A similar situation was recently reported for the canine AP3B1 gene [ 13 ]. A second situation in which transcription slippage has a positive outcome is when it leads to synthesis of more than one useful product from a single gene - during expression of the P gene in paramyxoviruses, for example.
The best-studied example is in Sendai virus, where a specific number of untemplated Gs are inserted at the position corresponding to the slippery site reviewed in [ 14 ]. Remarkably, this process depends on a hexanucleotide phasing of the slippery sequence relative to the end of genome and this is modulated by viral protein N [ 15 ].
In addition to its involvement in paramyxovirus decoding, transcriptional slippage is used for the synthesis of additional functional proteins in other viruses, such as Ebola virus [ 16 — 18 ]. Utilization of transcriptional slippage is not limited to viral genes. This gene has a run of nine As in its sense strand seven-eighths of the way through its coding sequence.
When the number of As is anything else, for example 8, 10, 11, 13, the translating ribosomes encounter a 3' stop codon located close to the poly A run. They terminate, resulting in the synthesis of a shorter product Figure 1 , the gamma subunit of DNA polymerase III, which has distinctive functional properties [ 20 , 21 ]. In some other bacteria such as E. The same end result can be achieved by nonstandard events at different levels of readout [ 19 ]. A scheme for the nonlinear expression of Thermus thermophilus dnaX via transcriptional slippage.
Another example of the use of transcriptional slippage was recently reported in the decoding of the Shigella flexneri mxiE gene which encodes a transcription activator [ 27 ]. Transcriptional insertion of an additional non-templated nucleotide at the run of Us results in a proportion of the mRNAs having mxiEa and mxiEb in the same reading frame [ 27 ].
Therefore, in contrast to T. Transcription slippage-prone sequences are expected to be under-represented in coding regions [ 2 ], because functional utilization of such sequences is unlikely to be common. The recent dramatic increase in the number of sequenced bacterial genomes provides an opportunity to perform wide-scale analysis of whole kingdoms of life [ 28 ]. The current study explores whether long runs of As or Ts are indeed avoided in the coding regions of sequenced bacterial genomes, and where such runs do occur, whether they play a positive functional role in gene expression.
If any sequence pattern is randomly distributed in a genomic sequence, the following equation should be satisfied:. Where Pc is the number of pattern copies in coding regions, Pg is the number of copies in the whole genome, Nc the number of nucleotides in coding regions and Ng the size of the whole genome.
An example of such an analysis for a few representative genomes is illustrated in Figure 2a. The position of the transition is different among the genomes analyzed. Therefore, they are more likely to occur if there is positive selection.
This suggests that such organisms have developed a mechanism to suppress transcriptional slippage at long runs of As or Ts. Analysis of the distribution of runs of As and Ts in selected genomes.
Run length is indicated on the x -axis; the ratio of pattern occurrence on the y -axis. Biases preserved during the randomization procedure are indicated above each pair of graphs. Comparison of poly A and poly T occurrence in genomic sequences versus coding regions has two disadvantages.
First, runs of As cannot be discriminated from runs of Ts at the level of genomic sequences. Second, such runs could have a positive or negative role s outside of coding regions. For example, long runs of Ts can serve as parts of transcriptional terminators, although poly T runs do not have to be uninterrupted for this purpose [ 30 ].
In addition, the occurrence of A and T runs can be affected by dinucleotide bias, codon usage and amino-acid composition of encoding proteins. To minimize the influence of these factors on our analysis, we used another approach to estimate the distribution of such patterns.
A thousand random genomes were generated for every genome shown in Figure 2a using the following rules: protein sequences from the real genomes were preserved, but the codons encoding the amino acids were randomized, taking into account codon usage. Such random genomes are relieved of selective pressure to avoid slippery sequences.
A similar approach was previously used for statistical analysis of frameshift-inducing patterns in E. In addition, we used randomization approaches that preserved dinucleotide bias and both dinucleotide bias and codon usage using the DiShuffle and CodonDishuffle programs developed by Katz and Burge [ 32 ]. If there were no selective pressure on a particular pattern, its occurrence in random genomes would be similar to its occurrence in a corresponding real genome. If there were negative selection against a particular pattern, it would occur more frequently in random genomes than in real ones.
This analysis confirmed our general conclusion that runs of As and Ts of a certain length are avoided in some prokaryotic genomes, but the length of the pattern that is likely to be harmful varies among different genomes. Consequently, such patterns are significantly under-represented in AT-rich genomes. However, when the occurrence of such patterns is compared with their occurrence in random genomes, a negative selection is evident for patterns of exceptional length.
This suggests that very long patterns have a negative effect not associated with transcriptional slippage. The next step was to find occurrences of transcriptional slippage and to investigate, using comparative sequence analysis, whether they are likely to have any functional role. The scheme of this analysis is shown in Figure 3. We searched for occurrences of 9As and 9Ts in protein encoding genes. Only those genes were selected where transcriptional slippage would result in synthesis of a protein which is larger than the counterpart generated by standard decoding.
When transcriptional slippage results in the synthesis of a truncated product, as in decoding T. The next filter was the exclusion of genes from bacteria where transcriptional slippage is unlikely to occur on runs of 9As and Ts.
Organisms with AT-rich genomes that do not demonstrate selection against 9A and 9T sequences within their coding regions may have evolved to suppress transcriptional slippage on 9A and 9T and are unlikely to exhibit it. To select bacteria in which transcriptional slippage on 9A and 9T is unlikely, we first determined the number of genes containing 9A and 9T.
For those bacteria where this number was higher than the threshold number 20 we assumed that it is unlikely that transcriptional slippage can be utilized by more than 20 genes in the same species we searched for evidence of negative selection against these sequences. If such sequences were not under-represented, corresponding bacteria were considered as those where transcriptional slippage is unlikely to occur on 9A or 9T runs.
Genes from such bacteria were excluded from further analysis. The remaining pool of genes contained some identical genes. Some of these exist in multiple copies inside the same genome whereas others are identical because they derived from genomes of highly related species. Such identical genes were combined to reduce redundancy.
In the list of these genes 2 only one representative is given for each group of identical genes. The products of those genes that can be generated by transcriptional slippage were compared to each other using tBLASTn [ 33 ], and to those derived from other sequences present in sequenced bacterial genomes. Genes that produced no significant sequence similarity were considered as ORFans [ 34 , 35 ]. Since ORFans are not suitable for comparative analysis, they were excluded from further analysis shown in gray in 2.
The number of gene groups for which homologs were found is The likelihood of functional utilization of transcriptional slippage was estimated using comparative sequence analysis.
According to the scheme utilized Figure 4 , we consider transcriptional slippage patterns likely to be functional if the organization of ORFs fused by transcriptional slippage is the same in at least two non-identical sequences sharing significant sequence similarity.
We have not found evidence of functional utilization of transcriptional slippage for 40 cases shown in blue in 2. Most probably, although transcriptional slippage is likely to occur during expression of these genes, it has no significant detrimental effect. This result is consistent with our previous finding that sequences that direct significant levels of frameshifting in the E.
Six cases were found where protein products expressed by transcriptional slippage have homologs encoded in a single ORF in genes from other species. One example is shown in Figure 5. Such genes are normally considered as pseudogenes, because their ORF is disrupted.
However, transcriptional slippage should result in the synthesis of normal functional protein and consequently such genes should not be treated as inactive as a result of frameshift mutation. These genes are shown in green in 2. In seven cases red in 2 homologs were found with both a conserved organization of the overlapping ORFs and a conserved pattern of 9As in the overlapping regions. Among them, six cases derive from IS elements whose total number of copies is One group is composed of the mapW genes from Staphylococcus aureus strains; mapW is a functional candidate derived from a non-mobile element.
In the DNA, an A causing a frameshift mutation is underlined. Transcriptional slippage was recently found in the S. We have obtained an initial view of the distribution and functional utilization of simple transcriptional slippage sites in bacterial genomes performed on a multiple-genome scale. The data obtained demonstrate that runs of As and Ts, which result in efficient transcriptional slippage, are significantly underrepresented in coding regions of AT-rich genomes.
One likely reason for this underrepresentation is the 'slippery' nature of such sites. In addition to transcriptional slippage, these sequences are likely to be hypermutable as a result of slippage during replication.
This also contributes to negative selection against these sequences. It has previously been shown that in eukaryotes short repetitive sequences of specific length are usually under-represented in coding regions compared to noncoding regions [ 36 ]. The implication is that such sequences are susceptible to frameshift errors at the DNA level.
We cannot distinguish whether the reason for negative selection against A or T runs is slippage at the replication or transcriptional level or at both. Our approach to finding genes where transcriptional slippage is functionally utilized can, however, discriminate it from replicational slippage in some instances. Since we deal with those cases where sequence extension after a slippery pattern in a shifted reading frame is conserved among several homologs, it is very likely that this extension is expressed.
In the first case, the result would be the existence of a population of bacteria with heterogeneous genomes where different members of such a population would have a different number of nucleotides within a repetitive run, as previously described for several occurrences in the Campylobacter jejuni [ 37 ].
We have found several such examples for the group of genes that we classified as 'pseudo pseudogenes' an example is in Figure 5. If a specific run of 9As or 9Ts occurs within a number of homologs and the length of such run is conserved among all homologs, then it is very likely that this specific run is used for purposeful transcriptional slippage to generate a set of heterogeneous mRNAs.
Acids Res. Copy-choice illegitimate DNA recombination revisited. Eckert, K. CrossRef Full Text. Engler, M. Two forms of the DNA polymerase of bacteriophage T7. Feschenko, V.
Slipped misalignment mechanisms of deletion formation: analysis of deletion endpoints. Gemayel, R. Variable tandem repeats accelerate evolution of coding and regulatory sequences.
Greenstein, D. Interaction between the replication origin and the initiator protein of the filamentous phage f1. Binding occurs in two steps. Gueguen, Y. Characterization of two DNA polymerases from the hyperthermophilic euryarchaeon Pyrococcus abyssi. Henneke, G. In vitro reconstitution of RNA primer removal in Archaea reveals the existence of two pathways.
Ishino, Y. A novel DNA polymerase family found in Archaea. Kim, J. The balancing act of DNA repeat expansions. Kroutil, L. Exonucleolytic proofreading during replication of repetitive DNA. Biochemistry 35, — Kunkel, T. DNA replication fidelity. Lechner, R. Characterization of strand displacement synthesis catalyzed by bacteriophage T7 DNA polymerase.
Regulation of interactions with sliding clamps during DNA replication and repair. Genomics 10, — Lovett, S. Encoded errors: mutations and rearrangements mediated by misalignment at repetitive DNA sequences. Lyons-Darden, T. Madsen, C. McMurray, C. DNA secondary structure: a common and causative factor for expansion in human disease. Michel, B. Replication fork arrest and DNA recombination. Trends Biochem. Mirkin, S. Expandable DNA repeats and human disease. Nature , — Moxon, E. Adaptive evolution of highly mutable loci in pathogenic bacteria.
Principles and concepts of DNA replication in bacteria, archaea, and eukarya. Cold Spring Harb. Palud, A. Intrinsic properties of the two replicative DNA polymerases of Pyrococcus abyssi in replicating abasic sites: possible role in DNA damage tolerance? Pearson, C. Repeat instability: mechanisms of dynamic mutations. Perales, C. Raymann, K. Global phylogenomic analysis disentangles the complex evolutionary history of DNA replication in Archaea. Genome Biol. Richards, R. Dynamic mutations: a new class of mutations causing human disease.
Cell 70, — Rouillon, C. Samadashwily, G. Trinucleotide repeats affect DNA replication in vivo. Shah, S. Defective mismatch repair, microsatellite mutation bias, and variability in clinical cancer phenotypes.
Cancer Res. Sinden, R. Slipped strand DNA structures. Soengas, M. Tran, H. Replication slippage between distant short repeats in Saccharomyces cerevisiae depends on the direction of replication and the RAD50 and RAD52 genes. Trinh, T. The influence of primary and secondary DNA structure in deletion and duplication between direct repeats in Escherichia coli. Genetics , — Viguera, E. Replication slippage involves DNA polymerase pausing and dissociation.
Username Please enter your Username. Password Please enter your Password. Forgot password? Don't have an account? Sign in via your Institution. You could not be signed in, please check and try again. Sign in with your library card Please enter your library card number. Show Summary Details Overview replication slippage.
0コメント