Redox-dependent stability of the γ-glutamylcysteine synthetase enzyme of Escherichia coli: A novel means of redox regulation

Biochem. J. (2013) 449, 783–794 (Printed in Great Britain) doi:10.1042/BJ20120204

783

Redox-dependent stability of the γ -glutamylcysteine synthetase enzyme

of Escherichia coli: a novel means of redox regulation

Shailesh KUMAR, Neha KASTURIA¹, Amit SHARMA², Manish DATT³and Anand K. BACHHAWAT⁴

Institute of Microbial Technology (CSIR), Sector 39-A, Chandigarh 160 036, India

Glutathione is a thiol-containing tripeptide that plays important

roles in redox-related processes. The ﬁrst step in glutathione

biosynthesis is catalysed by γ -GCS (γ -glutamylcysteine

synthetase). The crystal structure of Escherichia coli γ -GCS has

revealed the presence of a disulﬁde bond. As the disulﬁde-bonding

cysteine residues Cys³⁷²and Cys³⁹⁵are not well conserved among

γ -GCS enzymes in this lineage, we have initiated a biochemical

genetic strategy to investigate the functional importance of these

and other cysteine residues. In a cysteine-free γ -GCS that

was non-functional, suppressor analysis yielded combinations of

cysteine and aromatic residues at the position of the disulﬁde

bond, and one mutant that lacked any cysteine residues. Kinetic

analysis of the wild-type and mutant enzymes revealed that the

disulﬁde bond was not involved in determining the afﬁnity of

the enzyme towards its substrate, but had an important role in

determining the stability of the protein, and its catalytic efﬁciency.

We show that in vivo the γ -GCS enzyme can also exist in a reduced

form and that the mutants lacking the disulﬁde bond show a

decreased half-life. These results demonstrate a novel means of

regulation of γ -GCS by the redox environment that works by an

alteration in its stability.

Key words: degradation, γ -glutamylcysteine synthetase (γ -GCS),

glutathione, redox regulation, stability.

enzyme belongs to lineage I and includes other γ -proteobacteria.

It also includes the γ -GCS domain of the bifunctional gshF

enzymes. Lineage I enzymes are distinct from those of non-plant

eukaryotes (animal/fungal) (lineage II) and from the plant γ -GCS

lineage III that includes Brassica juncea, Arabidopsis thaliana

and some plant-associated bacteria [8].

γ -GCS is also the rate-limiting and regulated step of GSH

biosynthesis, being both transcriptionally and post-translationally

regulated. A post-translational regulation of the γ -GCS enzyme

by the redox environment was ﬁrst shown for the mammalian [9]

and Drosophila [10] enzymes (lineage II) through biochemical

studies. The γ -GCS enzymes from these organisms have been

shown to exist as heterodimers with catalytic and regulatory

subunits. Conserved cysteine residues in the catalytic and

regulatory subunit have been found to be involved in the disulﬁde

bond formation between the two subunits. The disulﬁde bond

that is formed under oxidizing conditions increases the substrate

afﬁnity and decreases the feedback inhibition by glutathione and

thus the enzyme also functions as a redox sensor [10,11].

A study of the regulation by the cellular redox environment was

carried out with the plant γ -GCS enzyme (lineage III), which also

includes some plant-associated bacteria. The crystal structure of

the γ -GCS enzyme from B. juncea [12] and studies with both the

B. juncea and A. thaliana enzymes [12,13] have revealed

the presence of two intramolecular disulﬁde bonds, CC1 and CC2.

CC1 had limited inﬂuence on enzyme activity, but disruption (or

reduction) of the CC2 disulﬁde bond severely affected the enzyme

activities, and was responsible for the enzyme’s redox sensitivity.

Both studies have clearly demonstrated that this disulﬁde bond

INTRODUCTION

Glutathione, the most abundant small molecular mass thiol-

containing compound present in living cells, carries out several

important functions that include a role as the cellular redox buffer,

as an antioxidant, and in the cellular response to different stress

conditions [1].

In prokaryotes, glutathione is not as universally present as in

eukaryotes. Furthermore, unlike eukaryotes, where glutathione

has been found to be an essential metabolite, in prokaryotes

glutathione appears to be dispensable under normal growth

conditions, although it is required in bacterial responses to

different kinds of osmotic, oxidative and metal stresses [2–4].

The biosynthesis of GSH occurs by two sequential ATP-

dependent enzymatic steps. The ﬁrst enzyme γ -GCS (γ -

glutamylcysteine synthetase, gshA) ligates glutamate and cysteine

to yield γ -glutamylcysteine. GS (glutathione synthetase, gshB),

the second enzyme, then catalyses the addition of glycine to

yield glutathione. In eukaryotes these activities are encoded by

two distinct enzymes as indicated above. In some prokaryotes,

such as Escherichia coli and Vibrio cholerae, separate enzymes

exist for these two reactions. However, in some prokaryotes, such

as Streptococcus agalactiae [5], Pasteurella multicoda [6] and

Listeria monocytogenes [7], both of these activities are encoded

by a single bifunctional enzyme (gshF).

The analysis of the γ -GCS enzyme from different organisms

has revealed that the evolution of γ -GCS has occurred by

convergent evolution in three different lineages with no signiﬁcant

sequence similarities between the lineages. The E. coli γ -GCS

Abbreviations used: AMS, 4-acetamido-4^ꢀ-maleimidylstilbene-2,2^ꢀ-disulfonic acid; CF, cysteine-free; DTT, dithiothreitol; DTT_Ox, trans-4,5-dihydroxy-

1,2,dithane; γ-GCS, γ-glutamylcysteine synthetase; IPTG, isopropyl β-^D-thiogalactopyranoside; LB, Luria–Bertani; LDH, lactate dehydrogenase; MM(PEG),

methyl-[poly(ethylene glycol)] maleimide; Ni-NTA, Ni^{2 +}-nitrilotriacetate; PK, pyruvate kinase; SD, synthetic deﬁne; SMG, S-methyl glutathione; TCA,

trichloroacetic acid.

1

Present address: Institute of Genomics and Integrative Biology (CSIR), Mall Road, Delhi 110007, India

Present address: Astbury Centre for Structural Molecular Biology, University of Leeds, Leeds LS2 9JT, U.K.

Present address: Center for Proteomics and Bioinformatics, Case Western Reserve University, Cleveland, OH 44106, U.S.A.

To whom correspondence should be addressed at the present address: Indian Institute of Science Education and Research, Mohali, S.A.S. Nagar,

2

3

4

Punjab 140306, India (email anand@iisermohali.ac.in).

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

784

S. Kumar and others

acts as a redox sensor to modulate the activity of γ -GCS [12,13].

Interestingly, the plant-associated bacteria, that are also part

of the plant lineage and include Xanthomonas campestris and

Agrobacterium tumefaciens, contained the equivalent of the CC2

disulﬁde. However, despite the presence of this disulﬁde bond,

these bacteria were, surprisingly, not found to display similar

redox-sensitive activities on reduction/oxidation of the disulﬁde

bond similar to that observed in plants [14].

The E. coli γ -GCS that belongs to lineage I has been shown to

exist as a monomeric enzyme. The crystal structure of the E. coli

γ -GCS enzyme has revealed one disulﬁde bond between Cys³⁷²

and Cys³⁹⁵[15], but the role of the disulﬁde bond, if any, has not

been investigated. On the one hand, the possibility of the disulﬁde

bond being involved as a redox sensor has seemed unlikely given

the ﬁndings with the plant lineage bacteria. However, on the other

hand, the very existence of a disulﬁde bond in the structure of E.

coli γ -GCS, a cytosolic enzyme, when most cytosolic enzymes

rarely have disulﬁde bonds, has suggested that it might still play

an important role. The present study addresses the need, if any, of

the disulﬁde-bonded cysteine residues Cys³⁷²and Cys³⁹⁵, as well

as other cysteine residues in E. coli γ -GCS using a genetic and

biochemical approach. Our ﬁndings reveal a novel means to redox

regulation that is mediated through stability of the reduced form

of γ -GCS, a key enzyme involved in glutathione homoeostasis.

as required. Yeast transformation was carried out by the lithium

acetate method [16].

Creation of CF (cysteine-free) E. coli gshA mutants

The E. coli gshA gene was ampliﬁed by PCR from genomic DNA

and cloned downstream of the single-copy yeast expression vector

pTEF416 using the restriction enzyme sites BamHI and XhoI. The

E. coli gshA open reading frame with surface cysteine residues

at positions 106, 164, 205 and 223 mutated to serine cloned into

pKGC20 vector was a gift from Dr Takao Hibi (Fukui Prefectural

University, Fukui, Japan). The fragment containing 4CF to

serine mutations (EcoRI–SnaBI) was excised from pKGC20 and

subcloned into E. coli gshA in pTEF416.

Site-directed mutagenesis was used to mutate the remaining

native cysteine residues at positions 357, 372, 395, 433

and 439 sequentially to serine by Splice Overlap Extension

PCR by using the 4CF E. coli gshA gene as a template

and pairs of complementary oligonucleotides (Supplementary

Table S1 at http://www.biochemj.org/bj/449/bj4490783add.htm).

Oligonucleotides were designed to introduce a single nucleotide

change resulting in an amino acid substitution from cysteine to

serine. Ampliﬁed plasmid DNA of the resulting mutants was

isolated and sequenced to conﬁrm the presence of the desired

nucleotide changes and to rule out any undesired mutations

introduced during the mutagenic procedure. In the process of

generating a 9CF gshA, a series of CF mutants bearing different

numbers and combinations of cysteine-to-serine mutations were

generated (Table 1).

MATERIALS AND METHODS

Chemicals and reagents

All of the chemicals used in the present study were either of analyt-

ical or molecular biology grade and were obtained from commer-

cial sources. Media components were purchased from Difco.

Oligonucleotides were purchased from Sigma. Restriction

enzymes, Vent DNA polymerase and other DNA-modifying en-

zymes were obtained from New England Biolabs. The enzymes

LDH (lactate dehydrogenase) and PK (pyruvate kinase) were

procured from Roche. The DNA sequencing kit (ABI PRISM 310

XL with dye termination cycle sequencing ready reaction kit) was

obtained from PerkinElmer. Gel extraction kits, plasmid miniprep

columns and the Ni-NTA (Ni^{2 +}-nitrilotriacetate) agarose resin

were obtained from QIAGEN. His-tagged mouse monoclonal

antibody was bought from Cell Signaling Technology and ECL

(enhanced chemiluminescence) Plus Western Blotting detection

reagents were purchased from Amersham Biosciences. The

radiolabelled ³⁵S-Eligmix (70% L-[³⁵S]methionine and 30% L-

[³⁵S]cysteine) was procured from BRIT. Other reagents were

from Sigma–Aldrich, HiMedia, Merck or USB Corporation. The

monobromobimane AMS (4-acetamido-4^ꢀ-maleimidylstilbene-

2,2^ꢀ-disulfonic acid) and MM(PEG) {methyl-[poly(ethylene

glycol)] maleimide} were from Invitrogen BioServices India.

Hydroxylamine mutagenesis

The in vitro mutagenesis of the 9CF gshA clone was performed

using hydroxylamine as described previously [17] with some

modiﬁcations [18].

Cloning, expression and puriﬁcation of E. coli wild-type and

mutant gshA

For protein expression studies, the E. coli and mutant gshA genes

were His-tagged and expressed in the pET23a vector. For the His-

tagging of wild-type and mutants, 1.52 kb of E. coli gshA gene

was PCR ampliﬁed with SmaIFw and EcGshA_CHis _BamHIRev

primers. The ampliﬁed 1.52 kb gene was digested with SmaI and

BamHI restriction enzymes and cloned into the NdeI (that was

blunted) and BamHI sites in the pET23a vector. The in-frame

fusions were conﬁrmed by sequencing.

All proteins were expressed as C-terminal His₆-tagged fusion

proteins in E. coli BL21 or E. coli Origami strains in LB (Luria–

Bertani) broth with 100 μg/ml ampicillin and at a temperature

of 37^◦C. At a D₆₀₀of ∼0.6, the cultures were induced with

1 mM IPTG (isopropyl β-D-thiogalactopyranoside). Cells were

harvested after 5 h by centrifugation at 2500 g for 5 min at 4^◦C

following which they were lysed by sonication in the presence of

50 mM Tris/HCl, pH 8.0, 300 mM NaCl and 1 mM PMSF. The

lysate was centrifuged at 15000 g for 30 min at 4^◦C to remove

the insoluble debris. Cleared lysate was then loaded on to an

equilibrated Ni-NTA column and subsequently washed with wash

buffer (50 mM Tris/HCl, pH 8.0, 300 mM NaCl, 1mM PMSF and

20 mM imidazole). Elution was performed in the presence of

elution buffer carrying 250 mM imidazole and the samples were

analysed on SDS/PAGE (10% gel). Protein quantiﬁcation was

Strains, media and growth conditions

The E. coli strain DH5α was used as a cloning host. For protein

expression, either E. coli BL21 (DE3) or E. coli Origami was used.

The Saccharomyces cerevisiae strain ABC1195 (MATα his3ꢀ1

leu2ꢀ-0 lys2ꢀ-0 ura3ꢀ-0 gsh1ꢀ:: LEU2) was used for the

genetic screens and for complementation assays. The S. cerevisiae

strain was maintained on YPD [1% (w/v) yeast extract/2%

(w/v) peptone/2% (w/v) glucose] medium. SD (synthetic deﬁne)

minimal medium contained yeast nitrogen base, ammonium

sulfate and dextrose, and was supplemented with histidine, leucine

and lysine at concentrations of 80 mg/l. Glutathione was added

performed by measuring absorbance at A₂₈₀

.

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

Redox-dependent stability of γ -GCS

785

Table 1 In vivo and in vitro functional status of different mutants

+++, yes; – , no; N.D., not determined; SDM, obtained by site-directed mutagenesis.

Strain

In vivo complementation

In vitro activity

Derived from

Wild-type GshA

2CF (C433S,C439S)

+++

–

+++

N.D.

SDM

9CF suppressor

SDM

9CF suppressor

SDM

9CF suppressor

SDM

9CF suppressor

Subcloning of R374QV375F

Subcloning of R347QV375F

SDM

4CF (C106S,C164S,C205S,C223S)

6CF (C106S,C164S,C205S,C223S,C433S,C439S)

7CF (C106S,C164S,C205S,C223S,C357S,C433S,C439S)

8CF (C106S,C164S,C205S,C223S,C357S,C372S,C433S,C439S)

9CF (C106S,C164S,C205S,C223S,C357S,C372S,C395S,C433S,C439S)

9CF+S372F+S395C

N.D.

+++

N.D.

–

+++

9CF+S372F+C395S

–

N.D.

9CF+S372C+S395W

+++

–

+++

N.D.

9CF+C372S+S395W

9CF+S372C+S395Y

+++

9CF+C372S+S395Y

–

N.D.

9CF+R374Q+V375F

+++

–

+++

–

+++

N.D.

9CF+V375F

9CF+R374Q

9CF+S372W+S395C

F357 P358 C372 W395

F357 P358 C372 W395S

F357 P358 C372S W395S

–

to a D₆₀₀= 0.2. Serial dilutions were prepared and 10 μl was

Multiple sequence analysis and phylogenetic analysis

spotted on to SD medium with or without glutathione.

The E. coli γ -GCS protein sequence was used as the query se-

quence to retrieve the γ -GCS homologues of different organisms

of lineage I (Supplementary Table S2 at http://www.biochemj.

org/bj/449/bj4490783add.htm). The multiple sequence alignment

of the protein sequences was generated using ClustalW

(http://www.ebi.ac.uk/Tools/msa/clustalw2/) using default para-

meters and the MEGA 3.1 software was used to construct the

phylogenetic tree using the neighbour-joining method.

CD spectroscopy

Thermal denaturation studies were performed by monitoring

the change in signal at 222 nm using CD on a Jasco

810 spectropolarimeter. Puriﬁed protein at a concentration of

0.15 mg/ml, in 50 mM Tris/HCl, pH 8.0, 20 mM NaCl and

0.5 M guanidinium chloride was used for these investigations,

with a temperature ranging from 25^◦C to 90^◦C. Data obtained

were plotted as a function of mean residue ellipticity against

temperature. Spectra were also collected for the wild-type and

mutant proteins with a wavelength range 205–255 nm in the

presence of buffer (50 mM Tris/HCl, pH 8.0, 20 mM NaCl and

0.5 M guanidinium chloride) to estimate the secondary structural

content of the samples. The graph was plotted using Origin

software and the data was smoothed by point 5 using the Savitzky–

Golay algorithm.

γ -GCS activity assay

The activity of E. coli γ -GCS was assayed by measuring the rate of

ADP formation using a coupled assay that involved PK and LDH

[19]. A standard reaction mixture of 800 μl contained 150 mM

Tris/HCl, pH 8.0, 150 mM KCl, 40 mM MgCl_2,2 mM sodium

phosphoenol pyruvate, 4 mM sodium ATP, 20 mM sodium L-

glutamate, 15 mM L-cysteine, 1.6 units of PK (rabbit muscle),

4 units of LDH (rabbit muscle type II) and 0.24 mM NADH.

The reaction was initiated by adding 1 μg of protein. The rate of

decrease in NADH concentration was monitored at 340 nm using a

spectrophotometer (UV-1800, Shimadzu, UV spectrophotometer)

for 5 min. The reaction was carried out at 25^◦C.

For the determination of the K_mvalue for L-glutamate, L-

cysteine was added at 15 mM and ATP at 4 mM. For determining

the K_mvalue for cysteine, L-glutamate was added at 20 mM and

ATP at 4 mM whereas the K_mdetermination for ATP was carried

out at 20 mM L-glutamate and 15 mM L-cysteine. The kinetic

parameters were calculated using GraphPad Prism software.

Identiﬁcation of the reduced form of γ -GCS protein in vivo

The wild-type (or mutant) gshA plasmid was transformed into

E. coli BL21, the secondary cultures were induced at D₆₀₀= 0.4

with 1 mM IPTG and cells were harvested 90 min post-induction.

Prior to harvesting, the culture was treated with 100% TCA

(trichloroacetic acid) to a ﬁnal concentration of 20% to trap the

in vivo thiols in their original status and to prevent aerial oxidation.

Protein extraction was performed following the glass bead lysis

method and the supernatants were centrifuged at 15000 g for

10 min at 4^◦C. The pellets obtained were washed three times

with chilled acetone and dried in the speedvac concentrator.

The dried pellet was dissolved in Hepes (50 mM Hepes, pH 7.0,

and 2.5% SDS) buffer and 5 mM MM(PEG) (alkylation agent)

was added. For the AMS treatment, the pellet was dissolved

in TES (100 mM Tris/HCl, pH 8.0, 10 mM EDTA and 1%

SDS) buffer and 30 mM AMS (alkylation agent) was added.

The MM(PEG)-treated samples were incubated in the dark at

25^◦C whereas AMS-treated samples were incubated at 37^◦C

for 2 h with 1400 rev./min agitation in a thermo mixer. Puriﬁed

protein was precipitated by adding an equal volume of 20%

Functionality assay by in vivo complementation in S. cerevisiae

The yeast strain ABC 1195 (Scgsh1ꢀ) was transformed with

wild-type or mutant gshA gene downstream of the TEF promoter

in pTEF416. Transformants were grown in minimal media

containing 0.2 mM glutathione and other amino acid supplements.

The overnight culture was used to reinoculate the secondary

culture and incubated until D₆₀₀= 0.6. Equal numbers of cells

were harvested, washed with water and suspended in sterile water

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

786

S. Kumar and others

TCA. Precipitated protein was washed twice with ice-cold acetone

and air-dried. The protein sample was then treated with 5 mM

MM(PEG) as described above. After 2 h the samples were

centrifuged at 15000 g for 15 min at room temperature (25^◦C)

and the supernatant was collected. The protein concentration was

determined by the Bradford method and 1 μg of protein was mixed

with non-reducing loading dye and boiled for 5 min. SDS/PAGE

(14% gel) was run followed by Western blotting [20].

Pulse–chase experiments

The E. coli BL21 strains were transformed with plasmids carrying

either wild-type or 9CF + R374Q + V375F gshA gene. The

primary culture was set up in M9 minimal medium supplemented

with all 20 amino acids except methionine and cysteine and grown

at 37^◦C overnight. The primary culture was used to seed the

secondary culture at D₆₀₀= 0.1 in M9 minimal medium and

allowed to grow until D₆₀₀= 0.5, when protein expression was

induced with 1 mM IPTG. The culture was induced for 2 h at

37^◦C followed by pulse-labelling by addition of 40 μCi/ml ³⁵S-

Eligmix (70% ^L-[³⁵S]methionine and 30% L-[³⁵S]cysteine) and

the culture was incubated at 37^◦C. After 5 min of pulse-

labelling, the medium was supplemented with 5 mg/ml of

non-labelled L-methionine and L-cysteine mixture (7:3 ratio).

A 0.2 ml aliquot of the culture was taken at 0, 2, 4 and 7 h

intervals and mixed with 50 μl of the protease inhibitor PMSF

(1 mM). Cells were lysed in 90 μl of 1× SDS-reducing dye and

15 μl of lysates were electrophoresed on reducing SDS/PAGE

(10% gel). The gel was exposed on a phosphoimager screen and

autoradiographed and the band corresponding to 58.2 kDa was

observed in induced cells. The band intensity was analysed by

using scion software. This protocol was modiﬁed from Parsell

and Sauer [21].

Figure 1 Phylogenetic tree analysis of lineage I

The E. coli γ -GCS protein sequences were used to retrieve the γ -GCS homologues

of different organisms of lineage I. The MEGA 3.1 software was used to construct the

phylogenetic tree. The conservation patterns of cysteine residues are shown. The representative

members used in multiple sequence alignment are listed in Supplementary Table S2 (at

http://www.biochemj.org/bj/449/bj4490783add.htm).

Midpoint redox potential of E. coli γ -GCS

The midpoint redox potential of E. coli γ -GCS was achieved

by titrating the protein with deﬁned ratios of DTT_Ox(trans-

4,5-dihydroxy-1,2,dithane) and DTT (dithiothreitol). The protein

(50 μg) was equilibrated in buffer containing 100 mM Hepes

(pH 7.0) and different ratios of DTT (80–0.01 DTT_Ox/DTT) with

a total concentration of 10 mM for 3 h at 25^◦C to achieve

the equilibrium. The experiment was carried out and midpoint

potential calculations were made as described previously [22,23].

conservation pattern of these cysteine residues. We observed

that, among the nine cysteine residues seen in E. coli γ -GCS,

three of the residues, Cys¹⁰⁶, Cys²⁰⁵and Cys³⁵⁷were conserved

in the majority of the orthologues, barring the bifunctional gshF

proteins that fell into a completely distinct cluster (Figure 1).

However, surprisingly, the two cysteine residues involved in

making the disulﬁde bond, Cys³⁷²and Cys³⁹⁵, were conserved

(along with Cys⁴³³) in only a small subset that primarily included

the enterobacteria (Supplementary Figure S2 at http://www.

biochemj.org/bj/449/bj4490783add.htm). The least-conserved

cysteine residues were Cys¹⁶⁴, Cys²³³and Cys⁴³⁹. In the analogous

enzymes of the other lineages, conservation of the cysteine

residues correlated with a functional role in disulﬁde-bonding

participation. The relative lack of conservation seen in the

disulﬁde-bonded cysteine residues in the E. coli γ -GCS lineages

was surprising.

To understand these aspects and also evaluate the role

of the different cysteine residues, we created a series of

cysteine-to-serine mutants, where eventually all of the cysteine

residues were mutated to serine. Functionality was assessed

by complementation of the S. cerevisaie gsh1Δ phenotype in

medium lacking glutathione. Unlike E. coli gshA deletions which

are viable even in the absence of glutathione, S. cerevisiae

gsh1Δ deletions are strict glutathione auxotrophs. As E. coli γ -

GCS can complement S. cerevisiae gsh1Δ, it enabled a simple

complementation-based growth assay for the mutant γ -GCS

RESULTS

The conservation patterns and functional importance of cysteine

residues in E. coli γ -GCS

Conservation patterns and co-evolution of pairs of cysteine

residues often reveals important insights on the functionality of

cysteine residues in proteins. To examine the extent to which the

nine cysteine residues present in E. coli γ -GCS are conserved,

we carried out multiple sequence alignments with the γ -GCS

enzymes of lineage I to which the E. coli γ -GCS belongs. This also

included the γ -GCS domain of the bifunctional enzymes (gshF)

observed in several bacteria [6]. To ensure that these putative

orthologues from other organisms might indeed be functioning

as γ -GCSs, we ﬁrst examined whether the active site residues

identiﬁed from the crystal structures were conserved in these

proteins. All of the active site residues were conserved in the

selected proteins, suggesting that they might be true orthologues

(Supplementary Figure S1 at http://www.biochemj.org/bj/449/

bj4490783add.htm). Even the bifunctional gshF proteins

contained these conserved residues. We then examined the

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

Redox-dependent stability of γ -GCS

787

Figure 2 Complementation analysis of wild-type and CF mutants of E. coli γ -GCS

S. cerevisiae strain ABC1195 (MATα his3ꢀ1 leu2Δ-0 lys2ꢀ-0 ura3ꢀ-0 gsh1Δ:: LEU2) was transformed with plasmids bearing genes for the wild-type gshA or the different CF gshA mutants

cloned under the TEF promoter. The transformants were serially diluted and spotted on to SD medium containing/lacking GSH as the sole source of organic sulfur. The vector pTEF416 was used as a

control.

in which both the Ser³⁷²and Ser³⁹⁵residues (originally cysteine)

had reverted back to cysteine. As these residues have been

shown to form a disulﬁde bond, the simultaneous appearance

of both these cysteine residues in the majority of suppressors

(nine suppressors among 14) strongly suggested that the disulﬁde

bond is a very important determinant for functionality. However,

in addition to these mutants, we interestingly also obtained the

double mutants 9CF + S372C + S395W, 9CF + S372F + S395C

and 9CF + S372C + S395Y. These mutants had a pattern where

only one of either Cys³⁷²or Cys³⁹⁵was restored as cysteine,

whereas the other change, although it also involved the

corresponding 372 or 395 residues, contained an aromatic residue

at the position (Figure 3A). Although some of the mutants

contained mutations in addition to these mutations (Table 1),

subcloning experiments that were done to separate the mutations

revealed that they were not required for functionality (results

not shown). With only a single cysteine residue in the mutants,

they lacked the ability to form an intramolecular disulﬁde bond

observed in the E. coli γ -GCS structure, clearly indicating that an

intramolecular disulﬁde mutant was not essential for activity.

In addition to these mutants that contained either a single

or double cysteine residue at one or other of these positions,

we were also able to isolate a single mutant that lacked

any cysteine residues. This mutant contained two changes,

R374Q and V375F, both changes proximal to the disulﬁde

bond (Supplementary Figure S3 at http://www.biochemj.org/bj/

449/bj4490783add.htm). To determine whether both changes

were needed for functionality, we separately reconstructed these

mutants, and observed that it was only the V375F mutation that

was responsible for the restoration of functionality (Figure 3B).

To investigate the cysteine-aromatic suppressors, and how

important the relative positioning of the aromatic and cysteine

residues were in this mutant, we created a mutant in which

the 9CF + S372C + S395W mutation position was swapped. The

9CF + S372W + S395C mutation so created was found to be

functionally similar to 9CF + S372C + S395W, suggesting that

although an aromatic and cysteine residue were important in

these positions, swapping the positioning was also acceptable

for function. This suggested that the nature of the interaction

rather than the precise positioning might be important

(Supplementary Figure S4A at http://www.biochemj.org/bj/449/

bj4490783add.htm). However, mutations at both of these

positions were necessary for the activity since a serine-aromatic

clones on plates lacking glutathione. We observed that the 2CF,

4CF, 6CF and even the 7CF mutants were all functional in this

assay (Table 1 and Figure 2). However, on introduction of further

cysteine-to-serine mutations by mutating the Cys³⁷²and then the

Cys³⁹⁵residues to yield the 8CF and 9CF mutants, activity was

immediately lost. The 8CF and 9CF mutants lack the disulﬁde

bond, whereas the 7CF mutant retains the cysteine residues that

form the disulﬁde bond. Thus, in spite of the fact that Cys³⁷²

and Cys³⁹⁵were not the most conserved, they still appeared to be

functionally very important on the basis of this analysis.

Isolation of intragenic functional suppressors of a non-functional

CF E. coli γ -GCS

The complete loss in activity seen in the 8CF and 9CF mutants

suggested that the disulﬁde bond was likely to be critical for

activity. We took an independent approach to conﬁrm if these

cysteine residues were essential for activity, or if we might identify

a functional molecule devoid of cysteine residues. This approach

exploited a genetic selection strategy in S. cerevisiae gsh1Δ where

functional suppressors could be selected.

We carried out in vitro hydroxylamine mutagenesis of the

CF E. coli gshA (9CF) gene expressed from a TEF promoter,

transformed this into an S. cerevisiae gsh1Δ strain, and selected

for functional clones on minimal medium lacking glutathione.

A total of 53 suppressors were isolated from two independent

mutagenic events. Among these mutants, 22 transformants

showing different levels of complementation were selected for

further analysis. Among the 22 selected mutated transformants,

only 14 behaved true on secondary screening and were chosen

for further investigation. The plasmids were isolated from those

yeasts, transformed in E. coli and, after puriﬁcation through E.

coli, reconﬁrmed by transformation in the S. cerevisiae gsh1Δ

strain. All 14 clones were able to functionally complement the S.

cerevisiae gsh1Δ and were subjected to sequencing. To eliminate

the possibility of mutations in the promoter being responsible

for the phenotypes, the coding regions of the functional mutants

were also excised and subcloned under a TEF promoter of a

non-mutated plasmid. In cases where more than one mutational

change was detected, we further subjected these to subcloning

to determine the minimal residues that were permitting the

molecule to regain activity. These data are summarized in

Table 1. The majority of the mutants carried double mutants

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

788

S. Kumar and others

Figure 3 Functional complementation displayed by suppressor of 9CF γ -GCS

E. coli 9CF γ -GCS suppressors were transformed in S. cerevisiae strain ABC1195 (MATα his3ꢀ1 leu2ꢀ-0 lys2ꢀ-0 ura3ꢀ-0 gsh1ꢀ:: LEU2) and spotted on to SD medium containing/lacking

glutathione as the sole source of organic sulfur. The vector pTEF416 was used as a control. (A) Functional complementation of cysteine aromatic suppressor mutants. (B) The complementation of

Scgsh1Δ by CF suppressors 9CF + R374Q + V375F, 9CF + R374Q + V375 and 9CF + R374 + V375F.

combination that we created was not functional (Supplementary

Figure S4B).

GSH the activity of wild-type and 7CF was reduced by 25%,

whereas at 10 mM GSH a 45% reduction in enzyme activity

was observed. However, the mutants 9CF + S372C + S395W and

9CF + R374 + V375F that lacked disulﬁde-bonding ability, did

not show any inhibition with glutathione (Figure 4A). As the

inhibitory effect of glutathione could be either a consequence

of its reductant property or as a consequence of its ability to

cause product inhibition, we investigated the effect of DTT on the

activity of these mutants. Interestingly, we observed inhibition by

DTT that paralleled the inhibitory effect of GSH in the wild-

type and 7CF mutant but caused no inhibition in activity in

the 9CF + S372C + S395W and 9CF + R374 + V375F mutants

(Figure 4B). This suggested that the inhibition by glutathione

was not by virtue of its being a ﬁnal product of the reaction or

of allosteric regulation, but as a consequence of the reductant

property of glutathione. To further conﬁrm this aspect we used

the glutathione analogue SMG (S-methyl glutathione). In this

analogue, the free thiol group is blocked by the methyl group.

The properties of the compound as a product analogue are largely

unaffected owing to the presence of only a very small methyl

group, but as a consequence of a blockage of the thiol group

it lacks the reductant property of glutathione. The wild-type

Wild-type and 7CF E. coli γ -GCS capable of forming a disulﬁde

bond show reduced activity in the presence of DTT/GSH, but the

mutants 9CF + S372C + S395W and 9CF + R374Q + V375F are

unaffected

As the in vivo complementation assay in S. cerevisiae was

a qualitative assay and did not give an indication of other

properties of the protein that might be affected by the mutations,

we resorted to in vitro puriﬁcation and characterization of

these mutant proteins from E. coli. The wild-type, 7CF (Cys³⁷²

and Cys³⁹⁵), 9CF + S372C + S395W, 9CF + S372C + S395Y and

9CF + R374Q + V375F mutants were His-tagged at their C-

terminal, and puriﬁed from E. coli using Ni-NTA afﬁnity chro-

matography. The proteins were puriﬁed to >90% homogeneity

(Supplementary Figure S5 at http://www.biochemj.org/bj/449/

bj4490783add.htm).

Previous studies with E. coli gshA have revealed an inhibition

by GSH [24]. We reinvestigated this inhibition by GSH with

the wild-type and mutant proteins and observed that at 2 mM

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

Redox-dependent stability of γ -GCS

789

at a concentration of 15 mM, was not expected to affect the

redox status of the protein since no prior pre-incubation was

involved and the kinetics were determined based on initial

velocities. The wild-type and 7CF mutant were found to have an

almost similar K_mvalue for the different substrates. Surprisingly,

the mutants 9CF + S372C + S395Y, 9CF + S372C + S395W and

9CF + R374Q + V375F also had very similar K_mvalues for

glutamate, cysteine and ATP. These results indicate that the

presence or absence of a disulﬁde bond in E. coli γ -GCS does not

determine the afﬁnity of the enzyme for any of its substrates to a

signiﬁcant extent (Table 2). However, we observed a lower V_max

of the mutants. Comparisons of k_catand k_cat/K_m(i.e. the turnover

number and catalytic efﬁciency) of wild-type and mutant revealed

that some mutants have a reduced turnover number and catalytic

efﬁciency. The catalytic efﬁciency of 9CF + S374C + S375Y is

reduced by 2-fold for cysteine and ATP. To conﬁrm that this is the

result of the absence of a disulﬁde bond, we incubated the wild-

type and 7CF protein with 2 mM DTT for 30 min at 4^◦C followed

by the assay of these enzymes for their kinetic parameters for all

three substrates. We observed that reduction of the disulﬁde bond

in the wild-type and 7CF γ -GCS protein had mild effects on the

K_mvalue of the enzyme for all substrates, but signiﬁcantly affected

the catalytic efﬁciency (k_cat/K_m) (Table 3). We observed that the

reduced wild-type and 7CF (DTT-treated) had a lower k_cat/K_m

value than the untreated enzyme. These observations on the wild-

type and mutant enzymes demonstrate that, although the absence

of the disulﬁde bond in the mutant does not affect the substrate

afﬁnities of enzyme, the mutant enzymes displayed reduced

turnover number and catalytic efﬁciency. More importantly, the

kinetic behaviour of the reduced protein was found to be very

similar to the pattern shown by the 9CF + R374Q + V375F and

9CF + S372C + S395Y mutants.

Decreased in vitro stability in the mutants lacking the capability to

form a disulﬁde bond

Figure 4 Inhibition study of E. coli γ -GCS wild-type and mutant enzymes

by GSH, DTT and SMG

During the initial characterization of the different wild-type and

mutant puriﬁed proteins, we observed a differential thermolability

of the different proteins. This prompted us to carry out more

detailed investigations on the stability assays of representative

mutants from each class. In this assay we pre-incubated the

wild-type and mutant proteins at 37^◦C, 25^◦C and 4^◦C for

1 h followed by an activity assay at room temperature. The

wild-type had almost similar activity at all three different pre-

incubated temperatures, whereas the 9CF + S372C + S395W and

9CF + R374Q + V375F mutants had signiﬁcantly lower activity.

9CF + S372C + S395W lost approximately 30% activity by pre-

incubation at 37^◦C, whereas the 9CF + R374Q + V375F mutant

lost 75% of its activity by pre-incubation at 37^◦C compared with

its activity when pre-incubated at 4^◦C (Figure 5). In contrast, the

7CF mutant, which retained the ability to form a disulﬁde bond,

showed almost similar thermal stabilities as the wild-type. To

investigate whether reduction of the disulﬁde bond would reduce

the stability of the wild-type and 7CF proteins, we incubated the

wild-type and 7CF γ -GCS protein with 5 mM DTT and GSH

for 1 h at 37^◦C, 25^◦C and 4^◦C followed by measurement of

enzyme activity at room temperature (25^◦C). The stability pattern

of these DTT-treated proteins was very similar to that obtained

with mutants lacking the ability to form a disulﬁde bond. The wild-

type and 7CF proteins that were pre-incubated at 4^◦C showed

maximum activity, whereas activity decreased with the increase

in pre-incubation temperature (Figure 6). The activity of wild-

type and 7CF in DTT or GSH showed no signiﬁcant change in

activity when pre-incubated at 25^◦C, but a drastic loss in activity

The puriﬁed wild-type and mutant proteins were treated with different concentrations of GSH

(A), DTT (B) and SMG (C) and incubated for 30 min at 4 C. The enzyme activity of the

◦

treated protein was assayed as described in the Materials and methods section. The saturated

substrate concentration was used. The enzyme activity of wild-type (WT, ᭹), 7CF (᭿),

9CF + R374Q + V375F (᭡) and 9CF + S372C + S395W (᭢) were considered as 100% at

0 mM GSH, DTT and SMG respectively and the activity at other concentrations was calculated

+

relative to that. The experiment was carried out in triplicate and means S.D. are shown.

−

γ -GCS enzyme was incubated with different concentrations of

SMG (0–10 mM) for 30 min and assayed for enzyme activity

(Figure 4C). We observed that SMG was unable to inhibit the

enzyme activity even at 10 mM concentrations, further conﬁrming

that the glutathione inhibition of E. coli gshA was a consequence

of its ability to reduce the disulﬁde bond.

Kinetic studies of wild-type and mutant γ -GCS enzymes reveal no

role of the disulﬁde bond in determining the afﬁnity of the enzyme

towards its substrates

In the human and plant γ -GCS enzymes, the reduction of

the disulﬁde bond led to an altered afﬁnity of the enzyme to

its substrates. To determine whether the presence or absence

of the disulﬁde bond in E. coli γ -GCS also inﬂuenced

the kinetic parameters towards the substrates, we evaluated the

mutants for their kinetic parameters in respect to all three

substrates, cysteine, glutamate and ATP. Cysteine, although added

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

790

S. Kumar and others

Table 2 Kinetic parameters of wild-type and mutants

+

All reactions were performed as described in the Materials and methods section. The values are presented as means S.D. for three independent experiments. The statistical signiﬁcance of the

differences was with respect to the wild-type protein, *P ꢀ 0.05.

−

Substrate

Parameters

Wild-type

7CF

9CF + R374Q + V375F

9CF + S372C + S395Y

9CF + S372C + S395W

+

Glutamic acid

K_m(mM)

3.9 0.4

32.3 2.6

4.0 1.6

34.9 11.6

4.0 0.5

30.4 5.7

3.2 0.4

22.7 9.7

4.1 1.8

31.2 2.9

−

+

V_max(nmol/min per μg)

−

k_cat(s^{− 1}

)

31.4 2.5

34.0 11.3

29.6 5.5

22.1 9.4

30.4 2.8

+

−

k_cat/K_m(s^{− 1}·mM^{− 1}

)

8.1

8.5

7.4

6.9

7.4

+

Cysteine

ATP

K_m(mM)

0.10 0.01

0.11 0.04

0.14 0.02*

0.11 0.05

0.16 0.08

−

+

V_max(nmol/min per μg)

47.9 15.7

40.8 20.9

30.0 10.0

22.5 5.8*

32.7 12.2

−

k_cat(s^{− 1}

)

46.6 15.3

39.7 20.3

29.2 9.8

21.9 5.7

31.8 11.8

+

−

k_cat/K_m(s^{− 1}·mM^{− 1}

)

466.0

396.9

208.6

199.1

198.8

+

K_m(mM)

0.14 0.04

0.10 0.04

0.19 0.03

0.53 0.15*

0.24 0.23

−

+

V_max(nmol/min per μg)

48.2 12.9

40.8 6.5

27.9 5.7

25.4 4.1*

31.3 1.2

−

k_cat(s^{− 1}

)

46.9 12.6

39.7 6.3

27.2 5.5

24.7 4.0*

30.5 1.2

+

−

k_cat/K_m(s^{− 1}·mM^{− 1}

)

335.0

397.0

142.9

46.6*

127.1

Table 3 Kinetic parameters of wild-type and 7CF in the presence of DTT

All reactions were performed as described in the Materials and methods section. The values are

+

presented as means S.D. for three independent experiments. The statistical signiﬁcance of

−

the differences was with respect to the unreduced wild-type protein, *P ꢀ 0.05.

Substrate

Parameters

Wild-type (reduced)

7CF (reduced)

+

Glutamic acid

K_m(mM)

5.1 0.7*

22.8 4.9*

5.3 2.0

24.4 7.7

−

+

V_max(nmol/min per μg)

−

k_cat(s^{− 1}

)

22.1 4.8*

23.8 7.5

+

−

k_cat/K_m(s^{− 1}·mM^{− 1}

)

4.3*

4.5*

+

Cysteine

ATP

K_m(mM)

0.14 0.01*

0.10 0.02

−

+

V_max(nmol/min per μg)

24.9 8.9

22.8 7.4

−

k_cat(s^{− 1}

)

24.2 8.7

22.2 7.2

+

−

k_cat/K_m(s^{− 1}mM^{− 1}

K_m(mM)

)

172.9*

222.0*

+

0.46 0.48

0.25 0.01*

−

+

Figure 6 Stability assay of wild-type and 7CF γ -GCS in the presence of

DTT and GSH

V_max(nmol/min per μg)

29.5 9.0

30.9 5.7

−

k_cat(s^{− 1}

)

28.7 8.8

30.1 5.6

+

−

k_cat/K_m(s^{− 1}·mM^{− 1}

)

62.4

120.4*

The puriﬁed wild-type (WT) and 7CF γ -GCS protein was treated with 5 mM DTT and GSH and

◦

pre-incubated at 37 C, 25 C and 4 C for 1 h followed by assay of activity at room temperature

◦

+

(25 C). The experiment was carried out in triplicate and means S.D. are shown.

−

their secondary α-helix structure compared with the wild-type

(Supplementary Figure S6A at http://www.biochemj.org/bj/449/

bj4490783add.htm). The melting temperature scan of mutants

along with wild-type, however, revealed that although the wild-

type had a T_mat 47.5^◦C, the 9CF + S372C + S395W mutant

showed lesser thermal stability with a T_mat 44^◦C, whereas the

mutant 9CF + R374Q + V375F had a T_mat 42.5^◦C. This was in

contrast with the 7CF mutant that had a T_msimilar to wild-type at

47^◦C (Supplementary Figure S6B). The similar T_mvalues of wild-

type and 7CF mutant strongly implicates these cysteine residues in

disulﬁde bond formation that provided greater thermostability and

higher T_mof unfolding transitions. To conﬁrm this fact, we carried

out T_mscans of the wild-type protein that was reduced with DTT

and found a decreased T_m(T_m= 43^◦C) (Supplementary Figure

S6C).

Figure 5 Effect of temperature on wild-type and mutant activity

The effect of temperature was studied with wild-type (WT) and mutant proteins. The wild-type

◦

and mutant proteins were pre-incubated at 37 C, 25 C and 4 C for 1 h. The enzyme activity of

the wild-type and mutants were assayed at saturated substrate concentration at room temperature

◦

(25 C) as described in the Materials and methods section. The activity was expressed as a

percentage of the highest activity at the pre-incubated temperature of 4 C. The experiment was

carried out in triplicate and means S.D. are shown.

◦

Reduced form of E. coli γ -GCS can be detected in vivo

+

−

The crystal structure of E. coli γ -GCS has shown the presence of

a disulﬁde bond between Cys³⁷²and Cys³⁹⁵[15]. The biochemical

data described above have suggested that if this disulﬁde bond

is reduced under reducing physiological conditions it might have

an altered stability and consequent physiological implications.

It was therefore important to establish whether E. coli γ -GCS

when pre-incubated at 37^◦C. This ﬁnding further suggests that

the disulﬁde bond is very important for stability.

To investigate the thermal instability of the mutant proteins,

we performed CD spectroscopy [25] of wild-type and mutant

proteins. The results revealed that all mutants maintained

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

Redox-dependent stability of γ -GCS

791

biochemj.org/bj/449/bj4490783add.htm). We then subjected the

7CF mutant to MM(PEG) and AMS treatment. The presence of

a shift in mobility of the protein band on MM(PEG) treatment

or AMS treatment would reﬂect the presence of a reduced

cysteine in the 7CF molecule. We observed that the MM(PEG)-

treated samples shifted compared with the un-treated controls.

Surprisingly though, we observed that in the MM(PEG)-treated

samples there were two shifted bands (this was seen reproducibly

in all of the experiments carried out). The presence of two bands,

both of them shifted up, probably reﬂects binding to either one or

two of the cysteine residues. Normally one expects both cysteine

residues to be conjugated to the alkylating reagent. However, it

is possible that steric hindrance owing to the proximity of the

two cysteine residues might be hindering the rate of reaction,

thereby enabling us to detect the singly conjugated and dually

conjugated samples. We also observed a discernable shift on

AMS treatment (Supplementary Figure S8A). We carried out the

MM(PEG) modiﬁcation experiments with puriﬁed wild-type and

mutant proteins (with either zero, one or two cysteine residues)

and observed similar shifts, validating the interpretations with the

crude extracts. One difference seen in the puriﬁed samples of 7CF

was the absence of a double shift that may be complete oxidation

of both cysteine residues in the puriﬁed samples (Supplementary

Figure S8B). These results suggested that the reduced form of

E. coli γ -GCS exists in vivo and can be detected under in vivo

physiological conditions. The presence of only the reduced form

might be a consequence of the overexpression of the γ -GCS

enzyme and might explain the detection of the reduced protein

despite the low midpoint redox potential. As controls we have

included the 9CF + R374Q + V375F mutant with and without

MM(PEG) or AMS treatment and, as expected, no differences in

mobility were observed in this mutant (either in the extracts or

in the puriﬁed controls).

Figure 7 Identiﬁcation of a reduced form of γ -GCS in vivo using MM(PEG)

The wild-type (WT), 7CF and 9CF + R374Q + V375F plasmid harbouring the E. coli strain was

grown in LB broth supplemented with 100 μg/ml ampicillin. The expression was induced at

D₆₀₀= 0.4 with 1 mM IPTG. At 90 min post-induction, the growing culture was treated with

100% TCA to a ﬁnal concentration of 20%. The extracted protein was treated with 5 mM

MM(PEG), an alkylation agent, as described in the Materials and methods section. The reduced

and oxidized proteins were separated by SDS/PAGE (14% gels) under non-reducing conditions.

ThedetectionofproteinwascarriedoutbyWesternblottingusingananti-Hisantibody .M olecular

masses in kDa are shown to the left-hand side.

was found only as an oxidized form in vivo as seen in the

structure, or if an equilibrium might exist between the reduced

and oxidized forms in vivo. To demonstrate whether the disulﬁde

bond can form under physiological conditions, we determined

the midpoint redox potential of the disulﬁde bond by redox

titration. We used the 7CF mutant that contains only two disulﬁde-

bonding cysteine residues for this analysis. We obtained an

approximate midpoint redox potential of E_m= − 330 mV for

the E. coli γ -GCS (Supplementary Figure S7 at http://www.

biochemj.org/bj/449/bj4490783add.htm). The redox potential of

the E. coli cytosol has been estimated in the range − 260 to

− 280 mV [26]. This suggests that the E. coli γ -GCS enzyme

must largely exist in the oxidized form in vivo and the detection

of the disulﬁde bond in the crystal structure is in agreement with

this observation.

To investigate whether a reduced form of γ -GCS could

nevertheless be detected in vivo, we carried out experiments

using MM(PEG) and AMS alkylation to determine whether the

disulﬁde bond of γ -GCS might also exist as a reduced form

in vivo. MM(PEG) is an alkylating agent that can bind to a

free thiol group of cysteine and increase the molecular mass

by 0.71 kDa, whereas AMS increased the molecular mass by

0.5 kDa. The procedure involves the rapid acidiﬁcation of E.

coli cells by TCA that rapidly trap the thiols in the native

in vivo form and helps protein precipitation. To validate the

method, the wild-type E. coli γ -GCS, which has nine cysteine

residues, seven of which in reduced form, was examined

in an MM(PEG) and AMS experiment and as expected a

signiﬁcant shift in mobility of approximately 5–6.5 kDa and

3.5–4.5 kDa respectively was observed on MM(PEG) and AMS

treatment corresponding to the expected number of free cysteine

residues (Figure 7 and Supplementary Figure S8A at http://www.

The mutant enzyme lacking cysteine residues shows a decreased

in vivo half-life

The in vitro studies with wild-type and mutants had clearly shown

the effects of disulﬁde on the stability of protein. We hypothesized

that the absence of a disulﬁde bond could possibly affect the in vivo

half-life of the protein. As it was difﬁcult to achieve completely

reduced or oxidized forms of the protein in vivo, we exploited

the mutants to evaluate this possibility. The in vivo protein

degradation of wild-type and mutant 9CF + R374Q + V375F

was studied by pulse–chase experiments. The wild-type and

9CF + R374Q + V375F clones were grown in M9 minimal

medium supplemented with all amino acids except methionine

and cysteine. Expression was induced for 2 h with 1 mM IPTG

followed by a pulse with ³⁵S-Eligmix. This was followed by a

chase with excess non-labelled methionine and cysteine mixture

at 3 7^◦C. Aliquots were taken out at different time intervals. The

wild-type protein appeared to be very stable even after 7 h of

chase. We observed that the mutant which lacked the disulﬁde

bond was signiﬁcantly more susceptible to degradation, with a

half-life of approximately 4 h in contrast with the wild-type which

was relatively stable during this period (Figure 8). The result

demonstrated that the disulﬁde bond was important for the in vivo

stability of E. coli γ -GCS.

DISCUSSION

The present study provides new insights into how a key redox

enzyme, γ -GCS of E. coli , which largely determines the levels of

the abundant and key redox metabolite glutathione in the cell, is

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

792

S. Kumar and others

Figure 8 Pulse–chase analysis for the degradation of wild-type and mutant 9CF + R374Q + V375F

The wild-type (WT) and mutant 9CF + R374Q + V375 gshA plasmid was transformed in E. coli BL21 strain and grown in M9 minimal medium. The exponential phase cells were induced with 1 mM

IPTG and after 2 h of post-induction the culture was pulse-labelled with 40 μCi/ml ³⁵S-Eligmix. The sample was collected at different time intervals after chase with a non-labelled methionine and

+

cysteine mixture and electrophoresis was carried out as described in the Materials and methods section. The SDS/PAGE represents a single experiment with the means S.D. for three experiments

provided in the bottom panel.

−

itself regulated by the redox status of the cell. In the analogous

enzymes of plants and humans a disulﬁde bond also exists either

between two subunits (humans) or within a subunit (plants) and

is a key regulator of the efﬁciency of this enzyme. In these cases,

the reduction of the disulﬁde bond led to an altered afﬁnity for the

substrates and altered feedback sensitivity leading to decreased

synthesis of GSH and this has a direct impact on the redox state of

the cell. In contrast with this situation, in E. coli we demonstrate

that, although a reduction in the disulﬁde bond of γ -GCS does not

lead to an alteration of the afﬁnity towards the substrates, it

did lead to an alteration of the stability and increased degradation

of the protein. The consequences are decreased catalytic

efﬁciency, reduced synthesis of glutathione and a consequent

effect on the cellular redox status.

The importance of the disulﬁde bond in proteins in imparting

stability to folded proteins is well known. The importance of

folding and stability in conferring resistance to intracellular

proteolytic digestion of proteins is also well known. However,

the fact that these two well-known features of proteins may be

conjoined to yield a mechanism by which the redox status of the

protein would lead to a differential degradation thereby enabling

the cells to alter the GSH level and therefore the redox state, had

not been considered so far, at least to our knowledge. The ﬁndings

of the present study thus demonstrate a new means by which redox

homoeostasis can be maintained.

inhibitory mechanism might parallel what was observed with

the human enzyme where non-allosteric substrate inhibition was

observed. However, the E. coli enzyme appears to be quite

distinct from the mammalian enzyme in several aspects. It is

insensitive to cystamine, and weakly sensitive to iodacetamide

and N-methylmaleimide. It is also insensitive to 4-methylene

glutamate, S-sulfocysteine and S-sulfohomocysteine, to which the

mammalian enzymes are very sensitive [24,27–29]. Although

the mammalian enzyme was also relatively insensitive to SMG,

it was also insensitive to DTT and 2-mercaptoethanol [27]. The

present study, with the E. coli γ -GCS wild-type and mutants, and

use of DTT and the glutathione analogue SMG, however, clearly

demonstrated that the inhibition in E. coli γ -GCS enzyme activity

was a consequence of the reduction of the disulﬁde bond leading

to both altered stability and activity.

In the bifunctional enzyme gshF of S. agalactiae, it is

interesting to note that no allosteric inhibition of the enzyme

by GSH was observed, although inhibition at high concentrations

of GSH was observed with the P. multocida enzyme. However, as

the structure of the gshF enzyme does not reveal the presence of

a disulﬁde bond even in the dimer [30], the inhibitory activity

of GSH in P. multocida is unlikely to be functioning through its

action on a disulﬁde bond as observed in the present study.

In our investigations with the E. coli γ -GCS enzyme, we have

taken a biochemical genetic strategy to investigate the role of the

cysteine residues and the disulﬁde bond. This was largely because

the cysteine residues of E. coli showed an unusual conservation

pattern, which, along with the lack of redox dependence of the

Xanthomonas and Agrobacterium enzymes of the plant lineage,

The inhibition of E. coli γ -GCS activity in the presence

of increasing concentrations of GSH was observed many

years ago [24]. However the mode of inhibition was never

investigated at that time, and it was largely assumed that the

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

Redox-dependent stability of γ -GCS

793

initially questioned the functional importance of the disulﬁde

bond in the enzyme. However, the suppressor strategy turned

out to be extremely rewarding as it not only yielded insights

into the disulﬁde bond, but also yielded a functional γ -GCS

variant that lacked any cysteine residues. This has proved very

useful in our analysis and has greatly facilitated our reaching the

above conclusions. Furthermore, the use of the 7CF mutant which

only contained the two cysteine residues involved in disulﬁde

bond formation, but lacking other cysteine residues, ensured

that no other non-physiological disulﬁdes might be forming to

stabilize the structure.

REFERENCES

1

2

Meister, A. and Anderson, M. E. (1983) Glutathione. Annu. Rev. Biochem. 52, 711–760

Masip, L., Veeravalli, K. and Georgiou, G. (2006) The many faces of glutathione in

bacteria. Antioxid. Redox Signaling 8, 753–762

3

4

McLaggan, D., Logan, T. M., Lynn, D. G. and Epstein, W. (1990) Involvement of

γ -glutamyl peptides in osmoadaptation of Escherichia coli. J. Bacteriol. 172, 3631–3636

Thorgersen, M. P. and Downs, D. M. (2009) Oxidative stress and disruption of labile iron

generate speciﬁc auxotrophic requirements in Salmonella enterica. Microbiology 155,

295–304

5

6

7

Janowiak, B. E. and Grifﬁth, O. W. (2005) Glutathione synthesis in Streptococcus

agalactiae. One protein accounts for γ -glutamylcysteine synthetase and glutathione

synthetase activities. J. Biol. Chem. 280, 11829–11839

Vergauwen, B., De Vos, D. and Van Beeumen, J. J. (2006) Characterization of the

bifunctional γ -glutamate-cysteine ligase/glutathione synthetase (GshF) of Pasteurella

multocida. J. Biol. Chem. 281, 4380–4394

Gopal, S., Borovok, I., Ofer, A., Yanku, M., Cohen, G., Goebel, W., Kreft, J. and

Aharonowitz, Y. (2005) A multidomain fusion protein in Listeria monocytogenes catalyzes

the two primary activities for glutathione biosynthesis. J. Bacteriol. 187, 3839–3847

Copley, S. D. and Dhillon, J. K. (2002) Lateral gene transfer and parallel evolution in the

history of glutathione biosynthesis genes. Genome Biol. 3, research0025

Tu, Z. and Anders, M. W. (1998) Identiﬁcation of an important cysteine residue in human

glutamate-cysteine ligase catalytic subunit by site-directed mutagenesis. Biochem. J.

336, 675–680

The appearance of the combination of cysteine and aromatic

amino acids at positions 372 and 395 in many of the suppressors

in place of the disulﬁde-bonding cysteine residues was initially

puzzling. However, it has been reported that cysteine–aromatic

aminoacidinteractions are stabilizedby π-type hydrogenbonding

between the aromatic ring and thiol group of cysteine and

the interaction between the aromatic ring and aliphatic group

(-CH₂) of cysteine [31]. This interaction might have been

sufﬁcient to stabilize the folding conformation of the catalytic

domain. It is interesting to note that the putative γ -GCS of

Erwinia pyrifoliae contains a Cys³⁷²/Tyr³⁹⁵combination, whereas

the putative γ -GCS of Candidatus hamiltonella and Baumanni

cicadellinicola carry a Tyr³⁷²/Thr³⁹⁵and Tyr³⁷²/Gly³⁹⁵combination

respectively, and it would be interesting to see the properties

of these enzymes in the light of the observations made in

the present study. In the E. coli CF γ -GCS enzyme, the con-

version of valine to phenylalanine at position 375 even in the

absence of any cysteine residues leading to the appearance of a

functional CF molecule is also intriguing. Once again, the change

occurred at a residue near the disulﬁde-bond-forming residues,

suggesting that a similar interaction might be stabilizing the active

site structure, but this awaits more detailed structural insights.

Although the ﬁndings and postulated mechanism of redox

regulation have been demonstrated for the E. coli γ -GCS enzyme,

it is likely that the mechanism may also, in part, function in

the case of the plant and human enzymes as well, but this

aspect of these enzymes has never been investigated, and actual

demonstration would await further experiments on these enzymes.

8

9

10 Fraser, J. A., Kansagra, P., Kotecki, C., Saunders, R. D. and McLellan, L. I. (2003) The

modiﬁer subunit of Drosophila glutamate-cysteine ligase regulates catalytic activity by

covalent and noncovalent interactions and inﬂuences glutathione homeostasis in vivo. J.

Biol. Chem. 278, 46369–46377

11 Fraser, J. A., Saunders, R. D. and McLellan, L. I. (2002) Drosophila melanogaster

glutamate-cysteine ligase activity is regulated by a modiﬁer subunit with a mechanism of

action similar to that of the mammalian form. J. Biol. Chem. 277, 1158–1165

12 Hothorn, M., Wachter, A., Gromes, R., Stuwe, T., Rausch, T. and Scheffzek, K. (2006)

Structural basis for the redox control of plant glutamate cysteine ligase. J. Biol. Chem.

281, 27557–27565

13 Hicks, L. M., Cahoon, R. E., Bonner, E. R., Rivard, R. S., Shefﬁeld, J. and Jez, J. M. (2007)

Thiol-based regulation of redox-active glutamate-cysteine ligase from Arabidopsis

thaliana. Plant Cell 19, 2653–2661

14 Gromes, R., Hothorn, M., Lenherr, E. D., Rybin, V., Scheffzek, K. and Rausch, T. (2008)

The redox switch of γ -glutamylcysteine ligase via a reversible monomer-dimer transition

is a mechanism unique to plants. Plant J. 54, 1063–1075

15 Hibi, T., Nii, H., Nakatsu, T., Kimura, A., Kato, H., Hiratake, J. and Oda, J. (2004) Crystal

structure of γ -glutamylcysteine synthetase: insights into the mechanism of catalysis by a

key enzyme for glutathione homeostasis. Proc. Natl. Acad. Sci. U.S.A. 101, 15052–15057

16 Ito, H., Fukuda, Y., Murata, K. and Kimura, A. (1983) Transformation of intact yeast cells

treated with alkali cations. J. Bacteriol. 153, 163–168

17 Rose, R. M. and Fink, R. G. (1987) KAR1, a gene required for function of both intranuclear

and extranuclear microtubules in yeast. Cell 48, 1047–1060

18 Kaur, J., Srikanth, C. V. and Bachhawat, A. K. (2009) Differential roles played by the native

cysteine residues of the yeast glutathione transporter, Hgt1p. FEMS Yeast Res. 9, 849–866

19 Kelly, B. S., Antholine, W. E. and Grifﬁth, O. W. (2002) Escherichia coli

γ -glutamylcysteine synthetase. Two active site metal ions affect substrate and inhibitor

binding. J. Biol. Chem. 277, 50–58

20 Sevier, C. S., Qu, H., Heldman, N., Gross, E., Fass, D. and Kaiser, C. A. (2007)

Modulation of cellular disulﬁde-bond formation and the ER redox environment by

feedback regulation of Ero1. Cell 129, 333–344

AUTHOR CONTRIBUTION

Shailesh Kumar isolated most of the genetic suppressors, carried out the biochemical,

biophysical and kinetic studies, analysed the results, and helped in writing the paper. Neha

Kasturia isolated and evaluated the functionality of a few of the suppressors. Amit Sharma

constructed the different CF mutants of E. coli γ -GCS. Manish Datt helped in building

a three-dimensional model of γ -GCS with the wild-type and the CF suppressor. Anand

Bachhawat supervised the project, analysed the data and contributed towards writing the

paper.

21 Parsell, D. A. and Sauer, R. T. (1989) The structural stability of a protein is an important

determinant of its proteolytic susceptibility in Escherichia coli. J. Biol. Chem. 264,

7590–7595

22 Kim, S. G., Chi, Y. H., Lee, J. S., Schlesinger, S. R., Zabet-Moghaddam, M., Chung, J. S.,

Knaff, D. B., Kim, S. T., Lee, S. Y. and Kim, S. K. (2010) Redox properties of a

thioredoxin-like Arabidopsis protein, AtTDX. Biochim. Biophys. Acta 1804, 2213–2221

23 Setterdahl, A. T., Goldman, B. S., Hirasawa, M., Jacquot, P., Smith, A. J., Kranz, R. G. and

Knaff, D. B. (2000) Oxidation-reduction properties of disulﬁde-containing proteins of the

Rhodobacter capsulatus cytochrome c biogenesis system. Biochemistry 39,

10172–10176

ACKNOWLEDGEMENTS

We thank Dr Takao Hibi for providing the 4CF E. coli gshA plasmid. We thank Dr

Purnananda Guptasarma for his help and advice with the CD experiments and analysis of

the CD data. We thank Dr Chitranshu Kumar for his advice on the AMS experiments.

24 Huang, C. S., Moore, W. R. and Meister, A. (1988) On the active site thiol of

γ -glutamylcysteine synthetase: relationships to catalysis, inhibition, and regulation.

Proc. Natl. Acad. Sci. U.S.A. 85, 2464–2468

FUNDING

25 Kelly, S. M., Jess, T. J. and Price, N. C. (2005) How to study proteins by circular

dichroism. Biochim. Biophys. Acta 1751, 119–139

26 Zheng, M., Aslund, F. and Storz, G. (1998) Activation of the OxyR transcription factor by

reversible disulﬁde bond formation. Science 279, 1718–1721

This work was supported, in part, by Grant-in-Aid projects to A.K.B. from the Departments

of Science and Technology and the Department of Biotechnology, Government of India.

S.K., N.K. and M.D. were the recipients of a Research Fellowship from the Council of

Scientiﬁc and Industrial Research, Government of India.

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

794

S. Kumar and others

27 Richman, P. G. and Meister, A. (1975) Regulation of γ -glutamyl-cysteine synthetase by

nonallosteric feedback inhibition by glutathione. J. Biol. Chem. 250, 1422–1426

28 Misra, I. and Grifﬁth, O. W. (1998) Expression and puriﬁcation of human

γ -glutamylcysteine synthetase. Protein Expression Purif. 13, 268–276

29 Watanabe, K., Murata, K. and Kimura, A. (1986) Puriﬁcation and characterization of

γ -glutamylcysteine synthetase of Escherichia coli B. Agric. Biol. Chem. 50, 1925–1930

30 Stout, J., De Vos, D., Vergauwen, B. and Savvides, S. N. (2012) Glutathione biosynthesis

in bacteria by bifunctional GshF is driven by a modular structure featuring a novel hybrid

ATP-grasp fold. J. Mol. Biol. 416, 486–494

31 Duan, G., Smith, V. H. and Weaver, D. F. (2001) Characterization of aromatic-thiol π-type

hydrogen bonding and phenylalanine-cysteine side chain interactions through ab initio

calculations and protein database analyses. Mol. Phys. 99, 1689–1699

Received 2 February 2012/29 October 2012; accepted 5 November 2012

Published as BJ Immediate Publication 5 November 2012, doi:10.1042/BJ20120204

ꢁ

c

ꢁ

c

The Authors Journal compilation 2013 Biochemical Society

Biochem. J. (2013) 449, 783–794 (Printed in Great Britain) doi:10.1042/BJ20120204

SUPPLEMENTARY ONLINE DATA

Redox-dependent stability of the γ -glutamylcysteine synthetase enzyme

of Escherichia coli: a novel means of redox regulation

Shailesh KUMAR, Neha KASTURIA¹, Amit SHARMA², Manish DATT³and Anand K. BACHHAWAT⁴

Institute of Microbial Technology (CSIR), Sector 39-A, Chandigarh 160 036, India

Figure S1 Conservation patterns of active site residues in γ -proteobacterium

The E. coli γ -GCS protein sequences were used to retrieve the γ -GCS homologues of different organisms of lineage I. The multiple sequence alignment of the protein sequences were generated

using ClustalW using default parameters. The conserved active site residues are highlighted.

1

Present address: Institute of Genomics and Integrative Biology (CSIR), Mall Road, Delhi 110007, India

Present address: Astbury Centre for Structural Molecular Biology, University of Leeds, Leeds LS2 9JT, U.K.

Present address: Center for Proteomics and Bioinformatics, Case Western Reserve University, Cleveland, OH 44106, U.S.A.

To whom correspondence should be addressed at the present address: Indian Institute of Science Education and Research, Mohali, S.A.S. Nagar,

2

3

4

Punjab 140306, India (email anand@iisermohali.ac.in).

ꢀ

c

ꢀ

c

The Authors Journal compilation 2013 Biochemical Society

S. Kumar and others

Figure S2 Multiple sequence alignment of E. coli γ -GCS with a member of lineage I

The E. coli γ -GCS sequences were used to retrieve the γ -GCS homologues of different organisms of lineage I. The multiple sequence alignment of the protein sequences was generated using

ClustalW program using default parameters. The conserved cysteine residues are highlighted.

ꢀ

c

ꢀ

c

The Authors Journal compilation 2013 Biochemical Society

Redox-dependent stability of γ -GCS

Figure S3 Cartoon representation of the crystal structure of E. coli γ -GCS

(PDB code 1V4G)

The position of the cysteine residues are labelled and shown as ball and stick. Four of these

cysteine residues (Cys¹⁰⁶, Cys¹⁶⁴, Cys²⁰⁵and Cys²²³) were mutated to serine to facilitate

crystallization; those rotamers were reverted to cysteine in silico. The co-ordinates of Cys¹⁶⁴are

missing in the crystal structure and hence not shown here. Two sites of mutations identiﬁed in

our experiments are shown as ball and stick and labelled in green. Amino acid residues lining

the glutamate- and cysteine-binding sites are shown as wires in magenta and lime respectively.

ꢀ

c

ꢀ

c

The Authors Journal compilation 2013 Biochemical Society

S. Kumar and others

Figure S4 Functional complementation displayed by suppressor of 9CF γ -GCS

E. coli 9CF γ -GCS suppressors were transformed in S. cerevisiae strain ABC1195 (MATα his3ꢀ1 leu2ꢀ-0 lys2ꢀ-0 ura3ꢀ-0 gsh1ꢀ:: LEU2) and spotted on to SD medium containing/lacking

glutathione as the sole source of organic sulfur. The vector pTEF416 was used as a control. (A) Swapping of the aromatic cysteine position from 9CF + S372C + S395W to 9CF + S372W + S395C.

(B) Functional complementation of serine aromatic mutants at position 372/395.

Figure S5 Coomassie Blue gel showing the purity of puriﬁed protein

Eluted fractions (20 μl) of the wild-type (WT) and mutant proteins from the Ni-NTA columns

were run on SDS/PAGE (10% gel) and stained with Coomassie Blue dye. Molecular masses are

indicated in kDa to the left-hand side.

ꢀ

c

ꢀ

c

The Authors Journal compilation 2013 Biochemical Society

Redox-dependent stability of γ -GCS

Figure S7 Redox midpoint potential measurement of 7CF γ -GCS protein

Theoxidation–reductiontitrationfor7CFγ -GCSproteinwascarriedoutin100 mMHepesbuffer,

pH 7.0, that contained DTT at a total concentration of 10 mM. Proteins (50 μg) were incubated

for 3 h to achieve redox equilibrium. The sample was then treated with monobomobimane

(mBBr) and the E_hvalue was calculated from the Nernst equation as described previously [1,2].

Figure S6 CD spectra and melting temperature scan of wild-type and

E. coli γ -GCS wild-type and mutant protein. The T_mscan of wild-type and mutants were carried

mutant γ -GCS

◦

out in the temperature range 25–90 C. The spectra of wild-type, 7CF, 9CF + R374Q + V375F

and 9CF + S372C + S395W was acquired at 222 nm. (C) CD spectroscopy temperature scan

of E. coli γ -GCS in the presence of DTT. The wild-type γ -GCS sample was prepared in

50 mM Tris/HCl, pH 8.0, 20 mM NaCl, 0.5 M guanidinium chloride and 2 mM DTT. Protein

concentration was 150 μg/ml. The temperature scan with DTT and without DTT was carried out

at 222 nm. MRE, mean residue ellipticity.

The sample preparation for CD spectra of puriﬁed wild-type (WT) and mutant protein was

carried out in 50 mM Tris/HCl, pH 8.0, 20 mM NaCl and 0.5 M guanidinium chloride. A

protein concentration of 150 μg/ml was used. (A) CD spectra of wild-type and mutant γ -GCS.

The CD spectra of wild-type, 7CF, 9CF + R374Q + V375F and 9CF + S372C + S395W were

acquired at a wavelength range 205–255 nm. (B) A CD spectroscopy temperature scan of

ꢀ

c

ꢀ

c

The Authors Journal compilation 2013 Biochemical Society

S. Kumar and others

Figure S8 Identiﬁcation of the reduced form of γ -GCS in vivo using AMS

and in vitro using MM(PEG)

(A) The wild-type (WT), 7CF and 9CF + R374Q + V375F plasmids harbouring the E. coli

strain were grown in LB medium supplemented with 100 μg/ml ampicillin. Expression

was induced at D₆₀₀= 0.4 with 1 mM IPTG. At 90 min post-induction, the growing culture was

treated with 100% TCA to a ﬁnal concentration of 20%. The extracted protein was treated with

30 mM AMS as described in the Materials and methods section of the main text. The extracted

protein was separated by SDS/PAGE (14% gel) under non-reducing conditions. The detection

of protein was carried out by Western blotting using an anti-His antibody. (B) The wild-type,

7CF, 9CF + R374Q + V375F and 9CF + S372C + S395Y proteins were puriﬁed from E. coli.

Puriﬁed protein (400–500 μg) was precipitated with 20% TCA by adding an equal volume.

Precipitated protein was washed with ice-cold acetone twice and air-dried. The protein sample

was treated with 5 mM MM(PEG) as described in the Materials and methods section of the main

text. Protein detection was carried out by Western blotting using an anti-His antibody. Molecular

masses are indicated to the left-hand side in kDa.

ꢀ

c

ꢀ

c

The Authors Journal compilation 2013 Biochemical Society

Redox-dependent stability of γ -GCS

Table S1 List of oligonucleotides and their sequences used in the present study

EC, E. coli; Fw, forward; Rev, reverse.

Oligomer name

Sequence (5^ꢁ→3^ꢁ)

EcGshAC357S Rev

EcGshAC372S Rev

EcGshAC395S Rev

EcGshAC433C439S Fw

EcGshAC433C439S Rev

EcGshA400Rev_seq

EcGshA400Fw_seq

EcGshABamHI Fw

CTGCTCATTTCCGGTGCATCAGCCAGCGCAGACCAGACCATAAACAGGTCGAGG

GGTTCCAGTTAACGCGTGTAGAGGCAAGTTCGCTACTGCTCATTTCCGGTCATC

GAACAGATCTTTACCCACCTGCGGTAACGGGAACTGTGCGGTTTCGGAGCCGATACCCAGCGTCGTCAGAC

GCGAAGCGTATCAGAAAGTGTCTGATGAACTGGTTGCCTCCTTCGATAATCCCGATCTGA

GTCAGATCGGGATTATCGAAGGAGGCAACCAGTTCATCAGACACTTTCTGTACGCTTCGC

CAGCCCTTCACGATACAGCG

CGCTGTATCGTGAAGGGCTG

CTGCAGGGATCCATGATCCCGGACGTATCACAG

CAGCTGCTCGAGTCAGGCGTGTTTTTCCAGCC

EcGshAXhoI Rev

EcGshA S372W _Rev

EcGshA S395F _ Fw

EcGshA S395F _ Rev

EcGshA CHis BamH1 _Rev

EcGshA_Sma1 Fw

R374Q F375V Rev

Q374R V375F Fw

Q374R V375F Rev

EcGshA C372A Rev

EcGshA C395A Rev

GTTCCAGTTAACGCGTGTCCAGGCAAGTTCGCTACTGCTCATTTC

GTCTGACGCTGGGTATCGGCTTCGAAACCGCACAGTTCCCG

CGGGAACTGTGCGGTTTCGAAGCCGATACCCAGCGTCAGAC

ACCTAGGGATCCTTAATGATGGTGATGGTGATGGGCGTGTTTTTCCAGCCACACC

GATATCCCCGGGATGATCCCGGACGTATCACAGG

CCGGTTCCAGTTAACTTGTGTAGAGGC

GCGAACTTGCCTCTACACGCTTTAACTGGAACCGGG

CCCGGTTCCAGTTAAAGCGTGTAGAGGCAAGTTCGC

GTACCAGTTAACGCGTGTAGCGGCAAGTTCGCTACTGCTCATTTCC

GTCAGTAGATCTTTACCCACCTGCGGTAACGGGAACTGTGCGGTTTCGGCGCCGATACCCAGCGTC

ꢀ

c

ꢀ

c

The Authors Journal compilation 2013 Biochemical Society

S. Kumar and others

Table S2 List of the representative members of lineage I used for sequence alignment to construct the phylogenetic tree in Figure 1 of the main text

Abbreviation used

NCBI accession number

Organism name

Esch_coli

Shig_ﬂex

Salm_typh

Citr_kose

Ente_canc

Kleb_pneu

Cron_turi

Serr_odor

Yers_pest

Pect_caro

Erwi_pyri

Soda_glos

Prot_mira

Baum_cica

Arse_naso

Cand_hami

Alii_salmo

Shew_deni

Idio_balt

Ocea

NP_417173.1

NP_838224.1

NP_461744.1

Escherichia coli str. K-12 substr. MG1655

Shigella ﬂexneri 2a str. 2457T

Salmonella enterica subsp. enterica serovar Typhimurium str. LT2

Citrobacter koseri ATCC BAA-895

Enterobacter cancerogenus ATCC 35316

Klebsiella pneumoniae subsp. pneumoniae MGH 78578

Cronobacter turicensis z3032

Serratia odorifera 4Rx13

Yersinia pestis KIM 10

Pectobacterium carotovorum subsp. carotovorum PC1

Erwinia pyrifoliae Ep1/96

Sodalis glossinidius str. ‘morsitans’

Proteus mirabilis HI4320

Baumannia cicadellinicola str. Hc

Arsenophonus nasoniae

Candidatus Hamiltonella defensa 5AT (Acyrthosiphon pisum)

Aliivibrio salmonicida LFI1238

Shewanella denitriﬁcans OS217

Idiomarina baltica OS145

Oceanospirillum sp. MED92

Acinetobacter calcoaceticus RUH2202

Alcanivorax sp. DG881

Chromohalobacter salexigens DSM 3043

Francisella novicida GA99-3548

Burkholderia phytoﬁrmans PsJN

Pseudomonas aeruginosa PA7

Allochromatium vinosum DSM 180

Thioalkalivibrio sp. HL-EbGR7

Nitrosococcus oceani ATCC 19707

Psychrobacter arcticus 273-4

YP_001455545.1

ZP_05969993.1

YP_001336655.1

YP_003211651.1

ZP_06192317.1

NP_668220.1

YP_003018766.1

YP_002649770.1

YP_454221.1

YP_002150150.1

YP_588658.1

CBA73578.1

YP_002925016.1

YP_002262162.1

YP_562228.1

ZP_01043504.1

ZP_01166769.1

ZP_06058310.1

ZP_05043145.1

YP_574700.1

ZP_04989211.1

YP_001893776.1

YP_001351264.1

YP_003442779.1

YP_002512775.1

YP_343611.1

Acin_calc

Alca

Chro_sale

Fran_novi

Burk_phyt

Pseu_aeru

Allo_vino

Thio

Nitr_ocea

Psyc_arct

List_mono

Clos_botu

Past_mult

Haem_inﬂ

Ente_faec

Stre_ther

Stre_agal

Ente_faeca

YP_263343.1

ZP_05300878.1

YP_001920910.1

NP_245985.1

ZP_01786436.1

ZP_05667624.1

ADH82115.1

Listeria monocytogenes LO28

Clostridium botulinum E3 str. Alaska E43

Pasteurella multocida subsp. multocida str. Pm70

Haemophilus inﬂuenzae R3021

Enterococcus faecium 1,141,733

Streptococcus thermophilus

Streptococcus agalactiae FSL S3-026

Enterococcus faecalis AR01/DG

EGS26792.1

ZP_05594212.1

REFERENCES

1

Kim, S. G., Chi, Y. H., Lee, J. S., Schlesinger, S. R., Zabet-Moghaddam, M., Chung, J. S.,

Knaff, D. B., Kim, S. T., Lee, S. Y. and Kim, S. K. (2010) Redox properties of a

thioredoxin-like Arabidopsis protein, AtTDX. Biochim. Biophys. Acta 1804, 2213–2221

Setterdahl, A. T., Goldman, B. S., Hirasawa, M., Jacquot, P., Smith, A. J., Kranz, R. G. and

Knaff, D. B. (2000) Oxidation-reduction properties of disulﬁde-containing proteins of the

Rhodobacter capsulatus cytochrome c biogenesis system. Biochemistry 39, 10172–10176

2

Received 2 February 2012/29 October 2012; accepted 5 November 2012

Published as BJ Immediate Publication 5 November 2012, doi:10.1042/BJ20120204

ꢀ

c

ꢀ

c

The Authors Journal compilation 2013 Biochemical Society

Article Doi

Redox-dependent stability of the γ-glutamylcysteine synthetase enzyme of Escherichia coli: A novel means of redox regulation

DOI: 10.1042/BJ20120204

Source and publish data:

Authors:

Article abstract of DOI:10.1042/BJ20120204

Full text of DOI:10.1042/BJ20120204

Products guided by the article

R&D Labs maybe for 636-58-8

Relevant to this article

Hot Product