Adaptation to Environmental Stimuli within the Host: Two ...terium species, including Mycobacterium...

MICROBIOLOGY AND MOLECULAR BIOLOGY REVIEWS, Dec. 2011, p. 566–582 Vol. 75, No. 41092-2172/11/$12.00 doi:10.1128/MMBR.05004-11Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Adaptation to Environmental Stimuli within the Host:Two-Component Signal Transduction Systems of

Mycobacterium tuberculosisDaniel J. Bretl,† Chrystalla Demetriadou,† and Thomas C. Zahrt*

Center for Infectious Disease Research and Department of Microbiology and Molecular Genetics,Medical College of Wisconsin, 8701 Watertown Plank Road,

Milwaukee, Wisconsin 53226-0509

INTRODUCTION .......................................................................................................................................................566TCSSs OF MYCOBACTERIUM TUBERCULOSIS ..................................................................................................567

senX3 (Rv0490)-regX3 (Rv0491)..............................................................................................................................567Rv0600c-Rv0601c-tcrA (Rv0602c) ...........................................................................................................................569phoP (Rv0757)-phoR (Rv0758) ...............................................................................................................................571narL (Rv0844c)-Rv0845...........................................................................................................................................572prrA (Rv0903c)-prrB (Rv0902c)..............................................................................................................................573mprA (Rv0981)-mprB (Rv0982) ..............................................................................................................................573kdpD (Rv1028c)-kdpE (Rv1027c)............................................................................................................................574trcR (Rv1033c)-trcS (Rv1032c)................................................................................................................................574devR-devS-Rv2027c or dosR-dosS-dosT...................................................................................................................574mtrA (Rv3246c)-mtrB (Rv3245c) ............................................................................................................................576tcrX (Rv3765c)-tcrY (Rv3764c) ................................................................................................................................577pdtaR (Rv1626)-pdtaS (Rv3220c) and Other Orphaned RRs and SKs ............................................................577

CONCLUDING REMARKS......................................................................................................................................578ACKNOWLEDGMENTS ...........................................................................................................................................578REFERENCES ............................................................................................................................................................578

INTRODUCTION

Adaptation to environmental stimuli in bacteria is mediatedprimarily through the expression of transcriptional regulators,including serine-threonine protein kinases (STPKs), extracyto-plasmic function (ECF) sigma factors, and two-component sig-nal transduction systems (TCSSs). Prototypical TCSSs arecomprised of a membrane-localized histidine sensor kinase(SK) and a cytoplasmically localized response regulator (RR).SKs and RRs are often identified based on their conserveddomain architectures and their ability to mediate signal trans-duction events via phosphorylation. SKs are typically com-prised of a sensor domain and one or more transmembranedomains connected via a flexible linker region to a cytoplasmictransmitter region containing a dimerization motif and a ki-nase domain. The kinase domain can be further divided intotwo subdomains, one containing the highly conserved site ofhistidine phosphorylation (H box) and the other containing 4regions of highly conserved amino acids (N, D, F, and G boxes)that collectively form the ATP-binding pocket. RRs are alsoorganized into discrete functional modules that include anN-terminal receiver domain containing a highly conserved as-

partic acid that serves as the site of phosphorylation and aC-terminal DNA-binding domain.

Initiation of signal transduction events between SKs andRRs is mediated following the recognition of environmentalstimuli by the SKs (Fig. 1A). Following signal recognition, SKsoften dimerize and autophosphorylate in trans at a single con-served histidine residue present within each SK monomer.Transfer of this phosphate by the SK to the conserved asparticacid residue in the receiver domain of the cognate RR resultsin conformation alterations that promote DNA binding andtranscriptional regulation, RNA binding, protein-protein inter-actions, or other enzymatic activities (46, 65, 152–154, 171). Inaddition to serving as phosphoryl donors, certain SKs alsofunction as RR phosphatases, an activity that helps regulateactivation levels of the system. Recently, more complex ver-sions of TCSSs with multiple histidine- or aspartate-containingphosphotransfer domains or phosphodonor intermediates(Fig. 1B) have been described (105).

While present in some plants, lower eukaryotes, and ar-chaea, TCSSs are ubiquitous in bacteria (77). The number ofsystems present in a given organism generally correlates withits genome size and the complexity of the environment(s) inwhich the bacterium typically resides (46). TCSSs are usuallygenetically linked and transcriptionally coupled, though or-phaned and independently transcribed systems also exist. Well-characterized TCSSs in bacteria include the Escherichia coliEnvZ-OmpR system, which regulates gene expression in re-sponse to osmotic stress (37), the E. coli and Bacillus subtilisCheA-CheY system involved in chemotaxis (81), and the Sal-monella enterica PhoQ-PhoP system required for gene regula-

* Corresponding author. Mailing address: Center for Infectious Dis-ease Research, Department of Microbiology and Molecular Genetics,Medical College of Wisconsin, 8701 Watertown Plank Road, Milwau-kee, WI 53226-0509. Phone: (414) 955-7429. Fax: (414) 955-6567. E-mail: [email protected].

† D.J.B. and C.D. contributed equally to this work.

566

on February 12, 2021 by guest

http://mm

br.asm.org/

Dow

nloaded from

http://mmbr.asm.org/

tion in response to magnesium concentrations (75). Beyondthese well-characterized systems, TCSSs have also been shownto regulate many physiological processes, including sporula-tion, competence, antibiotic resistance, transition into station-ary phase, virulence, and carbon, nitrogen, and phosphate uti-lization (82). The importance of TCSSs in bacterial survivaland the absence of TCSSs in higher eukaryotes also makethese systems attractive targets for therapeutic developmentagainst pathogenic organisms (7, 8). Consistent with this idea,mutant strains that are defective in specific TCSSs that lead tovirulence attenuation are now being investigated as potentialvaccine candidates (1, 102, 163).

TCSSs OF MYCOBACTERIUM TUBERCULOSIS

Mycobacterium tuberculosis is a facultative intracellularpathogen and is estimated to currently infect about one-thirdof the world’s population. It is the etiological agent of tuber-culosis (TB) and is responsible for over 9 million new infec-tions and 1.7 million deaths annually (177a). M. tuberculosisinfects and survives within professional phagocytes, includingmacrophages, and is able to persist in the human host fordecades within granulomatous lesions, where the organism islikely exposed to environmental stresses that include hypoxia,nutrient limitation, reactive oxygen and reactive nitrogen in-termediates, pH alterations, and cell wall/membrane stress. Toadapt to these and other stresses, M. tuberculosis encodes ap-proximately 190 regulatory proteins, including 11 geneticallylinked TCSSs, five orphaned RRs, and two orphaned SKs (30,161). The number of intact TCSSs in M. tuberculosis is lowerthan that typically found in other bacteria of similar genome

size, possibly reflecting the evolution of this bacterium as astrict human pathogen and its adaptation to a predominantlyintracellular lifestyle. Comparative genomic analyses of the 11genetically linked TCSSs in M. tuberculosis indicate that ho-mologs of these genes exist in other representative Mycobac-terium species, including Mycobacterium bovis, Mycobacteriumavium, Mycobacterium leprae, and Mycobacterium smegmatis,which is often used as a fast-growing surrogate for M. tubercu-losis (Table 1) (30, 183). All of the paired TCSSs found in M.tuberculosis are conserved in their genetic arrangement andlocation within the closely related Mycobacterium bovis BCGvaccine strain. In contrast, only four TCSSs are present andpredicted to be functional in M. leprae. Here, we discuss ourcurrent understanding of the 11 genetically linked TCSSs andthe orphaned RRs and SKs of M. tuberculosis H37Rv, includ-ing the environmental signal(s) to which these systems areresponsive, the regulons they control, and the importance ofindividual TCSSs in the physiology and/or virulence of M.tuberculosis.

senX3 (Rv0490)-regX3 (Rv0491)

senX3-regX3 was the first TCSS to be described in M. tuber-culosis (178). It is comprised of SK SenX3 and RR RegX3.Homologs of senX3 and regX3 are present in the genomes ofvarious mycobacterial species, including M. leprae (Table 1)(183), suggesting that this TCSS is evolutionarily conservedand may regulate fundamental physiological processes. senX3and regX3 are transcriptionally coupled in M. tuberculosis, withsenX3 encoding the first gene product in the operon (155). Theintergenic region of these genes contains repetitive DNA se-

FIG. 1. Organization of the prototypical TCSS and phosphorelay systems in bacteria. (A) The prototypical TCSS is comprised of a single sensorkinase (SK) and a single response regulator (RR). The input domain of the SK recognizes a specific signal(s) from the environment. Thisrecognition results in activation of the kinase domain and autophosphorylation in the output domain of the SK at a conserved histidine residue.The output domain of the phosphorylated SK interacts with the receiver domain of the RR, catalyzing the transfer of phosphate to a conservedaspartate residue within the receiver domain. Phosphorylation of the RR activates its output domain, resulting in conformational changes in theRR that help mediate specific biological activities, including DNA binding and transcriptional regulation. (B) Phosphorelay systems are comprisedof an SK and a terminal RR. These systems also contain an intermediate regulator (connector) lacking an output domain and a phosphotransferprotein with a conserved histidine for phosphorylation. In some phosphorelays, the intermediate phosphotransfer protein and the SK are fused.(Modified from reference 105 with permission of the publisher.)

VOL. 75, 2011 TCSS IN MYCOBACTERIUM 567


http://mm

br.asm.org/

Dow

nloaded from


quences termed mycobacterial interspersed repeat units(MIRUs) (49, 155, 178). While the physiological importance ofthese repeat units remains unclear, it has been proposed thatthese elements could be used as a diagnostic to differentiate M.tuberculosis from other mycobacterial species, as the numberand structure of MIRUs within the intergenic region appear tobe species specific (96, 97, 119, 155).

Biochemical and genetic studies indicate that SenX3-RegX3functions as a paired signaling system in vitro and in vivo (49,178). The cytoplasmic portion of SenX3 is autophosphorylatedat His-167, and this residue participates in phosphorelay reac-tions with Asp-52 of RegX3 (Tables 2 and 3) (49, 66) . WhileregX3 appears to be essential in M. smegmatis (49), this gene isdispensable for the growth of M. tuberculosis (115, 123, 124). In

contrast, senX3 is not essential in either M. smegmatis or M.tuberculosis (49, 115, 123, 124). Regardless, deletion of eitherregX3 or senX3 in M. tuberculosis does not substantially alterthe growth of the H37Rv strain under standard laboratorygrowth conditions (115, 123).

SenX3 has been shown to regulate adaptation programs inM. smegmatis in response to inorganic phosphate (Pi) (Table 2)(49). Under conditions of low environmental Pi, SenX3 phos-phorylates RegX3, leading to the upregulation of several ge-netic determinants, including phoA, encoding alkaline phos-phatase (49), pstSCAB and phnDCE, encoding components oftwo high-affinity, phosphate-specific transporter systems (48,49), phnF, encoding a negative regulator of phnDCE expres-sion (48), and senX3-regX3 itself (49, 66). Regulation by RegX3

TABLE 1. Conservation of M. tuberculosis H37Rv TCSSs in other Mycobacterium species

TCSSStatus in indicated Mycobacterium species and straina

Mab Mav Mb BCG Ml Mmar Mpar Msm Mtb MRa Mul

regX3-senX3 � � � � � � � � � � �Rv0600c-Rv0601c-tcrA � � � � � � ˆ � * � �phoP-phoR � � � � � � � � � � �narL-Rv0845 � � � � � � � � � � �prrA-prrB � � � � � � � � � � �mprA-mprB � � � � � � � � � � �kdpD-kdpE � � � � � � � � � � �trcR-trcS � � � � � � � � � � �dosR-dosS (dosT) � � � � � � � � � � �mtrA-mtrB � � � � � � � � � � �tcrX-tcrY � � � � � � � � � � �pdtaR-pdtaSb � � � � � � � � � � �

a Mab, Mycobacterium abscessus ATCC 19977; Mav, Mycobacterium avium 104; Mb, Mycobacterium bovis AF2122/97; BCG, Mycobacterium bovis BCG Pasteur1173P2; Ml, Mycobacterium leprae; Mmar, Mycobacterium marinum strain M; Mpar, Mycobacterium paratuberculosis K-10; Msm, Mycobacterium smegmatis; Mtb,Mycobacterium tuberculosis CDC1551; MRa, Mycobacterium tuberculosis H37Ra; Mul, Mycobacterium ulcerans Agy99. �, genes encoding the sensor kinase and responseregulator are present and are genetically linked; �, genes encoding both the sensor kinase and response regulator are absent; *, genes encoding the two sensor kinaseshave been fused and the gene encoding this fused sensor kinase is genetically linked to the response regulator; ˆ, one of the two sensor kinase genes is absent and thegene encoding the other kinase is genetically fused to the gene encoding the response regulator.

b pdtaR and pdtaS are not genetically linked but are present in all genomes listed.

TABLE 2. Sensor kinases in M. tuberculosis H37Rva

SK Gene Size ofSK (aa)

Predicted extracytoplasmicdomain(s)

(length of aa sequence)b

Site ofphosphorylation Metal ion cofactord Signale

SenX3 Rv0490 410 22–45, 288–410 His-167 Mg2� Pi (Msm); ND (Mtb)HK1 Rv0600c 168 None None Mg2� NDHK2 Rv0601c 157 1–44 His-131 NDPhoR Rv0758 485 38–157 His-259 Mn2� NDRv0845 Rv0845 426 63–66, 128–130, 175–183 ND ND NDPrrB Rv0902c 446 43–149 ND Mg2� or Mn2� NDMprB Rv0982 504 48–161 His-249 Mg2� or Mn2� SDS, alkaline pH, Triton X-100,

nutrient limitationKdpD Rv1028c 860 419–429, 453–475 ND ND NDTrcS Rv1032c 509 48–187 His-287c Ca2� or Mn2� NDDosS Rv3132c 578 None His-395 Mg2� Low O2, NO, CO, ascorbateDosT Rv2027c 573 None His-392 Mg2�, Mn2�, or Ca2�

MtrB Rv3245c 567 63–119 His-305c Mg2� NDTcrY Rv3764c 475 1–152 His-256c Mg2� or Ca2� NDPdtaS Rv3220c 501 None ND Mg2� ND

a aa, amino acids; ND, not defined.b The size(s) of the predicted extracytoplasmic domain(s) of each sensor kinase was determined using transmembrane prediction servers, including TMHMM2.0 (83)

and TMpred (67).c The site was predicted based on crystal structure, sequence alignment, or molecular modeling with other sensor kinases.d Magnesium (Mg), manganese (Mn), and calcium (Ca) divalent cations able to mediate autophosphorylation of the sensor kinase in vitro.e Environmental stimulus shown to activate the TCSS. Information in parentheses indicates that the signal activates only that species. Msm, Mycobacterium smegmatis;

Mtb, Mycobacterium tuberculosis; NO, nitric oxide; CO, carbon monoxide.

568 BRETL ET AL. MICROBIOL. MOL. BIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


is direct and is initiated through recognition of a loosely con-served, inverted-repeat element present in the promoter re-gions of these genes (Table 3) (49). Structural analysis ofRegX3 dimers supports the ability of this protein to recognizean inverted-repeat element when the protein is in an activestate (80). In the presence of elevated Pi levels, SenX3 acts asa RegX3 phosphatase, preventing the accumulation of phos-phorylated RegX3 and subsequent activation of downstreamgene targets (49). Interestingly, RegX3 and downstream tar-gets can be activated in the absence of SenX3, suggesting thatother mechanisms for RegX3 phosphorylation may exist (49).In M. tuberculosis, RegX3 both positively and negatively regu-lates a large and functionally diverse regulon comprised of�100 genes (115). Several of these genes encode hypotheticalproteins, while others are involved in important physiologicalactivities, including energy metabolism, cell envelope mainte-nance, and regulatory functions (115). RegX3 also positivelyregulates expression of itself in M. tuberculosis (Table 3) (115).The signals mediating activation of SenX3-RegX3 in M. tuber-culosis are unknown; however, this system is not responsive toexposure to antibiotics (kanamycin, tetracycline, gentamicin,isoniazid, ampicillin, rifampin), lysozyme, HCl, NaOH, di-methyl sulfoxide (DMSO), SDS, or H2O2 (115). WhetherSenX3 responds to Pi in M. tuberculosis as it does in M. smeg-matis has yet to be established.

In vitro and in vivo studies indicate that senX3-regX3 is re-quired for aspects of M. tuberculosis virulence (Table 4) (115,123, 124). An H37Rv �(senX3-regX3) mutant is attenuated forgrowth in the human THP-1 macrophage-like cell line and ingamma interferon (IFN-�)-activated, murine bone marrow-derived macrophages (115). Significant attenuation of this mu-tant strain is also observed in the lungs, spleens, and livers ofDBA/2 mice following intravenous infection with a high-doseinoculum (115) and in time-to-death assays following infectionof SCID mice (115). A similarly attenuated phenotype is ob-served in the lungs and spleens of BALB/c mice followingintravenous infection with individual �senX3 or �regX3 mutantstrains of H37Rv (123). Finally, an M. tuberculosis CDC1551regX3-transposon (Tn) insertion mutant is attenuated forgrowth in the lungs of BALB/c mice and Hartley guinea pigsfollowing a low-dose aerosol infection (124). Defining the spe-

cific contribution of the SenX3-RegX3 system to the M. tuber-culosis life cycle remains a priority and is expected to providenovel insights into determinants that contribute to the patho-genesis of this bacterium.

Rv0600c-Rv0601c-tcrA (Rv0602c)

Rv0600c-Rv0601c-tcrA encodes a genetically and structurallyunique TCSS that is present in the genomes of M. tuberculosiscomplex members but absent from other mycobacterial species(Table 1). Genes encoding this TCSS are part of a genomicisland that may have resulted from a horizontal gene transferevent between a Nocardia species and an ancestral Mycobac-terium species at some evolutionary point following the diver-gence of the M. tuberculosis lineage (10). Unlike in other TC-SSs in M. tuberculosis, Rv0600c and Rv0601c each encode oneof the two conserved domains typically present in SKs (141);Rv0600c (termed HK1 for histidine kinase 1) contains theATP-binding domain, while Rv0601c (termed HK2) contains anovel histidine phosphotransfer (Hpt) monodomain (141). Invitro biochemical studies of this system indicate that HK1 bindsATP with high affinity and mediates ATP hydrolysis (143).HK2 does not bind ATP, but it is phosphorylated at His-131 inthe presence of HK1 and ATP (Table 2) (143). Neither HK1nor HK2 undergoes autophosphorylation (143). Finally, HK2alone transfers phosphate to RR TcrA, establishing these threeproteins as a cognate TCSS (143). Functional interaction be-tween these proteins is also observed in vivo, as determined byyeast two-hybrid assays (142). It has been proposed that HK2exists as a dimer and that each monomer of HK2 is anchoredin the membrane by a single transmembrane domain. Follow-ing recognition of an as-yet-unknown environmental signal,HK1 phosphorylates HK2, followed by phosphorelay fromHK2 to TcrA (142).

Beyond analyses of the signaling cascade described above,very little is known about this TCSS. Indirect evidence suggeststhat HK2 is downregulated during the growth of M. tubercu-losis under hypoxic conditions (138) and upregulated by tetra-hydrolipstatin, a small-molecule antimicrobial agent (167).Neither observation has been expanded upon to date. It is alsointeresting to note that in M. tuberculosis strain CDC1551,

TABLE 3. Response regulators in M. tuberculosis H37Rva

RR Gene Size ofRR (aa) Recognition sequence Site of

phosphorylationNo. of genes

regulatedc Autoregulated

RegX3 Rv0491 227 23-bp inverted repeat Asp-52 �100 YesTcrA Rv0602c 253 ND ND ND NDPhoP Rv0757 247 22-bp direct repeat Asp-71 �150 YesNarL Rv0844 216 ND ND ND NDPrrA Rv0903c 236 ND Asp-58b ND YesMprA Rv0981 230 17-bp direct repeat Asp-48 �200 YesKdpE Rv1027c 226 ND ND ND NDTrcR Rv1033c 257 28-bp AT-rich region Asp-82b �50 YesDosR Rv3133c 217 18/20-bp palindromic sequence Asp-54 �100 YesMtrA Rv3246c 228 20-bp direct repeat Asp-53 4d YesTcrX Rv3765c 234 ND Asp-54b ND NDPdtaR Rv1626 205 ND Asp-65b ND ND

a aa, amino acids; ND, not defined.b The site was predicted based on crystal structure, sequence alignment, or molecular modeling with other response regulators.c Number of genes in the regulon in M. tuberculosis, as determined by DNA microarrays.d The number of genes in the regulon was determined via quantitative reverse transcription-PCR (qRT-PCR).



http://mm

br.asm.org/

Dow

nloaded from


TA

BL

E4.

Invi

tro

and

invi

vovi

rule

nce

char

acte

rist

ics

ofM

.tub

ercu

losi

sT

CSS

mut

ants

a

TC

SSM

utat

ion(

s)St

rain

bIn

vitr

om

odel

(s)c

Invi

tro

phen

otyp

e(s)

Invi

vom

odel

(s)d

Rou

te(s

)eIn

vivo

phen

otyp

e(s)

Invi

voph

enot

ype(

s)co

mpl

emen

tedf

Ref

eren

ce

senX

3-re

gX3

�(s

enX

3-re

gX3)

H37

Rv

mB

MD

Man

dhT

HP1

Att

enua

ted

for

both

mod

els

DB

A/2

mic

ean

dSC

IDm

ice

i.v.f

orbo

thm

odel

sA

tten

uate

din

lung

,spl

een,

and

liver

for

DB

A/2

mic

e,de

laye

dtim

eto

deat

hfo

rSC

IDm

ice

ND

for

both

mod

els

115

�se

nX3

or�

regX

3H

37R

vN

DN

DB

AL

B/c

mic

ei.v

.A

tten

uate

din

lung

and

sple

enY

es12

3

regX

3::T

nC

DC

1551

ND

ND

BA

LB

/cm

ice

and

guin

eapi

gsA

eros

olfo

rbo

thm

odel

s

Att

enua

ted

inlu

ngfo

rbo

thm

odel

sN

Dfo

rbo

thm

odel

s12

4

Rv0

600c

-R

v060

1c-t

crA

No

mut

ants

cons

truc

ted

phoP

-pho

Rph

oP::K

mr

MT

103

mB

MD

MA

tten

uate

dB

AL

B/c

mic

ei.v

.A

tten

uate

din

lung

,spl

een,

and

liver

ND

117

phoP

::Km

rM

T10

3N

DN

DSC

IDm

ice

i.v.

No

deat

hY

es10

2ph

oP::K

mr

H37

Rv

hTH

P1,m

AM

J2-C

11,

and

mJ7

74A

.1A

tten

uate

dfo

ral

lth

ree

mod

els

C57

/BL

6m

ice

Aer

osol

Att

enua

ted

Yes

168

�ph

oPR

::Hyg

r

�ph

oP::H

ygr

1237

hTH

P1A

tten

uate

dN

DN

DN

DN

D53

phoP

::Km

rM

T10

3N

DN

DB

AL

B/c

mic

ei.t

.N

ode

ath;

atte

nuat

edin

lung

ND

2ph

oP::K

mr

MT

103

NR

K-4

9F,a

ndH

Lan

dM

RC

-5N

oat

tenu

atio

nfo

rN

RK

-49F

,de

crea

sed

cyto

toxi

city

for

HL

and

MR

C-5

ND

ND

ND

ND

41

narL

-Rv0

845

�na

rLH

37R

vm

BM

DM

No

phen

otyp

eSC

IDm

ice

i.v.

No

phen

otyp

eN

D11

4

prrA

-prr

Bpr

rA::T

n956

3M

T10

3m

BM

DM

Att

enua

ted

earl

yB

AL

B/c

mic

ei.v

.and

aero

sol

No

phen

otyp

efo

rei

ther

rout

eY

esfo

ri.v

.,N

Dfo

rae

roso

l38

mpr

A-m

prB

mpr

A::K

mr

H37

Rv

J774

A.1

and

mB

MD

MH

yper

viru

lent

for

both

mod

els

BA

LB

/cm

ice

i.v.

Att

enua

ted

inlu

ngan

dsp

leen

,no

phen

otyp

ein

liver

ND

183

�m

prA

BH

37R

vH

uman

mon

ocyt

esH

yper

viru

lent

ND

ND

ND

ND

113

kdpD

-kdp

E�

kdpD

EH

37R

vN

DN

DSC

IDi.v

.E

nhan

ced

time

tode

ath

ND

114

trcR

-trc

Str

cS::T

nM

T10

3m

BM

DM

No

phen

otyp

eC

57/B

L6

Aer

osol

No

phen

otyp

eN

D38

�tr

cSH

37R

vN

DN

DSC

IDi.v

.E

nhan

ced

time

tode

ath

ND

114

dosR

(dev

R)-

dosS

(dev

S)-d

osT

devR

::Km

rH

37R

vH

uman

mon

ocyt

esN

oph

enot

ype

Gui

nea

pigs

s.c.

Att

enua

ted

insp

leen

ND

98�

dosR

TH

37R

vM

ouse

hollo

w-fi

ber

gran

ulom

am

odel

No

phen

otyp

eC

57B

L/6

and

BA

LB

/cm

ice,

guin

eapi

gs,

and

rabb

its

Aer

osol

for

allt

hree

mod

els

Att

enua

ted

inlu

ngs

and

sple

ens

ofC

57/B

L6

mic

e,at

tenu

ated

inlu

ngan

dsp

leen

for

guin

eapi

gs,

slig

htvi

rule

nce

defe

ctin

lung

sfo

rra

bbits

No

for

C57

BL

/6an

dB

AL

B/c

mic

ean

dgu

inea

pigs

,ND

for

rabb

its

31

�de

vRH

37R

vm

BM

DM

Hyp

ervi

rule

ntSC

IDm

ice

and

DB

A/2

mic

ei.v

.for

both

mod

els

Enh

ance

dtim

eto

deat

hfo

rSC

IDm

ice,

hype

rvir

ulen

tin

lung

,spl

een,

and

liver

for

DB

A/2

mic

e

Yes

for

SCID

mic

e,N

Dfo

rD

BA

/2m

ice

114

�R

v313

4c-d

osS

H37

Rv

ND

ND

C57

BL

/6m

ice

Aer

osol

Slig

hthy

perv

irul

ent

phen

otyp

eea

rly

insp

leen

;no

phen

otyp

ein

lung

Yes

9

�do

sR::K

mr

H37

Rv

ND

ND

C57

BL

/6m

ice,

DB

A/2

Jm

ice,

and

C3H

e/F

eJm

ice

Aer

osol

for

allt

hree

mod

els

No

phen

otyp

ein

lung

orsp

leen

for

any

mod

elN

D12

7



http://mm

br.asm.org/

Dow

nloaded from


Rv0600c and Rv0601c have been fused to form a single geneproduct (141); however, the significance of this difference fromother strains remains unclear. In addition to the lack of knowl-edge regarding recognized stimuli, nothing is currently knownabout the genes regulated by this TCSS or the contribution ofthis system to aspects of M. tuberculosis physiology or patho-genesis.

phoP (Rv0757)-phoR (Rv0758)

The phoP-phoR system was initially named for its sequencesimilarity to PhoP-PhoR from B. subtilis and PhoP-PhoQ fromSalmonella enterica serovar Typhimurium (30). This system iscomprised of RR PhoP and SK PhoR. phoP and phoR aregenetically linked and appear to be present in all Mycobacte-rium species examined to date except M. leprae (Table 1). In M.tuberculosis, PhoR is phosphorylated at residue His-259 follow-ing its activation (Table 2) (58). Once phosphorylated, PhoRtransfers this phosphate to Asp-71 of PhoP (Table 3) (58).

Biochemical and genetic studies in M. tuberculosis indicatethat phoP-phoR is autoregulated and that expression of thesegenes is transcriptionally coupled (Table 3) (52). However,phoR may also be expressed from its own promoter element(52). PhoP positively regulates the expression of itself andphoR through recognition of a 22-bp sequence that containstwo 9-bp direct repeat elements present in the phoP upstreamregion (Table 2) (52, 57). PhoP likely binds to its consensussequence in a cooperative manner, binding one direct repeatand then the other in a head-to-head fashion (57). Phosphor-ylation of PhoP results in a conformational change that isthought to enhance protein dimerization and subsequentDNA-binding stability (147). The structure of the PhoP DNA-binding domain has been solved. It contains a winged helix-turn-helix motif that includes three �-helices flanked by two�-sheets (169). Several amino acids within the �-helices appearcritical for DNA binding and sequence recognition, includingAsn-212, Val-213, Glu-215, Ser-219, Tyr-220, Tyr-222, Tyr-223,and Lys-224 (33). Interestingly, PhoP from the attenuated M.tuberculosis strain H37Ra (PhoPH37Ra) contains a single nu-cleotide polymorphism that results in an amino acid substitu-tion at position 219 from serine to leucine (27). Relative toPhoPH37Rv, PhoPH37Ra exhibits a significantly diminished ca-pacity to bind its recognition sequence from the phoP upstreamregion (27, 87). Consistent with this observation, a large num-ber of genes that are downregulated in a �phoP mutant of M.tuberculosis H37Rv are also expressed at lower levels in wild-type H37Ra (87). Additionally, of the 50 genes that are differ-entially expressed between wild-type M. tuberculosis H37Rvand wild-type M. tuberculosis H37Ra during intracellulargrowth in murine bone marrow-derived macrophages, 12 ofthem are regulated by PhoP-PhoR (91). The effect of thismutation in M. tuberculosis H37Ra on transcriptional regula-tion at the phoP-phoR promoter region remains unclear, asthere are conflicting reports regarding whether PhoPH37Ra

positively or negatively regulates its own expression (51, 52, 58,168).

The generation of defined mutations in phoP and/or phoRhas provided important insights into the genes regulated bythis TCSS and the role that this system plays in the physiologyand pathogenesis of M. tuberculosis. PhoP-PhoR regulates am

trA

-mtr

Bm

trA (o

vere

xpre

ssio

nco

nstr

uct)

H37

Rv

mJ7

74A

.1an

dhu

man

peri

pher

albl

ood

mon

ocyt

es

Att

enua

ted

for

both

mod

els

C57

BL

/6m

ice

Aer

osol

Att

enua

ted

inlu

ngs

and

sple

ens

ofm

ice

atla

ter

time

poin

tspo

stin

fect

ion

NA

43

mtr

A (ove

rexp

ress

ion

cons

truc

t)

H37

Rv

hMD

MA

tten

uate

dN

DN

DN

DN

D12

1

tcrX

-tcr

Y�

tcrX

Y::H

ygr

H37

Rv

ND

ND

SCID

mic

ei.v

.E

nhan

ced

time

tode

ath

ND

114

pdta

R-p

dtaS

�R

v322

0cH

37R

vN

DN

DSC

IDm

ice

i.v.

No

phen

otyp

eN

D10

7

aN

D,n

otde

term

ined

.b

M.t

uber

culo

sis

stra

inba

ckgr

ound

inw

hich

the

mut

atio

nw

asge

nera

ted

orth

eov

erex

pres

sion

cons

truc

tw

asin

trod

uced

.c

mB

MD

M,m

urin

ebo

nem

arro

w-d

eriv

edm

acro

phag

es;h

TH

P-1,

hum

anT

HP-

1m

acro

phag

e-lik

ece

lllin

e;m

AM

J2-C

11,m

urin

eal

veol

arm

acro

phag

e-lik

ece

lllin

eA

MJ2

-C11

;mJ7

74A

.1,m

urin

eJ7

74A

.1m

acro

phag

e-lik

ece

lllin

e;N

RK

-49F

,nor

mal

rat

kidn

eyfib

robl

ast-

like

cell

line

NR

K-4

9F;H

L,h

uman

lung

fibro

blas

ts;M

RC

-5,h

uman

feta

llun

gfib

robl

ast-

like

cell

line

MR

C-5

.d

SCID

,sev

ere

com

bine

dim

mun

odefi

cien

cy.

ei.v

.,in

trav

enou

s;i.t

.,in

trat

rach

eal;

s.c.

,sub

cuta

neou

s.fY

es,t

hem

utat

ion

was

gene

tical

lyco

mpl

emen

ted

and

the

resu

lting

phen

otyp

ew

asfu

llyor

part

ially

rest

ored

toth

atof

the

pare

ntal

wild

-typ

est

rain

;no,

the

mut

atio

nw

asge

netic

ally

com

plem

ente

d,bu

tth

ere

sulti

ngph

enot

ype

was

not

that

ofth

epa

rent

alw

ild-t

ype

stra

in;N

A,n

otap

plic

able

.



http://mm

br.asm.org/

Dow

nloaded from


diverse regulon of �150 genes in M. tuberculosis (Table 3) (51,168). Functional gene classes regulated by PhoP include thoseinvolved in lipid metabolism, general metabolism, and respi-ration (51, 168). A number of genes encoding membrane/secreted proteins and those comprising the initial hypoxic re-sponse (dosRS and constituents of its regulon) and enduringhypoxic response are also differentially regulated, as are genesfrom the PE/PPE/PE-PGRS protein families (51, 168). Inter-estingly, a number of genes from the virulence-associated re-gion of difference 1 (RD1), including espB and espR, are alsoregulated by PhoP (51, 91). Recent studies have indicated thatboth EspB and EspR are necessary for the secretion of pro-teins (notably, the major antigens ESAT-6 and CFP-10) via theESX-1 secretion pathway (104, 120). Consistent with this ob-servation, M. tuberculosis phoP::Kmr mutants and wild-type M.tuberculosis H37Ra synthesize ESAT-6 but only marginally se-crete this protein. These alterations in ESAT-6 secretion alsoresult in low T-cell responses to this protein following infectionby these strains (45).

Given the diverse regulon controlled by PhoP-PhoR, it is notsurprising that M. tuberculosis phoP and/or phoR mutants ex-hibit pleiotropic phenotypes, compared to their wild-typecounterparts (see reference 129 for a review). Morphologi-cally, these mutants are smaller than the wild type and exhibitaltered colony characteristics, including inabilities to form ser-pentine cords, fix neutral red, and stain acid-fast (53, 117, 168).Some of these differences are due to the lack of specific lipidswithin the cell membrane/cell wall of the phoP mutant, includ-ing 2,3-di-O-acyl-trehaloses (DATs), polyacyltrehaloses(PATs), and sulfolipids (SLs) (53, 168). Wild-type M. tubercu-losis H37Ra exhibits similar colony characteristics and alsolacks DATs, PATs, and SLs in its cell membrane/cell wall (27).Interestingly, these phenotypes are reversed by expression ofM. tuberculosis phoPH37Rv in H37Ra (27). M. tuberculosis phoPmutants are also more sensitive than the wild type to cumenehydrogen peroxide (an organic peroxide), CdCl2 (a superoxidegenerator and toxic heavy metal), and several cell wall-perturb-ing antibiotics, including vancomycin and cloxacillin (168).

Several lines of evidence also support a role for PhoP-PhoRin the virulence of M. tuberculosis (Table 4). phoP expression isupregulated following exposure of M. tuberculosis to iron(100), a metal that is essential for the in vivo growth of numer-ous bacterial pathogens, including M. tuberculosis. In addition,several genomic alterations in the phoP-phoR locus have oc-curred in various M. bovis BCG derivatives, perhaps explain-ing, in part, the virulence variability observed in these vaccinestrains (17, 89). Similarly, a multidrug-resistant strain of M.bovis BCG that carries an IS6110 insertion sequence in thephoP promoter region that enhances its transcription was re-sponsible for a large outbreak of tuberculosis in Spain (148).Importantly, phoP and phoP-phoR mutants of M. tuberculosisare attenuated for growth in various cultured or primary celltypes, including murine bone marrow-derived macrophages(117), murine alveolar macrophages (168), murine J774A.1macrophage-like cells (168), and human THP-1 macrophage-like cells (53, 168) (Table 4). Interestingly, this attenuationphenotype may be cell-type specific, as a phoP mutant of M.tuberculosis is not altered for growth in fibroblasts (41). Com-pared with wild-type M. tuberculosis, phoP mutants also exhibitlower bacterial burdens in target organs of infection when

assayed in several animal model systems of infection, includingmice and guinea pigs (Table 4) (1, 102, 168). Finally, miceinfected with phoP or phoP-phoR mutants also exhibit delayedtimes to death compared to those of mice infected with thewild-type parent (1, 102). Thus, a loss of PhoP-PhoR is detri-mental to the growth/survival of M. tuberculosis in vivo.

The observed attenuation of phoP and/or phoR mutantstrains has led to investigations of the potential utility of an M.tuberculosis phoP mutant (SO2 strain, a derivative of clinicalisolate MT103) as a possible live-vaccine candidate. Subcuta-neous vaccination of BALB/c mice with SO2 provides the samelevel of protection as vaccination with M. bovis BCG followingintravenous or intratracheal challenge with M. tuberculosisH37Rv (1, 102). Similarly, subcutaneous vaccination with SO2affords guinea pigs an increased level of protection against ahigh-dose aerosol challenge with M. tuberculosis H37Rv com-pared to that of animals vaccinated with BCG (102). Vaccina-tion with SO2 also induces less lung pathology than vaccinationwith BCG (102) following challenge with H37Rv and reducesthe production of several cytokines, including interleukin 4(IL-4), tumor necrosis factor alpha (TNF-�), and IFN-� (1).Furthermore, vaccination results in high levels of induciblenitric oxide synthase (iNOS) (1), increases numbers of CD4�

and CD8� T cells in the spleen, and increases the proportionof CD4�/CD8� cells that express IFN-� (1, 102). Importantly,the SO2 strain is fully susceptible to four frontline antimyco-bacterial drugs (ethambutol, isoniazid, rifampin, and strepto-mycin) in vivo and does not induce adverse effects followingvaccination in guinea pigs previously exposed to M. tuberculosis(20). The SO2 strain is also protective in rhesus monkeys,suggesting real promise that a phoP mutant could be furtherevaluated for use as an effective live vaccine in humans (163).

Despite extensive characterization of this TCSS, the environ-mental signal(s) to which PhoP-PhoR is responsive remains un-known. phoP mutants of M. tuberculosis are unable to grow in theabsence of the divalent cation magnesium (168). Interestingly,supplementing tissue culture medium with Mg2� partially re-stores intracellular growth characteristics to phoP mutants of M.tuberculosis within THP-1 macrophage-like cells (168). Whethermagnesium is the signal to which PhoP-PhoR responds has yet tobe delineated. Clearly, further work to define the molecularmechanism(s) underlying the attenuation of these strains in vitroand in vivo is required before phoP mutants can be fully assessedfor their potential as vaccine candidates.

narL (Rv0844c)-Rv0845

While genes encoding homologs of NarL-Rv0845 exist inmost Mycobacterium species sequenced to date, except M. lep-rae (Table 1), very little is known about the biochemical activityor functional significance of the narL-Rv0845 TCSS. In M.tuberculosis, narL is divergently transcribed from Rv0845, andits gene product exhibits sequences and structures in the re-ceiver domain that are homologous to those of NarL from E.coli (136). In E. coli, narL encodes the RR component of theNarQ-NarL TCSS. This system regulates genes in response tonitrate concentrations, including determinants involved in ni-trate metabolism during anaerobic respiration (150). WhetherRv0845 senses a similar signal remains to be demonstrated,



http://mm

br.asm.org/

Dow

nloaded from


and characterization of NarL-Rv0845 as a bona fide TCSS inmycobacteria awaits.

Several studies suggest that this system may not contributeto the virulence of M. tuberculosis. Expression of narL is belowdetection limits following infection of M. tuberculosis H37Rv inhuman macrophages (60), and exposure of peripheral bloodmononuclear cells isolated from tuberculosis (TB) patients torecombinant NarL does not stimulate significant levels ofIFN-� production (160). Importantly, a �narL mutant of M.tuberculosis H37Rv exhibits no apparent growth or survivaldefect following infection of activated murine bone marrow-derived macrophages (Table 4) (114), and SCID mice infectedwith this strain display time-to-death kinetics similar to thoseof the wild-type H37Rv parent (Table 4) (114). Thus, furtherstudies are needed to better define the NarL-Rv0845 TCSSand to delineate the role of this TCSS in M. tuberculosis phys-iology and pathogenesis.

prrA (Rv0903c)-prrB (Rv0902c)

prrA-prrB is one of four TCSSs that is present and geneticallylinked in all Mycobacterium species examined to date (Table 1).Bioinformatic and biophysical studies indicate that PrrA and PrrBpossess amino acid sequences and structural motifs highly char-acteristic of RRs and SKs, respectively (110, 111). In vitro, PrrA-PrrB has been biochemically characterized and shown to functionas an intact TCSS. The cytoplasmic domain of SK PrrB autophos-phorylates in the presence of Mn2� or Mg2�, and this domainparticipates in phosphotransfer reactions with RR PrrA (39, 111).However, sites of phosphorylation for both PrrB and PrrA haveyet to be defined. Electromobility shift assays (EMSAs) and re-porter fusions have demonstrated that PrrA binds to its ownpromoter region and positively autoregulates its own expression(Table 3) (39). The binding of DNA by PrrA is phosphorylationindependent, but it is enhanced by phosphorylation (39). It hasbeen suggested that PrrA initially assumes an inhibitory, closedconformation that is poorly phosphorylatable yet shifts to a moreopen, DNA-binding-favored conformation upon phosphorylation(110) (6). The stimuli recognized by PrrB and the genes regulatedby PrrA have yet to be defined.

Several studies indicate that prrA-prrB may play a role in M.tuberculosis virulence. prrA-prrB is expressed by M. tuberculosisH37Rv during growth in human peripheral blood monocyte-derived macrophages but not during growth in standard labora-tory medium, indicating that these genes may be specifically up-regulated following infection (54). In agreement with theseobservations, infection of murine bone marrow-derived macro-phages with an M. bovis BCG derivative containing an M. tuber-culosis prrA (prrAM. tb)::gfp promoter fusion plasmid is transientlyupregulated at 4 h postinfection (38). Consistent with these ob-servations, while strain MT103 with a Tn9563 transposon inser-tion mutation of prrA exhibits no phenotype during the growth ofM. tuberculosis in 7H9 or Sauton’s medium, this mutant strain isattenuated during initial time points following infection of murinebone marrow-derived macrophages in vitro (Table 4) (38). Inter-estingly, growth of the mutant recovers to wild-type levels at latertime points, suggesting that signaling through PrrA-PrrB may beimportant only for early stages of infection in this cell type. How-ever, when assayed in vivo, the prrA::Tn mutant exhibited nodifferences in virulence levels in the lungs, spleens, or livers of

BALB/c mice following either intravenous or aerosol infectionfrom those of the wild-type parent (Table 4) (38). Thus, the roleof prrA-prrB in M. tuberculosis pathogenesis remains unclear.

mprA (Rv0981)-mprB (Rv0982)

The Rv0981-Rv0982 TCSS system was originally describedas being necessary for the establishment and maintenance ofpersistent infection by M. tuberculosis in mice and was there-fore named mpr for mycobacterium persistence regulator(183). This system consists of RR MprA and SK MprB. Genesencoding these determinants are highly conserved in all Myco-bacterium species, including M. leprae (Table 1) (183). In ad-dition to mprA and mprB being genetically coupled, their chro-mosomal positioning adjacent to pepD (encoding an HtrA-likeserine protease) and moaB2 (encoding a predicted molybdot-perin biosynthetic gene) is also conserved among these species(183). Biochemical studies have established MprA and MprB as afunctional TCSS in M. tuberculosis H37Rv both in vitro and invivo. MprB autophosphorylates at residue His-249 and transfersthis phosphate to MprA at residue Asp-48 (Tables 2 and 3) (184).In addition to having kinase activity, MprB acts as an MprAphosphatase to regulate levels of phospho-MprA within the cell(184). In vivo, activation of downstream signaling pathways byMprA is dependent on the extracytoplasmic domain of MprB, asstrains of M. tuberculosis expressing mutant mprB alleles carryingdeletions in this domain are unable to initiate the signaling cas-cade (63).

The generation of mprA::Kmr and �mprAB mutant strains ofM. tuberculosis H37Rv has led to the identification of determi-nants regulated by the mprAB TCSS and the environmentalsignals to which this system is responsive (63, 113). Compara-tive microarray analyses between wild-type and mutant strainsindicate that MprA both positively and negatively regulates adiverse regulon of �200 genes (63, 113). MprA positively reg-ulates its own expression and that of downstream genes pepD andmoaB2 by directly binding a 17-bp sequence containing two 6-bpdirect-repeat motifs separated by 5 nucleotides (Table 3) (64, 112,173). MprA also directly regulates downstream genes pepD andmoaB2 by binding three tandemly positioned MprA recognitionsequences located in the end of mprB and extending into themprB-pepD intergenic region (64, 173), and MprA directly regu-lates, apart from its own locus, the expression of other determi-nants, including those encoding the Acr2 alpha-crystallin-like pro-tein (112), a predicted 18-kDa chaperone.

Defining the core regulon directly controlled by MprA in M.tuberculosis is complicated by the fact that MprA directly reg-ulates the expression of two ECF sigma factors, sigE and sigB,which have extensive regulons of their own (36, 63, 113). Reg-ulation of sigE by MprA seems especially critical, since MprAand SigE participate in a positive-feedback loop in response tocell wall/membrane-perturbing agents, such as SDS (63, 101,113, 173, 174). Consistent with this observation, both mprA andsigE mutants exhibit increased sensitivity to SDS (101, 173).SigE also upregulates other determinants that may influenceMprAB activation indirectly. ppk1 is regulated by SigE andgenerates polyphosphate molecules that are phosphoryl do-nors for MprB activation under conditions of low ATP con-centrations (156). Additionally, the MprA-regulated determi-nant PepD associates with and proteolytically cleaves the



http://mm

br.asm.org/

Dow

nloaded from


SigE-regulated determinant Rv2744c (the 35-kDa antigen)(174), an interaction that influences susceptibility to cell wall-targeting antibiotics (174). MprA is also responsive to stimuliother than SDS which are likely to affect cell wall/cell mem-brane homeostasis, including Triton X-100, alkaline pH, andnutrient limitation (12, 63, 113).

While several in vitro conditions activate MprA-MprB sig-naling, mprA-mprB expression is also upregulated in M. tuber-culosis in vivo. mprA-mprB is part of an in vivo-expressedgenomic island (iVEGI) of 49 genes that is highly upregulatedin M. tuberculosis within the lungs of BALB/c mice duringacute and chronic stages of infection (157, 158). mprA is alsoupregulated in M. tuberculosis within an artificial hollow-fibergranuloma model system (74). Finally, mprA expression is up-regulated following infection by M. tuberculosis in humanmonocyte-derived macrophages (60). Consistent with activa-tion of MprA-MprB by in vivo signals, mprA and/or mprA-mprB mutants of M. tuberculosis also exhibit altered virulencecharacteristics in model systems of infection (Table 4).mprA::Kmr and �mprAB mutants of M. tuberculosis H37Rv arehypervirulent in macrophages (113, 183) and are attenuatedfor virulence within the lungs and spleens of BALB/c micefollowing intravenous infection (183). Thus, MprA-MprB tran-scriptionally regulates adaptation programs in response to sev-eral environmental stimuli and is required by M. tuberculosisfor aspects of virulence in vivo.

kdpD (Rv1028c)-kdpE (Rv1027c)

SK KdpD and RR KdpE have been named based on thesimilarity of their amino acids to those of the KdpD-KdpETCSS in E. coli and on their conserved genetic location adja-cent to determinants (kdpFABC) encoding a predicted potas-sium uptake system. kdpD-kdpE is present in many species ofMycobacterium but is absent from M. leprae and M. ulcerans(Table 1). In M. tuberculosis and M. bovis, kdpD-kdpE is diver-gently transcribed from kdpFABC, while these genes are tran-scriptionally coupled in other species, including M. avium, My-cobacterium marinum, and M. smegmatis.

Only a few studies have examined the biochemical activity ofKdpD-KdpE in M. tuberculosis or the importance of this sys-tem in the physiology and/or pathogenesis of the bacterium.Expression of an M. tuberculosis kdpF promoter-lacZ fusion isreduced in M. smegmatis and M. tuberculosis when grown underconditions of increasing K� concentrations (151). However, itis unclear whether K� is the stimulus to which kdpD-kdpE isresponsive. Similarly, while yeast two-hybrid assays indicatethat KdpD and KdpE physically interact with each other (145,151), it is not yet established whether these proteins function asa cognate SK/RR pair, mediate phosphotransfer reactions be-tween each other, or autogenously regulate their own expres-sion. KdpD has been shown to associate physically with twomembrane lipoproteins, LprF (Rv1368) and LprJ (Rv1690)(151). While the significance of these interactions awaits fur-ther delineation, overexpression of lprF or lprJ enhances ex-pression of the kdpF promoter-lacZ fusion, suggesting thatthese proteins may modulate the activity of KdpD-KdpE sig-naling (151). KdpD has an unusual structure for an SK, withlarge N- and C-terminal cytoplasmic domains separated byfour small, transmembrane-spanning regions (151). This struc-

ture is consistent with a model that predicts that KdpD likelysenses osmotic changes in the cytoplasm. Physiological stimuliother than K� may also regulate activation of KdpD-KdpE.kdpD is induced in M. tuberculosis H37Rv under conditions ofstarvation (12), and kdpE is required for the optimal growth ofM. tuberculosis in vitro (134). Importantly, kdpD-kdpE may alsocontribute to M. tuberculosis virulence, as these genes are dif-ferentially expressed in human macrophages (60), and SCIDmice intravenously infected with an M. tuberculosis �kdpDEmutant exhibit enhanced times to death (Table 4) (114).Clearly, additional work is needed to further characterize thesegenes and delineate mechanisms underlying the hypervirulentphenotype seen in the mouse model of infection.

trcR (Rv1033c)-trcS (Rv1032c)

The trcR-trcS system is comprised of SK TrcS and RR TrcRand was one of the first M. tuberculosis systems shown tofunction as a cognate TCSS in M. tuberculosis (62). In vitro,TrcS autophosphorylates in the presence of Ca2� or Mn2� andacts as a TrcR kinase (62). Amino acid residues that are phos-phorylated in both TrcS and TrcR have yet to be defined.

trcR and trcS are genetically linked, transcriptionally cou-pled, and autogenously regulated (Table 3) (59). TrcR bindsdirectly to its own upstream region at an AT-rich stretch ofnucleotides (Table 4) (59). A bioinformatic search of the M.tuberculosis genome identified a similar 28-bp, AT-rich se-quence upstream of Rv1057, encoding a seven-bladed �-pro-peller protein (61). While TrcR negatively regulates expressionof Rv1057 by directly binding to the promoter region (61), thefunction of this gene product, as well as the significance of thisregulation, awaits further investigation. Microarray analysescomparing M. tuberculosis H37Rv and a �trcS mutant haveidentified approximately 50 genes that are differentially regu-lated by this TCSS (172). Whether these genes are directlyregulated by TrcR via binding to a similar AT-rich sequencehas yet to be determined. trcR-trcS is expressed during thegrowth of M. tuberculosis in broth medium and shortly afterinfection in primary human peripheral blood-derived macro-phages (59). These genes are also upregulated during thegrowth of M. tuberculosis in an artificial hollow-fiber granu-loma model system (74).

While trcR-trcS is expressed by M. tuberculosis in physiolog-ically relevant environments, there are conflicting reports as tothe importance of this system in the virulence of the tuberclebacillus (Table 4). While a trcS transposon insertion mutant ofM. tuberculosis MT103 is unaltered for growth/survival in mu-rine bone marrow-derived macrophages and in an aerosol in-fection model of C57BL/6 mice (38), a �trcS mutant of M.tuberculosis H37Rv is slightly hypervirulent in SCID mice(114). Therefore, further work on this system is still needed toidentify the stimulus processed by TrcS, to characterize theregulon controlled by TrcR, and to delineate the contributionof these proteins to M. tuberculosis host-pathogen interactions.

devR-devS-Rv2027c or dosR-dosS-dosT

The RR comprising the devR-devS-Rv2027c TCSS was ini-tially identified in a screen for determinants in the virulent M.tuberculosis strain H37Rv that were differentially expressed in



http://mm

br.asm.org/

Dow

nloaded from


its avirulent counterpart H37Ra, prompting its initial designa-tion as devR (differentially expressed in the virulent strain) (34,79). Subsequent studies have demonstrated a role for devR asa regulator of dormancy survival (i.e., dos) in M. bovis BCGwhen grown in the Wayne culture model for nonreplicatingpersistence (15, 16), leading to the alternate designation of thisdeterminant as dosR. Both gene designations remain in usetoday. Here, the dos designation will be utilized, given thedemonstrated functional role for these genes in mediating thetransition to nonreplicating persistence in vitro.

DosR-DosS-DosT represents an unusual TCSS in that itconsists of two soluble, full-length SKs (DosS and DosT) anda single RR (DosR). In M. tuberculosis, dosR and dosS aregenetically linked and transcriptionally coupled with eachother, as well as with upstream gene Rv3134c (34). In contrast,dosT is encoded elsewhere in the genome. While dosR anddosS are conserved and tandemly arranged in many mycobac-terial species (except M. leprae and M. ulcerans) (Table 1),dosT appears to be less well conserved. Interestingly, M. smeg-matis contains homologs of DosR and DosT but lacks a DosShomolog. Biochemical and genetic studies indicate that DosR-DosS-DosT constitutes a functional TCSS. DosS and DosTautophosphorylate at conserved histidine residues (DosS atHis-395 and DosT at His-392), and both proteins phosphorelayto Asp-54 of DosR (Tables 2 and 3) (125, 130–132).

The DosR-DosS-DosT system has been extensively studiedover the years due, in part, to its association with bacterialdormancy and the demonstration that environmental signalsactivating this system in vitro are likely to be encountered by M.tuberculosis in vivo. dosRS is upregulated in M. tuberculosis andM. bovis BCG when bacteria are cultured under conditions oflow-oxygen tension (Table 2), an observation that was initiallyreported in 2001 (16, 138). A similar phenotype is also ob-served when these M. tuberculosis genes and their upstreampromoter region are introduced into surrogate host M. smeg-matis cells that are cultured under microaerophilic or anaero-bic conditions (5, 103). Exposure of M. tuberculosis H37Rv tohypoxia alters the expression of more than 100 genes (138),including a subset of 48 genes whose upregulation is dependenton DosR (Table 3) (116). The same 48-gene regulon is alsoupregulated following exposure of M. tuberculosis to nitric ox-ide (NO) (165) or during the transition of M. tuberculosis fromaerobic respiration to nonreplicating persistence induced viagradual oxygen depletion (165, 166, 170). The 48 genes up-regulated by DosR have been collectively termed the “dor-mancy regulon.” They are generally clustered into a smallnumber of discrete transcriptional units that are scatteredaround the chromosome (116, 165). This type of arrangementmay facilitate a more rapid, pronounced, and coordinated re-sponse to environmental stimuli. It is predicted that DosRdirectly regulates the vast majority of genes comprising thisregulon. EMSAs, DNase I footprinting, and gene reporterexpression studies carried out thus far generally support thiscontention (4, 24–26, 42, 55, 116, 125, 162). In particular, DosRrecognizes a fairly conserved 18/20-bp palindromic sequencethat is present with some variation upstream of the transcrip-tional units for these genes (Table 3) (4, 25, 26, 42, 55, 116, 125,162). Similar sequences are also found upstream of other genesoutside this regulon, although their regulation by DosR re-mains unclear (116, 185). Promoter regions from DosR regu-

lon genes often contain two or more sets of recognition se-quences, and cooperative binding by DosR to these sites isnecessary for their full induction (24–26). Binding by DosR toits recognition sequences is mediated primarily through theC-terminal domain (176, 177), and amino acid residues withinthis region that are important for contacting DNA have beendefined and characterized (55, 176). However, the N-terminaldomain also regulates aspects of DosR binding and subsequentgene regulation. This domain influences the ability of DosR toact cooperatively (47), particularly at sites in which the recog-nition motif varies from the optimal consensus (25, 26). How-ever, the N-terminal domain of DosR also regulates DNAbinding by “locking” the C-terminal domain in a conformationthat limits its ability to bind DNA in the absence of phosphor-ylation (47, 177). Interestingly, a phenylcoumarin-based com-pound that inhibits DosR binding to its recognition sequencewas recently identified (56). Exposure of M. tuberculosis to thiscompound prevents the upregulation of genes from the DosRregulon, even under inducing conditions (56). Furthermore,exposure to this compound reduces the ability of M. tubercu-losis to establish nonreplicating dormancy in vitro followinggradual oxygen depletion (56). Thus, the development of smallmolecules inhibiting DosR-mediated binding or regulationmay represent novel therapeutic strategies for combating in-fection by M. tuberculosis.

Exposure of M. tuberculosis to carbon monoxide (CO), highNO levels, and possibly other stimuli also induce expression ofdosR-dosS and genes from this regulon (Table 2) (76, 84, 85,140). While the differential regulation of dosR-dosS and genesfrom the DosR regulon is dependent on dosR (15, 138), dosSis dispensable for this regulation (138). This observation led tothe discovery of the second DosR kinase, DosT, which is en-coded outside the DosR regulon (125, 131). dosT appears to beconstitutively expressed (60, 125, 131), and mutational analysishas suggested that this SK may mediate initial adaptationevents to hypoxia (69). Consistent with a role for DosT as asecond DosR kinase, M. tuberculosis �dosS �dosT double mu-tants fail to increase the expression of DosR-regulated deter-minants during hypoxia (125). More recently, dosR-dosS andmembers of its regulon were shown to be upregulated in M.tuberculosis in response to ascorbic acid (vitamin C) in thepresence of oxygen (68, 159). However, this effect is likely dueto the ability of this powerful cytochrome c reductant to inhibitrespiration, scavenge oxygen, and induce hypoxia (159). Fi-nally, it has recently been reported that M. tuberculosis H37Radisplays constitutive DosR-mediated regulation when grownaerobically in medium containing asparagine (99) and thatDosR-DosS-DosT signaling is regulated by the serine/threo-nine protein kinase PknH, which phosphorylates DosR onThr-198 and Thr-205 and enhances DosR-DNA interactions(22). While these observations await further characterization,they solidify the contention that DosR-DosS-DosT receivesand processes regulatory inputs from multiple sources.

Structural and biochemical analyses have provided impor-tant insights into the nature of gas ligand binding by SKs DosSand DosT (28, 29, 71, 72, 78, 85, 86, 118, 133, 149, 165, 179–181). Based on this extensive structural information, it hasbeen proposed that DosT is inactive when bound to oxygenunder normoxic conditions. Oxygen binding occurs via a hememolecule contained within the first of two cyclic GMP (cGMP)



http://mm

br.asm.org/

Dow

nloaded from


phosphodiesterases, adenylate cyclases, and FhlA (GAF) do-mains found near the N terminus of the protein. No ligand todate has been identified for the second GAF domain, althoughthe crystal structure suggests that it does not bind cyclic nu-cleotides, which are prototypical ligands of GAF domains (86).During hypoxia or, alternatively, upon displacement of oxygenby NO or CO, DosT is activated and initiates signaling throughDosR. Similar gas binding can also be attributed to DosS.However, DosS may respond to the redox state of the cell, asDosS is oxidized more rapidly than DosT following oxygenbinding (85, 149). Biochemical analysis supports this conclu-sion, as Fe3�-DosS does not autophosphorylate, while Fe2�-DosS does (85). Further evidence supporting a role for DosSas a redox sensor comes from studies demonstrating that flavinmononucleotides (FMNs) and flavin adenine dinucleotides(FADs) reduce the first GAF domain of DosS (29). In addi-tion, DosS-mediated activation of DosR is observed followingthe disruption of the electron transport chain, resulting inincreased concentrations of reduced components of this path-way (including FMNs, FADs, and quinones) (68).

Numerous studies have implicated DosR-DosS-DosT in as-pects of M. tuberculosis pathogenesis (for further discussion ofthe DosR-dependent response, see reference 128). dosRand/or members of the DosR regulon are upregulated in M.tuberculosis following infection of human monocytes (34), hu-man monocyte-derived macrophages (60), and murine bonemarrow-derived macrophages (135). Genes or proteins fromthis regulon are also expressed or produced in M. tuberculosiswithin an artificial hollow-fiber granuloma model (74). Finally,dosR-dosS and members of the DosR regulon are expressed inM. tuberculosis during late stages of infection in mice (139) andin the lungs of guinea pigs following infection with the tuberclebacillus (137). Consistent with the in vivo expression of thesedeterminants, a subset of DosR regulon determinants are im-munodominant antigens recognized during infection. Antibodyproduction and Th1 T-cell responses are observed in micevaccinated with plasmid DNA encoding select dormancy genes(1, 11, 126). Antibodies to the M. tuberculosis 14-kDa alpha-crystallin-like antigen HspX (Acr; Rv2031c) are also observedin the cerebrospinal fluid of humans with TB meningitis (21)and in the sera of BCG-vaccinated individuals (73). Further-more, purified proteins from several DosR regulon membersinduce IFN-� production in T-cell lines (90) and in peripheralblood monocytes from M. tuberculosis-infected individuals,with a greater response observed in latently infected individu-als (90). Immunological responses to DosR regulon antigensare also observed in TB-positive individuals from geographi-cally diverse regions of Africa (14), in individuals with priorexposure to M. tuberculosis or nontuberculosis mycobacteria(94, 95), and in individuals with remote latent tuberculosisinfection or who have been cured of the disease (50). In con-trast, peripheral blood mononuclear cells from adolescentsvaccinated with M. bovis BCG exhibit low responsiveness whenstimulated with proteins from the dormancy regulon (94). Thismay be due to observed differences in dormancy regulon ex-pression between the M. tuberculosis and M. bovis strains (70).Recent evidence also indicates that strains of the highly viru-lent M. tuberculosis W-Beijing lineage express dosR at elevatedlevels (3, 40, 122). Additionally, a large number of isolatessequenced from this lineage contain a frameshift mutation in

dosT that results in the introduction of a premature stop codon(40). Some of these strains also carry a large gene duplicationin the region of DNA that includes dosR-dosS (35). The impactof these genomic events on the virulence of these strains is notyet understood.

Despite evidence demonstrating that dosR-dosS-dosT andproteins from this regulon are produced during infection andare immunogenic, the role for this TCSS in the virulence of M.tuberculosis remains unclear. M. tuberculosis derivatives con-taining mutations in dosR, dosS, and/or dosT either are atten-uated (31, 98), are hypervirulent (9, 114), or exhibit no phe-notype (31, 127) when examined in in vitro and/or in vivo modelsystems of infection (Table 4). These observed virulence dif-ferences may reflect the use of different M. tuberculosis strainbackgrounds, mutation types, and/or model systems of infec-tion. There is also debate regarding the importance of theDosR regulon to bacterial dormancy (23). While the expres-sion of genes comprising this regulon is rapid following acti-vation of DosR-DosS-DosT by hypoxia (127, 138), the expres-sion of these genes wanes over time (127). Furthermore,growth attenuation of M. tuberculosis �dosR under hypoxicconditions occurs only after DosR-regulated gene expressionhas waned (15, 88, 127). An enduring hypoxic response inwhich numerous DosR-independent genes are upregulated fol-lowing prolonged exposure to hypoxia has been observed(127). In this response, expression of DosR and the dormancyregulon is predicted to “prime” the M. tuberculosis response tohypoxia and allow M. tuberculosis to consume oxygen at anappropriate rate, allow maintenance of appropriate NAD/NADH ratios (and thus the redox environment), and allow forimproved recovery following the reactivation of M. tuberculosisfrom anaerobic to aerobic growth (88). Furthermore, NO in-duction of the dormancy regulon does not sensitize M. tuber-culosis to a second NO response (165), suggesting that DosR-DosS-DosT may be important within the dynamic environmentof the granuloma.

Since its first description, the DosR-DosS-DosT system hasgenerated a lot of interest. Despite this, further work is stillneeded to better elucidate the contribution of this TCSS to M.tuberculosis physiology and pathogenesis and to define the bi-ological contribution of dosR regulon members in the variousphenotypes attributed to this system.

mtrA (Rv3246c)-mtrB (Rv3245c)

Genes comprising the mtrA-mtrB TCSS are present in allMycobacterium species characterized thus far (Table 1) (183)and regulate functions essential for the growth of M. tubercu-losis on laboratory medium (164, 182). This system was firstidentified in 1994 (32). At that time, RR Rv3246c and SKRv3245c were designated MtrA and MtrB (Mycobacterium tu-berculosis response regulator), respectively, based on the ob-served amino acid similarity of this system to other TCSSs(164). Biochemical studies have since confirmed that MtrBautophosphorylates and participates in phosphotransfer reac-tions with MtrA (92). In particular, MtrA is phosphorylated atAsp-56 (Table 3) (44). The site of MtrB autophosphorylationremains undescribed but is predicted to occur at His-305 basedon sequence alignment with other SKs (Table 2) (164). Like



http://mm

br.asm.org/

Dow

nloaded from


PrrA, MtrA is classified as a poorly phosphorylatable RRbased on protein structure (6, 44).

Gene expression studies indicate that mtrA-mtrB is ex-pressed in vivo and under physiologically relevant conditions invitro. An M. tuberculosis mtrAB promoter fusion is expressed atlow levels in M. bovis BCG during growth on laboratory me-dium (32, 164) and is induced in this strain following infectionof murine J774A.1 macrophage-like cells (164, 182). Interest-ingly, mtrA-mtrB is constitutively expressed in M. tuberculosis invitro, within J774A.1 macrophage-like cells, and in human pe-ripheral blood monocyte-derived macrophages (60, 182). Thus,expression of this locus may be differentially regulated betweenM. tuberculosis and BCG. Consistent with MtrA’s intracellularexpression, sera obtained from preclinical tuberculosis pa-tients, but not purified protein derivative-positive (PPD�) orhealthy controls, contain antibodies to MtrA (146). This sug-gests that mtrA is expressed and recognized by the immunesystem during acute infection when the bacteria are growingbut not in PPD� patients in whom the infection is latent andthere is little to no growth of the tubercle bacillus (146).

Due to the inability to generate mtrA and/or mtrB mutants ofM. tuberculosis, investigations into the function of the mtrA-mtrB TCSS have relied on the use of alternative model organ-isms containing homologs of MtrA-MtrB that are dispensablefor growth. Additionally, MtrA-MtrB function has been as-sessed directly in M. tuberculosis using genetic strategies thatresult in overexpression or repression of mtrA-mtrB. mtrA-mtrB mutants of Corynebacterium glutamicum, a related acti-nomycete, exhibit a number of pleiotropic phenotypes, includ-ing an altered transition into stationary phase, abnormalcellular division, and increased sensitivity to certain cell wall-targeting antibiotics or enzymes, including penicillin, vancomy-cin, and lysozyme (106). Similar phenotypes are also observedwith an mtrB::Tn mutant of M. avium (19). This mutant is alsoavirulent in human THP-1 macrophage-like cells (19). In con-trast, while overexpression of mtrA in wild-type M. tuberculosisdoes not result in observable growth phenotypes in vitro, thisderivative fails to prevent phagolysosomal fusion in THP-1cells. This strain also exhibits intracellular attenuation in thiscell type as well as other cell types, including murine J774A.1macrophages, human peripheral blood monocytes, and humanperipheral blood monocyte-derived macrophages (Table 4)(43, 121). These observed phenotypes are likely the conse-quences of elevated levels of phosphorylated MtrA, as M.tuberculosis strains overexpressing an allele of mtrA that cannotbe phosphorylated (MtrAD53N), or expressing mtrA and mtrBtogether, exhibit normal intracellular trafficking and growthcharacteristics (43, 121). Regardless, M. tuberculosis strainsoverproducing either wild-type MtrA or MtrAD53N are atten-uated in the lungs and spleens of C57BL/6 mice followinginfection even though these mutants maintain the ability topersist, albeit at lower levels (Table 4) (43). Thus, MtrA-MtrBalso participates in aspects of M. tuberculosis virulence in vivo.

MtrA recognizes a 20-bp sequence that contains two 9-bpdirect-repeat elements that are upstream of as many as 150genes in M. tuberculosis, including dnaA, an essential replica-tion factor, and fbpABC, encoding components of the antigen85B, which is critical for cord formation and cell wall biogen-esis (Table 3) (93, 121). MtrA is also autoregulatory, binding toits own promoter region (Table 3) (121). Additionally, MtrA

binds to the promoter region of iniB, an isoniazid-induciblegene that is a predicted transporter and may be involved inefflux (93). Perhaps most intriguingly, MtrA binds to the My-cobacterium origin of replication, oriC, suggesting that thisTCSS may directly influence replication.

While genes regulated by MtrA-MtrB continue to be eluci-dated, the signal detected and processed by this TCSS remainsunknown (Table 2). It appears that interaction between MtrBand the membrane protein LpqB, a determinant encoded im-mediately downstream of mtrA-mtrB, may be important (109).Consistent with this observation, an M. smegmatis lpqB trans-poson mutant exhibits phenotypes similar to those seen withmtrA-mtrB mutants of other organisms, including an alteredcolony morphology, increased cell wall permeability, defectiveseptation, and increased SDS sensitivity (109). Specifically, in-teraction between MtrB and LpqB may be necessary to acti-vate signaling through the MtrA-MtrB system, as decreasedlevels of phosphorylated MtrA are observed in the lpqB trans-poson mutant even though total protein levels are similar(109). Additionally, lpqB, like mtrA, is thought to be essentialin M. tuberculosis (134).

MtrA-MtrB signaling may represent a novel therapeutic target,given that this system regulates essential processes, including as-pects of DNA replication and cell wall integrity in the bacterium.However, there is still much to be delineated regarding this TCSS,including the environmental signals that MtrA-MtrB senses, themanner in which genes are regulated by MtrA, and the genes thatcomprise the MtrA-MtrB regulon. Importantly, additional studiesmay also help clarify the role that this TCSS plays in M. tubercu-losis physiology and pathogenesis.

tcrX (Rv3765c)-tcrY (Rv3764c)

Bioinformatic and biochemical studies have recently dem-onstrated that the tcrX and tcrY determinants encode a func-tional TCSS in M. tuberculosis (13). tcrY encodes an SK thatcontains an N-terminal sensing domain, a single transmem-brane domain, and a cytoplasmic C-terminal affector domain.TcrY dimerizes, autophosphorylates in the presence of Mg2�

or Ca2�, and transfers phosphate to the corresponding RRTcrX (13). While the sites of phosphorylation in TrcY andTcrX have not yet been determined biochemically, sequencealignments and molecular modeling predictions point to residuesHis-256 in TcrY and Asp-59 in TcrX as likely sites of modification(Tables 2 and 3) (13). Little is known about the conditions orenvironments which influence the expression and/or activation oftcrX-tcrY. tcrX is transiently expressed in human macrophages(60), and an M. tuberculosis tcrX deletion mutant exhibits in-creased virulence in SCID mice (Table 4) (114). Clearly, morework is necessary to define the mechanism underlying these phe-notypes, the regulon controlled by TcrX, and the environmentalsignals processed by TcrY.

pdtaR (Rv1626)-pdtaS (Rv3220c) and OtherOrphaned RRs and SKs

The original sequencing of M. tuberculosis H37Rv revealedthat, in addition to the 11 genetically linked TCSSs, there areat least five orphaned RRs (Rv0260c, Rv0818, Rv1626, Rv2884,and Rv3143) and two orphaned SKs (Rv2027c and Rv3220c)



http://mm

br.asm.org/

Dow

nloaded from


(30). Subsequent sequence analysis has also revealed the pres-ence of an open reading frame (Rv2998A) encoding a putativesensor portion of an SK (18). Compared to what is knownabout the genetically linked TCSSs, very little is known aboutthese orphaned RRs and SKs. However, it is likely that theseTCSS components influence the physiology and virulence ofM. tuberculosis.

Rv2027c was originally described as an orphaned SK; how-ever, it is now appreciated that Rv2027c, i.e., DosT, is a secondSK of the DosR-DosS TCSS. More recently, a novel TCSScomprised of the orphaned RR Rv1626 and the orphaned SKRv3220c has been elucidated (107, 108). Rv3220c is a solubleSK that autophosphorylates in the presence of Mg2� and cantransfer phosphate to Rv1626 but not to other M. tuberculosisRRs, including PrrA, NarL, and RegX3 (108). While Rv3220ccontains a single GAF domain and a single period clock pro-tein, aryl hydrocarbon receptor, and single-minded protein(PAS) domain, the environmental signals activating this SKhave yet to be defined. Similarly, Rv1626 has an unusual do-main structure. While the N-terminal receiver domain of thisprotein shares high homology with prototypical RRs, includingthe putative phosphorylation site (107), the C-terminal domainis unlike any other RR of M. tuberculosis. It contains an AmiRand NasR transcription antitermination regulator (ANTAR)domain, which is found only in prokaryotic proteins (144). Thisdomain has been shown to be important for RNA binding(175). Given this domain architecture, it has been speculatedthat Rv1626 may bind to RNA and regulate gene expression byacting as an antiterminator to prevent stem-loop structure forma-tion in the leader region of newly transcribed RNA (107). Con-sequently, Rv1626 was renamed the phosphorylation-dependenttranscriptional antitermination regulator (PdtaR), and Rv3220cwas renamed PdtaS (108). Whether Rv1626 indeed acts as anantiterminator remains to be defined. Interestingly, pdtaR is anessential gene in M. tuberculosis (134), but deletion of pdtaS canbe tolerated (114). Additionally, pdtaR and pdtaS appear to beconserved in all Mycobacterium species (Table 1) (108). Whilenothing is currently known about the genes/operons that are con-trolled by PtdaR-PdtaS, a pdtaS mutant of M. tuberculosis exhibitsno phenotype following infection of SCID mice (114), indicatingthat this system may not participate in aspects of virulence. Re-gardless, further investigations into this novel TCSS and the re-maining orphaned RRs are needed.

CONCLUDING REMARKS

The circulation of multidrug-resistant and extensively drug-resistant strains of M. tuberculosis that are recalcitrant to cur-rent antibiotics and the increase in comorbidity of patientsinfected with M. tuberculosis and HIV continue to threaten ourability to control tuberculosis. The development of new anti-mycobacterial agents and vaccines with improved efficacyagainst M. tuberculosis is an important goal for the effectivecontrol of this disease. The TCSSs of M. tuberculosis have beenproposed as targets for antimycobacterial drug design, andattenuated mutants have been suggested as possible vaccinecandidates. The recent expansion of M. tuberculosis TCSS re-search and the utilization of powerful techniques, such asglobal transcriptional profiling using DNA microarrays, havegenerally validated beliefs that the expression of TCSSs plays

an important role in facilitating the successful adaptation of M.tuberculosis to diverse environmental conditions encounteredwithin the host. Such stress signals include nutrient starvation,hypoxia, fluctuating Pi and K� concentrations, altered pH, NOand CO exposure, and cell wall/membrane stress. Undoubt-edly, more signals exist and await identification. Studies of M.tuberculosis SK and RR mutants in tissue culture and animalmodel systems of infection indicate that several TCSSs playsubstantive roles in virulence. Genetic and biochemical char-acterizations of these genes and the downstream determinantscomprising their regulons will continue to aid in our under-standing of the roles that individual TCSSs play in M. tubercu-losis physiology and virulence. Such studies are also expectedto provide new insights into the environmental signals recog-nized by these systems and the molecular mechanisms main-taining them in an inactive state in the absence of stimulation.Importantly, continued investigation of the TCSSs of M. tuber-culosis is likely to lead to novel therapeutics or vaccine candi-dates which may aid in the successful control of this significantglobal pathogen.

ACKNOWLEDGMENTS

We are grateful to members of the Zahrt laboratory and to ShelleyHaydel for useful discussions and critical reviews of this paper prior toits submission. We apologize to those authors whose studies in thisarea we were not able to include.

This work is sponsored by the Center for Infectious Disease Re-search at the Medical College of Wisconsin, an Investigator in Patho-genesis Award from the Burroughs Wellcome Fund, and the NIHNational Institute of Allergy and Infectious Diseases (grant RO1-AI-51669).

REFERENCES

1. Aguilar, D., et al. 2007. Immunological responses and protective immunityagainst tuberculosis conferred by vaccination of Balb/C mice with the at-tenuated Mycobacterium tuberculosis (phoP) SO2 strain. Clin. Exp. Immu-nol. 147:330–338.

2. Reference deleted.3. Badillo-Lopez, C., et al. 2010. Differential expression of dnaA and dosR

genes among members of the Mycobacterium tuberculosis complex underoxic and hypoxic conditions. Int. Microbiol. 13:9–13.

4. Bagchi, G., S. Chauhan, D. Sharma, and J. S. Tyagi. 2005. Transcriptionand autoregulation of the Rv3134c-devR-devS operon of Mycobacteriumtuberculosis. Microbiology 151:4045–4053.

5. Bagchi, G., Mayuri, and J. S. Tyagi. 2003. Hypoxia-responsive expression ofMycobacterium tuberculosis Rv3134c and devR promoters in Mycobacte-rium smegmatis. Microbiology 149:2303–2305.

6. Barbieri, C. M., T. R. Mack, V. L. Robinson, M. T. Miller, and A. M. Stock.2010. Regulation of response regulator autophosphorylation through inter-domain contacts. J. Biol. Chem. 285:32325–32335.

7. Barrett, J. F., et al. 1998. Antibacterial agents that inhibit two-componentsignal transduction systems. Proc. Natl. Acad. Sci. U. S. A. 95:5317–5322.

8. Barrett, J. F., and J. A. Hoch. 1998. Two-component signal transduction asa target for microbial anti-infective therapy. Antimicrob. Agents Che-mother. 42:1529–1536.

9. Bartek, I. L., et al. 2009. The DosR regulon of M. tuberculosis and anti-bacterial tolerance. Tuberculosis (Edinb.) 89:310–316.

10. Becq, J., et al. 2007. Contribution of horizontally acquired genomic islandsto the evolution of the tubercle bacilli. Mol. Biol. Evol. 24:1861–1871.

11. Ben Amor, Y., et al. 2005. Immunological characterization of novel secretedantigens of Mycobacterium tuberculosis. Scand. J. Immunol. 61:139–146.

12. Betts, J. C., P. T. Lukey, L. C. Robb, R. A. McAdam, and K. Duncan. 2002.Evaluation of a nutrient starvation model of Mycobacterium tuberculosispersistence by gene and protein expression profiling. Mol. Microbiol. 43:717–731.

13. Bhattacharya, M., A. Biswas, and A. K. Das. 2010. Interaction analysis ofTcrX/Y two component system from Mycobacterium tuberculosis.Biochimie 92:263–272.

14. Black, G. F., et al. 2009. Immunogenicity of novel DosR regulon-encodedcandidate antigens of Mycobacterium tuberculosis in three high-burden pop-ulations in Africa. Clin. Vaccine Immunol. 16:1203–1212.

15. Boon, C., and T. Dick. 2002. Mycobacterium bovis BCG response regulatoressential for hypoxic dormancy. J. Bacteriol. 184:6760–6767.



http://mm

br.asm.org/

Dow

nloaded from


16. Boon, C., R. Li, R. Qi, and T. Dick. 2001. Proteins of Mycobacterium bovisBCG induced in the Wayne dormancy model. J. Bacteriol. 183:2672–2676.

17. Brosch, R., et al. 2007. Genome plasticity of BCG and impact on vaccineefficacy. Proc. Natl. Acad. Sci. U. S. A. 104:5596–5601.

18. Camus, J. C., M. J. Pryor, C. Medigue, and S. T. Cole. 2002. Re-annotationof the genome sequence of Mycobacterium tuberculosis H37Rv. Microbi-ology 148:2967–2973.

19. Cangelosi, G. A., et al. 2006. The two-component regulatory system mtrABis required for morphotypic multidrug resistance in Mycobacterium avium.Antimicrob. Agents Chemother. 50:461–468.

20. Cardona, P. J., et al. 2009. Extended safety studies of the attenuated livetuberculosis vaccine SO2 based on phoP mutant. Vaccine 27:2499–2505.

21. Chandramuki, A., G. H. Bothamley, P. J. Brennan, and J. Ivanyi. 1989.Levels of antibody to defined antigens of Mycobacterium tuberculosis intuberculous meningitis. J. Clin. Microbiol. 27:821–825.

22. Chao, J. D., et al. 2010. Convergence of Ser/Thr and two-component sig-naling to coordinate expression of the dormancy regulon in Mycobacteriumtuberculosis. J. Biol. Chem. 285:29239–29246.

23. Chao, M. C., and E. J. Rubin. 2010. Letting sleeping dos lie: does dormancyplay a role in tuberculosis? Annu. Rev. Microbiol. 64:293–311.

24. Chauhan, S., and J. S. Tyagi. 2008. Cooperative binding of phosphorylatedDevR to upstream sites is necessary and sufficient for activation of theRv3134c-devRS operon in Mycobacterium tuberculosis: implication in theinduction of DevR target genes. J. Bacteriol. 190:4301–4312.

25. Chauhan, S., and J. S. Tyagi. 2008. Interaction of DevR with multiplebinding sites synergistically activates divergent transcription of narK2-Rv1738 genes in Mycobacterium tuberculosis. J. Bacteriol. 190:5394–5403.

26. Chauhan, S., and J. S. Tyagi. 2009. Powerful induction of divergent tgs1-Rv3131 genes in Mycobacterium tuberculosis is mediated by DevR interac-tion with a high-affinity site and an adjacent cryptic low-affinity site. J.Bacteriol. 191:6075–6081.

27. Chesne-Seck, M.-L., et al. 2008. A point mutation in the two-componentregulator PhoP-PhoR accounts for the absence of polyketide-derived acyl-trehaloses but not that of phthiocerol dimycocerosates in Mycobacteriumtuberculosis H37Ra. J. Bacteriol. 190:1329–1334.

28. Cho, H. Y., H. J. Cho, Y. M. Kim, J. I. Oh, and B. S. Kang. 2008. Crystal-lization and preliminary crystallographic analysis of the second GAF do-main of DevS from Mycobacterium smegmatis. Acta Crystallogr. Sect. FStruct. Biol. Cryst. Commun. 64:274–276.

29. Cho, H. Y., H. J. Cho, Y. M. Kim, J. I. Oh, and B. S. Kang. 2009. Structuralinsight into the heme-based redox sensing by DosS from Mycobacteriumtuberculosis. J. Biol. Chem. 284:13057–13067.

30. Cole, S. T., et al. 1998. Deciphering the biology of Mycobacterium tuber-culosis from the complete genome sequence. Nature 393:537–544.

31. Converse, P. J., et al. 2009. Role of the dosR-dosS two-component regula-tory system in Mycobacterium tuberculosis virulence in three animal models.Infect. Immun. 77:1230–1237.

32. Curcic, R., S. Dhandayuthapani, and V. Deretic. 1994. Gene expression inmycobacteria: transcriptional fusions based on xylE and analysis of thepromoter region of the response regulator mtrA from Mycobacterium tu-berculosis. Mol. Microbiol. 13:1057–1064.

33. Das, A. K., et al. 2010. A single-amino-acid substitution in the C terminusof PhoP determines DNA-binding specificity of the virulence-associatedresponse regulator from Mycobacterium tuberculosis. J. Mol. Biol. 398:647–656.

34. Dasgupta, N., et al. 2000. Characterization of a two-component system,devR-devS, of Mycobacterium tuberculosis. Tuber. Lung Dis. 80:141–159.

35. Domenech, P., G. S. Kolly, L. Leon-Solis, A. Fallow, and M. B. Reed. 2010.Massive gene duplication event among clinical isolates of the Mycobacte-rium tuberculosis W/Beijing family. J. Bacteriol. 192:4562–4570.

36. Dona, V., et al. 2008. Evidence of complex transcriptional, translational,and posttranslational regulation of the extracytoplasmic function sigmafactor E in Mycobacterium tuberculosis. J. Bacteriol. 190:5963–5971.

37. Egger, L. A., and M. Inouye. 1997. Purification and characterization of theperiplasmic domain of EnvZ osmosensor in Escherichia coli. Biochem.Biophys. Res. Commun. 231:68–72.

38. Ewann, F., et al. 2002. Transient requirement of the PrrA-PrrB two-com-ponent system for early intracellular multiplication of Mycobacterium tu-berculosis. Infect. Immun. 70:2256–2263.

39. Ewann, F., C. Locht, and P. Supply. 2004. Intracellular autoregulation ofthe Mycobacterium tuberculosis PrrA response regulator. Microbiology150:241–246.

40. Fallow, A., P. Domenech, and M. B. Reed. 2010. Strains of the East Asian(W/Beijing) lineage of Mycobacterium tuberculosis are DosS/DosT-DosRtwo-component regulatory system natural mutants. J. Bacteriol. 192:2228–2238.

41. Ferrer, N. L., et al. 2009. Intracellular replication of attenuated Mycobac-terium tuberculosis phoP mutant in the absence of host cell cytotoxicity.Microbes Infect. 11:115–122.

42. Florczyk, M. A., et al. 2003. A family of acr-coregulated Mycobacteriumtuberculosis genes shares a common DNA motif and requires Rv3133c(dosR or devR) for expression. Infect. Immun. 71:5332–5343.

43. Fol, M., et al. 2006. Modulation of Mycobacterium tuberculosis prolifera-tion by MtrA, an essential two-component response regulator. Mol. Micro-biol. 60:643–657.

44. Friedland, N., et al. 2007. Domain orientation in the inactive responseregulator Mycobacterium tuberculosis MtrA provides a barrier to activa-tion. Biochemistry 46:6733–6743.

45. Frigui, W., et al. 2008. Control of M. tuberculosis ESAT-6 secretion andspecific T cell recognition by PhoP. PLoS Pathog. 4:e33.

46. Gao, R., T. R. Mack, and A. M. Stock. 2007. Bacterial response regulators:versatile regulatory strategies from common domains. Trends Biochem. Sci.32:225–234.

47. Gautam, U. S., S. Chauhan, and J. S. Tyagi. 2011. Determinants outside theDevR C-terminal domain are essential for cooperativity and robust activa-tion of dormancy genes in Mycobacterium tuberculosis. PLoS One6:e16500.

48. Gebhard, S., and G. M. Cook. 2008. Differential regulation of high-affinityphosphate transport systems of Mycobacterium smegmatis: identification ofPhnF, a repressor of the phnDCE operon. J. Bacteriol. 190:1335–1343.

49. Glover, R. T., J. Kriakov, S. J. Garforth, A. D. Baughn, and W. R. Jacobs,Jr. 2007. The two-component regulatory system senX3-regX3 regulatesphosphate-dependent gene expression in Mycobacterium smegmatis. J. Bac-teriol. 189:5495–5503.

50. Goletti, D., et al. 2010. Response to Rv2628 latency antigen associates withcured tuberculosis and remote infection. Eur. Respir. J. 36:135–142.

51. Gonzalo-Asensio, J., et al. 2008. PhoP: a missing piece in the intricatepuzzle of Mycobacterium tuberculosis virulence. PLoS One 3:e3496.

52. Gonzalo-Asensio, J., et al. 2008. The Mycobacterium tuberculosis phoPRoperon is positively autoregulated in the virulent strain H37Rv. J. Bacteriol.190:7068–7078.

53. Gonzalo Asensio, J., et al. 2006. The virulence-associated two-componentPhoP-PhoR system controls the biosynthesis of polyketide-derived lipids inMycobacterium tuberculosis. J. Biol. Chem. 281:1313–1316.

54. Graham, J. E., and J. E. Clark-Curtiss. 1999. Identification of Mycobacte-rium tuberculosis RNAs synthesized in response to phagocytosis by humanmacrophages by selective capture of transcribed sequences (SCOTS). Proc.Natl. Acad. Sci. U. S. A. 96:11554–11559.

55. Gupta, R. K., S. Chauhan, and J. S. Tyagi. 2011. K182G substitution inDevR or C8G mutation in the Dev box impairs protein-DNA interactionand abrogates DevR-mediated gene induction in Mycobacterium tubercu-losis. FEBS J. 278:2131–2139.

56. Gupta, R. K., T. S. Thakur, G. R. Desiraju, and J. S. Tyagi. 2009. Structure-based design of DevR inhibitor active against nonreplicating Mycobacte-rium tuberculosis. J. Med. Chem. 52:6324–6334.

57. Gupta, S., A. Pathak, A. Sinha, and D. Sarkar. 2009. Mycobacterium tuber-culosis PhoP recognizes two adjacent direct-repeat sequences to form head-to-head dimers. J. Bacteriol. 191:7466–7476.

58. Gupta, S., A. Sinha, and D. Sarkar. 2006. Transcriptional autoregulation byMycobacterium tuberculosis PhoP involves recognition of novel direct re-peat sequences in the regulatory region of the promoter. FEBS Lett. 580:5328–5338.

59. Haydel, S. E., W. H. Benjamin, Jr., N. E. Dunlap, and J. E. Clark-Curtiss.2002. Expression, autoregulation, and DNA binding properties of the My-cobacterium tuberculosis TrcR response regulator. J. Bacteriol. 184:2192–2203.

60. Haydel, S. E., and J. E. Clark-Curtiss. 2004. Global expression analysis oftwo-component system regulator genes during Mycobacterium tuberculosisgrowth in human macrophages. FEMS Microbiol. Lett. 236:341–347.

61. Haydel, S. E., and J. E. Clark-Curtiss. 2006. The Mycobacterium tubercu-losis TrcR response regulator represses transcription of the intracellularlyexpressed Rv1057 gene, encoding a seven-bladed �-propeller. J. Bacteriol.188:150–159.

62. Haydel, S. E., N. E. Dunlap, and W. H. Benjamin, Jr. 1999. In vitro evidenceof two-component system phosphorylation between the Mycobacteriumtuberculosis TrcR/TrcS proteins. Microb. Pathog. 26:195–206.

63. He, H., R. Hovey, J. Kane, V. Singh, and T. C. Zahrt. 2006. MprAB is astress-responsive two-component system that directly regulates expressionof sigma factors SigB and SigE in Mycobacterium tuberculosis. J. Bacteriol.188:2134–2143.

64. He, H., and T. C. Zahrt. 2005. Identification and characterization of aregulatory sequence recognized by Mycobacterium tuberculosis persistenceregulator MprA. J. Bacteriol. 187:202–212.

65. Hess, J. F., R. B. Bourret, and M. I. Simon. 1988. Histidine phosphorylationand phosphoryl group transfer in bacterial chemotaxis. Nature 336:139–143.

66. Himpens, S., C. Locht, and P. Supply. 2000. Molecular characterization ofthe mycobacterial SenX3-RegX3 two-component system: evidence for au-toregulation. Microbiology 146(Part 12):3091–3098.

67. Hofmann, K., and W. Stoffel. 1993. TMbase—a database of membranespanning protein segments. Biol. Chem. Hoppe-Seyler 374:166.

68. Honaker, R. W., R. K. Dhiman, P. Narayanasamy, D. C. Crick, and M. I.Voskuil. 2010. DosS responds to a reduced electron transport system toinduce the Mycobacterium tuberculosis DosR regulon. J. Bacteriol. 192:6447–6455.



http://mm

br.asm.org/

Dow

nloaded from


69. Honaker, R. W., R. L. Leistikow, I. L. Bartek, and M. I. Voskuil. 2009.Unique roles of DosT and DosS in DosR regulon induction and Mycobac-terium tuberculosis dormancy. Infect. Immun. 77:3258–3263.

70. Honaker, R. W., et al. 2008. Mycobacterium bovis BCG vaccine strains lacknarK2 and narX induction and exhibit altered phenotypes during dormancy.Infect. Immun. 76:2587–2593.

71. Ioanoviciu, A., Y. T. Meharenna, T. L. Poulos, and P. R. Ortiz de Mon-tellano. 2009. DevS oxy complex stability identifies this heme protein as agas sensor in Mycobacterium tuberculosis dormancy. Biochemistry 48:5839–5848.

72. Ioanoviciu, A., E. T. Yukl, P. Moenne-Loccoz, and P. R. de Montellano.2007. DevS, a heme-containing two-component oxygen sensor of Mycobac-terium tuberculosis. Biochemistry 46:4250–4260.

73. Jackett, P. S., et al. 1988. Specificity of antibodies to immunodominantmycobacterial antigens in pulmonary tuberculosis. J. Clin. Microbiol. 26:2313–2318.

74. Karakousis, P. C., et al. 2004. Dormancy phenotype displayed by extracel-lular Mycobacterium tuberculosis within artificial granulomas in mice. J.Exp. Med. 200:647–657.

75. Kato, A., and E. A. Groisman. 2008. The PhoQ/PhoP regulatory network ofSalmonella enterica. Adv. Exp. Med. Biol. 631:7–21.

76. Kendall, S. L., et al. 2004. The Mycobacterium tuberculosis dosRS two-component system is induced by multiple stresses. Tuberculosis (Edinb.)84:247–255.

77. Kim, D., and S. Forst. 2001. Genomic analysis of the histidine kinase familyin bacteria and archaea. Microbiology 147:1197–1212.

78. Kim, M.-J., K.-J. Park, I.-J. Ko, Y. M. Kim, and J.-I. Oh. 2010. Differentroles of DosS and DosT in the hypoxic adaptation of mycobacteria. J.Bacteriol. 192:4868–4875.

79. Kinger, A. K., and J. S. Tyagi. 1993. Identification and cloning of genesdifferentially expressed in the virulent strain of Mycobacterium tuberculo-sis. Gene 131:113–117.

80. King-Scott, J., et al. 2007. The structure of a full-length response regulatorfrom Mycobacterium tuberculosis in a stabilized three-dimensional do-main-swapped, activated state. J. Biol. Chem. 282:37717–37729.

81. Kirby, J. R. 2009. Chemotaxis-like regulatory systems: unique roles indiverse bacteria. Annu. Rev. Microbiol. 63:45–59.

82. Krell, T., et al. 2010. Bacterial sensor kinases: diversity in the recognition ofenvironmental signals. Annu. Rev. Microbiol. 64:539–559.

83. Krogh, A., B. Larsson, G. von Heijne, and E. L. Sonnhammer. 2001. Pre-dicting transmembrane protein topology with a hidden Markov model:application to complete genomes. J. Mol. Biol. 305:567–580.

84. Kumar, A., et al. 2008. Heme oxygenase-1-derived carbon monoxide in-duces the Mycobacterium tuberculosis dormancy regulon. J. Biol. Chem.283:18032–18039.

85. Kumar, A., J. C. Toledo, R. P. Patel, J. R. Lancaster, Jr., and A. J. Steyn.2007. Mycobacterium tuberculosis DosS is a redox sensor and DosT is ahypoxia sensor. Proc. Natl. Acad. Sci. U. S. A. 104:11568–11573.

86. Lee, J. M., et al. 2008. O2- and NO-sensing mechanism through the DevSRtwo-component system in Mycobacterium smegmatis. J. Bacteriol. 190:6795–6804.

87. Lee, J. S., et al. 2008. Mutation in the transcriptional regulator PhoPcontributes to avirulence of Mycobacterium tuberculosis H37Ra strain. CellHost Microbe 3:97–103.

88. Leistikow, R. L., et al. 2010. The Mycobacterium tuberculosis DosR regulonassists in metabolic homeostasis and enables rapid recovery from nonre-spiring dormancy. J. Bacteriol. 192:1662–1670.

89. Leung, A. S., et al. 2008. Novel genome polymorphisms in BCG vaccinestrains and impact on efficacy. BMC Genomics 9:413.

90. Leyten, E. M., et al. 2006. Human T-cell responses to 25 novel antigensencoded by genes of the dormancy regulon of Mycobacterium tuberculosis.Microbes Infect. 8:2052–2060.

91. Li, A. H., et al. 2010. Contrasting transcriptional responses of a virulent andan attenuated strain of Mycobacterium tuberculosis infecting macrophages.PLoS One 5:e11066.

92. Li, Y., J. Zeng, and Z. G. He. 2010. Characterization of a functional C-ter-minus of the Mycobacterium tuberculosis MtrA responsible for both DNAbinding and interaction with its two-component partner protein, MtrB.J. Biochem. 148:549–556.

93. Li, Y., J. Zeng, H. Zhang, and Z. G. He. 2010. The characterization ofconserved binding motifs and potential target genes for M. tuberculosisMtrAB reveals a link between the two-component system and the drugresistance of M. smegmatis. BMC Microbiol. 10:242.

94. Lin, M. Y., et al. 2007. Lack of immune responses to Mycobacterium tuber-culosis DosR regulon proteins following Mycobacterium bovis BCG vacci-nation. Infect. Immun. 75:3523–3530.

95. Lin, M. Y., et al. 2009. Cross-reactive immunity to Mycobacterium tubercu-losis DosR regulon-encoded antigens in individuals infected with environ-mental, nontuberculous mycobacteria. Infect. Immun. 77:5071–5079.

96. Magdalena, J., P. Supply, and C. Locht. 1998. Specific differentiation be-tween Mycobacterium bovis BCG and virulent strains of the Mycobacteriumtuberculosis complex. J. Clin. Microbiol. 36:2471–2476.

97. Magdalena, J., A. Vachee, P. Supply, and C. Locht. 1998. Identification ofa new DNA region specific for members of Mycobacterium tuberculosiscomplex. J. Clin. Microbiol. 36:937–943.

98. Malhotra, V., et al. 2004. Disruption of response regulator gene, devR,leads to attenuation in virulence of Mycobacterium tuberculosis. FEMSMicrobiol. Lett. 231:237–245.

99. Malhotra, V., J. S. Tyagi, and J. E. Clark-Curtiss. 2009. DevR-mediatedadaptive response in Mycobacterium tuberculosis H37Ra: links to aspara-gine metabolism. Tuberculosis (Edinb.) 89:169–174.

100. Manabe, Y. C., B. J. Saviola, L. Sun, J. R. Murphy, and W. R. Bishai. 1999.Attenuation of virulence in Mycobacterium tuberculosis expressing a con-stitutively active iron repressor. Proc. Natl. Acad. Sci. U. S. A. 96:12844–12848.

101. Manganelli, R., M. I. Voskuil, G. K. Schoolnik, and I. Smith. 2001. TheMycobacterium tuberculosis ECF sigma factor E: role in global geneexpression and survival in macrophages. Mol. Microbiol. 41:423–437.

102. Martin, C., et al. 2006. The live Mycobacterium tuberculosis phoP mutantstrain is more attenuated than BCG and confers protective immunityagainst tuberculosis in mice and guinea pigs. Vaccine 24:3408–3419.

103. Mayuri, G. Bagchi, T. K. Das, and J. S. Tyagi. 2002. Molecular analysis ofthe dormancy response in Mycobacterium smegmatis: expression analysis ofgenes encoding the DevR-DevS two-component system, Rv3134c and chap-erone alpha-crystallin homologues. FEMS Microbiol. Lett. 211:231–237.

104. McLaughlin, B., et al. 2007. A mycobacterium ESX-1-secreted virulencefactor with unique requirements for export. PLoS Pathog. 3:e105.

105. Mitrophanov, A. Y., and E. A. Groisman. 2008. Signal integration in bac-terial two-component regulatory systems. Genes Dev. 22:2601–2611.

106. Moker, N., et al. 2004. Deletion of the genes encoding the MtrA-MtrBtwo-component system of Corynebacterium glutamicum has a strong influ-ence on cell morphology, antibiotics susceptibility and expression of genesinvolved in osmoprotection. Mol. Microbiol. 54:420–438.

107. Morth, J. P., V. Feng, L. J. Perry, D. I. Svergun, and P. A. Tucker. 2004. Thecrystal and solution structure of a putative transcriptional antiterminatorfrom Mycobacterium tuberculosis. Structure 12:1595–1605.

108. Morth, J. P., S. Gosmann, E. Nowak, and P. A. Tucker. 2005. A noveltwo-component system found in Mycobacterium tuberculosis. FEBS Lett.579:4145–4148.

109. Nguyen, H. T., K. A. Wolff, R. H. Cartabuke, S. Ogwang, and L. Nguyen.2010. A lipoprotein modulates activity of the MtrAB two-component sys-tem to provide intrinsic multidrug resistance, cytokinetic control and cellwall homeostasis in Mycobacterium. Mol. Microbiol. 76:348–364.

110. Nowak, E., S. Panjikar, P. Konarev, D. I. Svergun, and P. A. Tucker. 2006.The structural basis of signal transduction for the response regulator PrrAfrom Mycobacterium tuberculosis. J. Biol. Chem. 281:9659–9666.

111. Nowak, E., et al. 2006. Structural and functional aspects of the sensor histidinekinase PrrB from Mycobacterium tuberculosis. Structure 14:275–285.

112. Pang, X., and S. T. Howard. 2007. Regulation of the �-crystallin gene acr2by the MprAB two-component system of Mycobacterium tuberculosis. J.Bacteriol. 189:6213–6221.

113. Pang, X., et al. 2007. Evidence for complex interactions of stress-associatedregulons in an mprAB deletion mutant of Mycobacterium tuberculosis.Microbiology 153:1229–1242.

114. Parish, T., et al. 2003. Deletion of two-component regulatory systemsincreases the virulence of Mycobacterium tuberculosis. Infect. Immun. 71:1134–1140.

115. Parish, T., D. A. Smith, G. Roberts, J. Betts, and N. G. Stoker. 2003. ThesenX3-regX3 two-component regulatory system of Mycobacterium tuber-culosis is required for virulence. Microbiology 149:1423–1435.

116. Park, H. D., et al. 2003. Rv3133c/dosR is a transcription factor that medi-ates the hypoxic response of Mycobacterium tuberculosis. Mol. Microbiol.48:833–843.

117. Perez, E., et al. 2001. An essential role for phoP in Mycobacterium tuber-culosis virulence. Mol. Microbiol. 41:179–187.

118. Podust, L. M., A. Ioanoviciu, and P. R. Ortiz de Montellano. 2008. 2.3 ÅX-ray structure of the heme-bound GAF domain of sensory histidine kinaseDosT of Mycobacterium tuberculosis. Biochemistry 47:12523–12531.

119. Queipo-Ortuno, M. I., J. D. Colmenero, P. Bermudez, M. J. Bravo, and P.Morata. 2009. Rapid differential diagnosis between extrapulmonary tuber-culosis and focal complications of brucellosis using a multiplex real-timePCR assay. PLoS One 4:e4526.

120. Raghavan, S., P. Manzanillo, K. Chan, C. Dovey, and J. S. Cox. 2008.Secreted transcription factor controls Mycobacterium tuberculosis viru-lence. Nature 454:717–721.

121. Rajagopalan, M., et al. 2010. Mycobacterium tuberculosis origin of repli-cation and the promoter for immunodominant secreted antigen 85B are thetargets of MtrA, the essential response regulator. J. Biol. Chem. 285:15816–15827.

122. Reed, M. B., S. Gagneux, K. DeRiemer, P. M. Small, and C. E. Barry III.2007. The W-Beijing lineage of Mycobacterium tuberculosis overproducestriglycerides and has the DosR dormancy regulon constitutively upregu-lated. J. Bacteriol. 189:2583–2589.

123. Rickman, L., et al. 2004. A two-component signal transduction system with



http://mm

br.asm.org/

Dow

nloaded from


a PAS domain-containing sensor is required for virulence of Mycobacte-rium tuberculosis in mice. Biochem. Biophys. Res. Commun. 314:259–267.

124. Rifat, D., W. R. Bishai, and P. C. Karakousis. 2009. Phosphate depletion:a novel trigger for Mycobacterium tuberculosis persistence. J. Infect. Dis.200:1126–1135.

125. Roberts, D. M., R. P. Liao, G. Wisedchaisri, W. G. Hol, and D. R. Sherman.2004. Two sensor kinases contribute to the hypoxic response of Mycobac-terium tuberculosis. J. Biol. Chem. 279:23082–23087.

126. Roupie, V., et al. 2007. Immunogenicity of eight dormancy regulon-encodedproteins of Mycobacterium tuberculosis in DNA-vaccinated and tuberculo-sis-infected mice. Infect. Immun. 75:941–949.

127. Rustad, T. R., M. I. Harrell, R. Liao, and D. R. Sherman. 2008. Theenduring hypoxic response of Mycobacterium tuberculosis. PLoS One3:e1502.

128. Rustad, T. R., A. M. Sherrid, K. J. Minch, and D. R. Sherman. 2009.Hypoxia: a window into Mycobacterium tuberculosis latency. Cell. Micro-biol. 11:1151–1159.

129. Ryndak, M., S. Wang, and I. Smith. 2008. PhoP, a key player in Mycobac-terium tuberculosis virulence. Trends Microbiol. 16:528–534.

130. Saini, D. K., et al. 2004. DevR-DevS is a bona fide two-component systemof Mycobacterium tuberculosis that is hypoxia-responsive in the absence ofthe DNA-binding domain of DevR. Microbiology 150:865–875.

131. Saini, D. K., V. Malhotra, and J. S. Tyagi. 2004. Cross talk between DevSsensor kinase homologue, Rv2027c, and DevR response regulator of My-cobacterium tuberculosis. FEBS Lett. 565:75–80.

132. Saini, D. K., N. Pant, T. K. Das, and J. S. Tyagi. 2002. Cloning, overex-pression, purification, and matrix-assisted refolding of DevS (Rv 3132c)histidine protein kinase of Mycobacterium tuberculosis. Protein Expr. Pu-rif. 25:203–208.

133. Sardiwal, S., et al. 2005. A GAF domain in the hypoxia/NO-inducibleMycobacterium tuberculosis DosS protein binds haem. J. Mol. Biol. 353:929–936.

134. Sassetti, C. M., D. H. Boyd, and E. J. Rubin. 2003. Genes required formycobacterial growth defined by high density mutagenesis. Mol. Microbiol.48:77–84.

135. Schnappinger, D., et al. 2003. Transcriptional adaptation of Mycobacte-rium tuberculosis within macrophages: insights into the phagosomal envi-ronment. J. Exp. Med. 198:693–704.

136. Schnell, R., D. Agren, and G. Schneider. 2008. 1.9 Å structure of the signalreceiver domain of the putative response regulator NarL from Mycobacte-rium tuberculosis. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun.64:1096–1100.

137. Sharma, D., et al. 2006. Expression of mycobacterial cell division protein, FtsZ,and dormancy proteins, DevR and Acr, within lung granulomas throughoutguinea pig infection. FEMS Immunol. Med. Microbiol. 48:329–336.

138. Sherman, D. R., et al. 2001. Regulation of the Mycobacterium tuberculosishypoxic response gene encoding �-crystallin. Proc. Natl. Acad. Sci. U. S. A.98:7534–7539.

139. Shi, L., Y. J. Jung, S. Tyagi, M. L. Gennaro, and R. J. North. 2003.Expression of Th1-mediated immunity in mouse lungs induces a Mycobac-terium tuberculosis transcription pattern characteristic of nonreplicatingpersistence. Proc. Natl. Acad. Sci. U. S. A. 100:241–246.

140. Shiloh, M. U., P. Manzanillo, and J. S. Cox. 2008. Mycobacterium tuber-culosis senses host-derived carbon monoxide during macrophage infection.Cell Host Microbe 3:323–330.

141. Shrivastava, R., D. R. Das, H. G. Wiker, and A. K. Das. 2006. Functionalinsights from the molecular modelling of a novel two-component system.Biochem. Biophys. Res. Commun. 344:1327–1333.

142. Shrivastava, R., A. K. Ghosh, and A. K. Das. 2009. Intra- and intermolec-ular domain interactions among novel two-component system proteinscoded by Rv0600c, Rv0601c and Rv0602c of Mycobacterium tuberculosis.Microbiology 155:772–779.

143. Shrivastava, R., A. K. Ghosh, and A. K. Das. 2007. Probing the nucleotidebinding and phosphorylation by the histidine kinase of a novel three-proteintwo-component system from Mycobacterium tuberculosis. FEBS Lett. 581:1903–1909.

144. Shu, C. J., and I. B. Zhulin. 2002. ANTAR: an RNA-binding domain intranscription antitermination regulatory proteins. Trends Biochem. Sci. 27:3–5.

145. Singh, A., D. Mai, A. Kumar, and A. J. Steyn. 2006. Dissecting virulencepathways of Mycobacterium tuberculosis through protein-protein associa-tion. Proc. Natl. Acad. Sci. U. S. A. 103:11346–11351.

146. Singh, K. K., X. Zhang, A. S. Patibandla, P. Chien, Jr., and S. Laal. 2001.Antigens of Mycobacterium tuberculosis expressed during preclinical tuber-culosis: serological immunodominance of proteins with repetitive aminoacid sequences. Infect. Immun. 69:4185–4191.

147. Sinha, A., S. Gupta, S. Bhutani, A. Pathak, and D. Sarkar. 2008. PhoP-PhoP interaction at adjacent PhoP binding sites is influenced by proteinphosphorylation. J. Bacteriol. 190:1317–1328.

148. Soto, C. Y., et al. 2004. IS6110 mediates increased transcription of the phoPvirulence gene in a multidrug-resistant clinical isolate responsible for tu-berculosis outbreaks. J. Clin. Microbiol. 42:212–219.

149. Sousa, E. H., J. R. Tuckerman, G. Gonzalez, and M. A. Gilles-Gonzalez.

2007. DosT and DevS are oxygen-switched kinases in Mycobacterium tu-berculosis. Protein Sci. 16:1708–1719.

150. Stewart, V. 1993. Nitrate regulation of anaerobic respiratory gene expres-sion in Escherichia coli. Mol. Microbiol. 9:425–434.

151. Steyn, A. J., J. Joseph, and B. R. Bloom. 2003. Interaction of the sensormodule of Mycobacterium tuberculosis H37Rv KdpD with members of theLpr family. Mol. Microbiol. 47:1075–1089.

152. Stock, A. M., J. M. Mottonen, J. B. Stock, and C. E. Schutt. 1989. Three-dimensional structure of CheY, the response regulator of bacterial che-motaxis. Nature 337:745–749.

153. Stock, A. M., et al. 1988. Phosphoproteins involved in bacterial signaltransduction. Cold Spring Harb. Symp. Quant. Biol. 53(Part 1):49–57.

154. Stock, J. B., A. J. Ninfa, and A. M. Stock. 1989. Protein phosphorylation andregulation of adaptive responses in bacteria. Microbiol. Rev. 53:450–490.

155. Supply, P., J. Magdalena, S. Himpens, and C. Locht. 1997. Identification ofnovel intergenic repetitive units in a mycobacterial two-component systemoperon. Mol. Microbiol. 26:991–1003.

156. Sureka, K., et al. 2007. Polyphosphate kinase is involved in stress-inducedmprAB-sigE-rel signalling in mycobacteria. Mol. Microbiol. 65:261–276.

157. Talaat, A. M., R. Lyons, S. T. Howard, and S. A. Johnston. 2004. Thetemporal expression profile of Mycobacterium tuberculosis infection inmice. Proc. Natl. Acad. Sci. U. S. A. 101:4602–4607.

158. Talaat, A. M., et al. 2007. Mycobacterial bacilli are metabolically activeduring chronic tuberculosis in murine lungs: insights from genome-widetranscriptional profiling. J. Bacteriol. 189:4265–4274.

159. Taneja, N. K., S. Dhingra, A. Mittal, M. Naresh, and J. S. Tyagi. 2010.Mycobacterium tuberculosis transcriptional adaptation, growth arrest anddormancy phenotype development is triggered by vitamin C. PLoS One5:e10860.

160. Tavares, R. C., et al. 2006. Cell proliferation and interferon-gamma re-sponse to recombinant MBP-3, NarL, MT-10.3, and 16 kDa Mycobacteriumtuberculosis antigens in Brazilian tuberculosis patients. Mem. Inst. OswaldoCruz 101:857–861.

161. Tekaia, F., et al. 1999. Analysis of the proteome of Mycobacterium tuber-culosis in silico. Tuber. Lung Dis. 79:329–342.

162. Vasudeva-Rao, H. M., and K. A. McDonough. 2008. Expression of theMycobacterium tuberculosis acr-coregulated genes from the DevR (DosR)regulon is controlled by multiple levels of regulation. Infect. Immun. 76:2478–2489.

163. Verreck, F. A., et al. 2009. MVA.85A boosting of BCG and an attenuated,phoP deficient M. tuberculosis vaccine both show protective efficacy againsttuberculosis in rhesus macaques. PLoS One 4:e5264.

164. Via, L. E., et al. 1996. Elements of signal transduction in Mycobacteriumtuberculosis: in vitro phosphorylation and in vivo expression of the responseregulator MtrA. J. Bacteriol. 178:3314–3321.

165. Voskuil, M. I., et al. 2003. Inhibition of respiration by nitric oxide induces aMycobacterium tuberculosis dormancy program. J. Exp. Med. 198:705–713.

166. Voskuil, M. I., K. C. Visconti, and G. K. Schoolnik. 2004. Mycobacteriumtuberculosis gene expression during adaptation to stationary phase andlow-oxygen dormancy. Tuberculosis (Edinb.) 84:218–227.

167. Waddell, S. J., et al. 2004. The use of microarray analysis to determine thegene expression profiles of Mycobacterium tuberculosis in response toanti-bacterial compounds. Tuberculosis (Edinb.) 84:263–274.

168. Walters, S. B., et al. 2006. The Mycobacterium tuberculosis PhoPR two-component system regulates genes essential for virulence and complex lipidbiosynthesis. Mol. Microbiol. 60:312–330.

169. Wang, S., J. Engohang-Ndong, and I. Smith. 2007. Structure of the DNA-binding domain of the response regulator PhoP from Mycobacterium tu-berculosis. Biochemistry 46:14751–14761.

170. Wayne, L. G., and L. G. Hayes. 1996. An in vitro model for sequential studyof shiftdown of Mycobacterium tuberculosis through two stages of nonrep-licating persistence. Infect. Immun. 64:2062–2069.

171. Weiss, V., and B. Magasanik. 1988. Phosphorylation of nitrogen regulatorI (NRI) of Escherichia coli. Proc. Natl. Acad. Sci. U. S. A. 85:8919–8923.

172. Wernisch, L., et al. 2003. Analysis of whole-genome microarray replicatesusing mixed models. Bioinformatics 19:53–61.

173. White, M. J., H. He, R. M. Penoske, S. S. Twining, and T. C. Zahrt. 2010.PepD participates in the mycobacterial stress response mediated throughMprAB and SigE. J. Bacteriol. 192:1498–1510.

174. White, M. J., et al. 2011. The HtrA-like serine protease PepD interacts withand modulates the Mycobacterium tuberculosis 35-kDa antigen outer en-velope protein. PLoS One 6:e18175.

175. Wilson, S. A., S. J. Wachira, R. A. Norman, L. H. Pearl, and R. E. Drew.1996. Transcription antitermination regulation of the Pseudomonas aerugi-nosa amidase operon. EMBO J. 15:5907–5916.

176. Wisedchaisri, G., et al. 2005. Structures of Mycobacterium tuberculosisDosR and DosR-DNA complex involved in gene activation during adapta-tion to hypoxic latency. J. Mol. Biol. 354:630–641.

177. Wisedchaisri, G., M. Wu, D. R. Sherman, and W. G. Hol. 2008. Crystalstructures of the response regulator DosR from Mycobacterium tubercu-losis suggest a helix rearrangement mechanism for phosphorylation activa-tion. J. Mol. Biol. 378:227–242.



http://mm

br.asm.org/

Dow

nloaded from


177a.World Health Organization. 2010. Global tuberculosis control. WorldHealth Organization, Geneva, Switzerland.

178. Wren, B. W., S. M. Colby, R. R. Cubberley, and M. J. Pallen. 1992. De-generate PCR primers for the amplification of fragments from genes en-coding response regulators from a range of pathogenic bacteria. FEMSMicrobiol. Lett. 78:287–291.

179. Yukl, E. T., A. Ioanoviciu, P. R. de Montellano, and P. Moenne-Loccoz.2007. Interdomain interactions within the two-component heme-based sen-sor DevS from Mycobacterium tuberculosis. Biochemistry 46:9728–9736.

180. Yukl, E. T., A. Ioanoviciu, M. M. Nakano, P. R. de Montellano, and P.Moenne-Loccoz. 2008. A distal tyrosine residue is required for ligand dis-crimination in DevS from Mycobacterium tuberculosis. Biochemistry 47:12532–12539.

181. Yukl, E. T., et al. 2011. Nitric oxide dioxygenation reaction in DevS and the

initial response to nitric oxide in Mycobacterium tuberculosis. Biochemistry50:1023–1028.

182. Zahrt, T. C., and V. Deretic. 2000. An essential two-component signal trans-duction system in Mycobacterium tuberculosis. J. Bacteriol. 182:3832–3838.

183. Zahrt, T. C., and V. Deretic. 2001. Mycobacterium tuberculosis signal trans-duction system required for persistent infections. Proc. Natl. Acad. Sci.U. S. A. 98:12706–12711.

184. Zahrt, T. C., C. Wozniak, D. Jones, and A. Trevett. 2003. Functionalanalysis of the Mycobacterium tuberculosis MprAB two-component signaltransduction system. Infect. Immun. 71:6962–6970.

185. Zhang, Y., K. A. Hatch, J. Bacon, and L. Wernisch. 2010. An integratedmachine learning approach for predicting DosR-regulated genes in Myco-bacterium tuberculosis. BMC Syst. Biol. 4:37.

Daniel J. Bretl obtained a B.S. in biologyfrom the University of Wisconsin—StevensPoint. He then received an M.S. in clinicalmicrobiology from the University of Wis-consin—La Crosse, investigating aspects ofmethicillin-resistant Staphylococcus aureusvirulence in the lab of William Schwan. Heis currently seeking a Ph.D. in microbiologyand molecular genetics at the Medical Col-lege of Wisconsin in the lab of Dr. ThomasC. Zahrt, investigating the role of MprABgene regulation in M. tuberculosis physiology and virulence.

Chrystalla Demetriadou received her B.S.in chemistry from the University of Cyprusand her M.S. in clinical laboratory sciencesfrom the University of Wisconsin—Milwau-kee. She also received an M.S. in microbi-ology and molecular genetics in the labora-tory of Dr. Thomas Zahrt at the MedicalCollege of Wisconsin, where she studiedgenes regulated by the MprAB two-compo-nent signal transduction system in Mycobac-terium tuberculosis. Ms. Demetriadou is cur-rently an analyst at the General State Laboratory of Cyprus.

Thomas C. Zahrt is an associate professorin the Department of Microbiology and Mo-lecular Genetics and the Center for Infec-tious Disease Research at the Medical Col-lege of Wisconsin. He received his B.A.from Kalamazoo College and his Ph.D.from the University of Illinois in Urbana-Champaign, and he conducted postdoctoralwork at the University of Michigan MedicalSchool in the laboratory of Vojo Deretic.Dr. Zahrt’s interests include the study ofintracellular bacterial pathogens and the regulatory factors utilized bythese organisms to adapt within the host. In particular, Dr. Zahrtstudies Mycobacterium tuberculosis and the importance of two-compo-nent signal transduction systems in the establishment, maintenance,and reactivation of this organism from long-term persistent infection.Dr. Zahrt also studies Francisella tularensis pathogenesis, including thegenetic programs necessary for growth and/or survival within tissues ofthe host, including macrophages and lung pneumocytes.



http://mm

br.asm.org/

Dow

nloaded from


Adaptation to Environmental Stimuli within the Host: Two ...terium species, including Mycobacterium...

Documents

Transcript of Adaptation to Environmental Stimuli within the Host: Two ...terium species, including Mycobacterium...