Specific secondary structures in the capsid-coding region of giardiavirus transcript are required for its translation in Giardia lamblia 1 1Edited by D. E. Draper

Identification of an essential pseudoknot in the putative downstream internal ribosome entry site in giardiavirus transcript

“…In our previous investigation, we observed that any alteration of the sequence 59-UCUCC-39 (nt 216-220, Fig+ 1) in the 264-nt capsid-encoding region of GLV mRNA resulted in a drastic reduction in luciferase expression in the transfected Giardia (Garlapati et al+, 2001)+ The loss of activity in two of the previously analyzed mutants, U216A/C217A and C219U/C220G, was not due to disruption of any identifiable stem-loop structure+ It was due to alteration of the specific sequence, suggesting that the latter is probably involved in base pairing with another sequence in the transcript (Garlapati et al+, 2001)+ A search in the 264-nt region by visual inspection revealed a sequence, 59-GGAGA-39, in the loop of stem-loop II (nt 154-158) that could form a Watson-Crick base pairing with the sequence 59-UCUCC-39 (Fig+ 1)+ To test this possibility, a compensatory mutation was introduced at G157U/A158U in the mutant U216A/C217A+ The double mutation resulted in a significant recovery of the luciferase activity from 4+9% to 41+0% of the wild-type level (Table 1)+ Compensatory mutations were also introduced at G154C/G155A to determine if they could complement the mutations C219U/C220G and C219U/C220G/ C229A/A230C (Table 1)+ The luciferase activity in mutant C219U/C220G was restored from 2+5% to 55+0%, whereas a 90% recovery of luciferase activity was achieved in mutant C219/C220G/C229A/A230C (Table 1)+ The mere partial recovery in the former case could be due to a competition between the unaltered C229/A230 and the altered C154/A155 in compensatory mutation to base pair with U219/G220, whereas this competition was absent in the latter case and thus resulted in a much enhanced recovery+ Two double substitutions, U216A/C217G and G157C/A158U, were shown to each result in a drastic loss of luciferase activity to 3+8% and 4+2% of the wild type (Table 1) Dam et al+, 1990Dam et al+, , 1992Hilbers et al+, 1998), this putative pseudoknot in our hands is similar to the classical H-type pseudoknot with only minor variations+ By the classical H-type pseudoknot structure, the stem in stem-loop II could be considered as the pseudoknot stem 1, whereas the stem structure generated by pairing the two pentanucleotides could be regarded as pseudoknot stem 2 (Fig+ 6)+ The part of loop II sequence immediately upstream from the pentanucleotide 59-GGAGA-39 could be taken as pseudoknot loop 1, whereas pseudoknot loop 2 spans 58 nt, including stem-loops III and IVA separated by an 10-nt singlestranded region (Fig+ 6)+ For the ease of discussion, the stem formed by the two pentanucleotides will be referred to as the pseudoknot stem and the rest of the stem structures will remain as previously designated (Garlapati et al+, 2001)+ Another possible explanation, that the two pentanucleotides may have formed a "kissing complex," is less likely, because the latter has been defined as a basepair formation between two hairpin loops (Chang & Tinoco, 1994)+ Whereas 59-GGAGA-39 constitutes a part of the loop in stem-loop II, 59-UCUCC-39 is located in a single-stranded region immediately downstream from stem-loop IVA (Garlapati et al+, 2001)+ A pseudoknot will be a more appropriate description of such a basepairing complex+…”

“…The enzymatic probing experiments indicated that A221 and A222 immediately downstream from the pentanucleotide 59-UCUCC-39 could be slowly digested by RNaseV (Figs+ 3 and 6) suggesting that A221 could base pair with U153 in loop II, immediately upstream to the other pentanucleotide sequence, 59-GGAGA-39, to extend the length of base pairings by one more (Fig+ 6)+ An A152C/U153C substitution in the previous study was found without significant effect on translation, suggesting a lack of functional importance of the postulated U153/A221 base pair (Garlapati et al+, 2001)+ In the current investigation, two double substitutions, U153A/G154C and C220G/A221U, were each found to cause decrease of luciferase expression to 11% and 8%, which was restored to 78% of the wild-type level by combining the two ( S. Garlapati and C.C. Wang 153U and 221A does not make a significant contribution to the stability of the presumed pseudoknot structure+…”

“…The putative pseudoknot structure was also evaluated by probing the RNA with chemical agents+ DMS (dimethyl sulphate), which methylates unpaired A residue at N-1 and unpaired C residue at N-3, KE (kethoxal), which modifies only unpaired G residue at N-1, and CMCT (1-cyclohexyl-3-(-2-morpholinoethyl)carbodiimide), which modifies the N-3 group of unpaired U and the N-1 group of unpaired G residue, were used in the present study+ In our previous probing experiments, we probed the RNA molecule under semidenatured conditions, and observed that only the three C residues in the 59-UCUCC-39 pentanucleotide were inaccessible to chemical probing, whereas the two Us and the 5 nt in 59-GGAGA-39 were all chemically modified (Garlapati et al+, 2001)+ In our present study, we used the same slow renaturing protocol under native conditions as for the enzymatic probing (i+e+, in the presence of 10 mM MgCl 2 instead of 1 mM EDTA; Garlapati et al+, 2001), to try to stabilize the putative pseudoknot structure+ A major difference between the outcomes from under these two conditions is that bases in the pentanucleotide 59-GGAGA-39 in loop II were no longer labeled by the chemicals (Fig+ 4)+ Only a weak DMS labeling of A158 at the 39 end of the pentanucleotide was observed, suggesting that the bases are mostly in paired conformation+ The weak modification of the last residue in the pentanucleotide could still be due to partial denaturation of the pseudoknot stem under the current in vitro conditions+ Similarly, all the five residues in the other pentanucleotide, 59-UCUCC-39, were not labeled at all under the present conditions ( The bases G183, A184, U185, C186, G187, G189, and U192 in the previously identified single-stranded 10-nt region separating stem III and stem IVA also became chemically modified under the native condition (Fig+ 5), verifying its presence in a single-stranded form+ Its relative insensitivity toward digestion by RNaseT1, RNaseT2, and RNaseA shown in Figure 2 could thus be taken as an indication that the sequence, flanked by two stem loops, is inaccessible to these enzymes+…”

“…The pseudoknot is located 143 nt downstream from the initiation codon and yet its disruption leads to a drastic loss of translation efficiency (Yu et al+, 1998;Garlapati et al+, 2001 …”

“…Giardiavirus (GLV) is a double-stranded (ds)RNA virus of the Totiviridiae family, which specifically infects trophozoites of the most primitive protozoan parasite Giardia lambila (Wang & Wang, 1986)+ Its dsRNA genome of 6277 bp encodes two polypeptides; a major 100-kDa capsid protein (Gag) and a minor 190-kDa fusion protein (Gag-Pol) via a Ϫ1 ribosomal frameshift (Wang et al+, 1993;Li et al+, 2001)+ The coding region in GLV (ϩ)-strand RNA is flanked by a 367-nt 59 untranslated region (UTR) and a 301-nt 39 UTR + The ability of purified GLV to infect, and the capability of its (ϩ)-stranded RNA to transfect G. lamblia trophozoites, resulting in intracellular proliferation of infectious GLV particles, are the major distinguishing features of this virus among the totiviruses (Wang & Wang, 1986)+ It has turned the GLV (ϩ)-strand RNA into an effective transfection vector for high-level expression of foreign mRNAs in GLVinfected G. lamblia (Yu et al+, 1995;Yu & Wang, 1996)+ In vitro-transcribed chimeric mRNA containing a fulllength firefly luciferase transcript flanked by the 367-nt GLV 59 UTR and a 2022-nt 39 terminus of GLV (ϩ)-strand RNA can be introduced into GLV-infected G. lamblia trophozoites via electroporation (Yu et al+, 1995)+ The chimeric mRNA thus introduced undergoes vigorous replication and transcription but only a basal level of translation, resulting in an expressed luciferase activity barely above the background level (Yu et al+, 1995;Yu & Wang, 1996)+ This relatively poor translation efficiency was, however, enhanced by 5,000-fold when the initial 264 nt of the capsid-coding region in GLV mRNA were fused in frame with, and upstream from, the luciferase mRNA (Yu & Wang, 1996)+ In our earlier studies, we have identified a 13-nt downstream box (DB) sequence within the initial 264-nt capsid coding region of GLV mRNA at position 66-78 that complements a 15-nt sequence (with two gaps) between nt 1382 and 1396 in the V9 region near the 39 end of the 16S-like ribosomal RNA (rRNA) of G. lamblia (Yu et al+, 1998)+ Deletion or scrambling of this DB sequence led to a significant loss of translation efficiency+ However, although DB is located 66-78 nt downstream of the start codon, inclusion of the first 98 nt of the downstream region encompassing the entire DB exerted little enhancement of translation of the chimeric transcript (Yu & Wang, 1996)+ It was the increment of coding region from nt 111 to 264 that resulted in an exponential increase of translation efficiency up to 5,000-fold (Yu & Wang, 1996)+ The entire 264-nt segment of GLV mRNA has since been thoroughly analyzed for additional structural and sequence-specific elements that may contribute to the enhanced translation+ Results from chemical probing and mutational analysis of the MFOLD-predicted stemloops I (nt 11-35), verified their presence in the RNA molecule and indicated their essential involvement in enhanced translation (Garlapati et al+, 2001)+ Similar experiments also demonstrated the presence and ...…”