Eukaryotic 16S-like ribosomal RNAs contain 12 so-called expansion segments, i.e., sequences not included in the RNA secondary structure core common to eubacteria, archaea, and eukarya. Two of these expansion segments, ES3 and ES6, are juxtaposed in the recent three-dimensional model of the eukaryotic 40S ribosomal subunit. We have analyzed ES3 and ES6 sequences from more than 2900 discrete eukaryotic species, for possible sequence complementarity between the two expansion segments. The data show that ES3 and ES6 could interact by forming a helix consisting of seven to nine contiguous base pairs in almost all analyzed species. We, therefore, suggest that ES3 and ES6 form a direct RNA-RNA contact in the ribosome.Eukaryotic 18S ribosomal RNA is considerably longer than its prokaryotic homolog 16S rRNA (Clark 1987). Despite differences in size and sequence, 18S rRNA and 16S rRNA share a common structural core (Gutell et al. 1985) in which the additional nucleotides found in 18S rRNA are inserted at specific positions. The inserted nucleotides form extra sequence elements called variable regions (Neefs and De Wachter 1990) or expansion segments (ES; Gerbi 1996). The length and sequence of these expansion segments varies considerably between organisms.18S rRNA contains 12 expansion segments (Gerbi 1996). One of these expansion segments, referred to as ES6, is located in the central domain of 18S rRNA (Fig. 1). With an average length of 250 nucleotides (Neefs and De Wachter 1990), this is the largest expansion segment found in 18S rRNA. ES6 can be divided in two halves based on sequence variability. The 5Ј half exhibits extensive variability, whereas the sequence of the 3Ј half is more conserved. Sequences corresponding to the latter half are absent in eubacterial and archaeal 16S rRNA (Cannone et al. 2002). Several attempts to construct a secondary structure model for the 3Ј half have been made (Nickrent and Sargent 1991;Hancock and Vogler 1998;Wuyts et al. 2000). However, because of the sequence conservation, structure prediction is difficult, and this part of ES6 is therefore left unstructured in the phylogenetic models of 18S rRNA secondary structure ( Fig. 1; Cannone et al. 2002).The location of ES6 within the three-dimensional structure of the yeast 40S subunit has recently been determined using cryo-electron microscopy (Spahn et al. 2001). The expansion segment is located on the back of the so-called body of the 40S subunit (Fig. 1, inset). The segment is seen as two separate densities, one positioned across the back of the 40S body, whereas the second density is located at the side of the lower part of the body with its lower end close to the so-called left foot (Spahn et al. 2001).The left foot contains another expansion segment, called ES3 (Spahn et al. 2001), only found in eukaryotes (Cannone et al. 2002). ES3 is very variable in size and sequence (Gutell et al. 1985;Wuyts et al. 2002), but the phylogenetic data indicate a similar basic structure for ES3 in different organisms ( Fig. 1; Cannone et al...

The 18S rRNA of the small eukaryotic ribosomal subunit contains several expansion segments. Electron microscopy data indicate that two of the largest expansion segments are juxtaposed in intact 40S subunits, and data from phylogenetic sequence comparisons indicate that these two expansion segments contain complementary sequences that could form a direct tertiary interaction on the ribosome. We have investigated the secondary structure of the two expansion segments in the region around the putative tertiary interaction. Ribosomes from yeast, wheat, and mouse-three organisms representing separate eukaryotic kingdoms-were isolated, and the structure of ES3 and part of the ES6 region were analyzed using the single-strand-specific chemical reagents CMCT and DMS and the double-strand-specific ribonuclease V1. The modification patterns were analyzed by primer extension and gel electrophoresis on an ABI 377 automated DNA sequencer. The investigated sequences were relatively exposed to chemical and enzymatic modification. This is in line with their indicated location on the surface at the solvent side of the subunit. The complementary ES3 and ES6 sequences were clearly inaccessible to single-strand modification, but available for cleavage by double-strand-specific RNase V1. The results are compatible with a direct helical interaction between bases in ES3 and ES6. Almost identical results were obtained with ribosomes from the three organisms investigated.

Expansion segment ES6 in 18S ribosomal RNA is, unlike many other expansion segments, present in all eukaryotes. The available data suggest that ES6 is located on the surface of the small ribosomal subunit. Here we have analyzed the secondary structure of the complete ES6 sequence in intact ribosomes from three eukaryotes, wheat, yeast, and mouse, representing different eukaryotic kingdoms. The availability of the ES6 sequence for modification and cleavage by structure sensitive chemicals and enzymatic reagents was analyzed by primer extension and gel electrophoresis on an ABI 377 automated DNA sequencer. The experimental results were used to restrict the number of possible secondary structure models of ES6 generated by the folding software MFOLD. The modification data obtained from the three experimental organisms were very similar despite the sequence variation. Consequently, similar secondary structure models were obtained for the ES6 sequence in wheat, yeast, and mouse ribosomes. A comparison of sequence data from more than 6000 eukaryotes showed that similar structural elements could also be formed in other organisms. The comparative analysis also showed that the extent of compensatory base changes in the suggested helices was low. The in situ structure analysis was complemented by a secondary structure analysis of wheat ES6 transcribed and folded in vitro. The obtained modification data indicate that the secondary structure of the in vitro transcribed sequence differs from that observed in the intact ribosome. These results suggest that chaperones, ribosomal proteins, and/or tertiary rRNA interactions could be involved in the in vivo folding of ES6.

No abstract