Site-specific cleavage by T1 RNase of U-1 RNA in U-1

Transcription

1 Proc. NatL. cad. Sci. US Vol. 78, No., pp , March 98 Biochemistry Site-specific cleavage by T RNase of U- RN in U- ribonucleoprotein particles (small nuclear RN/RN processing/systemic lupus erythematosus antibodies) PUL EPSTEIN, RMHNDR REDDY, ND HRRIS BUSH Department of Pharmacology, Baylor ollege of Medicine, Houston, Texas 7700 ommunicated by Maxine F. Singer, December 5, 980 BSTRT The structures and functions of small nuclear ribonucleoprotein particles have become of interest because of their suggested role in processing heterogeneous nuclear RN [Lerner, M. R., Boyle, J.., Mount, S. M., Wolin, S. L. & Steitz, J.. (980) Nature (London) 28, ]. To determine the conformation of U- RN in U- ribonucleoprotein particles and whether proteins of these particles protect segments of U-I RN, intact particles and isolated U- RN were digested with Ti RNase. The digested particles were immunoprecipitated with anti-sm antibodies. 5'-end fragment containing nucleotides -7 and '-end fragments containing nucleotides 8-5 and 8-5 were recovered in nearly quantitative yield from digestion of the particles, suggesting that position 7 is the principal cleavage site in them. t the same Ti RNase concentrations, deproteinized U-I RN was cleaved into many fragments. t low TI RNase concentrations, a major cleavage site of deproteinized U- RN was at nucleotide 69. omparison of the cleavage sites of free U- RN and of U- RN in U- ribonucleoprotein particles suggested similar secondary structures. The resistance of the 5' end of U- RN to T RNase was unexpected inasmuch as this region has been implicated in hydrogen bonding with heterogeneous nuclear RN splice junctions. U- RN exists in the nucleus in small nuclear ribonucleoprotein (RNP, snrnp) particles (-). Suggestions that U- RN is associated with heterogeneous nuclear RN (hnrn) (2-) have led to studies that have shown that the 5' end of U- RN may base pair with intron-exon regions of hnrn (, ). s an approach to the structure of U- snrnp particles and to identify regions of U- RN that are bound to proteins, the present study was undertaken; it combines partial cleavage of the RNP with immunoprecipitation of the RNP particles. This study showed that the 5'-end 7 nucleotides and most of the ' end of U- RN are resistant to T RNase cleavage and are probably protected in several regions by the proteins of the U- RNP particle. MTERIL ND METHODS Harvested Novikoff hepatoma ascites cells were incubated in Eagle's medium with [2P]orthophosphate as described (5). The nuclei were isolated by homogenizing the cells in a Tissumizer in mm Nal/.5 mm Mgl2/0.5% Nonidet P-0/ mm The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 8 U. S.. 7 solely to indicate this fact. 562 Tris buffer, ph 7.2, sedimented by centrifugation at 000 x g for min (6), and then sonicated in 0 mm Nal/ mm Mgl2/ mm Tris buffer, ph 8.5 until no intact nuclei were visible (2). The sonicates were centrifuged at 20,000 X g for min, and the supernatant was used. The RN concentrations of the supernatants were determined spectrophotometrically: one OD2w was equivalent to 50,ug. The usual range of the sonicate concentration was kkg/ml. The supernatant was incubated with T RNase at an enzyme/substrate ratio of : or :5 (wt/wt) or without enzyme for 0 min at 0. Preliminary experiments showed identical RN patterns and yields after T RNase digestion when either antigen or antibody was used in excess. Immunoglobulin fractions of sera from patients with systemic lupus erythematosus previously demonstrated to precipitate U-, U-2, U-, U-5, and U-6 RNs (anti-sm sera) were then added at 2 mg/ml of sonicate, and the mixtures were incubated for 5 min on ice. The immune complexes were precipitated by addition of Pansorbin at ml/2 mg of immunoglobulin (2), followed by 5 min of incubation on ice. The precipitates were washed five times with 50 mm Nal/5 mm EDT/0.05% Nonidet P-0/50 mm Tris buffer, ph 7.. The RN was extracted from the Pansorbin pellets by using the hot phenovnadodso procedure, precipitated with 2 vol of ethanol, and then subjected to electrophoresis on % polyacrylamide/7 M urea, ph 8., gels. fter the gels were subjected to autoradiography, the bands were excised and the RN was extracted as described (7). omplete digestion of the RN with T or pancreatic RNase and separation of the resultant oligonucleotides was carried out as described by Brownlee et al. (8). Polyethylenimine-cellulose sheets were used for homochromatography in the second dimension. To determine the action of T RNase on deproteinized U- RN, gel-extracted 2P-labeled U- RN was added to unlabeled nuclear sonicates prepared as described above. The sonicates were then incubated with Ti RNase at enzyme/substrate ratios ranging from :5 to :5 for 5 min at 0. The RN was then extracted, subjected to gel electrophoresis, and identified as described above. The T and pancreatic RNases and Pansorbin were obtained from albiochem. The polyethylenimine-cellulose sheets were obtained from Brinkmann. bbreviations: RNP, ribonucleoprotein; snrnp, small nuclear ribonucleoprotein; hnrn, heterogeneous nuclear RN.

2 Biochemistry: Epstein et al 2 U-2 -U-._~ ~~UJ U-ID - U-lB - U-l U-B U-I FIG.. utoradiograph of RN extracted from Pansorbin pellets and subjected to electrophoresis on a % polyacrylamide/7 M urea, ph 8., gel. Nuclear sonicates were incubated with T RNase at enzyme/substrate ratios (wt/wt) of : (lane ) or :5 (lane 2) and anti- Sm antibody as described in Materials and Methods. RESULTS n autoradiograph of a % polyacrylamide gel on which RN from the Pansorbin-treated pellets was separated is shown in Fig.. Each slot of the gel contained the products obtained from an aliquot of the nuclear sonicate and an equal amount of each immunoglobulin fraction. utoradiographs of the complete T RNase fingerprints of U- RN and U- RN fragments, B,, and D are shown in Fig. 2. The average molar yields of these fragments (considering 0% as the 2P nucleotide of the U- RN precipitated when T RNase was not added) were 77%, 5%, 7%, and 9%, for U-, U-, U-B, and U-ID, respectively. Under these conditions, less than 5% of the U- RN was undigested. The Ti RNase fingerprint of the major fragment (U-I) (see Proc. Nati cad. Sci. US 78 (98) 56 Fig. 2B) contained spot 2, which is the 5' end of U- RN (9). Fragment U-I was also digested with nuclease P and subjected to electrophoresis on DEE paper at ph.5 (not shown); it contained the "cap." Therefore fragment U-I contained nucleotide of U- RN. Fragment U-i also contained oligonucleotide T-22 (see Fig. 5), which is composed of nucleotides 95-7 of U- RN. Inasmuch as fragment U-I contained all the Ti RNase fragments from the 5' end of U-i RN, including spot 22 (nucleotides 95-7), and the next guanosine phosphate (at position 9) was found in fragments U- and U-B in a combined molar yield of 9% (see below), fragment U- extended from nucleotide to nucleotide 7 of U- RN. The Ti RNase fingerprint of fragment U-lB (see Fig. 2) contained spot 7, the '-end fragment. This fragment did not contain spot 22, nucleotides 95-7 of U- RN, but did include spot, nucleotides 2-8. In addition, the complete pancreatic RNase fingerprint of fragment U-lB included the oligonucleotide G-G-G----p, which suggests that this fragment extends in the 5' direction to nucleotide 8. Therefore, fragment U-lB contained nucleotides 8-5 and thus represents the portion of the U- RN structure not included in fragment U-. The T RNase fingerprint of fragment U-l (see Fig. 2D) was identical to that shown in Fig. 2 except that it lacked spot 7. Fingerprint analysis of this fragment after complete pancreatic RNase digestion indicated that it also included the oligonucleotide G-G-G----p. Therefore, fragment U-l extends from nucleotide 8 to nucleotide 5 (or 55). The T RNase fingerprint of fragment U-D (see Fig. 2E) contained all of the spots of the Ti RNase fingerprint of whole U- RN (see Fig. 2) except spot 7 at the ' end of U- RN. Therefore, fragment U-ID included nucleotides -5 (or 55). Other experiments (not shown) with pancreatic RNase have produced several additional fragments, four of which were obtained in yields of >20%. The fragment in highest yield (6q% molar yield) extended from nucleotides 9 to the ' terminus of U- RN. Two fragments were products of cleavages within oligonucleotide T-2 (see Fig. ), the first nucleotides of U- RN, and both extended in the ' direction to nucleotide 8. Their combined molar yield was 5%. The other fragment, obtained in 2% molar yield, was cleaved in oligonucleotide T- at nucleotide, 2, or and extended to the ' end of U- RN. Other minor fragments were obtained in yields of % or less but sufficient for fingerprint analysis. Two obtained by T RNase digestion were products of a single cleavage at nucleotide 70 and contained nucleotides -69 and Of the four obtained by pancreatic RNase digestion, two were products of the cleavages described above but in different combinations; one fragment contained nucleotides -5, the other was cleaved in oligonucleotide T- and contained nucleotides 2 (or or 5)-8. The other two suggested two different sites of pancreatic RNase attack, nucleotide 2 and within oligonucleotide T- (see Fig. ), and contained nucleotides -5 and 9-50 (or 5 or 52) (in oligonucleotide T-). The cleavage patterns obtained from digestion of deproteinized U- RN under varying T RNase concentrations, including that shown in Fig., are shown in Fig.. Fragments larger than 20 nucleotides accounted for %, 20%, %, 5%, and 60% of the U- RN recovered from sonicates at enzyme/substrate ratios of :5, :25, :5, :625, and :5, respectively. When isolated U-i RN was digested under the same conditions but in the absence of added sonicate, similar results were obtained. Fingerprint analysis of the major fragments showed that fragments - and fragment 6 were cleaved at Gp 69.6 Fragment contained nucleotides 70-5, fragment 2 contained

3 56 Biochemistry: Epstein et al Proc. Nad cad. Sci. US 7&(98) 8 I. I# 7 la = N* ,20 f st 9* *S5 _tv 8 * 7 I.a a 2 b S 7 s,.% 0 -c D a 9 aw I & 8 a. 9 *lt. 0 st la eq 2 9 _OF 20 _ 22 Vw 5 N E 9 8, 20 * ,9,8 st an V _S. aw t FIG. 2. (Legend appears on following page.)

4 Biochemistry: Epstein et al. U-Ib ~~~~f FIG.. utoradiograph of products of T RNase digestion of isolated U- RN, separated as described in Fig.. Lane, isolated U- RN; lanes 2-7, fragments produced after incubation of isolated U- RN in nuclear sonicates with T RNase at enzyme/substrate ratios of 0, :5, :625, :5, :25, and :5 (wt/wt), respectively. t :5, there was marked fragmentation of the U- RN by comparison with the result noted in Fig.. nucleotides -69, fragment contained nucleotides 70-, fragment contained nucleotides 70-7, and fragment 6 contained nucleotides Fragment 5 was cleaved at Gp-5 and contained nucleotides -5. DISUSSION The sequences of U- RN (9-) and the major fragments recovered after digestion-of U- RN in U- RNP particles with T RNase are shown in Fig.. The fragment in highest (77%) molar yield (fragment U-) contained nucleotides -7. The remainder of the structure (fragment U-B, containing nucleotides 8-5) was obtained in 7% molar yield. Fragment U-, containing nucleotides 8-5, was obtained in 5% molar yield. The longest, fragment (U-iD), nucleotides -5, was Proc. Natl. cad. Sci. US 78 (98) 565 obtained in 9% molar yield. These fragments account for 86-0% of the U- RN precipitated by the antibody used in these experiments. In contrast to the limited digests obtained when U- RNP particles were used as substrate, isolated U- RN was more completely digested at much lower T RNase concentrations. t an enzyme/substrate ratio of :5, less than 5% of the original (undigested) radioactivity in isolated U- RN was recovered as fragments having 20 or more nucleotides. In contrast, almost 90% of the original (undigested) radioactivity was recovered in four large fragments after digestion of U- RNPs under the same conditions. n equivalent degree of degradation with deproteinized U- RN was not obtained, even when the enzyme/substrate ratio was reduced to :5, indicating a difference in sensitivity of 500-fold or more. The high resistance to cleavage of U- RN in U- RNP particles suggests that this RN is protected, probably by proteins. The identity and number of U- RNP proteins is uncertain at present. Earlier reports () suggested that proteins are associated with U- and U-2 RNs; more recent papers (2, ) have reported that 7 proteins are precipitated with U- RN by anti- RNP antibodies. The same seven proteins were also precipitated with U-, U-2, U-, U-5, and U-6 RNs when anti-sm antibodies were used. The sequences of U-, U-2, U-, U-5, and U-6 are being determined, and several general regions of homology exist (). Such regions are potential protein-binding sites. Our preliminary data on U-2 RN suggests that this RN is also present as a T RNase-resistant form in snrnps. omparison of the cleavage sites in isolated U- RN and in U- RNP shows significant differences. The major site of Ti RNase cleavage of U- RNP particles was nucleotide 7, which accounted for almost 90% of the products. In contrast, only one of the six fragments obtained on digestion of isolated U- RN was cleaved at this site and that only at the higher enzyme concentrations. The principal site of cleavage of isolated U- RN was nucleotide 69. However, this position accounted for <5% of the U- RNP digestion products. This result suggests that the region of nucleotide 69 in U- RN is shielded by protein binding in U- RNP particles. This also appears to be the case for nucleotides 5 and. Despite the differences in T RNase sensitivity indicated in this study, as well as the different sites of cleavage, it is notable that four of the five positions cleaved in isolated U- RN are at or near the sites where pancreatic or Ti RNase cleaved U- RN contained in U- RNP particles. This finding suggests that U- RN maintains a similar structure in U- RNP. nalysis of the partial fragments recovered suggested that only five sites or regions of U- RNP particles are attacked by ribonucleases. In searching for reasons for this specificity, we constructed several secondary structures for U- RN by using the computer program of Korn et al () and computed their stability numbers according to the method of Tinoco et al (). The structure having the highest stability number (+28) is shown in Fig. 5. It is notable that all of the cleavage sites obtained with U- RNP particles and isolated U- RN occur either in the loops of this structure or in the non-hydrogenbonded region at the 5' end. This proposed structure is also compatible with products reported after partial nuclease S digestion of isolated U- RN (ll). FIG. 2 (on preceding page). utoradiographs of two-dimensional separations of the products obtained by complete T RNase digestion of complete U- RN and fragments U-, U-B, U-, and U-D as described in Materials and Methods. () omplete U- RN. (B) Fragment U-, which contains nucleotides -7. () Fragment U-B, which contains nucleotides 8-5. (D) Fragment U-i, which contains nucleotides 8-5 (or 55). (E) Fragment U-D, which contains nucleotides -5 (or 55). Numbers next to the spots refer to the numbering of complete T fragments shown in Fig.. B, Bromphenol blue; Y, Pyramine Y.

5 566 Biochemistry: Epstein. et al. Troc. Nad cad. Sci. US 78 (98) pr 2'2 7G r- P~~~~~~.--- * -- Ī., r-- M r M., I M pm Ur U U U G G G G G G G U U G U G G G U G G U U U U G G U-ic / U-D / ri u-u*r-i * I) mr,. I, I G G G G U U U U U G U G G U G U G U G U G G U U U U G k U uu*ni i-u r - I I* -u * r --uu-uu-u i- n n m rr-. II G G G U G U G U U U U G U G G U G U G G G G G U G G U U G G U U U &H U-B / \ U-lB \ - U-D U- Z \ S U- FIG.. Primary nucleotide sequence of Novikoff; hepatoma U- RN. Numbers above the sequence refer to the fragments shown in Fig. 2. Lines below the sequence refer to fragments U-, U-B, U-, and U-D. Both the structure of the 5' end of U- RN and its resistance to T RNase would seem surprising in view of the hypothesis that the 5'-end 2 nucleotides of U- RN are involved in hy- 5 G G G U GU G.5 G..?7 G * img G p N ', NG. G.- 'U U U-,50 pig P.G m UmUUUGG.G+7 U Goof a U. UGGUGUU.. GU G U 7 FIG. 5. possible secondary structure of U- RN. rrow at nu- -cleotide 7 indicates the major site of cleavage and arrows at nucleotides 5 and 55 indicate possible positions of the minor cleavage by Ti RNase in U- RNP. Numbers at each stem or stem and loop refer to the stability number of that region, as calculated by the method of Tinoco et al. (). The stability number for the entire structure was +28. drogen bondingwith hnrn splice junctions (2, ). It is possible that U-i RN may undergo conformational changes or that U- RNP may separate from one or more proteins during association with hnrn (or both). We wish to express our appreciation to Dr. Martin Lidsky for generous gifts of sera. We also thank Mrs. Rose K. Busch for the supply of tumor-bearing rats. This work was supported by Grant 89 from the National ancer Institute. P.E. was supported by Houston Pharmacological enter Grant GMO Raj, N. B., Ro-hoi, T. S. & Busch, H. (975) Biochemistry, Lerner, M. R. & Steitz, J.. (979) Proc. Natl. cad. Sci. US 76, Lerner, M. R., Boyle, J.., Mount, S. M., Wolin, S. L. & Steitz, J.. (980) Nature (London) 28, Rogers, J. & Wall, R. (980) Proc. Natl. cad. Sci. US 77, Mauritzen,. M., hoi, Y.. & Busch, H. (970) Methods ancer Res. 6, Muramatsu, M., Hayashi, Y., Onishi, T., Sakai, M., Takai, K. & Kashiyama, T. (97) Exp. ell Res. 88, Winter, G. & Brownlee, G. (978) Nucleic cids Res. 5, Brownlee, G. G., Sanger, F. & Barrel, B. G. (968)J. Mo!. Bio!., Reddy, R., Ro-hoi, T. S., Henning, D. & Busch, H. (97)J. Bio!. hem. 29, Roop, D. R., Kristo, P., Stumph, W. E., Tsai, M. J. & O'Malley, B. W. (98) ell, in press.. Branlant,., Krol,., Ebel, J., Lazar, E., Gallinaro, H., Jacob, M., Sri-Widada, J. & Jeanteur, P. (980) Nucleic cids Res. 8, -5.. Reddy, R. & Busch, H. (980) in The ell Nucleus, ed. Busch, H. (cademic, New York), Vol. 8, pp Korn, L. J., Queen,. L. & Wegman, N. M. (977) Proc. Nat!. cad. Sci. US.7, Tinoco, I., Uhlenbech, 0.. & Levine, M. D. (97) Nature (London) 20,