Proc.
Natl.
Acad.
Sci.
USA
Vol.
93,
pp.
8841-8845,
August
1996
Biochemistry
TnrA,
a
transcription
factor
required
for
global
nitrogen
regulation
in
Bacillus
subtilis
(glutamine
synthetase)
LEWIS
V.
WRAY,
JR.,
AMY
E.
FERSON,
KELLIE
ROHRER,
AND
SuSAN
H.
FISHER*
Department
of
Microbiology,
Boston
University
School
of
Medicine,
80
East
Concord
Street,
Boston,
MA
02118
Communicated
by
Boris
Magasanik,
Massachusetts
Institute
of
Technology,
Cambridge,
MA,
May
13,
1996
(received
for
review
February
20,
1996)
ABSTRACT
Expression
of the
Bacillus
subtilis
nrgAB
operon
is
derepressed
during
nitrogen-limited
growth.
We
have
identified
a
gene,
tnrA,
that
is
required
for
the
activation
of
nrgAB
expression
under
these
growth
conditions.
Analysis
of
the
DNA
sequence
of
the
tirA
gene
revealed
that
it
encodes
a
protein
with
sequence
similarity
to
GlnR,
the
repressor
of
the
B.
subtilis
glutamine
synthetase
operon.
.The
tnrA
mutant
has
a
pleiotropic
phenotype.
Compared
with
wild-type
cells,
the
tnrA
mutant
is
impaired
in
its
ability
to
utilize
allantoin,
y-aminobutyrate,
isoleucine,
nitrate,
urea,
and
valine
as
ni-
trogen
sources.
During
nitrogen-limited
growth,
transcrip-
tion
of
the
nrgAB,
nasB,
gabP,
and
ure
genes
is
significantly
reduced
in
the
tnrA
mutant
compared
with
the
levels
seen
in
wild-type
cells.
In
contrast,
the
level
of
glnRA
expression
is
4-fold
higher
in
the
tnrA
mutant
than
in
wild-type
cells
during
nitrogen
restriction.
The
phenotype
of
the
tnrA
mutant
indi-
cates
that
a
global
nitrogen
regulatory
system
is
present
in
B.
subtilis
and
that
this
system
is
distinct
from
the
Ntr
regulatory
system
found
in
enteric
bacteria.
Bacillus
subtilis
is
a
Gram-positive
bacterium
that
can
sporulate
in
response
to
nutrient
limitation
(1,
2).
The
expression
of
many
enzymes
involved
in
carbon
(3),
nitrogen
(4,
5),
and
phosphorus
(6)
assimilation
is
also
regulated
in
response
to
nutrient
availability
in
this
bacterium.
Although
the
B.
subtilis
regulatory
systems
affecting
carbon
and
phosphorus
metabo-
lism
have
been
the
subject
of
much
study,
relatively
little
is
known
about
the
regulation
of
nitrogen
metabolism
in
B.
subtilis
(4,
5).
In
Enterobacteriaceae,
the
Ntr
system
regulates
the
expres-
sion
of
glutamine
synthetase
(GS)
and
many
degradative
pathways
in
response
to
nitrogen
availability
(7).
Two
key
regulatory
factors
in
this
system
are
NRI
(NtrC)
and
NRI,
(NtrB).
These
proteins
are
members
of
the
two-component
family
of
regulatory
proteins.
The
phosphorylated
form
of
the
DNA
binding
protein
NRI
activates
transcription
of
Ntr-
regulated
genes
transcribed
by
RNA
polymerase
containing
the
o-4
sigma
factor.
The
level
of
NRI
phosphorylation
is
determined
by
the
kinase/phosphatase
activities
of
NRII.
Some
degradative
genes
are
indirectly
regulated
by
the
Ntr
system.
Nitrogen
regulation
of
the
Klebsiella
aerogenes
histi-
dine
degradative
and
urease
operons
is
controlled
by
Nac
(8).
During
nitrogen-limited
growth,
the
Ntr
system
activates
ex-
pression
of
Nac,
and
Nac
activates
expression
of
these
operons.
In
B.
subtilis,
a
number
of
nitrogen-regulated
enzymes,
i.e.,
enzymes
whose
expression
is
activated
during
nitrogen-limited
growth,
have
been
identified.
These
enzymes
include
GS
(inA),
asparaginase
(ansB),
aspartase
(ansA),
urease
(ure),
'y-aminobutyrate
permease
(gabP),
the
nitrate
assimilatory
enzymes
(nasA
and
nasBCDEF),
and
the
nrgAB
operon
(9-13;
unpublished
data).
All
available
evidence
indicates
that
reg-
ulatory
systems
analogous
to
Ntr
and
Nac
are
not
present
in
B.
subtilis
(4,
5).
The
B.
subtilis
ginRA
operon
contains
genes
encoding
GS
and
the
negative
regulatory
protein,
GlnR
(14,
15).
A
single
crA
promoter
is
used
to
transcribe
the
ginRA
operon
under
all
growth
conditions
(9).
GlnR
is
an
operon-specific
negative
regulatory
protein
because
the
glnRA
operon,
but
no
other
nitrogen-regulated
genes,
is
expressed
at
high
levels in
GlnR-
mutants
grown
on
media
containing
excess
nitrogen
(10,
12,
15).
The
GlnR
protein
binds
to
two
adjacent
operators
over-
lapping
the
glnRA
promoter
(16,
17).
Since
mutations
that
inactivate
GS
result
in
high-level
glnRA
expression
during
growth
on
medium
containing
excess
nitrogen,
the
wild-type
GS
protein
has
been
proposed
to
monitor
the
nitrogen
state
of
the
cell
and
transduce
this
information
to
GlnR
(5,
15,
18,
19).
It
is
not
known
how
GS
monitors
nitrogen
availability
or
what
constitutes
the
nitrogen
regulatory
signal
in
B.
subtilis.
How-
ever,
because
the
expression
of
all
known
nitrogen-regulated
genes
is
constitutive
inglnA
mutants
(5,
10, 12,
19),
GS
appears
to
generate
a
global
nitrogen
regulatory
signal.
No
operon-specific
transcription
factors
have
been
identi-
fied
for
the
nasA,
nasBCDEF,
nrgAB,
gabP,
or
ure
genes
(10,
12,
13).
Transcriptional
activation
of
nasA
and
nasBCDEF
expression
during
nitrogen-limited
growth
has
been
shown
to
require
an
upstream
cis-acting
site
(TGTNAN7TNACA)
(12).
Since
the
same
nucleotide
sequence
also
lies
upstream
of
the
gabP
P2
and
nrgAB
promoters
(Fig.
1),
this
sequence
may
be
the
binding
site
for
a
global
nitrogen
regulatory
protein.
This
communication
describes
identification
of
the
tnrA
gene
which
encodes
a
regulatory
protein
required
for
the
activation
of
transcription
of
gabP,
nasB,
nrgAB,
and
ure
during
nitrogen
limitation.
Interestingly,
TnrA
negatively
regulates
ginRA
ex-
pression
under
these
same
growth
conditions.
MATERIALS
AND
METHODS
Cell
Culture,
Transformations,
and
Enzyme
Assays.
Mops
minimal
medium
containing
0.5%
glucose
was
used
for
growth
of
liquid
cultures
as
previously
described
(20).
Agar
plates
were
prepared
with
balanced
salt
solution
(BSS)
minimal
medium
as
described
previously
(21).
f3-galactosidase
expression
from
lacZ
fusions
was
visualized
on
BSS
plates
containing
40
,g/ml
5-bromo-4-chloro-3-indolyl
13-D-galactoside
(X-Gal).
B.
subti-
lis
transformations
were
performed
as
described
(22).
Spore
production
was
examined
by
the
nutrient
exhaustion
and
resuspension
methods
(23).
Extracts
for
enzyme
assays
were
prepared
from
cultures
grown
to
mid-log
growth
phase
and
harvested
as
described
(21).
Urease
activity,
was
measured
as
the
urea-dependent
production
of
NH4+
(10).
One
unit
of
urease
activity
produced
Abbreviations:
GS,
glutamine
synthetase;
X-Gal,
5-bromo-4-chloro-
3-indolyl
f3-D-galactoside;
ErmR,
erythromycin
resistance.
Data
deposition:
The
sequence
reported
in
this
paper
has
been
deposited
in
the
GenBank
data
base
(accession
no.
U55004).
*To
whom
reprint
requests
should
be
addressed.
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.
8841
Proc.
Natl.
Acad.
Sci.
USA
93
(1996)
m~~~3
nrgAB
CA
TGT
C
A
GGAAATC
T
T
ACA
TGAAAATGTTTTAT
nasA
CG
TGT
C
A
CAAAAAC
T
T
ACA
CATGTCTTTTCCAG
nasB
TG
TGT
A
A
GTTTTTG
T
G
ACA
CGTTTAATGCGTTA
gabP
P2
GC
TGG
T
A
TATTTTC
T
T
ACA
CGAATTTTCGACAA
consensus
TGT
A
T
ACA
FIG.
1.
Alignment
of
nitrogen-regulated
promoters.
Nucleotides
corresponding
to
the
upstream
consensus
sequence
are
boxed.
The
-35
region
for
each
promoter
is
underlined.
1
nmol
of
NH4+
per
min.
,-Galactosidase
was
assayed
as
described
(21).
One
unit
of
f3-galactosidase
activity
produced
1
nmol
of
o-nitrophenol
per
min.
Urease
and
,B-galactosidase
activities
are
the
average
of
two
to
five
independent
determi-
nations
which
did
not
vary
by
more
than
25%.
Plasmids.
pTVlts
is
a
temperature-sensitive
B.
subtilis
plas-
mid
that
contains
Tn917
(24)
and
was
obtained
from
P.
Youngman
(University.
of
Georgia,
Athens,
GA).
Tn917
in-
sertion
libraries
were
prepared
using
pTVlts
as
described
(24).
pJL73
contains
a
spectinomycin
resistance
(SpcR)
gene
in-
serted
into
the
polylinker
of
pBluscript
II
(Stratagene)
and
was
obtained
from
A.
Grossman
(Massachusetts
Institute
of
Tech-
nology,
Cambridge,
MA).
pSF42
contains
the
entire
ginRA
operon
cloned
into
pJH101
(9).
The
EcoRI
DNA
fragment
internal
to
the
ginA
gene
in
pSF42
was
replaced
with
the
SpcR
gene
from
pJL73
to
give
pGLN14.
TheglnA
chromosomal
gene
was
disrupted
by
transforming
our
standard
laboratory
strain,
strain
168
(trpC2),
to
spectinomycin
resistance
with
pGLN14
DNA.
Strains
containing
the
Aglnl4::spec
allele
were
identi-
fied
by
their
Gln-
phenotype
and
absence
of
the
pJH101-
encoded
chloramphenicol
resistance.
lacZ
Fusions.
pSFL6
is
a
neomycin-resistant
lacZ
transcrip-
tional
fusion
vector
that
integrates
into
the
B.
subtilis
chro-
mosome
by
homologous
recombination
with
the
amyE
gene
(unpublished
data).
pNRG416
was
constructed
by
inserting
a
TaqI-NspI
nrgAB
promoter
DNA
fragment
into
pSFL6.
A
DraI-HpaI
DNA
fragment
containing
the
glnRA
promoter
was
cloned
into
pSFL6
to
give
pGLN17.
pNRG706
contains
the
gabP
promoter
region
cloned
into
pSFL6.
Derivatives
of
strain
168
containing
these
nrgAB-,
ginRA-,
and
gabP-lacZ
fusions
were
constructed
as
described
(13).
pZS5,
which
contains
the
nasB-lacZ
fusion,
was
obtained
from
M.
Nakano
(Louisiana
State
University,
Shreveport)
and
transformed
into
strain
168
(12).
The
DNA
fragments
used
to
construct
the
gabP-,
nasB-,
and
nrgAB-lacZ
fusions
all
contain
the
upstream
sequence
(Fig.
1)
that
has
been
proposed
to
be
required
for
transcrip-
tional
activation
during
nitrogen-limited
growth.
Bacterial
Strains.
Escherichia
coli
strains
KE93,
DH12S,
and
MC1061
were
used
for
DNA
cloning
experiments
(13).
B.
subtilis
strain
1A680
(lacA17
lacRl
trpC2)
(25)
was
obtained
from
the
Bacillus
Genetic
Stock
Center
(Columbus,
OH).
SF416
is
a
derivative
of
1A680
that
contains
the
nrgAB-lacZ
416
fusion.
The
tnrA62::Tn917
insertion
was
transferred
into
strain
168
(trpC2)
by
transformation
with
selection
for
the
Tn917-encoded
erythromycin
resistance
(ErmR).
HJS31
(AgInR57)
contains
an
in-frame
deletion
of
the
glnR
gene
(15)
and
was
obtained
from
A.
L.
Sonenshein
(Tufts
University,
Boston,
MA).
The
AglnR
mutation
was
moved
into
strain
168
by
transforming
the
168
AglnA1l4::spec
strain
to
Gln+
with
HJS31
chromosomal
DNA.
DNA
Sequencing.
The
dideoxynucleotide
chain
termination
method
(26)
was
used
to
determine
the
DNA
sequence.
The
entire
sequence
was
obtained
for
both
DNA
strands
using
synthetic
oligonucleotide
primers.
RESULTS
Isolation
of
a
Mutant
Defective
in
nrgAB
Expression.
Ex-
pression
of
13-galactosidase
from
the
nrgAB-lacZ
416
fusion
is
activated
over
1000-fold
during
nitrogen-limited
growth
(ref.
13;
Table
1).
Colonies
of
strain
SF416,
which
contains
the
nrgAB-lacZ
416
fusion,
are
dark
blue
when
grown
on
glucose
minimal
medium
plates
containing
X-Gal
and
a
limiting
nitrogen
source
such
as
glutamate.
Strains
with
mutations
that
prevent
the
activation
of
nrgAB
expression
during
nitrogen-
limited
growth
were
sought
by
screening
for
white
SF416
colonies
on
these
plates.
Random
Tn91
7
transposon
insertions
were
transferred
into
strain
SF416
by
transformation
using
chromosomal
DNA
isolated
from
six
independent
Tn917
libraries
and
selection
for
the
Tn917-encoded
ErmR
marker.
The
ErmR
transformants
were
replica
plated
onto
glucose
minimal
BSS
plates
that
contained
X-Gal,
glutamate
as
the
nitrogen
source,
and
selective
amounts
of
erythromycin.
White
colonies
arising
on
these
plates
could
result
either
from
a
mutation
that
prevents
high-level
expression
of
the
nrgAlB-lacZ
fusion
or
from
replacement
of
the
neomycin-
resistant
nrgAB-lacZ
416
fusion
with
the
wild-type
chromo-
somal
amyE
gene.
The
latter
class
of
mutants
was
identified
by
their
sensitivity
to
neomycin.
To
determine
if
the
Tn917
insertion
in
neomycin-resistant
white
colonies
was
linked
to
the
mutant
phenotype,
chromosomal
DNA
was
isolated
from
each
candidate
and
used
to
transform
SF416
to
ErmR.
The
defect
in
nrgAB
expression
in
strain
SF662
was
found
to
be
100%
linked
with
the
Tn917
insertion.
The
mutation
in
this
strain
was
named
tnrA62::Tn917
for
trans-acting
nitrogen
regulation.
Phenotype
of
the
tnrA62::Tn917
Mutant.
During
growth
with
the
limited
nitrogen
source
glutamate,
f3-galactosidase
expression
from
a
nrgAB-lacZ
transcriptional
fusion
was
660-
fold
lower
in
the
tnrA
mutant
than
in
the
wild-type
strain
(Table
1).
To
determine
whether
the
tnrA
gene
product
is
required
for
nitrogen
regulation
of
other
genes
in
B.
subtilis,
gabP
and
nasB
expression
was
examined
in
wild-type
and
tnrA
mutant
strains.
The
levels
of
gabP
and
nasB
expression
were
17-
and
493-fold
lower,
respectively,
in
the
tnrA
mutant
than
in
wild-type
cells
during
nitrogen-limited
growth
(Table
1).
Un-
der
the
same
growth
conditions,
the
tnrA62
mutation
also
reduced
urease
expression
7-fold
(Table
1).
These
results
indicate
that
the
tnrA
gene
product
is
required
for
high-level
expression
of
all
four
of
these
nitrogen-regulated
genes
during
nitrogen-limited
growth.
Transcription
of
the
gabP,
nasB,
nrgAB,
and
ure
genes
is
derepressed
in
glnA
mutant
strains
(10,
12).
To
examine
the
relative
roles
of
glnA
and
tnrA
in
nitrogen
regulation,
the
expression
of
these
nitrogen-regulated
genes
was
examined
in
a
strain
containing
both
of
these
mutations.
Because
glnA
mutants
are
glutamine
auxotrophs,
cultures
were
grown
using
glutamine
as
the
nitrogen
source
in
these
experiments.
Ex-
pression
of
the
nasB,
nrgAB,
gabP,
and
ure
genes
was
dere-
pressed
in
the
Ag1nA14
mutant
strain
(Table
2),
reaching
levels
Table
1.
Expression
of
nitrogen-regulated
genes
in
wild-type
and
tnrA
mutant
strains
Enzyme
specific
activity
in
Relevant
cultures
grown
on
Gene,
operon
genotype
Glutamate
+
N
Glutamate
nrgAB
Wild
type
0.06
132.0
nrgAB
tnrA62
0.03
0.2
gabP
Wild
type
0.4
6.9
gabP
tnrA62
0.4
0.4
nasB
Wild
type
0.03
148.0
nasB
tnrA62
0.1
0.3
ure
Wild
type
-2.4
91.2
ure
tnrA62
<2.1
12.9
Enzymes
assayed
were
13-galactosidase
from
gabP-lacZ,
nasB-lacZ,
and
nrgAl-IacZ
fusions
and
urease
from
the
chromosomal
ure
operon.
Cultures
were
grown
in
glucose
minimal
Mops
medium
containing
the
indicated
nitrogen
sources.
N,
ammonium
chloride.
8842
Biochemistry:
Wray
et
al.
Proc.
Natl.
Acad.
Sci.
USA
93
(1996)
8843
Table
2.
Expression
of
nitrogen-regulated
enzymes
in
wild-type,
tnrA,
and
ginA
mutants
strains
Relevant
strain
Enzyme
specific
activity
genotype
gabP
nasB
nrgAB
ure
ginRA
Wild
type
0.3
0.06
0.04
s3.2
0.4
tnrA62
0.3
0.05
0.03
<3.1
0.4
Ag1nA14
8.6
264
135
55.0
11.7
tnrA62
AglnA14
0.4
0.1
0.2
-2.6
52.1
Cultures
were
grown
in
glucose
minimal
Mops
medium
containing
glutamine
as
the
nitrogen
source.
Enzymes
assayed
were
,3-galacto-
sidase
from
the
gabP-lacZ,
nasB-lacZ,
nrgAB-lacZ,
and
glnRA-lacZ
fusions
and
urease
from
the
chromosomal
ure
operon.
comparable
to
those
found
in
the
wild-type
strain
grown
with
glutamate
as
the
sole
nitrogen
source
(Table
1).
In
the
tnrA62
glnA14
double
mutant,
the
expression
of
all
four
genes
was
almost
as
low
as
in
the
tnrA62
single
mutant
(Table
2).
These
results
indicate
that
the
tnrA62
mutation
is
epistatic
to
glnA14
and
suggest
that
the
tnrA
gene
product
receives
a
signal
for
nitrogen
availability
that
is
generated
by
the
wild-type
GS
protein.
The
ability
of
the
tnrA
mutant
to
utilize
nitrogen
sources
was
examined
by
determining
growth
rates
of
wild-type
and
tnrA
mutant
strains
in
glucose
minimal
medium
containing
various
nitrogen
sources.
The
tnrA
mutant
had
longer
doubling
times
than
the
wild-type
strain
when
grown
with
urea,
y-aminobu-
tyrate,
or
allantoin
as
the
sole
nitrogen
source,
while
no
detectable
growth
was
observed
with
isoleucine,
valine,
or
nitrate
(Table
3).
There
was
no
significant
difference
in
the
growth
rates
of
the
tnrA
mutant
and
the
wild
type
strain
in
media
containing
glutamine,
glutamate,
ammonium,
aspar-
tate,
arginine,
proline,
or
glucosamine
as
sole
nitrogen
sources
(Table
3).
Surprisingly,
the
tnrA
mutant
had
shorter
doubling
times
than
the
wild-type
strain
when
alanine
or
threonine
were
used
as
nitrogen
sources
(Table
3).
The
wild-type
168
strain
and
the
tnrA
mutant
sporulated
at
similar
frequencies
in
either
nutrient
broth
sporulation
medium
or
Sterlini-Mandelstam
resuspension
medium
(data
not
shown).
Nucleotide
Sequence
of
the
tnrA
Gene.
The
DNA
adjacent
to
the
tnrA62::Tn917
insertion
was
cloned
by
plasmid
rescue
(24)
and
sequenced.
The
tnrA62::Tn917
transposon
was
found
Table
3.
Growth
of
wild-type
and
tnrA
mutant
cultures
on
various
nitrogen
sources
Doubling
time,
min
Nitrogen
source
Urea
L-Isoleucine
y-Aminobutyric
acid
L-Valine
Allantoin
KNO3
L-Glutamine
L-Arginine
NH4Cl
L-Proline
L-Aspartate
L-Glutamate
Glucosamine
L-Alanine
L-Threonine
168
(wild
type)
150
180
200
220
230
345
50
90
120
120
120
145
270
340
650
SF62
(tnrA62)
260
2810
365
.1080
550
>1000
52
102
130
125
120
138
260
200
240
Wild-type
and
tnrA62
cultures
were
grown
to
early
logarithmic
phase
(Klett
40)
in
glucose
minimal
Mops
medium
containing
glutamate
as
the
nitrogen
source,
pelleted,
washed
with
glucose
minimal
medium
lacking
a
nitrogen
source,
and
resuspended
in
glucose
minimal
me-
dium
containing
the
indicated
nitrogen
sources
(0.2%)
for
the
growth
rate
determinations.
Cultures
grown
with
allantoin
or
glucosamine
as
sole
nitrogen
sources
were
inoculated
with
cells
pregrown
in
glutamate
minimal
medium
containing
allantoin
or
glucosamine.
to
be
inserted
into
an
open
reading
frame
containing
110
codons.
The
TnrA
protein
has
significant
sequence
similarity
with
the
B.
subtilis
GlnR
protein
(Fig.
2).
The
TnrA
and
GlnR
proteins
belong
to
a
family
of
transcriptional
regulators
that
includes
the
MerR
(27),
SoxR
(28),
BmrR
(29),
and
TipAL
(30)
proteins.
The
common
region
of
sequence
similarity
among
these
proteins
lies
within
a
domain
that
is
67-70
residues
in
length
(Fig.
2).
Analysis
of
the
TnrA
protein
sequence
by
the
method
of
Dodd
and
Egan
(31)
revealed
that
the
amino
acid
residues
between
positions
14
and
35
have
the
potential
to
form
a
helix-turn-helix
DNA
binding
motif.
The
DNA
binding
domain
of
MerR
is
located
at
the
same
position
within
its
sequence
(32).
Interestingly,
the
tnrA
gene
has
also
been
cloned
as
a
multicopy
suppressor
of
an
E.
coli
AsecG
mutation
(Vesa
Kontinen,
personal
communication).
Repression
of
ginRA
Expression
by
tnrA.
The
conserved
DNA
sequence
found
upstream
of
the
nrgAlB,
nasA,
nasBC-
DEF,
and
gabP
promoters
(TGTNAN7TNACA)
is
also
located
within
the
two
glnRA
operators
(Fig.
3).
To
determine
if
TnrA
controls
glnRA
expression,
the
production
of
3-galactosidase
from
a
glnRA-lacZ
fusion
was
examined
in
wild-type
and
tnrA
mutant
cells.
The
tnrA62
mutation
did
not
affect
ginRA
expression
in
cells
grown
in
media
containing
excess
nitrogen,
e.g.,
glutamate
plus
ammonium
(Table
4).
However,
when
cultures
were
grown
in
media
containing
the
limiting
nitrogen
source
glutamate,
ginRA
expression
was
4.4-fold
higher
in
the
tnrA62
mutant
than
in
the
wild-type
strain
(Table
4).
Similarly,
glnRA
expression
was
4.4-fold
higher
in
glutamate-grown
AgnR57
tnrA62
cultures
than
in
AglnR57
cultures
(Table
4).
These
results
indicate
that
the
TnrA
protein
represses
glnRA
expression,
but
only
during
nitrogen-limited
growth.
Nitrogen-regulated
genes
are
expressed
constitutively
in
g1nA
mutants
(10,
12).
Since
TnrA
represses
ginRA
expression
under
growth
conditions
in
which
the
expression
of
other
nitrogen-regulated
genes
is
derepressed,
then
TnrA
would
also
be
expected
to
repress
glnRA
expression
in
glnA
mutants.
This
hypothesis
was
tested
by
examining
the
effect
of
the
tnrA62
mutation
on
glnRA
expression
in
a
glnA
mutant.
The
level
of
glnRA
expression
was
4.5-fold
higher
in
the
Ag1nA14
tnrA62
double
mutant
than
in
the
AginA14
strain
in
glutamine-grown
cultures
(Table
2).
In
contrast,
similar
levels
of
glnRA
expres-
sion
were
seen
in
the
wild-type
and
tnrA
mutant
strains
grown
in
media
containing
glutamine,
a
good
nitrogen
source
for
B.
subtilis
(Table
2).
These
results
are
consistent
with
the
obser-
vation
that
tnrA
regulates
glnRA
expression
only
during
nitro-
gen-limited
growth
in
wild-type
cells.
helix
turn
helix
TnrA
MTTEDHSYKDKK
VIS
IGIVSELTGLSVRQIRYYEERKLIYPQRSSR
Gl
nR
MSDNIRRSMP
LFPIGIVMQLTELSARQIRYYEENGLIFPARSEG
TnrA
GTRKYSFADVERLMDIANKREDGVQTAE
ILKDM
RKKEQMLKNDPQ
Gi
nR
NRRLFSFHDVDKLLEIKHLIEQGVNMAGIKQIL
AKAEAEPEQKQN
TnrA
VR-KKMLEGQLNAHFRYKNR
Gl
nR
EKTKKPVKHDLSDDELRQLLKNELMQAGRFQRGNTFRQGDMSRFFH
FIG.
2.
Comparison
of
the
amino
acid
sequence
of
TnrA
and
GlnR.
Similar
residues
are
indicated
with
colons.
The
conserved
amino
acid
domain
common
to
the
MerR
family
of
proteins
has
been
boxed.
The
proposed
helix-turn-helix
region
is
indicated
above
the
amino
acid
sequence.
Biochemistry:
Wray
et
al.
Proc.
Natl.
Acad.
Sci.
USA
93
(1996)
glnRAo,
glnRAo2
(--
-I--
-,
(--
-I-
-
-
ginRA
TGATGTTAAGAATCCTTACATCGTATTGACACATAATATAACATCACCTATAATG
TnrA
sites
TGT-A-------T-ACA
TGT-A-------T-ACA
FIG.
3.
Nucleotide
sequence
of
the
ginRA
promoter
region.
The
-35
and
-10
regions
of
the
ginRA
promoter
are
underlined.
The
glnRAol
and
glnRAo2
operators
are
indicated
by
the
divergent
arrows.
Semicolons
indicate
matches
between
the
glnRAo,
and
glnRAo2
operators
and
the
proposed
TnrA
binding
sites.
DISCUSSION
Transcription
of
the
nrgAB,
gabP,
nasB,
and
ure
genes
is
only
partially
derepressed
in
a
tnrA
mutant
during
nitrogen-
restricted
growth.
Because
the
nitrogen-regulated
promoters
for
the
nrgAB,
gabP,
and
nasB
genes
all
have
poor
homology
with
consensus
sequence
for
iA
promoters
(refs.
12
and
13;
unpublished
data),
expression
of
these
genes
is
likely
to
be
positively
regulated.
In
addition,
all
three
promoters
contain
a
similar
DNA
sequence
located
upstream
of
their
-35
regions
(Fig.
1).
We
propose
that
TnrA
is
a
transcription
factor that
binds
to
this
palindromic
sequence
and
activates
transcription
from
the
gabP,
nasA,
nasB,
and
nrgAB
promoters
during
nitrogen-limited
growth.
Transcription
of
the
glnRA
operon
is
derepressed
in
the
tnrA
mutant
under
nitrogen-limiting
growth
conditions.
The
two
GlnR
operators
within
the
ginRA
promoter
region
are
also
potential
TnrA
binding
sites
(Fig.
3).
There
is
evidence
that
proteins
other
than
GlnR
can
bind
to
the
glnRA
regulatory
region
(5,
16).
In
in
vivo
footprinting
experiments,
protection
and
hypermodification
of
bases
within
the
glnRAo2
site
were
observed
in
GlnR-
strains
grown
under
nitrogen-limiting
conditions,
but
not
under
conditions
of
nitrogen-excess
(16).
These
GlnR-independent
modifications
obtained
in
vivo
are
identical
to
the
GlnR-dependent
modifications
obtained
in
vitro.
Since
the
amino
acid
sequences
of
the
putative
DNA
binding
domains
in
TnrA
and
GlnR
are
nearly
identical
(Fig.
2),
these
two
proteins
would
be
expected
have
similar
inter-
actions
with
their
DNA
binding
sites.
Taken
together,
these
observations
suggest
that
the
TnrA
protein
binds
to
the
glnRAo2
operator
during
nitrogen-limited
conditions
and
in-
hibits
ginRA
transcription.
Since
the
TnrA
protein
does
not
fully
repress
expression
from
the
glnRA
promoter,
we
assume
that
the
glnRAo2
operator
sequence
does
not
correspond
to
an
optimal
TnrA
binding
site.
The
tnrA
mutation
also
causes
altered
growth
rates
in
media
containing
several
nitrogen
compounds
(Table
3).
The
growth
defect
of
the
tnrA
mutant
on
medium
containing
y-aminobu-
tyrate,
nitrate,
or
urea
as
sole
nitrogen
sources
is
likely
to
result,
at
least
in
part,
from
the
inability
to
activate
expression
of
the
gabP,
nas,
and
ure
genes.
When
the
nitrogen
source
in
the
media
is
allantoin,
isoleucine,
or
valine,
growth
of
the
tnrA
mutant
is
also
impaired.
Because
the
pathways
for
allantoin,
isoleucine
and
valine
degradation
have
not
been
determined
in
Table
4.
glnRA
expression
in
wild-type,
tnrA,
and
glnR
mutant
strains
f3-Galactosidase
specific
activity
in
cultures
grown
on
Relevant
strain
genotype
Glt
+
N
Glt
Wild
type
2.2
11.5
tnrA62
2.6
51.1
AglnR57
60.3
15.9
AglnR57
tnrA62
63.6
69.7
Cultures
were
grown
in
glucose
minimal
medium
containing
the
B.
subtilis,
the
precise
defect
in
the
transport
and
catabolism
of
these
compounds
in
the
tnrA
mutants
cannot
be
determined.
However,
because
B.
subtilis
mutants
lacking
urease
cannot
utilize
allantoin
as
a
nitrogen
source
(unpublished
data),
the
defect
in
urease
expression
is,
at
least
in
part,
responsible
for
the
allantoin
growth
phenotype
of
the
tnrA
mutant.
Interest-
ingly,
the
B.
subtilis
sigL
gene,
which
encodes
a
-54
homolog,
is
also
required
for
utilization
of
valine
and
isoleucine
as
sole
nitrogen
sources
(33).
The
tnrA
mutant
grows
more
rapidly
than
wild-type
cells
in
medium
containing
alanine
or
threonine.
Alanine
and
threo-
nine
are
both
degraded
directly
to
ammonium
in
B.
subtilis.
Because
the
tnrA
mutant
has
higher
levels
of
GS
than
wild-type
cells
during
growth
on
nitrogen-limited
medium,
increased
incorporation
of
ammonium
into
glutamine
at
higher
rates
in
the
tnrA
mutant
than
in
wild-type
cells
may
be
responsible
for
the
shorter
doubling
time
of
the
tnrA
mutant
on
these
two
nitrogen
sources.
It
is
also
possible
that
the
expression
of
genes
required
for
the
transport
and/or
catabolism
of
alanine
and
threonine
is
negatively
regulated
by
TnrA.
TnrA-dependent
regulation
of
gene
expression
was
only
observed
under
conditions
of
nitrogen-limitation.
This
sug-
gests
that
the
activity
of
the
TnrA
protein
and/or
transcription
of
the
tnrA
gene
is
inhibited
in
cells
grown
with
excess
nitrogen.
When
B.
subtilis
cells
are
grown
with
excess
nitrogen,
the
wild-type
GS
enzyme
is
hypothesized
to
generate
a
signal
that
prevents
the
expression
of
nitrogen-regulated
genes
(Fig.
4).
The
nature
of
this
nitrogen-regulatory
signal
is
not
under-
stood.
Because
the
activity
of
the
MerR
(32),
SoxR
(28),
BmrR
(29),
and
TipAL
(30)
proteins
is
mediated
by
small
compounds,
TnrA
activity
could
be
modulated
by
a
metabolite
that
is
produced
by
GS.
This
metabolite
is
unlikely
to
be
glutamine
because
the
level
of
the
intracellular
glutamine
pool
is
higher
in
the
glnA22
mutant,
which
synthesizes
all
nitrogen-regulated
genes
constitutively,
than
in
wild-type
cells
(34).
Since
expression
of
the
glnRA
operon
is
derepressed
in
glnA
mutants
(18,
19),
GS
may
produce
a
metabolic
signal
that
functions
as
a
corepressor
with
GlnR
to
repress
transcription
from
the
ginRA
promoter
(Fig.
4).
Mutational
analysis
of
the
GlnR
protein
indicates
that
the
carboxyl-terminal
region
of
this
protein
is
involved
in
signal
transduction
(15,
35).
There
is
little
amino
acid
similarity
between
the
carboxyl-terminal
portions
of
the
GlnR
and
TnrA
proteins
(Fig.
2).
This
raises
the
possibility
that
TnrA
and
GlnR
are
regulated
in
response
to
different
regulatory
signals,
both
of
which
are
generated
by
the
wild-type
GS
protein.
It
has
been
proposed
that
GS
regulates
the
activity
of
the
GlnR
protein
by
a
direct
protein-protein
interaction
(15,
36).
Thus,
it
is
possible
that
TnrA
activity
could
be
modulated
by
a
direct
interaction
with
GS.
P
|
gI
n
[1M
ginA
}IL
I
-GlnR
e
gabP
nasA
TnrA
nasBCDEF
nrgAB
ure
FIG.
4.
Model
for
nitrogen
regulation
in
B.
subtilis.
The
hypothet-
ical
nitrogen
regulatory
signal
produced
by
GS
has
been
boxed.
indicated
nitrogen
sources
and
,-galactosidase
expression
from
a
glnRA-lacZ
fusion
examined.
Glt,
glutamate;
N,
ammonium
chloride.
8844
Biochemistry:
Wray
et
al.
Proc.
Natl.
Acad.
Sci.
USA
93
(1996)
8845
The
observation
that
the
expression
of
the
nitrogen-
regulated
ginRA,
nrgAB,
gabP,
ure,
and
nasB
genes
is
altered
in
the
tnrA
mutant,
taken
together
with
the
impaired
growth
of
the
tnrA
mutant
on
media
containing
allantoin,
isoleucine,
and
valine
as
nitrogen
sources,
indicates
that
a
global
nitrogen
regulatory
system
is
present
in
B.
subtilis.
This
is
the
first
evidence
that
we
know
of
for
a
global
nitrogen
regulatory
system
in
a
Gram-positive
bacterium.
The
TnrA
protein
be-
longs
to
a
family
of
transcriptional
activators
that
are
distinct
from
the
enteric
NR,
and
Nac
proteins
and
from
the
Synecho-
coccus
global
nitrogen
regulator
NtcA
(37).
Therefore,
the
TnrA-dependent
regulation
of
gene
expression
in
B.
subtilis
represents
a
novel
paradigm
for
nitrogen
regulation
in
bacteria.
TnrA
regulates
the
expression
of
only
one
subset
of
genes
involved
in
nitrogen
metabolism
in
B.
subtilis.
Unlike
the
case
in
enteric
bacteria,
the
expression
of
several
B.
subtilis
nitrogen
catabolic
genes
is
regulated
in
response
to
metabolites
other
than
ammonia.
For
example,
the
synthesis
of
the
histidine
and
proline
degradative
enzymes
is
repressed
by
growth
in
the
presence
of
amino
acids
(21),
whereas
high
levels
of
both
glucose
and
glutamine
are
required
to
inhibit
synthesis
of
the
arginine
degradative
enzymes
in
B.
subtilis
(38).
Regulation
of
the
expression
of
gene
products
involved
in
nitrogen
metab-
olism
by
multiple
systems
in
B.
subtilis
may
result
from
amino
acid
degradation
being
required
for
the
production
of
energy
and
precursors
of
macromolecular
synthesis
during
spore
germination,
and
perhaps
during
sporulation.
Multiple
nitro-
gen
regulatory
systems
would
ensure
the
presence
of
amino
acid
degradative
enzymes
in
sporulating
cells
and
germinating
spores,
and
also
allow
adaptation
for
growth
and
survival
under
nonsporulating
nitrogen-limited
conditions.
We
thank
M.
Nakano
for
generously
providing
pZS5
and
for
communicating
unpublished
results,
P.
Youngman
for
pTVlts,
A.
Grossman
for
pJL73,
A.
L.
Sonenshein
for
HJS31,
and
the
Bacillus
Genetic
Stock
Center
for
lA680.
We
are
grateful
to
Boris
Magasanik
for
helpful
discussions.
This
work
was
supported
by
National
Science
Foundation
Grant
MCB
9408094.
1.
Schaeffer,
P.,
Millet,
J.
&
Aubert,
J.-P.
(1965)
Proc.
Natl.
Acad.
Sci.
USA
54,
704-71
1.
2.
Dawes,
I.
W.
&
Mandelstam,
J.
(1970)J.
Bacteriol.
103,
529-535.
3.
Chambliss,
G.
H.
(1993)
in
Bacillus
subtilis
and
Other
Gram-
Positive
Bacteria:
Biochemistry,
Physiology,
and
Molecular
Genet-
ics,
eds.
Sonenshein,
A.
L.,
Hoch,
J.
A.
&
Losick,
R.
(Am.
Soc.
Microbiol.,
Washington,
DC),
pp.
213-219.
4.
Fisher,
S.
H.
(1993)
in
Bacillus
subtilis
and
Other
Gram-Positive
Bacteria:
Biochemistry,
Physiology,
and
Molecular
Genetics,
eds.
Sonenshein,
A.
L.,
Hoch,
J.
A.
&
Losick,
R.
(Am.
Soc.
Microbiol.,
Washington,
DC),
pp.
221-228.
5.
Schreier,
H.
J.
(1993)
in
Bacillus
subtilis
and
Other
Gram-Positive
Bacteria:
Biochemistry,
Physiology,
and
Molecular
Genetics,
eds.
Sonenshein,
A.
L.,
Hoch,
J.
A.
&
Losick,
R.
(Am.
Soc.
Microbiol.,
Washington,
DC),
pp.
281-298.
6.
Hulett,
F.
M.
(1993)
in
Bacillus
subtilis
and
Other
Gram-Positive
Bacteria:
Biochemistry,
Physiology,
and
Molecular
Genetics,
eds.
Sonenshein,
A.
L.,
Hoch,
J.
A.
&
Losick,
R.
(Am.
Soc.
Microbiol.,
Washington,
DC),
pp.
229-235.
7.
Reitzer,
L.
J.
&
Magasanik,
B.
(1987)
in
Escherichia
coli
and
Salmonella
typhimurium:
Cellular
and
Molecular
Biology,
eds.
Neidhardt,
F.
C.,
Ingraham,
J.
L.,
Low,
K.
B.,
Magasanik,
B.,
Schaechter,
M.
&
Umbarger,
H.
E.
(Am.
Soc.
Microbiol.,
Wash-
ington,
DC),
pp.
302-320.
8.
Bender,
R.
A.
(1991)
Mol.
Microbiol.
5,
2575-2580.
9.
Fisher,
S.
H.,
Rosenkrantz,.
M.
S.
&
Sonenshein,
A.
L.
(1984)
Gene
32,
427-438.
10.
Atkinson,
M.
R.
&
Fisher,
S.
H.
(1991)
J.
Bacteriol.
173,
23-27.
11.
Sun,
D.
&
Setlow,
P.
(1991)
J.
Bacteriol.
173,
3831-3845.
12.
Nakano,
M.
M.,
Yang,
F.,
Hardin,
P.
&
Zuber,
P.
(1995)
J.
Bacteriol.
177,
573-579.
13.
Wray,
L.
V.,
Atkinson,
M.
R.
&
Fisher,
S.
H.
(1994)
J.
Bacteriol.
176,
108-114.
14.
Strauch,
M.
A.,
Aronson,
A.
I.,
Brown,
S.
W.,
Schreier,
H.
J.
&
Sonenshein,
A.
L.
(1988)
Gene
71,
257-265.
15.
Schreier,
H.
J.,
Brown,
S.
W.,
Hirschi,
K.
D.,
Nomellini,
J.
F.
&
Sonenshein,
A.
L.
(1989)
J.
Mol.
Biol.
210,
51-63.
16.
Gutowski,
J.
C.
&
Schreier,
H.
J.
(1992)
J.
Bacteriol.
174,
671-
681.
17.
Brown,
S.
W.
&
Sonenshein,
A.
L.
(1996)
J.
Bacteriol.
178,
2450-2454.
18.
Dean,
D.
R.,
Hoch,
J.
A.
&
Aronson,
A.
I.
(1977)J.
Bacteriol.
131,
981-987.
19.
Schreier,
H.
J.,
Fisher,
S.
H.
&
Sonenshein,
A.
L.
(1985)
Proc.
Natl.
Acad.
Sci.
USA
82,
3375-3379.
20.
Neidhardt,
F.
C.,
Bloch,
P.
L.
&
Smith,
D.
F.
(1974)
J.
Bacteriol.
119,
736-747.
21.
Atkinson,
M.
R.,
Wray,
L.
V.
&
Fisher,
S.
H.
(1990)
J.
Bacteriol.
172,
4758-4765.
22.
Sonenshein,
A.
L.,
Cami,
B.,
Brevet,
J.
&
Cote,
R.
(1974)
J.
Bacteriol.
120,
253-265.
23.
Nicholson,
W.
L.
&
Setlow,
P.
(1990)
in
Molecular
Biological
Methods
for
Bacillus,
eds.
Harwood,
C.
R.
&
Cutting,
S.
M.
(Wiley,
Chichester,
U.K.),
pp.
391-450.
24.
Youngman,
P.
(1990)
in
MolecularBiologicalMethodsforBacillus,
eds.
Harwood,
C.
R.
&
Cutting,
S.
M.
(Wiley,
Chichester,
U.K.),
pp.
221-266.
25.
Errington,
J.
&
Vogt,
C.
H.
(1990)
J.
Bacteriol.
172,
488-490.
26.
Sanger,
F.,
Nicklen,
S.
&
Coulson,
A.
R.
(1977)
Proc.
Natl.
Acad.
Sci.
USA
74,
5463-5467.
27.
Misra,
T.
K.,
Brown,
N.
L.,
Fritzinger,
D.
C.,
Pridmore,
R.
D.,
Barnes,
W.
M.,
Haberstroh,
L.
&
Silver,
S.
(1984)
Proc.
Natl.
Acad.
Sci.
USA
81,
5975-5979.
28.
Amabile-Cuevas,
C.
F.
&
Demple,
B.
(1991)
Nucleic
Acids
Res.
19,
4479-4484.
29.
Ahmed,
M.,
Borsch,
C.
M.,
Taylor,
S.
S.,
Vazquez-Laslop,
N.
&
Neyfakh,
A. A.
(1994)
J.
Biol.
Chem.
269,
28506-28513.
30.
Holmes,
D.
J.,
Caso,
J.
L.
&
Thompson,
C.
J.
(1993)
EMBO
J.
12,
3183-3191.
31.
Dodd,
I.
B.
&
Egan,
J.
B.
(1990)
Nucleic
Acids
Res.
18,
5019-
5026.
32.
Shewchuk,
L.
M.,
Helman,
J.
D.,
Ross,
W.,
Park,
S.
J.,
Summers,
A.
0.
&
Walsh,
C.
T.
(1989)
Biochemistry
28,
2340-2344.
33.
Debarbouille,
M.,
Martin-Verstraete,
I.,
Kunst,
F.
&
Rapoport,
G.
(1991)
Proc.
Natl.
Acad.
Sci.
USA
88,
9092-9096.
34.
Fisher,
S.
H.
&
Sonenshein,
A.
L.
(1984)
J.
Bacteriol.
157,
612-
621.
35.
Schreier,
H.
J.
&
Rostkowski,
C.
A.
(1995)
Gene
161,
51-56.
36.
Nakano,
Y.
&
Kimura,
K.
(1991)
J.
Biochem.
109,
223-228.
37.
Vega-Palas,
M.,
Flores,
E.
&
Herrero,
A.
(1992)
Mol.
Microbiol.
6,
1853-1859.
38.
Baumberg,
S.
&
Harwood,
C.
R.
(1979)J.
Bacteriol.
137,
189-196.
Biochemistry:
Wray
et
al.
