Prokaryotic Signal Transduction PDF

ANNUAL
REVIEWS Further
Quick links to online content
Annu. Rev. Genet. 1989. 23:311-36

Copyright © 1989 by Annual Reviews Inc. All rights reserved
PROKARYOTIC SIGNAL
TRANSDUCTION MEDIATED BY
Access provided by University of Hyderabad on 01/06/20. For personal use only.
Annu. Rev. Genet. 1989.23:311-336. Downloaded from www.annualreviews.org
SENSOR AND REGULATOR

PROTEIN PAIRS
Lisa M. Albright, Eva Huala, and Frederick M. Ausubel

Department of Genetics, Harvard Medical School and Department of Molecular
Biology, Massachusetts General Hospital, Boston, Massachusetts 02114
CONTENTS
INTRODUCTION . . . .... . . . ... . . . . 312

HOMOLOGY BETWEEN SIGNAL TRANSDUCING REGULATORY PAIRS . . . . . .. . . 316
Th e S enso rs . . . . . . . . . . . . . . . . . . . . .. . . . . . . . .... . . . . . . . . . .. . . . . . . . ...... . . . . . .. . . . . . . . . ...... . . . . . . . .
. 316
The Re gu la tors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
Ident ification of Ne w Signa l Tra nsdu cing Regu latory Pairs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
CONSERVED INTERACTION BETWEEN SENSOR AND REGULATOR . . . . . . . . . .. . . 317
Th e Nt rBINtrC System . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
The CheA IChe Y + Che B S ystem . . . . . . . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
The E nvZI OmpR System . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
Th e PhoR IPhoB S ystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
C ross-Ta lk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
FUNCTION OF THE SENSOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
Si gna l De te c tion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
Domain Structure of th e S enso rs .. . . . . .. . .. .. .. . . . .. . . . . . ................
. . . . . . . 323
FUNCTION OF THE REGULATOR . ............... . . ..... . . ........... . . . . . . . . . . . . . . . . . . . . . .. . 324
Nt rC Sub cla ss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
OmpR Subcla ss. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
CheB Subcla ss. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Che Y Subcla ss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
GENETIC ANALYSIS OF SIGNAL TRANSDUCTION IN REGULATORY
PAIRS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . 328
SUMMARY AND PROSPECTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... .... . . . . . . . . . . . . ... . 329
311
0066-4197/89/1215-0311$02.00
312 ALBRIGHT, HUALA & AUSUBEL
INTRODUCTION
In bacteria, signal transduction in response to a wide variety of environmental

stimuli is mediated by pairs of proteins that communicate with each other by a
conserved mechanism involving protein phosphorylation. Initial genetic and
biochemical analyses of the gene pairs envZ/ompR (33), ntrB/ntrC (11, 82),
and phoR/phoB (119) suggested that one member of each pair codes for a
sensor (modulator) protein (EnvZ, NtrB, PhoR) that modifies the activity of
its partner, a regulator (effector) protein (OmpR, NtrC, PhoB) . Subsequently ,
DNA sequence analysis revealed that these three sensor/regulator (modulator/
effector) pairs belong to a large superfamily of signal-transducing regulatory

pairs, characterized by the feature that the sensor proteins are all similar in
their C-domains while their regulator partners share homology in their N
domains (3, 14, 19, 35, 55, 86, 92, 110, 113, 126, 127, 131).
The finding that NtrB phosphorylates its partner NtrC (82) and the similar
ity in domain organization among other homologous regulatory pairs sug
gested the following general model for signal transduction for these systems
(Figure 1; 86, 94, 127): The N-domain of the sensor protein receives an
environmental signal, transducing that signal to its C-dom ain. Utilizing a
mechanism con served among the homologous regul atory pairs , the C-domain
of the sensor phosphorylates the N-terminal domain of the regulator .
Phosphorylation of the regulator alters the activity of its C-domain, which
ultimately carries out the appropriate response. Thus, diverse environmental
signals are transduced through a common mechanism to regulate a wide range
of processes in bacteria, including response to nutrient deprivat ion, osmolar
ity changes, chemotaxis, sporulation , symbiosis, and bacterial pathogenesis
(Table 1).
The term "sensor" as used here applies only to the function of the protein
within a regulatory pair. Because some of these regulatory pairs are located in
ENVIRONMENTAL
STIMULUS
,+
Nil i��C SENSOR
+
REGULATOR N ..
t!0"",0""m�0:1"",--
__ -,---,1 C
TARGET
RESPONSE
Fi gu re 1 Model for signal transduction in homologous regulatory pairs. Diagonal shading
indicates conserved domains.
KpNt.rB 116
RpNt.rB 126
R I DctB 392
StPgtB 359
AV QDLVQA�RLA
AM DELIQTAKLA
�
RLSQEQLQH�QQIAARD
KM RQLTHRSAAR I
E < C pxA 225 FN MVTALERMMTSQ

EcEnvl 220 FNHMAACV L DR �
EcPh R 190 VARDVTQM
o L RR
BsPhoR 337 VFHDMTETK L
Ec Ph oM 242 AllfISMRVKLfElC
AtWHVirA 451 VLARRLEHA/QJR
RrnF i xL 221 QIIQIAELARLAR
Ec Uhp 8 29D NQHLAERLLIEJT
B s De gS 166 KQIiiIFCLRIT1!;iA E
StCheA 370 LTLVCSSTELDKS
StCheA-N 25 QHLLDLVPESPDAE
KpNtrB 172
RpNtrB 181
RIDctB 461
StPgtB 418
EcCpxA 278
EcEnvZ 271
EcPhoR 247
BsPhoR 395
EcPhoM 298
AtWHY irA 508
RmF i xL 279
EcUhpB 348
B s De gS 216
s.� �;�� 61 :: i Li9iETT�LMENLqDE�R G�"iliLNJ!JD i i NLFIQETKD iMQE:
KpNtrB 225 LGPECGEITLRTRTAFQ

RpNtrB 236
RIDdi8 506
StP tB g 473
EcCpxA 336
EcEnvl 324
EcPhoR 3D4
BsPhoR 454
EcPhoM 355
AtWHY i r A 563
RmF i xl 330
EcUhpB 404
BsDogS 2 74
StCheA 404
KpNtrB 278 ... RE�

RpNtrB 295 ... KQ
RIDctB 552 ... KE
StPgtB 516 .. KAV
EcCpxA 377
EcEn vZ 3.4
E cPh R 347o
BsPhoR 497
Ec Ph oM 397
AtWHYirA 620
RmF ixl 384
EcUhpB 442
BsDegS 312
StCh eA 428
KpNtrB 311
RpNtrO 328 11 .. ·
RIDctB 58 5
StrgtS 549 7 .. ·
Ec (p x A 41. 2 .. ·
EcEnvl 401 1D .. ·
EcPhoR 386 6
B s P h oR 53. 4
EcP h oM 434 1
AtWHVirA 655 134
RmF i xl 417 8
EcUhpB 475 10
BsOegS 338
S tC h eA 487 145
F igu re 2 flomology among sensor components. Boxes surround amino acids with conservative
replacements according to the following groupings: ILVM, C, PAGST, HKR, QNED. Refer
ences for sequences are in Table 1. WHVirA: wide host range virA; CheA-N: N-terrninal region
of CheA.
w
......
�
�
Superfamily of signal-transducing regulatory pairs
�
Table 1
�
Or ganism' Environmeniai signal Sensorg Regu!ato� Target References :-,l
:I:
NtrC Regulator Subclass (utilizes (T54)d; c:::
:>
Ec, Kp, Nitrogen status NtrBb,j NtrCk Nitrogen assimilation genes 19, 53, 65, 86
Rm, RI, Bj, &:
Bsp
Pl>
RI bC OctO dcrA :>
C4-dicarboxylates DctB 92 c:::
C rn
St Phosphog1 ycerate PgtBb PgtA pgtP 34, 49, 115a, 131, 132
c:::
t:C
OmpR Regulator Subclasse;
tTl
t'"'
i
Ec, SI MeJiulll osmolarity EnvZ bc OmpRk ompC. ompF 26,60
c
Bs, Ec Phosphate status PhoRb j PhoBk phoA. phoE. psi 69, 69a, 96, 97, 99, 119
Ec Carbon status? Ph oM
bc ORF2 ?; can be phoA 3, 122
At Acetosyringone VirAc VirG vir genes 58, 106, 127
Ec Sexual state? CpxAc SfrAi Conjugation genes 2,20, 100
Ec Redox state? ArcB SfrAi Aerobic metabolism genes 46-48, Iuchi & Lin , personal
communication
bc
St pH?, C & N status? PhoQ PhoP pagC, other vir genes 74
Fix] Regulator Subclassf:
Rm O2 status FixUc Fix] niJA. JixK 14, 17, 118
Ec G lucose-6-phosphate UhpBbC UhpA uhpT 27, 125, 126

Bs Degsb DegU Exoenzyme genes 35, 55
Ec UvrC- ? 34, 98, 115a
23Kd
Bp ? BvgCbC BvgA Virulence genes 7, 73, 109
CheB Regulator Subclass (methylesterase):

"'d
Ec, St attractants, etc. CheAi CheBk Chemoreceptors (MCPs) 79, 89, 107, I l O :;:0
0
�
�
CheY Regulator Subclass (regulator consists solely of conserved N-terminal domain):
Ec, St attractants, etc. CheAi Cheyk Flagellar motor 79, 89, 107, 110
Bs nutrient status? ? SpoOF Sporulation genes 62, 117
�
-
('")
Miscellaneous Regulator: en
-
Bs nutrient status? ? SpoOA Sporulation genes 21, 62 0
Z
• At, Agrobacterium tumefaciens; Bj, Bradyrhizobium japonicum; Bsp, Bradyrhizobium species; Bp, Bordetella pertussis, Bs, Bacillus subtilis; Ec,
Escherichia coli; Kp, Klebsiella pneumoniae; RI, Rhizobium leguminosarum; Rm, Rhizobium meliloti; St, Salmonella typhimurium.
�
'":l
b Sensor and regulator genes are located in the same operon :;:0
C Sensor predicted to be membrane-bound :>
d Regulator central domain also homologous to NifA (12, 19, 92), TyrR (13, 34), and HrpS (32) z
en
'Regulator Codomain also homologous to ToxR (75) 0
'Regulator Codomain also homologous to UvrC-ORFI (35, 55, 98), MalT (34, 35, 55), LuxR (34), GerE (34), and RcsA (34) c:::
g Sensor component has also been called "modulator" (66) and its conserved domain has been called a "transmitter motif" (54)
h Regulator component has also been called "effector" (66) and its conserved domain has been called a "receiver motif" (5 4)
Q
-
'Also known as AreA, Dye. FexA, Msp, Seg 0
Z
J Sensor shown to be a protein kinase
k Regulator shown to be a substrate for sensor protein kinase
w
..-
VI
the middle of a signal-transduction pathway, the sensor is not alw ays the
component utilized by the cell to percei ve the primary environmental stimu
l us. The terms "modulator" and "effector" have also been used to refer to the
sensor protein kinase and its regulator substrate , respectively (66).
HOMOLOGY BETWEEN SIGNAL TRANSDUCING

REGULATORY PAIRS
Table I summarizes several features of gene products that have been identified
as of March 1989 to be homologous to the superfamily of sign al-transducing
regulatory pairs . Many of the sensor components are predicted to be trans

membrane proteins with a substantial periplasmic domain (Table 1). The
regul ator component is often , though not exclusively , a regulator of gene
expression. Most of the transcriptional regulators can be further subclassified
by homology between their C-domains (Table 1 ). The sensor and regulator
genes are often organized in an operon (Table 1); in at least one case
(envZ/ompR) (60), translational coupling has been shown.
The Sensors
The conserved domain of the sensor class extends over approximately 250
amino acids and is usually located at the C-terminus of the protein (Figure 2).
Although most members show significant ho mology extending throughout the
entire domain , five regions showing the strongest conservation can be de
lineated; four of these surround at least one invari ant amino acid . The
align ments between different sensors and regulators have been evaluated
statistically (14, 86, 126) using the computer program PIRALIGN. In most
cases scores are greater than 10 standard deviation units above the average
score of alignments of the same statistically jumbled sequences (16). For
UhpB , CheA and DegS, alignment values > + 5 standard devi ation units can
be achieved with only one or two other members of the set. On the other hand,
each of these sensor proteins has been shown genetically or biochemically to
f unction with their respective regulator partner(s) (35, 38, 105, 110, 126).
Because of the low alignment scores , CheA , UhpB and DegS have been
manually aligned in Figure 2 to the most highly conserved regions.
The Regulators
The conserved domain in the regulator class extends over approximately 120
amino acids and in most cases is located at the N-terminus of the protein
(Figure 3, see page 325). Most members show strong homology throughout
the entire domain; however, four regions with the strongest conservation can
be delineated . Regions 1 and 2 have an invariant aspartic acid residue that
may be i mportant since NtrC is phosphorylated at an aspartic acid residue
within its N-terminal domain (52, 124).
PROKARYOTIC SIGNAL TRANSDUCTION 3 17
Identification of New Signal Transducing Regulatory Pairs

Using a new algorithm to search for homology to the most highly conserved
amino acids of region 4 (Figure 3 ) , Henikoff et al . (34) have identified three
n ew potential members of the regulator class; PgtA from Salmonella typhi
murium (Table 1), "TrpO" in the Pseudomonas aeruginosa trpAB region , and
"PetO" in the Rhodopseudomonas capsu[ata petABC region . Kofoid & P ar
kinson (54) took a different approach to look for other pairs of proteins that
retain f undamental structural features similar to the conserved domains of the
signal-tran sducing regulatory pair s . Using relaxed matching criteria to search
for what they called "transmitter" and "receiver" motif s , they identified a
number of other pairs of proteins, such as PtsG/Crr , which interact in the

phosphoenolpyruvate-dependent glucose phosphotransferase uptake system
(PTF ) ( 1 12) and MaiE/MalF , which probably interact in maltose transport (4).
Interestingly, by Kofoid & Parkinson's criteria, CheA and NtrB appear to
carry both transmitter and receiver motifs. B oth of these sensor components
monitor the environment indirectly through interaction with other protein s
(53, 8 9, 110).
CONSERVED INTERACTION BETWEEN SENSOR AND

REGULATOR
The conservation of protein domains in the signal-transducing regulatory pairs

suggests that the interaction between each pair involves a mechanism con
served in evolution . In each of the regulatory pairs discussed below , the
sensor component is a protein kinase that phosphorylates its regulator partner .
In this section we summarize the evidence that phosphorylation of the regula
tor by the sensor represents a conserved mechanism for signal transduction
between partner s .
The NtrBINtrC System

The NtrB/NtrC pair regulates the expression of several genes required for
n itrogen assimilation . The most well-characterized target gene is gInA, which
codes for glutamine synthetase, the major enzyme in enteric bacteria for
assimilating low concentrations of NH3 . Transcription of gInA occurs from
tandem promoters glnApl, which is repressed by NtrC, and glnAp2, which is
activated by NtrC (91). Activation mediated by NtrC requires the sigma f actor
NtrA (al so called (]"54 or RpoN) (53). Under conditions of nitrogen starvation ,
h igh levels of transcription from the glnAp2 promoter are induced by NtrC
and NtrA-polymerase (53, 65-67, 90).
The protein kinase activity of the sen sor NtrB (also called GlnL or NRu)
consists f irst of an autophosphorylation at a histidine residue using ATP as a
phosphate donor and subsequent transfer of that phosphate to the regulator
NtrC (also called GlnG or NR,) (52, 82, 124). Phosphorylation of NtrC
correlates with its ability to act as a transcriptional activator in vitro; efficient

transcription from glnAp2 requires NtrB and ATP in addition to NtrC and
NtrA-RNA polymerase (53, 82, 85). The continuous presence of NtrB is
probably required because phospho-NtrC is unstable and undergoes auto
dephosphorylation (52, 124). Phospho-NtrC is also rapidly dephosphorylated
in a regulated manner in a reaction requiring NtrB, ATP, and another regula
tory protein called PH (product of glnB) (52, 82), resulting in reduced
transcriptional activity of NtrC (53, 82). This is consistent with the effect of
PH in vivo, which is to antagonize NtrB action in response to nitrogen
sufficiency (53, 65). Information about the nitrogen status of the cell is
transmitted to PH by a uridylyltransferase/uridylyl-removing enzyme (UTase

UR; product of glnD) which uridylylates PII, probably in response to a low
ratio of intracellular glutamine to 2-ketoglutarate. PH in its uridylylated form
appears to have no effect on NtrB activity, but when unmodified promotes
NtrB-mediated dephosphorylation (i.e. inactivation) of Phospho-NtrC.
The following regulatory cascade therefore operates in the NtrB/NtrC
system: The uridylyl-removing activity of UTase-UR is stimulated by a
signal, possibly a high intracellular glutamine to 2-ketoglutarate ratio, in
dicative of nitrogen sufficiency. This increases the concentration of un
modified Pu in the cell, which in tum increases the function of NtrB in
stimulating removal of phosphate from NtrC, leading to inactivation of NtrC
as a transcriptional activator and tum off of ginA expression. Conversely,
nitrogen starvation favors NtrB-catalyzed phosphorylation of NtrC, which is
then active in transcription of ginAp2 (66).
The CheAICheY + CheB System

Chemotactic behavior is achieved by biasing the direction of flagellar rotation
in response to changes in external stimuli over time (89, 107). External
stimuli are transduced across the inner membrane by a family of methyl
accepting chemotaxis proteins (MCPs; 5). Wild-type cells alternate between
smooth swimming and tumbling. "Gutted" chemotactic mutants, which lack
all of the k nown signal-transducing regulatory gene products, show smooth
swimming behavior because of counterclockwise rotation of the flagella.
Addition of the sensor CheA, the regulators CheY and CheB, and two other
proteins, CheW and CheZ, is sufficient to restore signalling to gutted mutants
(129).
CheW is thought to couple a primary signal received by the MCPs to the
sensor CheA (10) that in tum signals to both CheY and CheB. The regulator
CheY in its activated form biases the flagellar motor toward clockwise (CW)
rotation, resulting illl tumbling (15, 56, 129). The regulator CheB is a
methylesterase which methylates MCPs to regulate their signalling capability,
permitting adaptation of the signalllng machinery to an environmental stimu-
PROKARYOTIC SIGNAL TRANSDUCTION 319
Ius. Addition of repellent or removal of attractant increases the activity of the

CheB methylesterase (107, 108 ) .
I n vitro, CheA undergoes autophosphorylation at His48 using ATP as a
donor (37 , 3 8 , 39 , 130). Phosphate is transferred from phospho-CheA to
CheY or CheB (3 8 , 130). Phospho-CheY and phospho-CheB undergo a
relatively slow hydrolysis (38,130) . CheZ accelerates hydrolysis of Pi from
CheY (3 8 , 88), consistent with its in vivo role as an antagonist of CheY action
(56, 129).
Although the physiological role of the protein kinase activity of CheA has
not been shown directly, mutagenesis of the cheA gene has shown a strong
correlation between mutants defective in chemotaxis and those altered either

in autophosphorylation or in transfer of phosphate to CheB and CheY (37,
88). In addition, ATP depletion antagonizes the ability of CheY to generate a
tumble signal ( 103) , consistent with the model that phospho-CheY promotes
clockwise rotation of the flagellum, resulting in tumbling (although the ability
of CheY to bind ATP may indicate a more direct role (103)). Likewise, the
CheB-mcthylesterase activity is inhibited by ATP depletion (103 ) , suggesting
that phospho-CheB has increased methylesterase activity, resulting in net
removal of methyl groups from MCPs. Thus, phospho-CheY and phospho
CheB appear to mediate the response to negative chemotactic stimuli, and
CheA appears to act as an integrator of signals from the MCPs (37, 89).
The CheA/CheB, CheY system is an example of a signalling network
which is slightly more complex than the NtrB/NtrC system, in which one
sensor and one regulator transmit a single environmental signal. In the
CheA/CheB, CheY system, a single sensor (CheA) signals to two different
regulators (CheY and CheB). Conversely, the sensors CpxA and ArcB each
appear to be capable of transmitting a unique environmental signal through
the same regulator, SfrA (46, 47).
The EnvZ/OmpR System

The sensor EnvZ and the regulator OmpR control expression of the outer
membrane porins OmpC and OmpF (33) which are regulated in a reciprocal
manner by medium osmolarity: low osmolarity favors OmpF synthesis while
high osmolarity favors OmpC synthesis (26). The regulator OmpR is abso
lutely required for expression of both genes, while envZ null mutants show
significantly reduced expression of both genes (29, 77). The interaction of
OmpR with the ompF and ompC promoters is likely to be complex given that
the binding sites for the two promoters are different (51, 76 , 87) . In addition,
there appears to be post-transcriptional regulation of ompF through a small
antisense RNA called micF (26). EnvZ and OmpR are also required for
normal expression of the tripeptide permease gene,tppB, but this regulation is
independent of the medium osmolarity (30).
The biochemistry of the sensor EnvZ has been difficult because of its
membrane location. Autophosphorylation of wild-type EnvZ (41) and of
EnvZ containing N-terminal deletions that remove up to 179 amino acids (1,
24, 41, 42) has been demonstrated. This phosphate linkage exhibits a pH
dependent stability similar to that of phospho-CheA and phospho-NtrB ,
suggesting a similar phospho-histidine linkage (24, 42). I n vitro , EnvZ
containing a 38 amino acid N-terminal deletion phosphorylates OmpR, which
stimulates the ability of OmpR to activate transcription from the ompF
promoter (41). P hospho-OmpR is dephosphorylated in a reaction mediated by
EnvZ and requiring ATP ( l a).
The PhoRIPhoB System
The PhoRJPhoB pair regulate expression of the phosphate regulon, which

includes phoA and other phosphate starvation-inducible (psi) genes . In vitro,
PhoR protein lacking the N-terminal 83 amino acids autophosphorylates and
phospho-PhoR trans�ers phosphate to PhoB . Phosphorylation of PhoB corre
lates with an increase in its ability to bind DNA upstream of the pstS promoter
and to activate transcription at that promoter (K. M akino, personal com
munication).
Cross-Talk
If phosphorylation of the regulator by the sensor represents a conserved mode

of signal transduction, unless each sensor is tuned perfectly to its cognate
regulator , it seems likely that a particular sensor might "cross-talk" with (i . e .
phosphorylate or dephosphorylate) regulators from other systems . This occur s
in both directions for the CheA/CheY , CheB and NtrB/NtrC systems . NtrC is
phosphorlyated in vitro by CheA, and the phospho-NtrC produced i n this
m anner stimulates in vitro transcription from glnAp2 (84). In the other
direction, phospho-NtrB transfers phosphate in vitro to CheY (84). I n vivo,
overproduction of a mutant allele of ntrB (NtrB2302), which phosphorylates
but does not dephosphorylate NtrC , suppresses the smooth-swimming phe
notype of a cheA mutant. U nless NtrB has some other CheA-like activity , this
strongly suggests that phospho-CheY generates a tumbling signal (84). In one
other example of cross-talk in vitro, NtrB has been shown to phosphorylate
SpoOA (52).
I n vivo, cross-talk between signal-transducing regulatory pairs provides a
means for i ntegrating many types of environmental stimuli . If cross-talk
contributed significantly to target activity, one might expect that a null sensor
mutant in one system could have a pleiotropic effect in other systems .
Although no such dramatic evidence exist s , there is some genetic evidence
that cross-talk can occur in vivo under special circumstances.
PhoM and the P hoR/PhoB system provide the most compelling genetic
evidence for cross-talk. The PhoR/PhoB pair regulate expression of phoA and
other phosphate starvation-inducible (psi) genes in response to phosphate
level. In the absence of the PhoR sensor, phosphate-independent expression
of phoA (coding for alkaline phosphatase) can occur. This expression depends
on the function of an unlinked gene, phoM (70, 1 2 1) , whose product is
homologous to the sensor class. PhoM-dependent expression of phoA stilI
requires the regulator P hoB , consistent with phosphorylation of P hoB by
PhoM . The physiological relevance of PhoM -dependent expression of phoA is
unclear, because phoM mutations have no effect on phoA expression in a
phoR+ background (99 , 1 1 9) . In fact , the ORF directly upstream of phoM in

the phoM operon (ORF2) is homologous to the regulator class (3) . The normal
function of the putative PhoM/ORF2 pair therefore seems likely to be the
regulation of different class of genes. Recent evidence that PhoM responds to
catabolite repression suggests that the target of the PhoM/ORF2 pair may be a
class of catabolite-respon sive genes (120, 122) .
One other hint that cross-talk can occur in vivo is also suggested by the fact
that expression or overproduction of the regulator proteins NtrC (65) , OmpR
(25 , 29, 30, 77, 87, 1 02), DctD (L. Albright & c. Ronson , unpublished
data), CheY ( 1 5 , 56 , 1 03 , 1 29), UhpA ( 1 25 , 1 26) , PgtA (49, 132) , FixJ (36) ,
and PhoB (68, 96) in the absence of their sensor partners elicits a target
response that, in most cases, is constitutive. These observation s might be
explained by low-level cross-talk from the sensors of other systems. Alterna
tively, the regulator protein may have some inherent activity in the absence of
sensor modification/phosphorylation .
The general absence of pleiotropic phenotypes for null mutants in sensor
proteins and the suggestion by the PhoR-PhoM example that cross-talk is not
seen until normal interactions are interrupted indicates that a high degree of
specificity within each pair has been achieved, probably in large part by
mutational optimization. The genetic organization of many of the pairs into
operons (see Table 1 ) , in some cases with translational coupling (envZlompR;
60), may also contribute to the maintenance of a specific interaction .
FUNCTION OF THE SENSOR
The concentration of conserved amino-acid residues in the Codomain of the

sensors (Figure 2) and the large variation in the structure of the N-domains
suggests that the N-domain has the capacity to receive a particular environ
mental signal and that the conserved Codomain transmits the signal through
phosphorylation of the regulator. Several experiments described in this sec
tion address the question of whether the receiving and transmitting functions
of the sensor are carried out by two distinct domains.
Signal Detection
The signal-transducing regulatory pairs appear to be using one of two modes
of signal detection: either the sensor receives an environmental stimulus
directly , or the environmental stimulus is f irst perceived and transduced by a
primary environmental sensor into an intracellular (or intramembrane) signal
that is then detected by the sensor component of the two-component system.
The l atter is clearly the case for CheA and NtrB , which are cytoplasmic
proteins and f unction as part of an intracellular relay mechanism. In the
chemotaxis system, the detection of external ligands by transmembrane recep
tor proteins called MCPs (89) is coupled to the sensor protein CheA through
CheW (10, 89). In the NtrB/NtrC system, GInA (glutamine synthetase) can be
thought of as the primary sensor of environmental ammonia (66, 67). Gluta
mine synthetase generates an intracellular signal (glutamine) that is relayed to
the NtrB sensor prote:in by the UTase-UR and Pn proteins as described in an
earlier section. DegS is another example of a sensor component thought to be
a cytoplasmic protein. In the DegS/DegU system , two additional genes
encoding polypeptides of 46 and 60 amino acid arc implicated in regulation of
degradative exoenzymes in Bacillus subtilis, although the signal to which this
regulatory pair responds remains unknown (35, 55).
In some cases where the sensor component of a regulatory pair is a
membrane protein, there is also evidence for the involvement of proteins other
than the sensor component in signal perception . In the case of the D ctBIDctD
pair, the regulator DctD activates expression of the dctA dicarboxylate
transport gene in the presence of external C4-dicarboxylic acids. Insertions in
dctA result in constitutive synthesis of a dctA-lacZ fusion (92). It is possible
that the transport protein DctA and the putative protein kil}ase DctB cooperate
in detecting the presence of C4-dicarboxylic acids outside the cell (92). A
similar phenomenon is seen in the PhoR/PhoB system that activates phoA, the
gene coding for alkaline phosphatase. Mutations in the pst-phoU locus,
which, with the exception of phoU, codes for the high-affinity inorganic
phosphate transport system, lead to constitutive expression of phoA. This
suggests that phosphate repression is achieved by transport of inorganic
phosphate by the pst system, with PhoU f unctioning to transduce that signal to
PhoR (80, 116, 119).
Many other sensor components predicted by hydropathy profiles to have a
disposition acro ss the inner membrane may serve as the primary sensor of an
environmental stimullus. These l atter sensors appear to be similar in structure
to the chemoreceptors (MCPs) , which have a periplasmic l igand-binding
domain and a cytopla smic domain that interacts with the chemotactic signal
l ing apparatus (Tabk I; 6). For the most part, the presumptive cytoplasmic
domain of these transmembrane sensors corresponds to the conserved sensor
domain postulated to interact with the regulator protein. The sensors EnvZ
PROKARyOTIC SIGNAL TRANSDUCTION 323
(23, 59), VirA (58), UhpB (125) and PgtB (131) have been localized to
membrane fractions, and VirA appears to span the inner membrane in a
manner similar to that of MCPs (128). These sensor proteins may therefore
bind specific external ligands and have a function similar to that of MCPs.
Domain Structure of Sensors
For those sensors that appear to have a transmembrane location, the separa
tion of signal-detection and signal-transmission functions into separate peri
plasmic and cytoplasmic domains, respectively, makes sense both from a
functional and an evolutionary point of view. Experimental evidence supports

this concept. For example, deletions of 38, 55, and 179 amino acids at the
N-terminus of the EnvZ protein are predicted to remove EnvZ from its
membrane location. The presence of EnvZ carrying a 38 or 55 amino acid
N-terminal deletion partially complements an envZ null mutant but does not
restore normal regulation of OmpF and OmpC (42, 77). Significantly, EnvZ
carrying a 38 or 179 amino acid deletion retains autophosphorylation and
OmpR kinase activity (24, 41, 42), consistent with the separation of the
signal-reception and signal-transmission functions into two distinct domains.
In contrast to EnvZ, there is both biochemical and genetic evidence in the
case of CheA that some determinants for interaction with CheY and CheB
occur in the N-domain. Although the strongest homology between CheA and
other sensors lies near its C-terminus (Figure 2), autophosphorylation of
CheA occurs near the N-terminus of the protein at His48 (37). At a genetic
level, Oosawa et al (88) performed in vitro mutagenesis of the cheA gene,
screening for chemotaxis-defective mutants, and mapped each mutation to
one of five regions defined by restriction sites. Five mutations mapping to
region I (amino acids 1-170) gave rise to CheA proteins capable of undergo
ing autophosphorylation but defective in transfer of phosphate to CheY and
CheB. Four mutations produced full-size CheA protein that could not undergo
autophosphorylation. One of these mapped in region I and the other three
mapped in region IV, roughly amino acids 420-540, corresponding to regions
3, 4 and 5 of Figure 2. These results indicate that determinants at the
N-terminus and in conserved region IV are necessary for autophosphoryla
tion, but that interaction with CheB and CheY is determined by the N
terminus alone (88). This is consistent with the ability of an N-terminal l 8-kd
CheA peptide to transfer phosphate to CheB and CheY, but not to undergo
autophosphorylation (37). CheA has two translational starts 90 amino acids
apart, giving rise to CheAL and CheAs (104). Only CheAL undergoes auto
phosphorylation, indicating that at least some of the determinants for auto
phosphorylation are located within the first 90 amino acids of the protein (37,
38).
One way to reconcile the CheA data with the overall model of sensor!
regulator interactions (Figure 1) is to postulate that the conserved sensor
domain in CheAL has been split into two parts. One part would consist of
amino acids 1-170 of CheA, corresponding to region 1 of Figure 2, and the
second part would consist of amino acids 290-420 of CheA (region IV as
defined by Oosawa et al (88)), corresponding to regions 2-5 of Figure 2.
Although the sites of autophosphorylation in NtrB or EnvZ have not yet been
mapped, it is tempting to speculate, as others have, that phosphorylation of all
of the sensor components takes place at the invariant histidine in region 1 of
Figure 2 (124). This interpretation is consistent with the findings that NtrB
autophosphorylation occurs at a histidine (124) and that the pH stability of

phospho-EnvZ and phospho-PhoR is consistent with phospho-histidine (24,
42; A. Makino, personal communication). An alignment of the region sur
rounding His48 of CheA with region 1 of the sensors is shown in Figure 2.
This alignment shows no statistical significance by P1RALIGN criteria (align
ment values <2). Nevertheless, because of the data cited above, it appears
possible that determinants for recognition and interaction of the sensor with
the regulator lie in region 1 and adjacent portions of the conserved sensor
domain, while determinants for autophosphorylation are contained in both
parts of the domain. It is intriguing that the portion of the conserved sensor
domain that includes region 1 is in general more weakly conserved than the
rest of the domain, which might be expected if determinants for specific
recognition of the regulator lie in this region.
Because CheA is an unusual sensor in its overall structure (54) and in its
ability to communicate with two regulators, it may have unique features.
Therefore caution must be used in generalizing the results obtained with CheA
to other sensor proteins. In any case, because CheA is an intracellular sensor,
there are no architectural constraints that would necessarily prevent separation
of sensor subdomains to the N- and C-terminal positions of the protein. More
data on the location of autophosphorylation of other sensor components are
needed before a general model for the function of sensor proteins can be
formulated.
FUNCTION OF THE REGULATOR
The presence of a conserved N-domain is the distinguishing feature of the

regulators (Figure 3). The conservation of the N-domain points to this portion
of the regulator participating in the conserved interaction with a sensor
protein. This is borne out by experiments with NtrC discussed below. In
addition, good evidence exists for the division of signal reception and re
sponse functions into two separate domains within regulators containing a
second domain. Most regulators that function as transcriptional activators can
be grouped into subclasses based on homology within their C-domains (Table

I). These structural features of the regulator s suggest that a common mech
anism of transcriptional regulation exists for each subclass and that the
r egulators evolved by the "mixing and m atching" of domains. In support of
this latter idea, the C-domain of most of the subclasses is shared by at least
one other protein that does not function as a member of a signal-transducing
regulatory pair (Table 1). The relationships between the regulator subclasses
has been elegantly illustrated in a recent paper on methods for searching
nucleic acid databases (34).
NtrC Subclass
NtrC belongs to a group of transcriptional activators requiring the alternate
s igma factor NtrA (RpoN; (T54) (31, 57, 90, 93). In addition to NtrC,
members of this group include DctD, PgtA, and NifA (Table I). Of these,
PgtA is the only one for which an NtrA requirement has not been shown .
These four regulators (NtrC , DctD , NifA , and PgtA) share a strongly con
served block of 238 amino acids in the central region and a less well
conserved block of 45 amino acids at the C-terminal end that contains a
KpNtrC
BpNtrC
RmNtrC
BsSpoOF
RIDctD
EcSfrA
EcUmpR
EcPhoB
BsPhoP
EcPhoMORF2 1
AtV;,G 25
RmFixJ 1
EcUhpA 1
EcCher 1
EcCheB 1
B.!:SpoOA 1
StPgtA -23
EcUvrCORF2 1
BsDegU 1
KpNttC 62
BpNttC 62
RmNtrC 62
BsSpoOF 62
RIDctD 63
EcSfrA 62
EcOmpR 63
EcPhoB 61
BsPhoP 61
EcPhoMORF2 62
AtVirG 86
RmFixJ 62
EcUhpA 62
EcCheY 65
EcChee 64
BsSpoOA 64
StPgtA 41
EcUvrCORF2 55
BsOegU 64
F igu re 3 Homology among regulator components. See Fi gu re 2 for other details.

helix-turn-helix DNA-binding motif (19, 92). However, NifA, which acti

vates structural genes required for nitrogen fixation in a variety of Gram
negative bacteria (12), does not contain the conserved N-terminal domain
shared by the other regulators considered in this review.
There is good evid,�nce that both the N-terminal and central domains of the
NtrC subclass can function autonomously. In the case of the N-terminal
domain, NtrC is phosphorylated at an aspartic acid residue in the amino
terminal 12.5 kd of the protein (52, 124). This peptide alone, which es
sentially constitutes the conserved domain shared by all regulators, is an
efficient substrate for the NtrB kinase, and undergoes autodephosphorylation
and regulated dephosphorylation with kinetics comparable to that of the entire

protein (52). This indicates that the N-domain of the regulator component
contains sufficient de:terminants for interaction with and modification by the
sensor component.
The presence of a large highly conserved block of amino acids in the central
domain of NtrC, DctD, and NifA has led to the prediction that the central
domain interacts directly with NtrA-RNA polymerase holoenzyme to stimu
late transcription from NtrA-dependent promoters (12, 19, 92). Evidence in
support of this predktion comes from experiments with all three activators.
For example, the Rhizobium leguminosarum DctD regulator can drive low
level constitutive expression of its target dctA promoter in the absence of the
sensor DctB (L. AlbIight & C. Ronson, unpublished data). A modified DctD
protein that deletes the conserved N-domain yields high constitutive levels of
dctA expression. This not only indicates that the N-domain of DctD is
unnecessary for transcriptional activation but also that the N-terminal domain
has a regulatory role O. Stigter, L. Albright & F. Ausubel, unpublished data;
B. T. Nixon, personal communication). Similarly, deletion analysis of the
NifA proteins from Klebsiella pneumoniae (78), Rhizobium meliloti (8, 40) ,
and Bradyrhizobium japonicum (22) has shown that neither the N-domain of
NifA nor its C-terminal helix-turn-helix domain is required for transcriptional
activity, and that the central domain of NifA shared by NtrC, DctD, and PgtA
is sufficient to activate transcription (40). In addition, several point mutations
in NtrC that result in a specific defect in the ability to activate transcription
map to highly' conserved amino acid residues in the central domain (57).
The central domain of the NtrC class is also found in HrpS, a regulatory
protein in Pseudomonas syringae pv. phaseolicola which is required for
pathogenesis of host plants and induction of a hypersensitive defense response
on nonhost plants. HrpS consists of the central domain and a possible
helix-turn-helix motif at its C-terminus (32). The central domain is additional
ly found in TyrR, a repressor of general and specific aromatic amino acid
transport and general aromatic amino acid biosynthesis in E. coli (1 3, 34).
The homology of TyrR to the activators of the NtrC subclass is particularly
intriguing in that it appears to function mainly as a repressor, although a

positive role in regulation of tyrP may be indicated (50; C. Stephens, personal
communication).
Finally, NtrA has been implicated in the regulation of the Pseudomonas
putida xylABC (xylenes and toluene metabolism) (44), carboxypeptidase G2
(18), and pilin genes (45), of the Caulobacter crescentus flbG and flaN
flagellar genes (83), and of the E. coli hydB and fdhF genes required for
formate metabolism (9,95). The previous domain homologies predict that the
regulatory proteins for these genes will have homology to the central domain
of the NtrC subfamily (already shown for XylR (44».
OmpR Subclass
OmpR, PhoB, PhoM-ORF2, VirG and SfrA share a common C-domain (127)
that is also found as an N-domain in ToxR, a transcriptional activator for
cholera toxin, pilus, and outer membrane proteins in Vibrio cholerae (75).
OmpR (51,76,87),PhoB (68),and ToxR (75) each bind DNA upstream of
the transcriptional start site of their target promoters; each of the binding sites
contains short direct repeats. A 16-kd proteolytic fragment of the OmpR
protein corresponding to the C-domain binds to the same sequence upstream
of the ompF promoter as does the full-length protein, although the fragment is
about tenfold less efficient in binding (114). Experiments using a protein
fusion between ToxR and PhoA (alkaline phosphatase) have demonstrated
that the homologous N-domain of ToxR is directly responsible for DNA
binding and transcriptional activation (75). The C-domain of the ToxR is
thought to be localized in the periplasm on the basis of toxR::TnphoA fusions
and is implicated in the mediation by ToxR of a response to NaCI but not to
pH and amino acids (73, 75).
CheB Subclass
In the case of the CheB regulator, the N-domain of inactive CheB appears
able to compete with wild-type CheB and CheY for phospho-CheA (108).
When expressed at high levels, a CheB-LacZ fusion protein that deletes the
C-terminal half of CheB inhibits the swarming ability of wild-type cells as
well as the ability of the wild-type methylesterase activity to increase in
response to negative stimuli. The same phenomenon is seen with three
different mutant CheB proteins that are catalytically inactive due to missense
mutations that map in the C-terminal half of CheB.
A number of observations indicate that the catalytic esterase activity of
CheB resides in its C-domain, and that its N-domain regulates the esterase
activity in response to negative stimuli (101, 108). A 21-kd proteolytic
fragment of the CheB methylesterase, representing the C-terminal 60% of the
protein, has fifteenfold higher methylesterase activity in vitro than the entire
protein (101). In addition, N-terminal deletions of up to one half of CheB

retain methylesterase activity in vivo. The normal increase of methylesterase
activity in response to negative stimuli is impaired in these deletions although
the decrease in activity in response to positive stimuli is not (108). The same
work showed that the response to both negative and positive stimuli depends
on cheA function, although others (lOS) have suggested that only the response
to negative stimuli requires CheA.
CheY Subclass
The structure of the CheY and SpoOF regulators, which consists only of the
conserved N-domain found in other regulators, clearly does not allow for a
separation of their signal reception and response functions into separate
domains. The three··dimensional structure of CheY was recently reported
(111).
GENETIC ANALYSIS OF SIGNAL TRANSDUCTION IN

REGULATORY PAIRS
The model in Figure 1 predicts that each component receives an input in its
N-domain, and that the N- and C-domains interact in an intramolecular
reaction to generate the appropriate output through the C-domain. Input and
output mutants should therefore cluster in the N- and C-domains, respective
ly, of each component. Mutations defective in intramolecular transduction
could reside in either domain; moreover, it should be possible to find allele
specific intragenic suppressors within individual sensor and regulator genes.
In previous genetic analyses, the most intensively studied mutations have
been those resulting illl constitutive activation of promoters normally regulated
by a signal-transducing pair. These mutations most likely fall into the in
tramolecular transdw;;tion category and, as expected, map throughout both the
sensor and regulator genes. In the NtrB/NtrC system, mutants selected for
high constitutive expression of gInA map in either ntrB or ntrC (63, 64, 91).
NtrB prepared from one such mutant (ntrB2302) does not mediate in vitro
dephosphorylation of NtrC in the presence of Pu (82). The ntrC610 allele,
which in vivo results in elevated levels of ginA expression in the absence of
NtrB, maps to the central domain of NtrC but other NtrC constitutive mutants
map to the N-terminal domain (D. Popham & S. Kustu, personal communica
tion). A similar all(�le of ntrC that also maps in the central domain was
recently reported (123). Significantly, the NtrC610 protein cannot undergo
regulated dephosphorylation (52). This indicates that determinants in the
central region of NtrC mediate interaction of its N-domain with NtrB and Pu.
In the DegS/DegU system, mutations resulting in overproduction of a number
of degradative enzymes have been mapped to the conserved domain in each
component. Mutations in DegS map at amino acids 236 (V�M) and 218
(�E). Those in DegV map at amino acids 1 2 (H�L) and 1 07 (E�K) (35).
In the EnvZ/OmpR system, an ompR constitutive mutant with a phenotype of
OmpP' OmpC- (the ompR2 class) maps to the C-domain (V203�M) (81)
and its gene product binds to the ompF promoter but not the ompC promoter
(76). Conversely, an OmpR mutant with a OmpP- Ompe phenotype
(OmpR3) maps to the N-domain (81; R l � C) and its gene product binds to
both the ompF and ompC promoters , as does wild-type OmpR protein. The
same OmpP- OmpCc phenotype is seen with envZ alleles 160, 1 1 , and 473
(71, 115). envZ160 maps in the N-domain (L35�Q) (7 1) and envZl l maps in
the C-domain (T247�R) (71). Other unmapped mutations of this class are the
R Ib mutants of phoR, which give rise to phoM- independent constitutive

expression of phoA (119), and the pho5 1O allele of the phoM locus, which
produces cAMP-independent expression of phoA in a phoR-mutant back
ground (120).
The envZl 1 mutation is particularly interesting as it may provide an
example of a mutation in a sensor for which an allele-specific suppressor
mutation in the regulator has been obtained. ompR77 (L 1 �Q) partially
suppresses the OmpP- and the Ompec phenotypes of the envZl l mutation in
an allele-specific manner , consistent with a direct interaction between EnvZ
and OmpR (7 1) . However, a simple interpretation ofthese mutants is difficult
because both envZl l and envZ473 show pleiotropic repression of phoA,
phoE, maLE, and lamB ( 1 15), a phenotype not seen in envZ null alleles (29,
77) . The pleiotropic repression by the envZl l mutation is also relieved by
ompR77 (7 1) . In addition , suppression of the envZl l LamB - and OmpP
phenotypes can occur by mutations in the rpoex operon , coding for the ex
subunit of RNA polymerase and several ribosomal proteins (28), and a
mutation in rpoa can interfere with suppression of envZl l by ompR77 (72) .
This may indicate that interaction between OmpR and RNA polymerase can
"feed back" to affect the interaction between EnvZ and OmpR or that EnvZ
modification of OmpR affects the interaction between OmpR and RNA
polymerase.
Recently, it has been shown that EnvZ l l and OmpR3 are each defective in
EnvZ-mediated dephosphorylation of OmpR. In addition, OmpR77 is de
fective in phosphotransfer from EnvZ l l but not from EnvZ ( 1 a).
SUMMARY AND PROSPECTS
The evidence is now quite convincing that a common mechanism of interac

tion involving phospho-transfer has been conserved among the regulatory
pairs discussed in this review. In many of the systems , a more detailed
understanding of the interaction of the regulator with its target and the role of
phosphorylation in that interaction is needed. In addition , the mechan ism(s)
of intramolecular signal transduction within each component remains to be
determined. Nevertheless , the phospho-transfer reaction , which must still be

demonstrated in many of the signal-transducing regulatory pairs, has provided
a powerful assay for intermediate steps in the signal transduction from sensor
to regulator component. Modem mutagenesis schemes aimed at identifying
mutations and suppressors within each regulatory pair, coupled with the
phosphotransfer assay, should prove extremely useful in dissecting determi
nants involved in signal reception, transduction , and output in each com
ponent (88).
It should also be possible to design mutagenesis schemes based on the
domain structure of the sensor component to uncover other components (if
they exist) necessary for signalling in systems where the nature of the signals
being perceived is not understood. As has already been pointed out (86, 94),
the predictive value of the signal-transducing regulatory pair homology also
points the way in cases where regulator components have been found (such as
SpoOA and SpoOF) but where the sensor component(s) has not been identified
by conventional means.
ACKNOWLEDGMENTS
We thank our many colleagues who provided reprints and unpublished in
formation; Boris Magasanik, Sydney Kustu , John S. Parkinson, and Tom
Silhavy for critically reading the manuscript, Clive Ronson for initiating the
effort to write this chapter and for literature searches, and Rachel Hyde for
help in literature searches and manuscript preparation.
Literature Cited
1. Aiba, H. , Mizuno, T. , Mizushima, S . mechanism and evolution. Annu. R ev.

1989. Transfer o f phosphoryl group be Bioch em . 55:397-425
tween two regulatory proteins involved 5. Ames, P . , Chen, J . , Wolff, c . , Parkin
in osmoregulatory expression of the son, J. S. 1989. Structure-Function stud
ompF and ompC genes in Esch eri chia ies of bacterial c hemosensors. Cold
coli. J. Biol. Ch em . 264:8563-67 Sp ring Ha rbor S ymp. Quan t. Biol.
l a. Aiba, H. , Nakasai, F . , Mizushima, S . , 53:59-65
Mizuno, T . 1989. Evidence for the 6. Ames, P . , Parkinson, J. S . 1988. Trans
physiological importance of the phos membrane signalling by bacterial che
photransfer between the two regulatory moreceptors: Esch eri chia coli transduc
components, EnvZ and OmpR, in ers with locked signal output. C ell
osmoregulation in E sch erichia coli. J. 5 5 : 8 1 7-26
Bioi. Chem . , In pn:ss 7. Arico, B . , Stibitz, S . , Royo, C . , Miller,
2. Albin, R . , Weber, R . , Silverman, P. M . J . , Falkow, S, et al. 1989. Gene prod
1986. The Cpx proteins o f E scherichia ucts required for expression of Bo r
coli K 12 . J. Biol. Ch e m . 261:4698--705 d e te lla per tussis virulence factors share
3. Amemura, M . , Makino, K., S hinagawa, homology with prokaryotic signal
H . , Nakata, A . 1 986. Nucleotide se transduction proteins. , Pro c. Natl.
quence of the phoM region of Es A ca d. S ci . USA , In press
ch eri chia coli: four open reading frames 8. Beynon, 1. L . , Williams, M . K . , Can
may constitute an operon. J. Bacterial. non, F. C. 1988. Expression and func
168:294--302 tional analysis of the Rhizobium m eliloti
4. Ames, G. F.-L. 1986. Bacterial peri nifA gene. EM BO J. 7:7-14
plasmic transport systems: structure , 9. Birkmann, A . , Sawers, R. G . , Bock, A .
1 987. Involvement of the n trA gene Spence, J. , Ferrari , E . , et al. 1985.

product in the anaerobic metabolism of Characterization of the spoOA locus and
Es che rich ia col i. Mol. Gen. Gene t. its deduced product. Pr oc. Na tl. A cad.
2 1 0:535-42 S ci. USA 82:2647-5 1
10. Borkovich, K. A . , Kaplan, N . , Hess, J. 22. Fischer, H.-M. , Bruderer, T . , Hen
F. , Simon, M . I . 1989. Transmembrane necke, H. 1988. Essential and nonessen
signal transduction in bacterial chemo tial domains in the Bra dyrh izob ium
taxis involves ligand-dependent activa japonicum NifA protein: identification
tion of phosphate group transfer. P ro c. of indispensable cysteine residues poten
Natl. A cad, Sci. USA 86: 1 208- 1 2 tially involved in redox reactivity and/or
1 1 . Bueno, R . , Pahel, G . , M agasanik, B . metal binding. Nucleic A cids Res.
1 985. Role of glnB and glnD gene prod 16:2207-24
ucts in regulation of the glnALG operon 23. Forst, S . , Comeau, D . , Norioka, S . , In
of Esche rich ia col i. J. Bacte riol. ouye, M. 1 987. Localization and mem
164: 8 1 6-22 brane topology of EnvZ, a protein in

1 2 . Buikema, W. J . , Szeto, W. W . , Lem volved in osmoregulation of OmpF and

ley, P. V . , Orme-Johnson, W. H . , Au OmpC in Escherichia c oli. 1. B ioi.
subel, F. M. 1985 . Nitrogen fixation Chem . 262: 1 6433-38
specific regulatory genes of K lebsie lla 24. Forst, S . , Delgado, J . , Inouye, M .
pneumon iae and Rhizob ium me li loti 1 989. Phosphorylation o f OmpR b y the
share homology with the general nitro osmosensor EnvZ modulates the expres
gen regulatory gene ntrC of K. pneumo sion of the ompF and ompC genes in
n iae. N ucle ic A cids Res. 1 3:4539-55 Esche rich ia col i. Proc. Natl. A cad. S ci.
1 3 . Chye, M . - ! . , Pittard, J. 1 987. Transcrip USA . In press
tion control of the aroP gene in Es 25. Forst, S . , Delgado, J . , Ramakrishnan,
che rich ia coli K- 1 2 : analysis of operator G . , Inouye, M. 1 988. Regulation of
mutants. 1. Bacteriol. 1 69:386-93 ompC and ompF expression in Es
14. David, M . , Daveran, M . -L. , Batut, J . , cherich ia col i in the absence of envZ. J.
Dedieu, A . , Domergue, 0 . , e t a ! . 1988. Ba cteria l. 170:5080-85
Cascade regulation of nif gene expres 26. Forst, S . , Inouye, M. 1988. Environ
sion in Rh izobium melilo ti. Cell 54:671- mentally regulated gene expression for
83 membrane proteins in Esche rich ia co li.
15. David, S . , Matsumura, P., Eisenbach, Annu. Rev. Cell BioI. 4:21-42
M. 1986. Restoration of flagellar clock 27. Friedrich, M . J . , Kadner, R . J. 1987.
wise rotation in bacterial envelopes by Nucleotide sequence of the uhp region of
insertion of the chemotaxis protein Escherich ia col i. J. Bacteriol. 169:
CheY. P roc. Na tl. Acad. S ci. USA 3556-63
83:7157-61 28. Garrett, S . , Silhavy, T. J. 1 987. I sola
16. Dayhoff, M. 0 . , Barker, W. c . , Hunt, tion of mutations in the a operon of
L. T. 1983. Establishing homologies in Es cherich ia col i that suppress the tran
DNA sequences. Me tho ds Enz ymol. scriptional defect conferred by a muta
9 1 :524-45 tion in the porin regulatory gene envZ. J.
17. Ditta, G . , Virts, E . , Palomares, A . , Ba cte ria l. 169: 1 379-85
Kim, C.-k. 1 987. The nifA gene of Rh i 29. Garrett, S . , Taylor, R. K . , Silhavy, T.
zob ium mel ilo ti is oxygen regulated. J. J . , Berman, M. L. 1985. Isolation and
Ba cteriol. 169:32 1 7-23 characterization of .1ompB strains of Es
1 8. Dixon, R . 1986. The xylABC promoter cherichia c oli by a general method based
from the Pseudomonas pu tida TOL plas on gene fusions. J. Bacterial. 162:840-
mid is activated by nitrogen regulatory 44
genes in Escherichia c oli. M ol. Gen . 30. Gibson, M. M . , Ellis, E. M . , Graeme
Genet. 203 : 1 29-36 Cook, K. A . , Higgins, C. F. 1987.
19. Drummond, M . , Whitty, P . , Wootton, OmpR and EnvZ are plei otropic regula
J . 1986. Sequence and domain rela tory proteins: positive regulation of the
tionships of n trC and nifA from Klebs iel tripeptide permease ( tppB) of Sal
la pne umoniae: homologies to other monella typh imu rium . M ol. Gen. Gene t.
regulatory proteins. EMBO 1. 5:441- 207 : 1 20-29
47 3 1 . Gussin, G. N . , Ronson, C. W . , Au
20. Drury, L . S . , Buxton, R. S . 1985. DNA subel, F. M. 1 986, Regulation of nitro
sequence analysis of the dye gene of gen fixation genes. Annu. Rev. Gene t.
Escher ich ia col i reveals amino acid 20:567-91
homology between the Dye and OmpR 32. Grimm, c . , Panopoulos, N. J. 1 989.
proteins. J. B ioi. Chem. 260:4236-42 The predicted protein product of a
2 1 . Ferrari, F. A . , Trach, K . , Lecoq, D . , pathogenicity locus from Pseu domonas
syringae is homologous to a highly con from Pseudomonas putida . Gene

served domain of several prokaryotic 66:30 1-6
regulatory proteins. 1. Bacteriol. , In 45. lshimoto, K . , Lory, S. 1989. Formation
press of pilin in Pseudomonas aeruginosa re
33. Hall, M. N . , Silhavy, T. J. 1981. Genet quires the alternative U' factor (RpoN) of
ic analysis of the ompB locus in Es RNA polymerase. Proc. Natl. Acad.
cherichia coli K - 12 . 1. Mol. Biol. Sci. USA 86:1954-57
151:1-15 46. Iuchi, S . , Cameron, D. C. , Lin, E. C.
34. Henikoff, S . , Wallace, J . c . , Brown, J. C . 1 989. A second global regulator gene
P. 1989. Finding protein similarities (arcB) mediating repression of enzymes
with nucleotide s(�quence databases. in aerobic pathways of Escherichia coli.
Methods Enzymol. , In press J. Bacterio!. 171:868-'73
35 . Henner, D. J . , Yang, M . , Gerrari , E . 47. Iuchi, S . , Furlong, D . , Lin, E. C. C .
1988. Localization o f Bacillus subtilis 1989. Differentiation o f areA , arcB, and
sacU (Hy) mutations to two linked genes

cpxA mutant phenotypes of Escherichia

with similarities to the conserved pro coli by sex pilus formation and enzyme
caryotic family of two-component regulation. 1. Bacterio!. 171 :2889-93
signalling systems. 1. Bacterio!. 48. Iuchi, S . , Lin, E. C. C. 1988. arcA
170:5102-09 (dye), a global regulatory gene in Es
36. Hertig , c . , Ya Li, R . , Louarn, A . -M . , cherichia coli mediating repression of
Garnerone, A . -M . , David, M . , et al. enzymes in aerobic pathways . Proc.
1989. Rhizobium meliloti regulatory Nat!. Acad. Sci. USA 85:1888-92
gene fix] activates transcription of R. 49. Jiang, S . -Q. , Yu, G . -Q. , Li, Z. -G. ,
meliloti nifA and fixK gencs in Es Hong, J .-S. 1 988. Genetic evidence for
cherichia coli. 1. Bacteriol. 171: 1736- modulation of the activator by two reg
38 ulatory proteins involved in the ex
37. Hess, J. F. , Bourret, R. B . , Simon, M . ogenous induction of phosphoglycerate
I . 1988. Histidine phosphorylation and transport in Salmonella typhimurium. 1.
phosphoryl group transfer in bacterial Bacterial . 1 70:4304-08
chemotaxis. Nature 336: 139-43 50. Kasian, P . A . , Davidson, B. E . , Pittard,
38. Hess, J. F . , Oosawa, K . , Kaplan, N . , J. 1986. MolelOular analysis of the pro
Simon, M. I . 1988. Phosphorylation of moter operator region of the Escherichia
three proteins in the: signalling pathway coli K- 1 2 tyrP gene. 1. Bacterial.
of bacterial chemotaxis. Cell 53: 79-87 167 :556-61
39. Hess, J. F. , Oosawa, K . , Matsumura, 5 1. Kato, M . , Aiba, H . , Mizuno, T. 1989.
P. , Simon, M. I. 1 987. Protein phos Molecular analysis by deletion and site
phorylation is involved in bacterial che directed mutagenesis of the cis-acting
motaxis. Proc. Natl. Acad. Sci. USA upstream sequence involved in activa
84:7609- 1 3 tion of the ompF promoter in Es
4 0 . Huala, E . , Ausubel , F. M . The central cherichia coli. 1. Biochem. (Tokvo) .
domain of Rhizobium meliloti NifA is 105:341-47
sufficient to activate transcription from 52. Keener, J . , Kustu , S. 1988. Protein
the R. meliloti llifT-I promoter. J. Bac kinase and phosphoprotein phosphatase
terial. 171:3354-65 activities of nitrogen regulatory proteins
41. Igo, M . M . , Ninfa, A. J . , Silhavy, T. J . NTRB and NTRC of enteric bacteria:
1989. A bacterial environmental sensor roles of the conserved amino-terminal
that functions as a protein kinase and domain of NTRC. Prac. Natl. Acad.
stimulates transcriptional activation. Sci. USA 85:4976-80
Genes Dev. 3:598-605 53. Keener, J . , Wong, P . , Popham, D . ,
42. Igo, M . M . , Silhavy, T. J. 1988. EnvZ, Wallis, J . , Kustu, S . 1987. A sigma
a transmembrane environmental sensor factor and auxiliary proteins required for
of Escherichia coli K- 1 2 , is phosphory nitrogen-regulated transcription in enter
lated in vitro. 1. Bacterio!. 170:597 1-73 ic bacteria. In RNA Polymerase and the
43. Ikenaka, K. , Tsung, K . , Comeau, D. Regulation of Transcription, ed. W. S .
E . , Inouye, M. 1 9S11. A dominant muta Reznikoff, R. R . Burgess, J . E. Dahl
tion in Escherichia coli OmpR lies with berg, C. A. Gross, M. T. Record, Jr. ,
in a domain which is highly conserved in M . P . Wickens. pp. 159-75. New York:
a large family of bacterial regulatory Elsevier
proteins. Mol. Gen. Genet. 211:538-40 54. Kofoid, E. C . , Parkinson, J. S. 1988.
44. Inouye, S. , Nakazawa, A . , Nakazawa, Transmitter and receiver modules in
T. 1988. Nucleotide sequence of the reg bacterial signaling proteins. Proc. Natl.
u latory gene xylR of the TOL p lasmid Acad. Sci. USA 85:4981-85
55. Kunst, F . , Debarbouille, M . , Msadek, Regulation of carbon and nitrogen

T . , Young , M . , Mauel, c . , et al . 1 98 8 . utilization. In Escherichia coli and Sal
Deduced polypeptides encoded by the monella typhimurium: Cellular and Mo
Bacillus subtilis sacU locus share lecular Biology, ed. F. C. Neidhardt, J .
homology with two-component sensor Ingraham, K . B . Low, B . Magasanik,
regulator systems. 1. Bacterial. 1 70: M. Schaechter, H . E. Umbarger, pp.
5093- 1 0 1 1 3 1 8-25. Washington, DC: Am. Soc.
5 6 . Kuo, S . C . , Koshland, D . E. , Ir. 1 987. Microhiol
Roles of the cheY and cheZ gene prod 68. Makino, K . , Shinagawa. H . , Amemura,
ucts in controlling flagellar rotation in M . , Kimura, S . , Nakata, A. 1 98 8 .
bacterial chemotaxis of Escherichia coli. Regulation of the phosphate regulon o f
J. Bacterial. 1 69 : 1 307- 1 4 Escherichia coli: activation of pstS
57. Kustu; S . , Santero, E . , Popham, D . , transcription by PhoB protein in vitro. 1.
Keener, J . 1 989. Expression o f sigmaS4 Mol. Bioi. 203:85-95
(ntrA)-dependent genes is probably 69. Makino, K . , Shinagawa, H . , Amemura,
united by a common mechanism. Micro M . . Nakata, A . 1 986. Nucleotide se

bial. Rev. In press quence of the phoR gene, a regulatory
5 8 . Leroux, B . , Yanofsky, M. F . , Winans, gene for the phosphate regulon of Es
S. C . , Ward , J. E . , Zeigler, S. F . , et al. cherichia coli. 1. Mol. Bioi . 1 92:549-56

1 98 7 . Characterization of the virA locus 69a. Makino, K . , Shinagawa, H . , Ame
of Agrobacterium tumefaciens: a tran mura, M . , Nakata, A. 1 986. Nucleutide
scriptional regulator and host range de sequence of the phoB gene, the positive
terminant. EMBO 1. 6:849-856 regulatory gene for the phosphate reg
59. Liljestrom, P. 1 986. The EnvZ protein ulon of Escherichia coli K 1 2. 1. Mol.
-
of Salmonella typhimurium LT-2 and Bioi. 1 90:37-44

Escherichia coli K - 1 2 is located in the 70. Makino, K . , Shinagawa, H . , Nakata, A .
cytoplasmic membrane. FEMS Micro 1 984. Cloning and characterization of
bial. Lett. 36: 1 45-50 the alkaline phosphatase positive regula
60. Liljestrom, P . , Laamanen, I . , PaIva, E . tory gene (phoMJ of Escherichia coli.
T. 1 98 8 . Structure and expression of the Mol. Gen. Genet. 1 95 : 3 8 1 -90
ompB operon, the regulatory locus for 71. Matsuyama, S . -I . , Mizuno, T. , Mizush
the outer membrane porin regulon in ima, S. 1 986. Interaction between two
Salmonella typhimurium LT-2. 1. Mol. regulatory proteins in osmoregulatory
Bioi. 20 1 :663-73 expression of ompF and ompC genes in
6 1 . Liljestrom, P . , Luokkamaki, M . , PaIva, Escherichia coli: a novel ompR mutation
E. T . 1 987. Isolation and characteriza suppresses pleiotropic defects caused by
tion of a substitution mutation in the an envZ mutation. 1. Bacterial.
ompR gene of Salmonella Iyphimurium 1 68 : 1 309- 1 4
LT2. 1. Bacterial. 1 69:438-41 72. Matsuyama, S . -I . , Mizushima, S . 1 987.
62. Losick, R . , Youngman, P. , Piggot, P. J . Novel rpoA mutation that interferes with
1 986. Genetics o f endospore formation the function of OmpR and EnvZ, posi
in Bacillus subtilus. Annu. Rev. Genet. tive regulators of the ompF and ompC
20:625-69 genes that code for outer-membrane pro
63. MacFarlane, S. A . , Merrick, M. J . teins in Escherichia coli K 1 2 . 1. Mol.
1 987. Analysis of the Klebsiella Rial. 1 95 : 847-53
pneumoniae ntrB gene by site-directed 73. Miller, J. F . , Mekalanos, J. J, Falkow,
in vitro mutagensis. Mol. Microbial. S . 1 989. Coordinate regulation and sen
1 : 1 33-42 sory transduction in the control of bacte
64. MacNeil , T . , MacNeil, D . , Tyler, B . rial virulence. Science 243 : 9 1 6-22
1982. Fine-structure deletion map and 74. Miller, S. I . , Kukral, A. M . , Mekala
complementation analysis of the glnA nos, J. J . 1 989. A two-component reg
glnL-glnG region in Escherichia coli. 1. ulatory system (phoP-phoQJ controls
Bacterial. 1 50: 1 302- 1 3 Salmonella typhimurium virulence.
6 5 . Magasanik, B . 1 982. Genetic control of Proc. Natl. Acad. Sci. USA 86:5054-58
nitrogen assimilation in bacteria. Annu. 75. Miller, V. L . , Taylor, R. K . , Mekala
Rev. Genet. 1 6: 1 35-68 nos, J. J. 1 987. Chulera toxin transcrip
66. Magasanik, B. 1 98 8 . Reversible phos tional activator ToxR is a transmem
phorylation of an enhancer binding pro brane DNA binding protein. Cell
tein regulates the transcription of bacte 48:27 1-79
rial nitrogen utilization genes. Trends 76. Mizuno, T . , Kato, M . , Jo, Y . -L . , Mi
Biochem. Sci. J 3:475-79 zushima, S . 1 98 8 . Interaction of OmpR,
67. Magasanik, B . , Neidhardt, F . C . 1 987 . a positive regulator, with the osmo-
regulated ompC and ompF genes of Es coli components is facilitated b y enhanc
cherichia coli. Studies with wild-type ers. C ell 50: 1 039-46
and mutant OmpR proteins. 1. Bioi. 86. Nixon, B . T . , Ronson, C. W . , Ausubel,
Chem . 263 : 1008-1 2 F. M. 1986. Two-component regulatory
77 . Mizuno, T . , Mizush.ima, S . 1987. Isola systems responsive to environmental
tion and characterization of deletion stimuli share strongly conserved do
mutants of ompR and envZ, regulatory mains with the nitrogen assimilation reg
genes for expression of the outer mem ulatory genes nt rB and nt rC. P rac. Natl.
brane proteins OmpC and OmpF in Es Acad. Sci. USA 83:7850--54
ch erichia coli. 1. Biochem . 1 0 1 : 387-96 87. Norioka, S . , Ramakrishnan, G . , Ikena
78. Morett, E . , Cannon, W . , Buck, M . ka, K . , Inouye, M. 1 986. Interaction of
1988. The DNA-binding domain o f the a transcriptional activator, OmpR, with
transcriptional activator protein NifA re reciprocally osmoregulated genes, ompF
sides in its carboxy terminus, recognises and ompC, of E scherichia coli. 1. Bioi .
the upstream activator sequences of nif Chern. 261 : 1 7 1 1 3- 1 9

promoters and can b<! separated from the 88. Oosawa, K . , Hess, J . F. , Simon, M . I .
positive control function of NifA. Nucle 1988. Mutants defective i n bacterial che
ic Acids R es. 1 6: 1 l469-88 motaxis show modified protein phos
79. Mutoh, N . , Simon, M. I. 1986. Nucleo phorylation. Cell 53:89-96
tide sequence corresponding to five che 89. Parkinson, J. S. 1988. Protein
motaxis genes in Escherichia cali. 1. phosphorylation in bacterial chemotaxis.
Bacte rial. 1 65 : 1 6 1-66 Cell 53: 1-2
80. Nakata A . , Amemura, M . , Makino, K . , 90. Popham, D. L . , Szeto, D . , Keener, J . ,
Shinagawa, H . 1987. Genetic and Kustu, S . 1 989. Function of a bacterial
biochemical analysis of the phosphate activator protein that binds to transcrip
specific transport system in Esche richia tional enhancers . Science 243:629-35
coli. In Phosphat., Metabolism and 91. Reitzer, L J . , Magasanik. B. 1985. Ex
C ellula r Regulatio n i n Microo rga nisms, pression of gInA in Esch erichia coli is
ed. A. Torriani-Gorini, F. G. Rothman, regulated at tandem promoters . P rac.
S . Silver, A. Wright, E. Yagil, pp. 150-- Natl. Acad. Sci. USA 82:1979-83
5 5 . Washington, DC: Am. Soc. Micro 92. Ronson, C. W . , Astwood. P. M . , Nix
bioI. on, B. T . , Ausubel, F. M. 1987. De
81. Nara, F. , Matsuyarna, S . -I. , Mizuno, duced products of C4-dicarboxylate
T . , Mizushima, S. 1986. Molecular transport regulatory genes of Rhizobium
analysis of mutant ompR genes exhibit legumi nosa rum are homologous to nit
ing different phenotypes as to rogen regulatory gene products. Nucl eic
osmoregulation of the ompF and ompC Acids Res. 1 5 :7921-34
genes of Esch erichia coli. Mol. G en. 93 . Ronson, C . W . , Nixon, B. T . , Albright,
G enet. 202: 1 94-99 L. M . , Ausubel, F. M. 1987. Rhizobium
82. Ninfa, A. J . , Magasanik, B . 1986. Co m eliloti nt rA ( rpoN) gene is required for
valent modification of the glnG product, diverse metabolic functions. 1. Bacteri
NR" by the gl nL product, NR", reg al. 1 69:2424-3 1
ulates the transcription of the ginALG 94. Ronson, C. W . , Nixon, B. T . , Ausubel.
operon in Esch erichia coli. P roc. Natl. F . M. 1987. Conserved domains in
A ca d. Sci. USA 83 :5909- 1 3 bacterial regulatory proteins that respond
83 . Ninfa, A . J . , Mullin, D. A . , Ramakrish to environmental stimuli. Cell 49:579-
nan, Go', Newton, A. 1 989. E scherichia 81
coli 0-'4 RNA polymerase recognizes 95. Sankar, P . , Shanmugam, K . T . 1 988.
Caulobacte r c resc e ntus flbG and flaN Biochemical and genetic analysis of hy
flagellar gene promoters in vitro. 1. Bac drogen metabolism in Esche richia coli:
terial. 1 7 1 :383-9 1 the h ydB gene. 1. Bacterial. 1 70:5433-
84. Ninfa, A. J . , Ninfa, E. G . , Lupas, A . 39
N . , Stock, A . , Magasanik, B . , e t al. 96. Seki, T., Yoshikawa, H., Takahashi,
1988. Crosstalk between bacterial che H . , Saito, H. 1 987. Cloning and nucleo
motaxis signal transduction proteins and tide sequence of phoP. the regulatory
regulators of transcription of the Ntr reg gene for alkaline phosphatase and phos
ulon: evidence that nitrogen assimilation phodiesterase in Bacillus subtilis. 1.
and chemotaxis are controlled by a com Bact erial. 169:2913-16
mon phosphotransfer mechanism. Proc. 97. Seki, T., Yoshikawa, H., Takahashi ,
Natl. Acad. Sci. USA 85 :5492-96 H . , Saito, H. 1988. Nucleotide sequence
85 . Ninfa, A. J . , Reitzf:r, L. J . , Magasanik, of the Bacillus subtilis phoR gene. 1.
B. 1987. Initiation of transcription at the Bacterial. 1 70:5935-38
bacterial glnAp2 promoter by purified E. 98. Sharma, S . , Stark, T. F. , Beattie. W .
G . , Moses, R . E. 1986. Multiple control dimensional structure of CheY , the re

elements for the uvrC gene unit of Es sponse regulator of bacterial chemotax
cherichia coli . Nucleic Acids Res. 14: is. Nature 337 :745-49
230 1 - 1 8 1 1 2. Stock, J. B . , Waygood, E. B . , Meadow,
9 9 . Shinagawa, H . , Makino, K . , Amemura, N. D . , Postma, P. W . , Roseman, S .
M . , Nakata, A. 1 987. Structure and 1 982. Sugar transport by the bacterial
function of the regulatory genes for the phosphotransferase system. J. Bioi.
phosphate regulon in Escherichia coli. Chern. 257 : 14543-52
See Ref. 80, pp. 2G-25 1 1 3 . Tanaka, T . , Kawata, M. 1 988. Cloning
100. Silverman, P. M. 1 986. Host cell and characterization of Bacillus subtilis
plasmid interactions in the expression of iep , which has positive and negative
DNA donor activity by F+ strains of effects on production of extracellular
Escherichia coli K- 1 2 . Bioessays 2:254- proteases. J. Bacteriol. 170:3593-600
59 1 14. Tate, S . -I . , Kato, M . , Nishimura, Y .
1 0 1 . Simms, S. A . , Keane, M. G. , Stock, J. Arata, Y . , Mizuno, T . 1988. Location
1 98 5. Multiple forms o f the CheB of DNA-binding segment of a positive

methylesterase in bacterial sensing. J. regulator, OmpR, involved in activation
Bioi. Chem. 260: 1 0 1 6 1 -68 of the ompF and ompC genes of Es
1 02 . Siauch, J. M . , Garrett, S . , Jackson, D. cherichia coli. FEBS Lett. 242:27-30
E., Silhavy, T. J . 1 988. EnvZ functions 1 1 5 . Taylor, R. K . , Ha1l, M. N . , Silhavy, T.
through OmpR to control porin gene ex J . 1983. Isolation and characterization of
pression in Escherichia coli K 1 2 J.
- . mutations altering expression of the ma
Bacteriol. 1 70:439-4 1 jor outer membrane porin proteins using
103. Smith, J. M . , Rowsell, E. H . , Shioi, J . , the local anaesthetic procaine. J. Mol.
Taylor, B . L. 1988. Identification of a BioI. 166:273-82
site of ATP requirement for signal pro 1 1 5a. Timme, T. L. , Lawrence, C . B . ,
cessing in bacterial chemotaxis. J. Bac Moses, R . E . 1989. Two new members
teriol. 170:2698-704 of the OmpR superfamily detected by
104. Smith, R. A . , Parkinson, J. S. 1 980. homology to a sensor-binding core do
Overlapping genes at the cheA locus of main. J. Mol. Evol. 28:545-52
Escherichia coli. Proc. Natl. Acad. Sci. 1 1 6. Torriani-Gorini. 1987. The birth and
USA 77:537G-74 growth of the pho regulon. See Ref 80,
1 05 . Springer, M. S . , Zanolari , B . 1984. pp. 3-1 1
Sensory transduction in Escherichia 1 1 7 . Trach, K . , Chapman, J. W . , Piggot, P .
coli: regulation of the demethylation by J . , Hoch, J. A . 1 985. Deduced product
the CheA protein. Proc. Natl. Acad. Sci. of the stage 0 sporulation gene SpoOF
USA 8 1 :5061-65 shares homology with the SpoOA,
106. Stachel, S. E . , Zambryski, P. C. 1 986. OmpR, and SfrA proteins. Proc. Natl.
virA and virG control the plant-induced Acad. Sci. USA 82:7260--64
activation of the T-DNA transfer process 1 1 8 . Virts, E. L . , Stanfield, S. W . , Helinski ,
of A . tumefaciens. Cell 46:325-33 D. R . , Ditta, G. S. 1988. Common reg
1 07 . Stewart, R. C . , Dahlquist, F. W. 1987. ulatory elements contrul symbiotic and
Molecular components of bacterial che microaerobic induction of nifA in Rhizo
motaxis. Chem. Rev. 87:997-1 025 bium me/iloti. Proc. Natl. Acad. Sci.
1 08. Stewart, R. C . , Dahlquist, F. W. 1988. USA 85:3063-65
N-terminal half of CheB is involved in 1 1 9. Wanner , B . L. 1 987. Phosphate regula
methylesterase response to negative tion of gene expression in Escherichia
chemotactic stimuli in Escherichia coli. coli. See Ref. 67, pp. 1326-33
J. Bacteriol. 1 70:5728-38 120. Wanner, B . L. 1987. Control of phoR
109. Stibitz, S . , Aaronson, W . , Monack, D . , dependent bacterial alkaline phosphatase
Falkow, S. 1989. Phase variation in clonal variation by the phoM region. J.
Bordetella pertussis by frameshift muta Bacteriol. 169:90G-3
tion in a gene for a novel two-component 1 2 1 . Wanner, B . L . , Wilmes, M . R . , Hunter,
system. Nature 338:266-69 E. 1 988. Molecular cloning of the wild·
1 10. Stock, A . , Chen, T . , Welsh, D . , Stock, type phoM operon in Escherichia coli
J. 1988. CheA protein, a central regula K-12. J. Bacteriol. 170:279-88
tor of bacterial chemotaxis, belongs to a 122. Wanner, B . L . , Wilmes, M . R . , Young,
family of proteins that control gene ex D. C. 1988. Control of bacterial alkaline
pression in response to changing en phosphatase synthesis and variation in
vironmental conditions. Proc. Nail. an Escherichia coli K- 1 2 phoR mutant
Acad. Sci. USA 8 5 : 1403-7 by adenyl cyclase, the cyclic AMP re
1 1 1 . Stock, A. M . , Mottonen, J. M . , Stock, ceptor protein. and the phoM operon . J.
J. B . , Schutt, C. E. 1 989. Three- Bacteriol. 170: 1092-102
123. Weglenski, P . , Ninfa, A . , Ueno-Hishio, J . E . , Nester, E. W . 1989 . A protein

S . , Magasanik, B . 1989 . Mutations in required for transcriptional regulation of
the glnG gene of Escherichia cali that Agrabacterium virulence genes spans
result in increased activity of nitrogen the cytoplasmic membrane. 1. Bacterial.
regulator 1 . 1. Bacterial. 1 71:4479-85 17 1 : 1 616-22
124. Weiss, Y . , Magasanik, B . 1988. 129. Wolfe, A. J . , Conley, M. P . , Kramer,
Phosphorylation of nitrogen regulator I T. J . , Berg, H. B . 1987. Reconstitution
(NR,) of Escherichia cali. Prac. Natl. of signaling in bacterial chemotaxis. 1.
Acad. Sci. USA 85: 8919-23 Bacterial. 1 69:1878-85
125. Weston , L. A . , Kadner, R. J. 1 98 7 . 130. Wylie, D . , Stock, A . , Wong, c . -Y . ,
Identification of U h p polypeptides and Stock, J. 1988. Sensory transduction i n
evidence for their role in exogenous in bacterial chemotaxis involves phospho
duction of the sugar phosphate transport transfer between che proteins. Biachem .
system of Escherichia coli K-1 2 . 1. Bac Biaphys. Res. Camm. 151:891-96
terial. 169:3546-55 13 1 . Yang, Y . -L . , Goldrick, D . , Hong, J .-S .

126. Weston , L. A . , Kadner, R. J. 1988. 1988. Identification of the products and

Role of uhp genes in expression of the nucleotide sequences of two regulatory
Escherichia coli sugar-phosphate trans genes involved in the exogenous induc
port system. 1. Bacterial. 1 70:3375- tion of phosphoglycerate transport in
83 Salmonella typhimurium. 1. Bacterial.
127. Winans, S. c . , Ebert, P. R . , Stache l , S . 1 70:4299-303
E . , Gordon, M . P . , Nester, E. W. 1986. 1 32 . Yu , G . -Q. , Hong, J . - S . 1986. Identifi
A gene essential for Agrabacterium viru cation and nucleotide sequence of the
lence is homologous to a family of posi activator gene of the externally induced
tive regulatory loci. Proc. Natl. Acad. phosphoglycerate transport system of
Sci. USA 83:8278-82 Salmonella typhimurium. Gene 45 :51-
128. Winans, S. C . , Kerstetter, R. A . , Ward , 57

Prokaryotic Signal Transduction PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Prokaryotic Signal Transduction PDF

Uploaded by

Copyright:

Available Formats

ANNUAL

Annu. Rev. Genet. 1989. 23:311-36

SENSOR AND REGULATOR

Lisa M. Albright, Eva Huala, and Frederick M. Ausubel

INTRODUCTION . . . .... . . . ... . . . . 312

In bacteria, signal transduction in response to a wide variety of environmental

effector) pairs belong to a large superfamily of signal-transducing regulatory

E < C pxA 225 FN MVTALERMMTSQ

s.� �;��� 61 :: i Li9iETT�LMENLqDE�R G�"iliLNJ!JD i i NLFIQETKD iMQE:

KpNtrB 225 LGPECGEITLRTRTAFQ

KpNtrB 278 ... RE�

Ec G lucose-6-phosphate UhpBbC UhpA uhpT 27, 125, 126

CheB Regulator Subclass (methylesterase):

HOMOLOGY BETWEEN SIGNAL TRANSDUCING

regulatory pairs . Many of the sensor components are predicted to be trans­

Identification of New Signal Transducing Regulatory Pairs

number of other pairs of proteins, such as PtsG/Crr , which interact in the

CONSERVED INTERACTION BETWEEN SENSOR AND

The conservation of protein domains in the signal-transducing regulatory pairs

The NtrBINtrC System

correlates with its ability to act as a transcriptional activator in vitro; efficient

transmitted to PH by a uridylyltransferase/uridylyl-removing enzyme (UTase­

The CheAICheY + CheB System

Ius. Addition of repellent or removal of attractant increases the activity of the

correlation between mutants defective in chemotaxis and those altered either

The EnvZ/OmpR System

The PhoRIPhoB System

The PhoRJPhoB pair regulate expression of the phosphate regulon, which

If phosphorylation of the regulator by the sensor represents a conserved mode

phoR+ background (99 , 1 1 9) . In fact , the ORF directly upstream of phoM in

FUNCTION OF THE SENSOR

The concentration of conserved amino-acid residues in the Codomain of the

Domain Structure of Sensors

functional and an evolutionary point of view. Experimental evidence supports

autophosphorylation occurs at a histidine (124) and that the pH stability of

FUNCTION OF THE REGULATOR

The presence of a conserved N-domain is the distinguishing feature of the

be grouped into subclasses based on homology within their C-domains (Table

F igu re 3 Homology among regulator components. See Fi gu re 2 for other details.

helix-turn-helix DNA-binding motif (19, 92). However, NifA, which acti­

and regulated dephosphorylation with kinetics comparable to that of the entire

intriguing in that it appears to function mainly as a repressor, although a

protein (101). In addition, N-terminal deletions of up to one half of CheB

GENETIC ANALYSIS OF SIGNAL TRANSDUCTION IN

R Ib mutants of phoR, which give rise to phoM- independent constitutive

SUMMARY AND PROSPECTS

The evidence is now quite convincing that a common mechanism of interac­

determined. Nevertheless , the phospho-transfer reaction , which must still be

1. Aiba, H. , Mizuno, T. , Mizushima, S . mechanism and evolution. Annu. R ev.

1 987. Involvement of the n trA gene Spence, J. , Ferrari , E . , et al. 1985.

164: 8 1 6-22 brane topology of EnvZ, a protein in­

1 2 . Buikema, W. J . , Szeto, W. W . , Lem­ volved in osmoregulation of OmpF and

syringae is homologous to a highly con­ from Pseudomonas putida . Gene

sacU (Hy) mutations to two linked genes

cpxA mutant phenotypes of Escherichia

55. Kunst, F . , Debarbouille, M . , Msadek, Regulation of carbon and nitrogen

united by a common mechanism. Micro­ M . . Nakata, A . 1 986. Nucleotide se­

S. C . , Ward , J. E . , Zeigler, S. F . , et al. cherichia coli. 1. Mol. Bioi . 1 92:549-56

of Salmonella typhimurium LT-2 and Bioi. 1 90:37-44

the upstream activator sequences of nif Chern. 261 : 1 7 1 1 3- 1 9

G . , Moses, R . E. 1986. Multiple control dimensional structure of CheY , the re­

1 98 5. Multiple forms o f the CheB of DNA-binding segment of a positive

123. Weglenski, P . , Ninfa, A . , Ueno-Hishio, J . E . , Nester, E. W . 1989 . A protein

s.� �;�� 61 :: i Li9iETT�LMENLqDE�R G�"iliLNJ!JD i i NLFIQETKD iMQE:

regulatory pairs . Many of the sensor components are predicted to be trans

transmitted to PH by a uridylyltransferase/uridylyl-removing enzyme (UTase

helix-turn-helix DNA-binding motif (19, 92). However, NifA, which acti

The evidence is now quite convincing that a common mechanism of interac

164: 8 1 6-22 brane topology of EnvZ, a protein in

1 2 . Buikema, W. J . , Szeto, W. W . , Lem volved in osmoregulation of OmpF and

syringae is homologous to a highly con from Pseudomonas putida . Gene

united by a common mechanism. Micro M . . Nakata, A . 1 986. Nucleotide se

G . , Moses, R . E. 1986. Multiple control dimensional structure of CheY , the re