THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 267, No. 17, Issue of June 15, pp. 11972-11976,1992 Printed in U. S. A.

Hyperprocessing of tRNA by the Catalytic RNA of RNase P CLEAVAGE OF A NATURAL tRNA WITHIN THE MATURE tRNA SEQUENCE AND EVIDENCE FOR AN ALTERED CONFORMATION OF THE SUBSTRATE tRNA*

(Received for publication, October 14, 1991)

Yo KikuchiS and Noriko Sasaki From the DeDartment of Molecular Biologv. Mitsubishi Kasei Institute of Life Sciences, 11 Minamiooya, Machida-shi,

I“ I

Tokyo 194, Japan

In the transposon copia-related retrovirus-like par- ticles of Drosophila, a 39-nucleotide-long fragment from the S’-region of Drosophila initiator methionine tRNA (tRNAPt) is used as the primer for copia minus- strand reverse transcription. This primer tRNAyt fragment is thought to be produced by cleavage within the mature tRNAPt sequence. We call this cleavage hyperprocessing. We have previously reported that catalytic RNA of RNase P from Escherichia coli (MlRNA) cleaves the synthetic tRNAP precursor in vitro at several sites within the mature tRNA sequence. Based on this result, we proposed a model for formation of the primer tRNA fragment involving RNase P. Here we show that natural tRNAPt prepared from Drosoph- ila adult flies can be cleaved by MlRNA. Using mutant tRNA? substrates, we also show that these cleavages are dependent on the occurrence of an altered confor- mation of the tRNA substrate. This is evidence that a tRNA can exist in aqueous solution at least in part in an altered conformation.

Processing of RNA is a crucial stage in the production of various forms of RNA, in that precursor RNA is cleaved or modified to form mature functional RNA (Abelson, 1979). In a previous paper (Kikuchi et al., 1990), we have proposed a new concept, “hyperprocessing.” Hyperprocessing may be de- fined as further processing of mature RNA that produces another functional RNA. A t present, we know only one ex- ample of hyperprocessing in biological systems: the formation of primer molecules for minus-strand reverse transcription of copia retrovirus-like particles (RVLPs)’ of Drosophila (Kiku- chi et al., 1990).

Drosophila cells contain RVLPs that have about 5 kilobases of RNA, homologous to the transposon copia (Shiba and Saigo, 1983). In the RVLP, a 39-nucleotide-long fragment from the 5’-region of Drosophila initiator methionine tRNA (tRNA?) is used as the primer for copia minus-strand reverse transcription (Kikuchi et al., 1986). To function as the primer, the tRNA? must be cleaved in vivo at the site between nucleotides 39 and 40. This further processing is an example of hyperprocessing. We have previously demonstrated that

* This work was supported in part by a grant from the Ministry of Education, Science and Culture of Japan. 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 accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom correspondence should be addressed Mitsubishi Kasei Institute of Life Sciences, 11 Minamiooya, Machida-shi, Tokyo 194, Japan. Tel.: 427-24-6286; Fax: 427-29-1252.

The abbreviations used are: RVLPs, retrovirus-like particles; tRNA?, initiator methionine tRNA.

the catalytic RNA (MlRNA) of RNase P of Escherichia coli (reviewed by Altman, 1989) is able to cleave a synthetic precursor of tRNA? at several sites within the mature tRNA? sequence (Kikuchi et al., 1990). One of these cleavage sites is between nucleotides 39 and 40. Based on this result, we have proposed a possible model for formation of the primer tRNA fragment involving RNase P (Kikuchi et al., 1990). This in vitro cleavage appeared to be dependent on the occur- rence of a conformational change of the tRNA?. To explain the cleavage between nucleotides 39 and 40, we supposed a different secondary structure than the cloverleaf conforma- tion for tRNAYt (Kikuchi et al., 1990).

In this paper, we report that fully modified mature tRNA? prepared from Drosophila adult flies is also cleaved by MlRNA at five sites and that the tRNA”’t can exist in aqueous solution at least in part in the altered conformation that is recognized and cleaved by MlRNA.

MATERIALS AND METHODS

Enzymes and Chernicals-The catalytic RNA subunit of RNase P of E. coli (MlRNA) was prepared by in vitro transcription with T7 RNA polymerase of plasmid pYAY (Kikuchi and Ando, 1989) as describedpreviously (Kikuchi et al., 1990). This preparation was used as catalytic MlRNA throughout this study. Other enzymes and chemicals were purchased from commercial sources as described (Kikuchi et al., 1990).

Preparation of the Natural tRNA? from Drosophila Adult Flies- Drosophila adult flies were kindly supplied by K. Kuroda and T. Shinomiya of Mitsubishi Kasei Institute of Life Sciences. Total RNA was extracted by the hot phenol method according to Jowett (1986). The tRNA? was purified essentially as described (Silverman et al., 1979). As the final step for this purification, two-dimensional poly- acrylamide gel electrophoresis was used as described (Beier et al., 1984). The purity and sequence of the terminally labeled tRNA? was confirmed by enzymatic sequencing as described (Krupp and Gross, 1983).

Preparation of Synthetic Substrates-The wild type substrate was transcribed in vitro with T7 RNA polymerase from plasmid pDtY as described (Kikuchi et al., 1990). To obtain mutant substrates, the ApaI-PstI fragment (nucleotides 33-76 in tRNA sequence) of pDtY was replaced by synthetic DNA of the desired sequence. Three mutant plasmids, pV42, pV67, and pW, were constructed (see Fig. 3). The mutant DNAs were synthesized using an Applied Biosystems DNA synthesizer. The mutant substrates, V42, V67 and W, were also prepared by transcription in vitro with T7 RNA polymerase from these plasmids restricted by PstI as described (Kikuchi et al., 1990).

Cleavage Reactions with MlRNA-Cleavage reactions were per- formed essentially as described (Guerrier-Takada et al., 1983; Kikuchi et al., 1990). The complete reaction mixture contained 50 mM Tris- HCl (pH 7.6), 100 mM NH4Cl, 60 mM MgC12, 5% (w/v) polyethylene glycol, 1 pg of MlRNA (transcript of pYAY), and 2.5 fmol of 3’- or 5’-end-labeled substrate RNA in a total volume of 10 pl. Mixtures were incubated at 37 “C for the appropriate time, and reactions were stopped by addition of 2 pl of 0.5 M EDTA. The products were separated by electrophoresis through 20% polyacrylamide, 8 M urea gels and were detected by autoradiography or pictrography using a

11972

Hyperprocessing of tRNA by M I RNA 11979

Hio-Image Analyzer BAS2000 (Fuji Film Co.) (Amemiya and Miya- hnra, 1988).

Cloning, Nuclcic Acid Manipulations, and Anal.vtical M~thods- L)NA manipulations and cloning techniques were performed essen- tially as descrihed hy Maniatis et al. (1982). Laheling, controlled nlkaline or acid hydrolysis, gel electrophoresis of RNA, and other analytical methods were as descrihed by Krupp and Gross (1983).

RESULTS

Cleavage of the Natural Mature tRNA?-We have previ- ously reported that the RNA component of Escherichia coli RNase P (MlRNA) cleaves a synthetic Drosophila tRNA?"' precursor at nine sites within the mature tRNA sequence in vitro (Fig. 1) (Kikuchi et al., 1990). To test whether the natural tRNA?'' is also cleaved by MlRNA, we used the natural mature tRNA:'" prepared from Drosophila adult flies as sub- strate for the MlRNA reaction. As shown in Fig. 2, the mature tRNA?" was also cleaved by MlRNA at several sites. Five fragments visible in Fig. 2, lane b (arrows 5-9) indicate cleavage between nucleotides 37 and 38,38 and 39,39 and 40, 54 and 55, and 59 and 60, respectively. These sites correspond to arrows 5-9 in Fig. 1. This result indicates that the mature tRNA?" prepared from Drosophila adult flies can be cleaved by MlRNA in the same manner as the synthetic tRNA?" precursor with respect to cleavages at these five sites. In the same reaction condition, however, the cleavages of the mature tRNA?" at sites 1-4 (Fig. 1) could not be detected (data not shown). Although we did not pursue this point further, this is probably due to the presence of 1-methylguanosine and N-2- methylguanosine at nucleotides 9 and 10, respectively, in the mature t R N A y (Silverman et al., 1979).

Evidence for a n Alternative Conformation of tRNA?-To explain the cleavage between nucleotides 37 and 38, 38 and 39, and 39 and 40 (Fig. 1, arrows 5-7) by MlRNA, we supposed a stem structure consisting of the base pairing between nucleotides 40-44 and 65-69 in the tRNAp"' (Fig. 3 A ) (Kikuchi et al., 1990). If tRNA?" takes this altered conformation, this substrate is cleavable by MlRNA, because the smallest substrate for MlRNA has been reported to retain only the domain of the aminoacyl stem, the T W stem and loop, and the 3'-end CCA sequence (McClain et al., 1987; Altman, 1989). To test this prediction, we have constructed

c 3 '

5'

4 +lJ G c

G-C C-G

I G-C A - 3 9

" A

c

"t C

8

c 'r

FIG. 1. Synthetic Drosophila t R N A p precursor and the cleavage sites with MlRNA. Arrows 1-9 indicate the cleavage sites within the mature t R N A sequence. The thick arrow indicates the predominant normal cleavage site (hetween -1 and 1 ) .

S-

a b c d

"1c FIG. 2. Cleavage of the 5'-end-laheled mature tRNA?" pre-

pared from Drosophila adult flies: electrophoretic annlysin of reactions. Mixturrs wr're incr~hntc~tl for :I h nnd nnnlyzwl on n 20'; polvarrylamide. 8 M urea gel. An ;tr~torndiojirnm of t he gc.1 is shnwn. Imnr a, minus MlRNA control: lnnc h. the rompletr renrtinn mixture.; lanr c, partial nridir digest of the stlhstrnte: h n r . d. pnrtinl digest of the suhstrate hy RNase T l . 'The nurleotitlr sequenre o f reqidt~r~s .32- 6 2 is shown on the riflhf. Assignment o f the ntlrlentidr s e q u r " ~ wns done hv G mapping from lnnr, d and the tlntn from Silvrrmnn rf 01. (1979). Arron's 5-9, fragments found in lanr h. Thrse ntlmt)ers cnr- respond to the sites shown in Fiji. 1 . ,<, s~~t)s tmte .

three mutant substrates, V42, V67, and W (Fig. 3). V42 has a single base change of G"' to C. V67 has a single base change of C"' to G. These mutations, therefore, create C-C and G-G mismatches, respectively, in the putative stem and are thought to destroy this stem. W is the double mutant, havinE both V42 and V67 mutations and is thought to maintain the stem structure (Fig. 3A ). These mutant tRNA>"" precursors were labeled at their .?'-end by [S'-.'.'P]pCp and T4 RNA ligase and used as substrates for the MlRNA reaction. Fig. 4 shows the time course of these reactions. The wild t.ype substrate was cleaved at seven sites within the mature tRNA sequence besides the major site producing mature-sized tRNA (Fig. 4, lane W7'120). Fragments 5-7 indicate the cleavages shown by arrows fi-7 in Fig. 1 (the sites between 37 and 38, 38 and 39, and 39 and 40, respectively), as previously described (Kikuchi et al., 1990). As expected, V42 and V67 could not be cleaved by MlRNA at these sites (Fig. 4. arrou-s -5-7). The

11974 Hyperprocessing of tRNA by M I RNA

FIG. 3. Schematic representation of wild type and mutant substrates. These substrates were synthesized hy in oitro transcription from plasmids pDtY, pV42, pV67, and pW, as descrihed under "Materials and Methods." The 3'-ter- minal C derives from the PstI restriction enzyme site of the plasmid template. A , the secondary structure necessary for cleavages 5-7 (thick arrows). Thin ar- rows indicate the base changes in mutant substrates. V42 has the base change of G to C at position 42. V67 has the base change of C to G a t position 67. W has both changes. R, location of the muta- tions in the cloverleaf structure.

V 4 2

A

r L

5 6 7 A 5 '

V 6 7 WT

Substrate+

Mature slzed- IRNA+

4 - 5 C - 6 - 1

FIG. 4 . Cleavage of t he .?'-end- labeled tRNA:"" precursors hy M I R N A . The mutant (L.42, YG, and LV) and wild t.ype ( "7') substrates were incuhated with .MIHNA for 0. 10. 30, GO and 120 min (as indicated on the f o p of each I a n r ) . Mixtures were analyzed in 20% po~yacry~amide, 8 hl uren gels. A pictrogram of liio-Image-Analyzer HAS2000 is shown. 'The posit ions of suh- strate, mature-sized tRNA and the proti- ucts 5, f; and 7 are indicated, Arrorr~s 5. 6 and 7 correspond to the sites shown in Fig. 3A.

double mutant W, however, could be cleaved by MlRNA at the sites (Fig. 4, arrows 5-7) even more efficiently than the wild type substrate. These results support the existence of base pairing between nucleotides 40-44 and 65-69 and are strong evidence for the proposed stem structure.

Cleavage of the Mutant tRNA?" Precursors at the Site between Nucleotides 8 and 9-As shown in Fig. 4, the major product of the cleavage reaction of the wild type and V42 substrates was the mature sized tRNA. In the reaction using V67 and W substrates, however, the major product seemed to be smaller and appeared to be the product of cleavage at the site between nucleotides 8 and 9. To confirm this major cleavage site, the reaction mixtures of W, V42, and V67 were electrophoresed together with the ladders from alkaline par- tial digests of W and V42. Fig. 5 shows that the major cleavage in W and V67 occurred at the site between nucleotides 8 and 9 (Fig. 5 , lanes a and c ) . This cleavage can be explained by a conformational change of the mutant substrates. The intro- duction of a base change of Cfi7 to G in V67 and W makes i t possible for these mutant substrates to form 9 base pairs

between nucleotides 7-15 and 6 5 7 3 (including two G-U pairs, Fig. 6). By this mutation, the D stem of the tRNA may be unfolded and these 9 base pairs may be newly formed. In particular, the reaction using W substrate produces almost no mature-sized tRNA but exclusively the smaller one (Fig. 4). W contains the V42 mutation in addition to V67. The V42 mutation creates a C-C mismatch in the anticodon stem of the cloverleaf form. Therefore, the instability of the anticodon stem may enhance this conformational change.

DISCUSSION

We have demonstrated that MlRNA can cleave natural t R N A y prepared from Drosophila adult flies a t several sites within the mature tRNA sequence. The five sites cleaved in the natural tRNA?" (Fig. 2) were exactly the same as that of the synthetic tRNAp"' precursor (Kikuchi ~t al., 1990). We have also observed that, under the same conditions as de- scribed above for tRNA;"' reaction, no cleavage of Ilrosophila

Hyperprocessing of tRNA by M I RNA 1197s

a b c d e F

"E

FIG. 5. Determination of major cleavage sites in mutant substrates. The 3"end-labeled V42, V67 and W were cleaved by MlRNA and electrophoresed with partial alkaline digests of the substrates in a 15% polyacrylamide/8 M urea gel. A pictrogram of Hio-Image-Analyzer BAS2000 is shown. Lunes a, c, and d , the reaction mixtures of W, V67, and V42, respectively; lanes 6 and e, partial alkaline digests of W and V42, respectively. The nucleotide sequence of residues -7 to 13 is shown on the right. The nucleotide sequence was determined by comparison of the degradation patterns of these RNAs (lanes 6 and e ) with those of partial alkaline digests in enzy- matic sequencing gel (data not shown). These bands can easily be recognized by the characteristic spacing they produce in the ladders; ~ . g . removal of a G residue results in a more pronounced increase of electrophoretic mobility as compared with removal of A, U, or C (Krupp and Gross, 1983). The open and closed arrows indicate the cleavages between -1 and 1 and between 8 and 9, respectively. S, substrate.

I l1.G-C C -G G - u / C 1n wild type C - G

Id type

C A G - C

C A c A U.35

FIG. 6. Possible secondary structure of the substrate W. Thick arrow, the major cleavage site in the substrate W with MlRNA. Thin arrows, the positions of mutations in the substrate W.

glutamine tRNA by MlRNA occurs.' This indicates that the susceptibility of the mature tRNA;'" to MlRNA is a unique property of tRNA?"'. Fragment 7 (Fig. 2, arrow 7) is the 5'- fragment consisting of nucleotides 1-39 of tRNA?, which is identical with the primer for copia minus-strand reverse tran- scription (Kikuchi et al., 1986). The formation of fully modi- fied 5"fragment (nucleotides 1-39) from natural tRNA?''' supports the model previously proposed (Kikuchi et al., 1990) for copia primer formation.

Using the mutant substrates, we have also demonstrated that these cleavages are dependent on the occurrence of an altered conformation of the tRNA;'". Since the natural tRNA?" was cleaved by MlRNA at the same sites as that of the synthetic tRNA,"" precursor, it is suggested that this natural tRNA?" can also exist in aqueous solution in part in the same altered conformation as the synthetic tRNA;'" precursor. Recently, Hall et al. (1989) compared the structure of unmodified and fully modified yeast phenylalanine tRNA and reported that the structure of these two molecules is similar under high M$+ concentration. Our experimental results presented in this paper suggest that the minor altered conformation of modified and unmodified tRNA,"' may also be similar. In the reaction using wild type substrate (Fig. 4). the production of mature-sized tRNA reached almost a pla- teau in 60 min, but the cleavages 5-7 proceeded linearly up to 120 min. The time course curves (data not shown) of these reactions were confirmed by quantitative analyses of this radiogram (Fig. 4) using the Bio-Image-Analyzer RAS2000. The time course of the reaction indicates that the altered conformer necessary for the cleavages 5-7 may be dynamically supplied from the common L-shaped structure of tRNA dur- ing the MlRNA reaction, rather than only from the cleavage of a statically pre-existing alternate conformer, which may have been produced during substrate preparation. I t seems likely that all possible conformers are interconvertible into each other. In Fig. 4, we have detected more efficient cleavages of the substrate W at the sites (arrows 5-7) than that of the wild type substrate. This may be explained by the instability of the normal conformation (common L-shaped structure) of the substrate W caused by the mutations. Because of its instability, this mutant substrate may more likely accept the altered conformation responsible for the cleavages 5-7 (Fig. 4), compared with the wild type substrate.

We have incidentally found that the V67 mutation causes an abnormal processingof the tRNA;'"' precursor by MlRNA. V67 and W substrates were mainly cleaved at the site hetween nucleotides 8 and 9 (Figs. 4 and 5). From the substrate specificity of MlRNA reported by McClain et al. (1987) and the true existence of an altered conformation of tRNA;'" as evidenced in this paper (Figs. 3 and 4), we have proposed a secondary structure of the W substrate (Fig. 6). Although this secondary structure has not been directly confirmed, it seems reasonable to postulate that this conformation may be nec- essary for cleavage between nucleotides 8 and 9. This indicates that an extensive conformational change may be caused by only one single mutation in tRNA?"' sequence in oitro. It will be interesting to know whether this abnormal processing of tRNA? precursor containing the V67 mutation also occurs in vivo.

The hyperprocessing of tRNA? in the copia system, as well as other hyperprocessing systems, if any, may arise from aberrant processing. In case of copia, it is possible that, in a period when the copia was generated, one of the aberrant processing steps was picked up and included in the copia

'Y. Kikuchi and N. Sasaki, unpublished ohservation.

11976 Hyperprocessing of tRNA by M1 RNA

system. Aberrant processing may therefore sometimes have played an important role in the evolution of biological sys- tems.

Acknowledgments-We thank Dr. F. Hishinuma for his encourage- ment, Drs. K. Kuroda and T. Shinomiya for supplying Drosophila adult flies, and F. Ozawa for the synthesis of oligonucleotides. We are especially grateful to Prof. H. J. Gross for his careful review of the manuscript and helpful comments.

REFERENCES Abelson, J. (1979) Annu. Reu. Biochem. 48, 1035-1069 Altman, S. (1989) Adu. Enzymol. 62 , 1-36 Amemiya, Y., and Miyahara, J. (1988) Nature 336 , 89-90 Beier, H., Barciszewska, M., Krupp, G., Mitnacht, R., and Gross, H.

Guerrier-Takada, C., Gardiner, K., Marsh, T., Pace, N., and Altman,

Hall, K. B., Sampson, J. R., Uhlenbeck, 0. C., and Redfield, A. G.

J. (1984) EMBO J. 3 , 351-356

S. (1983) Cell 35,849-857

(1989) Biochemistry 28, 5794-5801

B., ed) pp. 275-286, IRL Press, Oxford Jowett, T. (1986) in Drosophila: A Practical Approach (Roberts, D.

Kikuchi, Y., and Ando, Y. (1989) Agric. Biol. Chem. 53, 2005-2008 Kikuchi, Y., Ando, Y., and Shiba, T. (1986) Nature 323,824-826 Kikuchi, Y., Sasaki, N., and Ando-Yamagami, Y. (1990) Proc. Natl.

Acad. Sci. U. S. A . 87, 8105-8109 Krupp, G., and Gross, H. J. (1983) in The Modified Nucleotides in

Transfer RNA 11: A Laboratory Manual of Genetic Analysis, Iden- tification and Sequence Determination (Agris, P. F., and Kopper, R. A., eds) pp. 11-58, Alan R. Liss, Inc., New York

Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

McClain, W . H., Guerrier-Takada, C., and Altman, S. (1987) Science

Shiba, T., and Saigo, K. (1983) Nature 302, 119-124 Silverman, S., Heckman, J., Cowling, G. J., Delaney, A. D., Dunn, R.

J., Gillam, I. C., Tener, G. M., Soll, D., and RajBhandary, U. Z. (1979) Nucleic Acids Res. 6, 421-433

238,527-530