosomal regions that appear to have arisen by duplications of large chromosomal domains followed by extensive gene duplication and divergence. Our studies show that OR genes expressed in the same zone map to numerous loci; moreover, a single locus can contain genes expressed in different zones. These findings raise the possibility that OR gene choice may be locus-independent or involve consecutive stochastic choices.