Premature termination of RNA polymerase II mediated transcription of a seed protein gene in Schizosaccharomyces pombe

Abstract

The poly(A) signal and downstream elements with transcriptional pausing activity play an important role in termination of RNA polymerase II transcription. We show that an intronic sequence derived from the plant seed protein gene (AmA1) specifically acts as a transcriptional terminator in the fission yeast, Schizosaccharomyces pombe. The 3'-end points of mRNA encoded by the AmA1 gene were mapped at different positions in S.pombe and in native cells of Amaranthus hypochondriacus. Deletion analyses of the AmA1 intronic sequence revealed that multiple elements essential for proper transcriptional termination in S.pombe include two site-determining elements (SDEs) and three downstream sequence elements. RT-PCR analyses detected transcripts up to the second SDE. This is the first report showing that the highly conserved mammalian poly(A) signal, AAUAAA, is also functional in S.pombe. The poly(A) site was determined as Y(A) both in native and heterologous systems but at different positions. Deletion of these cis-elements abolished 3'-end processing in S.pombe and a single point mutation in this motif reduced the activity by 70% while enhancing activity at downstream SDE. These results indicate that the bipartite sequence elements as signals for 3'-end processing in fission yeast act in tandem with other cis-acting elements. A comparison of these elements in the AmA1 intron that function as a transcriptional terminator in fission yeast with that of its native genes showed that both require an AT-rich distal and proximal upstream element. However, these sequences are not identical. Transcription run-on analysis indicates that elongating RNA polymerase II molecules accumulate over these pause signals, maximal at 611-949 nt. Furthermore, we demonstrate that the AmA1 intronic terminator sequence acts in a position-independent manner when placed within another gene.

Figure 1

(A) Schematic representation of the AmA1 genomic clone, pSB5.4, indicating two exons (E1 and E2) and one intron (I). (B) Northern blot of RNAs from S.pombe cells transformed with plasmids pREP1, pRAC and pRAG. The positions of RNA molecular weight markers are shown in the right track. (C) RT–PCR products of RNAs from cells transformed with pRAC and pRAG using F51 and R213 primers. (D) Immunoblot of total soluble protein from S.pombe cells transformed with pREP1 and pRAG. Aliquots of 200 µg protein were separated by 16.5% Tricine–SDS–PAGE and electroblotted onto a Hybond membrane (Amersham Pharmacia). The AmA1 protein was detected with a rabbit polyclonal anti-AmA1 antibody and anti-rabbit IgG antibody. The arrow indicates the truncated AmA1 protein.

Figure 2

(A) RT–PCR products of RNAs from S.pombe cells transformed with pRAG using F51 in combination with R261, R316, R370 and R680 primers. RT products were analyzed on ethidium bromide stained 1% agarose gel. (B and C) Poly(A) site mapping of the premature transcripts by 3′-RACE. The reactions were carried out with RNAs from pRAG-transformed S.pombe cells using F51/F190 and AUAP (Life Technologies) primers (B) and amaranth seeds using F190 and AUAP/R1044 primers (C). The RACE products were analyzed on ethidium bromide stained 1.2% agarose gel. (D) Schematic representation of the AmA1 gene showing positions of different primers and poly(A) sites as detected in S.pombe and amaranth.

Figure 3

(A) Diagram of the clones pRAC, pRAG and its deletion subclones along with the deleted region (dotted line) in the AmA1 intron. In pRAG and deleted subclones, the position +1 represents the translation start site of the AmA1 gene and 213 and 1747 as the 5′- and 3′-intron termini. (B) Northern blot showing 100% read-through transcript of 1.2 kb in pΔRAG, 50% each of 1.3 and 0.5 kb truncated transcripts in pRABB, with only 20% of 0.65 kb read-through and 80% 0.45 kb truncated transcripts in pRABN whereas there was no read-through transcript in pRANd, pRADr10.0 and pRADr10.9. (C) The 3′-RACE products of RNA isolated from S.pombe cells transformed with the constructs as indicated at the top of each lane. The reaction was carried out with F51 and AP (Life Technologies) primers and the products were analyzed on a 1.2% agarose gel. The differences in polyadenyaltion of transcripts are shown by arrows.

Figure 4

Analysis of putative SDE mutations of pSB5.4 and deletion subclones. (A) Northern blot of RNA from S.pombe cells harboring pRAG, deletion clones (pRABB, pRABN and pRANd) and mutants (pRAM275, pRABBM275, pRABNM275 and pRAND275) as detailed in Figure 3B. Positions of transcripts are shown by arrows. (B) An impression of the gel stained with ethidium bromide prior to transferring the RNA sample onto a nylon membrane. (C) Bar graph of the data obtained from the northern blot. The absolute values were corrected for background hybridization and plotted against respective clones each of which indicates the transcript size.

Figure 5

Detection of transcripts downstream of poly(A). (A) Schematic representation of positions of each of the antisense oligonucleotides used for RT–PCR analysis with respect to the 2.55 kb AmA1 genomic fragment. (B) Electrophoretic separation of transcripts detected across the downstream region of the poly(A) site. RT reactions were carried out with RNA isolated from S.pombe cells transformed with pRAG using end-labeled F51 and different antisense primers. As reference for semi-quantification of differential amplification, 0.1 ng of pRAG DNA was also amplified and the products were analyzed on a 1.5% agarose gel. (C) Bar graph of the data obtained from RT–PCRs. The amplified DNA fragments were excised, the radioactivity count (Cerenkov) was determined and plotted as the percentage of the relative counts obtained from plasmid DNA (as 100%).

Figure 6

TRO analysis of the 2.55 kb AmA1 genomic fragment in S.pombe. (A) Schematic representation of different overlapping fragments of 2.55 kb in pSB5.4 used as probes. Restriction enzymes used to prepare the fragments are shown along with their respective positions. (B) TRO blots containing 30 ng of each of the probes were blotted and hybridized with nascent transcripts isolated from pRAC- and pRAG-transformed S.pombe cells as described in Materials and Methods. (C–E) Graphical display of the TRO profile. The graphs are drawn to scale such that the width of each bar reflects the length of the probe and the height indicates the respective signal across the probe. (C and D) The representative TRO signals from pRAC- and pRAG-transformed S.pombe cells, respectively, and (E) the signals across the probes in the case of pRAG-transformed cells which are corrected and expressed relative to probe 1.

Figure 7

Functional analysis of the AmA1 intronic sequence recognized as a transcription terminator in S.pombe. (A) Sequence of the RsaI fragment (213–589, 376 nt) which was used to carry out functional analysis of the terminator. (B) Truncated transcripts detected in the northern blot. The blot was prepared with 25 µg of total RNA from each extract, separated on a 1.2% denaturing formaldehyde gel, transferred onto a nylon membrane and hybridized with 1.18 kb AmA1 cDNA.

Figure 8

Structural analysis of the AmA1 transcript in S.pombe. (A) Secondary structure prediction of the transcript flanking poly(A) site. The AmA1 sequence spanning 1–464 nt was analyzed using the RNA-fold program. Only structures of stem–loops predicted in the region are shown. (B) RT–PCR analysis with different reverse primers in the stem–loop region. Electrophoretic separation of RT–PCR products generated after 20, 25 and 30 PCR cycles from RNA isolated from S.pombe cells transformed with pRAG. The reactions were carried with 1 µg of RNA using end-labeled F51 forward primer and one of the four reverse primers R261, R316, R370 and R680. The amplified products of the 25-cycle category were excised and radioactivity (Cerenkov count) was determined and represented graphically in the bottom panel. (C) In vitro transcription termination competition assay using T3 RNA polymerase (Life Technologies) from NdeI-linearized pSBNd4.3 in the presence of different primers (the primer quantities are indicated above the tracks) using [α³²-P]UTP. The reactions were analyzed on denaturing polyacrylamide gel. The relative positions of the antisense primers used in the competition assay with respect to the RNA secondary structure are shown.

Figure 9

Modular architecture for the mechanism of 3′-end formation of the AmA1 gene in S.pombe. The diagram depicts a general architecture of poly(A) signals and terminator in S.pombe (A) and positions of various cis-elements responsible for the premature termination of AmA1 transcript in S.pombe (B). The clevage/polyadenylation site is indicated by an arrowhead. The drawing is not to scale.