Original Paper |
|
|||
|
Acta Biochim Biophys |
||||
|
doi:10.1111/j.1745-7270.2008.00462.x |
In vitro observation of the molecular
interaction between NodD and its inducer naringenin as monitored by
fluorescence resonance energy transfer
Fengqing Li1,
Bihe Hou1, Lei Chen2,
Zhujun Yao2, and Guofan Hong1*
1 State Key Laboratory of Molecular Biology,
Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological
Sciences, Chinese Academy of Sciences, Shanghai 200031, China
2 State Key Laboratory of Bioorganic and Natural
Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of
Sciences, Shanghai 200032, China
Received: April 21,
2008
Accepted: May 11,
2008
The work was
supported by grants from the National Basic Research Program of China (No.
G2002CB713808) and the PanDeng Plan of China
*Corresponding
author: Tel, 86-21-54921223; Fax, 86-21-54921011; E-mail, [email protected]
At initial
stages in the Rhizobium legume symbiosis, most nodulation genes are
controlled by NodD protein and plant inducers. Some genetic studies and other
reports have suggested that NodD may be activated by its direct interaction
with plant inducers. However, there has been no molecular evidence of such an
inducing interaction. In this paper, we used fluorescence resonance energy
transfer technique to see whether such an interaction exists between NodD and
its activator, naringenin, in vitro. The tetracysteine motif
(Cys-Cys-Pro-Gly-Cys-Cys) was genetically inserted into NodD to label NodD with
4‘,5‘-bis(1,3,2-dithioarsolan-2-yl) fluorescein (FlAsH). Naringenin was
labeled with fluorescein by chemical linking. In the fluorescence resonance
energy transfer experiments in vitro, the fluorescence intensity of one
acceptor, NodD(90R6)-FlAsH, increased by 13%. This suggests that NodD may
directly interact with inducer naringenin in vitro and that the reaction
centre is likely near hinge region 1 of NodD.
Keywords NodD; naringenin; fluorescence resonance energy transfer
(FRET)
The development of nitrogen-fixing plant root nodules by rhizobia
requires an exchange of signals between the two partners, and hence the
nodulation process exhibits high host specificity and coordination. Legume
roots secrete specific flavonoid or isoflavonoid compounds that induce the
transcription of bacterial nodulation (nod, nol and noe)
genes. During this process, NodD acts as both a sensor of the plant inducer and
an activator of nod gene transcription [1]. In Rhizobium leguminosarum
biovar viciae, NodD proteins are localized in the cytoplasmic
membrane [2], where the inducing flavonoid, naringenin, also accumulates [3].
It was proposed that the un-ionized form of naringenin accumulates in the
cytoplasmic membrane and activates, in a metabolically unaltered form, the NodD
protein [3]. Direct binding of inducers to NodD has not been demonstrated due
to technical difficulties, as flavonoids stick to all kinds of materials,
including proteins (K. Recourt, unpublished data). However, results with mutant
nodD genes [4–7], analysis of inducible nod gene transcription in an
isogenic background with nodD genes from various sources [8,9], and an
enhanced binding of nod box DNA by a 35 kDa protein in the presence of flavonoid
inducers [10] together strongly suggest that NodD functions as a specific
receptor for flavonoids. Our previous study also showed that NodD interacts
directly with naringenin in vitro when naringenin was at 4.0 mM, a
concentration thousands of times higher than in physiological conditions [11].
In R. meliloti, both NodD1 and NodD2 interact with plant flavonoids with
the help of GroESL in vitro [12], but whether the interaction exists in
vivo remains unclear. Fluorescence resonance energy transfer (FRET) is a
quantum phenomenon occurring between two dye molecules. When the fluorescence
spectrum of one fluorophore (the donor) overlaps with the excitation spectrum
of another fluorophore (the acceptor), the excitation of the donor induces
fluorescence of the acceptor, while its own fluorescence decreases. The extent
of FRET is extremely sensitive to the distance between the donor and the
acceptor and is inversely proportional to the sixth power of the distance. For
FRET to occur, the donor and acceptor molecules must be within 1–10 nm of each
other. This phenomenon has been exploited to study intermolecular and
intramolecular relationships in biophysical systems and cell biology [13–17]. In this
study, FRET technique was employed to observe the interactions between NodD
and naringenin in vitro. We used naringenin (YA6006) as the fluorescence
donor and NodD-FlAsH as the fluorescence acceptor to investigate the FRET
between labeled NodD and naringenin (Fig. 1). We found that NodD may
interact with inducer naringenin in vitro and the reaction centre is
likely near hinge region 1 of NodD.
Materials and Methods
Bacterial strains and plasmids
All the bacterial strains and plasmids used in this work and their
relevant characteristics are listed in Table 1 [1824]. Escherichia
coli strains were grown at 37 ºC in Luria-Bertani (LB) medium. Rhizobia
were grown at 28 ºC in TY medium [25]. If appropriate, antibiotics were added
at the following concentrations: for E. coli, 100 mg/ml of
ampicillin, 50 mg/ml of kanamycin, and 20 mg/ml of chloramphenicol or tetracycline; for
rhizobia, 100 mg/ml of streptomycin, 25 mg/ml of kanamycin, 20 mg/ml of tetracycline and
10 mg/ml
of chloramphenicol.
Site-specific mutagenesis of
NodD by overlapping extension polymerase chain reaction (PCR)
Site-specific mutagenesis by overlapping extension was used [26].
The primers used are listed in Table 2. To construct nodD(90R6),
PCR was performed twice. Two fragments, D901 and 90D2, were obtained in the
first PCR by using pIJ1518 plasmid as the PCR template and primer pairs BnodD
and 90R6 as well as 90R6R and EnodD. In the second PCR, the annealing fragments
of D901 and 90D2 were used as the PCR template and the primer pair BnodD and
EnodD was used. The resulting products, nodD(90R6), nodD(294I6)
and nodD(C-), were obtained in the same way but with primer pairs M294
and M294R as well I303 and I303R. These mutant nodD fragments were
cloned into the plasmid pKT230 at BclI and EcoRI sites. The
plasmids harboring these mutant nodD fragments were transferred by
biparental conjugation from Escherichia coli S17-1 to the 8401 strain
containing pMP221A. The resulting clones were sequenced by the Shenggong
Campany (Shanghai, China).
FlAsH-EDT2
synthesis
FlAsH is fluorescein with As(III) substituents on the 4- and
5-positions. FlAsH is the first complementary membrane-permeant small dye. The
rigid spacing of the two arsenics in FlAsH enables it to bind with considerable
affinity and specificity to the tetracysteine motif transplanted into a
variety of proteins. The synthesis of FlAsH-EDT2
(fluorescein arsenical helix binder, bis-EDT adduct) was as described by Adams et
al [27].
Fluorescence labeling of
naringenin
Using a direct and highly regioselective Mannich reaction recently
developed in our own lab [28], we introduced an ethylene-glycol linker 4 to the
C-6 position of the A ring of naringenin without any protection (Fig. 2).
Subsequent Cu(I)-catalyzed Huisgen cycloaddition of azido compound 2 with
the acetylene-bearing fluorescein derivative 3 afforded the
corresponding fluorescently labeled naringenin derivative YA6006 (Fig. 2).
To a mixture of compounds 2 (50 mg, 0.11 mmol) and 3 (43 mg, 0.11 mmol) in t-BuOH/H2O (4 ml, V/V=1:1), we added CuSO45H2O (2 mg, 0.008 mmol) and ascorbic acid (5 mg,
0.028 mmol). The mixture was stirred at room temperature for 24 h. EtOAc (50
ml) was added to dilute the reaction. The organic phase was washed by saturated
brine, and dried over anhydrous Na2SO4. The crude product was purified by silica gel chromatography
affording pure YA6006 (75 mg, 84%). Physical data for YA6006: IR (KBr): Vmax 3385, 2944, 1752, 1707, 1597, 1501, 1459, 1369, 1250, 1182, 1110,
1024, 836 cm–1. 1H NMR
(acetone-d6, 300 MHz): d 8.12 (1H, s), 7.99 (1H, d,
J=7.2 Hz), 7.78–7.71 (2H, m), 7.35 (2H, d, J=8.7 Hz), 7.26 (1H, d, J=7.5
Hz), 7.00 (1H, d, J=2.1 Hz), 6.88 (2H, d, J=8.4 Hz), 6.79–6.77 (1H, m),
6.74–6.69 (2H, m), 6.64 (2H, br), 5.78 (1H, s), 5.35 (1H, dt, J=12.9,
3.0 Hz), 5.23 (2H, s), 4.58 (2H, t, J=4.8 Hz), 3.89 (2H, d, J=4.8
Hz), 3.83 (2H, s), 3.66–3.55 (6H, m), 3.09 (1H, dd, J=17.1, 12.9 Hz), 2.83 (2H, t, J=5.4
Hz), 2.66 (1H, dd, J=17.1, 3.0 Hz), 2.39 (3H, s). 13C NMR (acetone-d6, 75 MHz): 195.7,
170.4, 168.6, 162.6, 161.2, 160.2, 159.7, 157.8, 153.0, 152.4, 152.4, 142.6,
135.2, 130.0, 129.9, 129.2, 129.1, 128.1, 126.9, 124.7, 124.5, 124.1, 115.3,
112.7, 112.4, 112.0, 110.5, 102.5, 101.7, 100.9, 100.4, 95.7, 82.7, 78.8, 70.1,
69.1, 67.2, 61.8, 55.7, 52.6, 49.9, 42.5, 40.3, 29.2. HRMS calcd for C46H42N4O12 (MH+): 843.2872, found 843.2897.
b-galactosidase
activity assay
b-galactosidase
activity was assayed as described by Miller [29], using 10 mM naringenin as
the nodA gene inducer. Three independent experiments were performed for
each strain at the same cells density (OD600=0.4).
Preparation of protein samples
A wet weight cell pellet (about 1 g) from a culture of 300 ml Rhizobium
leguminosarum 8401 strain was resuspended in 10 ml TEG buffer [10 mM
Tris-HCl (pH 8.0), 1 mM EDTA, 5% glycerol] (100 mM NaCl). It was then sonicated
30 times (10 s per time) in an Ultrasonics W375 sonicator at 80% output power
in a salt/ice bath. Phenylmethylsulfonyl fluoride was added to a 23 mg/ml final
concentration. The lysate was centrifuged at 15,000 g for 10 min to
remove unbroken cells and cell debris. The total protein concentration was
determined by Bio-Rad protein assay kit with standard bovine serum albumin
protein in TEG buffer.
Fluorescence spectroscopy
Fluorescence measurements were carried out with an F-3010
spectrofluorometer (Hitachi, Tokyo, Japan). All measurements were performed at
room temperature in TEG buffer (100 mM NaCl). Then, 4 mM FlAsH and 6 mM naringenin (YA6006)
were employed in each set of experiments. After the dyes were added, the
samples were kept in the dark at 28 ºC for 30 min. Three independent
experiments were performed for each protein extract at the same total protein
concentration (4 mg/ml). An excitation wavelength of 514 nm and an emission
wavelength of 537 nm were employed. The energy transfer efficiency was
determined from the enhancement of the fluorescence intensity of acceptor,
FlAsH.
Results
Three NodD mutants
containing tetracysteine motif
NodD mutants were constructed by
genetically inserting the tetracysteine motif (Cys-Cys-Pro-Gly-Cys-Cys) into NodD
through overlapping extension PCR. The resulting three mutants were named as nodD(90R6),
nodD(294I6) and nodD(C-) respectively. These nodD mutants
were cloned into pKT230 plasmid, and the resulting plasmids were introduced
into 8401/pMP221A to investigate their biological activities. The ability to
activate nodA transcription of NodD(294I6), NodD(90R6), and NodD(C-)
was about 2%, 50% and 78%, respectively, of native NodD (Table 3).
NodD(294I6) had almost no biological activity. It therefore acted as a negative
control.
FlAsH labeling of
tetracysteine-tagged NodD proteins
Three tetracysteine-tagged nodD mutants were cloned into
pKNDT plasmid [21], a high copy-expressing plasmid, and transferred into the
8401 strain by biparental conjugation. The resulting strains were named 8401/pU3D(90R6), 8401/pU3D(294I6) and 8401/pU3D(C-), respectively. These tetracysteine-tagged NodD proteins were fluorescent-labeled
with FlAsH. The fluorescence intensity of FlAsH at various concentrations is
listed in Table 4. Protein extracts from 8401/pKT230 were selected as a
negative control, since 8401/pKT230 strain does not harbor the nodD gene
on the symbiotic plasmid. The enhanced FlAsH fluorescence was likely caused by
non-specific binding of FlAsH-EDT2 to endogenous cysteine-containing
proteins [30]. At the same time, the fluorescence intensity of 8401/pKT230
protein extracts did not undergo much change with concentrations of FlAsH over
4 mM,
suggesting that the endogenous cysteinecontaining proteins were saturated by 4
mM
FlAsH-EDT2. So the fluorescence intensity of 8401/pKT230 protein extracts can
be regarded as background fluorescence. Surprisingly, the protein extracts of
8401/pU3D(294I6) almost exhibited the same fluorescence intensity as that of
8401/pKT230 at all concentrations of FlAsH, suggesting that NodD(294I6) cannot
be labeled with FlAsH. This may be caused by the tetracysteine motif being packaged
inside the 3-D structure. In contrast, when the concentration of FlAsH
increased to 6 mM, the fluorescence intensity did not change. This suggested that
NodD proteins in the extracts are saturated by FlAsH. We chose 4 mM FlAsH to
investigate the FRET between NodD-FlAsH and naringenin.
Fluorescein-labeled naringenin
YA6006
Fluorescein-labeled naringenin YA6006 was chemically synthesized.
8401/pIJ1518/pMP221A was constructed to examine the ability of YA6006 to
activate nodA transcription. Fig. 3(A) shown that its biological
activity was about 23% that of native naringenin. The fluorescence intensity
of YA6006 increased directly in proportion to its concentration within 0–10 mM [Fig. 3(B)].
We chose 6 mM naringenin YA6006 to investigate the FRET between NodD-FlAsH and
naringenin.
FRET in the NodD-naringenin
complex
Naringenin (YA6006) was selected as the fluorescence donor and
NodD-FlAsH was selected as the fluorescence acceptor. Since only acceptor
emission at 537 nm was observed on excitation at 514 nm for the donor-acceptor
pair, the energy transfer efficiency was determined from the enhancement of the
acceptor emission at 537 nm, where the influence of the donor emission was
negligible. The results are summarized in Table 5. Only the fluorescence
intensity of 8401/pU3D(90R6) protein extracts was enhanced when 6 mM YA6006 was
added. After excluding the influence of endogenous cysteine-containing proteins
in the protein extracts, the fluorescence intensity of 8401/pU3D(90R6) protein extract was 257 DRFI (delta relative
fluorescence intensity) without YA6006 and 291 DRFI with 6 mM YA6006. The
fluorescence intensity of the acceptor was increased by 13%, showing a typical
energy transfer.
Discussion
NodD (303 aa) is a regulatory protein belonging to the LysR family.
NodD binds to nod-box in tetramer form [31]. A previous study involving
CbnR (295 aa), another LysR family protein, showed that active tetramer CbnR is
approximately 130 Å´70 Å´60 Å [32]. Therefore, the size of active tetramer NodD may be
similar to active tetramer CbnR, and this size is suitable for FRET occurrence.
In this study, FRET was observed in vitro only between NodD(90R6)-FlAsH
and YA6006, but not between NodD(C-)-FlAsH and YA6006. This may have happened
because FlAsH labeled on the C-terminal of NodD is too far away from the
reaction center for FRET to occur with YA6006. However, FlAsH labeled at the
hinge region 1 (aa residues 67–93) of NodD is close enough to the reaction center for FRET to occur
with YA6006. Our results suggested that NodD can directly interact with
naringenin in vitro and that the reaction center is near the hinge
region 1 of NodD. Moreover, this is consistent with the proposal that NodD does
not contain separate functional domains for DNA binding and flavonoid
interaction [1].
Protein extracts from 8401 derivative strains harboring NodD
mutants, rather than purified NodD protein, were employed in the experiments in
vitro. As a result, two problems arose: the uncertain amount of NodD
protein and the high background in fluorescence measurement. Three
tetracysteine-tagged NodD were expressed with the same plasmid and introduced
into the same strain, and the fluorescence intensity of protein extracts of
8401/pU3D(90R6) and 8401/pU3D(C-) were basically equal at
the same concentration of FlAsH (Table 4); hence, it seemed likely that
the amount of NodD proteins in the extracts of three 8401 derivative strains
was basically equal. The fluorescence intensity of samples 8401/pU3D(90R6) and 8401/pU3D(C-) minus that of the
negative control (8401/pKT230) is the specific signal of NodD-FlAsH. Using this
data analysis, the high background could be partially excluded. However,
further efforts to decrease the signal-to-noise ratio ought to be done.
Our results showed that naringenin (YA6006) has only 23% biological
activity of native naringenin and that NodD(90R6) only has 50% biological
activity of wild-type NodD. Therefore, in vitro observation of the
interaction between impaired NodD and naringenin could not explain whether such
interaction exists in vivo. To further investigate the interaction
between NodD and naringenin in vivo, NodD and naringenin with higher
biological activity should be employed in FRET measurements in vivo.
In conclusion, we labeled NodD and naringenin with suitable dyes and
investigated the FRET between these molecules in vitro. We found that
the fluorescence intensity of acceptor, NodD(90R6)-FlAsH, increased by 13%,
suggesting that NodD may directly interact with inducer naringenin in vitro and
that the reaction center is likely near the hinge region 1 of NodD.
References
1 Schlaman
HR, Okker RJ, Lugtenberg BJ. Regulation of nodulation gene expression by NodD
in rhizobia. J Bacteriol 1992, 174: 5177–5182
2 Schlaman
HR, Spaink HP, Okker RJ, Lugtenberg BJ. Subcellular localization of the nodD
gene product in Rhizobium leguminosarum. J Bacteriol 1989, 171: 4686–4693
3 Recourt
K, van Brussel AA, Driessen AJ, Lugtenberg BJ. Accumulation of a nod gene
inducer, the flavonoid naringenin, in the cytoplasmic membrane of Rhizobium
leguminosarum biovar viciae is caused by the pH-dependent hydrophobicity of
naringenin. J Bacteriol 1989, 171: 4370–4377
4 Burn
J, Rossen L, Johnston AWB. Four classes of mutations in the nodD gene of
Rhizobium leguminosarum biovar. viciae that affect its ability to
autoregulate and/or activate other nod genes in the presence of
flavonoid inducers. Genes Dev 1987, 1: 456–464
5 Horvath
B, Bachem CWB, Schell J, Kondorosi A. Host-specific regulation of nodulation
genes in Rhizobium is mediated by a plant-signal, interacting with the nodD
gene product. EMBO J 1987, 6: 841–848
6 McIver
J, Djordjevic MA, Weinman JJ, Bender GL, Rolfe BG. Extension of host range of Rhizobium
leguminosarum biovar trifolii due to point mutants in nodD
that result in alterations in regulatory function and recognition of inducer
molecules. Mol Plant Microbe Interact 1989, 2: 97–106
7 Spaink
HP, Wijffelman CA, Okker RJH, Lugtenberg BEJ. Localization of functional
regions of the Rhizobium nodD product using hybrid nodD
genes. Plant Mol Biol 1989, 12: 59–73
8 Gyorgypal
Z, Kondorosi E, Kondorosi A. Diverse signal sensitivity of NodD protein
homologs from narrow and broad host range rhizobia. Mol Plant Microbe Interact
1991, 4: 356–364
9 Spaink
HP, Wijffelman CA, Pees E, Okker RJH, Lugtenberg BJJ. Rhizobium
nodulation gene nodD as a determinant of host specificity. Nature 1987,
328: 337–339
10 Goethals
K, Van Montagu M, Holsters M. Conserved motifs in a divergent nod box of Azorhizobium
caulinodans ORS571 reveal a common structure in promoters regulated by
LysR-type proteins. Proc Natl Acad Sci U S A 1992, 89: 1646–1650
11 Hong
GF, Cao H. Evidence of the specific molecular interactions between naringenin,
NodD and nod-promoter. Chin J Biotechnol. 1993, 9: 85–88
12 Yeh KC,
Peck MC, Long SR. Luteolin and GroESL modulate in vitro activity of
NodD. J Bacteriol 2002, 184: 525–530
13 Ota N,
Hirano K, Warashina M, Andrus A, Mullah B, Hatanaka K, Taira K. Determination
of interactions between structured nucleic acids by fluorescence resonance
energy transfer (FRET): selection of target sites for functional nucleic acids.
Nucleic Acids Res 1998, 26: 735–743
14 Madeira
C, Loura LM, Prieto M, Fedorov A, Aires-Barros MR. Effect of ionic strength and
presence of serum on lipoplexes structure monitorized by FRET. BMC Biotechnol
2008, 8: 20–29
15 David
PS, Tanveer R, Port JD. FRET-detectable interactions between the ARE binding
proteins HuR and p37AUF1. RNA 2007, 13: 1453–1468
16 Ozaki
H, Iwase N, Sawai H, Kodama T, Kyogoku Y. Detection of DNA bending in a
DNA-PAP1 protein complex by fluorescence resonance energy transfer. Biochem
Biophys Res Commun 1997, 231: 553–556
17 Qiu S,
Hua YL, Yang F, Chen YZ, Luo JH. Subunit assembly of N-methyl-d-aspartate
receptors analyzed by fluorescence resonance energy transfer. J Biol Chem 2005,
280: 24923–24930
18 Downie
JA, Ma QS, Knight CD, Hombrecher G, Johnston AW. Cloning of the symbiotic
region of Rhizobium leguminosarum: the nodulation genes are between the
nitrogenase genes and the nifA-like gene. EMBO J 1983, 2: 947952
19 Simon
R, Priefer U, P黨ler
A. A broad host range mobilization system for in vivo gentetic
engineering: transposon mutagenesis in Gram-negative bacteria. Biotechnology 1983,
1: 784–791
20 Bagdasarian
M, Lurz R, R點kert
B, Franklin F, Bagdasarian MM, Fry J, Timmis KN. Specific purpose plasmid
cloning vectors. II. Broad host range, high copy number RSF1010-derived vectors
and a host-vector system. Gene 1981, 16: 237–247
21 Rossen
L, Shearman, CA, Johnston AW, Downie JA. The nodD gene of Rhizobium
leguminosarum is autoregulatory and in the presence of plant exudate
induces the nodA, B, C genes. EMBO J 1985, 4, 3369–3373
22 Feng J,
Li FQ, Li Q, Hu LH, Hong GF. Expression and purification of Rhizobium
leguminosarum NodD. Protein Expr Purif 2002, 26: 321–328
23 Spaink
HP, Okker RJH., Wijffelman CA, Tak T, Goosen-de Roo L, Pees E, van Brussel AA et
al. Symbiotic properties of rhizobia containing a flavonoid independent
hybrid NodD product. J Bacteriol 1989, 171: 4045–4053
24 Chang
WZ, Hong GF. Two functional regions were discovered within nodA
promoter. Chinese Journal of Biotechnology 1997, 13: 83–87
25 Li Q,
Feng J, Hu HL, Chen XC, Li FQ, Hong GF. A HU-like gene mutation in Rhizobium
leguminosarum bv. viciae affects the expression of nodulation genes.
Mol Microbiol 2004, 51: 861–871
26 Ho SN,
Hunt HD, Horton RM, Pullen JK, Pease LR. Site-directed mutagenesis by overlap
extension using the polymerase chain reaction. Gene 1989, 77: 51–59
27 Adams
SR, Campbell RE, Gross LA, Martin BR, Walkup GK, Yao Y, Llopis J et al.
New biarsenical ligands and tetracysteine motifs for protein labeling in
vitro and in vivo: synthesis and biological applications. J Am Chem
Soc 2002, 124: 6063–6076
28 Chen L,
Hu TS, Zhu J, Wu HM, Yao ZJ. Application of a regioselective Mannich reaction
on naringenin and its use in fluorescent labeling. Synlett 2006, 8: 1225–1229
29 Miller
J. Experiment in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor
Laboratory Press 1972
30 Stroffekova
K, Proenza C, Beam KG. The protein-labeling reagent FLASH-EDT2 binds not only to
CCXXCC motifs but also non-specifically to endogenous cysteine-rich proteins.
Pflugers Arch 2001, 442: 859–866
31 Feng J,
Li Q, Hu HL, Chen XC, Hong GF. Inactivation of the nod box distal
half-site allows tetrameric NodD to activate nodA transcription in an
inducer-independent manner. Nucleic Acids Res 2003, 31: 3143–3156
32 Muraoka
S, Okumura R, Ogawa N, Nonaka T, Miyashita K, Senda T. Crystal structure of a
full-length LysR-type transcriptional regulator, CbnR: unusual combination of
two subunit forms and molecular bases for causing and changing DNA bend. J Mol
Biol 2003, 328: 555–566