J. Mol. Biol. (1992) 226, 335-347

Transcriptional Commitment of Mitochondrial RNA Polymerase from Saccharomyces cerevisiae Tapan K. Biswas Wayne State University School of Medicine Detroit, MI 48201, U.S.A. (Received 3 May

1991; accepted 2 March

1992)

The transcriptional commitment of mitochondrial RNA (mtRNA) polymerase and the conditions required for the formation of a stable ternary complex have been determined by in vitro transcription study. Four different transcription complexes were made in vitro by incubating purified mtRNA polymerase, cloned synthetic mitochondrial promoters and selective ribonucleotides. The responses of these complexes to heparin, an inhibitor of unbound mtRNA polymerase, have been examined to determine their involvement in transcription. This study leads to the following observations. (1) Under normal reaction 40 ni%-heparin completely inhibited mitochondrial transcription. conditions, (2) 9 preinitiation mitochondrial DNA-RNA polymerase complex (complex 0) showed partial resistance to heparin (~25% resistant to 40 nM-heparin) when heparin and ribonucleoside triphosphates (rNTPs) were added together to the preformed complex. This complex was rapidly inactivated when preincubated with heparin before the addition of rNTPs. (3) The early initiation (complexes 2 and 4) containing DNA template, RNA polymerase and a short RNA product showed more resistance (approx. 40 to 50%) to 40 nw-heparin but destabilized upon further incubation with heparin before addition of the rest of the rNTPs. (4) After generation of ten or more phosphodiester bonds (complex ll), the early transcription complex is converted into a stable initiation complex, leading to the polymerase consignment to elongation. On the basis of stability and heparin sensitivity, three initial steps of mitochondrial transcription have been defined: polymerase-promoter interaction, initiation, and the transition from initiation to elongation. The formation of preinitiation complex is the rate-limiting step (t 1,2 approx. 50 s), whereas the initiation and elongation reactions are very fast processes (t,,, > 5 s) in mitochondrial transcript’ion.

Keywords: mitochondrial

gene; Saccharomyces cerevisiae;

1. Introduction

nwleoside

regulation

core polymerase and a 43 kDa or 70 kDa specificity factor(s), recognizes an octanucleotide promoter and then initiates transcription on the rRNA, tRNA and mRNA genes, and at the putative origins of replication (Winkley et al., 1985; Shinkel et al., 1987; Ticho & Getz, 1988). The previous studies demonstrated that the yeast mtRNA polymerase closely resembles the bacteriophage T3/T7 RNA polymerases (Master et al., 1987), but differs from the eukaryotic and bacterial RNA polymerases (Levens et al., 198 1) . An eight nucleotide conserved sequence (TATAAGTA( + 1)) is present just upstream from each initiation site (Christianson & Rabinowitz, 1983) and functions as a promoter in yeast mitochondria (Biswas et al., 1985). Mutation and in vitro transcription studies determined that ( - 7)TAtAaGtN(+l) is the minimum DNA sequence requirement for mitochondrial promoter function in S. cerevisiae (Biswas & Getz 1986a,b; Schinkel et al.,

The 76,000 base-pair mitochondrial genome of the yeast Saccharomyces cerevisiae carries genes for two rRNAs (21 S and 15 S), 25 tRNAs and at least eight proteins involved in mitochondrial respiratory function (Grivell, 1989; Tzagoloff & Myers, 1986). The rest of the mitochondrial proteins including translation, transcription and replicating enzymes are nuclearly encoded and imported into mitochondria. The mitochondrial genes are transcribed singly or as a part of a multi-cistronic unit from different parts of, the mitochondrial DNA (mtDNAt). A single mtRNA polymerase, consisting of a 150 kDa t Abbreviations used: mtDNA, mitochondrial deoxyribonucleic acid; mtRPu’A, mitochondrial ribonucleio acid; nt, nucleotide(s); bp, base-pair(s);

transcriptional

NTP,

triphosphate.

335 tU,22-283s/92/14033513

to3.oojo

Q 1992 Academic Press Limited

T. K. Biswas

336

1986; Biswas et al., 1987). The lower-case letters indicate positions at which nucleotide variants are permissible, and N designates any nucleotide that is used as an initiating nucleotide. Although the octanucleotide sequence is sufficient for specific initiation of transcription, the rate of mitochondrial transcription is regulated by the downstream nucleotides (Biswas & Getz, 1986a, 1988; Biswas, 1991). The low level of transcription from a weak mitochondrial promoter is due to the slow rate of formation of the first phosphodiester bond between a purine and a pyrimidine (Biswas, 1990). Since promoter recognition and initiation of transcription is a multi-step process, dissection of these steps would be useful to understand both the initiation mechanism and the transition from initiation to elongation step. The events occurring during initiation of transcription in yeast mitochondria were analyzed by generating and characterizing different transcription complexes. It was of my particular interest to determine whether the early initiation complexes (complex 2 and complex 4) show different properties from the preinitiation complex (complex 0) or from the more elongated transcription complex (complex 11). In this paper, I demonstrate that these transcription complexes differ from each other in elongation eficiency, stability and heparin sensitivity. The mtRNA polymerase is fully committed to transcription only after formation of an RNA of size 11 nt or larger. In this regard, the to the mtRNA polymerase shows similarity bacterial and bacteriophage T3/T7 RNA polymerases, which move to the elongation step only after the formation of 10 to 12 nt RNA (Munson & Reznikoff, 1981; von Hippel et al., 1984; Morris et al., 1987), but differs from the human RNA polymerase II, which forms a stable ternary complex after making the first dinucleotide RNA (Luse et al.: 1987).

2. Materials

and Methods

(a) Chemicals Nucleoside triphosphates were purchased from P-L Biochemicals. Heparin (average molecular mass approx. 10,000) was from CalBiochem Corp. Restriction endonucleases were purchased from Bethesda Research Laboratories. [w~H]UTP (54 Ci/mmol) was from ICN Biomedicals, Inc., and [cr-32P]UTP (3000 Ci/mmol) was from Amersham Corp. or from New England Nuclear. (b) Plasmids and strains Plasmids containing mitochondrial promoters were as described (Biswas & Getz, 1986u,b; Biswas et al., 1987). Eseherichia coli strain RRlAM15 (Ruther et al., 1981) was used for the propagation of plasmid DNA. S. cerekiae strain D273-10B was used for the purification of mtRNA polymerase. (c) PuriJication

of mtRNA

polymeruse

The mtRNA polymerase from S. eerevisiae was purified by DEAE-cellulose, phosphocellulose, DEAE-Sephadex

A-50 and carboxymethylcellulose column chromatography as described (Ticho & Getz, 1988). The final preparation of enzyme was purified over 2500-fold and contained 80 pg protein/ml. The enzyme fractions containing specific polymerase activity were used in the in vitro transcription reaction. (d) Generation of transcription and in vitro transcription

complexes

A plasmid carrying a synthetic mitochondrial promoter sequence (5’ ATATAAGTaatagaattgac 3’) was used in this study (Biswas & Getz, 1986a). The upper-case letters designate the promoter sequence and the lower-case letters are the partial sequence of the transcribed region. Plasmid containing this mitochondrial promoter was linearized by PvuII digestion, and was used as a template in the in vitro kanscription reaction, which produced 117 RNA. Different transcription complexes were generated in vitro by incubation of 1 ~1 of mtRNA polymerase. 500 fmol of DNA template and with or without selective ribonucleotide(s) in 25~1 of transcription buffer (10 ma-Tris.HCl (pH 7.9), 10 miw-MgCl,, 20 rnM-KC]. 5% (v/v) glycerol and 0.2e/b (w/v) rabbit serum albumin) at 30°C for 10 min (see Fig. 1). These complexes were designated by the size of the nascent RNA formed in each complex (e.g. no RNA in complex 0. dinueleotide in complex 2. tetranucleotide in complex 4, and 11 nt RNA in complex 1I ). The preinitiation complex (complex 0) was made by incubating mtRNA polymerase and mtDNA template; the early initiation complex (complex 2) was made by incubation of mtRNA polymerase. mtDNA template and 100 PM-ATP; the initiation complex 4 was made by incubating mtRNA polymerase, mtDNA template, 100 PM-ATP and 5 ~M-UTP; transcription complex 11 was made by incubation of mtRNA polymerase. mtDNA template. 100 PM-ATP, 5 PM-UTP and 5 PM-GTP. To continue RNA synthesis on these complexes. 5 PM each of the remaining ribonucleotides and 5 to 10 &i of [c(-32P]UTP were added and incubated for an additional 7 min at, 30°C. The reaction was t,erminated by the addition of 25 ~1 of stop solution (0.3% (w/v) SDS, 200 pg tRNA/ml) followed by extraction with phenol. RNA molecules were separated from the unincorporated [32P]UTP by precipitation with 50 ~1 of -5 M-ammonium acetate and 250 ~1 of ethanol. Transcripts were resolved on 8 M-urea/s?& (w/v) polyacrylamide gel electrophoresis and then visualized by autoradiography. The entire gel has been displayed in some Figures to show the quality of the transcripts. In other Figures, gels are similar in yuality, so only the insets are shown. The RNU’A products were quantified by densitometric scanning of RNA bands on the autoradiograms or by counting the incorporated [32P]UMP in the transcripts. The amount of specific transcript synthesized under normal reaction conditions is equivalent t’o approx. 0.01 transcript/ molecule of ternplated DNA used. The results given are an average of 2 or 3 different experiments.

3. Results To characterize the commitment of mtRNA polymerase, different transcription complexes that are artificially arrested at a specific point of transcription were generated in vitro using specific reaction conditions. Heparin and a complete set of rNTPs were added to the preformed complexes to inactivate free polymerase and to permit a single round of

S. cerevisiae

lhnwi~tion

comDlex

Complex 0

Mitochondrial

‘hInscriDt seauence

Reaction conditions DNA

template + mtRNA pol.

Complex 4

Complex

11

+20

+lO

+l Complex 2

337

RNA Polymerase

DNA template + mtRNA pol. + ATP

AAUAQAAUUGACGGAUCCGG

DNA template + m&NA Pol. + ATP, UTP

AAUAQAAUUG

DNA template + mtRNA pol. + Al?, UTP, GTP

AAUAGAAUUGAcGGAUaGG

4 AWAUCCGG

4

4 Figure 1. Strategy for generation of different transcription complexes. The DKA template and the mtRKA polymerase were incubated with or without specified rPu’TP(s) at 30°C for 10 min to generate transcription complexes: 125 PM of the initiating nucleotide and 5 pM of other nucleotides were used. The partial RNA sequence of the transcribed region are shown at the right side of the Figure. The adenine nucleotide at position + 1 designates the initiating nucleotide. The underlined nucleotides indicate the positions at which nucleotide limitation could stop transcription. The nucleotide in bold face above the arrow refers to the expected 3’ end of the transcript in the complex.

transcription from the heparin-resistant complexes. Since each heparin resistant complex produced one RNA molecule, the relative number of transcription competent complexes was determined by measuring the amount of RNA products made by these complexes. Using this procedure, the involvement of mtRNA polymerase in transcription has been further characterized. (a) Generation

of transcription

complexes and

potential readthrough problem Different’ transcription complexes were made by preincubating mtRNA polymerase and template DNA with or without one, two or three of the four rNTPs. In principle, these reaction conditions would lead to elongation stopping at the nucleotide just prior to the first site at which the missing nucleotide would be used (Fig. 1). Since RNA chain elongation requires a low concentration of nucleoside triphosphate. a low level of contamination of the missing nucleotide will change the size of the transcript in the ternary complex and will affect the homogeneity of the complex. This possibility was examined in t’wo ways: first, the purity of a single nucleotide preparation was tested by CTP-less reaction (Fig. 2(a)): secondly, transcripts generated in the preformed transcription complexes were analyzed by polyacrylamide/urea gel electrophoresis and autoradiography (Fig. 2(b)). For CTP-less reaction, an A + T-rich mitochondrial tRNAfM” gene was used as a template in which cytosine base is located at positions + 31, + 52, + 63 ... . In principle, only a 30 nucleotide transcript is expected from this template in the absencewof CTP, since RNA synthesis will stop at the first cytosine nucleotide. The CTP-less reaction was carried out under standard reaction conditions except that CTP was not added.

The RNA products were separated by gel electrophoresis and identified by autoradiography. A 30 nucleotide major transcript. (approx. 90%) and a 51 nucleotide readthrough transcript (approx. 10 %) were detected in this reaction (Fig. 2(a)). The nascent transcripts generated by the preformed were also analyzed by 8 M-urea/ complexes polyacrylamide (20 y. (w/v) acrylamide, 3 y. (w/v) bisacrylamide) gel electrophoresis (Fig. 2(b)). The 4 nt and 11 nt transcripts were detected from complex 4 and complex 11, respectively. In addition to 11 nt transcript, a minor 17 nt RNA was seen in complex 11. To minimize the nucleotide contaminanucleotide and 5 PM of tion, 100 PM of the initiating other nucleotides were used for the generation of transcription complexes. (b) Promoter binding studied by gel retardation To study mtRNA polymerase-promoter interact*ions more directly, a DNA mobility shift experiment was carried out (Fig. 3). The DNA mobility shift experiment was selected due to the difficulty of obtaining convincing footprinting with mtRNA polymerase, which has a low binding affinity for its promoter. A 340 bp radio-labeled DNA fragment containing a mitochondrial promoter was used. The mtRNA polymerase and a labeled DNA fragment were incubated with or without selective rNTPs at 30°C for ten minutes to generate transcription complexes. Plasmid pUR 250 DNA (Fig. 3, lanes 5 t,o 7) or heparin (lanes 8 to 10) was then added before loading onto the gel to eliminate non-specific binding of polymerase to the template DNA. Polymerase binding to the template DNA should decrease the mobility of DNA. A slowly migrating DNA band was found (lanes 5 to 10) when plasmid DNA or heparin was added as a non-specific DNA

338

T. K. B&was

Transcription comDlex

-

0 4 110

4 11 0 4 11

M -CTP

- 11 nt 51 nt

1

30 nt

(a)

- 4nt

(b)

Figure 2. (a) In vitro transcription of a mitochondrial gene (tRNA Ne’) in the absence of CTP (CTP-less reaction). Reaction conditions were same as described in Materials and Methods, except that CTP was not added. RNA was labeled with [a3’P]UTP. Transcripts were analyzed by 8 M-urea/8O/b polyacrylamide gel electroend-labeled phoresis and autoradiography. An HinfI-digested pUR 250 plasmid DNA was used as a marker (M). (b) Analysis of transcripts generated by the transcription complexes. Transcription preformed complexes were generated as described above in the presence of 70 PCi of [32P]UTP (3000 Ci/mmol) in each reaction t,o label the RPU’A. The samples were extracted with phenol and RlVA was precipitated with an equal volume of 5 M-ammonium acetate and 5 volumes of ethanol. The precipitated RXA was analyzed by electrophoresis on an 8 M-urea/polyacrylamide (20 y0 acrylamide, 3 y0 bisacrylamide) gel. Lane 1: an RPjA ladder marker; lane 2, transcript from complex 4; and lane 3, transcript from complex 11.

competitor. In contrast, no such band shift was seen in the absence of competitor DNA (lanes 2 to 4). Instead, a band was observed near the top of the gel, which may be due to non-specific binding of the polymerase to the template DNA. Since non-specific binding but not specific binding of polymerase to the promoter can be competed by the plasmid DNA or heparin, the slow migration of DNA could be the result of specific interaction between the polymerase and the promoter.

2 3 4 5 6 I

8 9 10

Bottom (+ve) Figure 3. Gel retardation study of transcription complexes. Transcription complexes were made on an end-labeled 340 bp mtDNA template and then no competitor DNA (lanes 2 to 4). 200 nlvr-plasmid pVR 250 DRIA (lanes 5 to 7) or 40 nh%-heparin (lanes 8 to 10) was added before loading onto a native 8% polyacrylamide gel. Electrophoresis was carried out at 4°C for 3 to 4 h at 18 V/cm and the DNA visualized by autoradiography. The number above the lane indicates the type of transcription complex loaded onto that lane (e.g. 0, complex 0; 4. complex 4; 11, complex 11). Lane 1 shows the mobility of the DNA without incubation with the polymerase. The shifted band in lanes 5 to 10 is indicated by an arrow-headed line at the right side of the Figure.

(c) Time-course of transcription The time-course for transcription of different transcription complexes has been determined (Fig. 4). Transcription complexes were generated as described above and RNA synthesis was continued by the addition of the rest of the rNTPs for different periods of t’ime. The RNA products were identified by gel electrophoresis and autoradiography, and then quantified by densitometric scanning of the autoradiogram. Transcription from complexes 0 and 2 continued for seven to ten minutes, whereas transcription from the other complexes (4 and 11) was completed within one to three minutes. Since the formation of preinitiation complex is the ratelimiting step, and initiation of transcription is a fast process (see below), the longer transcription time of complexes 0 and 2 might be due to their unstable polymerase-promoter interactions. Probably a portion of these complexes started transcription immediately after the addition of rNTPs, whereas others took longer due to polymerase dissociation from the DNA template. In contrast, complexes 4 and 11 were more stable and efficient transcription complexes, and completed transcription in a shorter time.

S. cerevisiae

lOInin -

-t-

t 1. TemplateDNA 2. mtRNA polymemse 3. +/- SelectiverNTP(s)

t rNTPs

Mitochondrial

RNA

339

Polymeruse Reactmntime (min ) 0.5 1 3 5 7 10

Stop

2

4

6

Kewtion (a)

8

10

12

time (min b (b)

Figure 4. Time-course of transcription on different transcription complexes. Transcription complexes were generated and then the complete set of rPu’TPs was added to continue RNA synthesis. Reactions were terminated at different time points as indicated. and the transcripts were analyzed by urea/polyacrylamide gel electrophoresis. In a typical transcription reaction, approx. 5 fmol of transcript (corresponding to 6000 disints/min of [32P]UMP incorporated into the transcript) was made and was considered as 100% transcriptional activity. (a) The entire gel is exhibited to show the quality of the transcripts. (b) The insets from the same gel are shown; the transcripts were quantified the reaction time: (0) complex 0; (0) complex 2: (0) complex 4; (A) complex 11.

(d) Inhibition

qf mtRNA

polymerase by heparin

activity

The yeast mtRNA polymerase consists of at least two protein factors; a core polymerase and a specificity factor (Winkley et al., 1985; Schinkel et al., 1987; Ticho & Getz, 1988). The specificity factor is required to enable the core polymerase to initiate transcription at the promoter site; otherwise the core polymerase alone transcribes non-specific RNA from all over the DNA (Schinkel et al., 1987; Ticho & Getz, 1988). Since heparin inactivates free polymerase more rapidly than the DNA-bound enzyme, heparin was used to study the DNA binding ability of the core and holo mtRNA polymerases (Fig. 5). First, an RNA polymerase preparation containing both the core- and holoenzymes was incubated with template DNA at 30°C for ten minutes, and then rNTPs and heparin were added together (Fig. 5(a)). The core and holo polymerase showed different sensitivity to heparin inhibition. The core polymerase was completely inhibited by 8 nM-heparin, whereas the holoenzyme was resistant at least up to 40 nw-heparin. The polymerase resistance to 40 nMheparin was further examined using an enzyme preparation having mainly selective activity (Fig. 5(b)). The enzyme preparation mainly produced a promoter-specific RNA under normal reaction conditions lane 1). When (Fig. 5(b), heparin (40 nM) was included in the reaction, the polymerase activity was completely inactivated (Fig.‘5(b), lane 2). However, when polymerase was incubated with DNA template before the addition of rNTPs and heparin, approximately 30% of its normal activity (heparin-minus reaction) became resistent to heparin (lane 3). Since 40 nw-heparin inactivated only the polymerase that was not bound

to the promoter, used for further interactions.

and plotted

against

this concentration of heparin was study of the polymerase--promoter

(e) The effect of heparin concentrations on the activities of transcription complexes Various concentrations of heparin were used to determine the differences in sensitivity of these complexes to heparin inhibition. The protocol followed is summarized in Figure 6. In a control experiment DNA, enzyme, rNTPs and heparin were added together. In other cases. transcription complexes were generated first and then the rest of the rNTPs and various concentrations of heparin were added. The inhibition of transcription by heparin varied depending upon a concentration of heparin as well as the nature of the complexes. Heparin at 4 nM or less did not inhibit transcription, even when polymerase was incubated with heparin before the addition of DNA and rNTPs (data not shown). Heparin at higher concentrations (20 to 40 nM) inhibited transcription partially or completely. Transcription was almost completely inhibited by 40 nM-heparin when heparin was present from the beginning of the reaction, whereas RNA synthesis from the preformed complexes was partially resistant to heparin (40 nM). Transcription inhibition by heparin was reduced by increasing the enzyme concentration but not by increasing the DNA concentration. This result suggests that heparin is a non-competitive inhibitor of the mtRNA polymerase, as is the case for E. coli RNA polymerase (Zillig et al., 1971). In the presence of 40 nw-heparin, complexes 0, 2, 4 and 1 I produced approximately 25%, 40%, 55% and 7504. respec-

T. K. Biswas

340 Heparln

(nM)

template DNA for different lengths of time at 30°C or on ice before the addition of heparin and other rNTPs (Fig. 7). After preincubat,ion, heparin (40 nM) and ribonucleotides were added together t,o the preformed transcription complex to inactivate the free polymerase and to initiate t’ranscription from the heparin-resistant complex. The reaction was continued for an additional seven minutes and the RNA products were analyzed as above. The relative number of heparin-resistant complexes was determined from the amount of RNA products made. The formation of heparin-resist’ant complex was completed within three minutes and the t,, was approximately one minute. Formation of the heparin-resistant complex is temperature-depen dent, since it increased with increasing temperature of preincubation. i.e. heparin-resistant complex formation at 30°C is four to six times higher than that on ice. (g) Stability

(a)

(b)

Figure 5. Transcriptional activities of the selective and non-selective mtREA polymerases in the presence of heparin. (a) An enzyme preparation containing both the selective and non-selective RNA polymerase activities was used. The mtRPu‘A polymerase was preincubated with template D?;A at 30°C for 10 min, and then ribonucleotides and the indicated concentrations of heparin were added together. After 7 min of incubation. the R?U’A products were analyzed as above. (b) The effect of heparin (40 nM) on the selective mtRNA polymerase activity. Lane 1 (minus heparin), mtRl\jA polymerase and template DPiA were preincubated and then rNTPs were added for transcription; lane 2 (no preincubation, plus heparin), template DNA, mtRiYA polymerase, rNTPs and heparin were added together for transcription; lane 3 (preincubation, plus heparin). mtRNA polymerase and template DNA were preincubated, and then rNTPs and heparin were added together.

tively, of the transcript made in a reaction without heparin. Complex 0 showed the lowest and complex 11 showed the highest degree of heparin resistance. This result revealed that heparin resistance increases with increasing numbers of phosphodiester bonds formed. Similar results of heparin inhibition were obtained with bacteriophage T3/T7 RNA polymerases (McAllister et al.: 1973; Morris et al.. 1987). The formation of di- or tetranucleotide RNA by phage polymerase increases the level of heparin resistance but does not lead to a stable heparinresistant’ complex. Complete resistance of the phage polymerases occurs only after the formation of 10 to 12 nt RNA. (f) Time requirement complex

for heparin-resistant formation

The time requirement for formation of a heparinresistant complex (complex 0 or complex 11) was determined by preincubation of polymerase and

of transcription

complexes

To test the stability of transcription complexes. they were exposed to heparin (40 nM) for different, lengths of t’ime and then the rest of the rNTPs were added t,o complete transcription (Fig. 8). The stability of each complex was determined by estimating the amount of RNA made from each complex. The t,:, for inactivation of transcription complexes were found to be approximately one minute, one minute, five minutes and more than ten minutes for complexes 0, 2, 4 and 11, respectively. Complexes 0 and 2 were very unstable, and lost most, of their activities within t’hree minut’es of incubation with heparin, whereas complex 4 was relatively stable. and 60% of its initial activity survived under such conditions. Extended incubation with heparin further reduced the complex 4 act’ivity; e.g. 3094 activity remained after ten minutes of exposure to heparin. In contrast,. complex 11 was highly stable and retained about 75’!& of its initial activity even after ten minut’es of incubation with heparin. However, the remaining 25 “/& activity was inhibited by heparin, and that activity could ha,ve arisen from polymerase recycling during transcription (see below). This result suggests that the formation of ten phosphodiester bonds is required for conversion of the initiation into a stable elongation-oommit,ted complex complex. Complexes 2 and 4 are each considered as an intermediate between the preinitiation complex (complex 0) and the elongat’ion-committed complex 11. (h) Determination of the rate-limiting in the initiation of transcription

step

The initiation of transcription is a multi-step process requiring assembly of transcription components. conversion into an open complex. first phosphodiester bond formation, and then commitment of RNA polymerase to elongation. These reartions do not proceed at equal rates, so one of the steps must be rate-limiting. To identify the rate-

8. cerevisiae Mitochondrial

341

RNA Polymeruse Hepnrin

IOmln

-

__c

7 min

-

0 StCJp

t I. Template DNA 2. mtRNA polymerase 3. +/- Selective rNTP(s)

t rNTPs + Hepnrin

x

II)

40

Control Complex

0

Complex

2

Complex

4

Complex

II

0-i

J--F-.-. 10 10 [Hrparin]

(a)

(nu) 1x

30

41

’

(nM)

(b)

Figure 6. Effects of heparin concentrations

on transcription. In a control experiment, template DXA. mtRNA polymerase, rNTPs and the indicated concentrations of heparin were added together and incubated for 7 min at 30°C. In other cases, transcription complexes were generated first and then the rest of the rNTPs and different concentrations of heparin were added. RKB products were analyzed as above. (a) The complete gel is displayed to show the quality of the transcripts. (b) Insets of the same gel are shown and the amounts of RNA products were plotted against heparin concentration. ( x ) Control; (0) complex 0; (0) complex 2; (0) complex 4: (A) complex 11.

---

-

2. mtRNA

--) sl,>

Figure

-1.0 0.30

requirement for preinitiation complex formation. DNA template and the mtRNA polymerase were preincubated for different time periods to form a DNA-enzyme complex, and then rNTPs and heparin (40 nM) were added to complete transcription from the heparin-resistant complex. The amounts of RNA products were plotted against the initial incubation time t. (b) Time requirement for formation of the productive initiation complex. First, a preinitiation complex was made by incubating template DNA and the mtRNA polymerase for 10 min at 30°C. After addition of ATP, GTP and UTP, the preinitiation complex was incubated for different times to initiate and to form a short RNA chain. Before addition of the missing nucleotide, these initiation complexes were treated with heparin (40 nM) for 5 min to inactivate any early transcription complexes except the elongation-committed complex. Transcripts produced from the heparin-resistant complexes were measured and plotted against the initial incubation time t.

the gel.

polymerase availability to P2 from the most stable Pl-transcription complex 11 deserves further discussion. Since the initial transcription complexes were made using a molar excess of the Pl template, and no release of polymerase from the stable complex 11 was anticipated before the completion of the first round of transcription on Pl, transcription on P2 under such conditions might be carried out by the polymerase generated by the association of the free core polymerase and the free specificity factor. The mitochondrial specificity factor might fall off the template DNA after commitment of the core polymerase to elongation (e.g. complex ll), as is the case for E. co&i B factor (Travers & Burgess, 1969). These free specificity factors reassociate with some free core polymerases (a contaminant in polymerase preparation) and are then involved in a new set of transcription (Fig. 11). It’ was shown previously that the specific factor can bind to the mitochondrial core polymerase in vitro and then recognize its promoter (Winkley et al., 1985; Schinkel et aZ., 1987; Ticho & Getz, 1988). The

344

T. K. Biswas

Slow reoctlon

Fost reoctlon,

t,,>> 5 s

( G/P - I mm)

Templote DNA

I initiation

Holo mtRNA polymerose

complex

x -

t

I

sp. factor

I

L

RNA 4 Core mtRNA Tronscriptlonol

Figure 11. A proposed model for mitochondrial

difference in transcription from complex 11 in the presence or absence of heparin reflects such activity from polymerase recycling.

4. Discussion The processes of transcriptional initiation, elongation and termination are well defined in prokaryotes (Chamberlin, 1974; von Hippel et al., 1984; McClure, 1985) and partly known in eukaryotes (Schena, 1989; Johnson & McKnight, 1989). Recently, transcription and its regulation in yeast mitochondria have been studied by several laboratories. Two components, mtRNA polymerase (Winkley et al., 1985; Schinkel et al., 1987; Ticho & Getz, 1988) and its octanucleotide promoter (Biswas et al., 1985: 1987; Schinkel et al., 1986), are well characterized in vitro. The interaction between the mtRNA polymerase and its promoter during transcription has been further characterized and is described here. The central goal of this study was to determine the conditions required for the commitment of mtRNA polymerase to transcription. For this study, different mitochondrial transcription complexes were generated in vitro under selective reaction conditions. Since the formation of a transcription complex halted at a specific site of the template DNA may not be a routing process in normal transcription, the generation of different transcription complexes was confirmed by analyzing the transcripts formed in these complexes (Fig. 2) as well as by DNA mobility shift experiments (Fig. 3). The results show that the transcriptional progress of mtRNA polymerase in these complexes was stopped at a specific position, as expected; however, the polymerase retained its ability to produce a fulllength transcript in the presence of all four rNTPs. The stability and transcriptional competency of these complexes were examined using heparin, an inhibitor of free or loosely bound RNA polymerases (Zillig et aZ., 1971; McAllister et al., 1973). Two important findings were obtained from this experiment. First, inhibition of transcription by heparin is dependent on heparin concentration; secondly, each

transcription

pol

termlnotion

in S’accharomyce~sceraisiae.

complex responded differently to heparin inhibition. Since the in vitro generated mitochondrial transcription complexes are mostly homogeneous, the variable sensitivity of these complexes to heparin appears not to be due to their heterogeneity. Models for heparin inhibition are proposed. Since the early transcription complexes (e.g. complexes 0, 2 and 4) are not stable, it is expected that a part of these complexes start transcription immediately after the addition of rNTPs and heparin, whereas the other part is inactivated by heparin due to dissociation of polymerase from the template. Complex 11 is the most stable complex and produces the highest level of RNA (approx. 75%) in the presence of heparin. It is worth mentioning that the core mt,RNA polymerase is always sensitive to heparin, since it can not make a stable transcription complex. The other possibility of heparin inhibition could be due to the presence of two heparin binding sites in the polymerase molecule, as has been suggested for E. coli RNA polymerase (Chamberlin, 1976). These two sites are probably used for binding of DNA template and the RNA product under normal reaction conditions. The E. coZi RNA polymerase becomes resist’ant to heparin when polymerase is in a ternary complex and contains an RNA molecule (Zillig et al.. 1971). Similarly, both the DNA and RNA binding sites of the mtRNA polymerase in elongationcommitted complex 11 are occupied by the DNA template and the RNA product, and complex 11 is no longer sensitive to heparin. In contrast, the RNA binding site of mtRNA polymerase in complexes 0, 2 and 4 is still susceptible to heparin attack and these complexes are inhibited by heparin. If this is the case, 11 nt RNA should have the capacity to cover the RNA binding site of the mtRNA polymerase and protect from heparin inhibition. In this respect, the RNA binding site of mtRNA polymerase is similar to the RNA binding site of E. coli RNA polymerase, which has the capacity to hold a 10 to 12 nt RNA (Chamberlin, 1976; von Hippel et aZ., 1984). Based on these observations, three progressive steps of reaction heading towards the productive

S. cerevisiae

Mitochondrial

initiation of transcription are proposed. The first step is the formation of a preinitiation complex (complex 0), which is considered to be a template DNA-RNA polymerase assembled complex. In the presence of rNTPs, the preinitiation complex is converted into an early initiation complex (e.g. complexes 2 and 4) by catalyzing phosphodiester bond formation. The early initiation complex RNA synthesis and remains noncontinues committed until it forms an 11 nt or bigger RNA. Complex 11 is very stable, and rapidly starts RNA synthesis when four rNTPs are present (Fig. 4). In this regard, complex 11 is functionally analogous to the open promoter complex in E. coli (Chamberlin, 1976) and to the rapid start complex in higher eukaryotes (Hawley & Roeder, 1985, 1987). Since these reaction steps are defined on the basis of heparin resistance and stability of the artificially generated transcription complexes, it is not clear whether each step is an individual step or a combination of several sub-steps. The formation of preinitiation complex appears to be the slowest step in mitochondrial transcription, with a t,,* of 50 seconds. The conversion of a preinitiation complex int.0 a productive initiation complex is a fast process (i.e. tliz less than 5 s). Thus, in a manner similar to the E. coli transcription system (Chamberlin, 1974), the mammalian (Hawley & Roeder, 1985, 1987) and Drosophila (Kadonaga, 1990) RNA polymerase II transcription systems, formation of the preinitiation complex is the rate-limiting step in mitochondrial transcription. However, the rate of preinitiation complex formation in mitochondria (t,,, approx. 50 s) is more rapid than the Drosophila (t1,2 approx. 3 min) and mammalian syst’ems (t,,, approx. 8 min). This variation could be due to the differences between the simple mitochondrial transcription machinery versus the more complex nuclear system. Although the formation of preinitiation complex is the ratelimiting step in most transcription, there is at least one exception. The rate-limiting step of transcription on the 1acUV promoter by the E. coli RNA polymerase is the escape of polymerase from the futile abortive initiation process to the productive elongation, rather than preinitiation complex formation (Stefano & Gralla, 1979; Munson $ Reznikoff, 1981). An important parameter of gene regulation is the number of rounds of transcription that can occur at each promoter. Two pieces of evidence suggest that recycling of polymerase or its subunit(s) might occur in mitochondrial transcription. First, the mitochondrial RNA synthesis continues for seven to ten minutes under normal reaction conditions (Biswas $ Getz. 1990), although approximately two to three minutes is sufficient for preinitiation complex formation and less than five seconds is required for initiat’ion/elongation of transcription. Secondly, when heparin (40 nhr) was used to limit RNA synthesis on the preformed complexes to a single round, transcription was reduced to between 30 and 7O”/b of the control level: depending on the nature of

RNA Polymerase

345

the complexes. Polymerase recycling can occur before or after completion of the first round of transcription. The process, in which polymerase reinitiates before completing the first round of transcription, is called polymerase recycling at the promoter. In the case of multiple rounds of transcription, polymerase reinitiates RNA synthesis after completion of the first round of transcription. Multiple rounds of transcription can be demonstrated by showing that the number of transcripts is greater than the number of polymerase molecules. Since there was no definite information regarding the concentration of holoenzyme in the mtRNA polymerase preparation, a different strategy (a template challenge experiment: Fig. lo), rather than measuring the transcript to polymerase ratio, was selected. The mtRNA polymerase and saturation excess of the Pl template were incubated first to allow polymerase-promoter complex formation and then the second template (P2) was added to detect any free polymerase in the reaction. Interestingly, P2 transcript was obtained under these reaction conditions, even though a five-times saturation excess of Pl was used in the first incubation. To determine whether the availability of polymerase to P2 occurs before the first round of transcription on Pl, heparin was used to restrict a single round of transcription. In this case, P2 was preincubated with the preformed Pl-mtRNA polymerase complex before the addition of rest of the rNTPs and heparin. Some polymerases were still available to P2 under these conditions. Such polymerase activity from the most stable complex, complex 11, favors the hypothesis of recycling of specificity factor rather than the polymerase itself as is the case in E. coli transcription (Travers & Burgess, 1969; Helmann & Chamberlin, 1988). The overexpression of the core polymerase increases mitochondrial transcription by 30 to 4076 (Judith Jaehning, personal communication). leading to the notion that the specificity factor might recycle in vivo. Interestingly, the mitochondrial specificity factor and the E. coli specificity factor a7’ share three regions of amino acid sequence homology, including the conserved hydrophobic, and positively charged amino acids blocks (Lisowsky & Michaelis, 1989; Jang & Jaehning, 1991) that are required for the recognition of the E. coli promoters (Helmann & Chamberlin, 1988). In addition, the mitochondrial consensus promoter sequence shows a high degree of similarity to the consensus sequence of the major E. coli promoters (e.g. TAtAaG for mitochondrial promoter versus TAtAaT for the consensus Pribnow box: Biswas et aE., 1987). All these observations suggest that the mitochondrial specificity factor and the E. coli aTo might function in a similar way for promoter recognition. The availability of antibodies against the mitochondrial core polymerase and the specificity factor (Dr Godfrey Get’z’s laboratory, University of Chicago), provides an opportunity to test directly the hypothesis of subunit recycling by antibody-specific inhibition of transcription (Ticho & Getz. 1988) and that will be described elsewhere.

346

T. K. B&was

Similar studies of transcriptional commitment have been done with E. coli RNA polymerase on the ZacUV5 (Carpousis & Gralla, 1985; Hansen & McClure, 1980; Munson & Raznikoff, 1981), T3, T7 and A PL promoters (Levin et al., 1987), and with RNA polymerase II from higher eukaryotes (Howley & Roeder, 1987; Cai & Luse, 1987; Kadonaga, 1990). In E. coli, the most dramatic change in transcription occurs following the formation of ten phosphodiester bonds. After initiation of transcription, some E. coli RNA polymerases continue transcription, whereas the other polymerases lose the nascent RNA and recycle abortive initiation of transcription at the promoter without dissociating from the template. After formation of more than 10 nt RNA, the E. coli RNA polymerase loses 0 factor and fully commits to elongation. The bacterial RNA polymerase is therefore in an initiation step of transcription until it passes through the nucleotide at position 10. The RNA polymerase II in higher eukaryotes also passes through several steps of reaction before reaching the initiation/elongation stages of transcription (Howley & Roeder, 1987; Kadonaga, 1990). Like E. coZi RNA polymerase, after initiation of transcription at the adenovirus 2 major late promoter, the human polymerase II either goes to elongation or passes through several cycles of abortive initiation (Luse et al., 1987; Luse & Jacob, 1987). However, the human RNA polymerase II acquires most of the characteristics of an elongation-commited complex only after the formation of one or two phosphodiester bonds (Luse et al., 1987). As with the other RNA polymerases, the mtRNA polymerase passes through a few reaction steps before reaching the actual elongation process. After formation of ten or more phosphodiester bonds, the mtRNA polymerase probably undergoes a major conformational change and fully commits to transcription. Based on these observat.ions, a simple model for mitochondrial transcription in yeast has been proposed (Fig. 11). The results presented in this paper are in agreement with the previous report on mitochondrial transcription by Schinkel et al. (1988), who showed by DNA cleavage protection assay that synthesis of a 2, 4 or 8 nt transcript causes progressive protection of the downstream sequence of the mtDNA template without changing the upstream protected region, which is also the case for bacteriophage T7 RNA polymerase (Ikeda & Richardson, 1986). These results suggest that both the mtRNA polymerase and T7 RNA polymerase can synthesize 8 nt RNA without leaving their promoters; however, synthesis of this small RNA is not enough for the formation of a stable transcription complex. Considering amino acid sequence homology, a single block of promoter sequence adjacent to the transcriptional start site, rifampicin resistance and the pattern of transcription commitment, the mtRNA polymerase shows similarity to the T3 (McAllister et al., 1973) and T7 (Ikeda $ Richardson, 1986) bacteriophage RNA polymerases to a greater degree than any other characterized RNA polymerases.

I thank Dr William Zehring for reviewing the manuscript. This work is supported by American Heart Association grant 890859. References Biswas, T. K. (1990). Control of mitochondrial gene expression in the yeast Saecharomyces cerevisiae. Proc. Nat. Acad. Sci., U.S.A. 87, 9338-9342. Biswas, T. K. (1991). In vitro transcription analysis of the cerevisiae mitochondrial region of Saccharomyces DEjA containing the tRNAfMe’ gene. Nucl. Acids Res. 19, 5937-5942. Biswas, T. K. & Getz, G. S. (1986a). Pjucleotides flanking the promoter sequence influence the transcription of the yeast mitochondrial gene coding for ATPase subunit 9. Proc. Nat. Acad. Sci., U.S.A. 83, 270-274. Biswas, T. K. & Getz, G. S. (1986b). A critical base in the mitochondrial yeast nonanucleotide promoter. J. Biol. Chem. 261, 3927-3930. Biswas, T. K. & Getz, G. S. (1988). Promoter-promoter interactions influencing transcription of the yeast mitochondrial gene, Oli 1, coding for ATPase subunit 9: cis and trans. J. Biol. Chem. 263, 48444851. Biswas, T. K. & Getz, G. S. (1990). Regulation of transcription initiation in yeast mitochondria. J. Biol. Chem. 265, 19053-19059. Biswas, T. K., Edwards, J. C., Rabinowitz, M. & Getz, G. 8. (1985). Characterization of a yeast mitochondrial promoter by deletion mutagenesis. Proc. Nat. Acad. Sci., U.S.A. 82, 195441958. Biswas, T. K., Tieho, B. & Getz, G. S. (1987). In vitro characterization of the yeast mitochondrial promoter mutants. J. Biol. using single-base substitution Chem. 262, 13690-13696. Cai. H. & Luse, D. S. (1987). Transcription initiation by RNA polymerase II in vitro. Properties of preinitiation, initiation, and elongation complexes. J. Biol. Chem. 262, 298-304.

Carpousis, A. J. & Gralla, J. D. (1985). Interaction of RPU’Apolymerase with ZacUV5 promoter DKA during mRru’A initiation and elongation: Footprinting, changes methylation, and rifampicin-sensitivity accompanying transcription initiation. J. Mol. Biol. 183, 165177. Chamberlin, M. J. (1974). The selectivity of transcription. Annu. Rev. Biochem. 43, 721-775. Chamberlin, M. (1976). Interaction of RNA polymerase with the DPU’Atemplate. In RNA Polymerase (Losick, R. & Chamberlin, M., eds), pp. 159-191, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, ;“u’Y. Christianson, T. & Rabinowitz, M. (1983). Identification of multiple transcriptional initiation sites on the yeast mitochondrial genome by in vitro capping with guanylyltransferase. J. Biol. Chem. 258, 1402514033. Grivell, L. A. (1989). Nucleo-mitochondrial interactions in yeast mitochondrial biogenesis. Eur. J. B&hem. 182, 477-493. Hahn, S., Hoar, E. T. & Guarente, L. (1985). Each of three “TATA” elements specifies a subset of the transcription initiation sites at the CYC-1 promoter of Saccharomyces cerevisiae. Proc. Nat. Acad. Sci., C1.S.A. 82, 8562-8566. Hansen. U. M. & McClure, W. R. (1980). Role of the sigma subunit of Escherichia coli RNA polymerase in initiation: release of sigma from ternary complexes. .J. Biol. Chem. 255, 9564-9570.

S. cerevisiae Mitochondrial Helmann, J. D. & Chamberlin, M. J. (1988). Structure and function of bacterial sigma factors. Annu. Rev. Biochem.

57, 839-872.

Howley, D. K. & Roeder, R. G. (1985). Separation and partial characterization of three functional steps in transcription initiation by human RNA polymerase TI. .J. Biol. Chem. 260, 8163-8172. Howley, D. K. & Roeder, R. G. (1987). Functional steps in transcription initiation and reinitiation from the major late promoter in a HeLa nuclear extract. J. Biol. Chem. 262, 3452-3461. Ikeda. R. A. & Richardson, C. C. (1986). Interactions of the RNA polymerase of bacteriophage T7 with its promoter during binding and initiation of transcription. Proc. Nat. Acad. Sk., C7.S.A. 83. 3614-3618. ,Jang. S. H. & Jaehning. J. A. (1991). The yeast mitochondrial RNA polymerase specificity factor, MTFl, is similar to bacterial sigma factor. J. Biol. Chem. 266, 22671-22677. Johnson. P. F. & McKnight. S. L. (1989). Eukaryotic Annu. Rev. transcriptional regulation proteins. Biochem.

58: 799-840.

Kadonaga, J. T. (1990). Assembly and disassembly of the Drosophila polymerase II complex during transrription. J. Biol. Chem. 265, 2624-2631. Levene, D.. Lustig. A. & Rabinowitz, M. (1981). Purification of mitochondrial RNA polymerase from Saccharomyces cer&siae. J. Biol. Chenb. 256, 14741481. Levin. -1. It.. Krummel, B. & Chamberlin, M. J. (1987). Isolation and properties of transcribing ternary complexes of Escherichia coli RNA polymerase positioned at a single templat,e base. J. Mol. Biol. 196, 85-100. Lisowskp. T. & Michaelis, G. (1989). A nuclear gene essential for mitochondrial replication suppresses a defect of mitochondrial transcription in Saccharomycrs cerevike. Mol. Gen. Genet. 214, 218223. Luse. D. 8. &, Jarob. G. A. (1987). Abortive initiation by RNA polymerase II in vitro at the adenovirus 2 major late promoter. .I. Biol. Chem. 262, 1499014997. Luse. D. S.. Kochel, T.. Kuempel, D. E., Coppola, J. A. & (‘ai, H. (1987). Transcription initiation by RNA polymerase IT in vitro. At least two nucleotides must be added to form a stable ternary complex. J. Biol. (‘hem. 262, 289-297.

AMasters, B. S., Stohl, L. I,. & (‘layton, I). A. (1987). Yeast mit.ochondrial RNA polymerase is homologous to those encoded by bacteriophages T3 and T7. Cell, 51, 89-99. McAllister. W. T.. Kupper, H. & Bautz, E. K. F. (1973). Kinetics of transcription by the bacteriophage T3 RXA polymerase in vitro. Eur. J. Biochem. 34, 489501.

McClure. W. R. (1985). Mechanism and control of transcription initiat,ion in prokaryotes. Annu. Rev. Jl%ochpm. 54. 171-204.

RNA

Polymerase

347

Morris, C. E.. McGraw, N. J., Joho, K., Brown. J. E., Klement, J. F., Ling, M. L. & McAllist,er. W. T. (1987). Mechanisms of promoter recognition by the bacteriophage T3 and T7 RNA polymerases. In RNA Polymerase and the Regulation of Transcription. Proc. 26th Steenbock Symp. Univ. Wisconsin. Madison (Reznikoff et al.. eds), pp. 47-58, Elsevier, New York.

Munson, L. M. & Reznikoff, W. S. (1981). Abortive initiaand long acid synthesis. tion ribonucleic Biochemistry,

20, 2081-2085.

Ruther, U., Koenen, M.. Otto, K. & Muller-Hill. B. (1981). pUR 222, a vector for cloning and rapid chemical sequencing. Nucl. Acids Res. 9, 4087-4098. Schena, M. (1989). The evolutionary conservation of eukaryotir gene transcription. Experientia, 45. 972983.

Schinkel, A. H.. Groot Koerkamp. M. I. A.. Vander Horst. G. T. J., Touw, E. P. W.. Osinga. K. A., Vander Bliek, A. M.. Veenemam, G. H. & Tabak, H. F. (1986). Characterization of the promoter of the large ribosomal RNA gene in yeast mitochondria and separation of mitochondrial RNA polymerase into two different functional components. EMBO .J. 5, 1041-1047. Schinkel, A. H., Groot Koerkamp, M. ,J. A.. Touw, E. P. W. & Tabak, H. F. (1987). Specificity factor of yeast mitochondrial RNA polymerasc. Purification and interaction with core polymerase. .f. Kiol.