Research Article

Systematic discovery of cap-independent translation sequences in human and viral genomes

See allHide authors and affiliations

Science  15 Jan 2016:
Vol. 351, Issue 6270, aad4939
DOI: 10.1126/science.aad4939

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

Identifying the IRESs of humans and viruses

Most proteins result from the translation of 5′ capped RNA transcripts. In viruses and a subset of human genes, RNA transcripts with internal ribosome entry sites (IRESs) are uncapped. Weingarten-Gabbay et al. systematically surveyed the presence of IRESs in human protein-coding transcripts, as well those of viruses (see the Perspective by Gebauer and Hentze). Large-scale mutagenesis profiling identified two classes of IRESs: those having a functional element localized to one small region of the IRES and those with important elements distributed across the entire region. An unbiased screen across human genes suggests that IRESs are more frequent than previously supposed in 3′ untranslated regions.

Science, this issue p.10.1126/science.aad4939; see also p. 228

Structured Abstract


The recruitment of the ribosome to a specific mRNA is a critical step in the production of proteins in cells. In addition to a general recognition of the “cap” structure at the beginning of eukaryotic mRNAs, ribosomes can also initiate translation from a regulatory RNA element termed internal ribosome entry site (IRES) in a cap-independent manner. IRESs are essential for the synthesis of many human and viral proteins and take part in a variety of biological functions, such as viral infections, the response of cells to stress, and organismal development. Despite their importance, we lack systematic methods for discovering and characterizing IRESs, and thus, little is known about their position in the human and viral genomes and the mechanisms by which they recruit the ribosome.


Our method enables accurate measurement of thousands of fully designed sequences for cap-independent translation activity. By using a synthetic oligonucleotide library, we can determine the exact composition of the sequences tested and can profile sequences from hundreds of different viruses, as well as the human genome, in a single experiment. In addition, synthetic design enables the construction of oligos in which we carefully and systematically mutate native IRESs and measure the effect of these mutations on expression. This reverse-genetics approach enables the characterization of the regulatory elements that recruit the ribosome and provide specificity in translation.


We uncover thousands of human and viral sequences with cap-independent translation activity, which provide a 50-fold increase in the number of sequences known to date. Unbiased screening of cap-independent activity across human transcripts demonstrates enrichment of regulatory elements in the untranslated region in the beginning of transcripts (5′UTR). However, we also find enrichment in the untranslated region located downstream of the coding sequence (3′UTR), which suggests a mechanism by which ribosomes are recruited to the 3′UTR to enhance the translation of an upstream sequence. A genome-wide profiling of positive-strand RNA viruses ([+]ssRNA) reveals the existence of translational elements along their coding regions. This finding suggests that [+]ssRNA viruses can translate only part of their genome, in addition to the synthesis and cleavage of a premature polyprotein. Our analysis reveals two classes of functional elements that drive cap-independent translation: (i) highly structured elements and (ii) unstructured elements that act through a short sequence motif. We show that many 5′UTRs can attract the ribosome by Watson-Crick base pairing with the 18S ribosomal RNA, a structural RNA component of the small ribosomal subunit (40S). In addition, we systematically investigate the functional regions of the 18S rRNA involved in these interactions that enhance cap-independent translation.


These results reveal the wide existence of cap-independent translation sequences in both humans and viruses. They provide insights on the landscape of translational regulation and uncover the regulatory elements underlying cap-independent translation activity.

High-throughput bicistronic assay provides insights on translational regulation in human and viruses.

(A) A library of thousands designed oligonucleotides as synthesized and cloned into a bicistronic reporter. Measurements of eGFP production, representing cap-independent translation activity, were performed with fluorescence-activated cell sorting and deep sequencing (FACS-seq). (B) The landscape of cap-independent translation sequences in human and viruses and the identified cis-regulatory elements driving their activity.


To investigate gene specificity at the level of translation in both the human genome and viruses, we devised a high-throughput bicistronic assay to quantify cap-independent translation. We uncovered thousands of novel cap-independent translation sequences, and we provide insights on the landscape of translational regulation in both humans and viruses. We find extensive translational elements in the 3′ untranslated region of human transcripts and the polyprotein region of uncapped RNA viruses. Through the characterization of regulatory elements underlying cap-independent translation activity, we identify potential mechanisms of secondary structure, short sequence motif, and base pairing with the 18S ribosomal RNA (rRNA). Furthermore, we systematically map the 18S rRNA regions for which reverse complementarity enhances translation. Thus, we make available insights into the mechanisms of translational control in humans and viruses.

View Full Text