Members of the family include the pyruvate dehydrogenase complex PDHC, which catalyzes the conversion of pyruvate to acetyl-CoA and so links glycolysis and the citric acid cycle; the 2-o
Trang 1the archaeon Thermoplasma acidophilum ) recombinant expression, assembly and characterization
Caroline Heath1, Mareike G Posner1, Hans C Aass1, Abhishek Upadhyay2, David J Scott3,
David W Hough1and Michael J Danson1
1 Centre for Extremophile Research, Department of Biology and Biochemistry, University of Bath, UK
2 Department of Biology and Biochemistry, University of Bath, UK
3 National Centre for Macromolecular Hydrodynamics, School of Biosciences, University of Nottingham, Sutton Bonington, UK
In aerobic bacteria and eukaryotes, a family of
2-oxoacid dehydrogenase multi-enzyme complexes
(OADHCs) functions in the pathways of central
metabolism The complexes are responsible for the
oxi-dative decarboxylation of 2-oxoacids to their
corre-sponding acyl-CoAs Members of the family include
the pyruvate dehydrogenase complex (PDHC), which
catalyzes the conversion of pyruvate to acetyl-CoA
and so links glycolysis and the citric acid cycle; the
2-oxoglutarate dehydrogenase complex (OGDHC),
which catalyzes the conversion of 2-oxoglutarate to
succinyl-CoA within the citric acid cycle; and the
branched-chain 2-oxoacid dehydrogenase complex (BCOADHC), which oxidatively decarboxylates the branched-chain 2-oxoacids produced by the transami-nation of valine, leucine and isoleucine The complexes comprise multiple copies of three component enzymes: 2-oxoacid decarboxylase (E1), dihydrolipoyl acyl-trans-ferase (E2) and dihydrolipoamide dehydrogenase (E3) [1–3] E2 forms the structural core of the complex, with multiple polypeptide chains associating into octa-hedral (24-mer) or icosaocta-hedral (60-mer) configurations, depending on the particular complex and the source organism [2,4] E1 and E3 bind noncovalently to the
Keywords
Archaea; metabolism; multi-enzyme
complex; 2-oxoacid dehydrogenase;
thermophile
Correspondence
M J Danson, Centre for Extremophile
Research, Department of Biology and
Biochemistry, University of Bath, Bath,
BA2 7AY, UK
Fax: +44 1225 386779
Tel: +44 1225 386509
E-mail: M.J.Danson@bath.ac.uk
(Received 29 June 2007, revised 23 August
2007, accepted 24 August 2007)
doi:10.1111/j.1742-4658.2007.06067.x
The aerobic archaea possess four closely spaced, adjacent genes that encode proteins showing significant sequence identities with the bacterial and eukaryal components comprising the 2-oxoacid dehydrogenase multi-enzyme complexes However, catalytic activities of such complexes have never been detected in the archaea, although 2-oxoacid ferredoxin oxidore-ductases that catalyze the equivalent metabolic reactions are present In the current paper, we clone and express the four genes from the thermophilic archaeon, Thermoplasma acidophilum, and demonstrate that the recombi-nant enzymes are active and assemble into a large (Mr¼ 5 · 106) multi-enzyme complex The post-translational incorporation of lipoic acid into the transacylase component of the complex is demonstrated, as is the assembly of this enzyme into a 24-mer core to which the other components bind to give the functional multi-enzyme system This assembled complex
is shown to catalyze the oxidative decarboxylation of branched-chain 2-oxoacids and pyruvate to their corresponding acyl-CoA derivatives Our data constitute the first proof that the archaea possess a functional 2-oxo-acid dehydrogenase complex
Abbreviations
BCOADHC, branched-chain 2-oxoacid dehydrogenase complex; CoASH, coenzyme-A; DLS, dynamic light scattering; E1, 2-oxoacid
decarboxylase; E2, dihydrolipoyl acyl-transferase; E3, dihydrolipoamide dehydrogenase; FOR, ferredoxin oxidoreductase; IPTG, isopropyl thio-b- D -galactoside; Mr, relative molecular mass; OADHC, 2-oxoacid dehydrogenase complex; OGDHC, 2-oxoglutarate dehydrogenase multienzyme complex; PDHC, pyruvate dehydrogenase complex; TPP, thiamine pyrophosphate.
Trang 2E2 core E1 may occur as a homodimer or as an a2b2
hetero-tetramer, depending upon the source and the
type of complex, although in all cases E3 is a dimer of
identical subunits
E2 also forms the catalytic core of the complex; a
lipoyl moiety, covalently attached to a lysine residue in
the lipoyl domain of the E2, serves as a swinging arm,
connecting the active sites of each enzyme and
chan-nelling substrate through the complex [2,3] Thus, E1
catalyzes the thiamine pyrophosphate (TPP)-dependent
decarboxylation of the 2-oxoacid and the transfer
of the resulting acyl group to the lipoic acid of E2
E2 then transfers the acyl-group to coenzyme-A
(CoASH), after which E3 serves to reoxidize the
dihydrolipoyl moiety It does so by the reduction of
the noncovalently bound cofactor FAD, in conjunction
with a protein disulfide bond and an amino acid base,
all of which are themselves then reoxidized by NAD+
to form NADH
In Archaea [5] and anaerobic bacteria, the
equiva-lent oxidation of 2-oxoacids is catalyzed by an
un-related, and structurally more simple, family of
2-oxoacid ferredoxin oxidoreductases (FORs) This
comprises the pyruvate FOR, the 2-oxoglutarate FOR
and the 2-oxoisovalerate FOR, which catabolize the
oxidative decarboxylation of pyruvate, 2-oxoglutarate
and the branched-chain 2-oxoacids, respectively [6–8]
The pyruvate FOR from the halophilic archaeon
Halo-bacterium halobium is an a2b2 structure [9], whereas in
Sulfolobus solfataricus and Aeropyrum pernix it is an
ab dimer, and an octamer (a2b2c2d2) in Pyrococcus
furiosus, Methanothermobacter thermoautotrophicum
and Archaeoglobus fulgidus [reviewed in 5,8] The
cata-lytic reaction of FORs does not involve a lipoic acid
moiety or NAD+; rather, the acyl-moiety formed on
decarboxylation of the 2-oxoacid is handed on directly
to CoASH, and the reducing equivalents to ferredoxin
via the enzyme’s iron-sulfur centre [6–8]
No OADHC activity has ever been detected in any
archaeon [5] However, detection of E3 and lipoic acid
in various archaea [10–12] led to the discovery of a
putative OADHC operon in Haloferax volcanii [13]
and our subsequent detection of similar putative
ope-rons in the genome sequences of the aerobic archaea
Thermoplasma acidophilum, S solfataricus, Sulfolobus
acidocaldarius, A pernix, Pyrobaculum aerophilum,
Halobacterium NRC1 and Ferroplasma acidophilum
(M G Posner, unpublished data)
We have previously expressed the E1a and E1b
genes of the putative OADHC from T acidophilum in
Escherichia coli and shown the recombinant proteins
to assemble into an a2b2 enzyme that catalyzes the
decarboxylation of the branched-chain 2-oxoacids and
pyruvate [14] In the current paper, we report the clon-ing and expression of the E2 and E3 genes of the same operon from T acidophilum, and the in vitro assembly and characterization of an active 2-oxoacid dehydro-genase complex
This, then, is the first evidence that the putative OADHC operon in an archaeon encodes a 2-oxoacid dehydrogenase multi-enzyme complex that is func-tional in vitro and therefore may have physiological significance
Results
Expression and purification of the E1, E2 and E3 components
The E1 component The E1 a2b2 recombinant enzyme was produced as described previously, and was shown to be catalyti-cally active with the branched-chain 2-oxoacids 4-methyl-2-oxopentanoate, 3-methyl-2-oxopentanoate and 3-methyl-2-oxobutanoate, and with pyruvate [14]
By dynamic light scattering (DLS), its Mr was found
to be 168 000 (± 6000), consistent with the value
of 165 000 determined by gel filtration [14] and the expected value of 157 000 from the protein sequences
The E2 component The gene encoding the E2 component was PCR-ampli-fied from T acidophilum genomic DNA and cloned into the pET28a expression vector, as described in Experimental procedures The E2 protein was then expressed in two host systems: E coli BL21(DE3) cells,
in medium supplemented with lipoic acid but without isopropyl thio-b-d-galactoside (IPTG) induction, and
E coliBL21(DE3)pLysS cells without supplementation but with IPTG induction Both methods yielded solu-ble E2 protein, although the level of expression was significantly greater in the pLysS cells In each case, E2 was purified to > 95% homogeneity using His-Bind affinity chromatography followed by anion exchange chromatography
The E2 Mrvalues predicted from the published gene sequence are 46 276 for unlipoylated protein and
46 464 for the polypeptide possessing a single lipoyl residue Accordingly, mass spectrometric analysis revealed that the E2 expressed in E coli grown in lipoic acid-supplemented medium comprised an approximately equimolar mixture of unlipoylated (Mr¼ 46 273) and lipoylated (Mr¼ 46 461) protein Furthermore, from MS-analysis of tryptic fragments,
Trang 3lipoylation was shown to have occurred at K42, which
by sequence comparisons corresponds to the lipoylated
lysine residue in bacterial E2 components (Fig 1)
However, the E2 protein produced by expression in
induced cells, without lipoic acid supplement to the
growth medium, was < 5% lipoylated As reported
below, consistent with these data is the observation
that only the complex assembled using the lipoylated
E2 component showed detectable catalytic activity in
the overall complex assay
The E3 component
The gene encoding the E3 component was
PCR-amplified from T acidophilum genomic DNA and
cloned into the pET28a expression vector, as
described in Experimental procedures Expression of
the E3 gene was carried out in E coli BL21(DE3)
cells Small-scale expression trials with induced versus
uninduced host cells revealed a higher E3 activity in
the soluble cell extract from uninduced cells, and this
was confirmed by SDS⁄ PAGE The soluble protein
was purified to > 95% homogeneity by heat
precipi-tation at 60C and His-Bind affinity
chromatogra-phy The absorption spectrum of the purified protein
showed peaks at 375, 450 and 475 nm that are
char-acteristic of the presence of FAD in the enzyme;
using a molar absorption coefficient for FAD of
11 300 m)1Æcm)1 at 455 nm [15], the flavin content
was calculated to be 1.0 (± 0.1) FAD per
polypep-tide (Mr¼ 49 867) Analytical gel filtration revealed
an Mr¼ 100 000 for the purified enzyme, and DLS
gave a similar value (117 000 ± 2000), suggesting a
dimeric structure, as has been found for the bacterial
and eukaryotic enzymes The maximal specific activity
of the enzyme was found to be 22 (± 1) lmolÆ
min)1Æmg)1, a value that is considerably higher than
the overall complex activity (see below)
Assembly of OADHC from the recombinant components
Complex assembly When E1, E2 and E3 were incubated at 55C for
10 min in 20 mm sodium phosphate buffer, pH 7.5, containing 2 mm MgCl2 and 0.2 mm TPP, prior to assay, whole complex activity was subsequently detected with the branched-chain 2-oxoacids 4-methyl-2-oxopentanoate, 3-methyl-2-oxopentanoate and 3-methyl-2-oxobutanoate, and with pyruvate The 10-min incubation is necessary to allow E1-TPP bind-ing [16], but no increase in rate was seen when the incubation, after subunit mixing, was extended to 1 h
at 4C, 25 C or 55 C
With an E2 : E3 molar ratio fixed at 1 : 1 (E2 poly-peptide to E3 dimer), titration with E1 resulted in an increase in whole complex activity until a maximum was reached at a molar ratio (E1⁄ E2) of approximately 2–3 : 1 [E1 a2b2 tetramer to E2 polypeptide] (Fig 2) However, a reduction of the E3 ratio did not cause a significant change in whole complex activity, and by investigating the mixing of various amounts of the complex components it was found that maximum OADHC activity was achieved when the E1, E2 and E3 subunits were mixed in a molar stoichiometry of
3 : 1 : 0.1 However, as described below, this stoichio-metry does not equate to the amounts of the three enzymes in the assembled complex
B subtilis PDHC
B subtilis BCOADHC
Tp acidophilum OADHC
E coli OGDHC
E coli PDHC (lipoyl domain 3)
Fig 1 Alignment of various E2 sequences around the lipoylated
lysine residue The T acidophilum E2 protein sequence was aligned
with those of the E2 components of the following OADHCs using
the CLUSTALW multiple sequence alignment program: Bacillus subtilis
PDHC; B subtilis BCOADHC; E coli OGDHC; E coli PDHC (lipoyl
domain 3) Residues 27–53 of the T acidophilum sequence are
shown, along with the aligned regions of the other E2 sequences,
and the lysine residue that is lipoylated in each sequence is marked
by a (*).
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
Molar Ratio of E1:E2
Fig 2 Whole complex activity as a function of the E1 : E2 ratio Lipoylated E2 and E3 were mixed in a molar ratio of 1 : 1 (E2 poly-peptide to E3 dimer), and to this mixture varying amounts of E1 were added The mixture was then incubated at 55 C for 10 min
in the presence of 2 m M MgCl2and 0.2 m M TPP, following which samples were taken and assayed for whole complex activity (expressed as A (absorbance units).min)1) using the substrate 3-methyl-2-oxopentanoate The molar ratio of E1 : E2 is expressed
as E1 a2b2tetramer to E2 polypeptide.
Trang 4OADHC activity
The whole complex activity detected with
4-methyl-2-oxopentanoate, 4-methyl-2-oxopentanoate,
3-methyl-2-oxobutanoate, and pyruvate, but not with
2-oxogluta-rate, is entirely consistent with the substrate specificity
observed for the recombinant E1 enzyme in the
absence of the other complex components [14] The
relative activities are given in Table 1, along with those
for the isolated E1 component and for the BCOADHC
from bovine kidney Using 3-methyl-2-oxopentanoate
as substrate, the assembled T acidophilum complex
exhibited a hyperbolic dependence of velocity on the
2-oxoacid concentration, with KM¼ 250 (± 40) lm
and Vmax¼ 4 (± 0.1) lmolÆmin)1Æmg)1(E2) This
spe-cific activity is comparable to that of the Bacillus
stearothermophilus PDHC [8–10 lmo1Æmin)1Æmg)1(E2)
[17]
Relative molecular mass of the E2 core and the
assembled complex
Analytical centrifugation and DLS were used to
deter-mine the Mr of both the E2 core and the assembled
complex
E2 core
Sedimentation velocity analysis was carried out at
40C as described in Experimental procedures on
recombinant E2 that was 50% lipoylated and had been
purified by His Bind (Novagen, Merck Biosciences
Ltd., Nottingham, UK) and then anion-exchange
chro-matography The sedimentation coefficient
distribu-tions showed a major symmetrical peak constituting
95% of the total protein (Fig 3) Analysis of the data
by direct fitting using the finite element solution of the Lamm equation in sedfit [18], gave a sedimentation coefficient (corrected to water at 20C and infinite dilution; s20,w)¼ 27 (± 1) S (Svedberg units; 1S =
10)13 seconds) and an Mr¼ 1.1 (± 0.1) · 106 From the E2 polypeptide Mrof 46 276, the E2 protein there-fore comprises 23.8 polypeptides; that is, it assembles into a core of 24 subunits possessing octahedral sym-metry
DLS analysis at 55C, the optimum growth temper-ature of T acidophilum, gave an E2 Mr value¼ 1.0 (± 0.1) · 106, in good agreement with the value deter-mined by analytical centrifugation This E2 sample was nonlipoylated, demonstrating that lipoylation is not a prerequisite for assembly of the core Interest-ingly, the Mr of the same E2 at 25C, with no prior heat treatment at 55C, was estimated by DLS to be
55 000, close to the sequence-predicted monomer value (46 276); this indicates that the assembly of E2 into a 24-mer core is temperature dependent
The assembled complex Sedimentation velocity analysis at 40C of assembled complex, created by mixing the components in a
3 : 1 : 0.1 (E1 a2b2: E2 polypeptide: E3 a2) molar ratio, revealed three discrete protein peaks with s20,w values of 50S (24% of total protein), 19S (8%) and 6S
Table 1 Substrate specificities of the T acidophilum 2-oxoacid
dehydrogenase complex The T acidophilum assembled complex
was assayed as described in Experimental procedures For
3-methyl-2-oxopentanoate, Vmax¼ 4 lmolÆmin –1 Æmg –1 (E2), and
KM ¼ 250 l M ; other activities were determined at saturating
sub-strate concentrations.
2-Oxoacid substrate
Ratio of specific activities
T acidophilum
Bovine kidney BCOADHC b Assembled
a Data from [14] b Data from [25].
Fig 3 Sedimentation coefficient distributions of the E2 protein Sedimentation velocity of lipoylated E2 protein was carried out at
40 C as described in Experimental procedures The sedimentation velocity distribution was obtained by the c(s) method [34] using the program SEDFIT , and the data were then directly fitted using the finite element solution of the Lamm equation in SEDFIT [18] to give values of the sedimentation coefficient in the buffer of sedimenta-tion [20 m M Tris ⁄ HCl (pH 8.5), 0.4 M NaCl and 1 m M phenyl-methanesulfonyl fluoride].
Trang 5(68%), respectively Unfortunately, attempts to fit
these data using the finite element solution of the
Lamm equation in sedfit [18] were unsuccessful, thus
failing to give Mr values for these proteins Using the
relationship that s (Mr)2⁄ 3 for globular proteins, and
incorporating the values of Mr¼ 6.1 · 106 and
s20,w¼ 60S for the E coli PDHC [19], an
approxi-mate value of Mr¼ 4.7 · 106 was calculated for what
is assumed to be an assembled Thermoplasma complex
of 50S
DLS is better able to deal with the polydispersity in
the sample of assembled complex Analysis of the
auto-correlation curve using the volume distribution
algo-rithm gives an Mrof 5.0 (± 0.2)· 106for the complex
at 55C, in close agreement with that estimated by
sed-imentation velocity experiments described above
To investigate the identity of the species in the
assembled complex mixture, analytical gel filtration
on SuperdexTM200 (GE Healthcare, Chalfont St Giles,
UK) was carried out Three protein species were
observed (Fig 4), one eluting in the column exclusion
volume (where Mr> 1.3· 106), one of Mr 160 000,
and a third minor peak at Mr 100 000 Whole
com-plex activity was only detected in fractions at the
exclusion volume, whereas E1 activity was detected in
the second peak and E3 in the third SDS⁄ PAGE of
protein from the exclusion volume peak revealed three
protein bands, corresponding to E1a, E1b and E2⁄ E3
(the last two proteins run together due to their similar
polypeptide size) The second peak contained
predomi-nantly excess E1 protein, whilst the third minor peak comprised E1 and E3 protein that had not been com-pletely separated In repeat experiments, uncomplexed E3 was not always detectable
Thus it is concluded that the species of Mr¼ 5 · 106
is assembled, catalytically active whole complex Given that the molar ratio of E2 : E3 (E2 polypeptide: E3 a2)
is likely to be 1 : 0.1 at a maximum (that is, the ratio
of mixed components), then an Mr of 5.0· 106 would correspond to an approximate stoichiometry (E1 a2b2: E2 polypeptide: E3 a2) of 1 : 1 : 0.1 The other major peak in the sedimentation velocity analysis (6S) is pre-sumably excess E1 The minor 19S species remains unidentified, although it should be noted that a 19.8S species was also observed in the assembled E coli PDHC, where it was concluded that it had the proper-ties of an incomplete aggregate of the component enzymes based on a trimer of the E2 chain [19]
Discussion
As described in the introduction, the genomes of aero-bic archaea contain four genes whose translated pro-tein products show sequence identities to the E1a, E1b, E2 and E3 components of the 2-oxoacid dehydro-genase multi-enzyme complexes of aerobic bacteria and eukarya [reviewed in 5] Furthermore, the genes appear to be arranged in an operon, transcriptional evidence for which has recently been gained in the hal-ophilic archaeon, H volcanii [20] The presence of
Elution volume (mL)
1
3 1
3 1
2
3 1
2
3
Enzyme activity
Complex E1 E3
45.0
31.0
21.5
66.2 97.4 116.3 210.0
M 1 2 3
E2 and E3 E1α E1 β
A
280
Fig 4 Analytical gel filtration and SDS ⁄ PAGE analysis of assembled OADHC Assembled OADHC was created by mixing the recombinant protein components in a 3 : 1 : 0.1 (E1 a2b2: E2 polypeptide: E3 a2) molar stoichiometry, followed by incubation at 55 C for 10 min in the presence of 2 m M MgCl2 and 0.2 m M TPP (A) Gel filtration of the assembled complex was carried out at 25 C as described in Experimental procedures, and the fractions were assayed for whole complex, E1 and E3 catalytic activity Active fractions are indicated by bars above the elution profile (B) Fractions were also analyzed by SDS ⁄ PAGE M, marker proteins with their Mr values given in kDa Lanes 1–3 correspond
to protein peaks 1–3 from the gel filtration; the identity of the component polypeptides is indicated alongside the gel.
Trang 6these genes is unexpected, as all the archaea possess
ferredoxin oxidoreductases that catalyze the equivalent
oxidation of the same 2-oxoacids as those used by the
bacterial and eukaryal dehydrogenase complexes;
moreover, activity of the 2-oxoacid dehydrogenase
complexes has never been detected in any archaeon
The obvious question raised by these observations is
whether or not the archaeal ‘OADHC’ genes actually
encode functional proteins that assemble into a
multi-enzyme complex The E1 component defines the
sub-strate specificity of the whole complex, and we have
previously reported heterologous expression of the
a2b2 E1 enzyme from T acidophilum and shown it to
be catalytically active with the branched-chain
2-oxo-acids and pyruvate [14] In the current paper, the E2
and E3 genes have also been successfully cloned and
expressed as soluble proteins in E coli, and their
cata-lytic activities have been demonstrated
Generation of an active E2 enzyme requires
tion of a specific lysine residue In E coli the
lipoyla-tion of its own E2 components can occur via two
routes [21,22] The endogenous pathway involves the
covalent attachment of a C-8 intermediate of fatty acid
biosynthesis to the target lysine on E2 by enzyme LipB;
subsequently, LipA catalyzes the incorporation of
sul-fur atoms to generate the lipoic acid moiety However,
if lipoic acid is supplied in the growth medium, E coli
preferentially uses its exogenous pathway, which
employs lipoate protein ligase A; this enzyme catalyzes
the adenylation of lipoic acid after uptake into the cell
and its subsequent transfer to the E2 lysine residue
Knowing that the lipoylation process can take place
across the species barrier, albeit with varying
efficien-cies [2, and references therein], lipoylation of
recombi-nantly expressed Thermoplasma E2 was tested and
optimized Up to 50% lipoylation was achieved when
the rate and level of expression in E coli was slowed
down by decreasing the growth temperature and
avoid-ing induction, whilst at the same time supplementavoid-ing
the growth medium with lipoic acid
Clearly, therefore, the E coli machinery is able to
rec-ognize the lipoyl-domain of the Thermoplasma E2
enzyme, the target lysine of which is flanked by D and V
residues, as it is in the E2 protein of the E coli OGDHC
and of the Bacillus subtilis BCOADHC (Fig 1)
How-ever, in addition to the identity of the neighbouring
residues, it is the exact positioning of the lysine in the
lipoyl domain that is fundamental to target lysine
recog-nition [23], implying that the fold of the Thermoplasma
enzyme has been conserved in this region
Whilst the Thermoplasma recombinant E2 has not
been assayed for catalytic activity in isolation,
assem-bly of the whole active complex from its individually
expressed components shows that it is indeed a func-tional enzyme This assembly process, studied by both analytical ultracentrifugation and DLS, has been dem-onstrated to involve the formation of a 24-mer E2 core, which binds E1 and E3 components to give a complex that has the same substrate specificity as that determined for the isolated E1 enzyme; namely, it is a branched-chain 2-oxoacid dehydrogenase complex that
is also active with pyruvate An E2 core that comprises
an assembly of 24 polypeptide chains is consistent with other branched-chain 2-oxoacid multi-enzyme com-plexes from bacteria and eukarya [24,25], some of which also have activity with pyruvate Additionally, the E1 subunit in those branched-chain complexes is also an a2b2 oligomer [24,25] What is particularly interesting is that the assembly of the E2 core in the Thermoplasma enzyme is temperature dependent, incu-bation to at least 40C (the temperature of the ultra-centrifugation) being required
The data in Fig 2 show that overall complex activ-ity increased linearly with the ratio of E1 : E2 mixed together, as was found with the E coli PDHC for example [26], until maximal activity was achieved at a mixing ratio of 2 : 1 However, both ultracentrifuga-tion and DLS estimate the Mr of the assembled com-plex to be around 5· 106, closely fitting the tentative conclusion of an E1(a2b2)⁄ E2 ⁄ E3(a2) stoichiometry of
1 : 1 : 0.1 These experiments indicate that not all the E1 molecules may be tightly bound to the E2 core and⁄ or that assembly might be substrate enhanced Whilst uncertainties remain over the exact subunit composition of the assembled complex, the important conclusion from our data is that the four OADHC ORFs in the archaeon T acidophilum encode the com-ponents of a functional 2-oxoacid dehydrogenase multi-enzyme complex, the first to be identified in this domain of life Thus, OADHCs were probably present
in the common ancestor to the Bacteria and Archaea, and have been retained in aerobic members of each domain The Thermoplasma enzyme possesses catalytic activity with branched-chain 2-oxo acids and pyruvate, but it remains to be established whether other archaeal OADHCs have the same or different substrate specific-ities However, whatever specificity is found, the physi-ological role of these OADHCs in the archaea remains
a mystery given the presence of active 2-oxoacid FORs that catalyze the equivalent chemical reactions
In studies of the OADHC from the halophilic archa-eon H volcanii, we could find no growth substrate that would induce the expression of OADHC activity [20], nor was any physiological defect apparent in this organism when the E3 gene was inactivated by inser-tional mutagenesis [27] Interestingly, Wanner & Soppa
Trang 7[28] have found an additional gene cluster in H
volca-nii comprising three genes that would appear to code
for OADHC E1a and E1b subunits, and an
unat-tached lipoyl domain; however, no genes for a
com-plete E2 or an E3 were present Evidence for a
function during nitrate-respirative growth on
Casami-no acids was presented, although the metabolic
sub-strate could not be identified
In conclusion, with respect to the archaeal four-gene
OADHC cluster that we have studied in this paper, it
is highly unlikely that the genes would have been
retained in a highly sophisticated and functional state
without a physiological role We suggest that
proteo-mic studies on this thermoacidophile need to be
insti-tuted to reveal this role
Experimental procedures
Materials
Bacteriological media were purchased from Sigma-Aldrich
(Poole, UK) or Fisher Scientific (Loughborough, UK)
Expression vector pET28a, E coli expression strain
BL21(DE3), BugBuster Protein Extraction Reagent and
Benzonase nuclease were purchased from Novagen-Merck
(Nottingham, UK) E coli JM109 cells, pGEM-T vector,
restriction endonucleases, T4 DNA ligase and Taq
polymer-ase were purchpolymer-ased from Promega (Southampton, UK)
Vent DNA polymerase was from New England Biolabs
(Hitchin, UK) Lipoic acid, phenylmethanesulfonyl fluoride
and antibiotics were purchased from Sigma-Aldrich
SDS⁄ PAGE molecular mass markers were from Bio-Rad
(Hemel Hempstead, UK)
Plasmids pET19b-E1a and pET28a-E1b, which,
respec-tively, coexpress the a and b subunits of the T acidophilum
E1, have been described previously [14]
Bioinformatics
The putative OADHC operon was identified in the T
aci-dophilumDSM1728 genome from the ENTREZ Nucleotide
database (http://www.3.ncbi.nlm.nih.gov) E1a: Ta1438;
E1b: Ta1437; E2: Ta1436; and E3: Ta1435
Recombinant DNA techniques
E2 and E3 gene amplification
Preparation of genomic DNA from T acidophilum strain
DSM 1728 has been described previously [29] The E2 and
E3 genes were PCR-amplified from this genomic
DNA using primers that engineered restriction sites into the
5¢- and 3¢ ends of the gene products: NdeI and XhoI for the
E2 gene, and NheI and EcoRI for E3 Oligonucleotides
were as follows (restriction sequences are underlined):
E2 forward: CGCCATATGTACGAATTCAAACTGC CAGACATAGG
E2 reverse: CCGCTCGAGTCAGATCTCGTAGAT TATAGCGTTCGG
E3 forward: CTACGAGAGCTAGCATGTACGATGC AATAATAATAGGTTC
E3 reverse: TTTAAAAATGGAATTCAATGAGAT GGT
PCR amplification was carried out using Vent polymer-ase, and A-tails were added to the products with Taq poly-merase Both genes were then separately cloned into the intermediate vector pGEM-T using T4 DNA ligase, and the clones were amplified in E coli strain JM109 grown in Luria–Bertani LB media [1% (w⁄ v) tryptone, 1% (w ⁄ v) NaCl, 0.5% (w⁄ v) yeast extract] supplemented with carben-icillin (50 lgÆmL)1) Plasmids were extracted using the BD Biosciences (Palo Alto, CA) NucleoSpin Plasmid kit, and the genes were sequenced for fidelity The E2 and E3 genes were then excised from pGEM-T, using the appropriate restriction endonucleases, purified by electrophoresis in a 0.8% (w⁄ v) agarose gel, and extracted from the gel using the Qiagen (Hilden, Germany) Qiaex II Gel Extraction kit
Construction of expression vectors pET28a-E2 and pET28a-E3
Expression vector pET28a was prepared for recombinant ligation by NdeI⁄ XhoI restriction endonuclease digestion for E2, and NheI⁄ EcoRI digestion for E3, and purified by gel electrophoresis and gel extraction as already described E2 and E3 genes were separately ligated into the vectors using T4 DNA ligase, generating the recombinant expression vec-tors pET28a-E2 and pET28a-E3 This pET28a vector thus introduced a 20- and a 23-amino acid sequence at the N-ter-mini of the E2 and E3 recombinant protein products, respec-tively, with each containing a 6-histidine tag sequence Expression and purification of OADHC components
E2 expression and purification Two different methods of expression were used Several col-onies of E coli strain BL21(DE3), freshly transformed with plasmid pET28a-E2, were picked from LB agar plates and used to inoculate 2 L of LB medium supplemented with kanamycin (30 lgÆmL)1) and 0.2 mm dl-lipoic acid Incu-bation was at 30C for 20 h in darkness, with no induction
by IPTG, after which cells were harvested by centrifugation
at 6000 g Alternatively, an overnight culture (20 mL,
A6000.6) of freshly transformed E coli BL21(DE3)pLysS cells was used to inoculate 1 L of LB medium supple-mented with kanamycin (30 lgÆmL)1) After induction with IPTG (at A6000.6) and subsequent overnight incubation at
37C, cells were harvested as described before The former
Trang 8method allows production of recombinant E2 of which up
to 50% is lipoylated, whereas in the latter method only 5%
of the E2 is lipoylated
Purification of the recombinant E2 was carried out at
25C, unless otherwise stated Samples were analyzed by
SDS⁄ PAGE on a 10% (w ⁄ v) polyacrylamide gel at each
step of the purification, and protein concentrations were
determined from A280 values Frozen cells were disrupted
by resuspension in BugBuster (5 mLÆg)1 wet cells)
supple-mented with Benzonase nuclease (1 lLÆmL)1), incubated on
ice with gentle agitation for 30 min, and centrifuged at
16 000 g for 20 min at 4C to pellet cell debris The
solu-ble cell extract was subjected to His Bind Resin
chroma-tography, and fractions containing E2 protein were pooled
and dialyzed overnight into 20 mm Tris⁄ HCl buffer,
pH 9.0, 10% (v⁄ v) glycerol The protein was then subjected
to anion exchange chromatography on an Amersham
Bio-sciences (Chalfont St Giles, UK) A¨kta FPLC system, using
a 5 mL Q-Sepharose Hi-Trap column equilibrated with
50 mm Tris⁄ HCl buffer, pH 8.5 Protein was eluted over a
0–0.8 m gradient of NaCl in the same buffer, at a flow rate
of 1 mLÆmin)1 over 60 min Fractions containing E2 were
stored at 4C in the elution buffer supplemented with
1 mm phenylmethanesulfonyl fluoride
E3 expression and purification
For expression of the E3 enzyme, a 20 mL overnight
cul-ture of transformed BL21(DE3) (A6000.6) was used to
inoculate 1 L of LB medium supplemented with kanamycin
(30 lgÆmL)1) Cells were incubated at 37C until 5 h after
the A600had reached 0.6 (with no induction by IPTG), and
were then collected by centrifugation Cell disruption was
carried out as described for the purification of recombinant
E2 The soluble cell extract was subjected to heat
precipita-tion at 65C for 5 min, and precipitated material removed
by centrifugation at 16 000 g for 20 min at 4C E3 in the
remaining soluble fraction was purified by His-Bind Resin
chromatography, dialyzed overnight into 20 mm Tris⁄ HCl
buffer, pH 8.4, and then stored at 4C
Assembly of the OADHC multi-enzyme complex
OADHC was assembled in vitro by mixing together
recom-binant E1a, E1b, E2 and E3 proteins at 55C for 0–1 h
Molar ratios to enzyme E2 varied from 0.5 to 6.0 (E1a2b2)
and 0.01–1.0 (E3a2) Each assembled complex was assayed
for overall complex activity as described below
SDS-PAGE
Analysis of protein purity and determination of polypeptide
Mr values were carried out by SDS⁄ PAGE in a resolving
gel containing 10% (w⁄ v) acrylamide [30]
Enzyme assays E1 enzymic activity was assayed spectrophotometrically by following the 2-oxoacid-dependent reduction of 2,6-dichlo-rophenolindophenol (DCPIP) at 595 nm [31] Assays were carried out at 55C in 20 mm potassium phosphate (pH 7.0), 2 mm MgCl2and 0.2 mm TPP Buffer and recom-binant E1a2b2 enzyme were pre-incubated at 55C for
10 min; 50 lm DCPIP was then added and the assay started by the addition of the 2-oxoacid substrate (pyru-vate, 2-oxoglutarate, 4-methyl-2-oxopentanoate, 3-methyl-2-oxopentanoate or 3-methyl-2-oxobutanoate)
E3 was assayed at 55C in 50 mm EPPS buffer (pH 8.0) containing 0.4 mm dihydrolipoamide and 1 mm NAD+ The reaction, in a final volume of 1 mL, was started by the addition of enzyme, and activity was monitored by measur-ing the production of NADH at 340 nm
Overall complex activity was assayed at 55C in 50 mm potassium phosphate buffer (pH 7.0) containing 2.5 mm NAD+, 1 mm MgCl2, 0.2 mm TPP, 0.13 mm CoASH and 2.6 mm cysteine-HCl [32] Buffer and assembled enzyme complex were pre-incubated at 55C for 10 min to allow binding of TPP to E1 The assay, in a final volume of
1 mL, was started by the addition of the 2-oxoacid substrate
as for the assay of E1, and OADHC activity was monitored
by measuring the production of NADH at 340 nm
Kinetic parameters were determined by the direct linear method of Eisenthal & Cornish-Bowden [33]
Gel filtration Analytical gel filtration was carried out at 25C on the Amersham Biosciences A¨kta FPLC system, using a Superdex 200 10⁄ 300 GL column Protein standards were: b-amylase (Mr¼ 200 000), alcohol dehydrogenase (150 000), BSA (66 000), carbonic anhydrase (29 000) and cytochrome c (12 400) For analysis of E3, the column was equilibrated with 20 mm sodium phosphate (pH 7.0), 0.1 m NaCl and 10% (v⁄ v) glycerol; peak fractions were assayed for E3 enzymic activity Analysis of assembled complex was carried out in 20 mm sodium phosphate (pH 7.0), 2 mm MgCl2, 0.1 m NaCl and 10% (v⁄ v) glycerol Peak fractions were assayed for E1, E3 and OADHC activity
Analytical ultracentrifugation All analytical ultracentrifugation experiments were carried out on a Beckman XL-A analytical ultracentrifuge (Beck-man-Coulter, CA) Sedimentation velocity experiments were carried out at 15 000 r.p.m., and cells were scanned every
5 min at 280 nm For sedimentation of E2, the buffer was
20 mm Tris⁄ HCl (pH 8.5), 0.4 m NaCl and 1 mm phen-ylmethanesulfonyl fluoride; for whole complex, the buffer was the same as that used in the analytical gel filtration
Trang 9analysis, 20 mm sodium phosphate (pH 7.0), 2 mm MgCl2,
100 mm NaCl and 10% (v⁄ v) glycerol The set temperature
on the centrifuge was 40C, and the solution densities were
directly measured at this temperature using an Anton-Paar
DMA 5000 high-precision density-meter Sedimentation
velocity distributions were obtained by the c(s) method [34]
using the program sedfit Data were then directly fitted
using the finite element solution of the Lamm equation in
sedfit [18] to give values of the sedimentation coefficient
and of Mr
Dynamic light scattering
All DLS measurements were performed using a Zetasizer
Nano S from Malvern Instruments Ltd (Malvern, UK)
Prior to DLS measurements, protein solutions
(1 mgÆmL)1) in 20 mm sodium phosphate buffer, pH 7.5,
10% glycerol (v⁄ v) and 0.1 m NaCl were filtered through
a 0.02 lm membrane filter (Whatman Anotop 10, Fisher
Scientific, Loughborough, UK) to remove dust particles
However, it was found necessary to filter enzyme E3
through a 0.22 lm membrane filter (Millipore, Watford,
UK) DLS measurements were carried out at 25C or
55C Mr values were derived from the measured
hydro-dynamic radii using the Protein Utilities feature of the
dispersion technology software, version 4.10, supplied
with the instrument
Mass spectrometry
Determination of the E2 polypeptide mass
A 100 pmol sample of recombinant E2 was injected on to
a MassPrep on-line desalting cartridge (2.1· 10 mm)
(Waters, Milford, MA), eluted with an increasing
acetoni-trile concentration [2 vol acetoniacetoni-trile + 98 vol aqueous
for-mic acid (1%, v⁄ v) to 98 vol acetonitrile + 2 vol aqueous
formic acid (1%, v⁄ v)] and delivered to an electrospray
ion-ization mass spectrometer (LCT, Micromass, Manchester,
UK) that had previously been calibrated using myoglobin
An envelope of multiply charged signals was obtained and
deconvoluted using maxent1 software to give the molecular
mass of the protein
Mapping the lipoylation of E2
Recombinant E2 (50 pmol) was dialyzed against 50 mm
ammonium bicarbonate on a VS membrane disc (Millipore)
for 30 min Sequencing grade, modified porcine trypsin
(Promega) (60 ng) was added and the sample incubated at
37C for 16 h A portion of the sample was diluted in 5%
(v⁄ v) formic acid and the peptides separated using an
Ulti-Mate nanoLC (LC Packings, Amsterdam, the Netherlands)
equipped with a PepMap C18 trap and column The eluant
was sprayed into a Q-Star Pulsar XL tandem mass
spec-trometer (Applied Biosystems, Foster City, CA) and ana-lyzed in Information Dependent Acquisition mode The
MS⁄ MS data generated were analyzed using the Mascot search engine (Matrix Science, London, UK), with lipoyla-tion selected as a possible lysine modificalipoyla-tion The MS⁄ MS spectrum corresponding to the modified peptide was also interpreted ‘manually’ using BioAnalyst (Applied Biosys-tems) tools
Acknowledgements
MJD and DWH thank the US Air Force Office of Sci-entific Research (Arlington, VA, USA) for generous financial support CH and MGP gratefully acknowl-edge the receipt of Postgraduate Studentships from the
UK Biotechnology and Biological Sciences Research Council and from the University of Bath, respectively
We thank Dr Jean van den Elsen (University of Bath, UK) for allowing us to use the DLS Zetasizer Nano S, and Dr Catherine Botting, BMS Mass Spectrometry and Proteomics Facility, University of St Andrews,
UK, for carrying out the MS analyses
References
1 Perham RN (1991) Domains, motifs, and linkers in 2-oxo acid dehydrogenase multienzyme complexes – a paradigm in the design of a multifunctional protein Biochemistry 30, 8501–8512
2 Perham RN (2000) Swinging arms and swinging domains in multifunctional enzymes: catalytic machines for multistep reactions Ann Rev Biochem 69, 961–1004
3 Perham RN, Jones DD, Chauhan HJ & Howard MJ (2002) Substrate channelling in 2-oxo acid dehydrogenase multienzyme complexes Biochem Soc Trans 30, 47–51
4 Izard T, Ævarsson A, Allen MD, Westphal AH, Per-ham RN, de Kok A & Hol WGJ (1999) Principles of quasi-equivalence and Euclidean geometry govern the assembly of cubic and dodecahedral cores of pyruvate dehydrogenase complexes Proc Natl Acad Sci USA 96, 1240–1245
5 Danson MJ, Lamble HJ & Hough DW (2007) Central metabolism In Archaea: Molecular and Cell Biology (Cavicchioli R, ed.) Chapter 12, pp 260–287 ASM Press, Washington, DC
6 Kerscher L & Oesterhelt D (1982) Pyruvate – ferredoxin oxidoreductase: new findings on an ancient enzyme Trends Biochem Sci 7, 371–374
7 Schut GJ, Menon AL & Adams MWW (2001) 2-Keto acid oxidoreductases from Pyrococcus furiosus and Thermococcus litoralis Methods Enzymol 331, 144–158
8 Ragsdale SW (2003) Pyruvate ferredoxin oxidoreductase and its radical intermediate Chem Rev 103, 2333–2346
Trang 109 Plaga W, Lottspeich F & Oesterhelt D (1992) Improved
purification, crystallization and primary structure of
pyruvate: ferredoxin oxidoreductase from Halobacterium
halobium Eur J Biochem 205, 391–397
10 Danson MJ, Eisenthal R, Hall S, Kessell SR &
Williams DL (1984) Dihydrolipoamide dehydrogenase
from halophilic archaebacteria Biochem J 218, 811–818
11 Smith LD, Bungard SJ, Danson MJ & Hough DW
(1987) Dihydrolipoamide dehydrogenase from the
ther-moacidophilic archaebacterium Thermoplasma
acidophi-lum Biochem Soc Trans 15, 1097–1097
12 Pratt KJ, Carles C, Carne TJ, Danson MJ & Stevenson
KJ (1989) Detection of bacterial lipoic acid: a modified
gas chromatographic – mass spectrometric procedure
Biochem J 258, 749–754
13 Jolley KA, Maddocks DG, Gyles SL, Mullan Z,
Tang S-L, Dyall-Smith ML, Hough DW & Danson MJ
(2000) 2-Oxoacid dehydrogenase multienzyme complexes
in the halophilic Archaea? Gene sequences and protein
structural predictions Microbiology 146, 1061–1069
14 Heath C, Jeffries AC, Hough DW & Danson MJ (2004)
Discovery of the catalytic function of a putative
2-oxo-acid dehydrogenase multienzyme complex in the
ther-mophilic archaeon Thermoplasma acidophilum FEBS
Lett 577, 523–527
15 Massey V (1960) Identity of diaphorase and lipoyl
dehy-drogenase Biochim Biophys Acta 37, 314–322
16 Mann S, Melero CP, Hawksley D & Leeper FJ (2004)
Inhibition of thiamin diphosphate dependent enzymes
by 3-deazathiamin diphosphate Org Biomol Chem 2,
1732–1741
17 Chauhan HJ, Domingo GJ, Jung HI & Perham RN
(2000) Sites of limited proteolysis in the pyruvate
decar-boxylase component of the pyruvate dehydrogenase
multienzyme complex of Bacillus stearothermophilus and
their role in catalysis Eur J Biochem 267, 7158–7169
18 Schuck P (1998) Sedimentation analysis of
non-interact-ing and self-associatnon-interact-ing solutes usnon-interact-ing numerical
solu-tions to the Lamm equation Biophys J 75, 1503–1512
19 Danson MJ, Hale G, Johnson P, Perham RN, Smith J
& Spragg P (1979) Molecular weight and symmetry of
the pyruvate dehydrogenase multienzyme complex of
Escherichia coli J Mol Biol 129, 603–617
20 Al-Mailem DM, Hough DW & Danson MJ (2007) The
2-oxoacid dehydrogenase complex from Haloferax
vol-canii Extremophiles, doi: 10.1007/s00792-007-0091-0
21 Miller JR, Busby RW, Jordan SW, Cheek J,
Henshaw TF, Ashley GW, Broderick JB, Cronan JE &
Marletta MA (2000) Escherichia coli LipA is a lipoyl
synthase: in vitro biosynthesis of lipoylated pyruvate
dehydrogenase complex from octanoyl-acyl carrier
protein Biochemistry 39, 15166–15178
22 Morris T, Reed K & Cronan J Jr (1995) Lipoic acid metabolism in Escherichia coli: the lplA and lipB genes define redundant pathways for ligation of lipoyl groups
to apoprotein J Bacteriol 177, 1–10
23 Wallis NG & Perham RN (1994) Structural dependence
of post-translational modification and reductive acetyla-tion of the lipoyl domain of the pyruvate dehydrogenase multienzyme complex J Mol Biol 236, 209–216
24 Ævarsson A, Seger K, Turley S, Sokatch JR & Hol WGJ (1999) Crystal structure of 2-oxoisovalerate dehydroge-nase and the architecture of 2-oxo acid dehydrogedehydroge-nase multienzyme complexes Nature Struct Biol 6, 785–792
25 Pettit FH, Yeaman SJ & Reed LJ (1978) Purification and characterisation of branched-chain a-ketoacid dehy-drogenase complex of bovine kidney Proc Natl Acad Sci USA 75, 4881–4885
26 Bates DL, Danson MJ, Hale G, Hooper EA &
Perham RN (1977) Self-assembly and catalytic activity
of the pyruvate dehydrogenase multienzyme complex of Escherichia coli Nature 268, 313–316
27 Jolley KA, Rapaport E, Hough DW, Danson MJ, Woods WG, Dyal I & Smith ML (1996) Dihydrolipoa-mide dehydrogenase from the halophilic archaeon Halo-ferax volcanii: homologous overexpression of the cloned gene J Bacteriol 178, 3044–3048
28 Wanner C & Soppa J (2002) Functional role for a 2-oxo acid dehydrogenase in the halophilic archaeon Haloferax volcanii J Bacteriol 184, 3114–3121
29 Sambrook J & Russell DW (2001) Molecular Cloning: A Laboratory Manual, 3rd edn Cold Spring Harbour Laboratory Press, Cold Spring Harbour, NY
30 Laemmli UK (1970) Cleavage of structural proteins dur-ing the assembly of the head of bacteriophage T4 Nat-ure 227, 680–685
31 Lessard IAD & Perham RN (1994) Expression in Esc-herichia coliof genes encoding the E1a and E1b subun-its of the pyruvate-dehydrogenase complex of Bacillus stearothermophilusand assembly of a functional E1 Component (a2b2) in-vitro J Biol Chem 269, 10378– 10383
32 Domingo GJ, Chauhan HJ, Lessard IAD, Fuller C & Perham RN (1999) Self-assembly and catalytic activity
of the pyruvate dehydrogenase multienzyme complex from Bacillus stearothermophilus Eur J Biochem 266, 1136–1146
33 Eisenthal R & Cornish-Bowden A (1974) The direct linear plot A new graphical procedure for estimating enzyme kinetic parameters Biochem J 139, 715–720
34 Schuck P (2000) Size distribution analysis of macromole-cules by sedimentation velocity ultracentrifugation and Lamm equation modeling Biophys J 78, 1606–1619