Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Initiation of translation in bacteria by a structured eukaryotic IRES RNA


The central dogma of gene expression (DNA to RNA to protein) is universal, but in different domains of life there are fundamental mechanistic differences within this pathway. For example, the canonical molecular signals used to initiate protein synthesis in bacteria and eukaryotes are mutually exclusive1. However, the core structures and conformational dynamics of ribosomes that are responsible for the translation steps that take place after initiation are ancient and conserved across the domains of life2. We wanted to explore whether an undiscovered RNA-based signal might be able to use these conserved features, bypassing mechanisms specific to each domain of life, and initiate protein synthesis in both bacteria and eukaryotes. Although structured internal ribosome entry site (IRES) RNAs can manipulate ribosomes to initiate translation in eukaryotic cells, an analogous RNA structure-based mechanism has not been observed in bacteria. Here we report our discovery that a eukaryotic viral IRES can initiate translation in live bacteria. We solved the crystal structure of this IRES bound to a bacterial ribosome to 3.8 Å resolution, revealing that despite differences between bacterial and eukaryotic ribosomes this IRES binds directly to both and occupies the space normally used by transfer RNAs. Initiation in both bacteria and eukaryotes depends on the structure of the IRES RNA, but in bacteria this RNA uses a different mechanism that includes a form of ribosome repositioning after initial recruitment. This IRES RNA bridges billions of years of evolutionary divergence and provides an example of an RNA structure-based translation initiation signal capable of operating in two domains of life.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Translation initiation assays in bacteria.
Figure 2: IRES•70S ribosome structure.
Figure 3: Importance of IRES structure and ribosome binding.
Figure 4: Location of initiation on an IGR IRES in bacteria.

Accession codes

Primary accessions

Protein Data Bank

Data deposits

Atomic coordinates and structure factor amplitudes have been deposited in the Protein Data Bank under accession number 4XEJ.


  1. Hershey, J. W., Sonenberg, N. & Mathews, M. B. Principles of translational control: an overview. Cold Spring Harb. Perspect. Biol. 4, a011528 (2012)

    Article  Google Scholar 

  2. Melnikov, S. et al. One core, two shells: bacterial and eukaryotic ribosomes. . Nature Struct. Mol. Biol. 19, 560–567 (2012)

    CAS  Article  Google Scholar 

  3. Malys, N. & McCarthy, J. E. Translation initiation: variations in the mechanism can be anticipated. Cell. Mol. Life Sci. 68, 991–1003 (2011)

    CAS  Article  Google Scholar 

  4. Moll, I., Grill, S., Gualerzi, C. O. & Blasi, U. Leaderless mRNAs in bacteria: surprises in ribosomal recruitment and translational control. Mol. Micr. obiol. 43, 239–246 (2002)

    CAS  Article  Google Scholar 

  5. Berthelot, F., Bogdanovsky, D., Schapira, G. & Gros, F. Interchangeability of factors and tRNA’s in bacterial and eukaryotic translation initiation systems. Mol. Cell. Biochem. 1, 63–72 (1973)

    CAS  Article  Google Scholar 

  6. Hertz, M. I. & Thompson, S. R. Mechanism of translation initiation by Dicistroviridae IGR IRESs. Virology 411, 355–361 (2011)

    CAS  Article  Google Scholar 

  7. Schüler, M. et al. Structure of the ribosome-bound cricket paralysis virus IRES RNA. Nature Struct. Mol. Biol. 13, 1092–1096 (2006)

    Article  Google Scholar 

  8. Spahn, C. M. et al. Cryo-EM visualization of a viral internal ribosome entry site bound to human ribosomes: the IRES functions as an RNA-based translation factor. Cell 118, 465–475 (2004)

    CAS  Article  Google Scholar 

  9. Pestova, T. V., Lomakin, I. B. & Hellen, C. U. Position of the CrPV IRES on the 40S subunit and factor dependence of IRES/80S ribosome assembly. EMBO Rep. 5, 906–913 (2004)

    CAS  Article  Google Scholar 

  10. Jan, E. & Sarnow, P. Factorless ribosome assembly on the internal ribosome entry site of cricket paralysis virus. J. Mol. Biol. 324, 889–902 (2002)

    CAS  Article  Google Scholar 

  11. Nishiyama, T. et al. Structural elements in the internal ribosome entry site of Plautia stali intestine virus responsible for binding with ribosomes. Nucleic Acids Res. 31, 2434–2442 (2003)

    CAS  Article  Google Scholar 

  12. Jan, E. et al. Initiator Met-tRNA-independent translation mediated by an internal ribosome entry site element in cricket paralysis virus-like insect viruses. Cold Spring Harb. Symp. Quant. Biol. 66, 285–292 (2001)

    CAS  Article  Google Scholar 

  13. Thompson, S. R., Gulyas, K. D. & Sarnow, P. Internal initiation in Saccharomyces cerevisiae mediated by an initiator tRNA/eIF2-independent internal ribosome entry site element. Proc. Natl Acad. Sci. USA 98, 12972–12977 (2001)

    ADS  CAS  Article  Google Scholar 

  14. Wilson, J. E., Pestova, T. V., Hellen, C. U. & Sarnow, P. Initiation of protein synthesis from the A site of the ribosome. Cell 102, 511–520 (2000)

    CAS  Article  Google Scholar 

  15. Sasaki, J. & Nakashima, N. Methionine-independent initiation of translation in the capsid protein of an insect RNA virus. Proc. Natl Acad. Sci. USA 97, 1512–1515 (2000)

    ADS  CAS  Article  Google Scholar 

  16. Sasaki, J. & Nakashima, N. Translation initiation at the CUU codon is mediated by the internal ribosome entry site of an insect picorna-like virus in vitro. J. Virol. 73, 1219–1226 (1999)

    CAS  Article  Google Scholar 

  17. Costantino, D. A., Pfingsten, J. S., Rambo, R. P. & Kieft, J. S. tRNA–mRNA mimicry drives translation initiation from a viral IRES. Nature Struct. Mol. Biol. 15, 57–64 (2008)

    CAS  Article  Google Scholar 

  18. Jan, E., Kinzy, T. G. & Sarnow, P. Divergent tRNA-like element supports initiation, elongation, and termination of protein biosynthesis. Proc. Natl Acad. Sci. USA 100, 15410–15415 (2003)

    ADS  CAS  Article  Google Scholar 

  19. Zhu, J. et al. Crystal structures of complexes containing domains from two viral internal ribosome entry site (IRES) RNAs bound to the 70S ribosome. Proc. Natl Acad. Sci. USA 108, 1839–1844 (2011)

    ADS  CAS  Article  Google Scholar 

  20. Hellen, C. U. IRES-induced conformational changes in the ribosome and the mechanism of translation initiation by internal ribosomal entry. Biochim. Biophys. Acta 1789, 558–570 (2009)

    CAS  Article  Google Scholar 

  21. Yamamoto, H., Nakashima, N., Ikeda, Y. & Uchiumi, T. Binding mode of the first aminoacyl-tRNA in translation initiation mediated by Plautia stali intestine virus internal ribosome entry site. J. Biol. Chem. 282, 7770–7776 (2007)

    CAS  Article  Google Scholar 

  22. Fernández, I. S., Bai, X. C., Murshudov, G., Scheres, S. H. & Ramakrishnan, V. Initiation of translation by cricket paralysis virus IRES requires its translocation in the ribosome. Cell 157, 823–831 (2014)

    Article  Google Scholar 

  23. Deniz, N., Lenarcic, E. M., Landry, D. M. & Thompson, S. R. Translation initiation factors are not required for Dicistroviridae IRES function in vivo. RNA 15, 932–946 (2009)

    CAS  Article  Google Scholar 

  24. Stoneley, M., Paulin, F. E., Le Quesne, J. P., Chappell, S. A. & Willis, A. E. c-Myc 5′ untranslated region contains an internal ribosome entry segment. Oncogene 16, 423–428 (1998)

    CAS  Article  Google Scholar 

  25. Costantino, D. & Kieft, J. S. A preformed compact ribosome-binding domain in the cricket paralysis-like virus IRES RNAs. RNA 11, 332–343 (2005)

    CAS  Article  Google Scholar 

  26. Pfingsten, J. S., Costantino, D. A. & Kieft, J. S. Structural basis for ribosome recruitment and manipulation by a viral IRES RNA. Science 314, 1450–1454 (2006)

    ADS  CAS  Article  Google Scholar 

  27. Koh, C. S., Brilot, A. F., Grigorieff, N. & Korostelev, A. A. Taura syndrome virus IRES initiates translation by binding its tRNA-mRNA-like structural element in the ribosomal decoding center. Proc. Natl Acad. Sci. USA 111, 9139–9144 (2014)

    ADS  CAS  Article  Google Scholar 

  28. Nishiyama, T., Yamamoto, H., Uchiumi, T. & Nakashima, N. Eukaryotic ribosomal protein RPS25 interacts with the conserved loop region in a dicistroviral intergenic internal ribosome entry site. Nucleic Acids Res. 35, 1514–1521 (2007)

    CAS  Article  Google Scholar 

  29. Landry, D. M., Hertz, M. I. & Thompson, S. R. RPS25 is essential for translation initiation by the Dicistroviridae and hepatitis C viral IRESs. Genes Dev. 23, 2753–2764 (2009)

    CAS  Article  Google Scholar 

  30. Ben-Shem, A. et al. The structure of the eukaryotic ribosome at 3.0 Å resolution. Science 334, 1524–1529 (2011)

    ADS  CAS  Article  Google Scholar 

  31. Kabsch, W. Automatic processing of rotation diffraction data from crystals of initially unknown symmetry and cell constants. J. Appl. Crystallogr. 26, 795–800 (1993)

    CAS  Article  Google Scholar 

  32. Adams, P. D. et al. PHENIX: building new software for automated crystallographic structure determination. Acta Crystallogr. D 58, 1948–1954 (2002)

    Article  Google Scholar 

  33. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D 60, 2126–2132 (2004)

    Article  Google Scholar 

  34. DeLano, W. L. The PyMOL Molecular Graphics System (DeLano Scientific, 2002)

    Google Scholar 

  35. Karplus, P. A. & Diederichs, K. Linking crystallographic model and data quality. Science 336, 1030–1033 (2012)

    ADS  CAS  Article  Google Scholar 

Download references


We thank the members of the Kieft laboratory for insight and discussion and the staff at the Advanced Photon Source for their support. The original PSIV IGR IRES-containing plasmid was from N. Nakashima and the source of the luciferase genes was a plasmid from A. Willis. This work was supported by grants GM-17129 and GM-59140 from the National Institutes of Health (NIH) and MCB-723300 from the National Science Foundation (to H.F.N.), grant GM-103105 from the NIH (to A.A.K.), and grants GM-97333 and GM-81346 from the NIH (to J.S.K.). J.S.K. is an Early Career Scientist of the Howard Hughes Medical Institute. T.-D.M.P. was an American Heart Association Predoctoral Scholar (10PRE260143).

Author information

Authors and Affiliations



T.M.C. and J.S.K. designed the experiments and the constructs tested. T.M.C. and D.A.C. conducted the bacterial functional assays. Clones were generated by T.M.C., T.-D.M.P. and Z.A.J. J.S.K. performed the ribosome association assays. Ribosomes were purified, crystals grown, and the structure solved by J.P.D., J.Z. and A.A.K. under the supervision of H.F.N. J.S.K. provided overall supervision and guidance, and together with T.M.C. and D.A.C. wrote the manuscript with input from all authors.

Corresponding author

Correspondence to Jeffrey S. Kieft.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 Canonical translation initiation signals, characteristics of IGR IRESs, and experimental design.

a, Bacterial mRNAs (left) use a Shine–Dalgarno sequence (SDS; red) upstream of the AUG start codon and open reading frame (green) to recruit the 30S subunit (grey). The interaction is through the anti-SDS (A-SDS, yellow). Three initiation factors (magenta) are also important. Eukaryotic mRNAs (right) have a 5′ 7-methyl-guanosine ‘cap’ (red; 7 mG) that is bound by initiation factor 4E (4E, yellow). Multiple initiation factors (blue and magenta) serve to recruit the 40S subunit (grey) and allow it to scan to the start codon. b, Left, cryo-electron microscopy (cryo-EM) reconstruction of an IGR IRES RNA (magenta) bound to a human 40S subunit (yellow)8. The compact structure occupies the tRNA-binding groove of the subunit. Right, cryo-EM reconstruction of an IGR IRES RNA (magenta) bound to a human 80S ribosome8. The 40S subunit is yellow and the 60S subunit is cyan. The IRES RNA occupies the conserved intersubunit space. c, Cartoon representation of the secondary structure of a type 1 IGR IRES RNA (the type to which PSIV belongs). This structure is found between two open reading frames within the viral RNA genome. The two independently folded domains (domain 1+2 and domain 3) are indicated with dashed grey ovals. The locations of two pseudoknot interactions critical for inducing the correct IRES folded structure, and thus for function (PK 1 and PK 2), are shown. d, The structured IRES studied here is found in the intergenic region of the viral genome (red). It was placed into a dual-luciferase (LUC) reporter construct (blue, RLUC; yellow, FLUC) and this was cloned into bacterial expression vector pET30a. This vector was used to transform Escherichia coli. Induction of the culture leads to expression of the dual-LUC mRNA. Aliquots of the culture were harvested at defined time points and the amount of each LUC was measured. These data were used to determine the initial rate of LUC production (generally linear over the first 30–40 min post-induction) for each of the two reporters. RLUC served as a consistent internal control for different bacterial cultures, clones, growth rates, and so on.

Extended Data Figure 2 Verification of independent quantifiable LUC production in bacteria.

a, An empty pET30a vector (no inserted LUC reporter coding sequences) shows negligible signal. b, Traces of LUC activity as a function of time are shown from a construct in which the RLUC reporter was driven by the SDS and enhancer sequence from pET30a and FLUC was driven by an SDS only (Downstream SDS). The red octagon denotes stop codons. Both LUCs are generated, and RLUC production is higher, as expected. c, Removal of the SDS driving FLUC production (Downstream SDS_K/O) results in a loss of FLUC production, as expected. d, Insertion of the IRES from classical swine fever virus (CSFV) in a position to drive initiation of FLUC results in negligible FLUC activity. ad, The y-axis indicates relative light units (RLU). Error bars represent 1 s.d. from the mean from three biological replicates. Here and throughout this study, we observed different LUC versus time profiles with different constructs. For example, the RLUC traces for the Downstream SDS and Downstream SDS_K/O constructs are different, despite no change to the SDS driving RLUC production (one shows a decrease of RLUC in later time points, the other maintains RLUC levels). The reason for this effect is unknown, but it only appears 60 min after induction. e, Despite differences in longer time courses, LUC production was consistent and linear over the first 30–40 min post-induction. The RLUC and FLUC traces from the Downstream SDS and Downstream SDS_K/O constructs are shown. The consistency of these initial rates, before high levels of mRNA and reporter might build up and affect bacterial behaviour, justified their use as a means to quantitate LUC production (Extended Data Fig. 3).

Extended Data Figure 3 Determination of IRES activity from initial rates of LUC production.

a, Representative graphs of RLUC and FLUC levels at early time points from three cultures of bacteria transformed with an IRES-containing bicistronic vector, induced with isopropyl-β-d-thiogalactoside (IPTG) at time = 0. Data from the three cultures are shown as black, green, and blue points, and a linear fit is shown with a dashed line for each. The slopes of these fit lines were used as the initial rate of LUC production per minute. b, Representative table of data for one IRES construct. Data from six cultures are shown, with initial rates for RLUC and FLUC production in RLU min−1. Throughout this manuscript, the average rate for each LUC is shown in blue (RLUC) and yellow (FLUC) bar graphs. The ratio of these rates was determined from each culture, and these were averaged and shown in green bar graphs throughout the manuscript.

Extended Data Figure 4 Examination of leaky expression and cryptic promoter activity.

a, Traces of LUC production from the wild-type PSIV IRES-containing construct without induction with IPTG. Both RLUC and FLUC are produced due to ‘leaky expression’ of mRNA, a common observation with pET30a bacterial expression vectors. af, The y-axis shows RLU. b, Examination of the early time points of the traces from panel 1 show that both RLUC and FLUC are expressed to a low level without induction, and thus this leaky expression is not due to the IRES. c, Traces of wild-type PSIV IRES with IPTG induction at time = 0 (grey dashed line), showing the increase due to induction. d, Traces of a construct with the RLUC-driving SDS knocked out (Upstream SDS_K/O, same as in Fig. 1b), shown for comparison. e, To check for cryptic promoter activity due to transcription from a site other than the authentic T7 promoter, we cloned the full IRES-containing dual-LUC cassette into a pET30a vector in which the T7 promoter was mutated from 5′-TAATACGACTCACTATA-3′ to 5′-TAATGGTGTCTGAATTC-3′ (T7_K/O). Both RLUC and FLUC are produced to low levels, indicating some T7 promoter-independent expression exists in this vector, but the initial rates of producing upon induction are very low (see f and g). f, Initial rates of production of FLUC from the T7_K/O (induced), wild-type (uninduced), wild-type (induced), and Upstream SDS_K/O (induced) constructs. Rates of FLUC production from the T7_K/O and uninduced wild type are very low and not sufficient to account for apparent initiation from the IRES upon induction. This graph also illustrates the importance and utility of using the initial rates of LUC production for analysis, rather than the entire curve or an arbitrary later time point. g, Quantitated and graphed initial rate data for the four constructs in this figure. Error bars represent 1 s.d. of the mean from three biological replicates, except the uninduced control, which was done once.

Extended Data Figure 5 PSIV IGR IRES sequence, secondary structure, and design of mutants.

a, Secondary structure of the full-length IGR IRES from the PSIV. The specific changes that were introduced to generate the mutants and constructs described and tested in the main text are shown. For each, the altered region is boxed and the change is shown in red. For the uAUG and uSTOP constructs, the start and stop codons are underlined. RLUC and FLUC coding sequences are boxed cyan and yellow, respectively. b, Constructs without the IRES that contain various wild-type or mutant SDS and SDS-like sequences upstream of the FLUC open reading frame. c, Construct containing just domain 3 of the PSIV IGR IRES.

Extended Data Figure 6 Contributions of region upstream of AUG to initiation activity.

a, Diagram of constructs tested and traces of FLUC and RLUC production. The y-axis shows RLU. b, Quantitated initial rates from these constructs. Results from CSFV IRES (negative control) shown for comparison. ‘Downstream SDS’ contains an SDS driving FLUC production (in place of the IRES), ‘Downstream SDS-like’ contains the purine-rich sequence in place of the IRES and driving FLUC production. In ‘Downstream SDS-like_K/O’, this purine-rich sequence has been replaced by a pyrimidine-rich sequence. A PSIV IRES construct in which both pseudoknots are disrupted and the purine-rich SDS-like sequence just downstream of the IRES is mutated has essentially the same activity as the CSFV IRES (Downstream SDS-like_K/O+PK1+PK2_K/O). Error bars are 1 s.d. from the mean of three biological replicates.

Extended Data Figure 7 The position of domain 3 in the full-length PSIV IGR IRES•70S structure.

Crystal structure of a full-length PSIV IGR IRES bound to T. thermophilus 70S ribosomes. Cyan, small subunit; red, PSIV IRES domain 3; black, unbiased Fourier difference F− Fc map for domain 3 in the P site of the small subunit. The large subunit and domain 1+2 are not shown.

Extended Data Figure 8 Luciferase activity time courses for various constructs.

a, Time-course traces for constructs and bar graphs shown in Fig. 3. b, Time-course traces for constructs and bar graphs shown in Fig. 4. Error bars are 1 s.d. from the mean of three biological replicates. a, b, The y-axis shows RLU.

Extended Data Figure 9 Quantitated data for various constructs in the context of the PK1+PK2_K/O mutation.

a, Combination of knocking out the RLUC SDS (Upstream SDS_K/O) with the PK2_K/O or PK1+PK2_K/O. Initial rates of RLUC are greatly diminished. Rates of FLUC are lower, but less diminished than RLUC. This is probably attributable to the decreased competition for ribosomes and the presence of the SDS-like sequence upstream of the FLUC open reading frame and not to robust initiation on the IRES. b, The PK1+PK2_K/O dramatically reduced the initial rate of FLUC production on the IRES with the F-SHIFT(−1) mutation. c, The PK1+PK2_K/O dramatically reduced the initial rate of FLUC production on the IRES with the F-SHIFT(−2) mutation. d, The PK1+PK2_K/O dramatically reduced the initial rate of FLUC production on the IRES with the uSTOP and uAUG mutations. Error bars are 1 s.d. from the mean from three biological replicates.

Extended Data Table 1 Initial rates of RLUC and FLUC for all constructs tested and crystallographic data collection, phasing and refinement statistics

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Colussi, T., Costantino, D., Zhu, J. et al. Initiation of translation in bacteria by a structured eukaryotic IRES RNA. Nature 519, 110–113 (2015).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing