Functional Modulation of Escherichia Coli RNA Polymerase

P1: FQP/LVF P2: FQP

August 4, 2000 14:36 Annual Reviews AR110-15

Annu. Rev. Microbiol. 2000. 54:499–518Copyright c© 2000 by Annual Reviews. All rights reserved

FUNCTIONAL MODULATION OF

ESCHERICHIA COLI RNA POLYMERASE

Akira IshihamaNational Institute of Genetics, Department of Molecular Genetics, Mishima, Shizuoka411-8540, Japan; e-mail: [email protected]

Key Words transcription apparatus, sigma factor, transcription factor,stationary phase

■ Abstract The promoter recognition specificity ofEscherichia coliRNA poly-merase is modulated by replacement of theσ subunit in the first step and by interactionwith transcription factors in the second step. The overall differentiated state of∼2000molecules of the RNA polymerase in a single cell can be estimated after measurementof both the intracellular concentrations and the RNA polymerase-binding affinities forall seven species of theσ subunit and 100–150 transcription factors. The anticipatedimpact from this line of systematic approach is that the prediction of the expressionhierarchy of∼4000 genes on theE. coli genome can be estimated.

CONTENTS

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 499MODULATION OF RNA POLYMERASE SPECIFICITY BY

REPLACEMENT OF THEσ SUBUNIT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500SIGMA REPLACEMENT DURING GROWTH TRANSITION FROM

EXPONENTIAL TO STATIONARY PHASE . . . . . . . . . . . . . . . . . . . . . . . . . . . 502LEVEL AND ACTIVITY CONTROL OF SIGMA-S . . . . . . . . . . . . . . . . . . . . . . . 503ACTIVITY CONTROL OF THE SIGMA SUBUNITS BY ANTI-SIGMA FACTORS 507MODULATION OF RNA POLYMERASE SPECIFICITY BY

TRANSCRIPTION FACTORS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508FACTORS AFFECTING SELECTIVE UTILIZATION OF

TRANSCRIPTION APPARATUS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 509MULTIPLE PATHWAYS FOR STATIONARY-PHASE ADAPTATION. . . . . . . . . . 510CONCLUSION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511

INTRODUCTION

The RNA polymerase ofEscherichia coliis composed of the core enzyme (sub-unit composition,α2ββ ’) with the catalytic activity of RNA polymerization, andone of the seven different species ofσ subunit, each responsible for recognition

0066-4227/00/1001-0499$14.00 499

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


500 ISHIHAMA

of a specific set of promoters (28, 29, 31, 40). The total number of core enzymemolecules in a growingE. coli cell is∼2000 (38, 43), which is less than the totalnumber of genes (∼4000) on theE. coli genome (7). Together these findings ac-centuate the importance of RNA polymerase to choose which genes to transcribeand how often (39, 40). Because replacement of theσ subunit is the most efficientway to alter the promoter recognition properties of the transcription apparatus,the replacement of oneσ species of core enzyme-associatedσ subunit by anotheris believed to be the major mechanism for switching of the transcription pattern.Along this line, competition between availableσ subunits should be a key deter-minant of which group of genes is transcribed (22, 108; H Maeda, N Fujita, AIshihama, submitted for publication). To test theσ competition model, quanti-tative and comparative measurements of the intracellular concentrations and thecore enzyme-binding affinities of all sevenσ subunits are absolutely required.

The holoenzyme alone is able to recognize the simple promoters of genes thatare constitutively expressed and to initiate transcription at constant rates. Most ofthe genes in bacteria are, however, subject to regulation in response to changes inenvironmental conditions. Thus, for transcription of the majority ofE. coli genes,one or more additional accessory factors are required (39, 40). Among the samegroup of genes under the control of a singleσ species, the order of transcriptionlevel is therefore determined by the promoter strength and the initiation efficiency,assisted or inhibited by transcription factors. From the genome sequence ofE. coli,the total number of DNA-binding proteins that more or less influence transcriptioncan be estimated to be∼240–260 (85; N Fujita, unpublished data). Most of theseDNA-binding proteins, which either activate or repress transcription of specificgenes, interact directly with the RNA polymerase and modulate its specificityof transcription initiation (and elongation in some cases). The known transcrip-tion factors can be classified into groups by their structure and mode of function(40–42). Overall, therefore, the transcription specificity of RNA polymerase coreenzyme is modulated in two steps: by molecular interaction with theσ factors inthe first step and with the transcription factors, usually DNA bound, in the sec-ond step. The expression hierarchy of the∼4000 genes on theE. coli genomemust be determined to a great extent by the relative levels of transcription appara-tus composed of core enzymes combined with differentσ subunits and differenttranscription factors. This review summarizes our up-to-date knowledge of suchfunctional differentiation of RNA polymerase inE. coli, focusing on its alterationduring the growth phase transitions ofE. coli cultures.

MODULATION OF RNA POLYMERASE SPECIFICITY BYREPLACEMENT OF THE σ SUBUNIT

Seven different species ofσ subunits,σ 70, σN (also calledσ 54), σS(σ 38), σH (σ 32),σ F (σ 28), σE (σ 24), andσ FecI, have been identified inE. coli (28, 29, 31, 40), eachparticipating in transcription of a specific set of genes (Figure 1, see color insert).

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


RNA POLYMERASE FUNCTIONAL MODULATION 501

Most of the housekeeping genes expressed during exponential-phase growth aretranscribed by the holoenzyme containingσ 70 (therpoD gene product), while theholoenzyme EσS is essential for transcription of some stationary-phase specificgenes (32, 65). Stress response genes are transcribed by RNA polymerase holoen-zymes containing alternative minorσ subunits. The holoenzyme EσN transcribesthose genes, which are activated by a deficiency of nitrogen (70) and some otherstress response genes (90); the holoenzyme EσH transcribes the genes for heatshock proteins (103); Eσ F is needed for expression of the third wave of flagellarand chemotaxis genes (30); the holoenzyme EσE is responsible for transcriptionof genes whose products deal with protein defects such as misfolded proteins inthe periplasm, caused, for example, by heat shock (20, 81, 82, 87); and thefecIgene product, which was originally identified as a positive regulatory gene for theferric citrate transport system (80), is now known to be a member of the extra-cytoplasmic function subfamily ofσ factors (3, 67) based on its protein sequence(hereafter referred to asσ FecI) and is involved in transcriptional activation of thefecoperon (19, 25).

The intracellular concentration of RNA polymerase in the steady state of grow-ing E. coli W3350 cells is maintained at a constant level that is characteristic ofthe rate of cell growth (38, 43). In a rich medium, the total number of core en-zyme molecules is∼2000 per genome equivalent of DNA, among which aboutone third are disengaged from the DNA. Because RNA chain elongation is cat-alyzed by core enzyme without an associatedσ subunit, the combined number(1200 molecules) of all sevenσ subunits (46, 50) is more than the total number(600–700 molecules) of core enzyme molecules that are not involved in tran-scription and thus are available for bindingσ (Figure 2, see color insert). Themajority of free RNA polymerase molecules in the cytosol should therefore bein the holoenzyme form, associated with one of theσ subunits. These findingssupport the model that competition takes place between theσ subunits for bind-ing a limited supply of core enzyme. To estimate the relative levels of differentholoenzyme forms, two parameters must be determined: the intracellular concen-trations of all sevenσ subunits and the binding affinity of core enzyme to eachσ

subunit.The concentration of eachσ subunit is subject to variation depending on the cell

growth conditions, although the concentration of core enzyme stays constant at alevel that is characteristic of the rate of cell growth (38, 43). The first systematicdetermination of the intracellular concentrations of all sevenσ subunits has beenperformed for the laboratory strain W3110 type-A, which contains the intact formsof bothσS andσ F (47). The results indicate that the intracellular concentration ishighest for theσ 70subunit in both exponential and stationary phases and under vari-ous stress conditions (46, 50, 68). In exponential-phase cells, two of the alternativeσ subunits,σN andσ F, are present in significant concentrations, but, in addition,the level ofσS becomes detectable in stationary phase (Figure 1, see color insert).

The core-binding affinities have been determined in vitro by measuring theamount of core enzyme-boundσ subunit in the presence of various amounts of

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


502 ISHIHAMA

eachσ subunit and a fixed amount of core enzyme (H Maeda, N Fujita, A Ishi-hama submitted for publication). Among the sevenσ subunits fromE. coli, σ 70

was found to have the highest affinity to the core enzyme. The affinities of the othersix σ subunits ranged downwards by 16-fold fromσ 70 (0.26 nM) toσS, which hasthe weakest binding activity (4.28 nM) (Figure 2, see color insert). From thesetwo lines of experimental data, we can now estimate the intracellular concentra-tion of each holoenzyme. The numbers of each free holoenzyme formed in anE. coli cell during exponential growth are thus calculated to be 550 moleculesof Eσ 70, 95 molecules of Eσ F, and 55 molecules of EσN (Figure 2, see colorinsert).

SIGMA REPLACEMENT DURING GROWTH TRANSITIONFROM EXPONENTIAL TO STATIONARY PHASE

Bacterial populations in nature often exist in the stationary phase or a state ofpartial or complete starvation. The term stationary phase is used to denote a fixedphysiological state regardless of what factors led to growth cessation. The sta-tionary phase is synonymous with the starvation for only an ideal case in whichthe limiting nutrient leading to growth cessation can be specified, but even inlaboratory culture conditions the mechanism of cell growth cessation usually in-volves multiple factors. Bacteria are capable of sensing the maximum cell density(or “quorum”) and then can grow under a given condition and communicate thisto other members of the same species by production of extracellular signalingmolecules. This allows the single-celled prokaryotes to function in some respectsas if they are multicellular organisms (55, 86). After entry into stationary phase,a number of genes, that are not expressed during exponential phase are expressedin a sequential manner (reviewed in 41, 42).N-Acyl homoserine lactones are onegroup of the diffusible extracellular quorum-sensing signaling molecules (bacte-rial pheromones) that are used for cell-cell communication in some bacteria (88).The synthesis ofN-acyl homoserine lactones also arises as a natural response tostarvation and entry into stationary phase forE. coli (37), but the role of thesemolecules has not yet been established for this species. A family of diketopiper-azines such as cyclo(1Ala-Val) and cyclo(Pro-Tyr) appear to function as signalmolecules in the quorum-sensing systems in certain bacteria (35). Upon entry intostationary phase, a bacterial culture can divide into two populations, one enteringinto dormant phase (sporulation for gram-positive bacteria) and the other into pro-grammed cell death (or prokaryotic apoptosis) (34, 54). The choice between thesefates seems to be under genetic control (e.g. see 101). Mutations in therpoSgeneoften confer growth advantage for stationary-phase cultures (47, 104). Phages canexpress anti-cell death functions to favor their replication in infected cells (18).

The total number of detectably expressed genes among the>4000 putativegenes on theE. coli genome is<1000 in exponentially growing cells, as esti-mated from the protein patterns on two-dimensional gel electrophoresis (96). The

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP



accumulation of extracellular quorum-sensing signals ultimately leads to repres-sion of most of these growth-related genes and, instead, induces the expressionof a new set of genes that are specific for survival in the stationary phase. Up tonow,∼100 genes have been identified that are expressed only in stationary phase(27, 41, 42, 64, 65). The precise selection of genes expressed in stationary phase,however, differs depending on the factors that led to cessation of cell growth. Thisdrastic change in gene expression pattern is accompanied by a change in activityand specificity of both the transcription apparatus and the translation machinery(reviewed in 40, 41). The most significant changes are preceded by the appearanceof the stationary-phase-specificσS subunit (for reviews, see 32, 64, 65), which isinvolved in transcription of most, if not all, of the stationary-phase genes (Table 1).

The σS subunit is maximally expressed at the onset of stationary phase, themaximum level being∼30% that ofσ 70, or 230 molecules per cell (46, 50; Figure2, see color insert). The onset ofσSsynthesis is signaled by changes in metabolismthat lead to reductions in growth. Even during the exponential-growth phase,therefore, the synthesis ofσS can be induced when cells are exposed to conditionsthat are unfavorable for growth, so that, at very slow growth rates,E. colicontainsσS even in exponential phase.

LEVEL AND ACTIVITY CONTROL OF SIGMA-S

The synthesis and accumulation ofσS are controlled at multiple levels, includ-ing transcription, translation, protein turnover, and activity control (Figure 3, seecolor insert). Transcription control ofrpoSinvolves a number of factors, includ-ing ppGpp and polyphosphate as positive regulators and cyclic AMP (cAMP) andUDP-glucose as negative regulators (reviewed in 64). Upon increase in the con-centration of the stringent control signal ppGpp, transcription ofrpoSis enhanced(24, 62). ppGpp binds to theβ subunit of RNA polymerase (13) and thereby in-hibits transcription of stringently controlled genes. However, as for other genesunder the positive control of ppGpp such as those for the amino acid biosyntheticoperons, the molecular mechanism of transcription activation by ppGpp remainsunsolved. The mechanism(s) by whichrpoS transcription is stimulated by de-creases in the concentrations of the catabolite repression signal cAMP (32, 64) orUDP-glucose (8) also remain unknown.

Translation ofrpoSmessenger RNA (mRNA) is stimulated under various stressconditions by several regulatory factors, including the RNA-binding Hfq (HF-1)protein (10, 11, 73) and the small regulatory DsrA-RNA (63), and is repressed bythe histonelike protein H-NS (52). These factors modulate the secondary structureof the ribosome-binding region ofrpoSmRNA. Hfq was originally identified as ahost factor (HF-1) for phage Qβ replication (23). After gene cloning (51), it wasidentified as one of the most abundantE. coliproteins associated with both nucleoidand ribosomes (92, 93). Several lines of evidence indicate that Hfq binds a set ofmRNAs includingrpoSmRNA and regulates the efficiency of their translation by

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


504 ISHIHAMA

TABLE 1 Stationary-phase–specific genes under the control of σ S

Gene Gene function Reference(s)

Genes involved in cell morphology and divisionbolA Control of PBP6 synthesis 64, 65cfa Cyclopropane fatty acid synthase 64, 65csgBA Curly fimbriae formation 64, 65csgCDEF Curly fimbriae formation 64, 65fic Cell division control 64, 65ftsQAZ Septum formation 65gabP -Aminoburytic acid permease 88ahdeAB 65htrE Pili construction protein 64, 65osmB Outer membrane lipoprotein 64, 65osmC Outer membrane lipoprotein 9aosmE Outer membrane lipoprotein 14aosmY Periplasmic protein 64, 65pqi5 Membrane protein 65potF (o381) Periplasmic putrescine-binding protein 88aproU Glycine betaine and proline transport 83proP Glycine betaine and proline permease 65vanABCDF D-Ala-D-ala dipeptide transport 63avanX D-Ala-D-ala dipeptidase 63ayhiU Two-membrane drug efflux pump 89a

Genes involved in energy metabolism and anabolismacnA Aconitase 15aacs Acetyl-CoA synthetase 89aaldB Aldehyde dehydrogenase 65cbdAB Cytochrome bd-II oxidase 65cyxAB Cytochrome oxidase III 64, 65frd Fumarate reductase 65galEKT Galactose use 65glpD Glyacerol-3-phosphate dehydrogenase 65hmp Flavohemoglobin Hmp 65hyaABCDEF Hydrogenase I (hydrogen oxidation) 65nrz NRZ nitrate reductase 12co371 Glucose dehydrogenase B homolog 89apoxB Pyruvate oxidase (acetate synthesis) 65tam (o252) trans-Aconitate methyltransferase 12a

Genes involved in protein and nucleic acid breakdownappY Transcription factor for appABC 65clpA Clp protease subunit 65wrbA TrpR repressor binding protein 65xthA Exonuclease III 64, 65

(Continued )

Periplasmic proteinsγ

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP



Genes involved in gene regulation, DNA replication, and nucleoid configurationaidB Methylation damage repair of DNA 64, 65cbpA Curved DNA-binding protein 65dnaN DNA polymerase III β-subunit 96adps (pexB) DNA binding protein 64, 65hip (himD) IHF subunit 65hns (osmZ) H-NS 65rob (cbpB) Curved DNA-binding protein 51arpoS RNA polymerase 91atopA DNA topoisomerase I 65

Genes involved in production of storage productsglgCAF Glycogen synthesis 65glgS Glycogen synthesis 64, 65otsA (pexA) Trehalose synthesis 64, 65otsB (perX) Trehalose-6-phosphate phosphatase 64, 65

Genes for stress resistanceahpCF Alkyl hydroperoxide reductase 70aecp-thtrE Thermotolerance/osmotolerance 88acpxRA Disaggregation of misfolded proteins 16agadAB Glutamic acid decarboxylase 12b, 15bgor Glutathione oxidoreductase 65katE Catalase HPII 64, 65katG Catalase-peroxidase HPI 64, 65ldc Lysine decarboxylase 51boxyR Transcription factor for oxy regulon 70apcm L-Isoaspartyl protein methyltransferase 96bpspABCE Phage shock proteins 65sodC Periplasmic superoxide dismutase 25asprA Virulence to animals 29atreA (osmA) Periplasmic trehalase 64, 65treF Cytoplasmic trehalase 35ausp Universal stress proteins 21a

Unknown functionscsiDEF Carbon starvation-induced proteins 64, 65f186 89af253a 89ao215 89apexCDEF Carbon starvation-induced proteins 65yciG 65yohF 65

TABLE 1 (Continued )

Gene Gene function Reference (s)

σ S

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


506 ISHIHAMA

modulating the secondary structure of as-yet-unidentified sites near the translationinitiation codon (10, 11, 74). The regulatory RNA encoded byoxySis known tocompete for Hfq and, as a result, inhibits translation of therpoSmRNA (106).

Structural modification ofrpoS mRNA can also be mediated through directRNA-RNA interaction with DsrA RNA (63, 91). DsrA, an 87-nucleotide untrans-lated RNA, acts as a positive riboregulator of RpoS synthesis (91), by enhancingtranslation ofrpoSmRNA through RNA:RNA interactions withrpoSmRNA (63).DsrA is required for the optimal translation ofrpoSmRNA with the help of Hfq(11, 73, 91), and, accordingly,dsrA mutants fail to accumulateσS in stationaryphase. The product ofdksA(dnaK suppressor) is also required for the optimaltranslation ofrpoS mRNA, but the region ofrpoS mRNA required to see theeffects ofdsrAmutations was identified onrpoSbetween codons 8 and 73 (98),suggesting that the contact sequence of DksA is located downstream of the trans-lation initiation site onrpoSmRNA. The histone-like protein H-NS plays a dualrole by interfering with both transcription and translation ofrpoS(52, 102). Thetarget of H-NS binding on therpoSmRNA is located upstream from the initia-tion codon. Thus, DsrA RNA antagonizes the H-NS–mediated inhibition ofrpoSmRNA translation. LeuO, however, repressesdsrAexpression and thereby reducesrpoStranslation at low temperature (52).

TheσS protein is subject to rapid turnover in exponential-phaseE. coli cells.The increase inσS level in stationary-phaseE. coli results, at least in part, from alarge increase in the stability ofσSprotein (105). TheclpPgene product is one pro-tease that is responsible for degradation ofσS (89, 98). The target site for the ClpPprotease action apparently resides in a 20-amino-acid stretch near the middle oftheσS protein (89). TheσS factor also becomes stable inrssBmutants because theactivity of ClpP protease is enhanced by RssB [or SprE (MviA forSalmonella ty-phimurium)], a response regulator (73). Phosphorylation ofS. typhimuriumMviA(RssB/SprE forE. coli) is required for this regulation, but no evidence has beenobtained to support the model that acetyl phosphate contributes to the phosphoryla-tion of MviA, although theσSlevel increases about fivefold when acetate is used asa carbon source (15). Rss/SprE contains a unique C-terminal output domain and isthe first known response regulator involved in the control of protein turnover (73).On the other hand, the ClpP protease is activated by the ClpP-specific chaperoneClpX (26, 97, 99) and inhibited by the general chaperone DnaK (65). LrhA, a LysRhomolog, promotes degradation ofσS, indirectly activating the ClpP protease bymodulating the activity of RssB/SprE (25).

The promoter recognition specificity of EσS is not entirely understood becausethe promoters of the stationary-specific genes so far analyzed do not show anydistinctive consensus sequence and are mostly recognized by both Eσ 70 and EσS

holoenzymes in vitro (53, 94, 95). Several lines of evidence indicate that someadditional factors are involved in the control of activity and specificity of thesetwo different forms of RNA polymerase holoenzyme. Under stress conditions,there are marked increases in the intracellular concentrations of compatible solutes,including stress protectants such as potassium glutamate, trehalose, and glycine

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP



betaine, and of some storage products such as glycogen and polyphosphate (e.g. see78). These compounds influence the activity and specificity of RNA polymerase tovarious degrees. The activity ofσSholoenzyme is modulated by glutamate (17, 83)and trehalose (59) at the steps of EσS holoenzyme formation and EσS holoenzymebinding to certain promoters. Transcription by EσS of some osmoregulated genessuch asosmBandosmYtakes place in the presence of≥0.5 M concentrations ofpotassium glutamate, which completely inhibit the activity of Eσ 70 in vitro (17).The optimal concentrations of trehalose for maximal transcription by Eσ 70 andEσS are∼0.5 and 1.0 M, respectively (59).

RNA polymerase from stationary-phase cells ofE. coli is associated with anacidic compound(s) and exhibits an altered promoter selectivity (77). The RNApolymerase-associated acidic compounds were found to be inorganic polyphos-phate (poly P) (60), which is known to accumulate in stationary-phase cells (55).The ubiquitous occurrence of poly P is suggestive of some important physiologicalrole(s) for this polymer, and a number of hypotheses have been proposed for itsfunction (56, 84). Because mutants that are defective in theppk gene encodingpolyphosphate kinase (PPK) are also defective in survival in the stationary phase(60), poly P is now believed to play a role in bacterial adaptation to the stationaryphase. At low salt concentrations, poly P inhibits transcription in vitro by bothEσ 70 and EσS holoenzymes. Upon increasing the concentration of potassium glu-tamate, however, poly P inhibition is relieved for the EσS holoenzyme, but not forEσ 70, suggesting that poly P may play a role in the promoter selectivity controlof RNA polymerase inE. coli growing under high-osmolarity conditions and instationary phase (60).

Thecrl gene stimulates transcription ofcsgBA, the locus encoding the majorsubunits of curli (the surface structures in stationary phase).crl-null alleles influ-ence the expression ofσS-regulated genes in a fashion similar to anrpoS-null allele.Crl stimulates the activity of therpoSregulon by stimulatingσSactivity during sta-tionary phase (79). The mechanism ofσS activity control by Crl is not yet known.

ACTIVITY CONTROL OF THE SIGMA SUBUNITS BYANTI-SIGMA FACTORS

A new frame of transcription regulation has been discovered, in which the activityof RNA polymeraseσ subunit is controlled by so-called anti-σ factors (for areview, see 36). An anti-σ factor is defined by the ability to form a complexwith its cognateσ subunit and thereby inhibits theσ function. This definitionexcludes otherσ subunits that compete for available core enzyme. The control ofσ activities by anti-σ factors has been well established inBacillus subtilis(12, 57).Among sevenE. coli σ subunits, anti-σ factors have been identified forσ 70, σ F,andσE (cited in 48).

The fliA gene, one of the class II genes within the transcription hierarchy ofthe flagellar biosynthetic pathway, encodes the flagella-specific RNA polymerase

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


508 ISHIHAMA

σ subunitσ F, which is responsible for transcription of class III genes (30). Anegative regulatory gene,flgM, for class III genes encodes anti-σ F factor, whichinhibits σ F activity (61). FlgM exists as a binary complex withσ F in the earlyphase of flagella formation, but it is secreted out of the cell after completion ofthe flagella hook-basal body, resulting in reactivation of theσ F subunit. Thecontact site for FlgM on theσ F subunit is located within region 4 (45), as forclass II transcription factors (28, 40). Under the steady state ofE. coli growth, alarge fraction of theσ F subunit stays as a complex with FlgM (48), and thus theactual concentration of Eσ F holoenzyme must be considerably lower than theσ F

level.The first member of the extracytoplasmic function subfamily ofσ subunit iden-

tified in E. coli is σE, which was originally identified as a factor required forsurvival on exposure to extremely high temperature (19). Increasing the amountsof outer membrane proteins increasesσE-dependent gene expression, whereas re-ducing outer membrane proteins results in a decrease inσE-dependent transcription(69). Overexpression of thedsbgenes, coding for proteins involved in disulfidebond formation, and thesurAandfkpAgenes, coding for distinct peptidyl-prolylisomerases, can compensate for and/or complement some ofσE mutations (71),indicating thatσE is required for expression of the repair systems of denaturedproteins under certain stress conditions. The product ofrseA, located downstreamof therpoE gene within the same operon negatively regulatesσE (16, 72). The Nterminus of RseA interacts directly withσE.

After screening for proteins that are associated with theσ 70 subunit, Jishage& Ishihama (48) isolated Rsd (regulator of sigma D). Isolated Rsd forms a binarycomplex with purifiedσ 70 and thereby inhibits its function. It was thereforeproposed that Rsd is an anti-σ 70factor. Rsd is not present in exponentially growingE. coli cells (48, 49), and in these conditions most of the free RNA polymerase(not engaged in RNA synthesis) must be in the Eσ 70 holoenzyme form, which canbe immediately used for transcription of the growth-related genes. On entry intostationary phase, Rsd starts to be synthesized and reaches its maximum level inthe early stationary phase (49). As in the case of FglM-σ F interaction, the contacttarget for Rsd onσ 70 is located within region 4. Together, these observationsindicate that, in the stationary-phaseE. coli, some of the unusedσ 70 subunits arestored in inactive form as a complex with Rsd, the anti-σ 70 factor.

MODULATION OF RNA POLYMERASE SPECIFICITY BYTRANSCRIPTION FACTORS

Each holoenzyme recognizes and transcribes a different set of genes, but tran-scription of some genes requires, in addition, accessory proteins or nucleotidefactors. The functional specificity of RNA polymerase holoenzyme is modulatedthrough interaction with 1 or 2 (in some cases) of∼100–150 different transcrip-tion factors. These factors can be classified based on their contact subunit(s)

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP



on RNA polymerase (40; Figure 4, see color insert). Regulatory proteins wereoriginally classified as activators, or, if they were known to inhibit transcription, asrepressors. More recent studies, however, indicate that both activators and repres-sors can have dual functions, activating or inhibiting transcription from differentpromoters depending on where they bind to the DNA (e.g. 1, 14, 44). More-over, both activators and repressors seem to make direct contact with the RNApolymerase to function [both are hereafter referred to simply as transcription fac-tors (40)].

Many stationary-phase genes are transcribed by two different systems, de-pending on cell growth conditions; one isσS dependent but transcription factorindependent, and the other isσ 70 dependent but transcription factor dependent.Such genes usually carry multiple promoters, to be recognized by each form ofthe transcription apparatus. It is surprising, however, that the two systems ini-tiate transcription at the same site in some cases. For instance, some carboncatabolite genes such asdpsor pexB, encoding stationary-specific DNA-bindingprotein Dps with regulatory and protective roles, are transcribed by Eσ 70 in cAMP-receptor protein (CRP)–dependent fashion, but can also be transcribed by EσS inthe absence of cAMP-CRP; thus, the expression of these genes is cAMP-CRPdependent in growing phase, but becomes independent of cAMP-CRP in station-ary phase (66). Furthermore, thedpspromoter is activated by OxyR when it istranscribed by Eσ 70 during growth, but in the stationary phase it is transcribedby EσS without the support of OxyR [integration host factor (IHF) is requiredinstead (2)]. Thus, the same transcriptional start is achieved both by a combina-tion of Eσ 70 with stress-responsive transcription factors and by EσS holoenzyme.Other recent evidence substantiates the suggestion thatσS overlaps functionallywith σ 70 (53, 94, 95). The RNA polymerase holoenzymes carrying other minorσ subunits such as EσN, EσH, Eσ F, and Eσ FecI recognize and transcribe genesinvolved in adaptation to imbalance in nitrogen and other nutrients, in acquisitionof heat-shock resistance, in generation of flagella, and in an iron citrate uptakesystem, respectively. In sharp contrast to theσS-dependent system, the promotersunder the control of these minorσ subunits cannot be recognized by the Eσ 70

holoenzyme, and the holoenzymes containing these minorσ subunits are unableto transcribeσ 70-dependent genes.

FACTORS AFFECTING SELECTIVE UTILIZATION OFTRANSCRIPTION APPARATUS

More than 100 promoters have been identified, which can be recognized either invivo or in vitro by RNA polymerase holoenzyme containingσS, but no significantconsensus sequences have been identified. Instead, a number ofσ 70-dependentpromoters are transcribed in vitro by the EσS holoenzyme (33, 53, 94, 95). Onepossibility is that the EσS holoenzyme recognizes a specific DNA conformationsuch as bent DNA regions (21). The action of DNA gyrase and the association

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


510 ISHIHAMA

of DNA-binding proteins combine to cause growth-dependent changes in the con-figuration of theE. coli chromosome, often called the nucleoid, which affect thepattern of gene transcription as well as protecting the genome DNA from stress-induced damage. In agreement with the in vivo findings, transcription in vitro byEσS is much higher when directed by templates with low superhelical density (58).The enhancing effect of decreased superhelicity of template DNA on transcrip-tion by EσS is additive with that of trehalose and potassium glutamate, indicatingthat the changes in the cytoplasmic composition and the nucleoid conformationindependently produce growth-dependent changes in gene transcription.

A nucleoid-associated histone-like protein, H-NS, functions as a general si-lencer for gene transcription, repressing the expression of many and diverse genes(4, 83). Mutations in H-NS can lead to overexpression of stress-induced genes inthe absence of stresses. Because the binding of H-NS to target DNA sequences issensitive to both intracellular potassium glutamate levels (increased after osmoticshock) and the level of DNA superhelicity (decreased during stationary phase), theenhancing effect of high potassium glutamate concentration and low DNA superhe-licity on transcription ofσS-dependent genes may be caused not only by activationof EσS holoenzyme but also by release from the inhibition by H-NS (17, 83).

The protein composition ofE. coli nucleoid changes markedly depending oncell growth phase (93). Dps or PexB is a DNA-binding protein that appears inE. coli only in stationary phase and is involved, together with H-NS and IHF, inthe condensation of the nucleoid into a more compact configuration. ThedpsorpexBgene is recognized by EσS holoenzyme only when its promoter structure isaltered by the binding of another histone-like protein, IHF (2, 66), whose levelalso increases in stationary phase (5, 93). Both Dps and IHF are required for theefficient expression ofσS-dependent genes required for starvation survival (2, 76).On the other hand, the DNA-binding protein Fis, most abundant in exponentialphase (93), generally activates transcription of genes that are highly expressedin exponentially growing cells (9, 75, 107) and represses stationary-specific genetranscription (76, 100). Accordingly, the level of Fis protein changes dramaticallyupon nutrient upshift, from<100 molecules per cell in stationary phase to>50,000in exponential phase (6, 93).

MULTIPLE PATHWAYS FOR STATIONARY-PHASEADAPTATION

The highly significant difference in survival levels betweenrpoS+ andrpoSstrainsstrongly suggests a major role forrpoS (and/or the genes under its control) instationary-phase survival.E. coliW3110 is an attenuated laboratory strain that hasbeen widely used as a standard forE. coli genetics and as a model strain in thegenome program. However, there are at least five independent lineages of strainW3110 that differ in their content of twoσ subunits,σS andσ F (47). Becausemutations in therpoSgene render the wild-typerpoS+strain nonviable in stationaryphase, natural variants lacking the intactσS must presumably have acquired yet

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP



unidentified suppressor mutations. Thus, it should be noted that different geneticsystems and mechanisms must be present inE. coli for adaptation to stationaryphase. In this respect, care should be taken not to draw conclusions from resultsobtained by using different bacterial strains with different genetic backgrounds.

CONCLUSION

The stationary-phase–associated changes in the pattern of global gene expressionin E. coliare mainly mediated by modulation of the specificity of RNA polymeraseby replacement of the promoter recognition subunitσ 70 with σS. Accordingly, theintracellular concentration ofσSincreases when cultures stop growing. In addition,the activities of Eσ 70 and EσS are regulated in different ways by changes in boththe cytoplasmic composition and the conformation of the nucleoid. The overallfindings raise the possibility that each stationary-specific promoter carries a specificsequence that is recognized by EσSunder a specific condition or in the presence of aspecific factor. If so, the promoter sequence recognized by EσSmust vary betweengroups that require different conditions or factors for function. However, becausesomeE. colistrains lackingσS can successfully enter stationary phase, alternativepathways must exist, which allow bacterial adaptation to the stationary phase.

ACKNOWLEDGMENTS

Work at the National Institute of Genetics (NIG) was supported by grants-in-aidfrom the Ministry of Education, Science, Sports and Culture of Japan, and theCREST fund from the Japanese Science and Technology Corporation. The authorthanks his colleagues in NIG and collaborators from 70 laboratories in 20 countries.The author thanks Richard S Hayward for critical reading of the manuscript andhelpful comments.

Visit the Annual Reviews home page at www.AnnualReviews.org

LITERATURE CITED

1. Aiba H. 1983. Autoregulation of theEs-cherichia coli crp gene: CRP is a tran-scriptional repressor for its own gene.Cell32:141–49

2. Altuvia S, Almiron M, Huisman G, KolterR, Storz G. 1994. Thedpspromoter is ac-tivated by OxyR during growth and by IHFandσS in stationary phase.Mol. Microbiol.13:265–72

3. Angerer A, Enz S, Ochs M, Braun V. 1995.Transcriptional regulation of ferric citrate

transport inEscherichia coli. FecI belongsto subfamily ofσ 70-type factors that respondto extracytoplasmic stimuli.Mol. Microbiol.18:63–174

4. Atlung T, Sund S, Olesen K, Brondsted L.1996. The histone-like protein H-NS actsas a transcriptional repressor for expressionof the anaerobic and growth phase activa-tor AppY of Escherichia coli. J. Bacteriol.178:3418–25

5. Aviv M, Giladi H, Schreiber G, Oppenheim

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


512 ISHIHAMA

AB, Glaser G. 1994. Expression of thegenes coding for theEscherichia coliin-tegration host factor are controlled bygrowth phase,rpoS, ppGpp and by au-toregulation.Mol. Microbiol. 14:1021–31

6. Ball CA, Osuna R, Ferguson KC, JohnsonRC. 1992. Dramatic changes in Fis levelsupon nutrient upshift inEscherichia coli.J. Bacteriol.174:8043–56

7. Blattner FR, Plunkett G III, Bloch CA,Perna NT, Burland V, et al. 1997. Thecomplete genome sequence ofEscheri-chia coliK-12. Science277:1453–62

8. Bohringer J, Fischer D, Mosler G,Hengge-Aronis R. 1995. UDP-glucose isa potential intracellular signal moleculein the control of expression ofσSandσS-dependent genes inEscherichia coli. J.Bacteriol.177:413–22

9. Bokal AJ, Ross W, Gourse RL. 1995.The transcriptional activator protein FIS:DNA interactions and cooperative inter-actions with RNA polymerase at theEs-cherichia coli rrnBP1 promoter.J. Mol.Biol. 245:197–207

9a. Bouvier J, Gordia S, Kampmann G,Lange R, Hengge-Aronis R, et al. 1998.Interplay between global regulators ofEs-cherichia coli: effect of RpoS, Lrp andH-NS on transcription of the geneosmC.Mol. Microbiol. 29:971–80

10. Brown L, Elliott T. 1996. Efficient trans-lation of the RpoS sigma factor inSalmonella typhimuriumrequires hostfactor I, an RNA-binding protein encodedby thehfq gene.J. Bacteriol.178:3763–70

11. Brown L, Elliott T. 1997. Mutations thatincrease expression of therpoSgene anddecrease its dependence onhfq functionin Salmonella typhimurium. J. Bacteriol.179:656–62

12. Brown KL, Hughes KT. 1995. The roleof anti-sigma factors in gene regulation.Mol. Microbiol. 16:397–404

12a. Cai H, Clarke S. 1999. A novel methyl-

transferase catalyzes the methyl esteri-fication of transaconitate inEscherichiacoli. J. Biol. Chem.274:13470–79

12b. Castanie-Cornet MP, Penfound TA, SmithD, Elliott JF, Foster JW. 1999. Control ofacid resistance inEscherichia coli.J. Bac-teriol. 181:3525–35

12c. Chang L, Wei LI, Audia JP, Morton RA,Schellhorn HE. 1999. Expression of theEscherichia coliNRZ nitrate reductaseis highly growth phase dependent andis controlled by RpoS, the alternativevegetative sigma factor.Mol. Microbiol.34:756–66

13. Chatterji D, Fujita N, Ishihama A.1998. The mediator for stringent con-trol, ppGpp, binds to theβ-subunit ofEscherichia coliRNA polymerase.GenesCells3:279–87

14. Choy H, Park SW, Aki T, Parrack P, FujitaN, et al. 1995. Repression and activa-tion of transcription by Gal and Lac re-pressors: involvement of alpha-subunit ofRNA polymerase.EMBO J.14:4523–29

14a. Conter A, Menchon C, Gutierrez C. 1997.Role of DNA supercoiling andrpoSsigmafactor in the osmotic and growth phase-dependent induction of the geneosmEof Escherichia coliK12. J. Mol. Biol.273:75–83

15. Cunning C, Elliott T. 1999. RpoS synthe-sis is growth rate regulated inSalmonellatyphimurium, but its turnover is not de-pendent on acetyl phosphate synthesis orPTS function.J. Bacteriol.181:4853–62

15a. Cunningham L, Gruer MJ, Guest JR.1997. Transcriptional regulation of the as-conitase genes (acnA and acnB) of Es-cherichia coli. Microbiology 143:3795–805

15b. De Biase D, Tramonti A, Bossa F, ViscaP. 1999. The response to stationary-phase stress conditions inEscherichiacoli: role and regulation of the glutamicacid decarboxylase system.Mol. Micro-biol. 32:1198–211

16. De Las Penas A, Connolly L, Gross CA.

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP



1997. TheσE-mediated response to extra-cytoplasmic stress inEscherichia coliistransduced by RseA and RseB, two neg-ative regulators ofσE. Mol. Microbiol.24:373–85

16a. De Wulf P, Kwon O, Lin EC. 1999. TheCpxRA signal transduction system ofEs-cherichia coli: growth-related autoacti-vation and control of unanticipated targetoperons.J. Bacteriol.181:6772–78

17. Ding Q, Kusano S, Villarejo M, Ishi-hama A. 1995. Promoter selectivity con-trol of Escherichia coliRNA polymeraseby ionic strength: differential recogni-tion of osmoregulated promoters by EσD

and EσS holoenzymes.Mol. Microbiol.16:649–56

18. Engelberg-Kulka H, Reches M, Narasim-han S, Schoulaker-Schwarz R, Klemes Y,et al. 1998.rexBof bacteriophage lambdais an anti-cell death gene.Proc. Natl.Acad. Sci. USA95:15481–86

19. Enz S, Braun V, Crosa JH. 1995. Tran-scription of the region encoding the ferricdicitrate-transport system inEscherichiacoli: similarity between promoters forfecA and for extracytoplasmic functionsigma factors.Gene163:13–18

20. Erickson JW, Gross CA. 1989. Identifica-tion of sigma E onEscherichia coliRNApolymerase: a second alternative sigmafactor involved in high-temperature geneexpression.Genes Dev.3:1462–71

21. Espinosa-Urgel M, Tormo A. 1993.σS-Dependent promoters inEscherichia coliare located in DNA regions with intrinsiccurvature.Nucleic Acids Res.21:3667–70

21a. Farewell A, Kvint K, Nystr¨om T. 1998.uspB, a new σS-regulated gene inEs-cherichia coli which is required forstationary-phase resistance to ethanol.J.Bacteriol.180:6140–47

22. Farewell A, Kvint K, Nystr¨om T. 1998.Negative regulation by RpoS: a case ofsigma-factor competition.Mol. Micro-biol. 29:1039–51

23. Franze de Fernandez MT, Hayward WS,August JT. 1972. Bacterial proteins re-quired for replication of phage Qβ ribonu-cleic acid: purification and properties ofhost factor I, a ribonucleic acid-bindingprotein.J. Biol. Chem.10:824–31

24. Gentry DR, Hernandez VJ, Nguyen LH,Jensen DB, Cashel M. 1993. Synthesisof the stationary-phase sigma factorσS

is positively regulated by ppGpp.J. Bac-teriol. 175:7982–89

25. Gibson KE, Silhavy TJ. 1999. The LysRhomolog LrhA promotes RpoS degra-dation by modulating activity of theresponse regulatorsprE. J. Bacteriol.181:563–71

25a. Gort AS, Ferber DM, Imlay JA. 1999.The regulation and role of the periplasmiccopper, zinc superoxide dismutase ofEs-cherichia coli. Mol. Microbiol. 32:179–91

26. Gottesman S, Clark WP, de Crecy-LagardV, Maurizi MR. 1993. ClpX, an alterna-tive subunit for the ATP-dependent Clpprotease ofEscherichia coli: sequenceand in vivo activities.J. Biol. Chem.268:22618–26

27. Groat RG, Schultz JE, Zychlinsky E,Bockman A, Martin A. 1986. Starvationproteins inEscherichia coli: kinetics ofsynthesis and role in starvation survival.J. Bacteriol.168:486–93

28. Gross CA, Chan C, Dombroski A, GruberT, Sharp M, et al. 1998. The functionaland regulatory roles of sigma factors intranscription.Cold Spring Harbor Symp.Quant. Biol.63:141–55

29. Gross CA, Lonetto M, Losick R. 1992.Bacterial sigma factors. InTranscrip-tional Regulation, ed. K Yamamoto, SMcKnight, pp. 129–76. Cold Spring Har-bor, NY: Cold Spring Harbor Lab. Press

29a. Heiskanen P, Taira S, Rhen M. 1994. Roleof rpoS in the regulation of Salmonellaplasmid virlence (spv) genes.FEMS Mi-crobiol. Lett.123:125–30

30. Helmann JD. 1991. Alternative sigma

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


514 ISHIHAMA

factors and the regulation of flagellar geneexpression.Mol. Microbiol. 5:2875–82

31. Helmann JD, Chamberlin MJ. 1988.Structure and function of bacterial sigmafactors.Annu. Rev. Biochem.57:839–72

32. Hengge-Aronis R. 1996. Back to logphase: sigma S as a global regulator in theosmotic control of gene expression inEs-cherichia coli. Mol. Microbiol. 21:887–93

33. Hiratsu K, Shinagawa H, Makino K.1995. Mode of promoter recognition bythe Escherichia coli RNA polymeraseholoenzyme containing theσS subunit:identification of the recognition sequenceof the fic promoter. Mol. Microbiol.18:841–50

34. Hochman A. 1997. Programmed celldeath in prokaryotes.Crit. Rev. Microbiol.23:207–14

35. Holden MTG, Chhabra SR, de NysR, Stead P, Bainton NJ, et al. 1999.Quorum-sensing cross talk: isolation andchemical characterization of cyclic dipep-tides fromPseudomonas aeruginosaandother gram-negative bacteria.Mol. Micro-biol. 33:1254–66

35a. Horlacher R, Uhland K, Klein W,Ehrmann M, Boos W. 1996. Characteri-zation of a cytoplasmic trehalase ofEs-cherichia coli. J. Bacteriol.178:6250–57

36. Hughes KT, Mathee K. 1998. The anti-sigma factors.Annu. Rev. Microbiol.52:231–86

37. Huisman GW, Kolter R. 1994. Sens-ing starvation: a homoserine lactone-dependent signaling pathway inEs-cherichia coli. Science265:537–39

38. Ishihama A. 1981. Subunit assembly ofEscherichia coliRNA polymerase.Adv.Biophys.14:1–35

39. Ishihama A. 1988. Promoter selectivityof prokaryotic RNA polymerases.TrendsGenet.4:282–86

40. Ishihama A. 1997. Promoter selectivitycontrol of Escherichia coliRNA poly-merase. InNucleic Acids and Molecular

Biology, Vol. 11.Mechanism of Transcrip-tion, ed. F Eckstein, D Lilley, pp. 53–70.Heidelberg: Springer-Verlag

41. Ishihama A. 1997. Adaptation of gene ex-pression in stationary phase bacteria.Curr.Opin. Genet. Dev.7:582–88

42. Ishihama A. 1999. Modulation of the nu-cleoid, the transcription apparatus, and thetranslation machinery in bacteria for sta-tionary phase survival.Genes Cells3:135–43

43. Ishihama A, Taketo M, Saitoh T, Fukuda R.1976. Control of formation of RNA poly-merase inEscherichia coli. In RNA Poly-merase, ed. M Camberlin, R Losick, pp.475–502. Cold Spring Harbor, NY: ColdSpring Harbor Lab. Press

44. Ishizuka H, Hanamura A, Inada T, Aiba H.1994. Mechanism of the down-regulationof cAMP receptor protein by glucose inEscherichia coli: role of autoregulation ofthecrp gene.EMBO J.13:3077–82

45. Iyoda S, Kutsukake K. 1995. Moleculardissection of the flagelum-specific anti-sigma factor, FlgM, ofSalmonella ty-phimurium. Mol. Gen. Genet.249:417–24

46. Jishage M, Ishihama A. 1995. Regulationof RNA polymerase sigma subunit synthe-sis inEscherichia coli: intracellular levelsof σ70 andσ 38. J. Bacteriol.177:6832–35

47. Jishage M, Ishihama A. 1997. Variation inRNA polymerase sigma subunit composi-tion within different stocks ofEscherichiacoli W3110.J. Bacteriol.179:959–63

48. Jishage M, Ishihama A. 1998. A station-ary phase protein inEscherichia coliwithbinding activity to the majorσ subunit ofRNA polymerase.Proc. Natl. Acad. Sci.USA95:4953–58

49. Jishage M, Ishihama A. 1999. Transcrip-tional organization and in vivo role of theEscherichia coli rsdgene, encoding theregulator of RNA polymerase sigma D.J.Bacteriol.181:3768–76

50. Jishage M, Iwata A, Ueda S, IshihamaA. 1996. Regulation of RNA polymerasesigma subunit synthesis inEscherichia

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP



coli: intracellular levels of four species ofsigma subunit under various growth con-ditions.J. Bacteriol.178:5447–51

51. Kajitani M, Ishihama A. 1991. Identifi-cation and sequence determination of thehost factor gene for bacteriophage Qβ.Nucleic Acids Res.19:1063–66

51a. Kakeda M, Ueguchi C, Yamada H,Mizuno T. 1995. An Escherichia colicurved DNA-binding protein whose ex-pression is affected by the stationaryphase-specific sigma factorσS. Mol. Gen.Genet.248:629–34

51b. Kikuchi Y, Kurahashi O, Nagano T,Kamio Y. 1998. RpoS-dependent expres-sion of the second lysine decarboxylasegene inEscherichia coli. Biosci. Biotech-nol. Biochem.62:1267–70

52. Klauck E, Bohringer J, Hengge-Aronis R.1997. The LysR-like regulator LeuO inEscherichia coliis involved in the trans-lational regulation ofrpoSby affecting theexpression of the small regulatory DsrA-RNA. Mol. Microbiol. 25:559–69

53. Kolb A, Kotlarz D, Kusano S, IshihamaA. 1995. Selectivity of theEscherichiacoli RNA polymerase Eσ 38 for overlap-ping promoters and ability to support CRPactivation.Nucleic Acids Res.23:819–26

54. Kolter R, Losick R. 1998. One for all andall for one.Science10:226–27

55. Kolter R, Siegele DA, Tormo A. 1993.The stationary phase of the bacterial lifecycle.Annu. Rev. Microbiol.47:855–74

56. Kornberg A. 1995. Inorganic polyphos-phate: toward making a forgotten poly-mer unforgettable.J. Bacteriol.177:491–96

57. Kroos L, Zhang B, Ichikawa H, Yu YT.1999. Control of sigma activity duringBacillus subtilissporulation.Mol. Micro-biol. 31:1285–94

58. Kusano S, Ding Q, Fujita N, Ishihama A.1996. Promoter selectivity ofEscherichiacoli RNA polymerase Eσ70 and Eσ 38

holoenzymes: effect of DNA supercoil-ing. J. Biol. Chem.271:1998–2004

59. Kusano S, Ishihama A. 1997. Stimulatoryeffect of trehalose on the formation andactivity of Escherichia coliRNA poly-merase Eσ38 holoenzyme.J. Bacteriol.179:3649–54

60. Kusano S, Ishihama A. 1997. Functionalinteraction ofEscherichia coliRNA poly-merase with inorganic polyphosphate.Genes Cells2:433–41

61. Kutsukake K, Iino T. 1994. Role of theFliA-FlgM regulatory system on the tran-scriptional control of the flagellar regulonand flagellar formation inSalmonella ty-phimurium. J. Bacteriol.176:3598–605

62. Lange R, Fischer D, Hengge-Aronis R.1995. Identification of the transcriptionalstart sites and the role of ppGpp in theexpression ofrpoS, the structural genefor theσS subunit of RNA polymerase inEscherichia coli. J. Bacteriol.177:4676–80

63. Lease RA, Cusick ME, Belfort M. 1998.Riboregulation inEscherichia coli: DsrARNA acts by RNA: RNA interactions atmultiple loci.Proc. Natl. Acad. Sci. USA95:12456–61

63a. Lessard IA, Walsh CT. 1999. VanX, a bac-terialD-alanyl-D-alanine dipeptidase: re-sistance, immunity, or survival function?Proc. Natl. Acad. Sci. USA96:11028–32

64. Loewen PC, Hengge-Aronis R. 1994. Therole of the sigma-factor sigma S (KatF)in bacterial global regulation.Annu. Rev.Microbiol. 48:53–80

65. Loewen PC, Hu B, Strutinsky J, SparlingR. 1998. Regulation in therpoSregulonof Escherichia coli. Can. J. Microbiol.44:707–17

66. Lomovskaya OL, Kidwell JP, Matin A.1994. Characterization of the sigma 38-dependent expression of a coreEs-cherichia coli starvation gene,pexB. J.Bacteriol.176:3928–35

67. Lonetto M, Gribskov M, Gross CA. 1994.The sigma 70 family: sequence conser-vation and evolutionary relationships.J.Bacteriol.174:3843–49

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


516 ISHIHAMA

68. Maeda H, Jishage M, Nomura T, FujitaN, Ishihama A. 2000. Promoter selectiv-ity of the RNA polymerase holoenzymecontaining the extracytoplasmic function(ECF) sigma subunitE or σFecI. J. Bacte-riol. 182:1181–84

69. Mecsas J, Rouviere PE, Erickson JW,Donohue TJ, Gross CA. 1993. The ac-tivity of σE, an Escherichia coliheat-inducible σ -factor, is modulated by ex-pression of outer membrane proteins.Genes Dev.7:2618–28

70. Merrick MJ. 1993. In a class of its own:the RNA polymeraseσ54 (σN). Mol. Mi-crobiol. 10:903–9

70a. Michan C, Manchado M, Dorado G,Pueyo C. 1999. In vivo transcription of theEscherichia coli oxyRregulon as a func-tion of growth phase and in response tooxidative stress.J. Bacteriol.181:2759–64

71. Missiakas D, Betton JM, Raina S.1996. New components of protein fold-ing in extracytoplasmic compartmentsof Escherichia coli SurA, FkpA andSkp/OmpH.Mol. Microbiol. 21:871–84

72. Missiakas D, Mayer MP, Lemaire M,Georgopoulos C, Raina S. 1997. Modu-lation of theEscherichia coliσE (RpoE)heat-shock transcription-factor activityby the RseA, RseB and RseC proteins.Mol. Microbiol. 24:355–71

73. Muffler A, Fischer D, Altuvia S, StorzG, Hengge-Aronis R. 1996. The responseregulator RssB controls stability of theσS subunit of RNA polymerase inEs-cherichia coli. EMBO J.15:1333–39

74. Muffler A, Traulsen DD, Fischer D,Lange R, Hengge-Aronis R. 1997. TheRNA-binding protein HF-I plays a globalregulatory role which is largely, but notexclusively, due to its role in expressionof theσS subunit of RNA polymerase inEscherichia coli. J. Bacteriol.179:297–300

75. Muskhelishvili G, Travers AA, HeumannH, Kahmann R. 1995. FIS and RNA

polymerase form a specific nucleoproteincomplex at a stable RNA promoter.EMBOJ. 14:1446–52

76. Nystrom T. 1995. Glucose starvation stim-ulon of Escherichia coli: role of integra-tion host factor in starvation survival andgrowth phase-dependent protein synthesis.J. Bacteriol.177:5707–10

77. Ozaki M, Wada A, Fujitan N, Ishihama A.1991. Growth phase-dependent conversionof RNA polymerase inEscherichia coli.Mol. Gen. Genet.230:17–24

78. Poolman B, Glassker E. 1998. Regulationof compatible solute accumulation in bac-teria.Mol. Microbiol. 29:397–407

79. Pratt LA, Silhavy TJ. 1996. The responseregulator SprE controls the stability ofRpoS.Proc. Natl. Acad. Sci. USA93:2488–92

80. Pressler U, Staudenmaier H, Zimmer-mann L, Baun V. 1988. Genetics of theiron dicitrate transport system ofEs-cherichia coli. J. Bacteriol.170:2716–24

81. Raina S, Missiakas D, Georgopoulos C.1995. TherpoEgene encoding theσE(σ 24)heat shock sigma factor ofEscherichia coli.EMBO J.14:1043–55

82. Raivio TL, Silhavy TJ. 1999. The sigma Eand Cpx regulatory pathways: overlappingbut distinct envelope stress response.Curr.Opin. Microbiol.2:159–65

83. Rajkumari K, Kusano S, Ishihama A,Mizuno T, Gowrishankar J. 1996. Effectsof H-NS and potassium glutamate onσS-andσ70-directed transcription in vitro fromosmotically regulated P1 and P2 promotersof proU in Escherichia coli. J. Bacteriol.178:4176–81

84. Rao NN, Kornberg A. 1996. Inorganicpolyphosphate supports resistance and sur-vival of stationary-phaseEscherichia coli.J. Bacteriol.178:1394–1400

85. Robison K, McGuire AM, Church QM.1998. A comprehensive library of DNA-binding site matrices for 55 proteins ap-plied to the completeEscherichia coliK-12genome.J. Mol. Biol.284:241–54

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP



86. Roszak DB, Colwell RR. 1987. Survivalstrategies of bacteria in the natural envi-ronment.Microbiol. Rev.51:365–79

87. Rouviere PE, de Las Penas A, MearesJ, Lu CZ, Rudd KE, Gross CA. 1995.rpoE, the gene encoding the second heat-shock sigma factor,σE, in Escherichiacoli. EMBO J.14:1032–42

88. Salmond GPC, Bycroft BW, StewartGSAB, Williams P. 1995. The bacterial‘enigma’: cracking the code of cell-cellcommunication.Mol. Microbiol.16:615–24

88a. Schelhorn HE, Audia JP, Wei LI,Chang L. 1998. Identification of con-served, RpoS-dependent stationary-phasegenes ofEscherichia coli. J. Bacteriol.180:6283–91

89. Schweder T, Lee K-H, Lomovskaya O,Matin A. 1996. Regulation ofEscherichiacoli starvation sigma factor (σS) byClpXP protease.J. Bacteriol.178:470–76

89a. Shin S, Song SG, Lee DS, Pan JG, Park C.1997. Involvement oficlR andrpoSin theinduction ofacs, the gene for acetylcoen-zyme A synthetase ofEscherichia coliK-12. FEMS Microbiol. Lett.146, 103–08.

90. Shingler V. 1996. Signal sensing by sigma54-dependent regulators: derepression asa control mechanism.Mol. Microbiol.19:400–16

91. Sledjeski DD, Gupta A, Gottesman S.1996. The small RNA, DsrA, is essen-tial for the low temperature expression ofRpoS during exponential growth inEs-cherichia coli. EMBO J.15:3993–4000

91a. Takayanagi Y, Tanaka K, Takahashi H.1994. Structure of the 5′ upstream re-gion and the regulation of therpoSgeneof Escherichia coli. Mol. Gen. Genet.243:525–31

92. Talukder AZ, Ishihama A. 1999. Twelvespecies of DNA-binding protein fromEscherichia coli: sequence recognitionspecificity and DNA binding affinity.J.Biol. Chem.274:33105–13

93. Talukder AZ, Iwata A, Ueda A, Ishi-hama A. 1999. Growth phase-dependentvariation in the protein composition ofEscherichia colinucleoid. J. Bacteriol.181:6361–70

94. Tanaka K, Fujita N, Ishihama A, Taka-hashi H. 1993. Heterogeneity of the prin-cipal sigma factor inEscherichia coli:the rpoS gene product,σ38, is a princi-pal sigma factor of RNA polymerase instationary phaseEscherichia coli. Proc.Natl. Acad. Sci. USA90:3511–15

95. Tanaka K, Kusano S, Fujita N, IshihamaA, Takahashi H. 1995. Promoter deter-minants forEscherichia coliRNA poly-merase holoenzyme containingσ38 (therpoS gene product).Nucleic Acids Res.23:827–34

96. VanBogelen RA, Abshire KZ, Pertsem-lidis A, Clark RL, Neidhardt FC. 1996.Gene-protein database ofEscherichia coliK-12, edition 6. InEscherichia coli andSalmonella: Cellular and Molecular Bi-ology, ed. FC Neidhardt, R Curtiss III, JLIngraham, ECC Lin, KR Low, et al, pp.2067–17. Washington, DC: Am. Soc. Mi-crobiol.

96a. Villarroya M, Perez-Roger I, Macian F,Armengod ME. 1998. Stationary phaseinduction ofdnaNandrecF, two genes ofEscherichia coliinvolved in DNA repli-cation and repair.EMBO J.17:1829–37

96b. Visick JE, Cai H, Clarke S. 1998. TheL-isoaspartyl protein repair methyltrans-ferase enhances survival of agingEs-cherichia coli subjected to secondaryenvironmental stresses.J. Bacteriol.180:2623–29

97. Wawrzynow A, Wojtkowiak D, Marsza-lek J, Banecki B, Jonsen M, et al.1995. The CplX heat-shock protein ofEscherichia coli, the ATP-dependentsubstrate specificity component of theClpP-ClpX protease, is a novel molecu-lar chaperone.EMBO J.14:1867–77

98. Webb C, Moreno M, Wilmes-RiesenbergM, Curtiss R III, Foster JW. 1999. Effects

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FQP/LVF P2: FQP


518 ISHIHAMA

of DksA and ClpP protease on sigma Sproduction and virulence inSalmonellatyphimurium. Mol. Microbiol. 34:112–23

99. Wojtkowiak D, Georgopoulos C, ZyliczM. 1993. Isolation and characterization ofClpX, a new ATP-dependent specificitycomponent of the Clp protease ofEs-cherichia coli. J. Biol. Chem.268:22609–17

100. Xu JM, Johnson RC. 1995. Identifica-tion of genes negatively regulated byfis:Fis andrpoScomodulate growth-phase-dependent gene expression inEscherichiacoli. J. Bacteriol.177:938–47

101. Yamada M, Talukder AA, Nitta T. 1999.Characterization of thessnAgene, whichis involved in the decline of cell viabilityat the beginning of stationary phase inEs-cherichia coli. J. Bacteriol.181:1838–46

102. Yamashino T, Ueguchi C, Mizuno T.1995. Quantitative control of the station-ary phase-specific sigma factor,σS, inEscherichia coli: involvement of the nu-cleoid protein H-NS.EMBO J.14:594–602

103. Yura T, Nagai H, Mori H. 1993. Regula-

tion of the heat-shock response in bacte-ria. Annu. Rev. Microbiol.47:321–50

104. Zambrano MM, Siegele DA, Almiron M,Tormo A, Kolter R. 1993. Microbial com-petition: Escherichia colimutants thattake over stationary phase cultures.Sci-ence259:1757–60

105. Zgurskaya HI, Keyhan M, Matin A. 1997.The sigma S level in starvingEscherichiacoli cells increase solely as a result of itsincreased stability, despite decreased syn-thesis.Mol. Microbiol. 24:643–51

106. Zhang AX, Altuvia S, Tiwari A, Arga-man L, Hengge-Aronis R, Storz G. 1998.The OxyS regulatory RNA repressesrpoStranslation and binds the Hfq (HF-I) pro-tein.EMBO J.17:6061–68

107. Zhang XY, Bremer H. 1996. Effects of Fison ribosome synthesis and activity and onrRNA promoter activities inEscherichiacoli. J. Mol. Biol.259:27–40

108. Zhou YN, Walter WA, Gross CA. 1992. Amutantσ 32 with a small deletion in con-served region 3 ofσ has reduced affinityfor core RNA polymerase.J. Bacteriol.174:5005–12

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FDS

September 11, 2000 14:57 Annual Reviews AR110-CO

Figure 1 Sigma subunits ofEscherichia coli. E. coli contains seven species ofσ subunit.The RNA polymerase holoenzyme containing eachσ subunit recognizes and transcribes aspecific set of genes. The intracellular concentrations of all sevenσ subunits were deter-mined for both exponential and stationary phases ofE. coli W3110 (46, 50, 68).

Figure 2 Intracellular concentration of each holoenzyme form. The intracellular concen-trations of all sevenσ subunits were determined by quantitative Western blot analysis (46,50, 68), while the dissociation constant between theσ subunit and the core enzyme wasdetermined by FPLC gel chromatography (Maeda H et al, submitted for publication). Theconcentrations of all seven holoenzyme forms were calculated from these two values.

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

P1: FDS

September 11, 2000 14:57 Annual Reviews AR110-CO

Figure 3 Regulation of the level and activity ofσ s subunit. The synthesis ofσ s subunit isregulated at both transcription and translation steps by a number of regulatory factors. Themetabolic stability ofσ s subunit is also controlled. The activity of holoenzyme containingσ s subunit is specifically activated or repressed by a number of stationary-phase specificfactors. Transcription of someσ s-dependent genes is modulated by AppY, BolA, CRP, Fis,IHF, Lrp (for a review see 65).

Figure 4 Classification ofE. coli transcription factors. Transcription factors have beenclassified on the basis of the contact subunit of RNA polymerase (40, 42).

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

Annual Review of Microbiology Volume 54, 2000

CONTENTSTHE LIFE AND TIMES OF A CLINICAL MICROBIOLOGIST, Albert Balows 1ROLE OF CYTOTOXIC T LYMPHOCYTES IN EPSTEIN-BARR VIRUS-ASSOCIATED DISEASES, Rajiv Khanna, Scott R. Burrows 19BIOFILM FORMATION AS MICROBIAL DEVELOPMENT, George O'Toole, Heidi B. Kaplan, Roberto Kolter 49MICROBIOLOGICAL SAFETY OF DRINKING WATER, U. Szewzyk, R. Szewzyk, W. Manz, K.-H. Schleifer 81THE ADAPTATIVE MECHANISMS OF TRYPANOSOMA BRUCEI FOR STEROL HOMEOSTASIS IN ITS DIFFERENT LIFE-CYCLE ENVIRONMENTS, I. Coppens, P. J. Courtoy 129THE DEVELOPMENT OF GENETIC TOOLS FOR DISSECTING THE BIOLOGY OF MALARIA PARASITES, Tania F. de Koning-Ward, Chris J. Janse, Andrew P. Waters 157

NUCLEIC ACID TRANSPORT IN PLANT-MICROBE INTERACTIONS: The Molecules That Walk Through the Walls, Tzvi Tzfira, Yoon Rhee, Min-Huei Chen, Talya Kunik, Vitaly Citovsky 187PHYTOPLASMA: Phytopathogenic Mollicutes, Ing-Ming Lee, Robert E. Davis, Dawn E. Gundersen-Rindal 221ROOT NODULATION AND INFECTION FACTORS PRODUCED BY RHIZOBIAL BACTERIA, Herman P. Spaink 257ALGINATE LYASE: Review of Major Sources and Enzyme Characteristics, Structure-Function Analysis, Biological Roles, and Applications, Thiang Yian Wong, Lori A. Preston, Neal L. Schiller 289INTERIM REPORT ON GENOMICS OF ESCHERICHIA COLI, M. Riley, M. H. Serres 341ORAL MICROBIAL COMMUNITIES: Biofilms, Interactions, and Genetic Systems, Paul E. Kolenbrander 413ROLES OF THE GLUTATHIONE- AND THIOREDOXIN-DEPENDENT REDUCTION SYSTEMS IN THE ESCHERICHIA COLI AND SACCHAROMYCES CEREVISIAE RESPONSES TO OXIDATIVE STRESS, Orna Carmel-Harel, Gisela Storz 439RECENT DEVELOPMENTS IN MOLECULAR GENETICS OF CANDIDA ALBICANS, Marianne D. De Backer, Paul T. Magee, Jesus Pla 463FUNCTIONAL MODULATION OF ESCHERICHIA COLI RNA POLYMERASE, Akira Ishihama 499BACTERIAL VIRULENCE GENE REGULATION: An Evolutionary Perspective, Peggy A. Cotter, Victor J. DiRita 519

LEGIONELLA PNEUMOPHILA PATHOGENESIS: A Fateful Journey from Amoebae to Macrophages, M. S. Swanson, B. K. Hammer 567THE DISEASE SPECTRUM OF HELICOBACTER PYLORI : The Immunopathogenesis of Gastroduodenal Ulcer and Gastric Cancer, Peter B. Ernst, Benjamin D. Gold 615PATHOGENICITY ISLANDS AND THE EVOLUTION OF MICROBES, Jörg Hacker, James B. Kaper 641DNA SEGREGATION IN BACTERIA, Gideon Scott Gordon, Andrew Wright 681

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

POLYPHOSPHATE AND PHOSPHATE PUMP, I. Kulaev, T. Kulakovskaya 709ASSEMBLY AND FUNCTION OF TYPE III SECRETORY SYSTEMS, Guy R. Cornelis, Frédérique Van Gijsegem 735

PROTEINS SHARED BY THE TRANSCRIPTION AND TRANSLATION MACHINES, Catherine L. Squires, Dmitry Zaporojets 775HOLINS: The Protein Clocks of Bacteriophage Infections, Ing-Nang Wang, David L. Smith, Ry Young 799OXYGEN RESPIRATION BY DESULFOVIBRIO SPECIES, Heribert Cypionka 827REGULATION OF CARBON CATABOLISM IN BACILLUS SPECIES, J. Stülke, W. Hillen 849IRON METABOLISM IN PATHOGENIC BACTERIA, Colin Ratledge, Lynn G Dover 881

Ann

u. R

ev. M

icro

biol

. 200

0.54

:499

-518

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

SER

M-m

ulti-

site

acc

ount

on

03/0

4/09

. For

per

sona

l use

onl

y.

Functional Modulation of Escherichia Coli RNA Polymerase

Documents

Transcript of Functional Modulation of Escherichia Coli RNA Polymerase