Download Evidence for the Lac Repressor-Operator Complex as a Transcription Terminator and more Study Guides, Projects, Research Biochemistry in PDF only on Docsity!
Proc. Natl. Acad. Sci. USA Vol. 83, pp. 4134-4137, June 1986 Biochemistry
lac repressor blocks transcribing RNA polymerase
and terminates transcription
(transcription regulation/repressor-operator function)
ULRICH DEUSCHLE, REINER GENTZ, AND HERMANN BUJARDt
Central Research Units, F. Hoffmann-La Roche and Co. AG, CH-4002 Basel, Switzerland
Communicated by Werner Arber, February 5, 1986
ABSTRACT Operator sequences are essential elements in many negatively controlled operons. By binding repressors, they prevent the formation of active complexes between RNA
polymerase and promoters. Here we show that the Escherichia
coli lac operator-repressor complex also efficiently interrupts
ongoing transcription.^ This observation^ suggests^ a^ mechanism of action for operators located distal to promoter sequences.
It is generally believed that repressors of prokaryotic operons
act exclusively by preventing the onset of transcription. This
view is supported by a wealth of experimental data, of which
the most convincing are the structural analyses of regulatory
regions: operators are located within the DNA sequence
covered by a promoter-bound RNA polymerase (i.e., be-
tween positions +20 and -50, where +^1 is the first nucleotide
transcribed) (1-3) or, as in the case of promoter P1 of the
Escherichia coli gal operon, within the cAMP-CAP
(catabolite activator protein) binding sites (4). Thus, by
occupying an operator, a repressor may either obscure a
promoter sequence from being recognized by RNA polymer-
ase or prevent the formation of an active complex between
the enzyme and the promoter (5-7). However, within the lac
operon, as well as within the gal operon, additional operator
sequences were identified well downstream of the regulatory
region (8, 9), and although an in vivo function for such an
operator was^ demonstrated in^ the^ gal^ system^ (9),^ its^ mode^ of
action has not been elucidated. The most straightforward
mechanism, the direct interference of an operator-repressor
complex with the^ transcribing enzyme,^ is^ generally^ ruled^ out
(7, 10) despite suggestive genetic and^ biochemical^ data
(11-13). Here^ we^ present^ evidence^ that^ the lac^ repres-
sor-operator complex is indeed an efficient terminator of
transcription in vivo and in vitro, suggesting an obvious mode
of action for operator sequences found, for example, within
structural genes of operons.
We had observed that, when transformed with^ plasmid
pGBU207, E. coli cells showed differences in tetracycline
resistance depending upon the internal level of lac repressor.
In pGBU207 (14), a lac operator sequence is located between
promoter PH207 and the coding sequence of the tet region;
therefore, we^ analyzed^ the^ effect of^ an^ isolated^ lac^ operator
sequence inserted into a transcriptional unit distal to the
promoter.
MATERIALS AND METHODS
Plasmids and Bacteria. The pDS1 vector system and the
promoters P025 and^ PD/E20 have^ been described (15, 16). The
lac operator was obtained as a 54-base-pair (bp) Hpa II-Alu
I fragment from pBU10 (14). Plasmid pDM1.1, which carries
the lacIq gene and the pi5A replicon, was a gift of M. Lanzer
(ZMBH, Univ. of Heidelberg). All plasmids and in vivo
RNAs were prepared from transformed E. coli DZ 291 (14).
In Vitro Transcripts. In vitro transcription was carried out
under standard conditions (14, 16), whereby a 50-,ul assay
mixture contained 0.2 pmol of template (construct A, carry-
ing promoter PG25 or PD/E20 in plasmid pDS1; Fig. 1), 1 pmol
of E. coli RNA polymerase, and [a-32P]UTP whenever
labeling of the transcription products was required. The
reaction mixtures were incubated at 370C in the absence or
presence of lac repressor (gift of M. Lanzer). Repressor was
inactivated by addition of isopropyl 3-D-thiogalactoside
(IPTG) to a^ final^ concentration of^200 4M. In^ general,
incubation was for 3 min before samples were directly
prepared for PAGE.
In Vivo Transcripts. E. coli cells transformed with the
proper plasmid were grown to an OD60 of 0.5 in^ M9^ medium
containing 10%^ Luria^ broth^ (17). Labeled^ RNA^ was^ obtained
by adding 500 uCi (1 Ci =^37 GBq) of [3H]uridine to 10 ml of
the logarithmically growing culture. After 1 min at 370C, cells
were quickly chilled in liquid nitrogen and RNA was isolated
according to^ Glisin et^ al.^ (18).^ High^ intracellular^ levels^ of lac
repressor were achieved by the simultaneous presence of the
compatible plasmid pDM1.1. Repressor was^ inactivated^ by
addition of IPTG (200 ug/ml) to the cultures 60 min before
harvest.
Nuclease S1 Mapping (19). A suitable DNA fragment for the
characterization of the 3' ends of in vivo^ and^ in^ vitro
transcripts was obtained by cleaving construct A (Fig. 1) with
Acc I and Pvu II. The Acc I cleavage site located 147 bp
upstream of the operator sequence was filled in with [a-
32P]dATP, resulting in a 3'-labeled 318-bp fragment covering
the entire operator sequence. About 0.01 pmol of the labeled
DNA fragment was denatured and mixed with one-fourth of
an in vitro transcription assay mixture or with 10 ,g of total
cellular RNA. The nucleic acids were allowed to hybridize
(volume (^30) I.d, 80% formamide/0.4 M NaCl/40 mM Tris/HCl, pH 8) for 2 hr^ before (^300) p1 of S1 buffer (19) containing 20
units of nuclease S1 were added. After 2 hr at i4WC, the
Si-resistant material was analyzed by electrophoresis in 8%
polyacrylamide/8 M urea gels.
Quantitation of^ in^ Vivo^ RNA^ by^ Hybridization.^ RNA^ was
labeled with [3H]uridine and isolated as described above.
Dihydrofolate reductase (DHFR)- and chloramphenicol ace-
tyltransferase (CAT)-specific transcripts were quantified by
hybridization with an excess of single-stranded M13 DNA
carrying the^ proper DHFR^ ahd CAT gene sequences, respec-
tively. The hybridized material was collected by filtration
Abbreviations: bp, base pair(s); DHFR, dihydrofolate reductase;
CAT, chloramphenicol acetyltransferase; IPTG, isopropyl P-D-
thiogalactoside. *Present address: Zentrum fMr Molekulare Biologie, Heidelberg, Im Neuenheimer Feld^ 282, D-69 Heidelberg, Federal Republic of Germany. tTo whom all correspondence should be addressed.
4134
The publication costs of this article were defrayed in part by page charge payment. This article^ must^ therefore^ be^ hereby marked "advertisement" in accordance with 18 U.S.C. (^) §1734 solely to indicate this fact.
Proc. Natl. Acad. Sci. USA 83 (1986) 4135
P DHFR {1- B
A ---
CAT to
H 0
B { * 10 0
FIG. 1. Transcriptional unit used for operator insertion. The standard transcription unit of the pDS1 vector system (15) contains the coding sequence of the dihydrofolate reductase (DHFR) and the E. coli chloramphenicol acetyltransferase (CAT) genes, both of which can be brought under control of a single promoter (P). Transcripts of defined size are obtained by the function of terminator to of phage X, which also prevents extensive read-through into other parts of the plasmid (15). Expression of this unit in vivo under the
control of promoter PG2S or PD/E20 (16) yields exclusively CAT
protein, since only this sequence carries a functional translational
start signal (i). The lac operator (0) sequence (positions -17 to +34,
ref. 7) was fused to either HindIll or BamHI synthetic linkers and inserted into the HindIll (H) or BamHI (B) site, resulting in constructs A and (^) B, respectively. The distances between the pro- moter (P) and sites (^) B, H, and (^) to are about 100, 670, and 1700 bp, respectively, depending somewhat upon the position of the promoter within the cloned fragment.
through nitrocellulose and its radioactivity was monitored.
This method has been described previously (20, 21).
RESULTS
lac Repressor-Operator Complex Functions as a
Regulatable Terminator. The transcriptional unit used in
these experiments has been described earlier as part of the
pDS1 vector system (15). In these vectors, the coding
sequence of DHFR and CAT genes are under the control of
a single promoter, giving transcripts of "1700 nucleotides
due to the terminator to at the end of the CAT gene (Fig. 1).
The transcription units analyzed here were controlled by
either one of two promoters of coliphage T5 [PG25 or PD/E
(16)], and a lac operator sequence was inserted either
between the DHFR and the CAT sequence (construct A) or
into the BamHI site near the promoter (construct B in Fig. 1).
Since the DHFR sequence is not in-frame with any transla-
tional start site, the only protein expected from this expres-
sion unit is CAT.
With construct A (in pDS1), no CAT synthesis is observed
in E. coli cells containing high levels of lac repressor (Fig. 2).
However, CAT production is rapidly induced to a high level
by IPTG, as^ expected with,^ these^ promoters (16). This
experiment shows^ that^ the^ operator-bound lac^ repressor can
efficiently interfere with^ ongoing transcription. It^ raises^ the
question whether^ repressor merely blocks^ the^ transcribing
enzyme or causes^ a true^ termination^ event.^ Analysis of^ in
vitro and in vivo transcripts shows^ that^ the^ lac^ repres-
sor-operator complex acts as^ a^ transcription terminator.^ In
the absence of repressor, or^ in^ the^ presence of^ repressor and
IPTG, transcripts of^ around^1700 nucleotides^ are^ the^ major
products in vitro (Fig. 3A, lanes 1, 4, and 5). In contrast, when
active repressor is included in the transcription assay, the
vast majority of transcripts are terminated at three distinct
positions (a, b, and c in Fig. 3A), yielding RNAs^ about^750
nucleotides long. Of^ these, only the^ smallest^ species can^ be
converted into larger products (lanes 6 and 7). The others are
not affected by prolonged incubation with unlabeled
nucleoside triphosphates. When IPTG is added together with
unlabeled nucleoside triphosphates, no increase in radioac-
tivity is found in^ the^ 1700-nucleotide^ RNA^ species (data not
shown). These results^ show that^ the^ repressor does^ not
simply induce^ transcribing RNA^ polymerase to^ pause but
rather triggers an active process of termination. The results
(^9 )
31-
~1mEI - CAT
2 1-
...ia:?..,C3g J (^) Nov:-f_,; ,
M 0 0.5 1 2 3
Time of (^) induction, hr
FIG. 2. Effect of operator insertion on CAT synthesis in vivo. A
pDS1 plasmid carrying construct^ A^ with^ PD/E20 as^ promoter was^ used
to transform into E. coli cells carrying the compatible plasmid
pDM1.1.^ The^ latter^ plasmid contains the lacIq^ gene and^ provides high
intracellular levels of lac (^) repressor. Cultures of the transformed cells were (^) grown to (^) OD~w 0.7 before 1PTG (200 (^) Atg/ml) was added.
Aliquots of^ the culture^ were^ removed^ at^ times^ indicated^ and^ the
pattern of the total cellular^ protein was^ monitored^ by NaDodSO4/
PAGE. The Coomassie blue-stained gel shows that the CAT protein,
not visible at the time of IPTG addition, is the most prominent
product after^ only 30 min.^ The size markers^ (lane M) are^ given in^ kDa
at left.
of equivalent experiments carried out in vivo are shown in
Fig. 3B.^ Again, in the absence^ of^ operator or^ in^ the^ presence
of IPTG, the major plasmid-speciflied RNA is about 1700
nucleotides long (lanes 1 and 3). In cells containing high levels
of repressor, however, two short transcripts of about 750
nucleotides are synthesized (lane 2). Lanes 3 and 4 of Fig. 3B
show an additional RNA species of about 820 nucleotides
(labeled x). This transcript is only observed in the presence
of the operator-carrying fragment and when transcription is
allowed to proceed past the operator either by addition of
IPTG (lane 3) or by limiting amounts of intracellular repressor
(lane 4), suggesting that an additional sequence acting as a
terminator in vivo must be^ locatqd downstream^ of^ the^ 60-bp
operator fragment.
By quantifying DHFR- and^ CAT-specific RNA^ (refs. 20
and 21; unpublished work) we^ find as^ much^ As 90% termi-
nation in vivo (Table 1). This termination can be completely
reversed by IPTG. Our^ data also^ indicate^ that^ the^ repressor-
independent termination at^ position x^ (Fig. 3B) is^ "'17%
efficient (data not shown).
Topography of^ the lac^ Repressor-Operator Termination
Signal. Where does^ an^ operator-bound^ repressor^ force the
transcribing RNA^ polymerase to^ stop and^ to^ release the
nascent transcript? To answer this^ questions, we^ used^ con-
struct A (Fig. 1), containing promoter P625, to produce
transcripts in^ the^ presence or^ absence^ of^ lac^ repressor, and
the 3' ends of the RNAs terminated around the operator
sequence were^ characterized^ by nuclease^ Si-mapping with
3'-labeled DNA fragments (Fig. ,3C). When repressor is
bound to the operator, transcription is^ terminated^ in^ vivo^ and
in vitro at two sites upstream of the operator sequence (Fig.
4). The^ termination^ site observed^ in'vivo^ when^ repressor^ is
limiting or inactive has been mapped outside of but adjacent
to the cloned operator fragment. In^ Fig. 4, the^ different^ sites
are indicated by hatched columns. It^ appears- most^ likely. to us that the (^) repressor terminates (^) transcription at (^) precise positions and^ that^ the^ regions^ of^ 3-5^ nucleotides^ derived^ from S1-mapping experiments^ primarily^ reflect^ a^ heterogeneity^ of the (^) S1 digest. The (^) homogeneous transcript obtained^ in^ vitro
Biochemistry: Deuschle^ et^ al.
Proc. Natl. Acad. Sci. USA 83 (1986) 4137
-30 - 20 - 10 0 +10 +20 + 30 +40 + 50
1 _^1
CC G GCCAAGCTTGGCGGC TCGTATG TTGTGTGGAA TTG~TGAG CGGATAACAATTTCACACAGGAAACAGC CAAGC TTGGC GAGATTTTCAGGAGCTAAG GA lI
b a x
FIG. 4. Sequences involved in repressor-induced transcriptional termination.^ The^ central region^ of the operator sufficient to bind^ repressor (1, 22) is boxed, and the inverted repeat of the sequence is delineated by^ arrows.^ The^ G^ in the^ center^ of^ the operator sequence^ has^ been^ used to define position 0. The sites where transcription is terminated are indicated by the^ hatched columns. The^ width of the^ columns reflects^ the heterogeneity of the Si-resistant material and the height of the columns represents the relative^ frequency^ of^ termination^ at^ the respective^ site. The columns above and below the sequence describe the in vitro and in vivo results, respectively. Sites a, b, and x^ correspond^ to the^ designations used in Fig. 3. In the presence of repressor, transcripts are terminated upstream of the operator^ sequence. Termination^ at site^ x occurs^ outside of the original operator fragment. The HindIII cleavage sites used to insert the 54-bp fragment between the DHFR and the CAT sequence^ yielding construct A (Fig. 1) are underlined.
can only transiently block a transcriptional elongation com- plex. By contrast, our data demonstrate that a lac repres- sor-operator complex located distal to a promoter sequence can directly interfere with gene expression by efficiently terminating transcription. This sheds new light on the pos- sible role ofoperators found^ outside^ ofthe primary^ regulatory region. Thus, operator/repressor systems^ could^ have func- tions in addition to the^ one^ commonly^ considered-namely, (i) to prevent readthrough from upstream regions^ into^ the repressed operon and (ii) to establish a polarity pattern within an operon that is dependent on the level of inducer and the affinity between a particular operator sequence and a repres- sor. These properties could play a role in the fine tuning of^ gene expression at the transcriptional level and may be considered as a type of attenuation. Systems to examine^ this^ hypothesis could be the gal as well as the lac operon (8, 23, 24). Both operons contain a second operator sequence about 50 and 400 bp downstream of the RNA initiation site, respectively. In the gal operon, operator 2, which is located within the structural gene galE, may affect transcription from both promoters P1 and P2, but primarily from P2 by attenuation, whereas (^) operator 1 is the main control element ofP1. Finally, Sellitti and Steege (25) reported that transcription from the lacd promoter is^ "punctuated" within^ the^ lac^ control^ region and that this punctuation is strongly influenced by the lac repressor. These data^ suggest that the^ mechanism^ of^ action for repressor/operator systems proposed here is utilized in the E.^ coli^ lac^ system.
We thank M.^ Lanzer^ for^ supplying^ purified^ lac^ repressor^ and pDM1.1; W. Kammerer, M. Lanzer, and D. Stueber for helpful discussions; and J.^ Scaife^ and^ S.^ Le Grice^ for^ critical^ comments^ and reading of the manuscript. The help of Y. Kohlbrenner in preparing the manuscript is gratefully acknowledged.
- Gilbert, W. & Maxam, A. (1973) Proc. Natl. Acad. Sci. USA 70, 3581-3584.
- Gunsalus, R. P. & Yanofsky, C. (1980) Proc. NatI.^ Acad. Sci. USA 77, 7117-7121.
- (^) Adhya, S. & (^) Miller, W. (^) (1979) Nature (London) 279, 492-608.
- Shanblatt, S. H. & Revzin, A. (1983) Proc. Natl. Acad. Sci. USA 80, 1594-1598.
- Reznikoff, W. S. (1972) in RNA Polymerase, eds. Losick, R. & Chamberlin, M. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), pp. 441-454.
- Majors, J. (1975) Proc. Natl. Acad. Sci. USA 72, 4394-4399.
- Reznikoff, W. S. & Abelson, J. M. (1980) in The Operon, eds. Miller, J. H. & Reznikoff, W. S. (Cold Spring Harbor Labo- ratory, Cold Spring Harbor,^ NY),^ pp.^ 221-243.
- Gilbert, W., Gralla, J., Majors,^ J. &^ Maxam,^ A.^ (1975)^ in Protein-Ligand Interactions,^ eds.^ Sund,^ H.^ &^ Blauer,^ G.^ (de Gruyter, Berlin),^ pp.^ 193-210.
- Irani, M.^ H., Orosz,^ L.^ & Adhya,^ S.^ (1983)^ Cell^ 32,^ 783-788.
- Mitchell, D. H., Reznikoff, W. S. & Beckwith, J. (1975) J. Mol. Biol. 93, 331-350.
- (^) Reznikoff, W. S., Miller, J. H., Scaife, J. G. & Beckwith, J. R.^ (1969) J. Mol.^ Biol.^ 43, 201-213.
- Horowitz, H.^ & Platt, T.^ (1982)^ Nucleic^ Acids^ Res.^ 10, 5447-5465.
- Herrin, G.^ L. & Bennet,^ G.^ N.^ (1984)^ Gene^ 32,^ 349-356.
- Gentz, R., Langner, A., Chang,^ A.^ C.^ Y.,^ Cohen,^ S.^ N.^ & Bujard, H.^ (1981) Proc.^ Natl.^ Acad.^ Sci. USA^ 78,^ 4936-4940.
- Stueber, D.^ & Bujard, H.^ (1982)^ EMBO^ J.^ 1,^ 1399-1404.
- Gentz, R. & Bujard, H. (1985) J. Bacteriol. 164, 70-77.
- Miller, J. H. (1972) Experiments in Molecular Genetics (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), pp. 431-435.
- Glisin, V., Crkvenjakov, R. & Byus, G.^ (1974) Biochemistry 13, 2633-2637.
- Maniatis, T., Fritsch, E. F.^ & Sambrook, J.^ (1982) Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Labora- tory, Cold Spring Harbor, NY).
- Deuschle, U.^ (1986) Dissertation^ (Univ.^ of^ Heidelberg, Heidelberg).
- (^) Bujard, H., Deuschle, U., Kammerer, W., Gentz, R., Bann- warth, W.^ &^ Stueber,^ D.^ (1985)^ in^ Sequence^ Specificity^ in Transcription and^ Translation:^ UCLA^ Symposia,^ eds. Calen- dar, R.^ &^ Gold, L.^ (Liss, New^ York),^ pp.^ 21-29.
- von Hippel, P. H. (1979) in Biological Regulation and Devel- opment, ed.^ Goldberger, R. E.^ (Plenum, New^ York),^ Vol.^ 1, pp. 279-347.
- De (^) Crombrugghe, B., Busby, S. & Buc, H. (1984) Science 223, 831-838.
- Majumdar, A. & Adhya, S. (1984) Proc. Natl. Acad. Sci. USA 81, 6100-6104.
25. Selfitti, M. A. & Steege, D. A.^ (1985) J.^ Cell.^ Biochem.^ 85, 205.
Biochemistry: Deuschle^ et^ al.
