Capped mRNA with a Single Nucleotide Leader Is Optimally Translated in a Primitive Eukaryote, Giardia lamblia

The 5′-untranslated region (5′-UTR) of an mRNA plays an important role in translation initiation in eukaryotes. A minimal length of about 20 nucleotides is required to prevent leaky ribosome scanning. In one of the most primitive eukaryotes, Giardia lamblia, however, the mRNAs have 5′-UTRs mostly in the range of 0 to 14 nucleotides without a conserved sequence, which raises the question on how the ribosome could effectively scan such short 5′-UTRs for an accurate initiation of translation. In the present study, we expressed capped transcripts of luciferase gene in Giardia trophozoites via transfection and observed that when the 5′-UTR of the transcript was lengthened from 9 to 21 nucleotides, there was a corresponding decrease of translation efficiency. Conversely, shortening of the 5′-UTR from nine nucleotides down to a single nucleotide did not result in any reduced translation or leaky scanning. Translation appeared to initiate exclusively from the first initiation codon located downstream from the cap. Experimental evidence indicated also that a stem-loop structure immediately downstream from the initiation codon exerted significant inhibition on translation initiation when the 5′-UTR consisted of less than seven nucleotides. This inhibitory effect was abolished by increasing the distance between the stem-loop and the cap-G structure either upstream or downstream from the start codon, thus suggesting a spatial requirement for effective ribosome recruitment. Overall, our results suggest an absence of ribosome scanning for AUG in initiating translation in Giardia. A capped mRNA with a single nucleotide leader is apparently sufficient for recruiting ribosome and initiating translation. The 5′-untranslated region (5′-UTR) of an mRNA plays an important role in translation initiation in eukaryotes. A minimal length of about 20 nucleotides is required to prevent leaky ribosome scanning. In one of the most primitive eukaryotes, Giardia lamblia, however, the mRNAs have 5′-UTRs mostly in the range of 0 to 14 nucleotides without a conserved sequence, which raises the question on how the ribosome could effectively scan such short 5′-UTRs for an accurate initiation of translation. In the present study, we expressed capped transcripts of luciferase gene in Giardia trophozoites via transfection and observed that when the 5′-UTR of the transcript was lengthened from 9 to 21 nucleotides, there was a corresponding decrease of translation efficiency. Conversely, shortening of the 5′-UTR from nine nucleotides down to a single nucleotide did not result in any reduced translation or leaky scanning. Translation appeared to initiate exclusively from the first initiation codon located downstream from the cap. Experimental evidence indicated also that a stem-loop structure immediately downstream from the initiation codon exerted significant inhibition on translation initiation when the 5′-UTR consisted of less than seven nucleotides. This inhibitory effect was abolished by increasing the distance between the stem-loop and the cap-G structure either upstream or downstream from the start codon, thus suggesting a spatial requirement for effective ribosome recruitment. Overall, our results suggest an absence of ribosome scanning for AUG in initiating translation in Giardia. A capped mRNA with a single nucleotide leader is apparently sufficient for recruiting ribosome and initiating translation. The 5′-untranslated regions (UTRs) 1The abbreviations used are: UTR, untranslated region; RT, reverse transcription; DMS, dimethyl sulfate; CMCT, 1-cyclohexyl-3-(2-morpholinoethyl)-carbodiimide; nt, nucleotide; SL, stem-loop. of mRNA molecules in eukaryotes have an average length of about 90 nucleotides in mammals (1Kozak M. Nucleic Acids Res. 1987; 15: 8125-8148Crossref PubMed Scopus (4169) Google Scholar) and 52 nucleotides in yeast (2Cigan A.M. Donahue T.F. Gene. 1987; 59: 1-18Crossref PubMed Scopus (257) Google Scholar). Shortening the 5′-UTR was found to place the first AUG codon in an increasingly unfavorable context for in vitro translation initiation, resulting in leaky scanning (3Kozak M. Gene Expr. 1991; 1: 111-115PubMed Google Scholar). The yield of protein initiated from the first AUG codon was progressively decreased, whereas that derived from the second AUG downstream from it was correspondingly increased. In a systematic in vitro experiment, about half of the ribosomes bypassed the first AUG codon and initiated instead at a downstream AUG site when the first AUG was within 12 nucleotides from the m7GpppN cap (3Kozak M. Gene Expr. 1991; 1: 111-115PubMed Google Scholar). The leakiness was suppressed when the leader sequence was lengthened to 20 nucleotides. Further lengthening of the 5′-UTR beyond 20 nucleotides indicated a linear increase of the efficiency of translation up to 80 nucleotides (4Kozak M. Gene Expr. 1991; 1: 117-125PubMed Google Scholar). The scanning leakiness could be also reduced by introducing a modest amount of secondary structure downstream from the first AUG, presumably via slowing down the rate of ribosome scanning (5Kozak M. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 8301-8305Crossref PubMed Scopus (410) Google Scholar). Giardia lamblia is an anaerobic protozoan that infects the small intestines of humans and many other mammals. It is classified as one of the deeply branched and most primitive eukaryotes based on phylogenetic analysis of its small ribosomal RNA and many other proteins (6Adam R.D. Clin. Microbiol. Rev. 2001; 14: 447-475Crossref PubMed Scopus (921) Google Scholar, 7Inagaki Y. Ford Doolittle W. Mol. Biol. Evol. 2000; 17: 882-889Crossref PubMed Scopus (70) Google Scholar, 8Roger A.J. Svard S.G. Tovar J. Clark C.G. Smith M.W. Gillin F.D. Sogin M.L. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 229-234Crossref PubMed Scopus (244) Google Scholar, 9Sogin M.L. Gunderson J.H. Elwood H.J. Alonso R.A. Peattie D.A. Science. 1989; 243: 75-77Crossref PubMed Scopus (591) Google Scholar). Within its small 12-Mb genome, most of the genes reported so far do not contain introns (10Nixon J.E. Wang A. Morrison H.G. McArthur A.G. Sogin M.L. Loftus B.J. Samuelson J. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 3701-3705Crossref PubMed Scopus (123) Google Scholar). Their transcripts have exceedingly short 5′-UTRs ranging mostly from 0 to 14 nucleotides (11Adam R.D. Int. J. Parasitol. 2000; 30: 475-484Crossref PubMed Scopus (131) Google Scholar) and similarly short 3′-UTRs of 10 to 30 nucleotides (11Adam R.D. Int. J. Parasitol. 2000; 30: 475-484Crossref PubMed Scopus (131) Google Scholar). With such unusually short 5′-UTRs that do not share a consensus sequence (12Yu D.C. Wang A.L. Botka C.W. Wang C.C. Mol. Biochem. Parasitol. 1998; 96: 151-165Crossref PubMed Scopus (29) Google Scholar), it will be interesting to learn how an optimal accuracy, as well as efficiency of translation, can be maintained and leaky ribosome scanning prevented in Giardia. Similarly short 5′-UTRs have also been identified among the mRNAs from other deeply branched anaerobic protozoa such as Entamoeba histolytica (13Bruchhaus I. Leippe M. Lioutas C. Tannich E. DNA Cell Biol. 1993; 12: 925-933Crossref PubMed Scopus (119) Google Scholar) and Trichomonas vaginalis (14Davis-Hayman S.R. Shah P.H. Finley R.W. Lushbaugh W.B. Meade J.C. Parasitol. Res. 2000; 86: 608-612Crossref PubMed Scopus (8) Google Scholar). These short 5′-UTRs could reflect the presence of a unique protein synthetic machinery in these closely related primitive eukaryotes. In a previous study, we indicated that a priori decapping of the poly(A)+ RNA from G. lamblia was unnecessary for T4 polynucleotide kinase phosphorylation or ligation to oligonucleotides and suggested that much of it could be uncapped at the 5′-ends (12Yu D.C. Wang A.L. Botka C.W. Wang C.C. Mol. Biochem. Parasitol. 1998; 96: 151-165Crossref PubMed Scopus (29) Google Scholar). But a subsequent report on the natural abundance of antisense transcripts in Giardia up to 20% of the total poly(A)+ RNA cast some doubt on the real content of mRNA in the sample (15Elmendorf H.G. Singer S.M. Nash T.E. Nucleic Acids Res. 2001; 29: 4674-4683Crossref PubMed Scopus (49) Google Scholar). A recently completed Giardia Genome data base showed that many homologues of yeast and human translation initiation factors are present in Giardia. Genes encoding homologues of eIF1 (EAA42222), eIF1A (EAA41151), eIF2 (α, EAA42702; β, EAA38204; γ, AAD04236) (16Keeling P.J. Fast N.M. McFadden G.I. J. Mol. Evol. 1998; 47: 649-655Crossref PubMed Scopus (40) Google Scholar), eIF2B (α, EAA38784; β, EAA38691; δ, EAA40267; ϵ, EAA37510), eIF3 (δ, EAA39062), eIF4A (EAA36655), eIF5A (EAA37465), eIF4E (EAA39303), the cap-binding protein, and eIF5 (EAA37306), the putative GTPase-activating protein for eIF2, were identified, suggesting that capped mRNA may be present and translated in this organism. Upon a closer scrutiny, however, homologues of other initiation factors such as eIF4B, eIF4G, and eIF4H are apparently missing. and eIF4H are the proteins to eIF4A to the RNA secondary structure the of the mRNA presumably required for ribosome scanning Genes 2002; PubMed Scopus (410) Google Scholar). is a protein with other initiation factors T.E. Biochem. Sci. PubMed Scopus Google Scholar). It a RNA region of for ribosome scanning J. J. PubMed Scopus Google Scholar). a translation initiation in these protein may suggest a protein synthetic machinery that the and the ribosome an initiation at the site of without scanning down the mRNA in a to for the AUG start codon M. J. Cell Biol. 1989; PubMed Scopus Google Scholar). This could be used to the presence of unusually short and 5′-UTRs in Giardia In an to this we first a and much of translation of the capped than the uncapped mRNA in Giardia and to an absence of ribosome scanning in this organism. The AUG immediately downstream from the cap is as the initiation codon the short distance down to a single nucleotide in This which an optimal translation efficiency in a protein synthetic machinery that may the between and the eukaryotes. 5′-UTR H.G. Singer S.M. J. J. Nash T.E. Mol. Biochem. Parasitol. 2001; PubMed Scopus Google Scholar, Mol. Biochem. Parasitol. 1989; PubMed Scopus Google Scholar) located between and luciferase gene in the was with the 5′-UTR sequence from the from Giardia A. Alonso R.A. Peattie D.A. Mol. Biochem. Parasitol. PubMed Scopus Google Scholar) by the of the to the were with the by subsequent DNA The luciferase gene was by and a with a and an 5′-UTR located The thus was used for of its 5′-UTR by as In RNA DNA were at the of luciferase gene with a and the capped transcript was from of the DNA with RNA in the presence of 12 and were to the mRNAs the from a The was by the and for by in a The of mRNA sample was by its at in a of Giardia vitro of Giardia trophozoites was maintained as A.L. Wang C.C. Mol. Biochem. Parasitol. PubMed Scopus Google Scholar). of the in vitro were at an of to a of the trophozoites with RNA was by as Wang A.L. Wang C.C. J. 2001; PubMed Scopus Google Scholar, D.C. Wang A.L. Wang C.C. Mol. Cell Biol. 15: PubMed Scopus Google Scholar) with were at for 10 and with the Cell were in of with 20 of the in vitro transcript and of yeast and in a for 10 on was with a at and the the were of and at for of the luciferase in the of were by the of the The luciferase was with a and in was by the in a The luciferase expressed in of Giardia was from at with total RNA from Giardia trophozoites was and the of RNA sample was of RNA was as in of the the luciferase gene and to the transcript of luciferase gene The were at one of reverse at for 30 and 20 of at for for 30 and for The on the gene transcript in the Giardia was as an the of and for the gene the RNA by or and capped in vitro transcript of was a RNA The RNA was at for and down to in an for 30 in a 10 It was and its was by on a of the in vitro transcript was used in either dimethyl or 21 Wang A.L. Wang C.C. J. 2001; PubMed Scopus Google Scholar, S. J. Wang C.C. J. Mol. Biol. 2001; PubMed Scopus Google Scholar). The were at for 10 and 20 and by of the The RNA was by of of the RNA structure was as S. Wang C.C. 2002; PubMed Scopus Google Scholar). The capped in vitro transcript of with of yeast was in the RNA structure at for and to within 30 The RNA sample was with which RNA as at at for 20 The was by to the RNA which were by in the presence of and in The of or in the RNA were by the reverse with oligonucleotides to the region 20 downstream from the of the stem-loop in the transcript the were to of the or RNA by the at for by 10 on was at for as by the were in an with the by the The were identified as a reverse with a short of that in the corresponding DNA in of the The of in the of from RNA a nucleotide that is at its by the and thus the as that in the corresponding DNA The and Translation of mRNA in Giardia vitro of the by of the in vitro transcript the luciferase mRNA with in the This in vitro transcript was Giardia trophozoites via on the total RNA from the trophozoites and 20 transfection indicated that the capped luciferase transcript was present in Giardia at a which was and reduced from the luciferase in the of the indicated an and of expressed from to the transfection In a experiment, of an luciferase mRNA with in the 5′-UTR Giardia trophozoites in a of the It was reduced between and transfection and by luciferase could be in the from to the uncapped transcript was by suggesting that the uncapped mRNA was not or translated in to the capped the translation initiation is most in Giardia. the subsequent on translation initiation Giardia trophozoites were thus with capped mRNAs with the luciferase of 5′-UTR a in Translation in 5′-UTR of capped luciferase mRNA in of an structure as the cap from the initiation codon by a of which is the 5′-UTR of mRNA from Giardia by a and a A. Alonso R.A. Peattie D.A. Mol. Biochem. Parasitol. PubMed Scopus Google Scholar). This capped translation of its downstream to an of luciferase in Giardia This 5′-UTR was to and nucleotides by and the 5′-UTR sequence in In vitro transcripts of these were capped and and Giardia of the transcripts Giardia trophozoites transfection were by and to be to one a in The results on translation in to was observed in the in vitro (3Kozak M. Gene Expr. 1991; 1: 111-115PubMed Google Scholar), an increase in the length of 5′-UTR to a decrease in translation. A 5′-UTR to of the of that from a 5′-UTR whereas and 5′-UTRs reduced the of to and This is the first as far as we are that an increase of 5′-UTR from 9 to nucleotides an inhibitory instead of an on translation. The however, well with the that most 5′-UTRs in the mRNAs of Giardia are between 0 to 14 nucleotides in length (11Adam R.D. Int. J. Parasitol. 2000; 30: 475-484Crossref PubMed Scopus (131) Google Scholar). of the of these transcripts in Giardia within the were also by the in The results in that the rate of of the transcript with the of is to that with the of thus the that the of could be of of among the The a from the without Translation or length of 5′-UTR for an optimal translation initiation in Giardia the question on how much it can be without the translation efficiency. The AUG in luciferase mRNA was to whereas a AUG was nucleotides upstream from it and from the cap by a single This mRNA was Giardia via transfection and of the translation of that with the 5′-UTR is thus required between the cap and AUG for an optimal translation initiation in Giardia. the downstream in transcript was to AUG in transcript when a second AUG was nucleotides downstream from the first one that was one nucleotide from the the mRNA thus was expressed at of the in Giardia. when the distance between the was to nucleotides, the of was reduced to of the suggesting that not of the translation was initiated from the first AUG, which was of from the downstream luciferase This translation was not by or the second initiation codon nucleotides downstream from the first AUG and thus a of or ribosome scanning leakiness that could initiate translation from the downstream The absence of such leakiness and the short of a single nucleotide the cap and AUG for optimal translation suggest absence of ribosome scanning translation initiation in Giardia. The of and of transcripts in Giardia the transfection in the were also with as in and in that transcripts with of and have the and the The of of these transcripts are thus most to the translation A from and the to Translation on a mRNA with a absence of ribosome scanning that may an structure in the mRNA for ribosome and translation initiation in Giardia. of such a however, some downstream from the cap for of the initiation factors and of the instead of scanning leakiness in mammals (5Kozak M. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 8301-8305Crossref PubMed Scopus (410) Google Scholar), a secondary structure immediately downstream from the initiation codon is to have an inhibitory effect on initiating translation of a mRNA with a short 5′-UTR in Giardia. The luciferase mRNA an stem-loop structure immediately downstream from and the start codon the distance between the and the start codon of luciferase mRNA is from 9 to nucleotides, a of the mRNA is which however, reduced to with nucleotides and and with nucleotides and nucleotide in the with the that the reduced of transcript in could be to the content in the sequence, its were with in transcript of the in Giardia was of the which was than that of transcript suggesting that the base may have some effect on translation content in the sequence an inhibitory effect as the stem-loop is nucleotides closer to the cap The transcripts however, of than that of the transcripts and the of initiation codon in the structure may have to the reduced translation, an AUG initiation codon was immediately upstream from the luciferase AUG with the that the AUG, which is apparently not in the will be the real initiation codon was expressed at of the which is that of the transcript without the AUG thus that of AUG in a secondary structure translation This is by the of transcript in which an sequence was immediately downstream from the luciferase AUG in transcript to the initiation codon from the Translation initiation was however, when an was immediately upstream from AUG in transcript which in of the the previous that lengthening the distance between the cap and initiation codon not translation initiation it The of luciferase plays an role in its J. W. W. Int. J. Biochem. Cell Biol. 2002; PubMed Scopus Google Scholar). It be to the stem-loop structure in the encoding mRNA without the as The luciferase mRNA which not have a secondary structure within 10 downstream from the start codon, was used to the luciferase mRNA in the The result that when there is in of the initiation codon, the luciferase was expressed at of the 5′-UTR and when the stem-loop secondary structure in the luciferase mRNA was between the 5′-UTR and AUG in transcripts and in and the the of as of transcripts and whereas of transcript was reduced to of the and The of and of the transcripts in were also by The results indicated among any of and thus that the in luciferase observed in the in translation the at the of mRNA by and there is the an RNA secondary structure in we the secondary structure at the of luciferase mRNA a of and of the RNA by a The capped in vitro transcript was with or the A at and on at the whereas the of and the of Wang A.L. Wang C.C. J. 2001; PubMed Scopus Google Scholar, S. J. Wang C.C. J. Mol. Biol. 2001; PubMed Scopus Google Scholar). The thus can be by the reverse in a subsequent and base can be identified as a reverse in subsequent The at the base one nucleotide of the corresponding DNA immediately the The results from a and on the capped transcript are in the region the structure was by the the in it were as the for and suggesting that the region is in an The regions to the with a in the were found of for that is at the of the which could be to a of the structure at the region with the The AUG initiation codon at the of the in was of and is thus most in the The in the region by the and are in as base the in and in a to the presence of the stem-loop structure at the of luciferase mRNA with the AUG located at the base of the the from the transcript structure was also with to RNA in S. Wang C.C. 2002; PubMed Scopus Google Scholar). The results that the RNA reverse at and in the structure and at and in the The nucleotides in the that were found to however, also found to suggesting that could be from the by The nucleotides in the and with to thus suggesting the in the and and in the also to These could be thus in a between and The between data from and the presence of the stem-loop structure at the of luciferase mRNA the AUG start codon at the base of the In the we the of introducing mRNAs Giardia trophozoites via and its in The results thus on translation initiation from observed in the in vitro derived from yeast and mammals M. 2002; PubMed Scopus Google Scholar, M. Gene. 2002; PubMed Scopus Google Scholar). The mRNA translation efficiency in Giardia with increasing of 5′-UTR beyond nine nucleotides, whereas the in vitro reduced translation efficiency with increasing leakiness of ribosome scanning with a 5′-UTR than 20 (3Kozak M. Gene Expr. 1991; 1: 111-115PubMed Google Scholar). The 5′-UTR in Giardia mRNA can be reduced to a single nucleotide without its translation efficiency or any suggesting that ribosome scanning in translation initiation among the eukaryotes may not in Giardia. The absence from Giardia of eIF4B, and homologues that are in ribosome scanning J. J. PubMed Scopus Google Scholar, T.E. 2002; PubMed Scopus Google Scholar) this The minimal of a single in the 5′-UTR in our was to a in the RNA in vitro a of the first downstream from the start site the yield of the in vitro transcript Nucleic Acids Res. 1987; 15: PubMed Scopus Google Scholar). many Giardia mRNA are without any 5′-UTR at (11Adam R.D. Int. J. Parasitol. 2000; 30: 475-484Crossref PubMed Scopus (131) Google Scholar), it is not that a between the and the initiation codon may the and sufficient in the mRNA to initiate translation in Giardia. the of eIF4E to the the ribosome with other initiation factors and is to the T.E. 2002; PubMed Scopus Google Scholar). The AUG codon immediately downstream from the will be by the without a ribosome scanning from the to the the a spatial requirement for optimal translation initiation when a secondary structure is located immediately downstream from the initiation codon this These the protein in Giardia from in the other eukaryotes and In the of the start site is by base of the sequence in the mRNA with the of ribosomal RNA in the 30 J. J. Biochem. PubMed Scopus Google Scholar). mRNAs have however, also identified in are and translated by a that identified in eukaryotes I. S. U. Mol. Microbiol. 2002; PubMed Scopus Google Scholar). The which are the to the of eukaryotes, also have short or 5′-UTR in uncapped mRNAs PubMed Scopus Google Scholar). to translation initiation also a of initiation factors that are to of the eukaryotes that in the initiation the translation initiation factors identified in have been eIF2, and eIF4A without eIF4B, eIF4G, or eIF5 PubMed Scopus Google Scholar), which are a of the initiation identified in Giardia the These initiation factors may an for initiating translation of the uncapped mRNAs in M. apparently without either a 5′-UTR or ribosome scanning. Giardia mRNAs do not have a sequence or any other consensus sequence at the 5′-ends (11Adam R.D. Int. J. Parasitol. 2000; 30: 475-484Crossref PubMed Scopus (131) Google Scholar). from the a cap structure at the of mRNA is in recruiting the small ribosome and subsequent of the downstream initiation It is not an RNA base between the mRNA and the of the ribosomal RNA in Giardia is in the of In a previous (12Yu D.C. Wang A.L. Botka C.W. Wang C.C. Mol. Biochem. Parasitol. 1998; 96: 151-165Crossref PubMed Scopus (29) Google Scholar), we identified in a a downstream in the region of uncapped transcript that can a sequence at the of the ribosomal RNA in Giardia. This downstream is not in any secondary structure of the transcript S. J. Wang C.C. J. Mol. Biol. 2001; PubMed Scopus Google Scholar) is for initiating translation of the transcript (12Yu D.C. Wang A.L. Botka C.W. Wang C.C. Mol. Biochem. Parasitol. 1998; 96: 151-165Crossref PubMed Scopus (29) Google Scholar), suggesting that a could be in translation initiation, by a sequence downstream from the initiation Further analysis indicated that translation of the uncapped transcript in Giardia is initiated at an ribosomal site in the mRNA of the 5′-UTR and the downstream region the downstream S. Wang C.C. 2002; PubMed Scopus Google Scholar). Within the secondary of this ribosomal there is a stem-loop structure located 21 nucleotides upstream from the AUG and stem-loop nucleotides downstream from in translation The of the AUG between the be a upstream or downstream by nucleotides translation initiation S. Wang C.C. 2002; PubMed Scopus Google Scholar). the between the which could the of a small ribosome well Wang A. J. A. 2002; PubMed Scopus Google Scholar, M. J. PubMed Scopus Google Scholar), may be the site for ribosome small to the initiation codon in the mRNA may the site of the ribosome for initiating the translation. This from the ribosomal site in a transcript in Giardia to our that translation initiation from mRNAs may without ribosomal scanning in Giardia. A of about 20 Giardia indicated that there is second AUG within nucleotides downstream from the first initiation and C. C. It could be of the absence of ribosome scanning in the protein in Giardia. In we have identified in one of the most primitive eukaryotes Giardia a unique of translation initiation that between that of a and a with a with that in the The in these in and the in a of the protein synthetic machinery a from the the a for interesting in of for the and for this