Plant Cell Physiol. 43(12): 1558–1567 (2002)
JSPP © 2002
Molecular Properties of a Matrix Attachment Region-Binding Protein
Located in the Nucleoli of Tobacco Cells
Shiori Fujiwara 1, Nao Matsuda 1, Tomohiro Sato 1, Seiji Sonobe 2 and Masayoshi Maeshima 1, 3
1
2
Laboratory of Cell Dynamics, Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya, 464-8601 Japan
Faculty of Science, Himeji Institute of Technology, Harima Science Park City, Hyogo, 678-1297 Japan
;
We cloned a cDNA for matrix-attachment region
(MAR)-binding protein from Nicotiana tabacum cells to elucidate the structure and function of the nuclear matrix. The
cDNA encodes a protein of 555 amino acids (61,050 Da)
with an isoelectric point of 9.4. We named the protein
NtMARBP61. The sequence is 45% identical to yeast
Nop58p, which is involved in rRNA processing. The Cterminal part is unique and rich in lysine residues. The
recombinant C-terminal part had the ability to bind double-stranded DNAs of 12 tobacco MARs. The intracellular
localization was determined to be in the nucleolus by fluorescent microscopy using the antibody to the recombinant
NtMARBP61. The mRNA level was high in the lag and
early-log phases of cultured cells but low in the stationary
phase. The protein was accumulated only in the middleand late-log phases, suggesting that NtMARBP61 is essential for growing cells. The results suggest at least the structural and regulatory function of NtMARBP61 in the nucleolus as a MAR-binding protein in a growth-stage specific
manner.
Keywords: DNA-binding protein — Matrix attachment regionbinding protein — Nuclear matrix — Nucleolus — Tobacco.
Abbreviations: DAPI, 4¢,6-diamidino-2-phenylindole; GST, glutathione S-transferase; IPTG, isopropyl thio-b-D-galactoside; MAR,
matrix-attachment region of DNA; NtMARBP61; Nicotiana tabacum
MAR-binding protein of 61 kDa.
The nucleotide sequence reported in this paper has been submitted to the DDBJ, EMBL, and GenBank under accession number
AB059832 (NtMARBP61).
Introduction
The nuclear matrix has been defined as the insoluble
material left in the nucleus after sequential extraction with nonionic detergent, DNase and high concentrations of salt. This
matrix generally consists of a nuclear lamina, nuclear pore
complexes, nucleoli, and an internal fibrogranular network of
proteins. The major function of the nuclear matrix is to provide a structural framework upon which active DNA replication centers and transcription complexes are assembled. Chro3
mosomal DNAs have a large number of specific DNA
sequences called MAR (matrix-associated region) (Jenuwein et
al. 1997). MAR DNAs are more than 70% AT-rich with a
length of 200–2,000 bp. The MAR sequences occur in a ratio
of one motif per 50–200 kb. These DNA sequences have been
postulated to form the base of chromosomal loops, or to attach
chromosomes to the nuclear matrix (Holmes-Davis and Comai
1998, Holmes-Davis and Comai 2002, Laemmli et al. 1992).
Several kinds of proteins specifically bind to the MAR
DNA sequences and are called MAR-binding proteins
(MARBPs) (Dickinson and Kohwi-Shigematsu 1995,
Dworetzky et al. 1992). MARBPs recognize MARs in spite of
a lack of a consensus sequence. Many MARBPs have been isolated and characterized. MARBPs include several groups: (i)
MFP (MAR-binding filament-like protein) family (Harder et al.
2000, Meier et al. 1996, Meier and Gindullis 1999, Meier et al.
2000), (ii) AHM (AT hook-containing MAR binding protein)
family (Morisawa et al. 2000), (iii) Nop (nucleolar protein)
family, and (iv) others. The fourth group includes high mobility group chromosomal proteins (Reeves and Nissen 1990),
topoisomerase II (Adachi et al. 1989), histone H1, lamin B1,
and nucleolin. Members of the Nop family, such as Nop1p
and Nop5p (Nop58p) (Wu et al. 1998), have been identified
as components of the small nucleolar ribonucleoprotein complex (snoRNP) that is involved in RNA splicing (Lewis and
Tollervey 2000). Nop1p has been demonstrated to associate
with Nop5p in yeast snoRNPs (Wu et al. 1998). Nop1p and its
homologue, which is known as fibrillarin, have been reported
to participate in different aspects of ribosome biogenesis,
including pre-rRNA modification, processing, and ribosome
subunit assembly (Lewis and Tollervey 2000). Nop56p and
Nop58p of yeast (Gautier et al. 1997) and their orthologs of pea
(Hatton and Gray 1999) have been demonstrated to bind MAR
DNAs. To distinguish these proteins (Nop56p, Nop58p, and
their related proteins) from Nop1p and others, we call them the
Nop5 type.
A large number of protein components of gene expression
machinery, histones and the nuclear envelopes in the nuclei
have been extensively investigated (Blobel and Wozniak 2000,
Maniatis and Reed 2002, Reichheld et al. 1998). MARBPs are
also essential for understanding the nuclear function, namely,
structural organization, regulation of gene expression and
ribosome biogenesis. We are conducting a series of experi-
Corresponding author: E-mail, [email protected]; Fax, +81-52-789-4096.
1558
Tobacco MAR-binding protein
1559
Fig. 1 Alignment of deduced amino acid sequence of NtMARBP61 with the related sequences. The sequence of NtMARBP61 (Nt) was deduced
from a cDNA (AB087677) and compared with the sequences of the related proteins from Pisum sativum (Ps, MARBP-1, AF061962), A. thaliana
(At, Nop58-like protein, AF302491), S. cerevisiae (Sc, Nop58p, AF056070), and Homo sapiens (Hs, human Nop58, AF123534). Boldface letters
indicate the identical residues among five sequences. A possible nuclear localization signal is boxed. The Lys-rich region in the C-terminal part is
underlined. NtMARBP61 was divided into the N-terminal (N, 1–443) and the C-terminal parts (C) at the indicated site. The underlined amino
acid sequence (a) was identical to that of the direct sequence obtained from the protein purified from the tobacco nuclear matrix fraction. We prepared primers corresponding to the underlined sequence b (EWYGWHFP) for cDNA cloning.
ments to isolate and characterize the nuclear matrix proteins.
During the experiments, we isolated cDNAs for MARBP. In
this paper we describe the characterization and molecular cloning of cDNA encoding a tobacco MARBP, which seems to be
related to the Nop5 type. Interestingly, this protein was accumulated into the nucleoli in a growth-stage-specific manner.
Results
Isolation of a tobacco MARBP cDNA
To characterize the nuclear proteins, we subfractionated
the isolated nuclei into the extracts with 2 M NaCl and with
8 M urea. The proteins in the fractions were subjected to SDSPAGE and sequence analysis of the N-terminal part. A protein
in the urea extract corresponding to nucleolin was identified
and its cDNA was cloned (DDBJ/EMBL/GenBank accession
no. AB087677). In addition to nucleolin, the urea extract con-
1560
Tobacco MAR-binding protein
Fig. 2 Constructs of recombinant NtMARBP61 proteins. Four constructs were prepared: (F-His), full-length NtMARBP61 (555 residues) linked with a His tag; (GST-N-His), N-terminal part (1–443)
linked with GST and a His tag; (N-His), N-terminal part linked with a
His tag; (GST-C-His), C-terminal part (444–555) linked with GST and
a His tag. Apparent molecular masses on SDS-PAGE are shown on the
right.
tained a protein with an N-terminal sequence that is similar to
MARBPs of pea (Hatton and Gray 1999) and rice (AB015431).
Thus, we prepared degenerated primers that correspond to the
N-terminal sequence and internal conserved motif (EWYGWHFP) to obtain a cDNA fragment from the cDNA library of
tobacco BY-2. Three positive clones were obtained from the
library (4.5´105 pfu) and had the same nucleotide sequence.
The cloned cDNA of 1,665 bp encodes a protein of 555
amino acids (Fig. 1). The deduced sequence is 75% identical to
pea MARBP-1 (AF061962), 70% identical to Arabidopsis thaliana Nop58-like protein (AF007271), 45% identical to Saccharomyces cerevisiae Nop58p/Nop5p (X90565), and 50% identical to human Nop58p (AF123534). The molecular size of the
protein was calculated to be 61,050 Da. Thus we have named
this protein Nicotiana tabacum MARBP61 (NtMARBP61).
A marked characteristic of NtMARBP61 is the high content of charged amino acid residues (Lys, 79 residues; Arg, 19;
His, 11; Glu, 51; Asp, 35). The pI value was calculated to be
9.4. This protein has a basic C-terminus (475th to 555th, 112
residues) of pI 9.6, in which there are three clusters with more
than six Lys residues. A C-terminal sequence of RKHADEDEAETTDKKKD may serve as a nuclear localization signal,
since the sequence is a typical motif of the bipartite nuclear
localization signal (underlined residues) (Dehesh et al. 1995,
Lian and Clarke 1999). This protein does not contain a Zn-finger motif, Leu zipper, or repeated motif of KKE/D, which are
detected in several MARBPs. A typical plant RNA-recognition motif, which was found in small nuclear ribonucleoproteins and heterogeneous nuclear ribonucloproteins (Alba and
Pages 1998), is absent in NtMARBP61.
Preparation of recombinant NtMARBP61 proteins and
polyclonal antibodies
Recombinant NtMARBP61 proteins were prepared to
examine DNA-binding activity and to prepare the polyclonal
antibodies. Four constructs of chimeric proteins with GST and
Fig. 3 Preparation of recombinant NtMARBP61 proteins and the
antibody. (A) SDS-PAGE of GST-N-His. The cell extract of E. coli
was separated into the soluble (lane 1) and insoluble (lane 2) fractions
(30 ml of cell culture). The purified GST-N-His (3 mg) was applied to
lane 3. Gel was stained with Coomassie brilliant blue R. The arrowhead indicates the position of GST-N-His. (B) SDS-PAGE of GST-CHis. Lane 1, the cell extract (soluble fraction) of E. coli expressing the
protein; lane 2, GST-C-His purified by using glutathione agarose. The
arrowhead indicates the position of GST-C-His. (C) Immunoblot of the
recombinant proteins (lanes 1 and 2, GST-N-His; lanes 3 and 4, GSTC-His) with anti-NtMARBP antibody. E. coli was cultured in the presence (lanes 2 and 4) or absence (lanes 1 and 3) of IPTG. (D) Immunoreaction of the anti-NtMARBP with tobacco nuclear fraction. Lanes 1
and 3, the purified GST-N-His (14 ng); lanes 2 and 4, the nuclear
matrix fraction of tobacco BY-2 cells (50 mg of protein). Arrowhead
indicates the position of NtMARBP61 in nuclei (70 kDa).
a His-tag were prepared and expressed in Escherichia coli (Fig.
2). The GST-N-His protein (1–443) migrated at 74 kDa in
SDS-PAGE (Fig. 3A). The GST-C-His protein migrated as a
50-kDa protein. In the case of GST-N-His, the recombinant
protein was recovered in the inclusion bodies (insoluble fraction). Therefore, the protein was solubilized in 8 M urea and
Tobacco MAR-binding protein
1561
Table 1 Properties of 12 tobacco MAR DNAs (Michalowski
et al. 1999)
Name
MAR1
MAR2
MAR4
MAR5
MAR6
MAR7
MAR8
MAR9
MAR10
MAR11
MAR12
MAR13
a
Length (bp)
AT (%) a
Accession no.
437
866
635
1,087
704
306
685
899
999
1,499
587
383
71.8
65.7
68.3
64.2
71.4
77.1
71.7
65.2
62.0
73.1
72.2
61.6
AF065877
AF065878
AF065880
AF065881
AF065882
AF065883
AF065884
AF065885
AF065886
AF065887
AF065888
AF065889
% of AT content in MAR DNA.
purified by metal affinity resins (Fig. 3A). Polyclonal antibodies were prepared against the purified GST-N-His protein and
were routinely used in this study. Therefore, this anti-GST-NHis antibody recognizes the N-terminal part of NtMARBP61,
GST and a His tag. As shown in Fig. 3C, the antibody reacted
with the GST-N-His of 74 kDa and the GST-C-His at 50 kDa.
The synthesis of these recombinant proteins in E. coli was
induced by IPTG (Fig. 3C). The GST-N-His protein was also
recognized by both anti-His and anti-NtMARBP antibodies,
while the corresponding protein of 70 kDa in the nuclear fraction from BY-2 cells was reacted only with anti-NtMARBP
antibody (Fig. 3D).
The sizes of NtMARBP61 (70 kDa) and the GST-C-His
(48 kDa) observed in SDS-polyacrylamide gels were larger
than the predicted values of 61 kDa and 37 kDa, respectively.
The recombinant C-terminal part (444–555; estimated size,
13 kDa) without GST or His migrated to 25 kDa (data not
shown). Thus, the difference between the calculated and
observed molecular masses is likely due to the highly charged
C-terminal region.
DNA binding property of NtMARBP61
Twelve MAR DNAs were prepared from tobacco DNA by
PCR (Table 1). Fig. 4 shows the reaction of the full-length
NtMARBP61 to the MAR13 DNA. A negative control cDNA
for ornithine decarboxylase (ODC) was also bound to the protein but the reaction was cancelled by cold lDNA that was
added to prevent non-specific binding. The interaction of
NtMARBP61 with MAR13 was retained in the presence of
10 mg ml–1 lDNA. This result indicates a specific interaction
between NtMARBP61 and MAR13.
The binding capacity of the N-terminal and C-terminal
fragments of NtMARBP61 to MAR DNAs was examined by
SDS-PAGE of the purified proteins and affinity blotting with
radiolabeled MARs (Fig. 5). These MARs were not denatured
Fig. 4 DNA binding assay of recombinant NtMARBP61. Purified
preparation (50 mg per lane) of recombinant NtMARBP61 (F-His) was
applied to SDS-PAGE and subsequent DNA-binding assay. Proteins
were transferred to a membrane and the membrane sheet was overlaid
with 32P-labeled double-stranded MAR13 DNA (1 pmol ml–1) and
cDNA of ODC (non-MAR DNA, 10 ng ml–1). For the competition
assay, lambda phage DNA was added at final concentrations of 1.0
(lane 2) and 10 mg ml–1 (lane 3). The arrowheads indicate the position
of F-His (70 kDa).
and were used as double-stranded DNAs. MAR12 was bound
with the full-length NtMARBP61 (F-His, 70 kDa) and the Cterminal fragment (GST-C-His, 48 kDa). The signal was
observed even in the presence of a high concentration of
lDNA. However, the extracts of E. coli containing the vectors
(pET24-b and pMAT311) and the N-terminal fragment (GSTN-His, 50 kDa) did not bind MAR12 (Fig. 5A). In the case of
MAR1, MAR6, MAR8 and MAR10 (Fig. 5B–E), the F-His
and GST-C-His proteins were also bound to MARs, although
the signal intensity depended on the kind of MARs. The other
MARs in the 12 preparations were also bound to the F-His and
GST-C-His proteins (data not shown). The results indicate that
NtMARBP61 binds tobacco MAR DNAs with a broad specificity through the C-terminal basic region.
The nuclear matrix fraction gave a doublet of 70- and 75kDa MAR-labeled bands (Fig. 5A, lane 4). NtMARBP61 may
correspond to the band of 70 kDa that reacted with the antiNtMARBP antibody. The 75-kDa band may be another type of
MARBP in tobacco nuclei.
Immunolocalization of NtMARBP61 in tobacco cells
NtMARBP61 was detected in the nuclear matrix fraction
by immunoblotting (Fig. 3) and was confirmed to bind MAR
DNAs (Fig. 4, 5). Immunocytochemical analysis was carried
out to examine the intracellular localization of NtMARBP61
(Fig. 6). NtMARBP61 was visualized with anti-NtMARBP
antibody and the Cy3-labeled second antibody. All nuclei that
were stained blue with DAPI showed a red color of Cy3, indicating the nuclear localization of NtMABP61. In particular,
nucleoli in the nucleus in all cells were clearly immunostained
with the anti-NtMARBP antibody (Fig. 6B, D). The nuclei
were not stained with the pre-immune serum and the anti-GST
antibody (Fig. 6F, H). Thus, the fluorescent signal observed in
nucleoli was not due to GST or non-specific immunoreactions.
Fig. 7 shows a metaphase cell immunostained with anti-
1562
Tobacco MAR-binding protein
7th day). This was confirmed by using two different probe
DNAs (probe N and probe C) for NtMARBP61. On the other
hand, the NtMARBP61 protein was not detected for the first
3 d. In this experiment, the nuclear fractions with an equal
amount of DNA (Fig. 8B) or protein (Fig. 8C) were applied to
immunoblotting. Thus, Fig. 8B shows the relative level of
NtMARBP61 protein per cell. Fig. 8C shows the relative content of the protein in the nuclear fraction. In both cases, the
protein accumulated in the 4- and 5-day-old cells and then disappeared on day 6. By contrast, histone H1 was accumulated in
lag and early-log phases (from the 1st to 3rd day) and then
decreased as shown in Fig. 8D. This is consistent with a previous observation that the histone proteins were accumulated in
the S-phase cells of tobacco (Reichheld et al. 1998). The
present results indicate that NtMARBP61 was actively synthesized and accumulated in the middle- and late-log phases. The
amount of NtMARBP61 on the 4th day was roughly calculated
to be 0.02 mg mg–1 of the nuclear protein from the intensity of
the band compared with that of the purified recombinant protein (Fig. 8C).
Discussion
Fig. 5 Binding capacity of the C-terminal part of NtMARBP61 to
MAR DNAs. The recombinant proteins of F-His (Full, lane 2) and
GST-C-His (C, lane 3) were subjected to DNA-binding assay with 32Plabeled double-stranded DNA (1 pmol ml–1) of MAR12 (A, 0.39 mg
ml–1), MAR1 (B, 0.29 mg ml–1), MAR6 (C, 0.47 mg ml–1), MAR8 (D,
0.45 mg ml–1) and MAR10 (E, 0.66 mg ml–1). The E. coli cell extract
containing only the vector of pET24 (lane 1) or pMAT (lane 5) was
applied. The nuclear matrix fraction (Nuc) was applied to lane 4. The
recombinant protein of N-His (N) was applied to lane 6. The protein
amount applied to each lane was 50 mg. DNA-binding analysis was
done in the presence of lDNA (10 or 100 pmol ml–1). The arrowheads
indicate the position of NtMARBP61 (70 kDa) and the closed circles
indicate the position of GST-C-His (48 kDa). The upper bands
(75 kDa) observed in the nuclear matrix fraction (lane 4) are marked
by short lines (A).
NtMARBP antibody. At this stage of the cell cycle, the nuclear
envelopes were broken down and the condensed chromosomes
appeared (arrowhead, Fig. 7B). At metaphase, the NtMARBP61
protein seemed to have disappeared and/or diluted in the cell,
although the protein was clearly detected in nucleoli in neighboring cells at the prophase.
Changes in the level of NtMARBP61 during growth of cells
The mRNA and protein levels of NtMARBP61 in the cultured cells were determined by Northern and immunoblot analysis. The mRNA level of NtMARBP61 was constant for the
first 4 d after subculture and then markedly decreased on day 5
(Fig. 8A). The low level was kept in the mature stage (6th and
NtMARBP61 has a similarity to the Nop type MARBPs in the
primary sequence
Several MARBPs have been cloned from plants. Some of
them have been grouped into the MFP, AHM, and Nop families. In addition to these proteins, high mobility group chromosomal proteins (Reeves and Nissen 1990), topoisomerase II
(Adachi et al. 1989), histone H1, lamin B1, and nucleolin are
also included in the MARBPs. The MFP family members have
characteristics of a filamentous structure with an extended ahelical, coiled-coil-like domain. The members possess the
casein kinase phosphorylation site (T/SXXE) and a DNAbinding a-helical domain in the C-terminal part (Harder et al.
2000, Meier et al. 1996). It has been proposed that MFPs are
related to the nuclear lamins in animal nuclear matrix. AHM
family members possesses an AT hook, which interacts with
the minor groove of AT-rich stretches of DNA double strands
(Morisawa et al. 2000). In addition to the AT hook in the Cterminus, the proteins have both a J domain-homologous
region and a Zn finger-like motif in their internal parts. The J
domain is known to interact with Hsp70 chaperones and is
found in the DnaJ family proteins. NtMARBP61 does not
possess these characteristics.
NtMARBP61 resembles Nop5-type proteins from humans
and yeast in the primary sequences. Thus, NtMARBP61 may
be a member of Nop5 type proteins. The N-terminal and central
parts shows a high sequence identity of more than 43% among
the Nop5-type proteins from plants, yeast and human. There
are several long consensus motifs with more than 10 residues,
such as a LLDDLDKELNNYAMRVREWYGWHFPELAKI
motif. These motifs are peculiar to the Nop5 type including the
proteins from Pyrobaculum aerophilum (AE009919) and
Tobacco MAR-binding protein
1563
Fig. 6 Immunolocalization of NtMARBP61 in tobacco cells. Tobacco BY-2 cells subcultured for 4 d were harvested, fixed with formamide, and
then treated with anti-NtMARBP (B and D), control serum (F) and anti-GST (H) antibodies. The antigen was visualized with Cy3-coupled goat
anti-(rabbit IgG) antibody (B, D, F and H). The same cells were treated with DAPI to visualize nuclei (A, C, E and G). Images of Cy3 and DAPI
fluorescence were observed under a fluorescent microscope. Bar = 20 mm.
Sulfolobus solfataricus (AE006714) but are not found in other
proteins. The biochemical function of the motifs remains to be
determined.
It should be noted that NtMARBP61 does not possess
KKE/D repeat sequences, present in the N-terminal part of the
Nop58p of pea (Hatton and Gray 1999) and yeast (Gautier et
al. 1997). In yeast, the KKE/D domain has been thought to
confer microtubule-binding activity, which may be essential for
the accurate segregation of the rRNA processing machinery.
DNA binding through the C-terminal basic region
The notable property of NtMARBP61 and other Nop58p
proteins is the Lys-rich sequence in the C-terminal part. There
is no consensus sequence in the C-terminal region among the
Nop5-type proteins and the length of this part varies with the
proteins. By using the recombinant proteins, we demonstrated
that NtMARBP61 binds the double-stranded MAR DNAs
through the C-terminal region (444–555, Fig. 1). The recombinant N-terminal part (1–443) did not exhibit the MAR-DNA
binding activity in the present experiments. Both the full length
and C-terminal recombinant proteins showed a similar intensity in MAR-DNA binding experiments. Therefore, we estimate that the N-terminal region does not affect the substrate
specificity. At present, we cannot explain how the divergent
sequence in the C-terminal region recognizes the MAR DNAs.
MAR DNAs are known to occur in non-coding DNA and gen-
1564
Tobacco MAR-binding protein
Fig. 7 Immunostaining of the cell in metaphase with anti-NtMARBP
antibody. (A) NtMARBP61 was visualized by using anti-NtMARBP
antibody and Cy3-coupled second antibody. (B) Nuclei were stained
with DAPI. The arrowhead indicates the stationary chromosomes
aligned at metaphase plate.
erally have a base composition that is more than 70% AT-rich.
DAPI is also known to bind to the AT-rich region of the chromosomal DNA. Thus, it is reasonable that the fluorescent
image of the DAPI-staining in tobacco cells is similar to that of
immunostaining with the anti-NtMARBP antibody.
Nucleolus localization of NtMARBP61 and phase-specific
accumulation
Intracellular location of NtMARBP61 was examined by
the immunocytochemical analysis. The anti-NtMARBP antibody stained nuclei and, more strongly, nucleoli of tobacco
cells. Thus it was concluded that NtMARBP61 is accumulated
predominantly in the nucleolus in the nuclei. The nucleolus is a
small, dense body present in the nucleus of eukaryotes, singly
or in replicate. The nucleolus is rich in RNA and protein and is
not seen during mitosis. Its prime function is the transcription
of the nucleolar DNA into 45S ribosomal-precursor RNAs,
their processing, and the association of these processed rRNAs
with proteins synthesized outside the nucleolus.
The Nop5-type MARBPs have been demonstrated to bind
to precursors of rRNAs (Gautier et al. 1997). At present, structural properties of rRNA-binding motifs are not clear.
NtMARBP61 binds at least several MAR DNAs, although
whether or not NtMARBP61 binds pre-rRNAs has not been
determined. There is an example of a nuclear-matrix protein
binding both DNA and RNA (Fackelmayer et al. 1994). Pea
Nop58-type proteins (MARBP1 and MARBP2) in the nuclear
matrix have been reported to bind a MAR DNA (Hatton and
Gray 1999). MAR DNAs are known to occur in non-coding
DNA and generally have an AT-rich base composition. We estimated that NtMARBP61 has a high affinity for the AT-rich
Fig. 8 Changes in levels of mRNA and protein of NtMARBP61 during subculture of BY-2 cells. Total RNA and crude nuclear fractions
were prepared from tobacco cells at each stage of subculture. (A)
Northern hybridization analysis. Aliquots of total RNA (20 mg) were
subjected to Northern hybridization analysis with 32P-labeled probe N
(554 b) and probe C (336 b). The agarose gel was stained with ethidium bromide to detect rRNAs. (B and C) Immunoblot analysis of
NtMARBP61. Aliquots of the nuclear fraction (0.8 mg of DNA for
panel B, 80 mg of protein for panel C) were subjected to SDS-PAGE
and subsequent immunoblotting with anti-NtMARBP. For example,
the protein content in the nuclei on day 4 was 30 mg per 0.8 mg of
DNA. The purified GST-N-His (14 ng) was applied to lane 1 in both
cases. The positions of GST-N-His (open arrowhead, 74 kDa) and
native NtMARBP61 (closed arrowhead, 70 kDa) are indicated. (D)
Immunoblot analysis of histone H1. Aliquots of the nuclei (5 mg of
protein) were subjected to immunoblotting with polyclonal antibodies
to tobacco histone H1.
DNA sequences located in the nucleolus. With respect to the
nucleolar localization and its structural similarity to the Nop5
type proteins, there is a possibility that the protein may be
involved in the rRNA processing in growing cells in the same
way as Nop58 protein in yeast. It should be examined whether
or not NtMARBP61 binds to pre-rRNAs rather than MAR
DNAs.
In this study we demonstrated a reciprocal pattern of protein accumulation between NtMARBP61 and histone H1 during cell growth (Fig. 8). The decrease in the NtMARBP61 level
Tobacco MAR-binding protein
in the stationary phase is consistent with the mRNA level (Fig.
8). In the lag phase, however, the protein was not accumulated
even though the mRNA level was high. There must be a negative regulation of translation and/or a rapid turnover of the protein in the early phase. As another possibility, NtMARBP61
protein translated at the lag phase may be dispersed in the
cytosol. In this study, we analyzed the protein preparation
recovered as the nuclear matrix by immunoblotting. However,
it should be noted that we could not detect a clear signal in the
cytosolic space of the log-phase cells by immunocytochemistry. In any case, the transient accumulation of NtMARBP61 in
nuclei may be closely related to its physiological function in
tobacco cells. If NtMARBP61 is associated with ribosome biogenesis, such as rRNA processing, in growing cells, it is reasonable that the protein level is high in the middle- and late-log
phases, because ribosomes are actively synthesized in growing
cells.
In conclusion, we found Nop5-type MARBP in tobacco
cells, which is accumulated in the nucleoli in a log-phasespecific manner. This NtMARBP61 binds tobacco MARs. With
consideration of the primary sequence and the nucleolar localization, NtMARBP61 may have multifunctional roles in rRNA
processing and chromatin organization.
Materials and Methods
Plant materials
Tobacco (Nicotiana tabacum cv. BY-2) cells were used for preparation of RNAs, DNAs, and nuclei. Cells were subcultured at intervals
of 7 d (Matsuoka et al. 1997). The culture medium contains
Murashige-Skoog salt mixture, 1 mg liter–1 thiamine-HCl, 100 mg
liter–1 myo-inositol, 0.2 mg liter–1 2,4-D, and 3% sucrose, pH 5.5.
Isolation of nuclei and subfractionation
Cells subcultured for 4 d were washed with 0.4 M mannitol. Protoplasts were prepared by the method of Nagata (1987). Protoplasts
were suspended with 1.5 volumes of nuclear isolation buffer (NIB)
containing 0.5 M hexylenglycol, 10 mM HEPES-KOH, pH 7.0,
10 mM MgCl2, 1 mM EGTA, 10 mM 2-mercaptoethanol, 10 mM
APMSF, 0.5% Triton X-100, and 0.5% polyvinylpyrrolidone. The suspension was homogenized with a Polytron, and then centrifuged at
1,000´g for 5 min. The precipitate was washed with NIB and then suspended with a nuclear resuspension buffer containing 1 M hexylenglycol, 10 mM HEPES-KOH, pH 7.0, 10 mM MgCl2, 1 mM EGTA,
10 mM 2-mercaptoethanol, 1 mM APMSF, and 20% (v/v) glycerol.
After centrifugation at 1,000´g for 5 min, the precipitate was resuspended with the resuspension buffer. The suspension (10 ml) was layered on top of a step gradient of Percoll. The Percoll gradient consisted
of a 15-ml layer of 50% (w/v) Percoll, 10-ml layer of 90% Percoll, and
10-ml layer of 80% sucrose. Percoll solutions contained 28.5%
sucrose, 10 mM HEPES-KOH, pH 7.0, 10 mM MgCl2, 1 mM EGTA,
and 10 mM 2-mercaptoethanol. The gradients were centrifuged with a
swinging bucket rotor at 2,800´g for 20 min. The interface portion
between 90% and 50% Percoll solutions was collected as the nuclear
fraction. The nuclei were washed twice with the resuspension buffer.
The nuclei were washed with 10 mM BisTris-HCl, pH 6.5,
0.1 mM MgCl2, and 20% (v/v) DMSO, and then treated with 70 units
ml–1 of DNase I at 37°C for 30 min. The suspension was centrifuged at
1,000´g for 10 min, and then suspended with a high salt buffer (2 M
1565
NaCl and 10 mM BisTris-HCl, pH 6.5). The nuclear suspension was
incubated at 0°C for 30 min with gentle shaking. After centrifugation
at 2,800´g for 10 min, the supernatant was removed and stored as the
2 M NaCl extract. The precipitate was suspended with 1 M NaCl and
50 mM Tris-HCl, pH 9.0, and centrifuged at 2,800´g for 10 min. The
precipitate was suspended with 8 M urea, 50 mM Tris-HCl, pH 9.0,
incubated at 4°C for 30 min, and then centrifuged at 2,800´g for
10 min. The supernatant (urea extract) was pooled and used as the
nuclear matrix.
cDNA cloning
Total RNA was isolated from BY-2 cells cultured for 4 d to construct a cDNA library. Cells (1 g FW) were washed and frozen in liquid nitrogen, and then RNA was extracted by the phenol/SDS extraction method. RNA was precipitated with 2 M LiCl and then with cold
70% ethanol and 100 mM NaCl. RNA was resuspended in 10 mM
Tris-HCl, pH 8.0, and 1 mM EDTA. The poly (A) mRNAs were isolated from the RNA fraction with oligo (dT)-latex (Oligotex-dT30,
Roche), and converted to cDNAs using reverse transcriptase and oligodT adapter primer (ZAP-cDNA synthesis kit, Stratagene). During the
experiment, we found that the partial sequence (MLVLFETPAGFALFK) of a protein in the 8 M-urea extract of the tobacco nuclei was
coincident with the N-terminal sequences of MARBPs of pea and rice
(ref. Fig. 1, underlined sequence a). Thus the corresponding DNA was
amplified from the tobacco cDNA library by PCR using oligonucleotide primers (forward, TTTGAAACGCCCGCGGGAT; reverse,
AAGTGCCATCCGTACCACTC) based on a conserved sequence (ref.
Fig. 1, underlined sequence b) in the MAR-binding proteins of rice
(AB015431) and pea (AF061962). Amplified DNA fragments
(556 bp) were purified and ligated into the EcoRV site of pBluescript
SK(–) plasmid vector for transformation of E. coli MV1184. This
DNA fragment was labeled and used as a probe for isolation of a fulllength cDNA. Recombinant phages (5.0´104 pfu) were blotted to
nylon membranes (Biodyne, Pall) and then screened with the 32Plabeled DNA probe (Sambrook and Russell 2001). The DNA
sequences were determined with an ABI PRISM 310 capillary
sequencer and then aligned using the DNASIS program (Hitachi Software Engineering).
Preparation of tobacco MAR DNAs
The information of MAR DNA sequences of tobacco is available on the DNA database (Michalowski et al. 1999) (Table 1). MAR
DNAs were directly amplified by PCR from the tobacco DNA preparation using the corresponding DNA primers and LA Taq (Takara).
The amplified fragments were purified, ligated into plasmid vector
pZEro2.1 (Novagen), and then sequenced. The amplified DNAs were
used for the DNA-binding assay. A partial fragment of cDNA for
tobacco ODC was used as a negative control. MAR13 DNA and
cDNA of ODC were labeled with a-[32P]dCTP (110 TBq mmol–1,
Amersham) using a Megaprime DNA labeling kit (Amersham). Other
MAR DNAs were labeled with g-[32P]dATP (110 TBq mmol–1) using
T4 polynucleotide kinase (Takara). These samples were not denatured
and used as double-stranded DNAs for DNA-binding assay.
Preparation of recombinant proteins of NtMARBP61 and DNA binding
assay
Recombinant proteins of NtMARBP61 with a (His)6 tag; the fulllength sequence (F-His), the N-terminal part (N-His), and the Cterminal part (GST-C-His) were produced in E. coli (ref. Fig. 2). The
cDNAs for NtMARBP61 and N-terminal part were ligated into plasmid vector pET24-b (Novagen), which contained a (His)6 tag, and the
cDNA for the C-terminal into a vector pMAT311, which was prepared
by Dr. Ken Matsuoka, Nagoya University, and then transformed into
1566
Tobacco MAR-binding protein
an E. coli BL21(DE3)pLYsS strain using an Eppendorf electropolater
2510. IPTG (1 mM) was added to the culture medium to induce the
production of the recombinant proteins and then incubated at 20°C for
6 h (F-His), 37°C for 5 h (N-His), and 30°C for 5 h (C-His), respectively. Cell extracts were prepared from the transformed E. coli after
washing and used for the following experiments.
For the DNA binding assay, the cell extracts were subjected to
SDS-PAGE and then transferred onto nitrocellulose membranes (PROTRAN, Schleicher & Schuell) in 25 mM Tris, 192 mM glycine, 0.02%
SDS and 10% methanol. The membrane sheets were treated with 5%
defatted milk, 10 mM Tris-HCl, pH 7.5, 50 mM NaCl, and 2 mM
EDTA for 1 h. Then the membrane was incubated with 0.5% defatted
milk, 10 mM Tris-HCl, pH 7.5, 50 mM NaCl, 2 mM EDTA, and the
32
P-labeled MAR DNA (1 pmol ml–1). MARs were used as doublestranded DNAs for this assay. For the control experiments, lDNA
(Takara) was added into the reaction medium at 1, 10 or 100 nmol ml–1.
After incubation for 2 h at 25°C, the membrane was rinsed four times
with the same buffer at 25°C and then its radio-image was analyzed
with a Fuji image analyzer BAS 2500.
Preparation of antibody to NtMARBP61 and immunoblotting
For the benefit of protein isolation and detection, a cDNA for the
fusion protein of GST/NtMARBP61 (1–443)/His-tag (GST-N-His, ref.
Fig. 2) was constructed using a plasmid vector pMAT311. The
sequence of the construct was confirmed and then transformed into E.
coli BL21(DE3)pLysS cells as described above. The recombinant protein was induced by 1 mM IPTG for 5 h at 37°C. Cells were rinsed,
suspended in 20 mM Tris-HCl, pH 8.0, 100 mM NaCl, 8 M urea, and
1 mM PMSF (lysis buffer), and then sonicated. After centrifugation for
20 min at 10,000´g, the precipitate was suspended and mixed with
10% volume of TALON metal affinity resin (Clontech) and incubated
for 1 h at 25°C with gentle agitation. The resin was packed in a column and washed. Then the recombinant protein was eluted with
20 mM Tris-HCl, pH 8.0, 100 mM NaCl, 8 M urea, and 100 mM imidazole. The peak content fractions were collected and dialyzed against
2 liters of PBS (4.3 mM Na2HPO4, 1.4 mM KH2PO4, 2.7 mM KCl and
137 mM NaCl) containing 10% glycerol and 1 mM PMSF. The purified protein (4 mg) was injected into two rabbits. The obtained antibody was used as the anti-NtMARBP antibody. The antibody reacted
not only with the NtMARBP61 protein but also with the His tag and
GST. Thus, in some experiments the antibody was pre-treated with the
isolated recombinant GST prior to immunoblotting and immunocytochemistry. The antibody to GST was also prepared by injecting the
purified recombinant GST. Polyclonal antibodies to histone H1 was
prepared by injecting the purified histone H1 of tobacco nuclei into
rabbits.
SDS-PAGE was carried out in a 12% acrylamide gel. The proteins were transferred from the gel to a poly (vinylidene difluoride)
membrane (Immobilon, Millipore), and then the antigen on the membrane was detected with the primary antibodies, horseradish peroxidase-coupled protein A and chemiluminescence reagents (Amersham).
Immunocytochemical analysis
Tobacco BY-2 cells were collected by centrifugation, and rinsed
with PBS. BY-2 cells were fixed by incubating with 4% formaldehyde
in PBS for 1 h at 25°C in the dark and rinsed once with PBS and twice
with distilled water. Cells were incubated in 0.1% pectolyase Y-23 at
30°C for 2 h, and then rinsed with PBS. The fixed cells were treated
with 0.3% Triton X-100 in PBS for 15 min, and then incubated with
the primary antibodies resolved in 1% BSA and PBS for 1 h with gentle agitation. Then cells were treated with Cy3-linked goat anti-rabbit
IgG (Amersham), 0.05% NaN3, 1% BSA and PBS for 1 h. Cells were
washed and then suspended in 50% glycerol. In some experiments, the
immunostained cells were further stained with 6 mg ml–1 DAPI for
30 min. Finally, cells were rinsed twice with 0.05% NaN3, 0.5% BSA
and PBS and then suspended in 50% glycerol. The fluorescence image
of cells was observed under an Olympus BX51 microscope.
Acknowledgments
We are grateful to Dr. K. Nakamura for stimulating discussion
and for providing cDNA for ornithine decarboxylase, to Dr. K. Matsuoka for providing the original vector pMAT311 and valuable advice
on BY-2, to Dr. K. Maeo for providing tobacco genome DNA preparation. This work was supported by Grants-in-Aid for Scientific
Research (10219203, 13640645, 13CE2005) from the Ministry of
Education, Sports, Culture, Science and Technology of Japan.
References
Alba, M.M. and Pages, M. (1998) Plant proteins containing the RNA-recognition motif. Trends Plant Sci. 3: 15–21.
Adachi, Y.K., Kas, E. and Laemmli, U.K. (1989) Preferential, cooperative binding of DNA topoisomerase II to scaffold-associated regions. EMBO J. 8:
3997–4006.
Blobel, G. and Wozniak, R.W. (2000) Proteomics for the pore. Nature 403: 835–
836.
Dehesh, K., Smith, L.G., Tepperman, J.M. and Quail, P.H. (1995) Twin autonomous bipartite nuclear localization signals direct nuclear import of GT-2.
Plant J. 8: 25–36.
Dickinson, L.A. and Kohwi-Shigematsu, T. (1995) Nucleolin is a matrix attachment region DNA-binding protein that specifically recognizes a region with
high base-unpairing potential. Mol. Cell. Biol. 15: 456–465.
Dworetzky, S.I., Wright, K.L., Fey, E.G., Penman, S., Lian, J.B., Stein, J.L. and
Stein, G.S. (1992) Sequence-specific DNA-binding proteins are components
of a nuclear matrix attachment site. Proc. Natl. Acad. Sci. USA 89: 4178–
4182.
Fackelmayer, F.O., Dahm, K., Renz, A., Ramsperger, U. and Richter, A. (1994)
Nucleic-acid-binding properties of hn RNP-U/SAF-A, a nuclear-matrix protein which binds DNA and RNA in vivo and in vitro. Eur. J. Biochem. 211:
747–757.
Gautier, T., Bergès, T., Tollervey, D. and Hurt, E. (1997) Nucleolar KKE/D
repeat proteins Nop56pand Nop58p interact with Nop1p and a e required for
ribosome biogenesis. Mol. Cell. Biol. 17: 7088–7098.
Harder, P.A., Silverstein, R.A. and Meier, I. (2000) Conservation of matrix
attachment region-binding filament-like protein 1 among higher plants. Plant
Physiol. 122: 225–234.
Hatton, D. and Gray, J.C. (1999) Two MAR DNA-binding proteins of the pea
nuclear matrix identify a new class of DNA-binding proteins. Plant J. 18:
417–429.
Holmes-Davis, R. and Comai, L. (1998) Nuclear matrix attachment regions and
plant gene expression. Trends Plant Sci. 3: 91–97.
Holmes-Davis, R. and Comai, L. (2002) The matrix attachment regions (MARs)
associated with the heat shock cognate 80 gene (HSC80) of tomato represent
specific regulatory elements. Mol. Gen. Genet. 266: 891–898.
Jenuwein, T., Forrester, W.C., Fernandez-Herrero, L.A., Laible, G., Dull, M. and
Grosschedl, R. (1997) Extension of chromatin accessibility by nuclear matrix
attachment regions. Nature 385: 269–271.
Laemmli, U.K., Poljak, K.E. and Adachi, Y. (1992) Scaffold-associated regions:
cis-acting determinants of chromatin structural loops and functional domains.
Curr. Opin. Genet. Dev. Dev. 2: 275–285.
Lewis, J.D. and Tollervey, D. (2000) Like attracts like: getting RNA processing
together in the nucleus. Science 288: 1385–1389.
Lian, S.H. and Clarke, M.F. (1999) A bipartite nuclear localization signal is
required for p53 nuclear import regulated by a carboxyl-terminal domain. J.
Biol. Chem. 274: 32699–32703.
Maniatis, T. and Reed, R. (2002) An extensive network of coupling among gene
expression machines. Nature 416: 499–506.
Matsuoka, K., Higuchi, T., Maeshima, M. and Nakamura, K. (1997) A vacuolartype H+-ATPase in a nonvacuolar organelle is required for the sorting of solu-
Tobacco MAR-binding protein
ble vacuolar protein precursors in tobacco cells. Plant Cell 9: 533–546.
Meier, I. and Gindullis, F. (1999) Matrix attachment region binding protein
MFP1 is localized in discrete domains at the nuclear envelope. Plant Cell 11:
1117–1128.
Meier, I., Harder, P.A. and Silverstein, R.A. (2000) Conservation of matrix
attachment region binding filament-like protein 1 among higher plants. Plant
Physiol. 122: 225–234.
Meier, I., Phelan, T., Gruissem, W., Spiker, S. and Schneider, D. (1996) MFP1, a
novel plant filament-like protein with affinity for matrix attachment region
DNA. Plant Cell 8: 2105–2115.
Michalowski, S.M., Allen, G.C., Gerald, E.H., Thompson, W.F. and Spiker, S.
(1999) Characterization of randomly obtained matrix attachment regions
(MARs) from higher plants. Biochemistry 38: 12795–12804.
Morisawa, G., Han-Yama, A., Moda, I., Tamai, A., Iwabuchi, M. and Meshi, T.
(2000) AHM1, a novel type of nuclear matrix-localized, MAR binding pro-
1567
tein with a single AT hook and a J domain-homologous region. Plant Cell 12:
1903–1916.
Nagata, T. (1987) Interaction of plant protoplast and liposome. Methods Enzymol. 148: 34–39.
Reeves, R. and Nissen, M.S. (1990) The AT-DNA-binding domain of mammalian high mobility group I chromosomal proteins. J. Biol. Chem. 265: 8573–
8582.
Reichheld, J.P., Gigot, C. and Chaubet-Gigot, N. (1998) Multi-level regulation
of histone gene expression during the cell cycle in tobacco cells. Nucl. Acids
Res. 26: 3255–3262.
Sambrook, J. and Russell, D.W. (2001) Molecular Cloning, 3rd edn. Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, New York.
Wu, P., Brockenbrough, J.S., Metcalfe, A.C., Chen, S. and Aris, J.P. (1998)
Nop5p is a small nucleolar ribonucleoprotein component required for pre-18
S rRNA processing in yeast. J. Biol. Chem. 273: 16453–16463.
(Received August 19, 2002; Accepted October 3, 2002)