scieee Science in your language
[en] (orig)

In depth analysis of the mechanism of action of metal-dependent sigma factors: characterization of CorE2 from Myxococcus xanthus

Author: Marcos Torres, Francisco Javier; Pérez, Juana; Gomez Santos, Nuria; Moraleda-Munoz, Aurelio; Muñoz Dorado, Jose
Publisher: Zenodo
DOI: 10.1093/nar/gkw150
Source: https://zenodo.org/records/16632466/files/gkw150.pdf
Published online 6 Ma ch 2016 Nucleic Acids Resea ch, 2016, Vol. 44, No. 12 5571–5584
doi: 10.1093/na /gkw150
In dep h analysis o he mechanism o ac ion o
me al-dependen sigma ac o s: cha ac e iza ion o
Co E2 om
Myxococcus xan hus
F ancisco Ja ie Ma cos-To es, Juana P´
e ez, Nu ia G´
omez-San os,
Au elio Mo aleda-Mu˜
noz and Jos´
eMu˜
noz-Do ado*
Depa amen o de Mic obiolog´
ıa, Facul ad de Ciencias, Uni e sidad de G anada, A da. Fuen enue a s/n, E-18071
G anada, Spain
Recei ed July 30, 2015; Re ised Feb ua y 26, 2016; Accep ed Feb ua y 29, 2016
ABSTRACT
Ex acy oplasmic unc ion sigma ac o s ep esen
he hi d pilla o signal- ansduc ion mechanisms
in bac e ia. The a ie y o s imuli hey ecognize
and mechanisms o ac ion hey use ha e allowed
hei classi ica ion in o mo e han 50 g oups. We
ha e cha ac e ized Co E2 om
Myxococcus xan-
hus
, which belongs o g oup ECF44 and up egula es
he exp ession o wo genes when i is ac i a ed by
cadmium and zinc. Sigma ac o s o his g oup con-
ain a Cys- ich domain (CRD) a he C e minus which
is essen ial o de ec ing me als. Poin mu a ions a
he six Cys esidues o he CRD ha e e ealed he
con ibu ion o each esidue o Co E2 ac i i y. Some
o hem a e essen ial, while o he s a e ei he dis-
pensable o hei mu a ions only sligh ly a ec he
ac i i y o he p o ein. Howe e , impo an ly, mu a-
ion o Cys174 comple ely shi s he speci ici y o
Co E2 om cadmium o coppe , indica ing ha he
Cys a angemen o he CRD de e mines he me al
speci ici y. Mo eo e , he conse ed CxC mo i lo-
ca ed be ween he ␴2 domain and he ␴4.2 egion
has also been ound o be essen ial o ac i i y. The
esul s p esen ed he e con ibu e o ou unde s and-
ing o he mechanism o ac ion o me al-dependen
sigma ac o s and help o de ine new common ea-
u es o he membe s o his g oup o egula o s.
INTRODUCTION
Myxococcus xan hus is a soil ␦-p o eobac e ium o he
g oup o myxobac e ia used as a model o s udy mul icellu-
la beha io and di e en ia ion due o i s unique and com-
plex li e cycle. M. xan hus cells eed as coo dina ed g oups
un il nu ien s a e deple ed. Upon s a a ion hey ini ia e
a de elopmen al p og am, du ing which cells mus p oduce
and espond o se e al signals in o de o agg ega e and di -
e en ia e in o myxospo es, which a e esis an o a a ie y
o ad e se condi ions (1,2).
Bac e ia adap o en i onmen al changes by using a
la ge numbe o signal- ansduc ion sys ems which con-
nec ex acellula inpu s wi h he app op ia e cellula e-
sponses. The e a e h ee main common and uni e sally
p esen signal- ansduc ion mechanisms in bac e ia: one-
and wo-componen sys ems, and he ex acy oplasmic
unc ion (ECF) sigma ac o s (3–6). Mo eo e , he e is a
ou h signal- ansduc ion sys em less widesp ead among
p oka yo es which in ol es Se /Th p o ein kinases and
phospha ases (7,8).
ECF sigma ac o s belong o g oup 4 o he ␴70 amily
o sigma ac o s (9). Membe s o his g oup a e small p o-
eins ha con ain only wo o he ou conse ed domains
ound in sigma ac o s o g oups 1 and 2, he ␴2 and he
␴4 domains. The ␴2 domain is essen ial o ecogni ion o
he −10 p omo e sequences and coupling wi h he RNA
polyme ase co e enzyme, while he ␴4.2 egion (included in
he ␴4 domain) is equi ed o ecogni ion o he −35 p o-
mo e egions (10). ECF sigma ac o s a e abundan and
di e se in bac e ial genomes, especially in hose wi h a com-
plex li e cycle (11). Many ECF sigma ac o s unc ion wi h
a cogna e an i-sigma ac o . An i-sigma ac o s a e usu-
ally memb ane-ancho ed p o eins, co-exp essed wi h hei
cogna e sigma ac o , which con ain he senso domains o
hese signal- ansduc ion sys ems. In absence o he igh
en i onmen al s imulus, an i-sigma ac o s seques e hei
sigma ac o s in he memb ane and block he exp ession
o speci ic genes. When an i-sigma ac o s do de ec hese
ex e nal signals, sigma ac o s a e eleased, ec ui ing he
RNA polyme ase co e enzyme and binding o DNA o ini i-
a e ansc ip ion o he genes equi ed o espond o s imuli
(6,12–14).
*To whom co espondence should be add essed. Tel: +34 958243183; Fax: +34 958249486; Email: jdo ado@ug .es
P esen add ess: Nu ia G´
omez-San os, Max Planck Ins i u e o Te es ial Mic obiology, 35043 Ma bu g, Ge many.
C
The Au ho (s) 2016. Published by Ox o d Uni e si y P ess on behal o Nucleic Acids Resea ch.
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by-nc/4.0/), which
pe mi s non-comme cial e-use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Fo comme cial e-use, please con ac
jou nals.pe [email protected]
Downloaded om h ps://academic.oup.com/na /a icle/44/12/5571/2457481 by Uni e sidad de G anada - Biblio eca use on 31 July 2025
5572 Nucleic Acids Resea ch, 2016, Vol. 44, No. 12
The mechanism o ac i a ion o ECF sigma ac o s, o-
ge he wi h hei sequence simila i ies, has allowed he
classi ica ion o hese ansc ip ional egula o s in o mo e
han 50 g oups (13). E en hough he mechanism desc ibed
abo e is he main mode o ac i a ion o ECF sigma ac o s,
h ee o he mechanisms ha e been epo ed o hese egu-
la o s, in which an i-sigma ac o s do no pa icipa e. One
o hese o he mechanisms is used by g oups ECF32 and
ECF39, which consis s o di ec ansc ip ion o he sigma
ac o (15,16). A hypo he ical phospho elay in ol ing a
Se /Th p o ein kinase co- ansc ibed wi h he sigma ac-
o has been pos ula ed o g oups ECF43 and ECFSTK1–
4(5,17). Finally, some ECF sigma ac o s con ain a C-
e minal ex ension esponsible o he modula ion o hei
own ac i i y. To da e only ou g oups ha e been desc ibed
wi h C- e minal ex ensions: ECF41, ECF42, ECF01-Gob
and ECF44 (5,6,17,18).
Myxococcus xan hus Co E is he ounding membe and
he only cha ac e ized sigma ac o o he g oup ECF44.
This sigma ac o con e s coppe esis ance o M. xan hus
by egula ing he exp ession o he P1B- ype ATPases CopA
and CopB, and he mul icoppe oxidase CuoB (14,19–21).
In con as o mos ECF sigma ac o s, Co E only pa ially
egula es i s own exp ession, and i s ac i a ion s a e does
no depend on an an i-sigma ac o . Co E- egula ed genes
show a peak o exp ession a 2 h a e coppe addi ion ha
apidly dec eases due o Co E inac i a ion. I has been p o-
posed ha Cu(II) ac i a es Co E, allowing DNA-binding,
whe eas Cu(I) inac i a es he sigma ac o p e en ing DNA
binding. A conse ed C- e minal Cys- ich domain (CRD)
wi h 38 esidues in Co E con ols he ac i a ion and in-
ac i a ion media ed by coppe o his ECF sigma ac o .
Poin mu a ions a each Cys esidue o he CRD ha e e-
ealed ha ce ain key esidues play a ole in Co E ac i a-
ion and/o inac i a ion (14).
We ha e iden i ied a second membe o he ECF44 g oup
in he M. xan hus genome, which has been named co E2
(MXAN 5263). In he p esen wo k we dissec he mecha-
nism o ac ion o Co E2, which is ac i a ed by cadmium and
zinc, and egula es he exp ession o a leas a ca ion e lux
pump and a glyoxal oxidase and Kelch domain con aining
p o ein. We ha e compa ed he me al esponses and speci-
ici ies o Co E and Co E2, and ha e pe o med an in-dep h
in es iga ion o hei CRD Cys a angemen o unde s and
hei di e ences. We ha e ound ha a change in jus one
Cys esidue o he CRD o Co E2 shi s he esponse o
he sigma ac o om cadmium o coppe . Fu he mo e, we
ha e demons a ed ha a CxC mo i loca ed be ween he ␴2
domain and he ␴4.2 egion, conse ed in all he Co E-like
ECF sigma ac o s, is essen ial o he ac i i y o Co E2.
MATERIALS AND METHODS
Bac e ial s ains, plasmids and g ow h condi ions
Geno ypes o M. xan hus and Esche ichia coli s ains, plas-
mids and oligonucleo ides used in his s udy a e lis ed in
Supplemen a y Tables S1, 2 and 3, espec i ely. E. coli
s ains we e g own in lysogenic b o h (LB) (22)a 37
◦C.
Aga pla es con ained 1.5% Bac o-aga (Di co), which we e
supplemen ed wi h 40 ␮g/ml X-gal (5-b omo-4-chlo o-
3-indolyl-␤-D-galac opy anoside), kanamycin (25 ␮g/ml)
and/o e acycline (25 ␮g/ml) when necessa y. M. xan-
hus s ains we e g own in CTT medium (23)a 30
◦C wi h
igo ous shaking (300 pm). CTT aga pla es (1.5% aga )
we e supplemen ed wi h X-gal (100 ␮g/ml), galac ose (10
mg/ml), kanamycin (80 ␮g/ml) and/o e acycline (15
␮g/ml). When needed, di e en me als we e also added o
he medium a he concen a ions indica ed in each igu e.
To induce de elopmen , s a a ion medium CF (23)was
used. Cells exponen ially g owing o app oxima ely 3.0 ×
108cells/ml (op ical densi y a 600 nm [OD600]o 1)we e
concen a ed and esuspended o an OD600 o 15 in TM
bu e (10 mM T is–HCl [pH 7.6], 1 mM MgSO4). Ten
mic oli e d ops we e spo ed on o CF aga pla es supple-
men ed wi h he me als indica ed in he igu es and/o X-gal
(100 ␮g/ml) and incuba ed a 30◦C. F ui ing bodies we e
obse ed wi h an Olympus dissec ing mic oscope.
Nucleic acid manipula ions
Rou ine molecula biology echniques we e used o nucleic
acid manipula ions (22). The a ious plasmids we e in o-
duced in o E. coli by hea -shock ans o ma ion and in o M.
xan hus by elec opo a ion (24). To al RNA was ex ac ed
om M. xan hus wi h he High Pu e RNA Isola ion ki p o-
ided by Roche. Samples we e hen ea ed wi h DNAse I
(Sigma) o ensu e emo al o ch omosomal DNA. Comple-
men a y DNA (cDNA) was ob ained by e e se ansc ip-
ion (Supe Sc ip III e e se ansc ip ase, Li e Technolo-
gies) om he RNA empla e using he p ime 65RT, which
anneals o he gene MXAN 5265 (Supplemen a y Table S3).
A polyme ase chain eac ion (PCR) was hen pe o med
wi h he p ime s lis ed in Supplemen a y Table S3, using
o al RNA o cDNA as a empla e.
Cons uc ion o in- ame dele ion mu an s
The in- ame dele ion mu an s used in his s udy we e ob-
ained as p e iously epo ed (21). To gene a e he co e-
sponding plasmids (lis ed in Supplemen a y Table S2), se-
quences ups eam and downs eam o he M. xan hus e-
gions o be dele ed we e ampli ied by PCR wi h wild- ype
(WT) ch omosomal DNA as a empla e, he p ime s lis ed
in Supplemen a y Table S3, and he high- ideli y DNA-
polyme ase P imeSTAR HS (Taka a). The PCR p oduc s
we e diges ed and liga ed o ec o pBJ113 (25), which
had p e iously been diges ed wi h he same es ic ion en-
zymes o ob ain he desi ed plasmids (Supplemen a y Ta-
ble S2). The esul ing plasmids we e in oduced in o M.
xan hus s ains by elec opo a ion. Kanamycin esis an
(KmR) me odiploids we e selec ed om CTT aga pla es
supplemen ed wi h his an ibio ic and analyzed by Sou h-
e n blo hyb idiza ion o co obo a e he p ope ecombina-
ion e en s. Posi i e s ains we e g own on CTT aga pla es
wi hou kanamycin and con aining 1% galac ose, a o ing
he loss o he plasmid by a second homologous ecombina-
ion. Sou he n blo analysis was used o sc een kanamycin-
sensi i e (KmS) and galac ose- esis an (GalR) colonies o
he loss o he WT allele.
Downloaded om h ps://academic.oup.com/na /a icle/44/12/5571/2457481 by Uni e sidad de G anada - Biblio eca use on 31 July 2025
Nucleic Acids Resea ch, 2016, Vol. 44, No. 12 5573
Cons uc ion o s ains ha bo ing lacZ usions and ␤-
galac osidase assays
Plasmids ha bo ing lacZ usions (Supplemen a y Table S2)
we e cons uc ed as p e iously epo ed (20). In summa y,
o gene a e he co esponding plasmids, PCR was pe -
o med using ch omosomal DNA o M. xan hus as a em-
pla e and he p ime s lis ed in Supplemen a y Table S3.
PCR p oduc s we e hen diges ed wi h hei espec i e e-
s ic ion enzymes and liga ed in o ec o pKY481 (26) di-
ges ed wi h he same enzymes o gene a e ansc ip ional
lacZ usions. M. xan hus s ains we e elec opo a ed wi h
hese plasmids o gene a e he desi ed s ains. The esul an
KmR ecombinan s ains (Supplemen a y Table S1) we e
con i med by Sou he n blo analysis.
Fo quali a i e ␤-galac osidase ac i i y analyses, cells
we e concen a ed o an OD600 o 15 and spo ed on o CTT
o CF aga pla es con aining 100 ␮g/ml X-gal and he ad-
di i es indica ed in each igu e. Fo quan i a i e analysis,
cells g own on CTT liquid medium we e concen a ed a
an OD600 o 15 and spo ed on o CTT o CF aga pla es.
␤-galac osidase-speci ic ac i i y was de e mined in cell ex-
ac s ob ained a a ious ime poin s by sonica ion as
p e iously epo ed (21), and is exp essed as nmol o o-
ni ophenol p oduced pe min and mg o p o ein. All ex-
ac s we e assayed in iplica e, and alues we e a e aged
om h ee independen measu emen s.
Si e-di ec ed mu agenesis
Single amino acid subs i u ions in Co E and Co E2 we e
pe o med using he QuikChange II si e-di ec ed mu a-
genesis ki (Agilen ) as ecommended by he manu ac-
u e . Plasmids pNG00 and pMT00 (Supplemen a y Table
S2), con aining he WT co E and co E2 sequences, espec-
i ely, we e used as empla es. The oligonucleo ides used as
p ime s (lis ed in Supplemen a y Table S3) we e designed
using he QuikChange P ime Design P og am (h p://
www.genomics.agilen .com/). The ampli ied plasmids we e
diges ed wi h DpnI and ans o med in o E. coli o ob ain
he mu an plasmids lis ed in Supplemen a y Table S2. All
plasmids we e sequenced o con i m he p esence o he de-
si ed mu a ions and he absence o unwan ed mu a ions,
and we e in oduced by elec opo a ion in o M. xan hus
JM51EBZY (cuoB-lacZ-co E) o mu a ions in Co E,
and in o M. xan hus JM52IF3ZY5 (5265-lacZ-co E2) o
poin mu a ions in Co E2. The mu an s ains ob ained
a e lis ed in Supplemen a y Table S1. Plasmid pMT00 was
also elec opo a ed in JM52IF3ZY5 o cons uc he s ain
JM52SDM00, which was used as a con ol o Co E2 mu-
a ion analyses. All Te Rand KmR ecombinan s we e ana-
lyzed by Sou he n blo . As in p e ious s udies o he Co E
mu a ion, s ain JM00BZY was used as a con ol (14).
S udies on he s abili y o he p o eins ha bo ing poin mu a-
ions
In o de o de e mine whe he he p o eins ha bo ing poin
mu a ions o Co E and Co E2 we e s able in M. xan hus,
genes encoding hem and he WT genes we e cloned wi h
an N- e minal S ag and unde con ol o he cons i u i e
oa p omo e (27). A 781-bp agmen ups eam o he oa
gene (MXAN 1450) was ampli ied by PCR, using he ap-
p op ia e oligonucleo ide pai lis ed in Supplemen a y Ta-
ble S3 as p ime s. The PCR p oduc was in oduced in-
ame wi h he coding sequence o he S ag p esen in he
pRSFDue -1 ec o (No agen). The esul ing plasmid was
named pPOa -S and used o ampli y he oa p omo e used
o he S ag. Nex , he WT co E and co E2 genes we e ampli-
ied om he M. xan hus ch omosomal DNA, while he co -
esponding plasmids (Supplemen a y Table S2) we e used
o ampli y co E and co E2 genes ha bo ing poin mu a ions
and we e cloned in- ame wi h he S- agged oa p omo e
in he pBJ113 ec o (25). The esul ing plasmids we e in-
oduced by elec opo a ion in o a co E ( hose con ain-
ing he WT co E o poin mu a ions in co E)o co E2
M. xan hus s ain ( hose con aining he WT co E2 o poin
mu a ions in co E2). Se e al kanamycin- esis an (KmR)
colonies we e analyzed by PCR o con i m he p ope e-
combina ion e en . To de ec he S- agged p o eins, he di -
e en s ains we e g own o 24 h in CTT supplemen ed
wi h 200 ␮M o ZnNO3, a e which he cells we e ha -
es ed in TM bu e con aining a p o ease inhibi o cock ail
(P omega) and dis up ed by sonica ion. The sonica ed sam-
ples we e cen i uged and he p o ein concen a ion o he
supe na an s was de e mined by he Bio-Rad p o ein assay
ki using bo ine se um albumin as s anda d. P o eins we e
sepa a ed by sodium dodecyl sulpha e-polyac ylamide gel
elec opho esis and ans e ed on o a memb ane o PVDF.
S- agged p o eins we e de ec ed using an S-p o ein HRP
Conjuga e (No agen) an ibody, which is conjuga ed wi h
ho se adish pe oxidase, using 1-S ep Ul a TMB Blo ing
Solu ion (Pie ce) as he subs a e, ollowing he ins uc ions
speci ied by he manu ac u e .
Bioin o ma ic analysis
The lis o Co E-like ECF sigma ac o s was upda ed,
as p e iously epo ed (14), by BLASTP analysis o all
genome and p o ein sequences deposi ed in he da abase o
he Na ional Cen e o Bio echnology In o ma ion (h p://
www.ncbi.nlm.nih.go /genome/b owse/). P o ein sequence
alignmen s we e pe o med using he Clus alW p og am
(28), and a g aphic ep esen a ion o he esul s was gene -
a ed wi h ESP ip .cgi Ve sion 3.0 (29)(h p://esp ip .ibcp.
/ESP ip /cgi-bin/ESP ip .cgi). The domain a chi ec u e
o p o eins was analyzed agains he P am da abase (30).
RESULTS
Co E-like ECF sigma ac o s in myxobac e ia
A BLASTP analysis in sea ch o Co E-like ECF sigma ac-
o s esul ed in he iden i ica ion o 67 o hese egula o s in
bac e ia, 17 o which a e ound in species o he o de Myx-
ococcales (Supplemen a y Figu e S1). In e es ingly, all he
myxobac e ial genomes so a sequenced encode a leas one
Co E-like sigma ac o , wi h he excep ion o se e al species
o he genus Anae omyxobac e (whe e one is only ound in
he s ain Fw109) and Haliangium och aceum.TheMyx-
ococcus s ipi a us genome ha bo s h ee sigma ac o s o
his ype (Supplemen a y Figu e S1). In he case o M. xan-
hus, in addi ion o wo comple e Co E-like sigma ac o s,
Downloaded om h ps://academic.oup.com/na /a icle/44/12/5571/2457481 by Uni e sidad de G anada - Biblio eca use on 31 July 2025
5574 Nucleic Acids Resea ch, 2016, Vol. 44, No. 12
Figu e 1. The Myxococcus xan hus genome con ains h ee genes wi h con-
se ed C- e minal CRD domains. (A) Domain a chi ec u e o he h ee
p o eins wi h a CRD domain. Co E and Co E2 ha e he ␴2 domain
(Sigma70 2, PF04542) and he ␴4.2 egion (sigma70 4 2, PF08281) yp-
ical o ECF sigma ac o s. MXAN 0974 shows sequence simila i ies o
40.4% wi h Co E and 36.5% wi h Co E2 wi hin he ␴4.2 egion. (B)Se-
quence alignmen o he h ee CRDs showing he conse ed Cys esidues.
Numbe s indica e he posi ion o he i s and las esidue o he sequences
shown o each p o ein. Iden ical esidues in wo p o eins a e w i en in ed
and highligh ed in yellow, and hose ha a e iden ical in he h ee p o eins
a e w i en in whi e and highligh ed in ed. (C) Gene ic en i onmen o he
h ee genes wi h a CRD ( ep esen ed in ed). Genes o hese egions ha
encode p o eins associa ed wi h me als a e d awn ollowing he colo code
indica ed a he bo om o he panel.
a hi d gene (MXAN 0974) is p esen which encodes a p o-
ein wi h a egion esembling ha o ␴4.2 o sigma ac o s
andwi haCRD(Figu e1A and B). Howe e , his p o ein
is no expec ed o unc ion as an ECF sigma ac o because
i lacks he ␴2 domain. The pa alog o Co E ound in M.
xan hus has been designa ed as Co E2 (MXAN 5263), and
in his epo we ha e ocused on i s cha ac e iza ion. Anal-
ysis o he gene ic con ex o co E2 has e ealed a gene up-
s eam encoding a hypo he ical lipop o ein (MXAN 5262)
which has been p edic ed o be he cogna e an i-sigma ac-
o o Co E2 (12). Mo eo e , downs eam o co E2 he e
a e wo genes which encode p o eins associa ed wi h me als.
MXAN 5264 (wi h P am PF01545) is simila o ca ion e -
lux sys ems, such as CzcD o Cup ia idus me alidu ans (31),
while MXAN 5265 co esponds o a me alloenzyme wi h a
Kelch domain and a glyoxal oxidase domain (PF07250 and
PF01344, espec i ely), which esembles he de elopmen-
al p o ein Fb B o S igma ella au an iaca (32)(Figu e1C).
Genes encoding p o eins in ol ed in me al homeos asis and
de oxi ica ion a e also ound in he p oximi y o co E and
MXAN 0974 (Figu e 1C). This obse a ion, along wi h he
p esence o a CRD, sugges s ha Co E2, like Co E, migh
be me al- esponsi e.
Genes o he co E2 egion, bu no co E2, a e up egula ed by
me als
To analyze he me al esponse o genes in he co E2 e-
gion, ou genes (MXAN 5262,co E2,MXAN 5264 and
MXAN 5265) we e es ed o me al egula ion. Plasmids
con aining ansc ip ional usions be ween hese ou genes
and E. coli lacZ (Figu e 2A) we e elec opo a ed in o
he WT s ain o M. xan hus. The esul ing KmRs ains
JM52ZY2 (5262-lacZ), JM52ZY3 (co E2-lacZ), JM52ZY4
(5264-lacZ) and JM52ZY5 (5265-lacZ) (Supplemen a y
Table S1) we e con i med by Sou he n blo . The s ains con-
aining he usions, and he WT s ain as a nega i e con-
ol, we e spo ed on o CTT (g ow h) and CF (de elop-
men ) aga pla es con aining 100 ␮g/ml o X-gal and se -
e al me als o quali a i ely sc een o gene exp ession. Su -
p isingly, none o he ou genes was up egula ed by cop-
pe (Figu e 2B), as happened wi h Co E- egula ed genes
(14). Howe e , wo genes downs eam o co E2 we e up eg-
ula ed in he p esence o cadmium and zinc (Figu e 2B).
Bo h genes exhibi an exp ession p o ile wi h cadmium in
which a pla eau is eached se e al hou s a e addi ion o
he me al (Figu e 2C). In con as , he exp ession o hese
wo genes apidly inc eases a e he addi ion o zinc, ex-
hibi ing a maximum a 2 h. The ea e he exp ession le els
sligh ly dec ease, emaining qui e high o an ex ended pe-
iod (Figu e 2D). None o he me als es ed was able o a -
ec co E2 and MXAN 5262 exp ession (Figu e 2B). How-
e e , an up egula ion o he ou genes was obse ed du ing
de elopmen in he absence o me als (Figu e 2E), al hough
a di e en le els (Figu e 2F), indica ing egula ion o all
hese genes du ing ui ing-body o ma ion.
Due o hei p oximi y and o ien a ion, and he dual ex-
p ession p o ile o he genes in he clus e , co-exp ession o
hese ou genes was examined unde wo di e en condi-
ions: g ow h on CTT aga pla es wi h 0.1 mM cadmium
and de elopmen on CF aga pla es wi h no me als added.
To al RNA was ex ac ed a e 48 h o incuba ion unde
hese wo condi ions, and i was used as a empla e o syn-
hesis o cDNA using p ime 65RT (Supplemen a y Fig-
u e S2). Using he wo cDNAs as empla es and he s a -
egy depic ed in Supplemen a y Figu e S2A, i was ound
ha wo di e en ially egula ed ope ons a e ound in he
co E2 egion, one con aining he ou genes (MXAN 5262
h ough MXAN 5265), which is exp essed du ing de elop-
men (Supplemen a y Figu e S2B), and a second one, con-
aining only wo genes (MXAN 5264 and MXAN 5265),
which is up egula ed by cadmium (Supplemen a y Figu e
Downloaded om h ps://academic.oup.com/na /a icle/44/12/5571/2457481 by Uni e sidad de G anada - Biblio eca use on 31 July 2025
Nucleic Acids Resea ch, 2016, Vol. 44, No. 12 5575
Figu e 2. Exp ession o genes loca ed in he p oximi y o co E2.(A) Schema ic ep esen a ion o he co E2 con ex . Red a ows indica e Esche ichia coli
lacZ ansc ip ional usions. (B) Quali a i e analysis o lacZ usions o genes o he co E2 egion in he p esence o a ious me als. Cells ha bo ing usions
be ween genes o he co E2 clus e and lacZ we e inocula ed on CTT medium (g ow h) con aining X-gal and di e en me als. Pic u es we e aken a e
96 h o incuba ion. (Cand D) Quan i ica ion o ␤-galac osidase-speci ic ac i i y o he ou usions wi h lacZ. Cells we e g own on CTT wi hou me als
(o ange lines) o con aining ei he 0.1 mM cadmium (C, ed lines) o 0.5 mM zinc (D, g een lines), and samples we e ha es ed a he ime poin s indica ed
in each igu e. Speci ic ac i i y was de e mined as indica ed in ‘Ma e ials and Me hods’ sec ion. (E) Quali a i e analysis o gene exp ession o he co E2
clus e du ing de elopmen . Cells we e spo ed on o bo h CTT aga pla es (g ow h) and CF aga pla es (de elopmen ) con aining X-gal and no me als.
Pic u es we e aken a e 96 h o incuba ion. Ba in panels B and E ep esen s 0.5 mm. (F) Quan i ica ion o ␤-galac osidase-speci ic ac i i y du ing g ow h
(o ange lines) and de elopmen (blue lines) in he absence o me als. Cells o each s ain we e spo ed on o CF aga pla es, and a he ime poin s indica ed
in he igu es hey we e ha es ed and analyzed o ␤-galac osidase ac i i y. E o ba s in panels C, D and F indica e s anda d de ia ions. Ex ac s we e
assayed in iplica e, and alues we e a e aged om h ee independen measu emen s. Please no e ha he g aphs ha e di e en scales.
S2C). These da a a e in good ag eemen wi h he da a p e-
sen ed in Figu e 2.
The mu an co E2 is mo e sensi i e o cadmium and zinc
han he WT s ain, and exhibi s delay in de elopmen
To de e mine whe he Co E2 is in ol ed in me al de ox-
i ica ion, an in- ame dele ion mu an (co E2) was con-
s uc ed, lacking he essen ial ␴2 domain o he sigma ac-
o (Figu e 3A), and he pheno ype o he mu an was ana-
lyzed o me al sensi i i y. When he mu an co E2 and
he WT s ain we e cul u ed in liquid CTT medium, no
signi ican di e ence in g ow h was obse ed be ween he
wo s ains when se e al concen a ions o zinc o cad-
mium we e es ed (Supplemen a y Figu e S3). Since me al
homeos asis mechanisms equi e p eadap a ion o be ully
ac i e (33), WT and co E2 s ains we e i s g own o
24 h in CTT medium con aining ei he 10 ␮M cadmium
o 0.2 mM zinc. Cells we e hen ans e ed o CTT liq-
uid cul u es con aining a ying concen a ions o me als.
As shown in Figu e 3B, zinc ole ance was educed in he
co E2 p eadap ed cells. In con as , cadmium ole ance
was only sligh ly educed in he mu an (Figu e 3C). These
da a indica e ha Co E2 is in ol ed in con e ing esis ance
o some me als, and sugges ha his sigma ac o could be
egula ing he exp ession o he wo genes loca ed down-
s eam o i sel , which a e up egula ed by hese wo me als.
Downloaded om h ps://academic.oup.com/na /a icle/44/12/5571/2457481 by Uni e sidad de G anada - Biblio eca use on 31 July 2025

5576 Nucleic Acids Resea ch, 2016, Vol. 44, No. 12
Figu e 3. Pheno ype o he co E2 mu an . (A) Domains p esen in Co E2 in he WT s ain and in he mu an s co E2 and co E2CRD.(B) Cells o he
WT s ain (blue line) and he co E2 mu an ( ed line) we e g own o 24 h in CTT con aining 0.2 mM zinc and hen dilu ed o an OD600 o 0.05 in CTT
medium con aining 0, 0.2, 0.3 and 0.4 mM zinc. OD600 was de e mined a e 32 h o incuba ion. (C) As in panel B, cells we e g own o 24 h in 10 ␮M
cadmium and hen dilu ed o an OD600 o 0.05 in CTT medium con aining 0, 10, 25 and 40 ␮M cadmium. OD600 was de e mined a e 32 h o incuba ion.
E o ba s in panels B and C indica e s anda d de ia ions. Values a e a e ages o h ee expe imen s. (D) F ui ing body o ma ion o he WT s ain and
he co E2 mu an in he absence o me als. (E) F ui ing body o ma ion o he WT s ain and he co E2 mu an on CF medium wi h 200 ␮M zinc. (F)
F ui ing body o ma ion o he WT s ain and he co E2 mu an on CF medium wi h 20 ␮M cadmium. Pic u es in panels D, E, and F we e aken a he
ime indica ed. Ba in panels D, E, and F ep esen s 1 mm.
As he ou genes o he ope on a e induced du ing de el-
opmen , i was also es ed whe he he co E2 mu an ex-
hibi s de elopmen al de ec s. When d opped on o CF s a -
a ion medium he co E2 mu an exhibi ed a clea delay
in de elopmen , al hough no mal ui ing bodies we e ob-
se ed a 72 h o incuba ion (Figu e 3D). This delay in de el-
opmen indica es ha his sigma ac o mus be egula ing
he exp ession o some genes equi ed o he p ope im-
ing o ui ing body o ma ion. When de elopmen was an-
alyzed in he p esence o zinc, a sligh ly longe delay in ui -
ing body o ma ion was obse ed in bo h he WT s ain and
he mu an (Figu e 3E). In con as , ui ing bodies we e
no obse ed in he mu an s ain spo ed on CF medium
con aining 20 ␮M cadmium e en a e 72 h o incuba ion,
while he WT s ain de eloped almos no mally a 48 h (Fig-
u e 3F). These esul s co obo a e ha Co E2 plays a ole
in me al de oxi ica ion as well as in de elopmen .
Downloaded om h ps://academic.oup.com/na /a icle/44/12/5571/2457481 by Uni e sidad de G anada - Biblio eca use on 31 July 2025
Nucleic Acids Resea ch, 2016, Vol. 44, No. 12 5577
Co E2 is con olling he cadmium- and zinc-dependen ex-
p ession o genes MXAN 5264 and MXAN 5265
As we ha e demons a ed, genes MXAN 5264 and
MXAN 5265 o m an ope on which is up egula ed by
cadmium and zinc (Figu e 2and Supplemen a y Figu e
S2). To elucida e whe he Co E2 is esponsible o he
me al-dependen exp ession o his ope on, he co E2
mu an was used as gene ic backg ound o in oduce he
ansc ip ional usion 5265-lacZ, o ob ain he s ain
JM52IF3ZY5 (5265-lacZ-co E2). This s ain was used
o compa e he up egula ion o he gene MXAN 5265 by
cadmium and zinc o he WT s ain. As shown in Figu e
4A and B, lack o a unc ional Co E2 esul s in he absence
o bo h cadmium- and zinc-dependen exp ession o he
gene MXAN 5265, demons a ing ha he me al induc-
ion o his ope on is con olled by his sigma ac o . I
emains o be elucida ed why he pheno ype o he co E2
mu an du ing g ow h is mo e d ama ic wi h zinc han
wi h cadmium (Figu e 3), whe eas genes egula ed by his
sigma ac o exhibi highe exp ession le els wi h cadmium
(Figu e 2). Howe e , one explana ion could be ha he
p o eins esponsible o me al de oxi ica ion egula ed by
Co E2 exhibi a highe a ini y o zinc han o cadmium.
As he ou genes a e induced du ing de elopmen in he
absence o me al, i was also es ed whe he Co E2 was
also esponsible o his exp ession. The s ains 5265-lacZ-
co E2 and he WT ha bo ing he same lacZ usion we e
pla ed on o CF aga wi hou me als. As shown in Figu e
4C, Co E2 is no esponsible o he exp ession o gene
MXAN 5265 du ing de elopmen . This esul suppo s he
no ion o a complex egula ion o his ope on, wi h se e al
ansc ip ional egula o s in ol ed.
In es iga ing he mode o ac ion o Co E-like ECF sigma ac-
o s
In his sec ion, he mechanism o ac ion o Co E2 will be
compa ed wi h ha epo ed o Co E, in o de o iden i y
common ea u es o he whole ECF44 g oup o sigma ac-
o s and also he peculia i ies o each pa icula egula o .
Co E2 does no egula e i s own exp ession. As shown
abo e (Figu e 2and Supplemen a y Figu e S2), co E2 is
pa o an ope on ha is only induced du ing de elopmen ,
bu no by me als. Mo eo e , he exp ession o he gene
MXAN 5265 du ing de elopmen , ep esen ing he ope on,
does no depend on Co E2, indica ing ha Co E2 does
no egula e i s own exp ession, in con as o wha has
been epo ed o o he ECF sigma ac o s, including Co E
(9,14). Howe e , due o he low exp ession le els o co E2,
␤-galac osidase ac i i y was di ec ly quan i ied in he s ain
JM52ZY3 (co E2-lacZ) du ing de elopmen , bo h in he
p esence and in he absence o 7.5 ␮M cadmium. As shown
in Figu e 5A, co E2 exp ession is no up egula ed by cad-
mium, as no signi ican di e ence be ween he wo condi-
ions was obse ed. Fu he mo e, o suppo hese da a, he
s ain JM52IF3ZY3 (co E2-lacZ-co E2) was cons uc ed
in oducing he co E2-lacZ ansc ip ional usion in o he
co E2 mu an . Quan i a i e and quali a i e analyses o
his s ain du ing de elopmen wi h and wi hou cadmium
ende ed simila exp ession pa e ns o hose o he WT
Figu e 4. Co E2 egula ion o gene exp ession. (Aand B) Exp ession
o he gene MXAN 5265 in he WT s ain and he mu an s co E2 and
co E2CRD. The s ains we e incuba ed on CTT aga pla es con aining
ei he 0.1 mM cadmium (A) o 0.5 mM zinc (B), and X-gal o moni-
o ␤-galac osidase ac i i y. Pic u es we e aken a he ime poin s indi-
ca ed a he op o each pic u e. Numbe s below each pic u e indica e ␤-
galac osidase-speci ic ac i i y ob ained o each s ain. (C) Exp ession o
MXAN 5265 du ing de elopmen in he WT s ain and he co E2 mu-
an . Cells we e spo ed on o CF aga pla es wi hou me als. As in pan-
els A and B, he ime a which each pic u e was aken is shown abo e i ,
and numbe s below he pic u es indica e ␤-galac osidase-speci ic ac i i y
a ha ime poin . Values a e a e age o h ee expe imen s. Ba ep esen s
0.5 cm.
Downloaded om h ps://academic.oup.com/na /a icle/44/12/5571/2457481 by Uni e sidad de G anada - Biblio eca use on 31 July 2025
5578 Nucleic Acids Resea ch, 2016, Vol. 44, No. 12
Figu e 5. co E2 au o- egula ion. (A) Quan i a i e analysis o ␤-
galac osidase-speci ic ac i i y o he co E2-lacZ usion in he WT ( ed
line) and he co E2 (blue line) s ains du ing de elopmen wi hou me -
als (solid lines) and wi h 7.5 ␮M cadmium (dashed lines). Expe imen s
we e ca ied ou in iplica e, and e o ba s indica e s anda d de ia ions
o h ee measu emen s. (Band C) Quali a i e analysis o ␤-galac osidase
ac i i y o bo h s ains on CF aga pla es wi h X-gal and no me als (B), and
pla es supplemen ed wi h X-gal and 7.5 ␮M cadmium (C). Ba ep esen s
0.5 cm.
s ain (Figu e 5). These da a all suppo he conclusion ha
he exp ession o co E2 is no au o- egula ed.
MXAN 5262 does no unc ion as an an i-sigma ac o o
Co E2. E en hough i has been pos ula ed ha he ac-
i i y o Co E-like sigma ac o s is only egula ed by hei
own CRD (13,14), i has been p oposed ha a mem-
b ane lipop o ein encoded by a gene co-exp essed wi h
co E2 migh unc ion as an an i-sigma ac o o Co E2
(12). I he cogna e an i-sigma ac o o an ECF sigma
ac o is knocked ou , genes egula ed by such an ECF
sigma ac o would be exp essed e en in he absence o
he s imulus (14). In o de o es whe he MXAN 5262
ac s as he cogna e an i-sigma ac o o Co E2, a new
Figu e 6. The lipop o ein encoded by he gene MXAN 5262 does no
unc ion as he an i-sigma ac o o Co E2. The WT and 5262 s ains
ha bo ing he 5265-lacZ usion we e spo ed on o CTT aga pla es wi hou
me als (A) and con aining 0.1 mM cadmium (B). Fo quali a i e assays,
pla es also con ain X-gal o moni o ␤-galac osidase ac i i y. (C) Quan i-
a i e analysis o he cadmium-dependen exp ession o he Co E2 egu-
la ed gene MXAN 5265 in he WT and 5262 gene ic backg ounds. The
WT s ain (blue lines) and he 5262 mu an ( ed lines) we e incuba ed on
CTT aga pla es wi hou me als (dashed lines) and con aining 0.1 mM cad-
mium (con inuous lines). ␤-galac osidase-speci ic ac i i y was de e mined
in cell ex ac s ha es ed a he ime poin s indica ed in he igu e. Values
a e a e ages o h ee expe imen s. E o ba s indica e s anda d de ia ions.
Ba ep esen s 0.5 cm.
in- ame dele ion s ain, JM52IF2 (5262), was ob ained.
Nex , he usion 5265-lacZ was in oduced in o his mu-
an . When ␤-galac osidase ac i i y was de e mined in he
s ain JM52IF2ZY5 (5265-lacZ-5262), bo h quali a i ely
and quan i a i ely, and compa ed wi h ha o he WT cells
ca ying he 5265-lacZ usion, i was obse ed ha dele-
ion o MXAN 5262 does no inc ease Co E2 ac i i y, e en
in he absence o cadmium (Figu e 6A and C), demon-
s a ing ha his memb ane p o ein does no unc ion as a
Downloaded om h ps://academic.oup.com/na /a icle/44/12/5571/2457481 by Uni e sidad de G anada - Biblio eca use on 31 July 2025
Nucleic Acids Resea ch, 2016, Vol. 44, No. 12 5579
cogna e an i-sigma ac o o Co E2. Howe e , dele ion o
he gene MXAN 5262 delays he up egula ion o he gene
MXAN 5265 in he p esence o cadmium (Figu e 6Band
C), indica ing ha his lipop o ein mus somehow pa ici-
pa e in he same signal- ansduc ion pa hway as Co E2.
CRD is essen ial o he ac i i y o all Co E-like ECF
sigma ac o s. As a membe o g oup ECF44, Co E2
con ains a CRD consis ing o 28 esidues (as opposed o
he 38 amino acids o he CRD o Co E). I also con-
ains 6 Cys, bu in a di e en a angemen om ha o
Co E (Figu e 1B). To es whe he he Co E2 CRD is es-
sen ial o Co E2 ac i i y, an in- ame dele ion mu an ,
JM52IF3DCRD (co E2CRD), was gene a ed in which
mos o he CRD was dele ed (Figu e 3A). This mu an
was used o elec opo a e he usion 5265-lacZ o ob ain he
s ain JM52IF3DCRDZY5 (5265-lacZ-co E2CRD). Al-
hough Co E2CRD ha bo s he essen ial ␴2 domain and
␴4.2 egion o ECF sigma ac o s, i emains inac i e e en
in he p esence o 0.1 mM cadmium, as shown by he lack o
␤-galac osidase ac i i y in he s ain JM52IF3DCRDZY5
(Figu e 4A). Since he CRD o Co E is also equi ed o
ac i i y (14), i is plausible o pos ula e ha all he sigma
ac o s o he g oup ECF44 exhibi a simila mechanism o
ac ion.
In iew o he di e ences be ween he Cys a angemen in
he CRD domain o Co E2 and Co E, he me al ac i a ion
by each sigma ac o , and he exp ession p o ile o genes
egula ed by each egula o , i is plausible o expec each
Cys o he CRD o Co E2 o play a di e en ole in Co E2
ac i i y om he one epo ed o Co E (14). To analyze
which Cys a e esponsible o Co E2 ac i i y, all six esidues
we e mu a ed indi idually o Ala by si e-di ec ed mu agen-
esis (Figu e 7A). Each mu a ed co E2 was hen in oduced
in o he JM52IF3ZY5 s ain (5265-lacZ-co E2) oe al-
ua e he exp ession p o ile o he gene MXAN 5265 in he
p esence and absence o cadmium. Mo eo e , he WT co E2
gene was also elec opo a ed in o he s ain JM52IF3ZY5
(s ain JM52SDM00) o allow a eliable compa ison be-
ween exp ession o MXAN 5265 by he WT p o ein and
he six mu an p o eins. Al hough he s ain JM52SDM00
will be conside ed as a WT in hese expe imen s, i should
be emembe ed ha i is no he same WT s ain used in
he p e ious expe imen s, so he esul s ha we may ob-
ain wi h JM52SDM00 will no necessa ily be iden ical o
hose shown in Figu es 2–6. The analysis e ealed ha he
exp ession p o ile in he mu an C183A was nea ly iden-
ical o ha o heWTco E2’ s ain (Figu e 7B), indica -
ing ha his esidue is no c ucial o Co E2 ac i i y. Some
up egula ion by cadmium s ill emained in he C173A and
C181A mu an s. Howe e , he maximum exp ession le els
achie ed in hese wo mu an s we e only abou 15% o ha
o he WT s ain (Figu e 7C). These esul s indica e ha
Cys173 and 181 a e impo an , al hough no essen ial, o
Co E2 ac i a ion by cadmium. In e es ingly, when Cys174
was mu a ed o Ala, he e was a cons i u i e exp ession o
MXAN 5265 in he absence o cadmium, which did no sig-
ni ican ly change in he p esence o his me al (Figu e 7D).
These da a indica e ha Cys174 is in ol ed in bo h ac i a-
ion o Co E2 in he p esence o cadmium and inac i a ion
when he me al is absen . In e es ingly, Cys169 and Cys178
we e e ealed o be essen ial o Co E2 ac i i y, because no
exp ession was obse ed in bo h mu an s e en in he p es-
ence o cadmium (Figu e 7E).
To ule ou he possibili y ha he lack o ac i i y o some
poin -mu a ed p o eins migh be due o ins abili y, all he
mu a ed genes we e cloned unde con ol o he cons i u-
i e oa p omo e and used a he N- e minal egion o
an S ag, as desc ibed in ‘Ma e ials and Me hods’ sec ion.
Plasmids con aining he hyb id genes we e elec opo a ed
in o M. xan hus co E2, and he di e en s ains hus ob-
ained (Supplemen a y Table S1) we e analyzed o in es i-
ga e whe he he mu a ed p o eins a e p oduced and solu-
ble. The esul s ob ained ha e e ealed ha all he Co E2
poin -mu a ed p o eins a e s able (Supplemen a y Figu e
S4).
Cys a angemen in he CRD o me al-dependen sigma ac-
o s de e mines he me al speci ici y. As CRDs ha e been
pos ula ed o be he me al- ecogni ion si e o Co E-like
ECF sigma ac o s (14 and da a shown abo e), di e ences
be ween he CRDs o Co E and Co E2 we e analyzed o
esidues ha migh be esponsible o ecogni ion o coppe
by Co E and o cadmium by Co E2. E en hough he Cys
mo i is well conse ed wi hin he ECF44 g oup o sigma
ac o s, he e is one Cys (Cys174 in Co E2) ha ep esen s a
key di e ence be ween he wo cha ac e ized sigma ac o s
wi hin his g oup. The CRD o Co E lacks his Cys and an
Ala is ound in his posi ion (Figu es 1Band8A). As shown
abo e, cadmium ecogni ion was impai ed in he Co E2
C174A mu an (Figu e 7D), so i was plausible o hink ha
his mu an sigma ac o migh espond o coppe , in he
same way as Co E. To es his possibili y, he Co E2 C174A
mu an was assayed o coppe up egula ion o he gene
MXAN 5265. In e es ingly, he mu an C174A, which does
no espond o cadmium, exhibi s a change in me al speci-
ici y, and he exp ession o MXAN 5265 eaches qui e high
le els in he p esence o coppe (Figu e 8B). When o he
me als we e es ed, i was obse ed ha his mu a ed Co E2
is only ac i a ed by coppe (Supplemen a y Figu e S5A). I
emains o be elucida ed whe he Co E2 C174A esponds o
Cu(I) o Cu(II). Acco ding o hese da a, a change o jus
one esidue in he CRD o Co E2 is su icien o comple ely
shi he me al speci ici y o he sigma ac o om cadmium
o coppe .
To lea n mo e abou he signi icance o he esidue lo-
ca ed in posi ion 174 in Co E2 in me al ecogni ion, Co E
CRD was also mu a ed, in his case o eplace Ala185 wi h
a Cys, in o de o make he Co E CRD mo e simila o
ha o Co E2 (Figu es 1Band8A). As shown in Figu e
8C, he Co E A185C mu an was no signi ican ly impai ed
in e ms o coppe induc ion o cuoB (one o he genes
egula ed by his sigma ac o ). Howe e , up egula ion o
cuoB exp ession in he p esence o cadmium was h ee imes
highe in he mu an A185C han in he WT s ain (compa e
con inuous lines in Figu e 8C), indica ing ha his mu an
likely binds bo h me als. The e ec obse ed wi h cadmium
was no ob ained wi h any o he o he me als es ed (Sup-
plemen a y Figu e S5B). As he mu a ed Co E A185C is
also s able (Supplemen a y Figu e S4B), hese esul s con-
i m ha di e ences in only one esidue in he Cys a ange-
men o he CRD o sigma ac o s o he g oup ECF44 a -
Downloaded om h ps://academic.oup.com/na /a icle/44/12/5571/2457481 by Uni e sidad de G anada - Biblio eca use on 31 July 2025