THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1988 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 263, No. 34, Issue of December 5, pp. 18213-18219,1988 Printed in U. S. A.

Probing the RNA Structure within the Yeast 5 S RNA-Lla Protein Complex by Fluorescence and Enzymatic Digestion*

(Received for publication, June 8, 1988)

Lee-Chuan C. Yeh and John C. Lee From the Department of Biochemistry, The University of Texas Health Science Center, San Antonio, Texas 78284-7760

Conformational states of the ribosomal 5 S RNA molecule and its associated protein Lla in the yeast RNA-protein (RNP) complex were determined using controlled RNase T 1 digestion in conjunction with flu- orescence probes, ethidium bromide and bisanilino- naphthalenesulfonic acid. Fluorescence measurements indicated that the RNA molecule in the RNP complex appeared to exhibit a slightly lower degree of second- ary structure than that in the free form. Controlled digestion of the intact RNP complex with RNase T1 resulted in an initial increase in ethidium fluorescence followed by a gradual decrease. In free RNA, a similar profile, except that a larger increase in ethidium fluo- rescence at the initial stage of digestion, was observed. During digestion of the RNP complex, increases in bisanilinonaphthalenesulfonic acid fluorescence and in light scattering were observed. These findings implied that as regions of the 5 S RNA molecule were per- turbed, hydrophobic regions in the protein became ex- posed. Polyacrylamide gel analysis of the digestion products revealed a temporal appearance of discrete RNA fragments. Sequence analysis of these fragments generated information about the structural arrange- ment of the RNA molecule within the RNP complex. Results from the present investigation indicate that interactions between the 5 S RNA and protein L l a can stabilize functionally relevant conformations of the components that are individually labile. Properties of the separated components also suggest that special con- ditions, such as those suggested by Steitz et al. (Steitz, J. A., Berg, C., Hendrick, J. P., LaBranche-Chabot, H., Metspalu, A., Rinke, J., and Yario, T. (1988) J. Cell Biol. 106, 545-556) may be involved for these components to associate during ribosomal assembly.

RNA-protein interactions play critical roles in numerous recognition processes. Ribosomal 5 S RNA and its associated protein not only provides a convenient and amenable means for studying RNA-protein interactions in the ribosome (I), but also the complex has been shown recently to be a precursor to ribosome assembly in mammalian cells (2). Discrete 5 S RNA-protein complexes can be released from ribosomes of both prokaryotes and eukaryotes, although the number of protein species in the complex varies (3-11). In Saccharomyces cereuisiae, the ribonucleoprotein (RNP)’ complex consists of

* This work was supported by Grant GM 35851 from the National Institutes of Health (to. J . C. L.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accord- ance with 18 U.S.C. Section 1734 solely to indicate this fact.

The abbreviations used are: RNP, ribonucleoprotein; bis-ANS, bis-1-anilino-8-naphthalenesulfonic acid SDS, sodium dodecyl sul- fate.

one molecule each of 5 S RNA and protein Lla (yeast ribo- somal protein nomenclature of Ref. 12). The RNA-binding site on protein L l a appears to be located in a region consisting of 47 amino acid residues at the carboxyl terminus, although the amino-terminal region appears to confer specificity for the 5 S RNA molecule (8). The RNA-binding fragment is rich in basic amino acids, with 9 lysine and 3 arginine residues. Chemical modifications of the lysine and arginine residues abolish the ability of the protein to form stable RNA-protein complex (13).

Based on data from ribonuclease digestion of the RNA- protein complex, Nazar and his co-workers (14) proposed that the primary protein-binding sites in the yeast 5 S RNA resides in two separate locations of the molecule. One site is located in the 3’-end region encompassing residues 79-121 and the other in a small region of the molecule comprising residues 1-12 of the 5’-end (14). However, these RNA fragments by themselves are not sufficient to compete with intact 5 S RNA for the binding protein. The finding implies that additional RNA structures may contribute to the protein-binding do- main. Little is known about the nature of these additional features. The involvement of a third helical region, consisting of bases 14-20 and 58-66, in protein binding has been sug- gested later by RNA exchange studies with chemically modi- fied 5 S RNA (15). Recently, a variant 5 S RNA with one nucleotide shorter at the 3’-end than the mature RNA is shown to bind protein Lla with decreased affinity in vitro (16) and may not be assembled efficiently into ribosomes in vivo (17).

Because of the important potential role of the 5 S RNA- protein L la complex in ribosome assembly (2) and function (18), the present investigation is designed to elucidate further the molecular details of RNA-protein interaction in the 5 S RNA. Lla protein complex from S. cereuisiae using a combi- nation of procedures involving controlled enzymatic digestion, RNA sequencing, and fluorescence. Our results demonstrated that there are intimate interactions between regions of the RNA and the protein. Perturbation of these RNA regions resulted in an exposure of hydrophobic regions of the protein. A portion of the RNA molecule, uiz. the segment encompass- ing residues 50-85, appeared to contain sufficient information to displace intact 5 S RNA from the RNP complex.

EXPERIMENTAL PROCEDURES

Yeast Culture and Ribosome Preparation-S. cereuisiae (RNase 3A- strain) was grown in yeast extract/peptone/glucose medium and harvested in late exponential phase as described previously (19). Ribosomal subunits were isolated and purified by zonal centrifugation in 12-30% glycerol gradients (19). Purity of the ribosomal subunits was assayed by analytical centrifugation of the subunits in glycerol gradients and by polyacrylamide gel electrophoresis of the RNA isolated from the purified subunits. Only those ribosomal prepara- tions of greater than 96% purity were used for studies described herein.

18213

18214 RNA Structure in Yeast Ribonucleoprotein Complexes

Extraction of RNP from 60 S Ribosomes-RNP complexes contain- ing 5 S RNA were extracted from purified 60 S ribosomal subunits (200A2~/ml) with 25 mM EDTA, pH 7.0 at 4 "C for 60 min following mainly the procedure of Nazar (14). The RNP complex was further purified by centrifugation through a sucrose density gradient and stored at -70 "C until used.

Purification of 5 S RNA and Its Binding Protein LIa-5 S RNA and protein Lla were purified from isolated RNP complex using a DE52 column (1.5 X 2.0 cm) as described previously (14). Briefly, the column was preequilibrated with 10 mM Tris-HC1, pH 7.5, containing 0.15 M KCl. Protein Lla was eluted with 10 mM Tris-HC1, pH 7.5, 0.3 M KC1 and 6 M urea. The protein was recovered by lyophilization after dialysis. 5 S RNA was subsequently eluted with 10 mM Tris- HC1, pH 7.5, 1.0 M KCI, and 6 M urea and recovered by precipitation with ethanol.

Polyacrylamide Gel Electrophoresis of RNP Complexes, RNA, and Protein-5 S RNA and RNP complexes were identified by electro- phoresis on 8% polyacrylamide gels in TBE buffer (Tris/borate/ EDTA, pH 8.3) essentially as described (14). Total yeast RNAs were used as markers. The gels were stained with either methylene blue (0.2% in 0.8 M sodium acetate) for 16 h or EtBr (5 pg/ml) for 15 min and destained with water. Protein Lla was analyzed on SDS-poly- acrylamide gel (20). Either Coomassie Blue or silver were used to detect protein bands. Molecular weight standards used were pyruvate kinase (57,000), glyceraldehyde-3-phosphate dehydrogenase (36,000), and cytochrome c (12,500).

Fluorescence Spectroscopy-The fluorescence intensity of a solu- tion of RNA or RNP complex containing the indicated concentration of EtBr or bis-ANS was measured in a thermostated fluorescence cell with a Perkin-Elmer MPF-44A spectrofluorometer at 25 "C. Excita- tion and emission wavelengths were 540 and 600 nm for EtBr and 380 and 490 nm for bis-ANS, respectively. Ethidium fluorescence studies were carried out with 1 A260 unit of RNA or RNP complex in 1 ml of 25 mM EDTA and 25 p M of dye. Bis-ANS fluorescence studies were conducted with 1 A260 unit of RNP complex in 100 pl of 25 mM EDTA and 100 p~ dye.

Light Scattering-Light scattering measurement was made in a Perkin-Elmer MPF-44A spectrofluorometer at a wavelength of 354 nm. The RNP sample (1 A2~/0 .1 ml) was in 25 mM EDTA at 25 "C.

Determination of Nucleotide Concentration-RNA or RNP com- plex (I A260 unit) was hydrolyzed to mononucleotides in a sealed tube with 0.4 M NaOH at 37 "C overnight. Four volumes of 1 M NaP04, pH 7.0, were added. The absorbance of the mixture was determined at 260 and 290 nm. The concentration of nucleotides (mM) = (Azw-

Ribonuclease TI Digestion-5 S RNA or RNP complex was digested with RNase T1 (Calbiochem, final concentration 4 units of enzyme/ A260 unit RNA or RNP complex) in 25 mM EDTA at 25 "C. Digestion was monitored by fluorescence and electrophoresis on polyacrylamide gels. For analysis by electrophoresis, the sample was applied directly onto a nondenaturing polyacrylamide (8%) gel or a denaturing poly- acrylamide (12%) gel containing 8.3 M urea. Samples to be analyzed by denaturing gels were extracted with an equal volume of redistilled phenol prior to electrophoresis. RNA and its fragments were visual- ized under UV light after staining with EtBr. For sequence analysis and competition experiments, 5 S RNA and its fragments were extracted from the urea-polyacrylamide gels with phenol-chloroform and recovered by precipitation with ethanol.

RNA Sequence Determinution-Intact 5 S RNA and RNA frag- ments generated by RNase T1 digestion were treated with calf intes- tine alkaline phosphatase (21). Briefly, RNA molecules were dissolved in 30 pl of 25 mM Tris-HCl, pH 8.3, incubated for 5 min at 50 "C, and for an additional 30 min in the presence of calf intestine alkaline phosphatase (final concentration units of enzyme/pmol of RNA). The dephosphorylated RNA molecules were labeled with [3ZP]ATP (2 pmol/pmol of RNA, specific activity, >4,000 Cilmmol) and T4 polynucleotide kinase (5 units) in 50 mM Tris-HC1, pH 8.9, 10 mM MgClz, 10 mM dithiothreitol, and 5% glycerol at 37 "C for 60 min. After phenol-chloroform extraction and precipitation by ethanol, the labeled RNA fragments were purified by electrophoresis on urea- polyacrylamide gels (8%) in TBE buffer. Labeled molecules were located by autoradiography, extracted from the gel by phenol-chlo- roform, precipitated by ethanol in the presence of 2 pg of tRNA carrier, and stored at -20 "C until used.

Nucleotide sequence of RNA molecules was determined following basically the procedure of Donis-Keller et ul. (22). Briefly, labeled RNA molecules (about 25,000 cpm) with 5 pg of carrier tRNA was preincubated in the appropriate buffer at 50 "C for 5 min, quickly

A2w) X 0.1015.

chilled in ice, and digested with the appropriate RNase. For RNase T1 digestion, the buffer contained 20 mM sodium citrate, pH 5.0, 7 M urea, 1 mM EDTA, 0.02% xylene cyanol, and 0.02% bromphenol blue. Similar buffer was used for RNase U2 digestion, except the pH was adjusted to 3.5. Digestion was carried out with 0.5 or 0.25 unit of RNase T1 or U2, respectively, for 15 min at 50 "C. The digest was analyzed on a thin sequencing polyacrylamide (20%) gel containing 7 M urea in TBE buffer at constant power. A ladder marker containing a RNA sample partially hydrolyzed with alkali was also run on the same gel. RNA bands were detected by autoradiography with an intensifying screen at -20 "C.

Competition Experiments with RNA Fragments-The RNA frag- ment to be tested was labeled as described above and incubated with a known amount of RNP complex containing nonradioactive intact 5 S RNA in 25 mM EDTA, pH 7.0, at 25 "C for 30 min and at 4 "C for an additional 30 min. The reaction mixture was analyzed by electrophoresis on nondenaturing polyacrylamide (8%) gel. The gel was dried; radioactive RNA and RNP complexes were detected by autoradiography. Bands containing putative RNP complexes, as judged by shifted electrophoretic mobilities, were further analyzed for the presence of protein. The gel band of interest was extracted and analyzed on SDS-polyacrylamide gels together with protein Lla as a standard. Proteins were detected by silver staining.

RESULTS Characterization of Ribonucleoprotein Complex-As was

demonstrated previously the RNP complex extractable with 25 mM EDTA from purified yeast 60 S ribosomal subunit migrated as a single band on native gel with an electrophoretic mobility different from that of free 5 S RNA (Fig. 1, Lane 3) . Upon dissociation of the complex, only 5 S RNA and protein L l a could be detected (Fig. 1, Lanes 2 and 5).

Probing the RNA Conformation in the Ribonucleoprotein Complex by Ethidium Fluorescence-Published data indicated that strong fluorescence enhancement accompanies interca- lation of EtBr into double-stranded regions of nucleic acids (23, 24). In the present investigation, fluorescence was used, in conjunction with RNase T1 digestion, to monitor changes in the double-stranded character of the RNA in the RNP complex. As shown in Fig. 2, the ethidium fluorescence inten- sity changed as a function of time of RNase T1 digestion of the complex. An initial increase in fluorescence intensity was observed followed by a gradual decrease. For comparison, the fluorescence intensity of free 5 S RNA as a function of time of T1 digestion was also determined (Fig. 2). The fluorescence intensity of free RNA was about 1.3-fold higher than that of the RNP complex. Responses of free RNA and the RNP complex to RNase T1 as monitored by ethidium fluorescence appeared to be reproducibly distinct.

Probing the Conformation of Protein L la in the Ribonucle- oprotein Complex by Bis-ANS Fluorescence-The hydropho- bic probe, bis-ANS, has been shown to fluoresce intensely in hydrophobic environments and practically not at all in water (25). With this probe, we have shown previously presence of hydrophobic sites on protein L l a in the intact ribonucleopro- tein complex.' In the present study, the fluorescence intensity of bis-ANS was used to monitor conformation of the protein in the complex upon controlled digestion of the RNA with RNase T1. The fluorescence intensity increased as a function of time of T1 digestion (Fig. 3). A maximum of approximately 7-fold increase in bis-ANS fluorescence was observed. Hence, cleavage and subsequent removal of specific RNA regions from the RNP complex exposed additional hydrophobic sites on protein Lla. Exposure of these sites also led to increased insolubility of the complex, as measured by light scattering (Fig. 4). A slight lag period in the light scattering was noted. The protein appeared not to become insoluble until additional hydrophobic sites were exposed.

Yeh, L.-C. C., Horowitz, P. M., and Lee, J. C. (1988) J. Bid. Chem. 263, in press.

RNA Structure in Yeast Ribonucleoprotein Complexes 18215

5.0 S

5s

4s

1 2 3

57K

36K

12.5K

4 5

FIG. 1. Polyacrylamide gel electrophoresis showing RNP and its components. Lanes 1-3 show typical electrophoretic pat- terns of RNP complexes or RNA on 8% polyacrylamide gel in TBE buffer. Products were detected by staining with methylene blue. Lane 1, yeast 5.8 S RNA, 5 S RNA, and tRNA as markers; Lane 2, RNA isolated from RNP complexes; Lane 3, yeast RNP complexes isolated from purified yeast 60 S ribosomal subunits. Lane 4 shows a typical SDS-containing polyacrylamide gel electrophoretic pattern of molec- ular weight standards used in the calibration of molecular weight of proteins; pyruvate kinase (57,0001, glyceraldehyde-3-phosphate de- hydrogenase (36,0001, and cytochrome c (12,500). Lane 5 shows a typical electrophoretic pattern of the protein component in the ribo- nucleoprotein complex on SDS-containing Dolvacrvlamide gel. Direc- tion ofelectrophoresis was from top to bottom (- + +).

" - - ofelectrophoresis was from top to bottom (- + +). -

" - -

W

I- O

2 6SO!

0

Y z I -, E 550- ';.

W 0

6 450.- 0 v) W LK

\

5 350- -I LL

0 10 20 30 40

W

I- O

0 3 z

0

Y

I E ';. W 0 Z W

v) 0

W LK 0 -I 3 LL

TIME OF DIGESTION (rnin)

FIG. 2. Ethidium fluorescence of free 5 S RNA and 5 S RNA-Lla protein complex as a function of time of RNase T1 digestion. Isolated 5 S RNA ( 0 , l AZW) or ribonucleoprotein complex (0, 1 A2m) was digested with RNase T1 (4 units) in 1 ml of 25 mM EDTA a t 25 "C in the presence of 25 p~ ethidium bromide. Fluores- cence intensity is expressed in arbitrary units/mM nucleotide. Values shown represent the average of three independent determinations.

W D

3501 I

y W 1 5 0 ~ 0 , 0 ' ~ " : : : 1 W u v) W LK O 50 3 LL

0 20 40 60 80

TIME OF DIGESTION ( rnin)

FIG. 3. Bis-ANS fluorescence of ribonucleoprotein complex as a function of time of RNase T1 digestion. Ribonucleoprotein complex (1 A m in 0.1 ml of 25 mM EDTA) was digested with RNase T1 (4 units) a t 25 "C. Fluorescence intensity was monitored with 100 pM bis-ANS and expressed in arbitrary units/mM nucleotide. Values shown are the average of three independent determinations.

Fine Structure Analysis of the RNA in the Ribonucleoprotein Complex Using Controlled RNase TI and RNA Sequencing- A more detailed analysis of the T1 digestion products was carried out in order to obtain additional information about the arrangement of the RNA moiety in the RNP complex. Products produced during the initial 5 min of digestion were analyzed by electrophoresis on denaturing and nondenaturing polyacrylamide gels. As shown in Fig. 5A, a temporal appear- ance of discrete RNA fragments were produced by digestion of the R N P complex. The electrophoretic profile of RNA fragments generated by digestion of free 5 S RNA was dis-

700 7

600 -. o-o-o-o'o

500 -. 0' 400

-. 200

-. 0 300

-. 0' /

/ O/

100" o/ 0

0 6 9 ,

0 5 10 15

TIME OF DIGESTION (rnin)

FIG. 4. Light scattering of 5 S RNA-protein Lla complex as a function of RNase T1 digestion. Ribonucleoprotein complex (1 A2m in 0.1 ml of 25 mM EDTA) was digested with RNase T1 (4 units) a t 25 "C. Light scattering was measured a t 354 nm a t a 90" angle. Turbidity was expressed in arbitrary units.

A

1 2 3 4 5 6 7

FIG. 5. Electrophoretic patterns of RNase T1 digestion products of ribonucleoprotein complex ( A ) and free 5 S RNA ( B ) on denaturing polyacrylamide gels. RNP complex or RNA (1 A260 in 1 ml of 25 mM EDTA) was treated with the enzyme a t 25 "C. At the indicated time, aliquots were withdrawn, extracted with phenol, and applied onto a 12% denaturing polyacrylamide gel con- taining 8.3 M urea. RNA fragments were detected by ethidium stain- ing. Lane 1, molecular weight standards (5.8 S, 5 S, and 4 S RNA); Lane 2, RNP or RNA alone, no enzyme; Lanes 3-7: RNP complex ( A ) or RNA ( B ) digested for 0.5, 1.0, 2.0, 5.0, and 10 min. RNA fragments generated from digestion of RNP complex were numbered 21-1 starting from the top of the gel.

18216 RNA Structure in Yeast Ribonucleoprotein Complexes

tinctively different from that of the complex, although several RNA fragments were found in common (Fig. 5B). Base se- quences of a total of 21 fragments were determined, and representative autoradiograms of sequencing gels are shown (Fig. 6). Base sequences of all RNA fragments are shown in Fig. 7. Fig. 8A shows the electrophoretic patterns of digestion products on nondenaturing gels. Several distinct oligonucle- otide-protein complexes were detected which appeared in a temporal sequence. After 1 min of digestion, two major (la and 2a) and one minor (3a) oligonucleoprotein subparticles with shifted electrophoretic mobility appeared. After 2 min, the same bands were detected but the distribution was changed. After 5 min, band la became the prominent band. Nature of the oligonucleotides in individual oligonucleotide- protein complexes was examined by electrophoresis of these complexes in denaturing gels. A limited number of distinct oligonucleotides was evident (Fig. 8B). A comparison of Figs.

14 15 16 17 18 nnnnn

FIG. 6. Autoradiograms of sequencing gels of RNA frag- ments generated from RNase T1 digestion of RNP complex. Fragments 14-18 shown in Fig. 5A were analyzed. Nucleotide se- quences were determined with 5"end-labeled fragments and RNase T1 or U2 as described (22) on thin 20% polyacrylamide gels. Fragment numbers are indicated on the top. Four reactions were carried out for each fragment and are represented in four lanes (left to right): no enzyme, partial alkaline hydrolysis, RNase T1 digestion, and RNase U2 digestion.

5 6

I

111

FIG. 7. Nucleotide sequence of RNA fragments produced by RNase TI digestion of ribonucleoprotein complex. RNA frag- ments produced by T1 digestion of RNP complexes as described in Fig. 5A were resolved by electrophoresis on polyacrylamide gels containing 8.3 M urea. Sequences were determined with 5'-end- labeled fragments and RNase T1 or U2 as described (22).

RNA Structure in Yeast Ribonucleoprotein Complexes 18217

n

B

1 2 3 4

1 2 3 4 5

5A and 8B suggested that several oligonucleotides were re- leased from the complexes as a result of T1 digestion. Ribo- nucleoprotein subparticles la contained RNA fragments 3,4, and 6; subparticles 2a and 3a contained fragments 6, 8, and 12. Reasons for the slight difference in electrophoretic mobil- ities between subparticles 2a and 3a are not clear.

Exchange of 5 S RNA and Its Fragments into Ribonucleo- protein Complexes-The ability of each RNA fragment to exchange with intact 5 S RNA into the RNP complex con- taining intact 5 S RNA was measured by electrophoresis on nondenaturing gels. Radioactive RNA fragment was incu- bated with nonradioactive 5 S RNA-protein complex. The reaction mixture was analyzed by electrophoresis. Only those RNA fragments which contained residues 50-85 could ex- change with intact 5 S RNA in forming ribonucleoprotein complexes (Fig. 9 and Table I). The exchange was concentra- tion-dependent. In some cases, additional bands, whose elec- trophoretic mobilities were also shifted from that of the free oligomers, were observed. These extra bands did not contain protein; the exact nature of these minor bands is not clear a t present. That the major radioactive band was a ribonucleo- protein complex was confirmed by detection of one protein in these materials. The electrophoretic mobility of the protein in SDS-polyacrylamide gels was identical to that of standard L l a (data not shown).

FIG. 8. A, electrophoretic patterns of RNase T1 digestion products of ribonucleoprotein complexes on nondenaturing polyacrylamide gels. Digestion conditions were identical to those described in the legend of Fig. 5A, except that the digestion products were not ex- tracted with phenol and were applied directly onto a 8% nondenatu- ring gel. RNP subparticles were detected by ethidium staining and viewed under UV light. Lane I , molecular weight standards (5.8 S, 5 S, and 4 S RNA); Lane 2, RNP alone without enzyme; Lanes 3 and 4, RNP complexes treated with T1 for 1 and 2 min at 25 "C. B, electrophoretic patterns of RNA fragments isolated from ribonucle- oprotein subparticles generated from digestion of intact 5 S RNA. Lla protein complex. RNP subparticles as shown in A were treated with phenol-chloroform and analyzed on a denaturing 12% polyacryl- amide gel containing 8.3 M urea. Lane I , molecular weight standards (5.8 S, 5 S, and 4 S RNA); Lane 2, 5 S RNA alone; Lanes 3-5, RNA fragments isolated from RNP subparticles la, 2a, and 3a.

"

1 4

9 10 11 12

1 13 14 15 16 17

FIG. 9. Exchange of 5 S RNA and its fragments with 5 S RNA in intact ribonucleoprotein complexes. Radioactive 5 S RNA or its fragment was incubated with ribonucleoprotein complexes containing nonradioactive 5 S RNA for 30 min at 25 "C and an additional 30 min at 4 "C. The reaction mixture was loaded onto a 8% nondenaturing polyacrylamide gel. Free RNA and ribonucleoprotein complexes were detected by autoradiography. The following radioactive 32P-labeled RNA samples were used fragment 4 in increasing concentrations (Lanes 1- 4 ) ; fragment 6 in increasing concentrations (Lanes 5-8); fragment 15 in increasing concentrations (Lanes 9-12); fragment 19 in increasing concentrations (Lanes 13-16); and intact 5 S RNA (Lane 17). Arrows indicate the exchanged complex that contained protein Lla upon subsequent analysis.

18218 RNA Structure in Yeast Ribonucleoprotein Complexes

DISCUSSION

In the present study, the structure dynamics of the yeast 5 S RNA-protein complex was probed using a combination of techniques. Ethidium fluorescence indicated that free RNA was more accessible to the dye than RNA in the RNP com- plex. The present observation is in agreement with several other published observations. First, using the technique of hydrogen exchange, we previously demonstrated that 5 S RNA in the free form has slightly more secondary structure than that in the complex.* Secondly, similar results were obtained by CD analysis (26). On the other hand, the same study also reports that the complex binds almost as much ethidium as free RNA. We consistently observed that the free RNA bound an average of 20% more dye than the complex. In Escherichia coli (27) and Halobacterium cutirubrum (28), free 5 S RNA binds more ethidium than that in corresponding complexes.

TABLE I Exchangeability of intact 5 S RNA in complex

Fragment no. Residue no.

a. Exchangeable fragments 5 S RNA 21

1-121

20 8-121

22-121 19 26-121 18 31-121 17 38-121 16 42-121 15 50-121 14 26-91 13 31-91 12 31-87 11 38-91 10 42-91 9 42-89 8 42-81 7 50-91 4 50-85

516 81-121 3 81-115 112 90-121

b. Nonexchangeable fragments

\ \

It is noteworthy that the fluorescence data observed during RNase T1 digestion and the location of the sensitive sites correlates well with the published secondary structure model, i.e. ethidium fluorescence decreased as the RNA molecule was being digested and cleavage at the most T1-sensitive sites generated fragments which were released from the RNP. These fragments came from RNA regions that were involved in hydrogen bonds, thus ethidium-binding sites were de- stroyed reflecting in a decrease in fluorescence.

Kinetic studies of T1 digestion of intact RNP complexes revealed that several guanylic acid residues were extremely sensitive to the enzyme. These residues were located predom- inantly on the 5'-half segment of the RNA molecule. Cleavage of these residues resulted in the release of oligonucleotide fragments from the RNP complex. Although several residues which were located near the 3'-half of the molecule were very sensitive to RNase T1, cleavage of these residues did not result in release of oligonucleotide fragments from the RNP complex. A large number of guanylic acid residues were not cleaved by RNase T1 indicating that these residues might interact with the protein intimately and thus were protected from cleavage. Several of these nonsusceptible residues were previously shown to be involved in secondary and/or tertiary structure of the RNA molecule.

Previously, Nazar and his co-workers (11) reported that the protein binding site involved three helices, I, 11, and I11 (Fig. 10). Helix I encompasses approximately residues 1-12 and 110-121, whereas helix 11 encompasses residues 79-100 and helix I11 encompasses residues 14-20 and 58-66. However, none of these helical regions alone appear to contain sufficient information to compete with intact 5 S RNA for protein binding. The present study further defined that only the RNA region containing residues 50-121 constituted the protein- binding site (Fig. 10). Those residues containing the comple- mentary sequences, i.e. residues 1-12 and 14-20, were not part of the binding site. Presumably they were detected pre- viously because they were hydrogen-bound to the actual RNA sequence that formed the protein-binding site. The present results would suggest that the protein interacts with the RNA on one side of the RNA molecule. Such a suggestion would agree with the proposed tertiary structure model and the sidedness in the interaction between the RNA and the protein

1 .OCGAUC C U G G G A C U

00' c c c" I I I

I 1 1 1 II A I I<.*?,* : i . , 1 . ' I' I I

10' C G 0 C 0:UY G G P P(PI ik .

1 1

.. :...:.:::., ::: .:.,. . .. ... , .;,: ..:,, :: .:.::,::;-

I '' ....I'

. . . . . . . .

FIG. 10. A model of secondary structure of yeast 5 S RNA. The secondary structure was originally proposed by McDougall and Nazar (29) based on an estimation of base-pairing scheme of Nishikawa and Takemura (30). The RNase TI-sensitive guanylic acid sites are shown by arrows, together with an indication of the degree of sensitivity (the length of the arrow is proportional to the sensitivity of cleavage, - - - +, most sensitive; - - +, very sensitive; -+, moderately sensitive; +, sensitive; unmarked guanylic acid residues indicate that they were not cleaved under the present experimental conditions. The RNA region that remained sufficiently tightly associated with protein Lla after TI digestion is enclosed by broken lines. Also shown are shaded areas indicating sequences that were previously shown to be resistant to RNase digestion and remained in a RNP complex (14). solid indicate residues partially resistant to ethylnitrosourea modification in RNP (29). Solid circles indicate residues excluded from base modifications (15).

RNA Structure in Yeast Ribonucleoprotein Complexes 18219

as suggested by McDougall and Nazar (29). In addition, three bulges (A 51, A 64, and G 85) were found within the RNA fragment that interacted with the protein (Fig. 10). The portion of the RNA (residues 1-30) that was easily accessible to the ribonuclease did not contain any bulges. The physio- logical significance of this observation must await further experimentation.

Early reports indicated that the 5 S RNA-protein complex could efficiently exchange with intact 5 S RNA (11). The exchange was specific and neither yeast 5.8 S RNA, tRNA, nor H . cutirubrurn 5 S RNA would exchange. Nazar (14) also reported failure of the protein-protected RNA fragments to compete with intact 5 S RNA for protein binding. Results on the displacement experiments conducted in the present study indicated that fragments encompassing at least residues from 50 to 85 appeared to contain sufficient information to displace intact 5 S RNA from the RNP complex. The displacement was concentration-dependent. This contention is in agree- ment with our other observation that the 5’-half segment of the RNA remained associated with the protein after T1 diges- tion of the intact RNP complex.

The present study also revealed for the first time that removal of as much as 30 residues from the 5’-end of the RNA molecule did not seem to expose additional hydrophobic regions of the protein. However, further removal of residues, e.g. 31-49, resulted in a considerable exposure of hydrophobic sites on the protein. In fact, exposure of these hydrophobic sites led to insolubility of the RNP complex. This finding would suggest that the loop region containing residue 40 could be in direct contact with a hydrophobic region on the protein. Alternatively, disruption of this loop region perturbed the conformation of the protein sufficiently so as to expose hydrophobic regions not in direct contact with the RNA molecule. The present data could not discriminate these al- ternatives. Nature of these hydrophobic regions are being investigated.

REFERENCES 1. Nazar, R. N., Yaguchi, M., and Willick, G. E. (1982) Can. J.

Biochem. 60,490-496 2. Steitz, J . A,, Berg, C., Hendrick, J. P., LaBranche-Chabot, H.,

Metspalu, A., Rinke, J., and Yario, T. (1988) J. Cell Biol. 106, 545-556

3. Erdmann, V. A. (1976) Prog. Nucleic Acid Res. Mol. Biol. 18,45-

4. Blohel, G. (1971) Proc. Natl. Acad. Sci. U. S. A . 68, 1881-1885 5. Petermann, M. L., Hamilton, M. G., and Pavlovec, A. (1972)

6. Isoda, N., Tanaka, T., and Ishikawa, K. (1981) J. Biochem.

7. Marion, M. J., and Reboud, J. P. (1981) Biochim. Biophys. Acta

8. Smith, N., Matheson, A. T., Yaguchi, M., Willick, G. E., and

9. Terao, K., Takahashi, Y., and Ogata, K. (1975) Biochim. Biophys.

10. Chen-Schmeisser, U., and Garrett, R. A. (1977) FEES Lett. 74,

11. Nazar, R. N., Yaguchi, M., Willick, G. E., Rollin, C. F., and Roy,

12. Michel, S., Traut, R. R., and Lee, J. C. (1983) Mol. Gen. Genet.

13. Vioque, A,, Hernandez, F., and Palacian, E. (1987) Mol. Cell.

14. Nazar, R. N. (1979) J. Biol. Chem. 254, 7724-7729 15. Nazar, R. N., and Wildeman, A. G. (1983) Nucleic Acids Res. 11,

16. Brow, D. A. (1987) J. Biol. Chem. 262, 13959-13965 17. Piper, P. W., Lockheart, A., and Patel, N. (1984) Nucleic Acids

18. Nygard, O., and Nilsson, L. (1987) Biochim. Biophys. Acta 908,

19. Lee, J. C., Henry, B., and Yeh, Y. (1983) J. Biol. Chem. 258,

20. Laemmli, U. K. (1970) Nature 227,680-685 21. Vournakis, J. N., Celantano, J., Finn, M., Lockard, R. E., Mitra,

T., Pavlakis, G., Troutt, A., van den Berg, M., and Wurst, R. M. (1981) in Gene Amplification and Analysis (Chirikjian, J. G., and Papas, T. S., eds) Vol. 2, pp. 268-294, Elsevier/North- Holland, New York

22. Donis-Keller, H., Maxam, A. M., and Gilbert, W. (1977) Nucleic Acids Res. 4, 2527-2538

23. Waring, M. J. (1965) J. Mol. Biol. 13, 269-282 24. LePecq, J.-B., and Paoletti, C. (1967) J. Mol. Biol. 27, 87-106 25. Rosen, C. G., and Weber, G. (1969) Biochemistry 8, 3915-3920 26. Willick, G. E., Nazar, R. N., and Van, N. T. (1980) Biochemistry

27. Feunteun, J., Monier, R., Garrett, R., Le Bret, M., and Le Pecq,

28. Nazar, R. N., Willick, G. E., and Matheson, A. T. (1979) J. Biol.

29. McDougall, J., and Nazar, R. N. (1986) FEES Lett. 209, 52-56 30. Nishikawa, K., and Takemura, S. (1974) FEES Lett. 40, 106-

90

Biochemistry 11, 2323-2326

(Tokyo) 90,551-554

652,193-203

Nazar, R. N. (1978) Eur. J . Biochem. 89,501-509

Acta 402,230-237

287-291

C. (1979) Eur. J . Biochem. 102,573-582

191, 251-256

Biochem. 76, 141-146

3155-3168

Res. 12,4083-4096

46-53

854-858

19,2738-2742

J. B. (1975) J. Mol. Biol. 93,535-541

Chem. 254, 1506-1512

109