(Received for publication, June 5, 1996, and in revised form, November 19, 1996)
From the The dicluster-type ferredoxins from the
thermoacidophilic archaea such as Thermoplasma acidophilum
and Sulfolobus sp. are known to contain an unusually
long extension of unknown function in the N-terminal region. Recent
x-ray structural analysis of the Sulfolobus ferredoxin has
revealed the presence of a novel zinc center, which is coordinated by
three histidine ligand residues in the N-terminal region and one
aspartate in the ferredoxin core domain. We report here the
quantitative metal analyses together with electron paramagnetic
resonance and resonance Raman spectra of T. acidophilum
ferredoxin, demonstrating the presence of a novel zinc center in
addition to one [3Fe-4S] and one [4Fe-4S] cluster (Fe/Zn = 6.8 mol/mol). A phylogenetic tree constructed for several archaeal
monocluster and dicluster type ferredoxins suggests that the
zinc-containing ferredoxins of T. acidophilum and
Sulfolobus sp. form an independent subgroup, which is more distantly related to the ferredoxins from the hyperthermophiles than
those from the methanogenic archaea, indicating the existence of a
novel group of ferredoxins, namely, a "zinc-containing ferredoxin family" in the thermoacidophilic archaea. Inspection of the
N-terminal extension regions of the archaeal zinc-containing
ferredoxins suggested strict conservation of three histidine and one
aspartate residues as possible ligands to the novel zinc center.
Archaea (archaebacteria) represent deep and short lineages of the
universal phylogenetic tree, comprising the third independent domain of
life (1-3). They grow under various extreme environments, and contain
a variety of unique electron transfer proteins, which are currently
under extensive investigation (4-7). One of the characteristic
features in the archaeal central metabolic pathways is the involvement
of small iron-sulfur (FeS) proteins called ferredoxins in several key
steps, where NAD(P)+ usually substitutes in some bacteria
(eubacteria) and eukarya (eukaryotes) (5, 8-10). This is also the case
in several strictly aerobic archaea, including Sulfolobus
sp. strain 7, which contains nearly complete sets of the membrane-bound
proteins for aerobic respiration (7, 11-16), indicating that the
central metabolic pathways of aerobic archaea are probably more closely
related to the ferredoxin-linked fermentative pathways of anaerobic
archaea and bacteria than is the case for the aerobic bacteria and
eukarya (e.g. see Iwasaki et al. (7)).
In earlier studies, Oesterhelt and co-workers (17, 18) have established
that ferredoxins from the aerobic, thermoacidophilic archaea function
as an effective electron acceptor of the cognate coenzyme A-acylating
2-oxoacid:ferredoxin oxidoreductases, the key enzymes of the archaeal
oxidative tricarboxylic acid cycle and pyruvate oxidation (8). Later,
these ferredoxins were found to contain an unusually long N-terminal
extension region of unknown function, which was not detected in the
bacterial type ferredoxins from other sources (10, 18-22). This
probably represents a unique evolutionary event, but it has not been
pursued further.
Quite recently, T. Fujii, N. Tanaka, T. Oshima, and co-workers have
determined the 2.0-Å resolution crystal structure of a dicluster type
ferredoxin from Sulfolobus sp. strain 7 (optimal growth
conditions, pH 2.5-3 and 80 °C) (10, 11) with the R-factor of
17.9% by multiple isomorphous replacement method
(23).1 This led to an unexpected finding
that, in addition to two FeS clusters, a tightly bound zinc atom is
coordinated by His16, His19, His34,
and Asp78 in a tetragonal fashion, in the boundary between
the N-terminal extension region (containing three histidine ligand
residues) and the pseudo-2-fold symmetrical "ferredoxin
core-fold" portion (containing one aspartate ligand residue), fixing
these together.1 Thus, the Sulfolobus sp.
ferredoxin appears to be the first example that inherently contains an
additional metal, i.e. Zn2+, besides the iron
atoms. The tetragonal ligation of the novel zinc center in the
Sulfolobus sp. ferredoxin is similar to those of other
structurally unrelated zinc-containing proteins, e.g. adenosine deaminase and carbonic anhydrase in which a zinc center is
coordinated by three histidine residues and a water molecule (24,
25).
In order to investigate a possibility of the presence of an additional
metal center in other ferredoxins containing the long N-terminal
extension region, we have performed chemical and spectroscopic characterization of an another example of a bacterial type ferredoxin with a long N-terminal extension region from a thermoacidophilic archaeon, Thermoplasma acidophilum (optimal growth
conditions, pH 1.8 and 56 °C) (26). In this study, we report that
T. acidophilum ferredoxin (19) is in fact a 7Fe ferredoxin
containing a zinc center, as in the case of Sulfolobus sp.
ferredoxin. In addition, the amino acid sequence alignment and
phylogenetic tree analyses of several archaeal ferredoxins provide
strong evidence that ferredoxins with a long N-terminal extension
region derived from several thermoacidophilic archaea (10, 18-22) form
a novel separated group, namely, a "zinc-containing ferredoxin"
family. The possible zinc-binding ligand residues are strictly
conserved in the archaeal zinc-containing ferredoxins.
DEAE-Sepharose Fast Flow and Sephadex G-50 gels
were purchased from Pharmacia Biotech Inc. Water was purified by the
Milli-Q purification system (Millipore). Other chemicals used in this study were of analytical grade.
T.
acidophilum strain HO-62 cells, originally isolated from hot
sulfur springs at Owakudani solfataric field in Hakone, Japan, were
cultivated at pH 1.8 and at 56 °C as described by Yasuda et
al. (26). The Thermoplasma ferredoxin was purified
essentially as outlined by Kerscher et al. (18), using a
DEAE-Sepharose Fast Flow column (Pharmacia) connected to a Pharmacia
fast protein liquid chromatography and a Sephadex G-50 column
chromatography. Purified ferredoxin had a purity index
(A402/A280) of 0.53 (18) and showed a single band on 20% analytical polyacrylamide gel electrophoresis in the absence of sodium dodecyl sulfate.
Sulfolobus sp. strain 7 cells, originally isolated from
Beppu hot springs, Japan, were cultivated aerobically and
chemoheterotrophically at pH 2.5-3 and 75-80 °C, and the archaeal
ferredoxin was routinely purified as described previously (10, 23). The
cognate 2-oxoacid:ferredoxin oxidoreductase has been purified as
described elsewhere (10, 27).
Absorption spectra were recorded as
described previously (10). Electron paramagnetic resonance measurements
were carried out using a JEOL JEX-RE1X spectrometer equipped with an
Air Products model LTR-3 Heli-Tran cryostat system, in which
temperature was monitored with a Scientific Instruments series 5500 temperature indicator/controller, or a JEOL JES-FE3XG spectrometer
equipped with an Air Products model LTR-3-110 Heli-Tran cryostat
system, in which temperature was monitored with a Scientific
Instruments series 5500 temperature indicator/controller. Spin
concentrations were estimated by double integration, with 0.1 and 1 mM Cu-EDTA as standards.
Low temperature resonance Raman spectra were recorded at 77 K using
488.0 nm (500 mW) and 457.9 nm (200 mW) Ar+ laser
excitation as described previously (28). The sample was immersed into a
liquid nitrogen reservoir, and the scattered light near 45° to the
incident beam was collected. The spectral slit width was 4 cm Matrix-assisted laser desorption ionization-time of flight mass
spectrometry was performed by a Finnigan MAT VISION 2000 instrument at
an accelerating potential of 5.0 kV, using 2,5-dihydroxybenzoic acid as
a matrix. The average mass estimated by this method gave that of
apoferredoxin when the analysis was performed either with a holo- or
apoprotein preparation. Metal content analyses were carried out by
inductively coupled plasma atomic emission spectrometry with a Seiko
SPS 1500 VR instrument at the Tokyo Institute of Technology and a
Jobin-Yvon JY 38S instrument at Rigaku Ltd.
The phylogenetic trees were
constructed by the parsimony and neighbor-joining methods using the
phylogenetic analysis program package, Phylip 3.5c (29). All amino acid
sequence data of archaeal and hyperthermophile ferredoxins used for
phylogenetic calculations in this study were obtained from the PIR and
EMBL data bank, except for that from Sulfolobus sp. strain 7 (accession no. D78179[GenBank]).2 The distance
matrix was estimated using the Dayhoff PAM matrix for the
neighbor-joining method. Reliability of the phylogenetic trees thus
constructed was estimated through the bootstrap analysis with 100 replicates.
The x-ray crystal structure of the dicluster
ferredoxin from Sulfolobus sp. strain 7 at 2.0-Å resolution
has demonstrated the presence of a novel zinc center which is ligated
by His16, His19, His34, and
Asp78 in a tetragonal coordination, in the boundary between
the N-terminal extension and the FeS cluster-binding core regions,
fixing these together.1 The metal content analysis of the
isolated Sulfolobus sp. ferredoxin in solution also showed
the presence of a tightly bound zinc atom, in a ratio of ~6.4-6.9
Fe/Zn (mol/mol; data not shown). The zinc atom could not be removed by
dialysis against buffer containing 5 mM EDTA. These data
suggest that the Sulfolobus sp. ferredoxin inherently
contains an additional metal binding site specific to zinc atom, beside
two FeS clusters. For comparison, we have also purified a 7Fe
ferredoxin from a thermophilic bacterium Thermus thermophilus strain HB8 (30) using the same buffer system: The purified ferredoxin gave an average mass (M + H)1+ of
~8684 by matrix-assisted laser desorption ionization-time of flight
mass spectrometry, being consistent with the value estimated from the
primary structure (8687 Da) (30), and contained no zinc
atom.3
In order to investigate whether the zinc center is unique to the
Sulfolobus sp. ferredoxin, another bacterial-type ferredoxin with a long N-terminal extension has been purified to an
electrophoretically homogeneous state from a thermoacidophilic
archaeon, T. acidophilum strain HO-62 (26), according to the
guidelines of Kerscher et al. (18). The purified ferredoxin
of T. acidophilum strain HO-62 had a purity index
(A402/A280) of 0.53, and
showed the optical properties identical to those reported previously by
Kerscher et al. (18) (data not shown). The N-terminal 15 amino acid residues of the purified ferredoxin (VKLEELDFKPKPIDE) were
completely identical to that reported previously by Wakabayashi
et al. (19). Moreover, the matrix-assisted laser desorption
ionization-time of flight mass spectrometry of the purified protein
gave an average mass (M + H)1+ of ~15960 (data not
shown), which is significantly close to the value estimated from the
primary structure of T. acidophilum apoferredoxin (15963 Da)
(19). These data suggest that the ferredoxin purified from T. acidophilum strain HO-62 is essentially identical to that previously reported by Kerscher and co-workers (18, 19) and contains a
long N-terminal extension region. Metal content analysis by inductively
coupled plasma atomic emission spectrometry showed that the purified
ferredoxin contained a tightly bound zinc atom (6.8 Fe/Zn (mol/mol)).
The following metals were not detected: molybdenum, cobalt, nickel, and
copper. In conjunction with the primary structural evidence suggesting
two sets of FeS cluster-binding motifs (19), this indicates that
T. acidophilum ferredoxin contains a unique zinc center
beside two FeS clusters (see below).
The properties of the FeS clusters in T. acidophilum
ferredoxin were investigated by X-band EPR spectroscopy (Fig.
1). The EPR spectrum at 8.2 K of the isolated T. acidophilum ferredoxin elicited a sharp g = 2.02 signal
(~1.1 spin/mol), whose EPR lineshape and temperature dependence (data
not shown) are characteristic of a [3Fe-4S]1+ cluster
(Fig. 1, trace A). Upon reduction by excess sodium
dithionite at pH 6.8, the relative intensity of this EPR signal
decreased by ~85%, indicating the partial reduction of the
[3Fe-4S] cluster, but no additional EPR signal could be detected
(data not shown). Because the decrease of the
A402 nm peak in the optical spectrum of the
purified protein was ~20% under the conditions, these data suggest
the presence of a lower potential [4Fe-4S] cluster that is not
readily reduced by dithionite, in addition to a [3Fe-4S] cluster.
This was further supported by the EPR spectra at 8.2 K of T. acidophilum ferredoxin anaerobically reduced by excess dithionite
at pH 9.3 (Fig. 1, traces B and C). The
[3Fe-4S] cluster was fully reduced, thus giving rise to a very broad
low field resonance at g ~ 11, which is characteristic of the
reduced S = 2 [3Fe-4S]0 cluster (Fig. 1,
trace C). In addition, a rhombic EPR signal at
gz,y,x = 2.06, 1.94, and 1.88 attributed to the reduced
S = 1/2 [4Fe-4S]1+ cluster, was
detected together with additional wings on the high and low filed sides
of the main EPR signal (g = 2.13 and 1.78) which are due to the
magnetic interactions with the reduced S = 2 [3Fe-4S]0 cluster (Fig. 1, trace B). These
properties are characteristic of those of conventional reduced 7Fe
ferredoxins.
The properties of the oxidized FeS clusters of T. acidophilum ferredoxin were further investigated by the low
temperature resonance Raman spectroscopy, which was performed at 77 K
using 488.0 nm and 457.9 nm Ar+ ion laser excitation (Fig.
2, traces A and B). On the basis
of extensive assignments by Johnson and co-workers (31, 32), two weak
bands at 334 cm
As in the cases reported for the Sulfolobus and
Desulfolofulbus 7Fe ferredoxins (10, 21), anaerobic
reduction of T. acidophilum ferredoxin with the purified
Sulfolobus sp. 2-oxoacid:ferredoxin oxidoreductase (27)
suggested that, of two FeS clusters, only the higher potential
[3Fe-4S] cluster was reduced enzymatically, while the lower potential
[4Fe-4S] cluster was not (data not
shown).4
Taken together, the chemical and spectroscopic analyses suggest the
presence of one zinc center, one higher potential [3Fe-4S] cluster,
and one lower potential [4Fe-4S] cluster in T. acidophilum ferredoxin. For reasons discussed below, the zinc center probably represents an inherent metal rather than an adventitiously bound atom.
Of all amino acid sequences of
the bacterial type ferredoxins reported to date, only four ferredoxins,
all of which have been derived from thermoacidophilic archaea, are
known to possess a long N-terminal extension region (10, 18-22). Fig.
3 shows the amino acid sequence alignment of the
N-terminal extension region of these ferredoxins. The N-terminal
extension region of archaeal ferredoxins from Sulfolobus sp.
strain 7 (10),2 S. acidocaldarius (20, 22), and
Desulfolofulbus ambivalens (21) are considerably homologous
to each other, leaving that from T. acidophilum to be less
homologous (19). We have found, however, by careful amino acid sequence
alignment that all four ligand residues which have been identified as
ligands to the tetragonal zinc center in Sulfolobus sp.
ferredoxin (His16, His19, His34,
and Asp78)1 are strictly conserved among these
ferredoxins (Fig. 3). Thus, these archaeal ferredoxins show the overall
amino acid sequence homology. Since both the purified
Sulfolobus and Thermoplasma dicluster ferredoxins
contained a stoichiometric amount of a zinc atom as mentioned in the
preceding section, we suggest that a bacterial-type ferredoxin
containing a long N-terminal extension region with the consensus three
histidine motif (Fig. 3) would bind a zinc ion, beside one or two
conventional FeS clusters. A tentative model of possible metal-binding
ligand residues of T. acidophilum zinc-containing 7Fe
ferredoxin is schematically presented in Fig. 4, in
light of the three-dimensional structure of the Sulfolobus
sp. ferredoxin.1 It should be remembered that this is a
putative schematic drawing, and the detailed characterization of the
coordination environments of the zinc center of T. acidophilum ferredoxin would require the extended x-ray absorption
fine structure or the x-ray crystallography analyses. However, at this
stage limited availability of the archaeal ferredoxin, due largely to
extremely slow growth and poor yield of T. acidophilum
strain HO-62 cells (26), has made such analyses practically
difficult.
The bacterial type ferredoxins investigated so far
possess conserved FeS cluster binding motifs and overall homologies at least at the primary structural level (33-35). Using the amino acid
sequences of eight archaeal monocluster and dicluster ferredoxins (from
Sulfolobus sp. strain 7 (accession nos. PC2290 and
D78179[GenBank]),2 S. acidocaldarius (accession no.
P00219[GenBank]), T. acidophilum (accession no. P00218[GenBank]),
Methanococcus thermolithotrophicus (accession no. P21305[GenBank]),
Methanosarcina thermoaceticum (barkeri)
(accession no. P00202[GenBank]), Methanosarcina thermophila (accession no. A42960[GenBank]), Thermococcus litoralis (accession no. P29604[GenBank]), and Pyrococcus furiosus (accession no. X79502[GenBank])) and a 4Fe ferredoxin from the hyperthermophilic bacterium
Thermotoga maritima (accession no. X82178[GenBank]), putative
phylogenetic relations were estimated, taking the positions of
conserved cysteine ligand residues and charged and hydrophobic amino
acid residues into account (Fig. 5). Some of these
ferredoxins have been shown to serve at least as an electron acceptor
of the cognate 2-oxoacid:ferredoxin oxidoreductases (10, 18, 21, 27,
36, 37). Both distant geometry and parsimony analyses of the amino acid
sequence alignment gave essentially identical topologies of the
trees.5
The three-dimensional structures of all bacterial-type ferredoxins so
far determined exhibit a remarkable pseudo-2-fold symmetry, indicating that a putative initial gene duplication event might have
taken place in the earliest stage of the molecular evolution of a
primordial bacterial type ferredoxin (10, 38, 39). The basic concepts
of this have been generally accepted, and the extreme cases may be
emphasized for the unusual polyferredoxins typically found in certain
methanogenic archaea (40) (not considered in the phylogenetic tree
shown in Fig. 5). The pseudo-2-fold symmetry of the
three-dimensional structures of both the archaeal and bacterial ferredoxins suggests that a putative initial gene duplication event
might have occurred long before the divergence of the
Archaea and the Bacteria domains (1, 2). The
archaeal ferredoxin tree shown in Fig. 5 clearly suggests their
monophyletic origin, and, in a supposed first evolutionary event,
divergence of the monocluster and dicluster ferredoxins occurred prior
to the divergence of the archaeal genera.
Intriguingly, the ferredoxins containing a long N-terminal extension
region and a zinc center phylogenetically form a separate group, and
they are supposed to be evolved from a common ancestor. Thus, the
phylogenetic analysis clearly suggests that they can be classified as a
novel, separate group, namely, the "zinc-containing ferredoxin"
family. This is supported from our findings that the overall topologies
of the calculated trees are essentially identical whether or not
information of the long N-terminal extension and the central additional
loop regions of the zinc-containing ferredoxins is taken into account
for the analysis (data not shown).
Another interesting feature of the archaeal ferredoxin tree (Fig. 5) is
that it unexpectedly well reflects the differences in the growth
conditions of the archaeal species, rather than their phylogenetic
relationships based on the universal phylogenetic tree calculated on
the basis of the 16 S rRNA base sequences (2). Thus, ferredoxins from
hyperthermophiles, methanogens, and thermoacidophiles are clustered
together in separate groups as the monocluster type, the dicluster
type, and the novel zinc-containing dicluster type, respectively.
Moreover, the zinc-containing ferredoxins have been found only in the
"fast clock" archaea (such as Sulfolobales and Thermoplasma) (3), while ferredoxins from the "slow
clock" hyperthermophilic archaea and bacteria (3) are primarily the monocluster ferredoxins. Given the physiological importance of ferredoxins in the archaeal central metabolisms (5, 8-10, 18), these
observations seem to be somewhat meaningful (e.g. see
Darimont and Sterner (39)), although further discussion should await comparison at the base sequence level in order to preclude a
possibility of lateral gene transfer
event.6 One possibility is that the
molecular evolution of the archaeal ferredoxins might have largely been
affected by environmental pressure which should have made considerable
influences on the central and energy metabolisms as well as the growth
of the cells. The closer relationship between the zinc-containing
ferredoxins of thermoacidophilic archaea and the dicluster ferredoxins
of methanogens in the archaeal ferredoxin tree also raises an
interesting question of whether methanogens contain a putative
zinc-binding polypeptide of ~50-100 amino acids long, which could
potentially interact with the cognate ferredoxin. However, preliminary
search against the complete genome sequence of the methanogenic
archaeon Methanococcus jannaschii (41) failed to identify
the presence of such a sequence.3
A specific role of the novel zinc center of the zinc-containing
ferredoxin family is currently unclear, besides its obvious structural
contribution. It is very unlikely that ferredoxin itself serves as a
zinc sensor or a zinc storage protein in vivo, because the
bound zinc center is not released upon dialysis against buffer containing 5 mM EDTA. Nevertheless, formation of a mature
ferredoxin molecule should be correlated to intracellular levels of
both Zn2+ and Fe2+ ions at least at the
translation level, thus raising an interesting possibility that the
ferredoxin-dependent central metabolism of the fast clock
archaea (17, 18) could be potentially regulated by the availability of
zinc. In this connection, a zinc-containing ferredoxin identified in
this study may be useful as a simple model for more complex,
mixed-metal metalloenzymes, to investigate how different kinds of metal
ions are inserted into distinct metal binding sites of the same
apoprotein to give mature protein conformation. Moreover, since the
redox partners of the zinc-containing ferredoxins are known (10,
18-22), it will be tempting to construct mutant proteins with modified
N-terminal extension region by protein engineering and to determine the
effects on the protein stability and redox activity.
Since the first discovery of a bacterial type
ferredoxin by Mortenson et al. in 1962 (42), we have
recognized for the first time the existence of a novel ferredoxin
family inherently containing a zinc center besides conventional FeS
clusters, in several thermoacidophilic archaea. These ferredoxins,
namely, zinc-containing ferredoxins, can be characterized by a long
N-terminal extension region with a consensus of three histidine
residues, which probably function as parts of ligands to a zinc center.
Chemical and spectroscopic analyses of one such example from T. acidophilum have shown that the purified ferredoxin contains one
zinc center, one [3Fe-4S] cluster, and one [4Fe-4S] cluster, as in
the case of Sulfolobus sp.
ferredoxin.7 The presence of a unique zinc
center in these ferredoxins might represent a unique evolutionary event
in early thermoacidophilic archaea.
A part of this work was initiated by interest
in the preliminary x-ray structural work of the Sulfolobus
sp. ferredoxin initiated at the Department of Life Science, Tokyo
Institute of Technology, by Tomomi Fujii and co-workers. We gratefully
acknowledge Tomomi Fujii (Kyoto University), Takayoshi Wakagi, and
Nobuo Tanaka (Tokyo Institute of Technology) for valuable discussion on
the structural aspects of the Sulfolobus sp. dicluster
ferredoxin. We also thank Yasuhiro Isogai and Tetsutaro Iizuka
(Institute of Physical and Chemical Research) for their kind help in
initial EPR analysis, Hidenori Ikezawa (Finnigan MAT Instruments, Inc.)
for the matrix-assisted laser desorption ionization-time of flight mass
spectrometry, and Naoki Wakiya (Tokyo Institute of Technology) and
Hidehiro Daidohji (Rigaku Ltd.) for the metal content analyses by the
inductively coupled plasma atomic emission spectrometry.
Department of Biochemistry and Molecular
Biology,
Department of Chemistry,
Department of Molecular Biology,
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
FOOTNOTES
Acknowledgments
REFERENCES
Materials
1, and a multiscan averaging technique was
employed.
Purification and Characterization of T. acidophilum Dicluster
Ferredoxin
Fig. 1.
The low temperature EPR spectra of purified
T. acidophilum ferredoxin in the air-oxidized (trace
A) and the dithionite-reduced states at pH 9.3 (traces
B and C). The sample was dissolved in 600 mM N-cyclohexyl-3-aminopropanesulfonic acid
(CAPS) buffer, pH 9.3. Instrument settings for the EPR spectroscopy:
temperature, 8.2 K; microwave power, 0.4 mW; modulation amplitude, 0.63 millitesla; the g values are indicated.
[View Larger Version of this Image (13K GIF file)]
1 and ~358 cm
1, whose
signal relative intensities were enhanced upon 457.9 nm Ar+
ion laser excitation (Fig. 2, trace B), were assigned to be
associated primarily with Fe-S bridging mode and Fe-S terminal mode,
respectively, of a conventional biological [4Fe-4S]2+
cluster with the D2d structure. On the other
hand, the [3Fe-4S]1+ cluster exhibited three bands
associated primarily with Fe-S bridging modes at 265 cm
1,
~288 cm
1, and 347 cm
1, and at least two
bands associated primarily with Fe-S terminal modes at 369 cm
1 and 387 cm
1 (31, 32). Thus, although
7Fe ferredoxin of T. acidophilum strain HO-62 contains a
zinc center, the EPR and resonance Raman spectral properties are very
similar to those of conventional 7Fe ferredoxins reported earlier (31),
clearly demonstrating the existence of common [3Fe-4S] and [4Fe-4S]
core structures.
Fig. 2.
Resonance Raman spectra at 77 K of the
isolated Thermoplasma ferredoxin recorded using 488.0 nm
(trace A) and 457.9 nm (trace B)
Ar+ ion laser excitation. The spectral bandwidth was 4 cm1.
[View Larger Version of this Image (23K GIF file)]
Fig. 3.
Amino acid sequence alignment of the
N-terminal extension regions of ferredoxins from several
thermoacidophilic archaea. The strictly conserved cysteine ligand
residues to the FeS clusters (*) are boxed, and possible
ligand residues to the zinc center (+) are darkly shaded.
The consensus residues identified in this region are also
indicated. Key: Sulfolobus sp. strain 7 ferredoxin (accession nos. PC 2290 and D78179[GenBank]); S. acidocaldarius strain DSM 639 ferredoxin (20, 22); D. ambivalens ferredoxin (21); T. acidophilum ferredoxin (19) (T. Iwasaki and T. Oshima, unpublished result). Although not shown in the figure, both
Sulfolobus sp. and S. acidocaldarius ferredoxins
contain a monomethyl-lysine residue at position 29 (20).
[View Larger Version of this Image (22K GIF file)]
Fig. 4.
A schematic model of possible ligand residues
of a zinc-containing 7Fe ferredoxin from T. acidophilum
strain HO-62. This putative model was deduced on the bases
of the sequence alignments of the N-terminal extension regions (Fig. 3)
and the three-dimensional structure of Sulfolobus sp.
ferredoxin (accession no. 1XER). Possible ligand residues to the metal
centers are indicated by bold letters with a residue
number.
[View Larger Version of this Image (15K GIF file)]
Fig. 5.
The unrooted phylogenetic tree of archaeal
ferredoxins. All amino acid sequence data shown in the figure were
obtained as described under "Experimental Procedures." Comparison
of sequence pairs and multiple sequence alignments of ferredoxins from
8 archaeal and 1 hyperthermophilic bacterial species were performed as
described in the text, using the following data set:
Sulfolobus sp. strain 7, accession nos. PC2290 and D78179[GenBank];
S. acidocaldarius, accession no. P00219[GenBank]; T. acidophilum, accession no. P00218[GenBank]; M. thermolithotrophicus, accession no. P21305[GenBank]; M. thermoaceticum (barkeri), accession no. P00202[GenBank];
M. thermophila, accession no. A42960[GenBank]; T. litoralis, accession no. P29604[GenBank]; P. furiosus, accession
no. X79502[GenBank]; and T. maritima, accession no. X82178[GenBank]. The
resulting ferredoxin tree was constructed by the neighbor-joining
method. The identical topological tree was obtained by the parsimony
analysis (data not shown). Bootstrap values for selected nodes are
indicated as "(a value estimated by distance analysis)/(a value
estimated by persimony analysis)" in the figure. The zinc-containing
dicluster ferredoxins are darkly shaded, the other dicluster
ferredoxins are lightly shaded, and the monocluster
ferredoxins are not marked.
[View Larger Version of this Image (21K GIF file)]
*
This investigation was supported in part by Grants-in-Aid
from the Ministry of Education, Science, Sports and Culture of Japan. The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
§
To whom all correspondence should be addressed: Dept. of
Biochemistry and Molecular Biology, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo 113, Japan. Tel.: 81-3-3822-2131 (ext. 5216);
Fax: 81-3-5685-3054.
1
Fujii, T., Hata, Y., Wakagi, T., Tanaka, N., and
Oshima, T. (1996) Nat. Struct. Biol. 3, 834-837.
2
Wakagi, T., Fujii, T., and Oshima, T. (1996)
Biochem. Biophys. Res. Commun. 225, 489-493
(accession no. D78179[GenBank]).
3
T. Iwasaki and T. Oshima, unpublished
results.
4
The reduction level at
A402 nm of T. acidophilum ferredoxin
by the purified Sulfolobus 2-oxoacid:ferredoxin
oxidoreductase (27) was ~20% under the steady-state conditions in
the presence of 4 mM 2-oxoacid and 100 µM
coenzyme A for 60 min at pH 6.8 and at 50 °C. No EPR signal for the
reduced [4Fe-4S]1+ cluster was detected under these
conditions. The same results have been obtained with
Sulfolobus sp. ferredoxin (10).
5
T. Suzuki, T. Iwasaki, and T. Oshima,
unpublished data. The details of the amino acid sequence alignment of
archaeal ferredoxins are available upon request.
6
For instance, a putative lateral gene transfer
event from halophilic cyanobacterial origin has been well documented
for the case of Halobacterium salinarium [2Fe-2S]
ferredoxin (43).
7
T. Iwasaki, T. Imai, A. Urushiyama, D. Ohmori,
and T. Oshima, manuscript in preparation.
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.