Journal of Heredity 97 , — Emerson, J. Extensive gene traffic on the mammalian X chromosome. Science , — Fechotte, C. DNA transposons and the evolution of eukaryotic genomes.
Annual Review of Genetics 41 , — Gilbert, W. Origin of genes. Proceedings of the National Academy of Sciences 94 , — Hahn, M. Accelerated rate of gene gain and loss in primates.
Genetics , — a. Gene family evolution across 12 Drosophila genomes. PLoS Genetics 3 , e b. Harrison, P. Identification of pseudogenes in the Drosophila melanogaster genome.
Nucleic Acids Research 31 , — Transcribed processed pseudogenes in the human genome: An intermediate form of expressed retrosequence lacking protein-coding ability.
Nucleic Acids Research 33 , — He, X. Rapid subfunctionalization accompanied by prolonged and substantial neofunctionalization in duplicate gene evolution. Hotopp, J. Widespread lateral gene transfer from intracellular bacteria to multicellular eukaryotes. Jiang, N. Pack-MULE transposable elements mediate gene evolution in plants.
Jones, C. Parallel evolution of chimeric fusion genes. Lahn, B. The human Y chromosome, in the light of evolution. Nature Reviews Genetics 2 , — link to article. Levine, M. Novel genes derived from noncoding DNA in Drosophila melanogaster are frequently X-linked and exhibit testis-biased expression. Li, W. Evolutionary analyses of the human genome. Long, M. Natural selection and the origin of jingwei, a chimeric processed functional gene in Drosophila.
Science , 91—95 Lynch, M. The probability of duplicate gene preservation by subfunctionalization. Masly, J. Gene transposition as a cause of hybrid sterility in Drosophila. Morgante, M. Gene duplication and exon shuffling by helitron-like transposons generate intraspecies diversity in maize. Nature Genetics 37 , — link to article. Petrov, D. Pseudogene evolution and natural selection for a compact genome. Journal of Heredity 91 , — Presgraves, D. Evolutionary genomics: New genes for new jobs.
Current Biology 15 , R52—R53 Ranz, J. Principles of genome evolution in the Drosophila melanogaster species group. PLoS Biology 5 , e Roger, A.
Reconstructing early events in eukaryotic evolution. American Naturalist , S—S Rozen, S. Abundant gene conversion between arms of palindromes in human and ape Y chromosomes. Sasidharan, R. Genomics: Protein fossils live on as RNA.
Sebat, J. Large-scale copy number polymorphism in the human genome. Torgerson, D. Rapid evolution through gene duplication and subfunctionalization of the testes-specific alpha-4 proteasome subunits in Drosophila. Weber, M. Mammalian small nucleolar RNAs are mobile genetic elements. PLoS Genetics 2 , e Zhang, Z.
Comparative analysis of processed pseudogenes in the mouse and human genomes. Trends in Genetics 20 , 62—67 Millions of years of evolution preserved: A comprehensive catalog of the processed pseudogenes in the human genome. Genome Research 13 , — Origins of New Genes and Pseudogenes. Evolutionary Adaptation in the Human Lineage. Genetic Mutation. Negative Selection.
Interestingly, low-risk and high-risk patients can be stratified according to their overall survival using clinical variables alone. However, the low—medium risk group and the medium—high risk group can be distinguished only by the combination of clinical data and risk score based on the pseudogene expression levels.
These results represent a paradigmatic example of the added value that pseudogenes can bring to prognostic predictions. Finally, in the recently reported study by Ganapathi and colleagues 35 , 21 primary and 24 recurrent high-grade serous ovarian cancer HGSOC samples were subjected to RNA-seq analysis.
Twenty-one protein-coding genes and one non-coding gene were found to be differentially expressed between the two sample groups. Furthermore, low levels of SLC6A10P were found to be associated with longer time to progression, especially if considered together with high levels of one among the differentially expressed protein-coding genes COL2A1 This article offers a paradigmatic example of the ability of high-throughput techniques such as RNA-seq to point toward novel protein-coding and non-coding genes that not only contribute to our understanding of tumor pathogenesis but can also be exploited as useful biomarkers.
There are multiple lines of evidence in support of a causal link between altered pseudogene expression and the pathogenesis of human cancer 7 , First, pseudogenes often show cancer-specific deregulated expression, for example in the case of the OCT4 and NANOG pseudogenes, which are aberrantly expressed in cancer cells instead of their parental genes 7 , 11 , 18 , Second, similar to protein-coding genes, pseudogenes can undergo chromosomal rearrangements, amplification, deletion, and epigenetic silencing.
HMGIY pseudogenes have been shown to be affected by chromosomal rearrangements in benign human tumors PTENP1 , the processed pseudogene of PTEN phosphatase, is a paradigmatic example of an oncosuppressive pseudogene that is deleted or hypermethylated in human cancer 23 , 30 , 39 , Third, it has been reported that several variations in the sequence of pseudogenes such as those of PARP and CK2 can predispose to cancer 7.
Recently, a new and well-supported line of evidence for an oncogenic role of pseudogenes has been described. Cooke et al. Processed pseudogenes evolve from a retrotransposition event: the cDNA of the parental gene is retrotranscribed into DNA and is inserted randomly in the genome.
As a consequence, pseudogenes do not contain introns, are located in a different region of the genome, and are subjected to a different regulation compared to that of their parental genes 7. The authors analyzed genome sequencing data of cancer samples spanning 18 tumor types and identified 42 processed pseudogenes that are acquired somatically namely, present in the cancerous tissue and absent in the matched normal tissue. These pseudogenes are mostly derived from highly expressed transcripts by LINE retrotransposition and are most frequently found in two cancer types: non-small cell lung cancer and colorectal cancer.
In order to assess the consequences of pseudogenes retrotransposition on the expression of the pseudogenes themselves and of their host genes, the authors performed RNA sequencing on five samples in which they had identified 16 somatically acquired processed pseudogenes. Among them, 10 were found to have landed in intergenic regions, three in introns and three in exons.
The last group of pseudogenes was further analyzed in light of the possible deleterious consequences on the expression of the host genes. Even more strikingly, the authors report that the landing of PTPN12 pseudogene in the first exon of MGA , which is a likely oncosuppressor gene, causes an 8 kb deletion. This deletion spans the first exon itself, as well as the promoter region, and is ultimately responsible for the abrogation of MGA expression Because of their sequence similarity, the pseudogenes share multiple microRNA recognition elements MREs with their parental genes and can compete for the binding of common microRNA molecules Figure 1 D.
As a consequence, pseudogenes sustain the expression of their parental genes and hence can acquire oncogenic or oncosuppressive functions when deregulated.
For a comprehensive overview of ceRNA networks discovered in human cancer, both those that involve pseudogenes and other non-coding RNA classes, as well as those that involve protein-coding genes, please refer to Ref. Since the discovery of PTENP1 , the list of oncogenic and oncosuppressive pseudogenes that act as ceRNAs for their parental genes has been greatly expanded those that have been characterized up to July are listed in Table 1. Table 1. Pseudogenes that function as ceRNAs for their parental genes or other genes in cancer.
Additionally, it has been shown that pseudogenes can act as ceRNAs not only for their parental genes but also for other genes Figure 1 K and Table 1. This is because the microRNAs that they share with their parental genes have also other targets and because they can be targeted by additional microRNAs. Unitary pseudogenes represent an ultimate example of the ability of pseudogenes to exert ceRNA-based functions that are parental gene independent. The pseudogenes belonging to this class do not have parental counterparts because they derive from the progressive acquisition of mutations in protein-coding genes 7.
Interestingly, Marques et al. Furthermore, it has been shown that the loss or dysregulation of such networks is among the hallmarks of cancer cells compared to normal cells Besides the lines of evidence described in Section 5, formal proof of the causal link existing between pseudogenes and the pathogenesis of human cancer has recently come from genetically engineered mouse models 49 , The article by Karreth et al.
This observation prompted the authors to evaluate if the pseudogene functions as a ceRNA. Indeed, they found that the overexpression of Braf-rs1 in mouse NIH3T3 cells causes the up-regulation of Braf, the hyperactivation of the Erk pathway and, as a consequence, an increase in cell growth. Furthermore, the authors show that these effects are abolished when Braf-rs1 is overexpressed in mouse cells that lack Dicer or that lack the parental gene Braf KO cells , which suggests that the availability of mature microRNAs and the expression of Braf are required by Braf-rs1 in order to exert its activity.
Next, the authors sought to investigate the consequences of Braf-rs1 overexpression in vivo. This line was crossed with the CAG-rtTA line, and the compound transgenic animals were placed on a dox-containing diet at 3 weeks of age.
The effects of the induction of Braf-rs1 expression were dramatic: after approximately 4 months of treatment, the mice started to die and their median survival was as short as days. Moribund mice were characterized by splenomegaly and enlarged lymph nodes, all symptoms that an in-depth flow cytometric analysis revealed to be the result of an aggressive form of diffuse large B-cell lymphoma DLBCL.
Braf-rs1 -induced lymphomas were further characterized in multiple ways. Furthermore, if doxycycline was removed once the splenomegaly became apparent, the lymphomas largely regressed, indicating that the tumors are markedly addicted to Braf-rs1 , and its aberrant expression is required for tumor maintenance. Second, histological analyses proved that, consistent with the ceRNA hypothesis, Braf-rs1 -driven tumors are indeed characterized by increased Braf and pErk levels.
Furthermore, when transplanted NSG mice were treated with a Mek inhibitor, a marked impairment of the ability of lymphoma cells to infiltrate other organs was observed, which indicates that the tumors are addicted to Braf-rs1 because they are addicted to the Erk pathway.
On the contrary, the overexpression of Braf-rs1 coding sequence does not cause a marked increase in Braf levels and induces a milder phenotype.
These results provide further confirmation that Braf-rs1 is a non-coding ceRNA that exerts an oncogenic function by increasing Braf levels. These genetically engineered mouse models represent a proof of principle that aberrantly expressed pseudogenes are necessary and sufficient to cause cancer by working as ceRNAs for their oncogenic parental genes. Furthermore, even if the study is based on the mouse Braf pseudogene, it is of relevance to human cancer for two reasons.
On the one hand, in the same study, Karreth et al. Together these results support the notion that, analogously to PTEN tumor suppressor 41 , also BRAF , with its transcript variants and its pseudogenes, is under tight microRNA-mediated regulation and is likely involved in complex ceRNA-based networks. This in turn suggests that at the basis of the oncogenic potential of such a powerful protein kinase, there might be not only the widely known and extensively studied mutation at the Val residue but also an aberrant gene expression Thus, in order for a personalized treatment to be effective, it is crucial to achieve a detailed classification of the cancer genome and epigenome.
To this end, a classification solely based on the tissue of origin and on pathological features has shown its limitations. Conversely, large-scale projects that have merged the output of multiple omics techniques whole-exome sequencing, DNA copy number variations, DNA methylation, mRNA-seq, microRNA-seq, and proteomics have shown the power of a classification based on molecular alterations.
Such an approach allows the subdivision of virtually every cancer type into multiple subtypes and offers guidance in the treatment of each patient with the drug or drug combination that has the highest chance to be effective. Furthermore, it shows that molecular level similarities exist among cancer types of different tissue of origin and offers the rationale for proposing the use of non-standard therapeutic strategies 5 , The possibility to exploit such accurate markers of cell identity could represent a crucial refinement and a further step toward an effective personalized medicine.
If this is the case, the rescue of the pseudogenome from the genetic junk will be complete and hopefully will pave the way for the establishment of the causal link existing between other classes of non-coding RNAs and human cancer.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. We thank L. Fawls and K. Doherty for editorial assistance and A. Tuccoli for assistance with the figures. Landscape of transcription in human cells. Nature —8. A draft map of the human proteome. Nature — The emerging role of pseudogene expressed non-coding RNAs in cellular functions.
Int J Biochem Cell Biol 54 —5. Milligan MJ, Lipovich L. Pseudogene-derived lncRNAs: emerging regulators of gene expression.
Front Genet 5 About one year ago, a comprehensive paper on polymorphism among PPs in human beings appeared. Ewing et al. Using discordant reads not present in reference genomes, they found 48 novel PP insertion sites among low pass genomes from the 1, genomes project [ 22 ]. These PPs came from a wide variety of source genes, and were spread throughout the human chromosomes Figure 1. All 48 of these polymorphic PPs were confirmed by locating the precise genomic insertion site.
This group also studied the genome sequences of 85 human cancer-normal tissue pairs representing a variety of cancers. Among these cancers they found the first instances of somatic insertion of PPs; three PPs were predicted to occur in lung cancers that were absent from paired normal tissue. Locations of 48 non-reference gene processed pseudogene insertions sites in the human genome based on reads mapped to source genes. Discordant read mappings are represented by links colored based on chromosome of the source gene.
Insertion sites are represented by black circles and the gene labels are based on the position of the source gene. Republished with permission from Nature Communications. Among these, Mus musculus castaneus , M.
However, on average, each of the 12 inbred strains derived from C57Bl6 were genetically closer, but still differed from one another by 68 PPs on average. The much greater number of polymorphic PPs in mouse strains compared to individual human beings may be due to the much larger number of active L1s present in the mouse approximately 3, versus approximately in humans [ 23 , 24 ].
This paper represented the first comprehensive look at the question of PP insertions in humans, mice and chimpanzees, and the first study of somatic insertion of PPs in cancer. Two other papers demonstrating polymorphism of PPs in humans have now appeared. Using exon-exon junction spanning reads, Abyzov et al. Thirty-six of these were confirmed as polymorphic in humans by detection of the genomic insertion point. Interestingly, the parental genes of non-reference PPs were significantly enriched among genes expressed at the M-to-G1 transition in the cell cycle.
Schrider et al. They found 21 PPs not present in the reference genome and presumably polymorphic; 17 of these 21 were confirmed by PCR See [ 27 ] for a recent review of these papers.
Recently, Cooke et al. They analyzed cancer-normal pairs of sequenced samples at Wellcome Trust representing a variety of different cancers. In 17 or 2. The authors noted the presence of five PPs in non-small cell lung cancer among 27 cancers studied, similar to the Ewing et al. Additionally, they found two PPs in eleven colorectal cancer samples. The PP insertions in cancer were thoroughly characterized and all had the molecular signatures of germ line L1 insertions.
In a lung adenocarcinoma, one insertion was associated with an 8 kb deletion of the promoter and exon 1 of a tumor suppressor gene, MGA1. The deletion knocked out expression of that allele as determined by RNA-seq. Among the PPs in cancer, most were derived from highly expressed transcripts, yet many were not. In addition, many PP insertions appeared to be early events in tumor formation, being present in an early lesion along with the tumor or in multiple sections of the same tumor.
However, some PP insertions were shown to be later events in tumor progression because they were not detected in all sections of the same tumor. A final paper nailed down the potential for PP formation during early development in humans. This paper by de Boer et al. This man, now a young adult, had suffered from multiple bouts of pulmonary aspergillosis as a child. There are three interesting aspects of this case.
A PP had not been observed previously as a new insertion among previous insertions L1, Alu, SVA in human Mendelian disease or cancer etiology [ 32 ]. After insertion of this semi-processed pseudogene in reverse orientation into intron 1 of CYBB , splicing had occurred into an excellent acceptor splice site and out of an excellent donor site in exon 2 of TMF1.
The newly created bp exon also contained a nonsense codon that caused the CYBB gene to be non-functional Figure 2. Republished with permission from Human Mutation published by Wiley.
To date, somatic retrotransposition in Mendelian disease has been rarely found. Among the cases mentioned above, there is only a somatic insertion into the adenomatous polyposis coli APC tumor suppressor gene in a colorectal cancer case [ 33 ] and somatic and germ line mosaicism in the mother of a patient with the X-linked disease, choroideremia [ 34 ].
0コメント