While attempting to obtain large crystals of nitrogenase CrFe protein, brown crystals and brick red crystals were simultaneously or independently obtained from CrFe protein preparation, which was partially purified from a mutant UW3 of Azotobacter vinelandii Lipmann grown on Mo-, ammonia-free but Cr-containing medium. SDS-PAGE and anoxic native-PAGE analysis consistently showed that the protein of the brown crystal was mainly composed of subunits (~60 kD) similar to those of Av1 (MoFe protein), while the protein of the brick red crystal was composed of ~20 kD subunits. And only the larger subunits rather than the smaller ones were detectable by Western blot to the antibody of Av1. Comparing with the large subunits, the amount of the small subunits in the partially purified CrFe protein solution was much smaller, indicating that the protein composed of the smaller subunits was one of contamination proteins for CrFe protein. Detection by 3, 5-diaminobenzoic acid of native-PAGE gels showed that the proteins forming the brick red crystal and the brown crystal were two kinds of iron-containing proteins with different electrophoretic mobility on the gel. The analysis of matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) proved that the protein forming the brick red crystal was bacterioferritin of A. vinelandii (AvBF). X-ray diffraction to 2.34 Å resolution showed that the crystal belonged to space group H3, with unit-cell parameters a = 124.965 Å, b=124.965 Å and c = 287.406 Å. The detailed structural analysis published in the near future has confirmed that the brick red crystal is that of 24-meric bacterioferritin.
全 文 :Received 27 Apr. 2004 Accepted 4 Aug. 2004
Supported by the State Key Basic Research and Developmental Plan of China (001CB1089-06) and the National Natural Science Foundation
of China (30270296).
* Author for correspondence. E-mail:
http://www.chineseplantscience.com
Acta Botanica Sinica
植 物 学 报 2004, 46 (11): 1331-1337
Crystal Growth and Characterization of Residual Bacterioferritin
in Partially Purified Nitrogenase CrFe Protein Solution from
a Mutant UW3 of Azotobacter vinelandii
ZHAO Jian-Feng1, LIU He-Li2, ZHOU Hui-Na1, WANG Zhi-Ping3, ZHAO Ying1, BIAN Shao-Min1,
LI Shu-Xing2, BI Ru-Chang2, HUANG Ju-Fu1*
(1. Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany,
The Chinese Academy of Sciences, Beijing 100093, China;
2. Institute of Biophysics, The Chinese Academy of Sciences, Beijing 100101, China;
3. Institute of Nuclear Agriculture, Zhejiang University, Hangzhou 310029, China)
Abstract: While attempting to obtain large crystals of nitrogenase CrFe protein, brown crystals and
brick red crystals were simultaneously or independently obtained from CrFe protein preparation, which was
partially purified from a mutant UW3 of Azotobacter vinelandii Lipmann grown on Mo-, ammonia-free but Cr-
containing medium. SDS-PAGE and anoxic native-PAGE analysis consistently showed that the protein of
the brown crystal was mainly composed of subunits (~60 kD) similar to those of Av1 (MoFe protein), while
the protein of the brick red crystal was composed of ~20 kD subunits. And only the larger subunits rather
than the smaller ones were detectable by Western blot to the antibody of Av1. Comparing with the large
subunits, the amount of the small subunits in the partially purified CrFe protein solution was much smaller,
indicating that the protein composed of the smaller subunits was one of contamination proteins for CrFe
protein. Detection by 3, 5-diaminobenzoic acid of native-PAGE gels showed that the proteins forming the
brick red crystal and the brown crystal were two kinds of iron-containing proteins with different electro-
phoretic mobility on the gel. The analysis of matrix-assisted laser desorption ionization time-of-flight
(MALDI-TOF) mass spectrometry (MS) proved that the protein forming the brick red crystal was
bacterioferritin of A. vinelandii (AvBF). X-ray diffraction to 2.34 Å resolution showed that the crystal
belonged to space group H3, with unit-cell parameters a = 124.965 Å, b=124.965 Å and c = 287.406 Å. The
detailed structural analysis published in the near future has confirmed that the brick red crystal is that of
24-meric bacterioferritin.
Key words: crystal growth and characterization; bacterioferritin; partially purified CrFe protein solution;
mutant strain UW3 of Azotobacter vinelandii
The biological reduction of molecular nitrogen (N2) is
one of the most fundamental processes in nature. This pro-
cess is catalyzed by nitrogenase which consists of two
protein components (components Ⅰ andⅡ). During the
separation of nitrogenase from Azotobacter vinelandii, a
nitrogen-fixing bacterium, some important proteins includ-
ing bacterioferritin had been purified and characterized. Like
the bacterioferritins isolated from Escherichia coli and
Pseudomonas aeruginosa, etc., bacterioferritin from A.
vinelandii (AvBF) was shown to be a ferritin containing
haem in addition to the non-heme iron core (Andrews et
al., 1991; Wai et al., 1995). It was said that AvBF is a haem-
containing multisubunit protein that performs the same
functions of iron storage and iron detoxification as animal
ferritins being studied more extensively (Stiefel and Watt,
1979). And now it is proposed that the ferritins’ dual func-
tions of storing iron and detoxification of iron or protec-
tion against oxygen are likely to be bacterioferritins’ pri-
mary function (Carrondo, 2003). As a bacterioferritin,
AvBF’s significant difference from other ferritins, such as
the presence of 12-heme groups and structurally disor-
dered phospho-hydroxy mineral core, made great sense
to elucidate how it functions. It is obvious that X-ray
diffraction of AvBF crystal would help us to understand
its structure-function relationship. However, its crystal
structure has not been reported for a long time, though its
preliminary X-ray crystallographic studies were described
to show that the AvBF crystals belong to the cubic system,
space group I432, with cell dimension 230 Å (Zhao et al.,
1984).
Acta Botanica Sinica 植物学报 Vol.46 No.11 20041332
Mutant strain UW3 from A. vinelandii was unable to fix
N2 in the presence of Mo (Nif-) but able to grow either
under condition of Mo starvation or in a Mo-deficient me-
dium containing Re or V (Bishop et al., 1982). Two new
nitrogenase components Ⅰ(CrFe protein and MnFe protein)
have been partially purified from the mutant grown on Mo-
deficient medium containing Cr and Mn, respectively
(Huang et al., 2001; 2002). While attempting to crystallize
CrFe protein, large AvBF crystals were accidentally obtained.
X-ray diffraction analysis shows that these crystals be-
long to space group H3, rather than space group I432 as
reported before (Zhao et al., 1984). The present studies
focus on the growth and characterization of AvBF crystals
as well as the identification of them from CrFe protein
crystals.
1 Materials and Methods
1.1 Bacterial strains and growth conditions
The mutant strain UW3 of Azotobacter vinelandii
Lipmann was kindly provided by Prof. Bishop. Its growth
was carried out according to Bishop et al. (1982) except the
addition of Na2CrO4 (Huang et al., 2002). Precautions were
taken to minimize contamination by Mo as Mo-deficient
media were prepared. The chemicals of analytic grade and
redistilled water were used. All solutions of inorganic chemi-
cals were passed through a column with active carbon in
order to decrease the residual Mo content in the chemicals
(Schneider et al., 1991).
1.2 Purification and Crystallization of nitrogenase CrFe
protein
The purification of the protein was anaerobically per-
formed according to Huang et al. (2002).
The crystals were obtained by the liquid/liquid diffu-
sion method in a capillary. In a plexi-glass box fully filled
with Ar, 15 mL precipitant solutions and the same volume
of protein solution were carefully added into the capillary.
The mouth of the capillary was greased. The capillaries
were put into a large glass tube with a plastic stopper.
The sealed tubes were put into a glass bottle with its
tight cap, then stood for several months on the base-
ment at 20 ℃.
All operations were performed under Ar atmosphere.
All of the used solutions were rigorously degassed and
filled with Ar before and after the addition of 0.3 mg/mL and
1.6 mg/mL DT for protein purification and crystallization,
respectively.
Determination of protein concentration and anaerobic
absorption spectrum were carried out by the method of
Huang et al. (2001).
Anoxic native-PAGE, SDS-PAGE and Western blott were
performed according to Shah and Brill (1973), Paustian et
al. (1990) and Luo et al. (1995), respectively.
1.3 Mass spectrometry
The ~ 20 kD proteins from brick red crystals, in advance
run on SDS-PAGE gel and developed with Coomassie bril-
liant blue staining, were excised, sequentially digested with
trypsin in gel, and then applied to matrix-assisted laser de-
sorption ionization time-of-flight (MALDI-TOF) mass spec-
trometry (MS). Peptide mass fingerprints (PMFs) were ac-
quired in the reflectron mode on a Bruker Autoflex MALDI-
TOF mass spectrometer (Bruker Daltonics) using an accel-
eration voltage of 20 kV. Monoisotopic peptide masses
obtained from MALDI-TOF were queried against entries
for NCBInr protein databases using a protein search
program, Mascot (Matrix Science Ltd., London). The mass
accuracy of 50 ppm in the parent ion mass and 0.1 Da in the
product ion mass was set as search parameters.
1.4 Single crystal X-ray diffraction
The single crystal X-ray diffraction experiments were
performed at the Beijing Synchrotron Radiation Facility
(Beijing, China) beamline 3W1A at a wavelength of 0.980 1
Å using a MAR345 (MAR Research, Hamburg) image plate
detector. No good cryocondition was obtained after lots of
trials, most of which resulted in large mosaicities. As a result,
all the data were collected at room temperature, leading to a
relatively small mosaicity of 0.28°. A large crystal-to-detec-
tor distance 300 mm and a small oscillation 1.0° were
adopted to reduce the overlap of reflections.
2 Results and Discussion
2.1 Formation of brick red crystals in partially purified
CrFe protein solution
The precipitant composition is important for formation
and subsequent growth of protein crystals. In order to ob-
tain crystals of CrFe protein suitable for X-ray diffraction,
the optimal composition of precipitant has been screened.
The crystal number decreased and crystal size increased
with both the decrease of PEG and NaCl concentrations
under the same conditions (Table 1). Under the given
conditions, brick red crystals and brown crystals were si-
multaneously observed in a capillary for crystallization of
the partially purified CrFe protein solution (Fig.1). After
precipitant was extensively optimized, there was only one
large brick red crystal in a capillary incubated for 128 d. It is
shown that the composition of the precipitant had in-
deed an important effect on the number, size and color
of crystals formed from the partially purified CrFe pro-
tein solution.
ZHAO Jian-Feng et al.: Crystal Growth and Characterization of Residual Bacterioferritin in Partially Purified Nitrogenase CrFe
Protein Solution from a Mutant UW3 of Azotobacter vinelandii 1333
2.2 Characterization of brick red crystals
In order to clarify the reason about different colors of
crystals, the subunit composition of the crystal proteins
was analyzed. The brown and brick red crystals were picked
out, then washed and dissolved under atmosphere of Ar
with 25 mmol/L Tris-HCl buffer containing 0.3 mg/L DT,
respectively. After centrifugation, the supernatants were
analyzed by SDS-PAGE. Both brown crystals and the brick
red crystals had one band at ~14 kD and two bands at ~20
kD, but only the former had bands (~60 kD) similar to those
of Av1 (Fig.2). On SDS-PAGE gel of the partially purified
CrFe protein solution before crystallization, the two main
bands were indeed at ~60 kD position even though there
were several bands with low molecular weights. And the
two main bands were detectable by Western blotting analy-
sis of SDS gel, while other bands with lower molecular weight
were not detectable (Fig.3). Like crystalline Av1, a main
band at the same position as that of Av1 on the anoxic
native gel of the partially purified protein could be detected
as an iron-containing protein (Fig.4). This indicates that
CrFe protein in the partially purified CrFe protein solution
seemed to be an iron-containing protein composed of the
similar subunits to those of Av1. As reported earlier (Huang
et al., 2002), the partially purified CrFe protein was shown
to contain Fe and Cr (atom ratio of Fe to Cr was 11.60) and
to have ~70% of C2H2- and H
+-reduction activity of Av1.
Therefore, it is reasonable to consider that the protein
should be a nitrogenase component Ⅰ composed of the
~60 kD subunits. The smaller subunits of the brown crystal
shown in lane 4 of Fig.2 could be one of the following
subunits. (1) Subunits forming the brick red crystal. Some
Table 1 Effect of PEG 8000 and NaCl concentrations on crystallization of proteins in the partially purified CrFe protein solution by
liquid/liquid diffusion method
Precipitant Crystals
Solution(1) NaCl (mmol/L) PEG (%) Number Size(2)
#1 987.20 7.00 40 Small
#2 987.20 6.00 4/2 Large/middle
#3 987.20 5.57 2/1 Large/middle
#4 1 073.04 5.57 1(3) Large
#5 1 158.89 5.57 3/10 Middle/small
#6 987.20 4.71 3(4) Large
(1), concentrations of MgCl2, glycerin, Hepes buffer (pH 8.2) in the precipitant and the protein in 25 mmol/L Tris buffer (pH 7.4) containing
250 mmol/L NaCl were 599.75 mmol/L, 11.57 % (V/V), 74.57 mmol/L and 6.53 mg/mL, respectively; (2), the longest side of large, middle and
small crystal was >0.20 mm, 0.05-0.20 mm and <0.05 mm, respectively; (3), brick red in color; the others were brown or brick red; (4),
incubation time was 163 d; the others were 128 d.
Fig.1. Brick red crystal (A) and brown crystals (B) formed in
the partially purified CrFe protein solution incubated with the
solution #4 and the solution #6 of Table 1 for 128 d and 163 d,
respectively. The crystal diagonals in A are 0.28 mm and 0.20
mm, respectively; and both of longer sizes for the crystal in B are
0.16 mm.
Fig.2. SDS-PAGE (12%) of partially purified CrFe protein
solution before and after crystallization. 1, partially purified CrFe
protein solution before crystallization (5.17 mg); 2, crystalline
Av1 (4.79 mg); 3, crystals picked out, most of which were brick
red; 4, crystals picked out, most of which are brown; 5, marker
(low molecular standard).
Acta Botanica Sinica 植物学报 Vol.46 No.11 20041334
brick red crystals could be mixed with the brown ones be-
cause it was rather difficult to exactly differentiate the crys-
tals’ kind. (2) The possible third subunit of CrFe protein. It
has been reported that Apo-Av1 from some mutants could
be composed of a, b and g subunits which could be decom-
posed to fractions with molecular weights of ~20 kD and
~14 kD (Homer et al.,1995). Even though the final identifi-
cation of the subunits is on the way, it is reasonable to
suppose that the brown crystal was from CrFe protein and
the brick red crystal from another protein.
It is well shown in Fig.4 that there were only two iron-
containing proteins including CrFe protein in the partially
purified CrFe protein solution. CrFe protein migrated quicker
than the other protein did on native gel. The protein of the
brick red crystal was detected at the same position as that
of the protein with smaller electrophoretic mobility, show-
ing that the protein seemed to be the same as that of the
brick red crystal. It is reasonable to consider that the brick
red crystal was formed from an iron-containing protein,
rather than CrFe protein. An absorption spectrum of the
partially purified CrFe protein solution (in DT-containing
solution) had three peaks at 416 nm, 526 nm and 556 nm,
respectively (Fig.5). These absorption peaks were similar
to those of AvBF from wild type strain of A. vinelandii (Li
et al., 1980). It was shown that a protein similar to AvBF
indeed existed in the partially purified CrFe protein solution.
As it is known, bacterioferritin is composed of 24 subunits
with molecular weight of ~20 kD, half of which contains
porphyrin, resulting in a small difference between the sub-
units (Andrews et al., 2003). It is shown in Fig.2 that the
molecular weights of the subunits composing the brick red
crystal were approximately equal to that of bacteroferritin
subunits. Therefore, it was reasonable to deduce that the
brick red crystal was crystallized from the residual
bacterioferritin.
Based on a comparison of protein amount with iron
amount on the bands corresponding to bacterioferritin, CrFe
protein and Av1 in Fig.4, it was impossible to estimate that
the bacterioferritin contained much more iron atoms than
CrFe protein or Av1 (30 irons per molecule) did. Hence, it
could be speculated that the crystallized bacterioferritin
was, to a large extent, a hollow tetracosamer without iron
Fig.3. SDS gel (13%) of partially purified CrFe protein solution
stained with Coomassie R-250 (A) and Western blot of SDS gel
developed with antibody to Av1 (B). Lane 1, partially purified
CrFe protein solution (5.5 mg in A, 11.0 mg in B); lane 2, crystal-
line Av1 (5.0 mg in A, 10.1 mg in B).
Fig.4. Anoxic native gel (6%) of partially purified CrFe protein
solution stained with Coomassie R-250 (A) and 3,5-
diaminobenzoic acid (B). Lane 1, partially purified CrFe protein
solution (34.5 mg in A, 344.8 mg in B); lane 2, brick red crystals
fromed from partially purified CrFe protein solution; lane 3, crys-
talline Av1 (50. 30 mg in A, 150.96 mg in B).
Fig.5. Absorption spectrum of partially purified CrFe protein
solution. Concentration of protein and path of cell are 3.56
mg/mL and 4.98 mm, respectively.
ZHAO Jian-Feng et al.: Crystal Growth and Characterization of Residual Bacterioferritin in Partially Purified Nitrogenase CrFe
Protein Solution from a Mutant UW3 of Azotobacter vinelandii 1335
core, which is subsequently confirmed by X-ray diffraction
analysis (Liu et al., 2004). The reason why the crystallized
bacterioferritin is hollow needs further investigations.
2.3 Identification of bacterioferritin crystal
To investigate the protein composition of the crystals,
we used MALDI-TOF MS to obtain sequence information
of their subunits in SDS gel. The two bands of about 20 kD
were cut and digested in-gel with trypsin, and mass spectra
of the resulting peptides (peptide mass fingerprints) were
acquired. The list of apparent peptide masses was then
used to screen databases for correspondence to predict
tryptic digests of known proteins. The analysis result of
the larger subunit (Table 2) showed that 13 fragments
matched 100% with the tryptic peptides from 18 207 Da
bacterioferritin of A. vinelandii (AvBF), and the matched
peptides accounted for 69% (107/156 amino acid residues)
of total sequence. A similar conclusion was drawn from the
analysis result of the smaller subunit, except that 1 538.54
Da peptide (positions 144-156 amino acids without
modification) was not detected. It can be concluded that
the protein compositions of two bands were identical, which
was in accordance with the result of Grossman et al. (1992).
Therefore, it was proved that protein composition of the
crystals was AvBF.
The X-ray diffraction data integration and scaling were
performed with the programs DENZO and SCALEPACK
(Otwinowski et al., 1997). As shown in Fig.6, the brick red
crystal can diffract to 2.34 Å resolution. The crystals be-
long to space group H3, with unit-cell parameters a =
124.965 Å, b = 124.965 Å and c = 287.406 Å. A typical “432”
point group symmetry was found by calculating the self-
rotation function with MOLREP (Vagin et al., 1997), con-
firming that the crystal protein used for diffraction was 24-
meric bacterioferritin. The detailed structure determination
and analysis will be introduced in a forthcoming publica-
tion (Liu et al., 2004).
3 Conclusions
From the partially purified CrFe protein solution, both
the brown crystals and brick red crystals were observed
simultaneously under given conditions. The analyses by
many biochemical techniques have shown that the pro-
teins forming the two crystals are nitrogenase CrFe protein
and bacterioferritin,respectively. The determination by
mass spectrometry and preliminary crystallographic analy-
sis of the latter crystal further shows that the brick red
crystal was that of bacterioferritin.
In the partially purified CrFe protein solution, CrFe pro-
tein and bacterioferritin were different in both protein kind
and content. The former was the main protein composed of
subunits similar to a2b2 of Av1 while the latter was one of a
contamination protein composed of 24 subunits similar to
those of AvBF. However, it is unexpected that two different
proteins could be simultaneously crystallized under some
Table 2 Correlation of the peptide masses of larger subunit as determined by MALDI-TOF analysis to calculated mass data of AvBF
Peptide Experimental mass Theoretical mass DDa(1) Position(2) Peptide sequence Modification
1 909.52 909.53 -0 . 0 1 7 7 - 8 4 (K)LLIGEHTK(E) None
2 963.56 963.59 -0 . 0 3 6 - 1 3 (K)IVIQHLNK(I) None
3 1 082.57 1 082.60 -0 . 0 3 93 -1 0 2 (K)LEQAGLPDLK(A) None
4 1 169.49 1 169.51 -0 . 0 2 3 1 - 3 9 (R)MYEDWGLEK(L) None
5 1 185.47 1 185.50 -0 . 0 3 3 1 - 3 9 (R)MYEDWGLEK(L) 1-Met-ox(3)
6 1 416.57 1 416.60 -0 . 0 3 4 3 - 5 3 (K)HEYHESIDEMK(H) None
7 1 432.54 1 432.59 -0 . 0 5 4 3 - 5 3 (K)HEYHESIDEMK(H) 1-Met-ox(3)
8 1 538.54 1 538.69 -0 . 1 5 144 -15 6 (K)IGLENYLQSQMDE None
9 1 554.57 1 554.69 -0 . 1 2 144 -15 6 (K)IGLENYLQSQMDE 1-Met-ox(3)
10 1 631.69 1 631.72 -0 . 0 3 103 -11 7 (K)AAIAYCESVGDYASR(E) None
11 1 682.93 1 682.96 -0 . 0 3 6 2 - 7 6 (R)ILFLEGLPNLQELGK(L) None
12 1 839.08 1 839.06 0.02 6 1 - 7 6 (K)RILFLEGLPNLQELGK(L) None
13 1 984.08 1 984.09 -0 . 0 1 1 4 - 3 0 (K)ILGNELIAINQYFLHAR(M) None
(1), DDa means mass experimented-mass calculated; (2), positions refer to the positions of amino acid residues in the polypeptide chain; (3),
Met-ox refers to oxidation of methionine in the peptide sequence.
Fig.6. A diffraction pattern of brick red crystal and a zoomed-in
region with a reflection at 2.34 Å.
Acta Botanica Sinica 植物学报 Vol.46 No.11 20041336
conditions. It indicates that a protein with low purity still
has an opportunity to be crystallized under some conditions.
In general, protein solubility value is not unique but of
several kinds, which makes a protein to form crystals with
different shapes under several conditions (Mcpherson
et al., 1983). Crystal formation of two or more proteins in
one solution depends mainly on protein properties and
other factors. During diffusion of a protein into a precipitant,
the local concentration of the protein and the precipitant in
different position of a capillary are different from one
another, leading to form crystals of different proteins with
different shape and size. Therefore, it is not unexpected
that CrFe protein and the residual bacterioferritin in the
partially purified CrFe protein solution were simultaneously
crystallized or only the latter was crystallized under a given
condition. The crystallization of the residual bacterioferritin
should be an example for crystallization of the minor con-
tamination protein.
References:
Andrews S C, Robinson A K, Rodriguez-Quinones F. 2003. Bac-
terial iron homeostasis. FEMS Microbiol Rev, 27: 215-237.
Andrews S C, Smith J M, Yewdall S J, Guest J R, Harrison P M.
1991. Bacterioferritins and ferritins are distantly related in
evolution. Conservation of ferroxidase-centre residues. FEBS
Lett, 293: 164–168.
Bishop P E, Donna M L, Hetherington D R. 1982. Expression of
an alternative nitrogen fixation system in Azotobacter
vinelandii. J Bacterial, 150: 1244-1251.
Carrondo M A. 2003. Ferritins, iron uptake and storage from the
bacterioferritin viewpoint. EMBO J, 22: 1959-1968.
Grossman M J, Hinton S M, Minak-Bernero V, Slaughter C, Stiefel
E I. 1992. Unification of the ferritin family of proteins. Proc
Natl Acad Sci USA, 89: 2419–2423.
Homer M J, Dean D R, Roberts G P. 1995. Characterization of
the γ protein and its involvement in the metallocluster as-
sembly and maturation of dinitrogenase from Azotobacter
vinelandii. J Biol Chem, 270: 24745-24752.
Huang J-F, Dong Z-G, Zhang H-F, Lü Y-B , Zhao Y, Wang Z-P .
2002. Purification and characterization of Cr-containing nitro-
genase component Ⅰ. Acta Bot Sin , 44: 297-300.
Huang J-F , Wang D-Y, Dong Z-G, Wang Z-P , Zhang H-F, Lü Y-
B , Xu Y-M , Zhao Y. 2001. Purification and characterization
of Mn-containing nitrogenase component Ⅰ. Acta Bot Sin ,
43: 918-922.
Li J-D, Wang J-W, Zhong Z-P , Tu Y, Dong B . 1980. Presence of
a cytochrome b-containing ferritin in Azotobacter vinelandii.
Sci Sin.Ser B , 23: 897-904.
Liu H-L , Zhou H-N, Xing W-M, Zhao J-F, Li S-X, Huang J-F, Bi
R-C. 2004. 2.6 Å resolution crystal structure of the
bacterioferritin from Azotobacter vinelandii. FEBS Lett, (in
press)
Luo A-L, Wang J-W, Li J-G . 1995. A simple method for protein
blotting and immunoassay. Chin Bull Bot , 12: 63-64. (in Chi-
nese with English abstract)
McPherson A J. 1983. Group of Crystalline Growth, Institute of
Biophysics, The Chinese Academy of Sciences trans. The
growth and preliminary investigation of protein and nuclei acid
crystals for X-ray diffraction analysis. Glick D. Methods of
Biochemical Analysis. Vol. l23. Beijing: Chinese Science and
Technology Press. 249-345. (in Chinese)
Otwinowski Z, Minor W. 1997. Processing of X-ray diffraction
data collected in oscillation mode. Meth Enzymol, 276: 307–
326.
Paustian T D, Shah V K, Roberts G P. 1990. Apodinitrogenase:
purification, association with a 20-kilodalton protein, and ac-
tivation by the iron-molybdenum cofactor in the absence of
dinitrogenase reductase. Biochemistry, 29: 3515-3522.
Richards T D, Pitts K R, Watt G D. 1996. A kinetic study of iron
release from Azotobacter vinelandii bacterial ferritin. J Inorg
Biochem, 61: 1-13.
Schneider K, Muller A, Johannes K U, Diemann E, Kottmann J.
1991. Selective removal of molybdenum traces from growth
media of N2-fixing bacteria. Anal Biochem, 193: 292-298.
Shah V K, Brill W J. 1973. Nitrogenase. Ⅳ. Simple method of
purification to homogeneity of nitrogenase components from
Azotobacter vinelandii. Biochim Biophys Acta, 305: 445-454.
Stiefel E I, Watt G D. 1979. Azotobacter cytochrome b-557.5 is a
bacterioferritin. Nature, 279: 81–83.
Vagin A, Teplyakov A. 1997. MOLREP: an automated program
for molecular replacement. J Appl Cryst, 30: 1022-1025.
Wai S N, Takata T, Takade A, Hamasaki N, Amako K. 1995.
Purification and characterization of ferrit in from
Campylobacter jejuni. Arch Microbiol, 164: 1-6.
Watt G D, Frankel R B, Jacobs D, Huang H Q, Papaefthymiou G
C. 1992. Fe2+ and phosphate interactions in bacterial ferritin
from Azotobacter vinelandii. Biochemistry, 31: 5672-5679.
Zhao B-G,Lin Z-J,Tang Y-Q,Li J-D, Wang J-W,Tu
Y.1984. Preliminary crystallographic studies on
bacterioferritin from Azotobacter vinelandii. Sci Sin.Ser B ,
27: 1002-1007.