Molecular Microbiology (2002) 46(5), 1247–1257

Governor of the glnAp2 promoter of Escherichia coli
Mariette R. Atkinson, Narinporn Pattaramanon and
Alexander J. Ninfa*
Department of Biological Chemistry, University of
Michigan Medical School, Ann Arbor, MI 48109-0606,
USA.
Summary
~P (NtrC~
~P) that
Low-affinity sites for the activator NRI~
map between the enhancer and the glnAp2 promoter
were responsible for limiting promoter activity at high
~P in intact cells and in an in
concentrations of NRI~
vitro transcription system consisting of purified bacterial components. That is, the low-affinity sites constitute a ‘governor’, limiting the maximum promoter
activity. As the governor sites are themselves far from
the promoter, they apparently act either by preventing
the formation of the activation DNA loop that brings
the enhancer-bound activator and the promoterbound polymerase into proximity or by preventing a
productive interaction between the enhancer-bound
activator and polymerase. The combination of potent
enhancer and governor sites at the glnAp2 promoter
provides for efficient activation of the promoter when
the activator concentration is low, while limiting the
maximum level of promoter activity when the activator
concentration is high.
Introduction
The s54-dependent promoters of bacteria are unique in
their requirement for distant activator sites that function as
enhancer elements (Reitzer and Magasanik, 1986; Ninfa
et al., 1987; reviewed by Kustu et al., 1989; Atkinson and
Ninfa, 1994). These enhancer elements bind the activator,
which stimulates the rate of open complex formation at
the promoter (Popham et al., 1989). The enhancer-bound
activator interacts with the promoter-bound polymerase by
means of a DNA loop that brings the activator and polymerase into proximity (Su et al., 1990). In the absence of
activator, polymerase binds the promoter and forms a
stable closed complex in which the DNA strands are not
melted (Ninfa et al., 1987; Sasse-Dwight and Gralla,
Blackwell Science, LtdOxford, UKMMIM; olecular Microbiology0950-382XBlackwell Science, 200246Original ArticleGovernor of glnAp2
M. R. Atkinson, N. Pattaramanon and A. J. Ninfa

Accepted 10 September, 2002. *For correspondence. E-mail
aninfa@umich.edu; Tel. (+1) 734 763 8065; Fax (+1) 734 763 4581.

© 2002 Blackwell Publishing Ltd

1988; Popham et al., 1989). Interaction of the activator
with the polymerase in the closed complex results in the
isomerization of the closed complex to the transcriptionally active open complex (Popham et al., 1989). After formation of the open complex, activator appears to have no
additional role in transcriptional activation, such as activation of promoter clearance or transcription elongation
(Ninfa et al., 1989). Recent studies suggest that reinitiation follows the same kinetic pathway as the first initiation
event, that is no subassembly of the transcriptional apparatus remains at the promoter after initiation (Bondarenko
et al., 2002).
Several lines of evidence suggest that the role of the
enhancer element is to increase the local concentration
of the activator in the vicinity of the closed complex of
polymerase and promoter. Studies of the nitrogenregulated glnAp2 promoter, which expresses the nitrogen
assimilation enzyme glutamine synthetase (GS), indicated that activation from this promoter was only dependent on the enhancer when the concentration of the
activator in the reaction mixtures was low; at high activator
concentrations, the enhancer was not required for promoter activity (Ninfa et al., 1987). DNA binding by the
activator is apparently not required for transcriptional activation, as a mutant form of the activator that is unable to
bind to DNA is able to initiate transcription in vitro when
provided in sufficient concentration (Porter et al., 1993).
Kustu and colleagues showed that enhancer-dependent
activation could occur when the enhancer and promoter
were located on separate, concatenated circular DNA
molecules (Wedel et al., 1990). This experiment suggests
that activator interacts directly with the polymerase in
the closed complex and appears to exclude ‘tracking’
mechanisms or mechanisms that require the activator to
send a signal through the DNA, such as by altering its
topography.
Nitrogen-regulated Ntr promoters of Escherichia coli
are activated by the phosphorylated form of the glnG
(ntrC) protein NRI~P (NtrC~P) (Ninfa and Magasanik,
1986), which binds to enhancer sequences found
upstream from the regulated promoters. At the glnAp2
promoter, there are two adjacent high-affinity NRI binding
sites (sites 1 and 2) centred at position -110 relative to
the site of transcription initiation (Hirschman et al., 1985)
that serve as the enhancer (Reitzer and Magasanik, 1986;
Ninfa et al., 1987). In experiments with purified components, these sites were able to stimulate transcriptional

1248 M. R. Atkinson, N. Pattaramanon and A. J. Ninfa
activation when moved to new locations far upstream or
downstream from the promoter (Ninfa et al., 1987). Other
nitrogen-regulated promoters also contain a pair of NRI
binding sites upstream from the promoter. In the case of
the nitrogen-regulated glnK and nac promoters, one of the
two sites is degenerate in sequence, and these enhancers
do not bind NRI as avidly as the glnAp2 enhancer (Feng
et al., 1995a; Atkinson et al., 2002). Phosphorylation of
NRI results in the oligomerization of NRI and highly cooperative binding to adjacent NRI binding sites (Weiss
et al., 1992). The oligomerized NRI~P bound to the
enhancer has high ATPase activity that is somehow
involved in melting the DNA strands to form the open
complex (Popham et al., 1989; Weiss et al., 1991).
The phosphorylation and dephosphorylation of NRI is
regulated by a complex signal transduction system in
response to changes in the intracellular nitrogen status
(reviewed by Ninfa et al., 2000). In addition, the intracellular concentration of NRI is regulated in part by activation
of the glnAp2 promoter of the glnALG operon (Pahel et al.,
1982; Reitzer and Magasanik, 1985; Atkinson et al.,
2002). The effect of these regulatory mechanisms is that
the intracellular concentration of NRI~P is increased in
nitrogen-starved cells. This amplitude modulation of the
NRI~P concentration results in the sequential activation
of Ntr promoters as cells become starved. That is, promoters with potent enhancers, such as the glnAp2 promoter and the glnHp2 promoter, are activated first,
followed by the activation of the glnK and nac promoters
only when the NRI~P concentration reaches the level
required for interaction with their enhancers. Studies with
purified components have indicated that activation of
glnHp2 required a slightly higher concentration of NRI~P
than did activation of glnAp2 (Carmona and Magasanik,
1996), and activation of nacp and glnKp required a much
higher concentration of NRI~P than did activation of
glnAp2 (Feng et al., 1995a; Atkinson et al., 2002). Also,
studies with intact cells have shown that glnAp2 is activated before glnKp and nacp as cells become nitrogen
starved (Atkinson et al., 2002). Recently, we examined the
expression profile of glnK, nac and glnA promoter fusions
to lacZYA in mutant cells unable to reduce the phosphorylation state of NRI in response to signals of nitrogen
status (Blauwkamp and Ninfa, 2002a). We were surprised
to observe that, in such cells, the expression of the glnKp–
lacZYA and nacp–lacZYA fusions proceeded unchecked
as the cells became nitrogen starved, whereas the
expression of a glnAp–lacZYA fusion did not. This result
indicated that a mechanism for limiting the expression of
the glnAp–lacZYA fusion was operating, but did not
address the nature of this mechanism.
Footprinting of the glnAp2 promoter has revealed that,
in addition to the high-affinity NRI binding sites that serve
as the enhancer, the glnA control region contains three

additional sites (sites 3, 4 and 5, centred at -90, -65 and
-45 bp from the transcription start site) that are occupied
only at very high concentrations of NRI~P (Hirschman
et al., 1985; Ninfa et al., 1987). Experiments with purified
components indicated that activation of glnAp2 was
reduced at very high NRI~P concentration (Feng et al.,
1995b). Also, experiments with intact cells indicated that
overexpression of NRI caused a reduction in transcription
from glnAp2 under nitrogen-limiting conditions. For example, hyperexpression of NRI from the multicopy plasmid
pgln53 resulted in reduced expression of the glnA gene
encoding glutamine synthetase (Chen et al., 1982). In
another study, hyperexpression of NRI from a multicopy
plasmid under the control of the lac promoter resulted in
reduced expression of glnA (Shiau et al., 1992). This
study used a glnA promoter reporter construct that
contained a 700 bp insertion between the enhancer and
promoter. It was observed that, in such a context, overexpression of NRI resulted in reduced glnA expression in
intact cells regardless of the presence or absence of sites
3, 4 and 5 (Shiau et al., 1992). These results led to the
conclusion that neither sites 3, 4 and 5 nor regulation of
the flexibility of the DNA were responsible for the reduction
in glnA expression at high activator concentration (Shiau
et al., 1992).
Here, we examined the role of the low-affinity NRI binding sites in limiting the activity of the glnA control region
when NRI was expressed under physiological conditions.
Specifically, we examined the effect of scrambling the
sequences of the low-affinity sites 3 and 4 (using the
nomenclature of Hirschman et al., 1985) on the glnAp2
expression profile. Our results indicated that, under physiological conditions, these sites act as a governor responsible for setting the upper boundary of expression from
the promoter when the intracellular NRI~P concentration
was high. However, under non-physiological conditions,
such as when NRI was hyperexpressed from a multicopy
plasmid, limitation of the glnAp2 promoter was only partially dependent on sites 3 and 4. Thus, under the latter
conditions, multiple mechanisms for restraining expression of the wild-type promoter were indicated. We also
examined the role of sites 3 and 4 in activation of the
glnAp2 promoter in an in vitro transcription system, and
found that these sites limited promoter activation at high
concentrations of NRI~P.

Results
Sequences downstream from the promoter were not
required to govern glnAp–lacZYA fusions
We described previously a fusion of the glnA promoter
region to promoterless lacZYA which, when located
in single copy within a chromosomal ‘landing pad’,
© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

Governor of glnAp2
expressed b-galactosidase co-ordinately with the expression of GS from the natural wild-type glnA gene [Table 1
(Apf2), Fig. 1]. Studies of this fusion showed that the glnA
promoter was expressed before similar fusions to the glnK
and nac promoters as cells became starved (Atkinson
et al., 2002) but that, unlike the glnKp and nacp fusions,
it did not display ‘runaway’ expression in starved cells
that cannot limit the extent of NRI phosphorylation
(Blauwkamp and Ninfa, 2002a).
The glnAp–lacZYA fusion described above contains a
portion of the glnA transcribed leader sequence and a
portion of the glnA structural gene, followed by a synthetic
stop codon and the lac leader mRNA and lac structural
genes. Thus, the properties of the fusion may not have
been entirely caused by the properties of the promoter.
We therefore constructed another glnAp–lacZYA fusion
(Apf), in which the first transcribed nucleotide from
glnAp2 is from the lac leader (Experimental procedures).
Expression of both fusions was observed to parallel the
expression of glutamine synthetase in various genetic
backgrounds (Table 1). More importantly, hyperexpression of NRI from pgln53 resulted in a reduction in the level
of expression of both fusions (Table 2). We interpret this
as indicating that regulation of the fusions by high NRI~P
did not require sequences downstream from the promoter.
Low-affinity NRI~P binding sites 3 and 4 of the glnA
control region were responsible for limiting promoter
expression under physiological conditions
To study the effects of sites 3 and 4, we engineered
promoters in which the sequence of each site was multiply
mutated to a sequence having no similarity to an NRI
binding site, separately and together (Fig. 1). DNase I
footprinting studies indicated that these mutations eliminated detectable binding of NRI~P to the altered sites 3

1249

and 4, even at very high concentrations of NRI~P (data
not shown). Specifically, we observed that, at 150 nM NRI,
in the presence of excess NRII and ATP to phosphorylate
the NRI, weak footprinting of sites 3 and 4 was observed
with a wild-type template and that, at 300 nM NRI (with
excess NRII and ATP), sites 3 and 4 were completely
footprinted when a wild-type template was examined.
Under the same conditions, the region containing scrambled sites 3 and 4 on a DNA template in which both sites
were altered showed no detectable footprinting when NRI
was at 150 nm, 300 nM or 600 nM along with excess NRII
and ATP (data not shown).
The mutant promoters were used to form fusions analogous to the largest glnAp–lacZYA fusion (i.e. Apf2) and
incorporated into the E. coli chromosome in single copy
within the trp operon. In each case, limitation of the promoters by hyperexpression of NRI from pgln53 was defective, indicating a role for sites 3 and 4 in this process
(Table 2). Indeed, the level of expression from these promoters was elevated relative to the wild type, even in the
absence of NRI hyperexpression. This suggested that
some governing of glnAp2 expression by sites 3 and 4
occurs in nitrogen-limited wild-type cells at physiological
levels of NRI~P. Reduction in glnAp2 expression upon
hyperexpression of NRI from pgln53 was not completely
eliminated by mutations to sites 3 and 4 in these experiments (Table 2).
A more physiological assay for regulation of the glnA
promoter region is to examine the expression profile of
fusions as cells deplete a limiting supply of ammonia and
become nitrogen starved. Previous studies showed that
governing of glnAp2 was most obvious in cells lacking the
GlnK signal transduction protein. These mutant cells are
unable to activate the dephosphorylation of NRI~P once
starved, and display unlimited expression of glnKp–lacZYA and nacp–lacZYA upon starvation (Blauwkamp and

Table 1. Glutamine synthetase and b-galactosidase expression in adapted cultures.
Glutamine synthetasea (n b)
Strain
e

YMC10Apf (WT)
BApf (DglnB2306)
KcApf (Dmdl-glnK::camr )
YMC10Apf2 (WT)
BApf2 (DglnB2306)
KcApf2 (Dmdl-glnK::camr )

b-Galactosidasec

Ggtrpd

GNgtrp

Ggtrp

GNgtrp

1690
1310
1620
1240
1630
1700

180
1350
210
193
1620
240

2330
2770
2080
2290
3780
3020

110
1820
200
170
2380
230

(4.1)
(3.1)
(3.4)
(3.8)
(3.7)
(3.6)

(8.2)
(11.3)
(9.4)
(8.9)
(11.4)
(9.5)

a. GS transferase activity. Cultures were grown overnight in the indicated medium, diluted to an OD600 of ª0.02 and grown at 30∞C to an OD600
of ª0.5.
b. Adenylylation state expressed as average number of adenylylated subunits per GS dodecamer.
c. b-galactosidase in Miller units. Cells were permeabilized using chloroform and SDS.
d. Media used were: Ggtrp, glucose–glutamine–tryptophan; GNgtrp, glucose–ammonia–glutamine–tryptophan. In all cases, tryptophan was
present at 0.004% (w/v), glucose 0.4% (w/v) and the nitrogen sources at 0.2% (w/v).
e. Strains designated Apf contain the glnA promoter fused to lacZ at the transcription +1. Strains designated Apf2 contain the glnA promoter
fused to lacZ at +165.
© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

1250 M. R. Atkinson, N. Pattaramanon and A. J. Ninfa
Fig. 1. Features of the wild-type and mutant
glnA promoters. The DNA sequence of the glnA
promoter fused to lacZ is shown including the
PCR-generated restriction sites. The NRI and
s54 binding sites are underlined. The s70 transcription start site is indicated with a small
arrow. The s54 transcription start site is indicated with a large arrow. Mutant DNA
sequences introduced to NRI binding sites 3
and 4 are below the wild-type sequence. The
glnA start codon and the stop codon introduced
by PCR are shown in bold type.

Ninfa, 2002a). However, glnAp–lacZYA expression was
limited in cells lacking GlnK as the cells became starved
(Blauwkamp and Ninfa, 2002a; Fig. 2A). In contrast,
fusions in which site 3, site 4 or both sites 3 and 4 had
been mutated displayed an obvious defect in limiting glnA
expression in starved cells lacking GlnK (Fig. 2A). Cells
in which both sites 3 and 4 were mutated seemed to be
more severely affected than cells lacking either site 3 or
Table 2. Effect of hyperexpression of NRI from a multicopy plasmid
on b-galactosidase expression from glnA promoter fusions.
b-galactosidasea
Strain

Fusion

NRI sites
present

YMC10 (WT)

glnApf
glnApf2
glnApf5
glnApf6
glnApf7

1,
1,
1,
1,
1,

2,
2,
2,
2,
2,

3,
3,
4,
3,
5

4, 5
4, 5
5
5

pBR322

pgln53

2330
2910
3620
3660
4480

430
540
1005
1410
2190

a. Miller units. Cells were permeabilized using chloroform and SDS.
Cultures were grown overnight in glucose–glutamine–tryptophan
medium, diluted to an OD600 of ª0.02 in the same medium and grown
at 30∞C to an OD600 of ª0.7. Results are the average of three
experiments.

4 alone (Fig. 2A). These results show that sites 3 and 4
play a role in limiting glnA expression under physiological
conditions. Interestingly, the presence or absence of sites
3 and 4 had no apparent effect on the level of fusion
expression during growth on ammonia or on the kinetics
of induction upon depletion of the limiting ammonia in cells
lacking GlnK (Fig. 2A).
We could also observe an effect of sites 3 and 4 on
limiting promoter expression in wild-type cells (Fig. 2B).
All the fusions displayed a similar level of glnA expression
during growth on ammonia, which was increased upon
exhaustion of the ammonia (Fig. 2B). After overnight starvation, the fusions with mutations in sites 3, 4 or both 3
and 4 were expressed at a higher level than the wild type
(Fig. 2B), but at a considerably lower level than in cells
lacking GlnK (Fig. 2A).
As the mutations to sites 3 and 4 resulted in an increase
in the expression of the lacZYA fusions, the possibility
existed that the mutations had created new promoters. To
examine this, we studied the effects of a mutation that
deletes glnL and glnG, eliminating NRII and NRI, on the
expression of the fusions (Experimental procedures).
Deletion of glnL and glnG reduced the expression of the
© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

Governor of glnAp2

1251

Fig. 2. Induction of glnAp2 and glnAp2* upon
shift to nitrogen-starved growth. Cells were
incubated 30∞C overnight in defined minimal Wsalts medium containing 0.4% (w/v) glucose,
0.2% (w/v) ammonium sulphate and 0.004%
(w/v) tryptophan. The cells were then washed
and diluted into fresh media as above, only
containing 0.005% (w/v) ammonium sulphate.
Samples were assayed for expression of the
reporter b-galactosidase using the Miller assay
on SDS–chloroform-permeabilized cells.
Solid symbols, OD600; open symbols, bgalactosidase.
A. KgApF2 [wild-type glnAp–lacZ fusion (filled
circles)], KgApF5 [NRI site 3 mutant glnAp–
lacZ fusion (filled squares)], KgApF6 [NRI site
4 mutant glnAp–lacZ fusion (filled diamonds)],
KgApF7 [NRI sites 3 and 4 mutant glnAp–lacZ
fusion (filled triangles)].
B. YMC10ApF2 [wild-type glnAp–lacZ fusion
(filled circles)], YMC10ApF5 [NRI site 3 mutant
glnAp–lacZ fusion (filled squares)],
YMC10ApF6 [NRI site 4 mutant glnAp–lacZ
fusion (filled diamonds)], YMC10ApF7 [NRI
sites 3 and 4 mutant glnAp–lacZ fusion (filled
triangles)].

wild-type fusion glnApf2 from 2910 Miller units of bgalactosidase to a basal level of 90 Miller units. Similarly,
the fusions glnApf5–glnApf7 displayed basal levels of
expression of 110, 100 and 120 Miller units, respectively,
in cells deleted for glnL and glnG. This result indicates
that, in all cases, fusion expression was dependent on
NRII and NRI, implying that it resulted from expression
from glnAp2. In another set of experiments, we examined
the effect of the glnD99::Tn10 mutation on reporter
expression. Previous results indicated that, under the conditions in our experiments, this mutation reduces the
expression of glutamine synthetase by about two-thirds
(Bueno et al., 1985). We observed that, for all the fusions
studied here, the glnD99::Tn10 mutation resulted in a
similar decrease in fusion expression (data not shown).
These experiments are also consistent with the idea that
expression of the fusions results in all cases from the
activity of the glnAp2 promoter.
© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

Sites 3 and 4 play a role in the reduction of transcription
from glnAp2 at high activator concentration in an in vitro
transcription system
To examine the effects of sites 3 and 4 in vitro, we
constructed supercoiled plasmid vectors based on the
plasmid pTE103 (Elliot and Geiduschek, 1984; Experimental procedures). These transcription templates contain a strong rho-independent transcriptional terminator
sequence derived from bacteriophage T7 positioned
downstream from the glnAp2 promoter, such that a transcript of 329 nucleotides (nt) is formed. Unexpectedly, we
observed that both the wild-type template, pglnApOG7,
and a template with mutations in both site 3 and site 4,
pglnApOG8, produced two glnAp2 transcript bands in our
in vitro system that seemed to result from initiation at sites
a few nucleotides apart (Fig. 3; data not shown). As a
control in our in vitro transcription assays, we included the

1252 M. R. Atkinson, N. Pattaramanon and A. J. Ninfa
supercoiled plasmid template pLR100 (Ninfa et al., 1987),
which has the wild-type glnAp2 promoter driving the
expression of a 420 nt transcript.
In vitro transcription assays were performed as
described previously (Ninfa et al., 1987; Feng et al.,
1995a; Liu et al., 2002), using a combination of pLR100
and either pglnApOG7 or pglnApOG8 in the reaction mixtures (Experimental procedures). In single-cycle transcription assays in which transcription complexes were formed
for a fixed time, we observed that high concentrations of
NRI~P resulted in a reduced yield of transcripts from the
wild-type promoter and a less severe reduction in the
number of transcripts from the template bearing mutations
in sites 3 and 4 (pglnApOG8; data not shown). The difference in yield of transcripts from the wild-type and mutant
templates at high activator concentration was magnified
when a multiple-cycle transcription assay was used
(Fig. 3). Previous studies showed that the multiple-cycle
transcription assay results in about four cycles of
transcription/template under the conditions used here (Liu
et al., 2002).
Discussion
Our work has shown that the combination of high-affinity
activator binding sites (enhancer) and low-affinity activator
binding sites (governor) at the glnAp2 promoter was
responsible for the expression characteristics of this promoter under physiological conditions in which the expression of NRI was from the single-copy wild-type gene. The

Fig. 3. Multiple-cycle transcription assay analysis of the role of sites
3 and 4 in activation of glnAp2 at high activator concentration. Each
transcription reaction mixture contained two templates, pLR100 (wildtype), which produces a 420 nt transcript (top band), and either
pglnApOG7 or pglnApOG8, as indicated, which produce a doublet of
about 329 nt (bottom band). NRI concentrations are in nM as indicated. Multiple-cycle transcription was initiated by the addition of
[a-32P]-UTP after preincubation of all other components at 37∞C for
10 min. Transcription was for 15 min, after which reactions were
stopped by the addition of EDTA to 0.05 M, NaCl to 0.5 M and tRNA
to 100 mg ml-1. Reactions were then extracted with phenol–chloroform, and RNA was precipitated with ETOH, resuspended in formamide and subjected to electrophoresis on a 6% polyacrylamide-7 M
urea gel.

glnAp2 promoter was very sensitive to activation by
NRI~P, yet the maximum level of expression was set by
the governor sequences. In the absence of the governor
sequences, the kinetics of activation appeared to be normal as cells became nitrogen starved and the intracellular
concentration of NRI~P increased, but expression from
the mutant promoters was dramatically elevated over that
observed with the wild-type promoter in starved cells.
Similar results were obtained in an in vitro transcription
system, in which the yield of transcripts was higher from
a template that lacked the governor sites than from the
wild-type template at high activator concentrations.
Previous results have indicated that factors binding
between the enhancer and the promoter may regulate
expression from s54-dependent promoters. For example,
binding of IHF at a site between the enhancer and the
promoter is required for efficient activation of the E. coli
glnHp2 promoter (Claverie-Martin and Magasanik, 1991)
and certain nif promoters of Klebsiella pneumoniae
(Hoover et al., 1990). Binding of IHF at these promoters
is apparently required to facilitate the activation DNA loop
and position the enhancer-bound activator and promoterbound polymerase properly (Claverie-Martin and
Magasanik, 1991). Similarly, the Nac protein of Klebsiella
aerogenes represses transcription of the nac promoter
by binding to a site located between the enhancer and the
promoter (Feng et al., 1995b). Repression by Nac
appeared to result from bending of the DNA that adversely
affected the formation of the activation DNA loop (Feng
et al., 1995b). Indeed, by altering the distance between
the Nac site and the promoter, positions were identified at
which the DNA bend induced by Nac stimulated transcription instead of repressing it (Feng et al., 1995b). Together,
the previous experiments with IHF and Nac show that
topological regulation of the activation DNA loop is an
important mechanism for the regulation of s54-dependent
promoters.
The simplest model to account for all the available data
is that binding of NRI~P to the low-affinity (governor)
sites adversely affects the formation of the activation
DNA loop and, by so doing, reduces the probability of
interaction between the enhancer-bound activator and
the promoter-bound polymerase. For example, binding of
the governor sites may limit the flexibility of the DNA,
reducing DNA looping. Alternatively, the binding of
NRI~P to sites 3 and 4 may somehow block the ability of
the activator bound to the enhancer to interact with the
promoter-bound polymerase. A unique aspect of the glnA
system is that the same transcription factor, NRI~P, is
required for activation and governing of the promoter. A
consequence of this arrangement is that cells may activate glnA expression first upon becoming nitrogen limited, and this activation may be limited in severely
starved cells co-ordinately with the activation of other Ntr
© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

Governor of glnAp2
genes. Thus, amplitude modulation of the activator concentration may bring about complex patterns of gene
activation and inactivation.
Interestingly, in our experiments in which NRI was
hyperexpressed from a multicopy plasmid based upon
pBR322, governing of glnAp2 was defective but not eliminated by mutation of sites 3 and 4. Thus, it seems that,
under these conditions, multiple mechanisms were
responsible for limiting the promoter. In a previous study,
in which NRI was expressed from the lac promoter on a
very high-copy-number plasmid, it was concluded that
promoter limitation resulted from a mechanism that did not
depend on sites 3 and 4 (Shiau et al., 1992). Our results
may be reconciled with the earlier results if the levels of
NRI expressed from pgln53 in our experiments was lower
than that obtained when it was expressed from a very
high-copy-number plasmid bearing the lac promoter, as
seems likely. Apparently, at very high activator concentration, the governor-independent mechanism dominates,
and the contribution of sites 3 and 4 to limiting promoter
activity becomes less evident.
The s54-dependent promoters of E. coli have several
features that are reminiscent of eukaryotic promoters,
such as the relative position independence of the
enhancer sequences, regulation by covalent modification
of the activator and regulation of promoter activity by
factors that influence the topology of the DNA. These
promoters form a gene cascade, in which amplitude modulation of the activator is used to effect a programme of
gene expression (Atkinson et al., 2002). The use of governor sites to limit activation of the first gene in the cascade upon amplification of the activator may prove to be
a common feature of such gene cascades.
Experimental procedures
Plasmids, site-directed mutagenesis and
strain constructions
The plasmids, primers and strains used in this study are listed
in Table 3. The plasmid pgln53 contains the glnA promoter
fused to glnG (Chen et al., 1982). The plasmid pglnAplac2
containing the glnA promoter fused to lacZ at the transcriptional +1 was constructed by polymerase chain reaction
(PCR) amplification of the glnA promoter using primers 5¢CCGGAATTCATCCTCCGCAAACAAGTATTGCAGAG and
5¢-CGCGGATCCTAAAAAAGATAAAGCGAAATCTGT
GCCAAC and cloned as an EcoRI–BamHI fragment into
pRS551 (Simons et al., 1987). To mutagenize the glnA promoter fused to lacZ at the transcriptional +165, it was first
amplified by PCR using primers 5¢-CCGGAATTCATCCTC
CGCAAACAAGTATTGCAGAG and 5¢-CGCGGATCCTTA
CACCTGATGAGCAGGGATAGTGAC and cloned as EcoRI–
BamHI into pUC18 (Yanisch-Perron et al., 1985), resulting in
plasmid pglnAp1. Mutations were introduced to this
plasmid using the Stratagene Quickchange mutagenesis
system. The sequences of the mutations introduced are
© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

1253

shown in Fig. 1. The glnA promoter mutant plasmids were
pglnAp2 (NRI site 3), pglnAp3 (NRI site 4) and pglnAp5 (NRI
sites 3 and 4). The mutant promoters were then fused to lacZ
by cloning as EcoRI–BamHI fragments into pRS551, resulting in pglnAplac5 (NRI site 3), pglnAplac6 (NRI site 4) and
pglnAplac7 (NRI sites 3 and 4).
To place the glnA promoter–lacZ fusions in single copy
within the trp operon of the E. coli chromosome, all
plasmids (pglnAplac2, pglnAplac5, pglnAplac6 and
pglnAplac7) were digested with PstI, the linear DNA transformed into TE2680 (Elliot, 1992) by electroporation using a
Bio-Rad gene pulser apparatus and recombinants selected
for kanamycin resistance. The glnA promoter–lacZ fusions
were moved into desired strains by P1vir-mediated transduction of YMC10 (Backman et al., 1981), RB9060 (Bueno et al.,
1985), Kc (Atkinson et al., 2002) and K3 (Blauwkamp and
Ninfa, 2002a). The glnA promoter–lacZ fusions were amplified by PCR, and the DNA was sequenced at the University
of Michigan DNA sequencing core to confirm that they were
as designed.

Growth media, glutamine synthetase and
b-galactosidase assays
Cells were grown at 30∞C in defined W-salts media (Pahel
et al., 1978) supplemented with carbon and nitrogen sources
as indicated. The g-glutamyl transferase activity of glutamine
synthetase (GS) was measured as described previously
(Rhee et al., 1985). Total GS was determined in reactions in
which Mn2+ was the sole metal ion, and unadenylylated GS
was determined in reactions in which Mg2+ was present in
large excess over Mn2+. In the presence of Mn2+, both adenylylated and unadenylylated GS should be active, whereas
only unadenylylated GS should be active when supplied
with Mg2+. Results were expressed as nmol of glutamylhydroximate formed min-1 mg-1 cell protein. b-Galactosidase
levels were expressed in Miller units and were assayed
with cells permeabilized using SDS and chloroform
(Silhavy et al., 1984).

DNase I footprinting and in vitro transcription assays
DNase I footprinting was conducted as described previously
(Ninfa et al., 1987), except that the plasmid DNA used was
prepared using a Concert midi-prep system (Marligen)
instead of by density gradient centrifugation. DNase I was
from Invitrogen Life Technologies, and optimal results were
obtained using 0.1 unit of enzyme for 45 s at 22∞C in reactions containing 5 nM linear, end-labelled DNA. The highest
concentration of NRI used in our footprinting experiments
was 600 nM, with NRII present at 40 nM and ATP present at
400 mM.
Single-cycle in vitro transcription assays were performed
as described previously (Ninfa et al., 1987; Feng et al.,
1995a), and multiple-cycle transcription assays were performed as described previously (Liu et al., 2002). Core RNA
polymerase was from Epicentre and was used at 50 nM. s54
and NRI were kind gifts from S. Monje and L. Field and were
purified as described previously (Ninfa et al., 1987). s54 was
used at 50 nM, and NRI was used as indicated. NRII was a
kind gift from A. Pioszak and was purified as described

1254 M. R. Atkinson, N. Pattaramanon and A. J. Ninfa
Table 3. Strains, plasmids and primers used in this study.
Strain

Relevant genotype

Construction or source

YMC10
RB9060
RB9040
TH16
SN24
Kc
K3
YMC10A
YMC10LG
TE2680
JM109

endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs DglnB2306
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs Mu lysogen glnD99::Tn10
glnA::Tn5
endA1 thi-1 hsdR17 supE44 DlacU169 DglnLG lacIq lacL8/lgln105
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs Dmdl-glnK::camr
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs DglnK1 amtB+
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs glnA::Tn5
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs DglnLG
recD1903::Tn10 trpDC700::putPA1303 [kans camr lac]
F¢ traD36 proA + B + lacIq D(lacZ)M15/D(lac-proAB)glnV44
e14 gyrA96 recA1 relA1 endA1 thi hsdR17
recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1
lac [F¢proAB lacIqZDM15 Tn10]
recD1903::Tn10 trpDC700::putPA130
[f(glnAp-lacZ1) kanr cams]
recD1903::Tn10 trpDC700::putPA130
[f(glnAp-lacZ2) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
trpDC700::putPA130 [f(glnAp-lacZ1) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnB2306 trpDC700::putPA130 [f(glnAp-lacZ1) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
Dmdl-glnK::camr trpDC700::putPA130 [f(glnAp-lacZ1) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
trpDC700::putPA130 [f(glnAp-lacZ2) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnB2306 trpDC700::putPA130 [f(glnAp-lacZ2) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
Dmdl-glnK::camr trpDC700::putPA130 [f(glnAp-lacZ2) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnK1 amtB+ trpDC700::putPA130 [f(glnAp-lacZ2) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
glnD99::Tn10 trpDC700::putPA130 [f(glnAp-lacZ2) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnLG trpDC700::putPA130 [f(glnAp-lacZ2) kanr cams]
recD1903::Tn10 trpDC700::putPA130
[f(glnAp-lacZ5) kanr cams]
recD1903::Tn10 trpDC700::putPA130
[f(glnAp-lacZ6) kanr cams]
recD1903::Tn10 trpDC700::putPA130
[f(glnAp-lacZ7) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
trpDC700::putPA130 [f(glnAp-lacZ5) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
trpDC700::putPA130 [f(glnAp-lacZ6) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
trpDC700::putPA130 [f(glnAp-lacZ7) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnK1 amtB+ trpDC700::putPA130 [f(glnAp-lacZ5) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnK1 amtB+ trpDC700::putPA130 [f(glnAp-lacZ6) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnK1 amtB+ trpDC700::putPA130 [f(glnAp-lacZ7) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
glnD99::Tn10 trpDC700::putPA130 [f(glnAp-lacZ5) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
glnD99::Tn10 trpDC700::putPA130 [f(glnAp-lacZ6) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
glnD99::Tn10 trpDC700::putPA130 [f(glnAp-lacZ7) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnLG trpDC700::putPA130 [f(glnAp-lacZ5) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnLG trpDC700::putPA130 [f(glnAp-lacZ6) kanr cams]
endA1 thi-1 hsdR17 supE44 DlacU169 hutCklebs
DglnLG trpDC700::putPA130 [f(glnAp-lacZ7) kanr cams]

Backman et al. (1981)
Bueno et al. (1985)
Bueno et al. (1985)
Reitzer and Magasanik (1986)
Schneider et al. (1991)
Atkinson et al. (2002)
Blauwkamp and Ninfa (2002b)
YMC10 ¥ TH16 P1vir
YMC10A ¥ SN24 P1vir
Elliot (1992)
Yanisch-Perron et al. (1985)

XL1-Blue
MAAplac2
MAAplac3
YMC10Apf
BApf
KcApf
YMC10Apf2
BApf2
KcApf2
KgApf2
DApf2
LGApf2
MAAplac5
MAAplac6
MAAplac7
YMC10Apf5
YMC10Apf6
YMC10Apf7
KgApf5
KgApf6
KgApf7
DApf5
DApf6
DApf7
LGApf5
LGApf6
LGApf7

Stratagene
TE2680 ¥ PstI pglnAplac2
Atkinson et al. (2002)
YMC10 ¥ MAAplac2 P1vir
RB9060 ¥ MAAplac2 P1vir
Kc ¥ MAAplac2 P1vir
Atkinson et al. (2002)
RB9060 ¥ MAAplac3 P1vir
Kc ¥ MAAplac3 P1vir
Blauwkamp and Ninfa (2002b)
RB9040 ¥ MAAplac3 P1vir
YMC10LG ¥ MAAplac3 P1vir
TE2680 ¥ PstI pglnAplac5
TE2680 ¥ PstI pglnAplac6
TE2680 ¥ PstI pglnAplac7
YMC10 ¥ MAAplac5 P1vir
YMC10 ¥ MAAplac6 P1vir
YMC10 ¥ MAAplac7 P1vir
K3 ¥ MAAplac5 P1vir
K3 ¥ MAAplac6 P1vir
K3 ¥ MAAplac7 P1vir
RB9040 ¥ MAAplac5 P1vir
RB9040 ¥ MAAplac6 P1vir
RB9040 ¥ MAAplac7 P1vir
YMC10LG ¥ MAAplac5 P1vir
YMC10LG ¥ MAAplac6 P1vir
YMC10LG ¥ MAAplac7 P1vir

© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

Governor of glnAp2

1255

Table 3. cont.
Plasmid

Relevant features and construction

Source

pRS551
pUC18
pgln53
pTE103
pLR100
pglnAplac2

lacZ fusion vector
Ampicillin-resistant cloning vector
glnA promoter fused to glnG
Transcription vector
Wild-type glnAp2 promoter in pTE103
PCR-amplified glnA promoter with 3¢ end at +1 cloned as EcoRI–BamHI
fragment into pRS551
PCR-amplified glnA promoter with 3¢ end at +165 cloned as EcoRI–
BamHI into pUC18
NRI site 3 of pglnAp1 mutagenized
NRI site 4 of pglnAp1 mutagenized
Mutation of NRI site 4 added to pglnAp2
EcoRI–BamHI fragment from pglnAp2 containing mutation of NRI site 3
cloned into pRS551
EcoRI–BamHI fragment from pglnAp3 containing mutation of NRI site 4
cloned into pRS551
EcoRI–BamHI fragment from pglnAp4 containing mutation of NRI sites 3
and 4 cloned into pRS551
EcoRI–BamHI fragment amplified from pglnAp1 using primers
glnApLRUS1 and glnApLRDS1 and cloned into pTE103
EcoRI–BamHI fragment amplified from pglnAp4 using primers
glnApLRUS1 and glnApLRDS1 and cloned into pTE103

Simons et al. (1987)
Yanisch-Perron et al. (1985)
Chen et al. (1982)
Elliot and Geiduschek (1984)
Ninfa et al. (1987)

pglnAp1
pglnAp2
pglnAp3
pglnAp4
pglnAplac5
pglnAplac6
pglnAplac7
pglnApOG7
pglnApOG8

Primer

Sequence and description

glnApU.S.2

CCGGAATTCATCCTCCGCAAACAAGTATTGCAGAG
Upstream primer for cloning glnA promoter (EcoRI)
CGCGGATCCTAAAAAAGATAAAGCGAAATCTGTGCCAAC
Downstream primer for cloning glnA promoter at +1 (BamHI)
CGCGGATCCTTACACCTGATGAGCAGGGATAGTGAC
Downstream primer for cloning glnA promoter at +165 (BamHI)
CTATATTGGTGCAACATTGTCTTCGACGTCGAGCCCTTTTGCACG
(mutagenesis primer for NRI site 3)
CGTGCAAAAGGGCTCGACGTCGAAGACAATGTTGCACCAATATAG
(mutagenesis primer for NRI site 3)
GTGGTGCAGCCCTTTTCGAGGAACGTCCCCATGATAACGCCTTTTAG
(mutagenesis primer for NRI site 4)
CTAAAAGGCGTTATCATGGGGACGTTCCTCGAAAAGGGCTGCACCAC
(mutagenesis primer for NRI site 4)
GACGTCGAGCCCTTTTCGAGGAACGTCCCCATGATAACGCCTTTTAG
(mutagenesis primer for NRI sites 3 and 4)
CTAAAAGGCGTTATCATGGGGACGTTCCTCGAAAAGGGCTCGACGTC
(mutagenesis primer for NRI sites 3 and 4)
CGCCAGGGTTTTCCCAGTCACGAC
(New England Biolabs M13/pUC sequencing primer)
GGCTGTGGGATTAACTGCGCGTCGCCG
(pRS551 sequencing primer)
CGGGGTACCGGATCCAATTGTGAGCGCTCACAATTGCACCAACATGGTGCTTAATGTTTCC
Upstream PCR primer to clone glnA promoter into pTE103 (KpnI–BamHI)
GGGAATTCAAGCTTAATTGTGAGCGCTCACAATTAAAAAAGATAAAGCGAAATCTGTGCC
Downstream PCR primer to clone glnA promoter into pTE103 (KpnI–BamHI)

glnApD.S.3
glnApD.S.5
mutNRI#3US1
mutNRI#3DS1
mutNRI#4US
mutNRI#4DS
mutNRI#4US2
mutNRI#4DS2
S1224S
pRS551rev1
glnApLRUS
glnApLRDS

previously (Ninfa et al., 1986). In single-cycle assays, NRII
was used at 25 nM, whereas in multiple-cycle assays, NRII
was used at 100 nM.
The transcription template pLR100, containing wild-type
glnAp2 promoter, has been described previously (Ninfa et al.,
1987). Templates pglnApOG7 and pglnApOG8, containing
the wild-type promoter and promoter with NRI binding sites
3 and 4 mutated, respectively, were constructed by PCR
amplification of either pglnAp1 or pglnAp4, using primers
© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

glnApLRUS and glnApLRDS (Table 3). The PCR products
were cloned into pTE103as BamHI–HindIII fragments.

Acknowledgements
We thank Leslie Field, Sarah Monje and Augen Pioszak for
the kind gifts of the proteins used in these experiments,
Vladamir Bondarenko for assistance with the in vitro tran-

1256 M. R. Atkinson, N. Pattaramanon and A. J. Ninfa
scription system, and Tim Blauwkamp for helpful discussions.
This work was supported by grant GM63642 from the NIHNIGMS to A.J.N.

References
Atkinson, M.R., and Ninfa, A.J. (1994) Mechanism and
regulation of transcription from bacterial s54-dependent
promoters. In Transcription: Mechanisms and Regulation.
Conaway, R.C., and Conaway, J.W. (eds). New York:
Raven Press, pp. 323–342.
Atkinson, M.R., Blauwkamp, T.A., Bondarenko, V., Studitsky,
V., and Ninfa, A.J. (2002) Activation of the glnA, glnK, and
nac promoters as Escherichia coli undergoes the transition
from nitrogen-excess growth to nitrogen starvation. J Bacteriol 184: 5358–5363.
Backman, K., Chen, Y.-M., and Magasanik, B. (1981) Physical and genetic characterization of the glnA–glnG region
of the Escherichia coli chromosome. Proc Natl Acad Sci
USA 78: 3743–3747.
Blauwkamp, T.A., and Ninfa, A.J. (2002a) Physiological role
of the GlnK signal transduction protein of Escherichia coli:
survival of nitrogen starvation. Mol Microbiol 46: 203–214.
Blauwkamp, T.A., and Ninfa, A.J. (2002b) Nac-mediated
repression of the serA promoter of Escherichia coli. Mol
Microbiol 45: 351–363.
Bondarenko, V., Liu, Y., Ninfa, A., and Studitsky, V.M. (2002)
Action of prokaryotic enhancer over a distance does not
require continued presence of promoter-bound sigma54
subunit. Nucleic Acids Res 30: 636–642.
Bueno, R., Pahel, G., and Magasanik, B. (1985) Role of glnB
and glnD gene products in regulation of the glnALG operon
of Escherichia coli. J Bacteriol 164: 816–822.
Carmona, M., and Magasanik, B. (1996) Activation of transcription at s54-dependant promoters on linear DNA templates requires intrinsic or induced bending of DNA. J Mol
Biol 261: 348–356.
Chen, Y.-M., Backman, K., and Magasanik, B. (1982) Characterization of a gene, glnL, the product of which is
involved in the regulation of nitrogen utilization in Escherichia coli. J Bacteriol 150: 214–220.
Claverie-Martin, F., and Magasanik, B. (1991) Role of integration host factor in the regulation of the glnHp2 promoter
of Escherichia coli. Proc Natl Acad Sci USA 88: 1631–
1635.
Elliot, T. (1992) A method for constructing single-copy lac
fusions in Salmonella typhimurium and its application to
the hemA–pfrA operon. J Bacteriol 174: 245–253.
Elliot, T., and Geiduschek, E.P. (1984) Defining a bacteriophage T4 late promoter: absence of a ‘-35’ region. Cell 36:
211–219.
Feng, J., Goss, T.J., Bender, R.A., and Ninfa, A.J. (1995a)
Activation of transcription initiation from the nac promoter
of Klebsiella aerogenes. J Bacteriol 177: 5523–5534.
Feng, J., Goss, T.J., Bender, R.A., and Ninfa, A.J. (1995b)
Repression of the Klebsiella aerogenes nac promoter. J
Bacteriol 177: 5535–5538.
Hirschman, J., Wong, P.-K., Sei, K., Keener, J., and Kustu,
S. (1985) Products of nitrogen regulatory genes ntrA and
ntrC of enteric bacteria activate glnA transcription in vitro:

evidence that the ntrA product is a s factor. Proc Natl Acad
Sci USA 82: 7525–7529.
Hoover, T.R., Santero, E., Proter, S.. and Kustu, S. (1990)
The integration host factor stimulates interaction of RNA
polymerase with NifA, the transcriptional activator of nitrogen fixation operons. Cell 63: 11–22.
Kustu, S., Santero, E., Pophan, D., and Keener, J. (1989)
Expression of s54 (ntrA)-dependent genes is probably
united in a common mechanism. Microbiol Rev 54: 367–
376.
Liu, Y., Bondarenko, V., Ninfa, A., and Studitsky, V.M. (2002)
DNA supercoiling allows enhancer action over a large distance. Proc Natl Acad Sci USA 98: 14883–14888.
Ninfa, A.J., and Magasanik, B. (1986) Covalent modification
of the glnG product, NRI, by the glnL product, NRII, regulates the transcription of the glnALG operon in Escherichia
coli. Proc Natl Acad Sci USA 83: 5909–5913.
Ninfa, A., Ueno-Nishio, S., Hunt, T.P., Robustell, B., and
Magasanik, B. (1986) Purification of nitrogen regulator II,
the product of the glnL (ntrB) gene of Escherichia coli. J
Bacteriol 168: 1002–1004.
Ninfa, A.J., Reitzer, L.J., and Magasanik, B. (1987) Initiation
of transcription at the bacterial glnAp2 promoter by purified
E. coli components is facilitated by enhancers. Cell 50:
1039–1046.
Ninfa, A.J., Brodsky, E., and Magasanik, B. (1989) The role
of NRI-phosphate in the activation of transcription from
the nitrogen regulated promoter glnap2 of Escherichia coli.
In DNA–Protein Interactions in Transcription. Gralla, J.D.
(ed.). New York: Alan R. Liss, pp. 43–52.
Ninfa, A.J., Jiang, P., Atkinson, M.R., and Peliska, J.A. (2000)
Integration of antagonistic signals in the regulation of nitrogen assimilation in Escherichia coli. Curr Topics Cell Regul
36: 31–75.
Pahel, G., Zelentz, A.D., and Tyler, B.M. (1978) gltB gene
and regulation of nitrogen metabolism by glutamine synthetase in Escherichia coli. J Bacteriol 133: 139–148.
Pahel, G., Rothstein, D.M., and Magasanik, B. (1982) Complex glnA–glnL–glnG operon of Escherichia coli. J Bacteriol 150: 202–213.
Popham, D.L., Szeto, D., Keener, J., and Kustu, S. (1989)
Function of a bacterial activator protein that binds to transcriptional enhancers. Science 243: 629–635.
Porter, S.C., North, A.K., Wedel, A.B., and Kustu, S. (1993)
Oligomerization of NTRC at the glnA enhancer is required
for transcriptional activation. Genes Dev 7: 2258–2273.
Reitzer, L.J., and Magasanik, B. (1985) Expression of glnA
in Escherichia coli is regulated at tandem promoters. Proc
Natl Acad Sci USA 82: 1979–1983.
Reitzer, L.J., and Magasanik, B. (1986) Transcription of glnA.
E. coli is stimulated by activator bound to sites far from the
promoter. Cell 45: 785–792.
Rhee, S.G., Chock, P.B., and Stadtman, E.R. (1985)
Glutamine synthetase from Escherichia coli. Methods
Enzymol 113: 213–241.
Sasse-Dwight, S., and Gralla, J.D. (1988) Probing the
Escherichia coli glnALG upstream activation mechanism in
vivo. Proc Natl Acad Sci USA 85: 8934–8938.
Schneider, B.L., Shiau, S., and Reitzer, L.J. (1991) Role of
multiple environmental stimuli in control of transcription
from a nitrogen-regulated promoter in Escherichia coli with
© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

Governor of glnAp2
weak or no activator-binding sites. J Bacteriol 173: 6355–
6363.
Shiau, S.-P., Schneider, B.L., Gu, W., and Reitzer, L.J.
(1992) Role of nitrogen regulator I (NtrC), the transcriptional activator of glnA in enteric bacteria, in reducing
expression of glnA during nitrogen-limiting growth. J Bacteriol 174: 179–185.
Silhavy, T.J., Berman, M.L., and Enquist, L.W. (1984) Experiments with Gene Fusions. Cold Spring Harbor, NY: Cold
Spring Harbor Laboratory Press, pp. 107–111.
Simons, R.W., Houman, F., and Kleckner, N. (1987)
Improved single copy and multicopy lac-based cloning vectors for protein and operon fusions. Gene 53: 85–96.
Su, W., Porter, S., Kustu, S., and Echols, H. (1990)
DNA-looping and enhancer activity. association between
DNA-bound NtrC activator and RNA polymerase at the
glnA promoter. Proc Natl Acad Sci USA 87: 5504–5508.

Wedel, A., Weiss, D.S., Popham, D., Droge, P., and Kustu,
S. (1990) A bacterial enhancer functions to tether a transcriptional activator near a promoter. Science 248: 486–
490.
Weiss, D.S., Batut, J., Klose, K.E., Keener, J., and Kustu, S.
(1991) The phosphorylated form of the enhancer-binding
protein NtrC has an ATPase activity that is essential for
activation of transcription. Cell 67: 155–167.
Weiss, V., Claverie-Martin, F., and Magasanik, B. (1992)
Phosphorylation of nitrogen regulator I of Escherichia coli
induces strong cooperative binding to DNA essential for
activation of transcription. Proc Natl Acad Sci USA 89:
5088–5092.
Yanisch-Perron, C., Vieira, J., and Messing, J. (1985)
Improved M13 cloning vectors and host strains. nucleotide
sequences of the M13mp18 and pUC19 vectors. Gene 33:
103–119.
insert Table 1 near hereinsert Figure 1 near hereinsert Table 2 near hereinsert Figure 2 near hereinsert Figure 3 near hereinsert Table 3 near hereaninfa@umich.edu

© 2002 Blackwell Publishing Ltd, Molecular Microbiology, 46, 1247–1257

1257