RNA-polymerase binding at the promoters of the rRNA genes of Escherichia coli

435

Biochimica et Biophysica Acta, 609 (1980) 435--447 © Elsevier/North-Holland Biomedical Press

BBA 99743

RNA-POLYMERASE BINDING AT THE PROMOTERS OF THE rRNA GENES OF ESCHERICHIA COLI

IBOLYA KISS a, IMRE BOROS a, A N D O R U D V A R D Y a PAL VENETIANER a and HAJO DELIUS b a Institute of Biochemistry, Biological Research Center, Szeged (Hungary) and b European Molecular Biology Laboratory, Heidelberg (F.R.G.)

(Received April 24th, 1980) Key words: RNA polymerase binding; rRNA gene; Promoter site; (E. coil)

Summary The promoter region of two bacterial r R N A genes was investigated by electron-microscopic analysis of polymerase binding, transcription initiation and nitrocellulose filtration of RNA-polymerase-DNA complexes, using restriction endonuclease generated fragments of recombinant plasmids and a transducing phage. The following observations have been made: 1. T w o transcription initiation sites have been located approximately 200 and 300 base pairs upstream from the beginning of the sequence coding for mature 16 S rRNA. 2. Polymerase binding at these sites can be observed electronmicroscopically and a 360 base-pair fragment containing these sites binds to nitrocellulose in the presence of RNA-polymerase. This complex dissociates even at moderately high (0.1--0.2 M) Salt concentrations. Although transcription initiation is reported to be more frequent at the first of these sites,the binding is m u c h stronger at the second site. 3. In the case of the rrnD gene, B a m H I cleaves a few base pairs upstream from the frrst transcription start site. This cleavage destroys polymerase binding at this site but does not influence binding at the second site. 4. At higher polymerase/DNA ratio four weak but distinct and regularly spaced binding sites can be observed preceding the two initiation sites at approximately 1000, 820, 640 and 440 base pairs before the mature 16 S r R N A sequence. 5. A n extremely strong binding site is located about 1300 base pairs up-

Abbreviations: Hepes, N-2-hydroxyethylpiperazine-NP-3.propanesulfonic

acid.

436 stream from the beginning of the 16 S rRNA sequence. Very little (if any) initiation occurs at this site. The possibility is discussed that the noninitiating binding sites preceding the two transcription start points might functionally belong to the promoter region.

Introduction Ribosomal RNA genes are the most actively transcribed genes of the bacterial genome. It can be calculated that at high growth rates all seven rRNA genes are transcribed at the maximal speed allowed by the rate of chain elongation, thus the rate of initiation does not limit transcription. As preferential transcription of rRNA genes can be observed in purified in vitro systems as well, it is generally believed that the extremely high rate of transcription initiation on these genes is an inherent property of the rRNA promoters. On the basis of in vitro transcription experiments, several laboratories proposed that this high frequency of transcription could be explained by assuming the existence of multiple promoters at the beginning of each rRNA transcription unit. The numbers varied from 4--5 [1,2] to 30 [3]. However, similar transcription studies on transducing phages carrying rRNA genes did not support this hypothesis [4,5]. Rifampicin challenge experiments suggested the existence of two initiation sites per rRNA gene [4]. Recently the nucleotide sequences of the promoter regions of five different rRNA genes have been determined [6-8]. In vitro transcription studies on isolated DNA fragments confirmed that there are only two initiation sites per rRNA gene and localized the positions of these sites [7,9,10]. In these experiments, however, no direct attempt was made to visualize the sites of polymerase binding. The electron microscopic studies reported here were undertaken in order to resolve the apparent contradiction between the number of established initiation sites (two) and the assumed binding sites (4--5). They demonstrate that the two initiation sites are indeed preceded by weak binding sites which do not initiate, but could functionally belong to the rRNA gene. Materials and Methods Enzymes Sigma saturated E. coli RNA-polymerase was prepared according to Burgess and Jendrisak [11] by chromatography on phosphocellulose in glycerol. Restriction endonucleases were prepared in this laboratory following established protocols [12]. DNA Phage DNA was purified according to Miller's manual [13]. Plasmid DNA was purified by a fast procedure involving hydroxyapatite [14]. Restriction endonuclease generated fragments were isolated from agarose gels by the method of Koller et al. [15].

437

Electron-microscopic techniques In polymerase binding experiments 0.3--1 ~g DNA fragment was incubated for 5 min at 37°C in a final volume of 50/~1 binding buffer (10 mM potassium phosphate (pH 7.5)/10 mM MgC12/0.1 mM EDTA/0.1 mM dithiothreitol/50 mM KC1). RNA-polymerase was added in 2--4.fold molar excess and incubated for another 5 rain. The enzyme-DNA complexes were fixed in-the presence of 0.1% glutardialdehyde and after 15 rain at 37°C the mixture was passed through a small Sepharose CL-2B column, equilibrated with 4 mM magnesium acetate and 0.01% glutardialdehyde. The adsorption of enzyme-DNA complexes to freshly cleaved mica surface and specimen preparation was carried out as described by Portmann and Koller [16] except that only 4 mM magnesium acetate and 0.01% glutardialhyde were present during adsorption. For the electron microscopy of transcriptional complexes, 30 /~g/ml DNA and 20--30/~g/ ml RNA-polymerase were incubated at 37°C for 2 and 5 min in 20 mM TrisHC1 (pH 7.9)/10 mM MgC12/0.1 mM EDTA/50 mM KC1/200 pM each of ATP, GTP, UTP, CTP. Preparation of transcriptional complexes, glutardialdehyde fixation in the presence of T4 phage 32-protein, spreading, specimen-preparation, measurements on photographic negatives and computation of initiation sites were all carried out as described by Stiiber et al. [17]. R-loops on plasmids were formed by incubating 5--10 ~g/ml DNA and 15 ~g/ml E. coli rRNA in 80% formamide/0.1 M Hepes (pH 7.3)/0.33 mM NaC1/10 mM EDTA at 45°C for 1 h in 70 ~1 final volume [18]. R-loops on short linear DNA fragments were unstable under these conditions. For this reason R-loops were formed on intact plasmid circles as described above and fixed after the addition of 40 ~g/ml T 4 32-protein with 0.1% glutardialdehyde to complex the single-stranded DNA. After passage through a Sepharose CL-2B column, equilibrated with 20 mM Tris-HC1 (pH 7.5)/7 mM MgC12/2 mM 2-mercaptoethanol, the DNA was digested with BamHI restriction endonuclease. Samples were spread by the formamide variation of the basic protein film technique [19]. 0.5 ug/ml DNA in 0.1 M Tris-HC1 (pH 8.5)/1 mM EDTA/30-40% formamide/100 ~g/ml cytochrome c was spread on a hypophase containing 10--15% formamide/10 mM Tris-HC1 (pH 8.7)/0.1 mM EDTA. The protein film was picked, up on Parlodion~oated grids, stained with uranyl acetate (5 mM uranyl acetate/0.05 M HC1/95% ethanol, diluted 1000-fold into 90% ethanol just before use), rinsed with 90% ethanol and isopentane, and rotary shadowed with platinum at an angle of 8 ° . Pictures were taken with a Philips electron microscope at a magnification of 5000 or 10 000 at 40 kV. Length measurements were made on photographic negatives using an X-Y measuring stage connected with a Wang calculator system. Length distributions of restriction fragments were determined and molecular weights calculated by using the 10 kilobase phage PM2 DNA as internal standard [20]. 50--100 DNA molecules were measured to obtain the final histograms. Filter binding The method of Seeburg and SchaUer [21] was used to detect specific RNApolymerase-DNA complexes by retention on nitrocellulose filters.

438

Results Physical maps of the different DNAs used in this study are shown in the schematic drawings of Fig. 1. The maps give the positions of restriction enzyme sites relevant for this paper on the DNA of the transducing phage krifdl8 carrying the rrnB operon [22], of the recombinant plasmid 2/12 derived from this phage [23], and the recombinant plasmids pBK8 and pBK18 carrying the rrnD and rrnB regions cloned from BamHI fragments of the bacterial chromosome [24]. In an earlier paper [4] we described that most of the in vitro transcription of krifd18 DNA represents the transcription of the rRNA operon, and on the basis of rifampicin challenge experiments we determined two rRNA initiation sites. However, in these experiments the promoter sites were not accurately mapped, and the possibility of read-through from other promoter sites could not be rigorously excluded. Fig. 2b shows the micrograph o f an in vitro transcription complex of krifdl8 DNA using an electron-microscopic method described earlier [17,25]. The summary of measurements on these complexes is given in Fig. 3. The highest frequency of initiation of RNA synthesis is found in a position around 56--57% of

"r E rif d 18

~' rplLrplA

tufB

~' ~ ~I' int

rrn B

--

o....~~ / u / H H 1 / / l ~ PL

"/if/,,:,'/" r PR

all

o nn

; 't

2 112

i

L. . . . . . .

o..

nnuJ "r

Pst I

" PstI Pst I

.........

pBR

o f f ' \,, \

..................

.x

313 E=

"r E

I=1 "o ~.__.

o

~ ~

I=1 ~

-,-

E o

p B K 17

-r

pBR 322

e

pBK 8

,,

~

-r

o

o E

,,,

=

r r n DEc ° R I t

pBR 322

Fig. 1. Physical m a p s o f the transducing phage ;Lrifd18, and t h e r e c o m b i n a n t plasmids used in this w o r k . Plasmid 2 / 1 2 candes the 7 . 7 4 k i l o b a s e BamI-II fragment o f ;Lrifdl8 w i t h t h e rrnB gene inserted i n t o the v e c t o r p B R 3 1 3 . Plasmids p B K 8 a n d p B K 1 7 c o n t a i n 8 . 3 a n d 1 5 k i l o b a s e B a m H I f~agments of the b a c t e r i a l c h c o m o s o m e carrying r m B and r~nD, respectively, inserted i n t o the v e c t o r p B R 3 2 2 . Only t h o s e restriction sites w h i c h w e r e u s e d for fragment i s o l a t i o n are i n d i c a t e d . H a t c h e d bars: D N A o f l a m b d a phage or plasmid v e c t o r origin. Open bars: D N A originating f r o m the bacterial c h r o m o s o m e . Solid bars: D N A c o d i n g for the m a t u r e 1 6 S a n d 2 3 S R N A .

439

Fig. 2. E l e c t r o n m i c r o g r a p h s o f R - l o o p s a n d t r a n s c ~ p t i o n a l c o m p l e x e s , a . R - l o o p s w e r e f o ~ m e d b y L ' ~ a h a t i n g 8 /~g/ml k r i f d 1 8 D N A a n d 1 6 # g / m l E. coli r R N A i n 8 0 % f o r m a m i d e / 0 . 3 3 M N a C 1 / 0 . 1 M H e p e s (pH 7.5)/10 mM EDTA at 45°C for 1 h. b. In vitro transcription was carried out at 37°C for 2 rain on k r i f d 1 8 D N A a t a p o l y m e r a s e : D N A w e i g h t r a t i o , 0 . 8 . P r o c e d u r e w a s as d e s c r i b e d i n M a t e r i a l s a n d M e t h o d s . S i n g l e - s t r a n d e d p h a g e M 1 3 D N A m o l e c u l e s c a n also b e seen. c, R - l o o p s f o r m e d o n i n t a c t 2 / 1 2 p l a s m i d D N A w e r e f i x e d in t h e p r e s e n c e o f T 4 3 2 p r o t e i n w i t h 0 . 1 % g l u t a x a l d e h y d e a n d c l e a v e d w i t h BamHI enzyme.

the kri•18 genome (Fig. 3b). krit~18 DNA containing ribosomal R-loops was prepared (Fig. 2a). The comparison of the transcription map (Fig. 3b) with the R-loop map (Fig. 3c) indicates that the position showing the highest frequency of RNA synthesis starts is very close to the start of the 16 S RNA. A higher accuracy of the measurements was achieved using a 7.74 kilobase BamHI fragment derived from plasmid 2/12 which carries the same rrnB operon. The analysis of transcription complexes prepared with this fragment (Fig. 4b) shows that most of the RNA chains are initiated at a site located 1.2 to 1.3 kilobases away from the BarnHI cut. Very few chains originate from a site approximately 0.2 kilobase away from the BamHI site. A comparison with the R-loop measurements on this fragment (Figs. 2c and 4c) shows that the major promoter

440 20 i

rpoB,C

a

i

i

L,J A K 'U'' ~

40 i

i

E

rrnB

60 i

i

'

i

"

'"

~

I'

~'"

'

'"

11'I ~''

80 i

i

i

'

~I'I

i

~00 2 %

i

''

i

I

'

Fig. 3. L o c a l i z a t i o n o f i n v i t r o R N A i n i t i a t i o n sites o n ~krifdl8 D N A . a. S c h e m a t i c m a p o f t h e p h a g e showing the position of known promoters and direction of transcription, b. Statistical evaluation of transcriptional complexes. Start points for rightward transcription are shown above the bar, start points for l e f t w a r d t r a n s c r i p t s b e l o w t h e bax. c. S t a t i s t i c a l e v a l u a t i o n o f R - l o o p s f o r m e d w i t h E. coli r R N A .

site is near the start of the 16 S R N A sequence. This localization was confirmed by the filter-binding experiments shown on Fig. 5. BspI cleavage produces four fragments from the 1.33 kilobase PstIHindIII fragment of plasmid 2/12 (710, 200, 360, 34 base-pair). The 360 basepair fragment is specifically b o u n d to nitrocellulose in the presence of RNApolymerase. In order to correlate the result of the analysis of transcription complexes with the filter binding studies the binding sites of R N A polymerase were determined using a mica adsorption technique [26] which yields very precise electron microscopic measurements. A map of the binding sites of R N A polymerase to an EcoRI subfragment of the 7.74 kilobase BamHI fragment of plasmid 2/12 is shown in Fig. 4a. The major binding site is located 0.25 kilobases away from the BamHI end, and should therefore correspond to the site which initiates R N A synthesis only with very low frequency. A binding site of comparable affinity is positioned at a distance o f 1.35 kilobases from the BamHI end. This position should correspond to the major p r o m o t e r region for the rrnB operon on the 2/12 plasmid DNA. The positions marked $1 and $2 in the graph below the binding map indicate initiation positions as determined by sequence analysis [8]. It is evident that there is distinct binding to a site corresponding to $1 b u t it occurs with a much lower frequency. Plasmid pBK8 contains a BamHI fragment carrying the rrnD operon from E. coli [24]. From the published sequence of this operon [7] it is known that

441

8

Ai m

-

=

=

,J-

I

im,,|L -1.5

-1.0

-0.5

0

+0.5

kb

t_.L, S1 S 2

m

16 S

rRNmA

t

T

Z

Z

E_

P -1 | w

0 i

i

+2 i

!

+4 i

i

+6 k b i I

C

IeS

23S

Fig. 4. Localization of RNA-polymerase binding and initiation sites in the p r o m o t e r region of the rrnB gene of ~'Lfd18. a. RNA-polymerase binding on the 2.2 kflobase B a m H I - E c o R I fragment of plasmid 2/12 at 2.5 : 1 enzyme/DNA molar ratio, b. In vitro transcription initiation sites on the 7.74 kilobase B a m H I fragment of plasmid 2/12. Enzyme/DNA molar ratin, 5 : 1. Transcription was carried out at 37°C for 2 and 5 rain. Only rightward transcripts were detected, c. R-loop mapping on the same 7;74 kflobase B a m H I fragment of plasmid 2/12. R-loops were formed on the intact plasmid, before cleavage with B a m H I as described in Materials and Methods. S 1 and S 2 are the transcription start sites determined by in vitro transcription [10] and sequencing [8].

442

1

2

3

4

5

6

7

8

9

10

11

12

13

Fig. 5. R N A - p o l y m e r a s e b i n d i n g t o BspI g e n e r a t e d f r a g m e n t s o f p l a s m i d 2 / 1 2 d e t e c t e d b y r e t e n t i o n o n n i t r o c e l l u l o s e filters. P l a s m i d 2 / 1 2 w a s d i g e s t e d w i t h BglII a n d PstI, t h e 1 . 9 k i l o b a s e f r a g m e n t s i s o l a t e d b y a g a r o s e gel e l e c t r o f o r e s i s ( t h e PstI-BglII f r a g m e n t c a r r y i n g t h e r r n B p r o m o t e r c a n n o t b e s e p a r a t e d f r o m t h e s i m i l a r PstI f r a g m e n t o f t h e v e c t o r ) d i g e s t e d w i t h BspI a n d e n d - l a b e l e d w i t h [ ' y - 3 2 p ] A T P a n d p o l y n u c l e o t i d c k i n a s e . B i n d i n g e x p e r i m e n t s w e r e c a r r i e d o u t as d e s c r i b e d b y S e e b u r g a n d S c h a l ] e r [ 1 9 ] . P o l y m e r a s e / D N A r a t i o s a n d KC1 c o n c e n t r a t i o n s w e r e v a r i e d as s h o w n b e l o w . S l o t s 1 - - 3 : E t h i d i u m b r o m i d e s t a i n e d a g a r o s e gels. 1. p B R 3 2 2 v e c t o r D N A d i g e s t e d w i t h BspI, 2. 1 . 3 3 k b PsfI-HindIII f r a g m e n t o f r e c o m b i n a n t p l a s m i d 2 / 1 2 , d i g e s t e d w i t h BspI. T h e 3 6 0 b a s e p a i x m i d d l e f r a g m e n t c a r r i e s t h e PI a n d P2 p r o m o t e r s . 3. T h e t w o 1 . 9 k i l o b a s e f r a g m e n t s o f p l a s m i d 2 ] 1 2 g e n e r a t e d b y PstI a n d BglII d i g e s t i o n , d i g e s t e d b y BspI. S l o t s 4 - - 1 3 : A u t o r a d i o g r a m s o f t h e 3 2 p - l a b e l e d f r a g m e n t s a f t e r a g a r o s e gel eleetxop h o r e s i s . 4. S a m e as 3. 5 - - 1 3 . T h e s a m e d i g e s t as in s l o t s 3 a n d 4. R e t e n t i o n o n n i t r o c e l l u l o s e a f t e r i n c u b a t i o n w i t h R N A - p o l y m e r a s e . In p a r e n t h e s e s is t h e e n z y m e / D N A w e i g h t r a t i o . 5. 4 0 m M KC1, ( 1 . 3 : 1); 6. 4 0 m M KC1 ( 1 . 7 : 1); 7. 4 0 m M KCI ( 2 . 6 : 1); 8. 4 0 m M KC1 ( 4 . 3 : 1); 9. 4 0 m M KC1 ( 8 . 6 : 1); 1 0 . 1 0 0 m M KC1, (1.1 : 1); 1 1 . 1 0 0 m M KC1 ( 2 . 2 : 1); 1 2 . 2 0 0 m M KC1, (1.1 : 1); 1 3 . 1 0 m M KCI ( 2 . 2 : 1).

i8°I

;,°i

:,°I

; °li Oi

I

i I

-1

,

i

i

,

i

i

i

i

L_>~_>o, S1

S2

i

i

•

+1 kb rn

16 S

rRNA

Fig. 6. R N A - p o l y m e r a s e b i n d i n g o n t h e 1 . 3 7 k i l o b a s e E c o R I f r a g m e n t o f p l a s m i d p B K 8, c o n t a i n i n g t h e r r n D p r o m o t e r r e g i o n . E n z y m e / D N A m o l a r , r a t i o 2 : 1. T h e D N A t o t h e r i g h t o f t h e Bam H I site o r i g i n a t e s f r o m t h e v e c t o r p l a s m i d . S 1 a n d S 2 d e n o t e s t h e k n o w n sites o f i n i t i a t i o n , t h e BarnHI site is 1 9 b a s e p a i r t o t h e l e f t f r o m S 1.

443

Fig. 7. E l e c t r o n m i c r o g r a p h s of R N A o p o l y m e r a s e m o l e c u l e s b o u n d t o t h e r r n B p r o m o t e r r e g i o n , a a n d b . 4 . 0 5 k i l o b a s e BgiII f r a g m e n t o f p l a s m i d p B K 1 7 . c. 1 . 3 3 k i l o b a s e PstI-HindIII f r a g m e n t o ~ p l a s m t d p B K 17. The e x p e r i m e n t s w e r e p e r f o r m e d a t 6 : 1 o r 3 : 1 e n z y m e / D N A r a t i o m o l a r as d e s c r i b e d in Materials a n d M e t h o d s . P h a g e PM 2 D N A c a n also b e s e e n , it w a s u s e d as a c a l i b r a t i o n s t a n d a r d .

the BamHI cleavage truncates the first promoter. A binding map of an EcoRI fragment from plasmid pBK8 which covers the region of the BamHI insertion site is shown in Fig. 6. Strong RNA polymerase binding is observed near the end of the fragment corresponding to binding at the tet promoter site of the pBR322 vector DNA [27]. A second RNA polymerase binding site is found at a position at which the second promoter site should be located according to the sequence analysis [7]. Practically no binding is observed at the position of the first promoter, in agreement with the expectation that this site was destroyed by the BamHI cleavage. The experiments described above were carried out at a low polymerase/DNA ratio (2--3-fold molar excess). In the earlier transcription studies which led to the conclusion that 4--5 binding sites should exist per rRNA gene [2] a severaltimes higher polymerase/DNA ratio had been used in order to reach saturating levels. At such high ratios, however, the RNA polymerase-DNA complexes could not be measured in the electron microscope since the DNA was aggregated and could not be traced. However, a doubling of the RNA polymerase/ DNA ratio (6 : 1) was already sufficient to increase the multiple binding of polymerase to the DNA significantly. Examples of RNA polymerase bound to fragments of pBK17 (which contains the rrnB operon. [24]) at higher RNA polymerase concentrations are shown in Fig. 7. An analysis of the transcription complexes was carried out on the total BarnHI insert of plasmid pBK17. The

444 II

....

-35 *

. . . .

-25 i

j

-20

j

-15

-1.0

-0.5

0

+0.5

kb

~r

~,

J _l S,

S2

m

16 S

T

rRNA

T

m

g.

"

-8

-6

-4

-2

2: E

0

+2

+4

¢

+6 kb

L L I

~

J.

,,r

I

,L .I ,, k..,I "r,

?

I

-Lll

I

I

!

Is,,,,,I J" I

1

Fig. 8. R N A - p o l y m e r a s e b i n d i n g a n d t r a n s c r i p t i o n i n i t i a t i o n o n D N A f r a g m e n t s i s o l a t e d f r o m p l a s m i d p K B 17 c a r r y i n g t h e r r n B g e n e . a. p o l y m e r a s e b i n d i n g o n t h e 1 . 3 3 k i l o b a s e PstI-HindIII f r a g m e n t o f p B K 17. b. P o l y m e r a s e b i n d i n g o n t h e 4 . 0 5 k i l o b a s e B g l n I f r a g m e n t o f p B K 17. T h e h i s t o g r a m s u m m a r i z e s m e a s u r e m e n t s o n a large n u m b e r o f p i c t u r e s f r o m t h e e x p e r i m e n t s i l l u s t r a t e d w i t h Fig. 7a, b. c. I n v i t r o t r a n s c r i p t i o n i n i t i a t i o n sites o n t h e 1 5 k i l o b a s e B a m H I f r a g m e n t f r o m p l a s m i d p B K 17. E x p e r i m e n t a l c o n d i t i o n s w e r e t h e s a m e as in Fig. 1. R i g h t w a r d t r a n s c r i p t s a b o v e , l e f t w a r d b e l o w t h e b a r .

major R N A initiation is indicated b y the peak visible in Fig. 8c. A BglII fragment comprising the central region of the insert was used to construct an R N A polymerase binding map. As shown in Fig. 8b, most of the binding is observed in the center of this fragment. The position of this site corresponds to the strong binding site seen in Fig. 4a which shows the map of the analogous region derived from krifdl8. Binding in the r R N A p r o m o t e r region is comparatively low. A more stringent selection of DNA molecules carrying polymerases in this region was mado b y measuring complexes of R N A polymerase with a subfrag-

445 ment (PstI/HindIII) which does not contain the strong binding site next to the PstI cleavage site. The map shown in Fig. 8a now clearly shows the strongest binding in a position expected for the $2 site (as indicated in the graph), less binding at a site corresponding to the $1 site, and in addition four weak binding sites ahead of these two promoter positions. The terminal binding on both fragments is probably nonspecific and might be due to the presence of short singlestranded ends. Discussion

By the electron microscopic and filter-binding studies reported here, we hoped to complement the available DNA sequence and transcription information on the structure and function of bacterial rRNA promoters. The results indeed correlate well with other types of data and help to formulate a model of rRNA-promoter function free of contradictions. It is generally agreed that in functional terms the rRNA promoters are perhaps the strongest of all prokaryotic promoters. It is much less clear, however, what 'strong promoter' means in molecular terms. Promoter strength may be quantitatively correlated with: 1. Rate of formation of polymerase-DNA complexes. 2. Stability (rate of dissociation) of polymerase-DNA complexes. 3. Rate of transition from 'closed' to 'open' complexes. 4. Number of polymerase binding and initiation sites. The first of these factors was emphasized by the results of Seeburg et al. [28], who found good correlation between the functional strength of phage fd promoters and the rate of complex formation. In contrast, experiments from this laboratory [29] could not demonstrate any difference among the promoters of phage krifdl8 with respect to rate of complex formation, although as shown in Fig. 3, transcriptionally the rRNA promoters are far more active than any of the lambda promoters, whereas the filter-binding ability of these latter promoters was higher. Stability is also an insufficient explanation. In agreement with Glaser and Cashel [10] we found that the ribosomal promoters are quite unstable, even 0.1 M salt concentration prevents their formation (or causes the dissociation of preformed complexes) (Fig. 5). The rate of transition from the 'closed' to the 'open' complex as a determinant factor of promoter strength is emphasized in the generalized model of Chamberlin [30]. Although direct evidence for the existence of 'closed' complexes in the case of rRNA promoters is lacking, recent results of Hamming et al. [31] suggest that such a transition indeed exists. The role of this transition in the activity of rRNA promoters remains to be elucidated. Several lines of evidence -- including the results described here -- suggest the importance of the multiplicity of promoters as an important factor of functional 'strength'. The existence of two complete promoters per rRNA operon is proven by direct sequencing, analysis of in vitro and in vivo transcripts and in the present study by locating electron-microscopically these polymerase binding and transcription initiation sites. It is interesting to note that whilst binding is much stronger at the second start site ($2) than at the first (S~), chain initia-

446

tion occurs more frequently at the first site both in vitro [ 10] and in vivo [ 34]. T w o sites, however, are n o t enough to explain the extremely high functional strength of the r R N A p r o m o t e r and those earlier results which suggested the existence of 4--5 binding sites per r R N A gene. In the discussion of this problem it must be clarified that binding and initiation might be spatially distinct events and therefore the number of binding sites do n o t necessarily have to agree with the number of initiation sites. It is conceivable that the transition from the closed to the open complex involves the translocation of the polymerase along the DNA chain. The functional separation of binding from initiation had been demonstrated in the case of the tet p r o m o t e r of the plasmid p B R 3 2 2 [32]. We propose that the four distinct binding sites observed on Fig. 8b which precede the t w o initiation sites are functional parts of the r R N A gene and are responsible for its high frequency of transcription initiation. Although the evidence refers only to the rrnB gene it may be generally true for the other genes as well. As these sites do not initiate they may have remained undetected by other methods. As transcription of the region encompassing these sites was not observed it is very unlikely that they form part of any other gene. The very strong but nonproductive binding site near the PstI cleavage site merits some discussion. It m a y or may not belong to the r R N A gene. In any case it might explain those earlier findings [33] that randomly sheared DNA fragments were enriched in r R N A genes b y selecting for salt-resistant polymerase binding. As the real r R N A promoters are salt-lab•, as shown in Fig. 5, these nearby binding sites might have been responsible for these results. If the strong binding site would be a functional part of the r R N A gene then the presence of such sites would be expected in similar locations in the other r R N A opersons. Preliminary observations suggest that this might indeed be the case for the rrnX gene located on phage kdilv5. References 1 Travers, A. ( 1 9 7 5 ) i n Symposia of the 9 t h F E B S M e e t i n g (Siimegi, J., Venetianer, P. and Chambon, P., eds.), pp. 373--376, North-Holland, A m s t e r d a m 2 Siimegi, J., Udvardy, A. and Venetianer, P. (1977) Mol. Gen. Genet. 151, 305--312 3 Mueller, K., Oebbecke, C. and FSrster, G. (1977) Cell 10, 121--130 4 Kiss, I., Slaska, K., Siimegi, J., Udvardy, A. and Venetianer, P. (1978) Biochim. Biophys. Acta 518, 257--266 5 0 o s t r a , B.A., van Ooyen, A.J.J. and Gruber, M. (1977) Molec. Gen. Genet. 152, 1--6 6 deBoer, H.A., Gilbert, S.F. and Nomura, M. (1979) Cell 17, 201--209 7 Young, R.A. and Steitz, J.A. (1979) Cell 17, 225--234 8 Csordas-T6th, I~., Boros, I. and Venetianer, P. (1979) Nucleic Acids Res. 7, 2189--2197 9 Gilbert, S.F., deBoer, H.A. and N o m u r a , M. (1979) CeB 17, 211--224 10 Glaser, G. and Cashel, M. (1979) Cell 16, 111--121 11 Burgess, R.R. and Jendrisak, J.J. (1975) Biochemistry 14, 4634--4638 12 Roberts, R.J. (1978) Gene 4, 183--193 13 Miller, J.H. (1972) E x p e r i m e n t s in Molecular Genetics, Cold Spring H a r b o r L a b o r a t o r i e s 14 Udvardy, A., Le,udvay, K., SIlmegi, J. and Venetianer, P. (1979) Acta Biophys. Biochim. Acad. Sci. Hung. 14, 143--146 15 Roller, B., DeHus, H., Bflnemann, H. and Mitller, W. (1978) Gene 4, 227--239 16 Portmann, R. and Koller, T. (1976) 6 t h E u r o p e a n Congr. E l e c t r o n M i c r o s c o p y Jerusalem, Vol. 2., pp. 546--548, Tal I n t e r n a t i o n a l Publishing Co. 17 Sttiber, D., Delius, H. and Bujard, H. (1978) Mol. Gen. Genet. 166, 141--149 18 Chow, L.T., Roberts, J.M., Lewis, J.B. and Broker, T.R. (1977) Cell 1 1 , 8 1 9 - - 8 3 6 19 Davis, R.W., Simon, M. and Davidson, N. (1971) in Methods, K., eds.), pp. 413--428, Academic Press, New Yo rk 20 Sttiber, D. and Bujard, H. (1977) Mol. Gen. Genet. 154, 299--303

447 21 22 23 24 25 26 27 28 29 30 31 32 33 34

Seebuxg, P. and Schaller, H. (1975) J. Mol. Biol. 52, 261--277 Kirschbaum, J.B. and Konrad, E.B. (1973) J. Bacteriol. 116, 517--526 Kiss, A., Sain, B., Kiss, I., Boros, I., Udvardy, A. and Venetianer0 P. (1978) Gone 4, 137--152 Boros, I., Kiss, A. and Venetianer, P. (1979) Nucleic Acids. Res. 6, 1817--1830 Delius, H., Westphal, H. and Axelrod, N. (1973) J. Mol. Biol. 74, 677--687 Koller, Th., Sogo, M. and Bujard, H. (1974) Biopolymers 13,995--1009 Sutcliffe, J.G. (1978) Nucleic Acids Res. 5, 2721--2728 Seeburg, P., Ntlsslein, Ch. and Schaller, H. (1977) Eur. J. Biochem. 74, 107--113 Slaska, K., Sain, B., Udvardy, A. and Venetianer0 P. (1978) Acta Biochem. Biophys. Acad. Sci. Hung. 13,127--131 Chamberiin, M.J. (1976) in RNA polymerase (Losick, R. and Chamberlin, M.J., eds.), pp. 17--67, Cold Spring Harbor Laboratories Hamming, J., Gruber, M. and Geert, AB. (1979) Nucleic Acids Res. 7, 1019--1033 Rodriquez, R.L., West, R.W. and Heyneker, H.L. (1979) Nucleic Acids Res. 6, 3267--3287 Udvardy, A., Silmegi, J. and Venetianer, P. (1974) Nature 249, 548--550 deBoer, H. and Nomura, M. (1979) J. Biol. Chem. 254, 5609--5612