J. Mol. Biol. (1990) 211, 7-9

An Intron-containing Schizosaccharomyces pombe U6 RNA Gene Can Be Transcribed by Human RNA Polymerase III Ann M. Kleinschmidt’, Thoru Pederson’?, Tokio Tani’ and Yasumi Ohshima2 ‘Cell Biology Group, Worcester Foundation for Experimental Biology, Shrewsbury, MA 01545, U.S.A. and ‘Department

of Biology, Kyushu Fukuoka 812, Japan

University

(Received 2 August 1989) A Nchizosaccharomycespombe U6 small nuclear RNA gene containing an intron has been described. We find that the S. pombe U6 gene is transcribed in a human (HeLa) cell 8100 extract with an a-amanitin sensitivity characteristic of RNA polymerase III. The S. pwmbe l’6 gene is also t’ranscribed after transfection into human cells. The transcription of vertebrate U6 RNA genes bv” RNA polymerase III does not require intragenic control elements. The intron of the S. pombe U6 gene disrupts a “box A”-like intragenic sequence t#hat is typically an RNA polymerase III transcription control element. This, together with the transcription of the S. pombe C6 gene by human RNA polymerase III, suggeststhat it is recognized by human U6 gene-specific transcription machinery.

C6 is one of five nucleoplasmic small nuclear RSAs that function as cofactors in mRNA splicing (Maniatis & Reed, 198’7; Steitz et al., 1988). Unlike Ul , U2, 1:4 and US RNAs, which are transcribed by RNA polymerase II, U6 RNA is transcribed by RNA polymerase III in human (Kunkel et al., 1986)> mouse (R’eddy et al., 1987) and Xenopus (Krol et al., 1987). Unlike virtually all other RNA polymerase III-transcribed genes, these vertebrate U6 genesdo not cont’ain typical intragenic control elements (Carbon et ~1.. 1987; Das et al., 1988; Kunkel & Pederson. 1989). Recently a fission yeast (Schizosaccharomyces pombe) U6 RNA gene has been described that contains an intron (Tani & Ohshima, 1989). Because the S. pombe U6 gene’s intron disrupts an internal sequence similar to the “box A” element, of typical pol III intragenic promoters (Tani & Ohshima, 1989), and becausethe 5’-flanking region of the S. pombeU6 gene displays only limited similarity to the corresponding region of vertebrate V6 genes (Tani & Ohshima, 1989), we have investigated vvhether the S. pombe U6 gene can be transcribed by human RNA polymerase III. Figure 1 shows the result of incubating S. pombe or human c’6 DNA templates in a HeLa cell SlOO extract that is efficient for RNA polymerase IIIt

Author

to whom

all

correspondence

should

mediatedtranscription (Weiletal., 1979).TheS,pombe U6 DNA plasmid pU6-2, which contains - 1500 nucleotides of 5’-flanking sequence (Tani & Ohshima, 1989; T. Tani & Y. Ohshima, unpublished results), or a human U6 plasmid pGEM/UG (Kunkel et al., 1986; Kunkel & Pederson, 1988), was incubated in HeLa SlOO extract in the presence of [a-32P]GTP as described (Kunkel et al.. 1986). The RNA was recovered from the reaction and U6 RNAs were isolated by hybridization selection (Kunkel et al., 1986) and analyzed by electrophoresis in 10% (w/v) polyacrylamide gels containing %3 M-urea. It can be seen (Fig. 1, lane 4) that the S. pombe U6 DNA yielded a product, that migrated with a mobility corresponding to - 150 nucleotides, which is the length expected for a polymerase III transcription product from this gene (Tani & Ohshima, 1989). The level of transcription from both the S. pombe and human U6 templates was not reduced by 1 pg a-amanitinlml, a concentration that selectively inhibits polymerase IT (Fig. 1, lanes 2 and 5), but was completely eliminated by 200 pg a-amanitinlml, which inhibits polymerase III (Fig. 1, lanes 3 and 6). These results establish that the S. pombe U6 RNA gene can be transcribed by human RNA polymerase III. We next asked whether the S. pombe U6 RNA gene could be transcribed and spliced within human

be addressed.

7 0022~2836/90/010~M~7-03

$03.00/O

0

1990

Academic

Press

Limited

A. M.

8

Human

(I -Amanitin

(pg/ml)

1 0

Kleinschmidt

et al.

S, porn bc

1

200

I 0

I

DNA 2001

primer

M

sp sp

HU sp

sp HU HU HU M

169

s.pon 124

Maxi

Enda lganoul

US-

US-

65

I

2

3

4

5

6

M

Figure 1. The S. pombe U6 gene is transcribed

bq human RNA polymerase III. Human (Kunkel et al.. 1986) or S. pombe (Tani & Ohshima, 1989) U6 DNA (50 pg/ml) was incubated in SlOO extracts in the presence of [N-~‘P]GTP (500/.&i/ml) and 0. 1 or 2OOpg cc-amanitini ml. Total RNA was recovered and U6 RPiA was isolated by hybridization selection. Lane M, HindI restriction fragments of pGEM-2’run as markers.

cells. For transfections the Hind111 U6-containing insert of pU6-2 (Tani & Ohshima, 1989) was recloned into pGEM-2. Transcription from this plasmid was compared with that from a human U6 maxigene (Kunkel & Pederson, 1988). Transfection of the S. pombe or human U6 plasmid DNA into human 293 cells, RNA extraction and primer extension analysis were all carried out as described (Kunkel & Pederson, 1988). For analysis of S. pombe U6 expression an oligodeoxynucleotide complementary to nucleotides 130 to 144 of the unspliced RNA was used (GGGTTTTCTCTCAAT), which contains no sequence complementarity to human U6 RNA (Fig. 2, lanes 1 and 2). In cells transfected with the S. pombe U6 DNA, an RNA was expressed that generated a N 144 nucleotide primer extension product with the S. pombe U6-specific oligomer (lane 1); this corresponds to the predicted size of the unspliced S. pombe U6 transcript. No primer extension product was observed having the length (94 nucleotides) expected for spliced S. pombe U6 RNA. We cannot rule out the possibility that spliced S. pombe U6 RNA is unstable. However, this seems unlikely since the unspliced S. pombe U6 is

64

I

2

3

4

M

Figure 2. The S. pombr

I!6 gene is transcribed iti human 293 cells. Dishes (100 mm) of subconfluent 293 cells were transfected with 20 pg of ,Y. pombe (SJ)) or human maxigene (Hu) U6 DNA (Kunkel & I’rdrrson. 1988), and total cell RNA was isolat,ed. Primer extension reactions were carried out with oligonucxleotides vomplementary t,o nucleotides 130 to 144 of unspliced S. pOrnbu (Sp) or 88 to 102 of human (Hu) I!6 RNA: 2.5-fold more RNA was used for the primer extension analyses of t)ranscription products from cells transfected with 9. pombe ITA DNA. Lane M, HaeTII restriction fragments of pGEM-2 used as markers. S. pombe: primer exiension product of unspliced transcript of the transfected AS.pombe U6 gene. Maxi U6: primer extension product of transcripts from human U6 maxigene. Endogenous U6: primer extension product of cellular U6.

obviously stable in human cells and since the S. pombe U6 RNA sequence is 779;) similar to that of mammalian UA. (In the analysis shown in Fig. 2 it would have been possible to visualize an R,NA species that is 2O-fold less abundant t,han the unspliced S. pombe U6 RNA.) The N 144 nucleotidr primer extension product was not observed when RNA from cells transfected with the human Ii6 maxigene was used (Fig. 2. lane 2). There was also no primer extension product of this length seen a human UB-complementar)oligomer when (Kunkel & Pederson, 1988) was used to analyze t,he RNA from the cells transfected with S. pombr U6 DNA (lane 3). The fact that the steady-state Ievel of spliced lJ6 RNA is lower in N. pombr mutants defectivr in mRNA splicing (Potashkin et aZ.. 1989) argues that

Communications the removal of this mRNA-like intron from U6 gene transcripts is mediated by the usual S. pombe premRNA splicing machinery. It is possible that the lack of splicing of S. pombe U6 RNA in human cells reflects a difference in the splicing machinery of S. pombe and mammalian cells, or that in human cells pol III transcripts are placed in an intranuclear with compartment or pathway incompatible splicing (Sisoida et al., 1987). Alternatively, the lack of splicing may reflect the very short distance (N 13 nucleotides) between the probable S. pombe U6 branchpoint and the 3’ splice site as compared to most mammalian introns. A previous investigation of splicing of an X. cerevisiae mRNA gene in a human nuclear extract revealed the preferential use of alternative lariat branchpoints lying 3’ of the canonical UACUAAC box (Ruskin et al., 1986). Interestingly, the S. pombe U6 gene intron contains one adenosine 3’-ward of the UACUAAC box, but it lies only two nucleotides 5’-ward from the end of the intron (Tani & Ohshima, 1989). Although it has been shown that the UACUAAC sequence is a preferred branch site in mammalian mRNA introns (Zhuang et al., 1989) and that at least one (67 nucleotide long) mammalian mRNA intron is spliced in S. pombe (Kaufer et al., 1985), it is possible that the extreme proximity of this adenosine to the 3’ splice site would preclude its use as a lariat branchpoint in the unspliced S. pombe U6 RNA. The U6 Rh’A genes of vertebrates are remarkable, not simply because they are transcribed by R?U’A polymerase III but because, even more surprisingly, their transcription does not depend on intragenic sequences (Das et al., 1988; Kunkel & Pederson, 1989). This capacity for transcription, independent of intragenic control elements, may reflect the existence of a distinctive, U6 genespecific polymerase III control element in the %-flanking DNA, the presence of a U6 gene-specific polymerase TIT transcription factor(s) (Reddy, 1988), or both. Our results show that human RNA polymerase III can transcribe an S. pombe U6 RNA gene in which the putative intragenic control element is disrupted by an intron. This suggests that human RNA polymerase III and human U6 transcription factors may be operating on the 5’-flanking DNA of the S. pombe U6 gene. The facts that the vertebrate U6 RNA genes studied have polymerase II-like transcription control elements in their 5”-flanking DNA (Bark et al., 1987; Carbon et al., 1987; Krol et al., 1987; Kunkel & Pederson, Edited by M.

9

1988, 1989; Mattaj et al., 1988; Das et al., 1988), and that the S. pombe U6 gene has an upstream TATA box in a typical polymerase II location (Tani & Ohshima, 1989), raises the beguiling question of why U6 genes are transcribed by polymerase III altogether. Supported by NIH grant GM21595-14 to T.P. We thank Gary Kunkel for critically reading the manuscript and Susan Olszta for secretarial assistance.

References Bark,

C., Weller, P., Zabielski, J., Janson, L. & Pettersson, U. (1987). Nature (London), 328, 356-359. Carbon, P., Murgo, S., Ebel, J.-P., Krol, A., Tebb, G. & Mattaj, I. W. (1987). Cell, 51, 71-79. Das, G., Henning, D., Wright, D. & Reddy, R. (1988). EMBO J. 7, 503-512. Kaufer, N. F., Simanis, V. & Nurse. P. (1985). Nature (London), 318, 78-80. Krol, A., Carbon, P., Ebel, J.-P. & Appel, B. (1987). Nucl. Acids Res. 15, 2463-2478. Kunkel, G. R. & Pederson, T. (1988). Genes Develop. 2, 196-204. Kunkel, G. R. t Pederson, T. (1989). Nucl. ,4cids Res. 17, 7371-7379. Kunkel, G. R., Maser, R. L., Calvet, J. P. & Pederson, T. (1986). Proc. Nat. Acad. Sci., U.S.A. 83, 8575-8579. Maniatis, T. & Reed, R. (1987). Nature (London), 325, 673-678. Mattaj, I. W., Dathan, N. A., Parry, H. D., Carbon, P. & Krol, A. (1988). Cell, 55, 435442. Potashkin, J., Li; R. & Frendewey, D. (1989). EMBO J. 8, 551-559. Reddy, R. (1988). J. Biol. Chem. 263, 15980-15984. Reddy, R., Henning, D., Das, G., Harless, M. & Wright, D. (1987). J. Biol. Chem. 262, 75-81. Ruskin, B., Pikielny, C. W., Rosbash, M. & Green, M. R. (1986). Proc. Nat. Acad. Sci., U.S.A. 83, 2022-2026. Sisoida, S. S., Sollner-Webb, B. 8: Cleveland, D. W. (1987). Mol. Cell. Biol. 7, 3602-3612. Steitz, J. A., Black, D. L., Gerke, V., Parker, K. A., Kramer, A.! Frendewey, D. & Keller, W. (1988). In Small Nuclear Ribonucleoprotein Particles (Birnstiel, M. L., ed.), pp. 115-154, Springer-Verlag, Berlin. Tani, T. & Ohshima, Y. (1989). Nature (London), 337, 87-90. Weil, P. A., Segall, J., Harris, B., Ng, S. $ Roeder, R. G. (1979). J. Biol. Chem. 254, 6163-6173. Zhuang, Y., Goldstein, A. M. & Weiner, A. M. (1989). Proc. Nat. Acad. Sci., U.S.A. 86, 2752-2756. Yanagida