The EMBO Journal vol. 1 1 no. 1 3 pp. 5101 - 5109, 1992
E.coHi MukB protein involved in chromosome partition
forms a homodimer with a rod-and-hinge structure
having DNA binding and ATP/GTP binding activities
Hironori Niki, Ryu Imamura,
Mitsuhiko Kitaokal, Kunitoshi Yamanaka,
Teru Ogura and Sota Hiraga
Department of Molecular Cell Biology, Institute of Molecular
Embryology and Genetics, Kumamoto University School of Medicine,
Kuhonji 4-24-1, Kumamoto 862, and 'Department of Developmental
Neurobiology, Kumamoto University School of Medicine, Kuhonji
4-24-1, Kumamoto 862, Japan
Communicated by K.Nordstrom
mukB mutants of Escherichia coli are defective in the
correct partitioning of replicated chromosomes. This
results in the appearance of normal-sized anucleate
(chromosome-less) cells during cell proliferation. Based
on the nucleotide sequence ofthe mukB gene, the MukB
protein of 177 kDa was predicted to be a filamentous
protein with globular domains at the ends, and also
having DNA binding and nucleotide binding abilities.
Here we present evidence that the purified MukB protein
possesses these characteristics. MukB forms a homodimer
with a rod-and-hinge structure having a pair of large,
C-terminal globular domains at one end and a pair of
small, N-terminal globular domains at the opposite end;
it tends to bend at a middle hinge site ofthe rod section.
Chromatography in a DNA-cellulose colunm and the
gel retardation assay revealed that MukB possesses DNA
binding activity. Photoaftmity cross-linking experiments
showed that MukB binds to ATP and GTP in the
presence of Zn2+. Throughout the purification steps,
acyl carrier protein was co-purifiled with MukB.
Key words: acyl carrier protein/chromosome partitioning/
DNA binding protein/nucleotide binding protein/purification
It is still not known how replicated chromosomes are orderly
partitioned into each of the daughter cells in bacteria, of
which the mitotic apparatus has not been found yet. Jacob
et al. (1963) proposed a model for bacterial chromosome
partitioning. In this model, the chromosome is attached to
the cell envelope and the replicated daughter chromosomes
move apart from each other driven by a force provided by
the insertion of new cell envelope material at mid-cell; thus
resulting in the partitioning of replicated chromosomes into
each of the two daughter cells. However, this model is now
uncertain (for a review see Hiraga, 1992). Hiraga et al.
(1990) described how,
daughter chromosomes can rapidly migrate from mid-cell
to cell quarters regardless of cell elongation. This movement
or positioning of daughter chronosomes may be promoted
by events that are dependent on post-replicational protein
synthesis (Donachie and Begg, 1989; Hiragaetal., 1990).
in Escherichia coli, replicated
O Oxford University Press
Daughter chromosomes are rapidly moved apart by a fixed
distance (unit length) immediately after the completion of
replication (Begg and Donachie, 1991).
The oriC DNA (the replication origin of the E.coli
chromosome) is especially enriched in the DNA-membrane
complex fraction (Hendrickson et al., 1982; Yoshimoto
et al., 1986). Hemimethylated oriC DNA associates with
the cell membrane just after replication, and the association
the hemimethylated DNA
methylated by Dam methylase (Ogden et al., 1988). It was
thought that hemimethylated oriChad an equivalent function
to a centromere during partitioning of the chromosome.
produced at high frequencies in dam mutants which lack Dam
methylase. Therefore, hemimethylation of DNA is not
essential for chromosome partitioning (Vinella et al., 1992).
Conditionally lethal par mutants of E.coli were isolated
to analyze the mechanisms of chromosome segregation or
partition. These mutants produce elongated cells having a
large chromosome mass(es) at mid-cell after incubation at
non-permissive temperatures. Some ofthem have mutations
in the genes coding for the subunits ofDNA gyrase (Hussain
et al., 1987a,b; Kato et al., 1989) or DNA topoisomerase
IV (Kato et al., 1988, 1990). Mutations in the DNA gyrase
genes inhibit chromosome decatenation (Steck and Drlica,
1984). These par mutants are therefore defective in
decatenation and/or other topological events of replicated
daughter chromosomes at non-permissive temperatures.
On the other hand, we have isolated another type ofE.coli
mutant which is viable but defective in the active partitioning
of chromosomes, resulting in the appearance of normal-sized
proliferation (Hiraga et al.., 1989). In this type of mutant
decatenation of replicated chromosomes may be normal, but
the active positioning of daughter chromosomes from mid-
cell to cell quarter positions is defective. One of them, the
mukAl mutation is located in the tolC gene (Hiraga et al.,
1989; Niki et al., 1990), which encodes an outer membrane
protein (Morona et al., 1983) required for hemolysin
secretion from the cell to the medium (Wandersman and
Another mutation, mukB106 is located at 21 min on the
E. coli chromosome. The wild type mukB gene was cloned,
sequenced and found to code for a 177 kDa protein (Niki
et al., 1991). Based on the deduced amino acid sequence,
we previously speculated the molecular structure of MukB
as follows. The N-terminal globular domain (domain I)
contains a consensus nucleotide binding sequence. Domain
II presumably forms an et-helical coiled-coil structure.
Domain III is globular. Domain IV is rich in a-helix,
probably forming a coiled-coil structure. The C-terminal
domain (domain V) is also globular, having a C-terminal
subregion rich in cysteine, arginineandlysine;zincfinger-
like structures may be formed in this subregion. Two MukB
molecules may form a homodimer in the coiled-coil regions.
H.Niki et al.
Fig. 1. Purification of MukB in SDS-PAGE. (A) Overproduction of MukB. Lane 1, whole cell lysateof the wildtypestrain(W3110); lane 2,
whole cell lysate of MukB overproducing strain (W3110/pAX814). Cells collected from 0.2 ml of anexponentially growingculture weresuspended
in the sample buffer and boiled for 3 min. The samples were analyzed by 10% SDS-PAGE. (B) Proteins in eachpurification step. Lane 1, the
cleared cell lysate (2.5 ,ug of protein); lane 2 , the AS fraction (2.0 ,ug); lane 3, the S-400 fraction (1.5 Mg); lane 4, the DEAE fraction (1.5 Mg);
lane 5, the DNA-cellulose fraction (1.5 ug). Lane M shows marker proteins. Arrows at the left indicate molecular weights of markerproteins, and
open arrows at the right indicate the MukB protein. These gels were stained with Coomassie brilliant blue.
The predicted secondary structure of MukB suggests that
the protein is the first candidate in eubacteria for a force-
generating enzyme which moves daughter chromosomes
from mid-cell to cell quarters, and the N-terminal domain
is expected to act as a 'motor' domain (Niki et al., 1991;
for reviews see Hiraga, 1991; Hiraga et al., 1991). The
amino acid sequence of the N-terminal domain ofMukB is
significantly homologous to dynamin (D100) of rat brain,
which is a microtubule-associated mechanochemical enzyme
(Obar et al., 1990; Niki et al., 1991).
In this study, we have purified the MukB protein from
the extract of MukB overproducing cells. We examined its
properties and have found that the purified MukB protein
has some of the previously speculated characteristics. We
propose a revised structural model of the MukB protein.
Purification of the MukB protein
To purify the MukB protein, we used the E.coli strain
W3110 harboring the high copy-number plasmid pAX814
which carries the mukB gene. This strain overproduced
MukB during exponential growth (Figure IA, lane 2).
Growing cells were collected and sonicated. The cell lysate
was fractionated by precipitations with ammonium sulfate
in Materials and methods. The sample
containing MukB was further purified by gel filtration
through a column of Sephacryl S-400 (Figure 2A). Fractions
containing MukB (fractions 47-58) were pooled and
purified through a column of DEAE Sephacel, and then
eluted with buffer A containing 0.5 M NaCl from the DEAE
Sephacel column. After dialysis against buffer A, the MukB
sample was loaded on to a DNA-cellulose column. The
MukB protein adsorbed to DNA-cellulose, and then it was
eluted with buffer A containing 0.2 M NaCl (Figure 2B).
Figure lB shows proteins in each purification step. In
addition to MukB, an apparent 21 kDa protein (named
tentatively p21) was detected in the final sample of
DNA-cellulose column (Figure 2B). p21 was co-purified
with MukB through the purification steps.
Acyl camer protein was co-purified with the MukB
The N-terminal amino acid sequence of MukB was
determined. The first 14 amino acids of MukB were
MIEXGKFXSLTLIN (X represents an unidentified amino
acid residue). The sequence was consistent with the amino
acid sequence, MIERGKFRSLTLIN, which was deduced
from the nucleotide sequence ofthe mukB gene (Niki et al.,
The sequence of the first 40 amino acids of p21 was
determined and searched for homology in a protein sequence
database (SWISS-PROT). We found that the N-terminal
sequence ofp21 is consistent with that ofacyl carrier protein,
ACP (Figure 3A; Vanaman et al., 1968; Jackowski and
phospholipid biosynthesis (for a review see Cronan and
Rock, 1987). It is known that ACP is 8.8 kDa, however,
it migrates like an -20 kDa protein in SDS-PAGE (Rock
and Cronan, 1979). To confirm that p21 is identical to ACP,
we performed a Western blotting analysis using an anti-ACP
antibody. In SDS-PAGE, the mobility ofp21 was identical
to that ofan authentic ACP sample (purchased from Sigma)
and it cross-reacted strongly with the anti-ACP antibody,
too. These results indicate that p21 protein is identical to
ACP. ACP was thus co-purified with MukB through the
A sample from the DNA-cellulose fraction, containing
MukB and p21, and the authentic ACP sample were
independently analyzed in a Superose 6 column. In the
Properties of the MukB protein
1.2 M \I
Fig. 2. SDS -PAGE analysis of proteins in fractions eluted from a gel filtration column and a DNA-cellulose column. (A) The elution profile of
proteins in the Sephacryl S-400 gel filtration. TwentyAlsamples from each fraction were analyzed by 10% SDS-PAGE. Lane M shows molecular
weight markers. Vo, the void volume; TG, the elution peak of a marker, thyroglobulin (669 kDa). (B) The elution profile of MukB with 0.2 M
NaCl from a DNA-cellulose column. Samples (20 yl) from each fraction were analyzed by 12% SDS-PAGE. The fraction numbers are indicated
at the top. Arrows at the left indicate the marker proteins. Open arrows at the right indicate the MukB protein and the 21 kDa protein (p21),
sample from the DNA-cellulose fraction, ACP and MukB
were eluted in fractions 42-54 (Figure 3B). The authentic
ACP was eluted in fractions 78-88 (Figure 3C). Moreover,
the authentic ACP did not bind to DNA-cellulose (data not
shown). The results suggest that ACP associates with MukB
throughout the purification process.
MukB molecules observed by electron micrography
To analyze the ultrastructure of MukB, we observed the
purified MukB protein using the low angle rotary shadowing
method. The shape ofthe MukB molecule was a filamentous
rod having globular domains at both the ends (Figure 4, first,
second and third rows). The length between the centers of
the globular domains was 59.2 ± 4.7 nm. The globular
domains differed from each other in size. Their diameters
were 20.4 ± 2.5 and 26.4 ± 3.5 nm, respectively. Some
molecules were bent at a middle site of the rod, displaying
V-shaped molecules (Figure 4, fourth, fifth and sixth rows).
We found that the MukB molecule tends to bend at the
middle site of the rod.
H.Niki et al.
weightof a MukB dimer. On the otherhand, it wasmainly
950 kDa in a low salt condition of 25 mM KCI.
Binding of the MukB protein to DNA
The MukB protein bound to DNA-cellulose as described
above (Figure 2B), but it did not bind to cellulose alone(data
not shown). These results suggest that MukB possesses a
DNA binding activity. To confirm this property, we
performed the gel retardation assay using the purified MukB
protein and 32P-labeled DNA fragments (0.2 pmol) of
pUC13 digested with HaeIII and AluI. Five bands ofDNA
fragments were clearly observed in the absence of MukB
(Figure 6A, lane 1). When MukB was added to the DNA
sample at a range of 0.01-0.1 pmol, no retardation was
observed (Figure 6A, lanes 2-5). However, when MukB
was added at a range of 0.25-5.0 pmol, retardation of
labeled DNA fragments increased as the amount of MukB
was increased (Figure 6A, lanes 6-10). Addition of more
than 150 mM NaCl to the reaction mixture inhibited the
retardation oflabeled DNA fragments by MukB (Figure 6B).
When unlabeled DNA fragments of pUC13 were added to
the reaction mixture as competitors, the retardation oflabeled
DNA fragments by MukB was also inhibited (Figure 6C).
These results indicated that MukB possesses the DNA
Fig. 3. Determination of the N-terminal amino acid se
kDa protein, and immunoblotting analysis of the 21 k]
the anti-ACP antibody. (A) The N-terminal amino aci
the 21 kDa protein (p21) and ACP (Vanaman et al.,
and Rock, 1987). (B) and (C) the MukB protein-conta
the DNA-cellulose fraction (B) and the authentic AC
were independently fractionated through a column of '
fraction was analyzed in SDS-PAGE and immunoblo
anti-ACP antibody. Numbers at the bottom indicate fr
Vo, void volume; TG, thyroglobulin (669 kDa); F, fe
C, catalase (232 kDa).
1ining sample of
'P sample (C)
Superose 6. Each
Itted with the
Physical properties and the molecular wei
native MukB protein
The Stokes' radius ofthe native MukB protein v
by HPLC gel filtration; it was 11.8 nm in buf
KCl) and 9.7 nm in buffer A with 500 mM Ki
formation ofMukB aggregates. The sedimental
of MukB in buffer A lacking glycerol was
be 14.3S by a Beckman ultracentrifuge mo(
It is known that the sedimentation coeffi
Stokes' radius depend on the protein's shape.
tried to determine the molecular weight of the
protein using the low angle laser light scati
where the molecular weight can be measured
the protein's shape (Takagi et al., 1980;Takal
MukB protein was 365 kDa in a high salt coi
mM KCl (Figure 5), which coincides with
Photoaffinity cross-linking of nucleotides to the MukB
The consensus nucleotide binding sequence was previously
found in the N-terminal globular domain of MukB (Niki
etal., 1991). We performed photoaffinity cross-linking
experiments to detect nucleotides bound to MukB. The
MukB protein did not cross-link with [a-32P]ATP in the
absence, or the presence, of MgCl2, CaCl2, or EDTA
(Figure 7A). In contrast, it cross-linked with [a-32P]ATP
in the presence of 10 mM ZnCl2 (Figure 7A, lane 5) and
also with [ca-32P]GTP in the presence of 10 mM ZnCl2
(Figure 7C). When various amounts oflabeled ATP or GTP
were added to the reactionmixture, cross-linkingbetween
[a-32P]ATP and MukB increased as the amount of
unlabeled ATP or GTP was increased (Figure 7B, lanes 2-5
and 6-9). The largest amount of[ca-32P]ATP that cross-
linked with MukB was observed in the presence of 0.1 mM
unlabeled ATP or GTP(Figure 7B, lanes5and9). When
1 mM of unlabeled ATP or GTP was added to the mixture,
cross-linking between MukB and [ct-32P]ATP was inhibited
(Figure 7B, lanes 6 and 10). Analogous results were obtained
in the cross-linking between MukB and [Ca-32P]GTP
(Figure 7C). Thus, MukB binds to ATP and GTP in the
presence of Zn2+.
Nquenceof the 21
Da protein using
d sequences of
ight of the
Ffer A (25 mM
icient and the
gi, 1981). The
ndition of 500
Electron micrography revealed that MukB is a filamentous
protein having globular domains, of different sizes, at both
ends; the length between the centers of the two globular
and the large globular domains was 1.3, suggesting that the
ratio ofapparent masses must be
suggests that the small globular domain consists of domain
I (338 amino acid residues), and the large globular domain
consists of domains III, IV and V (total of 869 amino acid
residues). The central rod consists ofdomain II (327 amino
-60 nm. The diameter ratio between the small
-2.2. This ratio ofmasses
Properties of the MukB protein
-; ~ ~~~
w tte t
Fig. 4. Electron micrography of MukB molecules stained with platinum-carbon by the low angle rotary shadowing method. Upper rows display
extended molecules and lower rows display bent molecules. The bar indicates 100 nm.
acid residues) having long stretches of heptad repeats. The
length of the a-helical coiled-coil structure ofdomain H can
be estimated to be 48 nm according to the result that the
rod regions of coiled-coil proteins extend
amino acid residues (Bourne, 1991). This length of rod is
consistent with the above experimental value of 60 nm for
the length of rod between the centers of the two globular
Although it is not clear whether the MukB molecules
shown in Figure 4 are homodimers or monomers, the result
obtained by the low angle laser light scatter technique
indicates that the native MukB protein is 365 kDa in a high
salt concentration (500 mM KCl), thus suggesting that the
native MukB protein is a homodimer. Cytoplasmic KCl
concentration in living E.coli cells ranges from 100 to 600
mM (Epstein and Schultz, 1965). Under low salt conditions,
MukB homodimers presumably tend to associate with each
other and to form large aggregates. Our recent observation
by electron micrography (unpublished data) also supports
that MukB forms homodimers.
We show a revised structural model of MukB in Figure 8.
Two MukB molecules form a homodimer in domain H
(Leu339-Gln665) forming the a-helical coiled-coil rod
structure. The a-helical coiled-coil stretch is interrupted at
the Gly487-Pro488 site, which may form a hinge of the a-
helical coiled-coil structure to bend the MukB homodimer
into a V-shape. The N-terminal domain (Metl -Asn338)
forms the small globular domain. The large globular domain
corresponds to the part of Pro666-Glu1534. Although
- 15 nm per 100
M. W. (kd)
Fig. 5. Determination of the molecular weight of the native MukB
protein using the low angle laser light scatter technique. The vertical
axis represents the ratio between LS and RI (LS, output of a low
angle laser light scattering photometer; RI, output of a differential
refractometer). The horizontal axis represents the molecular weight.
Open circles represent authentic proteins; thyroglobulin (669 kDa),
catalase (232 kDa), aldolase (158 kDa) and albumin (67 kDa). The
closed circle shows the MukB protein.
domain IV is rich in ax-helix, heptapeptide repeats in the
domain are interrupted at two subregions (Niki et al., 1991).
This is consistent with the above speculation that domain
IV forms the large globular domain together with domains
la 4: &t
Fig. 6. Gel retardation assay for the DNA binding activity of MukB. (A) 32P-labeled pUC13 DNAfragments (0.2 pmol) were treated with various
amounts of the purified MukB protein for 10 min at 0°C and analyzed in PAGE. Amounts of MukB added to the reaction mixture are indicatedat
the top. (B) Effect of NaCl concentration on the DNA binding activity of MukB. 32P-labeled DNA (0.2 pmol) and MukB (5 pmol) were mixed in
the presence of various concentrations of NaCl; concentrations of NaCl are indicated at the top. (C) Effect of unlabeled DNA fragments ofpUC13
on the DNA binding activity of MukB. 32P-labeled DNA (0.2 pmol) and MukB (5 pmol) were mixed in thepresence of various concentrations of
unlabeled pUC13 DNA fragments. Amounts of unlabeled DNA fragments added to the mixture are indicated at thetop.Lane numbers are shownat
Fig. 7. Photoaffinity cross-linking between MukB and a-32P-labeled nucleotides. (A) The purified MukB protein (50 pmol) and [a-32P]ATP (3.3
pmol) were mixed and irradiated with UV light as described in Materials and methods. Lane 1, no addition; lane 2, 10 mM MgCI2; lane 3, 10 mM
CaC12; lane 4, 10 mM EDTA; lane 5, 10 mM ZnCl2. (B) Effect of unlabeled ATP or GTP on the photoaffinity cross-linking between [CI-32P]ATP
and MukB. The reaction mixture contained MukB (50 pmol) and [c-32P]ATP (3.3 pmol), and 10 mM ZnCl2. Lane 1, without UV irradiation; lanes
2-10, with UV irradiation; lanes 3-6, the addition of an indicated amount of unlabeled ATP; lanes 7-10, the addition of an indicated amount of
unlabeled GTP. (C) Effect of unlabeled ATP or GTP on the photoaffinity cross-linking between [cs-32P]GTP (3.3 pmol) and MukB (50 pmol). The
reaction mixture contained 10 mM ZnCI2 The open arrow on the right in each picture indicates the MukB protein.
IH and V. The MukB homodimer thus shows a rod-and-hinge
structure having a pair of small, N-terminal globular domains
at one end and a pair of large, C-terminal globular domains
at the opposite end.
A subregion (Cys1355 -Glu 1534) in domain V is rich in
cysteine residues and positively charged residues, arginine
and lysine. We speculated previously three putative zinc
finger-like structures in this subregion. These structures
H.Niki et al.
U...L,li ft *6
j .I 5X
Properties of the MukB protein
1 :54) 4
* (XI) -1½
Fig. 8. A revised structural model of MukB homodimer. See the text.
might be involved in the interaction with DNA (Niki et al.,
1991). We show here that the purified MukB protein
possesses DNA binding activity. However, it is not clear
as yet whether these zinc finger-like structures really exist
and act directly in DNA binding.
It is likely that MukB recognizes a specific DNA sequence
in the E.coli chromosome in vivo, and the specific DNA
sequence functions as a centromere in active chromosome
positioning. Low copy number plasmids in E. coli, such as
the F plasmid (Ogura and Hiraga, 1983; Mori et al., 1986),
P1 (Austin, 1984; Abeles et al., 1985; Martin et al., 1987),
and pSC101 (Meacock and Cohen, 1980; Tucker et al.,
1984) have a plasmid-specific cis-acting region which acts
as a centromere in the process of plasmid partitioning into
daughter cells (for a review see Hiraga, 1992). However,
partitioning of daughter chromosomes has not been found
E. coli chromosome.
that MukB recognizes many
nucleotide sequence motif which are scattered on the E. coli
chromosome, and it folds the replicated chromosomes into
a compact structure which is available for the active
The purified MukB protein binds to ATP and GTP in the
presence of Zn2+.
concentration (-0.1 mM) to obtain the highest binding
activity under experimental conditions (Figure 7B and C).
There was no remarkable difference between the binding
activities of MukB to ATP and GTP. We failed to detect
ATPase or GTPase activity in the MukB sample from the
requires an unknown cellular factor for ATPase and/or
essentialfor the active
sites of a
a high ATP
It is unclear whether MukB
GTPase activity. The amino acid sequence of the MukB N-
nucleotide binding sequence is significantly homologous with
the N-terminal part of the microtubule-associated protein
dynamin (D100) of rat brain (Obar et al., 1990; Niki et al.,
1991). Dynamin was originally identified as a microtubule-
activated ATPase (Shpetner and Vallee, 1989). However,
recent data showed that dynamin possesses GTPase activity
rather than ATPase activity, and that the GTPase activity
ofdynamin depends on the microtubules (Obar et al., 1990;
Shpetner and Vallee, 1992).
phospholipid biogenesis, initiation and elongation of fatty
acid synthesis, and acylation of sn-glycerol-3-phosphate to
phosphatidic acid (for a review see Cronan and Rock, 1987).
The 4'-phosphopantetheine prosthetic group is attached to
Ser36 of ACP, and intermediates of fatty acid biosynthesis
are attached to the sulfhydryl of the prosthetic group. ACP
is a very abundant protein (
and it is distributed throughout the cytoplasm in E.coli
prohemolysin (Issartel et al., 1991), which is a non-toxic
precursor of hemolysin, a membrane-targeted toxin. The
activation of prohemolysin is achieved by the transfer of a
fatty acyl group from ACP to prohemolysin in thepresence
ofHlyC protein. This suggeststhepossibilityofACPacting
as a modification factor of other proteinsbesidesbeingthe
co-factor of phospholipid biosynthesis. It is not clear yet
whether the interaction between ACP and MukB has a
biological function in vivo. ACP has an acidicpl (pH 4.1),
a co-factor for membrane
5 x I04 molecules per cell)
.~ ._. A i- ,,-
H.Niki et a/.
so it tends to associate with some protein by ionic interaction.
It is therefore possible that ACP associated with MukB in
the purification steps merely due to ionic interaction.
To elucidate molecular mechanisms of chromosome
partition in E. coli, it is important to examine whether the
purified MukB protein recognizes specific DNA sequences
of the E. coli chromosome
concentrations, and also to search for a cellular factor(s)
which interacts with MukB and endows it with the activity
of ATPase or GTPase.
Materials and methods
Bacterial strains and plasmids
The prototrophic strain W31 10 (Kohara et al., 1987) was used for this study.
Plasmid pAX814, which carried the mukB gene of E.coli, was previously
described (Niki et al., 1991).
Purification of the MukB protein
W31 10 harboring pAX814 was grown at 37°C in 10 1 of L medium (Niki
et al., 1991) containing ampicillin (200ytg/ml). Exponentially growing cells
were collected at 4'C by centrifugation. Harvested cells were washed with
ice-cold 25 mM HEPES-KOH buffer (pH 7.6) and suspended in 50 ml
of buffer A (25 mM HEPES-KOH pH 7.6, 25 mM KCI, 0.1 mM EDTA,
2 mM DTT and 20% glycerol) containing 0.2 mg/mil of lysozyme, 20 mM
spermidine and 0.5 mM phenylmethylsulfonyl fluoride (PMSF). The cell
suspension was kept on ice for 30 min and then sonicated at 20 s intervals
for 10 min. The cell lysate was centrifuged at 27 000 g for 60 min to remove
unbroken cells. The cleared cell lysate was centrifuged at 150 000 g for
2 h. Ammonium sulfate was added to the supernatant to a final concentration
of 35% (w/v), and stirred at 0°C for 30 min. After centrifugation at 12
000 g for 30 min, the supernatant was pooled and ammonium sulfate was
added to the supernatant to a final concentration of 55% (w/v). The
supernatant was stirred at 0°C for 30 min and then centrifuged. The pellet
was resuspended in 5 ml of buffer A, and this sample was called the AS
fraction. One ml of the AS fraction was loaded on to a column (1.2 x97
cm) of Sephacryl S-400 (Pharmacia LKB Biotechnology, Uppsala, Sweden)
which had been equilibrated with buffer A. Samples were fractionated into
4 ml fractions. Fractions were analyzed by SDS-PAGE (Laemmli, 1970).
Fractions containing MukB were collected (called S-400 fraction) and loaded
on to a column (2.8x8 cm) of DEAE Sephacel (Pharmacia) which had
been equilibrated with buffer A. The column was washed with 5 column
volumes of buffer A and eluted stepwise with 1 column volume of buffer
A containing 0.1, 0.2, 0.5 and 1.0 M NaCl. Fractions containing MukB
were collected and dialyzed overnight against 11 of buffer A (called DEAE
fraction). Fifty ml of DEAE fraction was loaded on to 2 ml of a column
of calf thymus DNA-cellulose (Pharmacia) equilibrated with buffer A. The
column was washed with 50 ml of buffer A and then eluted with buffer
A containing 0.2 M NaCl. Fractions containing MukB were dialyzed as
described above, and then concentrated 10- to 15-fold with a Centriprep
100 (Amicon Division, W.R.Grace & Co.-Conn., Beverly, MA). This
sample was called the DNA-cellulose fraction. All purification steps
described above were carried out at 4°C.
Western blotting and immunoluminescence method
The anti-ACP antibody was a gift of Drs S.Jackowski and C.O.Rock. An
ECL Western blotting system (Amersham) was used for immunoblotting
Determination of N-terminal sequences of proteins
The MukB protein and the p21 protein of the DNA-cellulose fraction were
separated in SDS-PAGE and electroblotted on to a PVDF membranefilter,
ProBlottTM (Applied Biosystems Inc., USA). Bands of proteins blotted on
to the membrane filter were detected by staining with Coomassie brilliant
blue and then excised. N-terminal amino acid sequences of the MukB protein
and the p21 protein were determined by a protein sequencer, Applied
Biosystems 477A (Applied Biosystems Inc.).
Low angle rotary shadowing electron micrography
The low angle rotary shadowing was performed according to Tyler and
Branton (1980) with minor modifications. Glycerol was added to the purified
MukB protein solution to a final concentration of 50% (v/v). The protein
solutionsupplementedwithglycerol wassprayedon to afreshly cleaved
micablake, which wassubsequently dried under vacuumusing a JEOL
freeze vacuumshadowing apparatus model JFD-700(JEOL Ltd, Tokyo,
Japan). Rotary shadowing with platinum-carbon was performed at an angle
of 3 -4°, and then the sample was treated with carbon at an angle of 90°.
The platinum-carbon replica was released from the mica blake on to the
surface of distilled water, collected on grids, and observed using a Hitachi
H300 transmission electron microscope.
Low angle laser light scatter technique
The molecular weight of the native MukB protein was determined using
the low angle laser light scatter technique (Sato et al., 1992). Prior to the
analysis, the sample of DNA-cellulose fraction was eluted through a column
of Superose 6 HR 10/30 with buffer A or buffer A containing 500 mM
KCI at flow rate of 0.2 ml/min to remove dust particles. The effluent was
successively monitored by an L-4200 UV-VIS detector (Hitachi), an LS-8000
low angle laser light scattering photometer (Tosoh) with a He-Ne laser (632.8
nm), and an RI-8011 differential refractometer (Tosoh) with a light-emitting
diode (660 nm). All measurements were carried out at room temperature
and the authentic proteins, thyroglobulin (669 kDa), catalase (232 kDa),
aldolase (158 kDa) and albumin (67 kDa) were used as standards.
Gel retardation assay
pUC13 DNA was digested with HaeIll and AluI and labeled with
[y-32P]ATP (3000 Ci/mmol) by T4 polynucleotide kinase. The purified
protein in buffer A was mixed with the labeled DNA fragments in a final
volume of 10
to electrophoresis at 4°C in 4% polyacrylamide gel containing 2.8% (w/v)
glycerol, 25 mM Tris-HCI (pH 8.3),
(Yamanaka et al., 1990). The gel was analyzed by autoradiography.
tl and kept at 0°C for 10 min. The mixture was subjected
1 mM EDTA and 190 mM glycine
Photoaffinity cross-linking of the MukB protein with 32P-labeled
The purified MukB protein (50 pmol) in buffer A was kept with 3.3 pmol
of [C_-32P]ATP or [a-32P]GTP (3000 Ci/mmol) for 15 min at 0°C in a final
volume of 50A1.The reaction mixture was subjected to photoaffinity cross-
linking for 40 min at 0°C using a 254 nm UV lamp at a distance of 1 cm
(Matsuyama et al., 1990). The UV-irradiated samples were precipitated
with 10% ice-cold TCA, washed with acetone cooled at -70°C, and
analyzed by 5% SDS-PAGE. The gel was stained with Fast Stain (Zoion
Research, Allston, MA) and then dried. 32P-labeled nucleotides cross-
linking with MukB were detected by autoradiography.
We are grateful to Drs S.Jackowski and C.O.Rock for providing the anti-
ACP antibody, Dr H.Maki for useful suggestions, Drs K.Sato and K.Shiga
for low angle laser light scatter technique, K.Nagata for analysis of the
sedimentation coefficient, and Dr S.Tanase for analysis of amino acid
sequences. We also thank C.Ichinose, T.Tsunenari and A.Matsusaka for
assistance in this laboratory. This work was supported by a Grant-in-Aid
for Scientific Research on Priority Areas, a Grant-in-Aid for Scientific
Research B and a Monbusho International Scientific Research Program for
Joint Research to S.H., and by a Grant-in-Aid for Encouragement of Young
Scientists to H.N., from the Ministry of Education, Science and Culture
Abeles,A.L., Friedman,S.A. and Austin,S.J. (1985) J. Mol. Biol., 185,
Austin,S.J. (1984) J. Bacteriol., 158, 742-745.
Begg,K.J. and Donachie,W.D. (1991) New Biol., 3, 475-486.
Bourne,H.R. (1991) Nature, 351, 188-190.
Cronan,J.E.,Jr and Rock,C.O. (1987) In Neidhardt,F.C., Ingraham,J.L.,
Low,K.B., Magasanik,B. and Schaechter,M. (eds), Escherichia coli and
Salmonella typhimurium: Cellular and Molecular Biology. American
Society for Microbiology, Washington, DC, pp. 474-497.
Donachie,W.D. and Begg,K.J. (1989) J. Bacteriol., 171, 5405-5409.
Epstein,W. and Schultz,S.G. (1965) J. Gen. Physiol., 49, 221-234.
Hendrickson,W.G., Kusano,T., Yamaki,H., Balakrishnan,R., King,M.,
Murchie,J. and Schaechter,M. (1982) Cell, 30, 915-923.
Hiraga,S. (1991) In Ishihama,A. and Yoshikawa,H. (eds), Control of Cell
Growth and Division. Japan Sci. Soc. Press, Tokyo/Springer-Verlag,
Berlin, pp. 47-60.
Hiraga,S. (1992) Annu. Rev. Biochem., 61, 283-306.
Hiraga,S., Niki,H., Ogura,T., Ichinose,C., Mori,H., Ezaki,B. and Jaffe,A.
(1989)J. Bacteriol., 171, 1496-1505.
Hiraga,S., Ogura,T., Niki,H., Ichinose,C. and Mori,H. (1990) J. Bacteriol.,
Propertiesof the MukB protein Download full-text
Hiraga,S., Niki,H., Imamura,R., Ogura,T., Yamanaka,K., Feng,J.,
Ezaki,B. and Jaffe,A. (1991) Res. Microbiol., 142, 189-194.
Hussain,K., Begg,K.J., Salmond,G.P.C. and Donachie,W.D. (1987a) Mol.
Microbiol., 1, 73-81.
Hussain,K., Elliott,E.J. and Salmond,G.P.C. (1987b) Mol. Microbiol., 1,
Issartel,J.-P., Koronakis,V. and Hughes,C. (1991) Nature, 351, 759-761.
Jackowski,S. and Rock,C.O. (1987) J. Bacteriol., 169, 1469-1473.
Jackowski,S., Edwards,H.H., Davis,D. and Rock,C.O. (1985) J. Bacteriol.,
Jacob,F., Brenner,S. and Cuzin,F. (1963) Cold Spring Harbor Symp. Quant.
Biol., 28, 329-348.
Kato,J., Nishimura,Y., Yamada,M., Suzuki,H. and Hirota,Y. (1988) J.
Bacteriol., 170, 3967-3977.
Kato,J., Nishimura,Y. and Suzuki,H. (1989) Mol. Gen. Genet., 217,
Kato,J., Nishimura,Y., Imamura,R., Niki,H., Hiraga,S. and Suzuki,H.
(1990) Cell, 63, 393-404.
Kohara,Y., Akiyama,K. and Isono,K. (1987) Cell, 50, 495-508.
Laemmli,U.K. (1970) Nature, 227, 680-685.
Martin,K.A., Friedman,S.A. and Austin,S.J. (1987) Proc. Natl. Acad. Sci.
USA, 84, 8544-8547.
Matsuyama,S., Kimura,E. and Mizushima,S. (1990) J. Biol. Chem., 265,
Meacock,P.A. and Cohen,S.N. (1980) Cell, 20, 529-542.
Mori,H., Kondo,A., Ohshima,A., Ogura,T. and Hiraga,S. (1986) J. Mol.
Biol., 192, 1-15.
Morona,R., Manning,P.A. and Reeves,P. (1983) J. Bacteriol., 153,
Niki,H., Imamura,R., Ogura,T. and Hiraga,S. (1990) Nucleic Acids Res.,
Niki,H., Jaffe,A., Imnamura,R., Ogura,T. and Hiraga,S. (1991) EMBO J.,
Vallee,R.B. (1990) Nature, 347, 256-261.
Ogden,G.B., Pratt,M.J. and Schaechter,M. (1988) Cell, 54, 127-135.
Ogura,T. and Hiraga,S. (1983) Cell, 32, 351-360.
Rock,C.O. and Cronan,J.E.,Jr. (1979) J. Biol. Chem., 254, 9778-9785.
Sato,K., Nishina,Y. and Shiga,K. (1992) J. Biochem., 111, 359-365.
Shpetner,H.S. and Vallee,R.B. (1989) Cell, 59, 421-432.
Shpetner,H.S. and Vallee,R.B. (1992) Nature, 355, 733-735.
Steck,T.R. and Drlica,K. (1984) Cell, 36, 1081-1088.
Takagi,T. (1981) J. Biochem., 89, 363-368.
Takagi,T., Miyake,J., and Nashima,T. (1980) Biochim. Biophys. Acta, 626,
Tucker,W.T., Miller,C.A. and Cohen,S.N. (1984) Cell, 38, 191-201.
Tyler,J.M. and Branton,D. (1980) J. Ultrastruct. Res., 71, 95-102.
Vanaman,T.C., Waild,S.J. and Hill,R.L. (1968) J. Biol. Chem., 243,
Vinella,D., Jaffe,A., D'Ari,R., Kohiyama,M. and Hughes,P. (1992) J.
Bacteriol., 174, 2388-2390.
Wandersman,C. and Delepelaire,P. (1990) Proc. Natl. Acad. Sci. USA,
Yamanaka,K., Ishihama,A. and Nagata,K. (1990) J. Biol. Chem., 265,
Yoshimoto,M., Kambe-Honjoh,H., Nagai,K. and Tamura,G. (1986) EMBO
J., 5, 787-791.
Received on August 3, 1992