Evolution of viral codon usage preferences:manipulation of translation accuracy and evasion of immune response

Project Information

CODOVIREVOL

Grant agreement ID: 647916

DOI

10.3030/647916

Project closed

EC signature date 9 December 2015

Start date 1 January 2016

End date 30 June 2022

Funded under

• EXCELLENT SCIENCE - European Research Council (ERC)

Total cost

€ 1 997 100,00

EU contribution

€ 1 997 100,00

Coordinated by

CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRS
France

Periodic Reporting for period 5 - CODOVIREVOL (Evolution of viral codon usage preferences:manipulation of translation accuracy and evasion of immune response)

Reporting period: 2022-01-01 to 2022-06-30

Summary of the context and overall objectives of the project

The problem. Biological instructions for building an organism are encoded in the genome. Genes are translated into proteins by specialised molecular machines, the ribosomes. Translation implies associating a block of three letters of the genomic message (called a codon) with a single protein building block (an amino acid). One amino acid is often associated to several codons, and different organisms have different codon usage preferences for the same amino acid. A large proportion of viruses infecting humans display a poor adaptation to the codon usage preferences of their host. This observation is a paradox, as viral genes completely depend upon the cellular translation machinery for protein synthesis. The poor match between codon usage preferences of virus to the host negatively affects speed and accuracy of viral protein translation.

The hypothesis. We propose that such apparent mismatch of viral to host codon usage preferences may provide with an adaptive value to the virus, as it reduces protein synthesis, decreases translational fidelity and provides globally with way to escape immune surveillance. Such strategy would be especially beneficial for viruses with a chronic lifestyle.

The importance. We use papillomaviruses as a model system. They are a large family of viruses with a small genome, with many members infecting humans, and with very different lifestyles, from self-limiting acute infections to chronic infections with malignant potential. Human papillomaviruses cause around 5% of all human cancers worldwide. They are also responsible for warts and genital warts, whose treatment imposes an important personal and societal burden. Our hypothesis aims to understand the differential adaptive strategies of papillomaviruses causing acute and chronic infections.

The objectives. We aim at understanding the impact of codon usage preferences on the quality and quantity of protein synthesised from a viral transcript. Our results will help solve the evolutionary puzzle of codon usage bias for viruses causing chronic infections, and will have implications for the development of mathematical models of virus-host interaction, and for the design of therapeutic vaccines to guide the immune response towards the identification and targeting of the main protein species.

The conclusion. We have shown that differences in codon usage bias strongly impact the molecular and cellular phenotype: (i) they result in large differences in mRNA and in protein levels, as well in mRNA-to-protein ratio; (ii) they introduce unpredicted splicing events; (iii) they lead to reproducible phenotypic heterogeneity; and (iv) they lead to a trade-off between the benefit of antibiotic resistance and the burden of heterologous expression.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

We have first analysed the impact of codon usage preferences in key amino acids involved in enzyme reactions on the overall enzyme performance. We have synthesised several synonymous versions of a secreted phosphatase gene and quantified phosphatase activity and mRNA transcripts. We observe that the codon usage preferences in the enzyme active centre have an impact on enzyme activity, beyond variations on the mRNA concentration. These results show the importance of differential translation quality associated to different synonymous codons.

We have then analysed the impact of codon usage preferences in a full protein, using the protein that confers resistance to the antibiotic bleomycine. We have synthesised several synonymous versions of this gene and transfected them into human cells. For each version we have monitored DNA levels, mRNA levels, protein levels, cellular growth and cytometry cell phenotype. We are analysing all this information aiming to understand the different filters associated to codon usage preferences that drive the connection between genotype-phenotype. We observe enormous differences in cellular phenotype associated to specific codon usage preferences, often in non-anticipated directions.

We have also launched experimental evolution and competition experiments using these synonymous versions of the bleomycine resistance gene, trying to evaluate variations in cellular fitness associated to variations in protein fitness.

We have analysed the impact of codon usage preferences on viral oncogenes. We are discovering gene expression patterns that are related to codon usage preferences and that may be major players in the large differences in phenotype associated to very closely related viral genotypes.

We have finally studied the molecular evolution of these oncogenic viruses, trying to understand the large differences in oncogenic potential that are not associated to specific viral lineages. We have discovered four novel papillomaviruses in animals and we have proposed for the first time an evolutionary hypothesis explaining the exacerbated oncogenic potential of HPV16, the most oncogenic human papillomavirus, which appeals to a transfer between human species during the interbreeding between modern and archaic humans.

We have communicated the results produced in this project in a number of articles in open access, and we have released the raw data for reuse in open repositories, such as zeonodo.org in the case of phylogenetic analyses, the GenBank in the case of RNASeq data and PRIDE in the case of protein data.

We have also integrated the tool COUSIN into a web-server for providing the research community with access to the tool (https://cousin.ird.fr).

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

We have generated the, to our knowledge, largest combination of genotypic and phenotypic information about the impact of codon usage preferences in human cells at several levels, spanning genomic, transcriptomic, proteomic and cell viability and growth data, immediately upon transfection and after experimental evolution and lineage competition. Only this dataset will allow us understand the implications of gene codon usage preferences on the connection between genotype and phenotype, on differential gene evolvability and on differential protein robustness.

Finally we have built on the information and knowledge generated in vitro at the cellular level to extend our research to organism-based level. We will have thus replaced this approach as far as possible for all questions and hypotheses that do not imply the local immune response and the virus-host interaction during chronic infection. We will generate by this approach invaluable, clinically relevant information about the differential progression, potential for malignisation and response to repeated infections depending on codon usage preferences of the viral oncogenes.

By exploring the transcription and translation outcomes from a broad diversity of synonymous sequences, we have been able to show that codon usage preferences strongly impact gene expression in human cells by the introduction of non-predicted splicing events, differentially promoting translation reinitiation upon termination and leaky scanning, it all leading to an increased diversity of transcription and translation products, resulting in an increased phenotypic diversity.

Our results will help disentangle the evolutionary forces underlying the largely divergent clinical, phenotypic presentation of the infections. Our ultimate aim is to understand why most papillomaviruses cause asymptomatic infections, some of them cause bening, wart-like lesions, and only a few are associated to human cancers.

graphic-summary-copy.png