plant$0940
The Plant Journal (1997) 11(1), 93–103
Characterization of proteins that interact with the
GTP-bound form of the regulatory GTPase Ran in
Arabidopsis
Thomas Haizel1, Thomas Merkle1,†, Aniko Pay2,
Erzsebet Fejes2 and Ferenc Nagy1,2,*
1Friedrich Miescher Institute, PO Box 2543, CH-4002
Basel, Switzerland, and
2The Biological Research Centre of the Hungarian
Academy of Sciences, Plant Biology Institute,
PO Box 251, H-6701 Szeged, Hungary
Summary
Ran, a small soluble GTP-binding protein, has been shown
to be essential for the nuclear translocation of proteins
and it is also thought to be involved in regulating cell
cycle progression in mammalian and yeast cells. Genes
encoding Ran-like proteins have been isolated from different higher plant species. Overexpression of plant Ran
cDNAs, similarly to their mammalian/yeast homologues,
suppresses the phenotype of the pim46-1 cell cycle mutant
in yeast cells. The mammalian/yeast Ran proteins have
been shown to interact with a battery of Ran-binding
proteins, including the guanidine nucleotide exchange
factor RCC1, the GTPase-activating Ran-GAP, nucleoporins
and other Ran-binding proteins (RanBPs) specific for RanGTP. Here, the characterization of the first Ran-binding
proteins from higher plants is reported. The yeast twohybrid system was used to isolate cDNA clones encoding
proteins of approximately 28 kDa (At-RanBP1a, AtRanBP1b) that interact with the GTP-bound forms of the
Ran1, Ran2 and Ran3 proteins of Arabidopsis thaliana. The
deduced amino acid sequences of the At-RanBP1s display
high similarity (60%) to mammalian/yeast RanBP1 proteins
and contain the characteristic Ran-binding domains.
Furthermore, interaction of the plant Ran and RanBP1
proteins, is shown to require the acidic C-terminal domain
(-DEDDDL) of Ran proteins in addition to the presence of
an intact Ran-binding domain. In whole cell extracts,
the GST–RanBP1a fusion protein binds specifically to
GTP–Ran and will not interact with Rab/Ypt-type small
GTP-binding proteins. Finally, in good agreement with
Received 13 May 1996; revised 16 September 1996; accepted 14 October
1996.
*For correspondence (fax 136 62 933939;
e-mail [email protected]).
†Present address: Department of Biology II, University of Freiburg,
Schanzlerstrasse 1, D-79104 Freiburg, Germany.
their proposed biological function, the At-Ran and the
At-RanBP genes are expressed coordinately and show the
highest level of expression in meristematic tissues.
Introduction
Higher plants, similarly to other eukaryotic organisms,
contain a superfamily of genes coding for Ras-related,
small GTP-binding proteins (Terryn et al., 1993). According
to the postulated functions of these plant proteins, this
superfamily is subdivided into three subfamilies that represent (i) the membrane-bound Rab/Ypt-type-proteins
involved in regulating vesicular trafficking (Terryn et al.,
1993), (ii) the membrane-bound Rho/Rac-type proteins
thought to be important for cytoskeleton organization
(Yang and Watson, 1993) and (iii) the soluble Ran (Rasrelated nuclear protein) that is most probably necessary
for protein import into the nucleus and for the onset of
mitosis (Ach and Gruissem, 1994; Merkle et al., 1994). The
mammalian Ran cDNA, originally identified as TC4 (Drivas
et al., 1990) codes for a 25 kDa nuclear protein and has
been shown to bind to the chromatin-associated protein
RCC1 (Bischoff and Ponstingl, 1991a). A temperature-sensitive tsBN2 hamster cell line containing a defective RCC1
gene (the regulator of chromosome condensation) has
been shown to form micronuclei and is unable to complete
mitosis properly (Nishimoto et al., 1978; Nishitani et al.,
1991). Several Schizzosaccharomyces pombe, Saccharomyces cerevisae and Drosophila mutants defective in the
RCC1 homologue pim1, PRP20/SRM1/MTR1 and BJ1 genes,
respectively, have been isolated. However, these mutants
exhibit pleiotropic phenotypes, and, unlike RCC1 and pim1,
PRP20/SRM1/MTR1 were discovered initially as genes
essential for a variety of nuclear functions including pheromone response, nuclear structure, chromosome stability
and mRNA metabolism and transport (Aebi et al., 1990;
Amberg et al., 1993; Clark et al., 1991; Fleischmann et al.,
1991; Kadowaki et al., 1993). Furthermore, it has been
shown that SRM1 and BJ1 genes complement the tsBN2
phenotype (Ohtsubo et al., 1991), and, conversely, expression of the mammalian RCC1 gene rescues the S. cerevisae
prp20, srm1 mutants (Clark et al., 1991; Fleischmann
et al., 1991).
Overexpression of various wild-type Ran homologues,
including the tomato and tobacco Ran genes, suppresses
93
94 Thomas Haizel et al.
the tsRCC1 phenotypes in S. cerevisae and in S. pombe
(Ach and Gruissem, 1994; Belhumeur et al., 1993;
Matsamuto and Beach, 1991; Merkle et al., 1994), but not
that of null mutants of the pim1 gene. These data indicate
that these proteins interact directly and that these RCC1
and Ran homologues are essential for normal functioning
of eukaryotic cells and have been well conserved during
evolution.
Like other GTP-binding proteins, Ran exerts its biological
function by binding and hydrolysing GTP. However,
intrinsic nucleotide exchange and GTP hydrolysis on Ran,
as for other Ras-related small GTP-binding proteins, is very
low. A number of different types of regulatory proteins
that interact specifically with Ran have recently been identified and shown to stimulate these rates. The biologically
active, GTP-bound form of Ran is converted into the the
inactive GDP-bound form by GTP hydrolysis. In HeLa
cell extracts, this process is stimulated by a Ran-specific
GTPase protein, designated as RanGAP1 (Bischoff et al.,
1994) which is homologous to cytoplasmic Rna1p from S.
pombe and S. cerevisae (Bischoff et al., 1995a). Another
of these interacting proteins is RCC1, which specifically
enhances the rate of exchange of guanine nucleotides on
Ran by about five orders of magnitude, but does not
distinguish between GDP and GTP (Bischoff and Ponstingl,
1991b). The RCC1-mediated guanine nucleotide exchange
on the GTP-bound form on Ran is specifically inhibited by
RanBP1 (Ran-binding protein, Bischoff et al., 1995b). This
protein was originally isolated by its ability to bind to the
GTP-bound form of Ran (Coutavas et al., 1993). The binding
domains of RanBP1 that are involved in the interaction
with GTP-bound Ran have been identified (Beddow et al.,
1995). Nucleoporins, such as Nup2p from S. cerevisae and
F59A2.1 from Caenorhabditis elegans contain very similar
domains and have been shown to interact with Ran in the
yeast two-hybrid system (Dingwall et al., 1995). These data
further underline the importance of Ran for protein import
into the nucleus, first suggested by Moore and Blobel
(1993), and indicate a direct link between Ran and nuclear
pore complexes.
We are interested in defining the physiological function of
plant Ran-like proteins. Genes encoding proteins displaying
high similarity to mammalian/yeast Ran proteins have
recently been identified in various plant species (Ach and
Gruissem, 1994; Merkle et al., 1994; Saalbach and Christov,
1994). In addition, it has been shown that overexpression
of plant Ran genes suppresses the phenotype of the cell
cycle regulatory mutant pim1-46 in fission yeast (Ach and
Gruissem, 1994; Merkle et al., 1994). In this paper, we
report the structure and expression pattern of the three
genes from Arabidopsis thaliana encoding Ran-like proteins. Furthermore, we describe two cDNA clones, isolated
by the yeast two-hybrid method, that encode proteins that
bind specifically to A. thaliana Ran proteins and display
significant homology to the mammalian/yeast RanBPs,
including RanBP1 and various nucleoporins.
Results
Isolation of A. thaliana cDNA and genomic clones
encoding Ran1, Ran2 and Ran3 proteins
Using as a probe the full-length tobacco NtRan1 cDNA
(Merkle et al., 1994), we screened various Arabidopsis
cDNA and genomic libraries to isolate cDNAs and genomic
clones encoding Ran-like small GTP-binding proteins. The
isolated cDNA and genomic clones code for three distinct
proteins, designated Ran1, Ran2 and Ran3. The AtRan1
and AtRan2 genes are located, in head to tail orientation,
on a 5.68 kb genomic fragment. AtRan3 is located on a
separate 6.87 kb fragment together with an unidentified
ORF that shows significant homology to β-1,3-glucanase.
Figure 1 shows the structure of the three different genomic
clones identified, as well as the deduced amino acid
sequence of the corresponding cDNAs. Figure 1a indicates
that AtRan1 and AtRan2 contain five exons and four introns,
while AtRan3 contains eight exons and seven introns. The
deduced amino acid sequences of the three genes are
nearly identical, differing from each other only at the Cterminal regions, and each protein consists of 221 amino
acid residues (Figure 1b). Figure 1b also demonstrates that
the characteristic domains of the Ran proteins known to
be involved in GTP-binding and hydrolysis, as well as
the acidic C-terminal domain and the so called ’effector
binding’ domain, have been highly conserved during
evolution.
Identification of AtRanBP1a and AtRanBP1b cDNA clones
by yeast two-hybrid assay
We employed the yeast two-hybrid system to isolate cDNA
clones encoding proteins that interact specifically with the
A. thaliana Ran1 regulatory GTPase. We fused the fulllength AtRan1 cDNA to the DNA-binding domain of GAL4
transcription activator (vector pGBT9) and used this construct as a ’bait’ to isolate cDNAs, encoding interacting
proteins, from a cDNA library of A. thaliana fused to the
GAL4 activation domain (vector pGAD10). Positive clones
were isolated by selecting for His1 prototrophs followed
by the β-galactosidase assay. Screening was performed
as described by Bartel and Zhu (1993) (for details see
Experimental procedures). By screening approximately
6 3 106 transformants, we isolated and partially sequenced
six clones. We found that these cDNAs encoded two
different but very closely related proteins. Detailed characterization of the isolated cDNA clones resulted in the
identification of two full-length cDNAs, designated
AtRanBP1a and AtRanBP1b, encoding AtRanBP1a and
Ran-binding proteins in Arabidopsis 95
Figure 1. (a) Structure of genomic clones
encoding the AtRan1, AtRan2 and AtRan3
proteins. Exons are indicated by filled boxes,
introns by thin lines. (b) Amino acid alignment
of tobacco (Merkle et al., 1994), tomato (Ach
and Gruissem, 1994) and Arabidopsis Ran
proteins. Conserved regions involved in GTPbinding/hydrolysis and the C-terminal region
are underlined, and the effector binding
domain implicated in protein–protein
interaction is indicated with asterisks.
AtRanBP1b proteins, respectively. We have demonstrated
that the AtRanBP1a and AtRanBP1b proteins interact specifically with the AtRan1 protein in vivo (Table 1). Amino
acid comparison of AtRanBP1a (234 amino acid residues)
and AtRanBP1b (216 amino acid residues) proteins showed
that these proteins display significant homology to each
other (80% identity) and are homologous (~60% identity)
to RanBP1 proteins characterized in other eukaryotes
(Figure 2). Figure 2 also shows that the so-called ’Ranbinding domain’, implicated in the interaction with Ran
(Beddow et al., 1995), is highly conserved among these
RanBP1 proteins, including the plant RanBP1s.
AtRanBP1s interact specifically with GTP-bound forms of
the AtRan1–3 proteins
The specificity of interaction between the AtRan and
AtRanBP proteins was characterized in several experi-
ments. To this end, we constructed recombinant pGBT9
plasmids containing cDNA fragments encoding the AtRan1,
AtRan2, AtRan3 proteins, as well as plasmids representing
AtRab6, NtRab5 and NtRab7 proteins. These clones were
then transformed into yeast cells harbouring the pGAD10
plasmid containing AtRanBP1a or AtRanBP1 cDNAs. Transformants containing both types of plasmids were then
selected and interaction of these proteins was determined.
Table 2 shows the results of these experiments. Firstly, we
found that both AtRanBP1a and AtRanBP1b proteins are
able to bind to AtRan1, as well as AtRan2 and AtRan3
proteins in vivo. Secondly, our results showed unambiguously that neither the AtRanBP1a nor the AtRanBP1b protein could interact with Rab type proteins in this
experimental system. Mutation of the glycine amino acid
residue at position 19 into valine (G19V) has been shown
to reduce the GTPase activity greatly and stabilized these
mutant Ran proteins in GTP-bound form, whereas replace-
96 Thomas Haizel et al.
ment of the threonine amino acid residue at position 24
with asparagine (T24N) decreased the affinity for GTP
and stabilized these mutant Rans in GDP-bound form
(Kornbluth et al., 1994). We used a PCR-assisted mutagenesis method to introduce these mutations into the cDNAs
encoding the AtRan1–3 proteins and cloned these mutated
genes into the pGBT9 vector. Interaction of these mutant
proteins with the previously isolated AtRanBP1a and
AtRanBP1b proteins was then tested in the yeast twohybrid system in vivo. Table 3 shows that only the GTPbound form of the AtRan proteins are able to bind to the
AtRanBP proteins. The GDP-bound form of the AtRan
Table 1. Binding of AtRan to AtRanBP1a in vivo
Plasmid 1
Plasmid 2
pGBT9
AtRan1
pGBT9
AtRan1
pGBT9
AtRan1R
pGBT9
AtRan1M
pGBT9
No insert
pGAD10
No insert
pGAD10
AtRanBP1
pGAD10
AtRanBP1
pGAD10
AtRanBP1
pGAD10
AtRanBP1
β-galactosidase
activity assay
Growth test
on His– plate
–
–
1
1
–
–
–
–
–
–
The β-galactosidase activity assay and growth test on His– medium
was performed according to Bartel and Zhu (1993). AtRan1R cDNA
representing the coding region of AtRan1 was inserted in the
reverse oritentation; AtRan1M, a frame-shift mutation after the
start codon ATG was introduced into the cDNA representing the
coding region of AtRan1.
Table 2. Binding of both AtRanBP1a and AtRanBP1b but not
NtRab6 by AtRan1–3 proteins
Plasmids
pGHT9
AtRan1
pGBT9
AtRan2
pGBT9
AtRan3
pGBT9
NtRab6
pGBT9
No insert
pGAD10
AtRanBP1a
pGAD10
AtRanBP1b
pGAD10
No insert
1
1
–
1
1
–
1
1
–
–
–
–
–
–
–
1/– indicate the results of both the β-galactosidase activity assay
and the growth test on His– medium.
proteins were found, uniformly, to be inactive in these
assays.
The ‘Ran-binding domain’ of the AtRanBPs and the
C-terminal region of the AtRan proteins are required for
interaction
We were interested in defining the position and size of
AtRanBP and AtRan domains required for binding. To this
end, we constructed several 59 and 39 deletion mutants of
AtRanBP1a and AtRan1 and transferred them into the
vector pGBT9. Interactions of the mutant AtRanBP1a proteins with the full-length AtRan1 protein, as well as interaction of the mutant AtRan1 protein with the full-length
AtRanBP1a protein, were then determined in vivo in the
Figure 2. Amino acid alignment of Ranbinding proteins isolated from A. thaliana
(AtRanBP1a, AtRanBP1b), mouse (MsRanBP1;
Coutavas et al., 1993) and human (HmRanBP1;
Bischoff et al., 1995b) cells.
The Ran-binding domain shown to be
essential for interaction with Ran (Beddow
et al., 1995) is indicated by asterisks.
Conserved amino acid residues are shown in
shaded boxes.
Ran-binding proteins in Arabidopsis 97
Table 3. Specific interaction of AtRanBPs with the GTP-bound
form of AtRan proteins
Plasmids
pGAD10
AtRanBP1a
pGAD10
AtRanBP1b
pGAD10
No insert
pGBT9
AtRan1
pGBT9
AtRan1M(GTP)a
pGBT9
pGBT9
AtRan1M(GTP)b No insert
1
1
–
–
1
1
–
–
–
–
–
–
aSubstitution
of the glycine residue (G19) with valine was
performed by PCR mutagenesis. Ran proteins containing this
mutation are defective in GTP hydrolysis, i.e. they stay
predominantly in the GTP-bound form (Kornbluth et al., 1994).
bSubstitution of the threonine residue (T24) with asparagine was
performed by PCR mutagenesis. Ran proteins containing this
mutation have a decreased affinity for GTP, i.e. they stay
predominantly in the GDP-bound form (Kornbluth et al., 1994).
1/– indicate the results of the β-galactosidase activity assay and
the growth test on His– medium.
also shows that removal of additional N-terminal and/or
C-terminal domains of the AtRanBP1a abolishes interaction. In good agreement with these data, we find that the
most truncated AtRanBP1a, which still interacts efficiently
with the full-length AtRan1 protein in vivo, contains only
the so-called ’Ran-binding domains’ (Beddow et al., 1995)
located between amino acid residues 27 and 154. In contrast, Figure 3b indicates that the short, conserved Cterminal region of the AtRan1 protein, containing only 7
amino acid residues, is required for binding, since deletion
of this region will prevent interaction between these proteins in vivo.
NtRan and AtRan proteins interact with the recombinant
GST–AtRanBP1a in vitro
We were also interested in characterizing the interaction
of the AtRanBPs and AtRan proteins in vitro. To
this end, we constructed two chimeric genes, encoding GST–AtRanBP1a (glutathione-S-transferase–full-length
Figure 3. Interaction of AtRanBP1 and AtRan
proteins is mediated by specific domains.
(a) Deletion analysis indicates that an 127
amino acid region of the AtRanBP1,
containing the conserved Ran-binding
domains (indicated by capital letters A–E) is
required for Ran-binding. (b) Deletion of the
conserved C-terminal region of AtRan1
abolishes interaction with AtRanBPs. The 59
and 39 endpoints of the mutant AtRanBP1a
and AtRan1 proteins tested are indicated by
numbers.
yeast two-hybrid system as described previously. Figure 3a
shows that deletion of the 27 amino acid residues from
the N-terminal region (1–27) or deletion of 80 amino acid
residues (154–234) from the C-terminal region of
AtRanBP1a does not effect binding. However, Figure 3a
ATRanBP1a) or the GST–AtRanBP1aM (glutathione-Stransferase–truncated AtRanBP1a) fusion proteins, and
expressed them in Escherichia coli. The overexpressed
recombinant fusion proteins were then bound to
glutathione–Sepharose B beads and mixed with Ran-
98 Thomas Haizel et al.
Figure 4. AtRan1 and the recombinant glutathione-S-transferase–
AtRanBP1a interact in vitro.
One µg of recombinant GST–AtRanBP1a protein (lanes 1–3) or 1 µg of
recombinant GST–AtRanBP1aM (lane 4) bound to glutathione–Sepharose B
beads were mixed with 30 µg of (1) GTP-loaded, (2) GTP-γS-loaded or (3)
crude high-speed supernatant fractions of whole-cell extracts prepared as
described by Merkle et al. (1994). These mixtures were then incubated for
30 min at 20°C, washed and pelleted by brief centrifugation at 1000 rev
min–1 for 2 min. Sepharose-bound proteins were denatured and separated
by SDS–PAGE. The amount of AtRan protein present in the untreated
100 µg cell extract (lane 5) or retarded by the Sepharose-bound GST–
AtRanBP1a protein was determined by Western blot analysis. The expected
molecular mass of AtRan1–3 proteins is approximately 26 kDa.
containing cell extracts prepared from tobacco or
Arabidopsis suspension culture. Whole-cell extracts were
prepared as described by Merkle et al. (1994). Interaction
of the Sepharose-bound recombinant GST–AtRanBP fusion
proteins with AtRan or NtRan was then assayed by determining the amount of retarded AtRan proteins by Western
blot analysis. Figure 4 shows that AtRan binds efficiently
to Sepharose-bound fusion protein containing full-length
AtRanBP1a (lanes 1–3). In good agreement with data
obtained by the yeast two-hybrid assay, we found that pretreatment of cell extracts with GTP or GTP-γS (i.e. increasing
the level of the GTP-bound form of Ran proteins) elevated
the amount of retarded AtRan proteins significantly (lanes
1 and 2, respectively) as compared with the untreated cell
extract (lane 3). Pre-treatments of cell extracts with the
same buffer lacking GTP did not significantly influence
the amount of the retarded Ran proteins. In contrast,
Sepharose-bound fusion protein containing the truncated
AtRanBP1a (a C-terminal region of 100 amino acid residues
is deleted) did not bind AtRan or NtRan either from
untreated (data not shown) or from GTP-loaded cell extracts
(lane 4). Taken together, these data suggest that AtRanBP1a
interacts with the GTP-bound form of AtRan and NtRan
proteins in vitro and that the Ran-binding domain of
AtRanBP1a is required for binding.
AtRan and AtRanBP genes are coordinately expressed
We characterized the expression level and pattern of the
AtRan1–3 genes by various methods. Firstly, we constructed chimeric genes containing the promoter regions
of the AtRan1–3 genes (480 bp, 900 bp and 940 bp,
respectively) fused to the GUS reporter gene and transferred them into Arabidopsis by A. tumefaciens mediated
transformation. The expression level and pattern of the
Ran–GUS transgenes (GUS histochemistry and activity)
were then analysed in various selected transgenic plants.
Figure 5 illustrates the expression pattern and level of the
AtRan1–GUS transgene in light- or dark-grown seedlings
and in fully developed transgenic plants. It also shows the
expression level of AtRan1–3 transgenes determined in 3day-old transgenic seedlings. Table 4 shows the number
of independently raised transgenic plants and the level of
GUS enzyme activity measured in leaf extracts. These
data indicate that the three AtRan–GUS transgenes are
expressed at different levels. The expression level of the
AtRan1–GUS is the highest. It is about twofold higher than
that of AtRan3–GUS and about 20-fold higher than of
AtRan2–GUS. However, independent of their expression
levels, the expression patterns of these genes are very
similar. Although their expression level is clearly higher in
meristematic tissues (i.e. root tip, developing embryos
etc.), they do not exhibit characteristic tissue-specifity but
are expressed ubiquitously in all tissues examined.
Secondly, we also determined the ratio of the endogenous AtRan1–3 mRNAs in total RNA samples isolated from
different tissues of mature Arabidopsis plants. To this
end, we prepared cDNAs and amplified by PCR a 138 bp
fragment representing the C-terminal regions of the
AtRan1–3 cDNAs, by using gene-specific oligonucleotide
primers. Amplified DNA samples were then digested with
different restriction endonucleases. Restriction site polymorphism of AtRan1–3 genes, within this 138 bp region,
allowed us to generate three gene-specific DNA fragments
of different length which were then detected by hybridization and quantified. According to this method, the ratio of
AtRan1–3 mRNAs in the original RNA sample is represented
by the ratio of the amplified DNA fragments, i.e. by the
ratio of the detected hybridization signals. Data obtained
by this approach demonstrate again that all of the three
endogenous AtRan genes are expressed, although at different levels. Figure 6, lane 2, shows that, in good agreement
with results obtained by analysing the transgenic AtRan1–
3–GUS plants, the expression level of the AtRan1 gene
(108 bp fragment) is the highest. It is about twofold higher
than that of AtRan3 (138 bp fragment) and about 30-fold
higher than that of AtRan2 (70 bp fragment). Figure 6 also
shows, however, that the expression levels of the AtRan1
and AtRan3 genes are nearly identical in other tissues
analysed, whereas the expression level of the AtRan2 gene
is uniformly low.
We also determined the expression patterns of the
AtRan1 and AtRanBP1a genes by in situ hybridization. To
this end, we prepared gene-specific, digoxigenin-labelled
antisense and sense riboprobes representing the
39-untranslated regions of these genes. We found that the
expression patterns of the AtRan1 and AtRanBP1a genes
are very similar in all tissues examined. Figure 7 illustrates
that both of these genes are highly expressed in root tips
Ran-binding proteins in Arabidopsis 99
Figure 5. AtRan–GUS transgenes are expressed ubiquitously but at different levels in transgenic tobacco and Arabidopsis.
(I) Expression patterns and levels of the (a,b) AtRan1–GUS, (c,d) AtRan3–GUS and (e,f) AtRan2–GUS transgenes in 3-day-old (a, c, e) and 6-day-old (b, d, f)
transgenic tobacco seedlings grown under 16 h light/8 h dark cycles. (II) Expression pattern of the AtRan1–GUS transgene in 3-, 5- and 7-day-old transgenic
Arabidopsis seedlings grown in constant light (a, b, c) or in constant dark (d, e, f) and in flowers of mature transgenic plants (g, h).
Table 4. The expression level of the AtRan1–3–GUS transgenes in
independent transgenic Arabidopsis plants
Construct
AtRan1–GUS
AtRan2–GUS
AtRan3–GUS
Number of
plants
18
15
20
GUS enzyme activitya
111
11
1
1/–
13
–
10
3
0
5
2
8
5
0
7
0
enzyme activity: 1111000–750; 11, 750–500; 1100–25,
1/–, , 25 pmol MU min–1 µg–1 leaf protein.
aGUS
(whole-mount method) and in the gynoecium (formed by
two fused carpels and separated by the central septum,
clearly recognizable on 7 µm cross-sections of the flower)
containing the ovule primordia developing into embryo
sacs. Lower-level expression could also be detected, at this
developmental stage, in petal and stamen. Expression of
the AtRan1 and AtRanBP1a genes was detectable in all
tissues examined (stem, leaves etc., data not shown);
therefore, we concluded that, in good agreement with their
Figure 6. The endogenous AtRan1–3 genes are expressed at different levels.
The ratio of AtRan1–3 specific mRNA levels was determined in flower (lane
1), leaf (lane 2), stem (lane 3) and whole flowering plants (lane 4).
Control experiments (lane 5–8), all performed on the same RNA sample,
demonstrate the gene specificity and completeness of restriction digestions:
undigested PCR product (lane 8), EcoRV digestion generating the AtRan1specific 108 bp fragment (lane 7), PvuII digestion generating the AtRan2specific 70 bp fragment (lane 6), co-digestion with EcoRV, PvuII and TaqI
(lane 5) generating only fragments specific to AtRan1 and AtRan2 (108 bp
and 70 bp fragments respectively) but eliminating the 138 bp fragment
specific to AtRan3, indicating that the previous digestions were complete.
Molecular weight markers (lanes 9, 10) were produced by PCR amplification
and restriction digestions from the isolated AtRan1–3 cDNA clones.
100 Thomas Haizel et al.
Figure 7. Cell-specific expression patterns and levels of AtRan1 and AtRanBP1 genes determined by in situ hybridization.
Cross-sections (7 µm) of flower tissues (a–d) or whole-mount root tips (e–h) were hybridized with AtRan1-specific antisense (a, e) or sense (b, f) as well as
with AtRanBP1a-specific antisense (c, g) or sense (d, h) riboprobes. 1, pollen; 2, anther; 3, gynoecium; 4, ovule. Bar575 µm.
Ran-binding proteins in Arabidopsis 101
proposed biological function, these genes are expressed
ubiquitously and coordinately at all stages of development.
Discussion
We have identified and characterized three Arabidopsis
genes encoding very similar, yet distinctly different Ran
GTPases. AtRan1p, AtRan2p and AtRan3p represent new
members of the group of plant Ran proteins and display
significant homology to other plant (Ach and Gruissem,
1994; Merkle et al., 1994), yeast (Belhumeur et al., 1993)
and mammalian Ran proteins (Drivas et al., 1990). In good
agreement with previous findings, we observed that the
Arabidopsis Ran proteins, similar to tobacco (Merkle et al.,
1994) and tomato (Ach and Gruissem, 1994) Ran proteins
are (i) soluble, (ii) detected in both the cytoplasm and
nucleus, and that (iii) overexpression of the Arabidopsis
cDNAs suppressed the phenotype of the pim1–46 cell cycle
regulatory mutant in Schizosaccharomyces pombe cells
(data not shown). Ran, like other small Ras-like GTPases,
has a very low intrinsic GTPase activity. To fulfil its proposed biological functions by coupling a controlled GTPase
cycle to the regulation of cellular processes, Ran has been
shown to interact with a number of accessory proteins
including GEF (guanine nucleotide exchange factor), GAP
(GTPase-activating protein) and putative effector proteins
such as RanBP1 and RanBP2 (nucleoporin-like proteins)
(for reviews see Rush et al., 1995; Sazer, 1995). We used
the yeast two-hybrid system and the AtRan1 protein as
’bait’ to isolate Arabidopsis genes encoding proteins interacting with AtRan1–3. Among the identified clones were
several cDNAs which code for two highly similar plant
proteins displaying significant homology to mammalian
RanBP1 proteins shown to interact with the GTP-bound
form of Ran. The Arabidopsis AtRanBP1a and AtRanBP1b
proteins contain the ’Ran-binding domain’, present also in
various nucleoporins (Beddow et al., 1995) and interact
only with the GTP-bound forms of AtRan1–3. Interestingly,
AtRanBP1a binds to each of the identified AtRan proteins
in vivo. The interaction is specific, it can be reconstituted
in vitro and AtRanBPs do not bind to other small, Rab/YPT
type GTP-binding proteins. We have shown that deletion
of the Ran-binding domain of the AtRanBP1 proteins or
removal of the C-terminal domain of the AtRan1 protein
abolishes this interaction. These findings strongly suggest
that the AtRanBP proteins represent the plant counterparts
of the mammalian RanBP1s and are the first identified
members of a new family of proteins interacting with Ran
(effector proteins).
The Arabidopsis genome contains three Ran (AtRan1–3)
and at least two RanBP1 genes. In situ hybridization assays
and analysis of transgenic plants containing AtRan1–3/
GUS transgenes clearly showed that these genes are all
transcribed, albeit at different levels. We also found that,
although they are expressed ubiquitously, their expression
level is highest in meristematic, metabolically active cells.
The AtRan and AtRanBP genes seem to be expressed
coordinately; however, the exact subcellular localization of
the encoded proteins is yet to be determined.
The biological functions of the plant Ran and RanBP
proteins remain elusive. The high degree of amino acid
conservation, the interchangeability and the similar patterns of intracellular localization of these proteins from
different species suggest that the biological role of Ran
and Ran-binding proteins is evolutionarily conserved. Ran
proteins were found to interact with a variety of accessory
and effector proteins and the RanGTPase system was
shown to affect numerous cellular processes including
cell cycle progression, DNA replication and chromosome
structure as well as nucleocytoplasmic transport of proteins
and RNA in various eukaryotic systems.
The molecular mechanisms underlying these obviously
important cellular processes are poorly understood in plant
cells. To gain information about the biological role of
the plant RanGTPase system, a central factor of these
processes in other eukaryotic cells, is naturally one of our
major interests. Therefore, we initiated different lines of
experiments which are in progress in our laboratory. Firstly,
we are using the yeast two-hybrid assay to isolate additional proteins interacting with Ran. Apart from the
AtRanBP1 cDNAs, as expected, we have already isolated
several other cDNAs encoding proteins that interact
specifically with AtRan1 in vivo. We assume that characterization of these proteins and the development of a plant
in vitro nuclear import system (Merkle et al., 1996) will make
it possible to determine their role in the nucleocytoplasmic
transport of proteins in higher plants.
Experimental procedures
Screening, isolation and characterization of cDNA and
genomic clones
AtRan1–3 cDNAs and genomic clones were isolated from
Arabidopsis leaf cDNA and genomic libraries (constructed in ZAPII
vector and purchased from Stratagene) by plaque hybridization
using as probe the full-length NtRan1 cDNA clone (Merkle et al.,
1994). Full-length AtRanBP1a–b cDNA and genomic clones were
isolated from the same libraries by the same method using a
partial (720 bp) cDNA fragment identified by the yeast two-hybrid
system as a probe. Labelling of the probes, hybridization and
washing of the filters were performed as described by Dallman
et al. (1992). Nucleotide sequences of both strands of the isolated
cDNA and genomic clones were determined by the dideoxy
chain termination method. All DNA manipulations were performed
according to Sambrook et al. (1989).
RNA isolation and PCR amplification
Total RNA was isolated from different tissues of flowering
Arabidopsis plants grown in the greenhouse under 16 h light/8 h
dark cycles as described by Nagy et al. (1988).
102 Thomas Haizel et al.
For the measurement of AtRan1–3 mRNA levels, 10 µg of
total, DNA-free RNA samples were reverse-transcribed. A 138 bp
fragment representing the C-terminal part of the coding regions
of the AtRan1–3 genes was then amplified by 12 PCR cycles using
a mixture of oligonucleotide primers representing all three AtRan
genes. Amplified DNA fragments were digested to completion by
EcoRV and/or PvuII restiction endonucleases. EcoRV restriction
generates an AtRan1-specific 108 bp fragment, since this enzyme
digests only DNA representing the AtRan1 gene. PvuII digests
only DNA representing the AtRan2 gene and yields a 70 bp
fragment specific for AtRan2. The remaining 138 bp fragment is
thus AtRan3-specific, because the amplified region of the AtRan3
is not digestible by these enzymes. The completeness of restriction
digestion by EcoRV and PvuII was verified by additional digestion
with the restriction enzyme TaqI that cuts only the amplified region
of AtRan3.
Fragments obtained by these digestions were separated by
electrophoresis on a 6% polyacrylamide gel containing 6 M urea
and electroblotted onto a Hybond N1 nylon membrane (Amersham) in 0.5 3 TBE (Sambrook et al., 1989) for 40 min at 36 V. Filters
were then hybridized with the mixture of labelled oligonucleotides
used previously in PCR amplification reactions. After hybridization,
filters were washed, dried and exposed for 20 h.
Two-hybrid screening
Vectors (pGBT9 and pGAD10), yeast strains (HF7c, SFY526) and
the Arabidopsis cDNA library cloned into the pGAD10 vector
(MATCHMAKER) were purchased from Clontech. Screening, selection of putative positive clones and β-galactosidase assays were
performed according to the manufacturer’s protocols with minor
modifications. Briefly, the ’bait’ AtRan1 (amino acids 2–221) was
cloned into the pGBT9 vector and was transformed into the
HF7c strain by a PEG/lithium acetate transformation method. This
transformant was then transformed with library cDNA and grown
on selective plates for 8 days at 30°C. Positive colonies were
picked and tested for β-galactosidase activity. Plasmids from
positive library clones were then rescued into E. coli HB101 cells
and the isolated plasmids were sequenced. Specific point mutants
of the AtRan1 gene (G19V, T24N), and deletion mutants of the
AtRanBP1a and AtRan1 genes were generated by oligonucleotideassisted PCR assays. A pGAD10 vector harbouring the NtRab6
gene (amino acids 1–216, Haizel et al., 1995) was used as control.
GST fusion and the in vitro protein binding assay
Gluthatione-S-transferase (GST) fusion proteins (GST–AtRanBP1a,
GST–AtRanBP1aM, GST–NtRab6) were expressed in E. coli and
purified partially, using gluthatione–Sepharose beads, as
described by Dallman et al. (1992). High-speed supernatant fractions of whole-cell extracts from tobacco and Arabidopsis plants
were isolated according to Merkle et al. (1994). Loading of Ran
proteins with GTP or GTPγS (guanosine-59-O-[γ-thiotriphosphate])
was performed as reported by Haizel et al. (1995). GST–AtRanBP1abound proteins were analysed by SDS–PAGE, electrophoretically
transferred onto PVDF membranes and probed with a mouse
polyclonal antiserum, raised against the NtRanA1 protein. This
antiserum cross-reacted with all AtRan proteins.
In situ hybridization
Probes employed in the in situ hybridization assays were cloned
into the pSP18 plasmid (sense orientation) and pSP19 plasmid
(antisense orientation) and purified on CsCl gradients. Purified
plasmids containing the AtRan1 insert (210 bp fragment, representing the 39-untranslated region) and AtRanBP1a insert (240 bp
fragment representing the last 100 bp of the coding region and
140 bp of the 39 untranslated region) were linearized by the
appropriate restriction enzymes, phenolized and precipitated.
These linearized plasmids were then used as templates to synthesize the digoxigenin-labelled antisense and sense RNA probes.
Synthesis and purification of the riboprobes were performed by
using a DIG Nucleic Acid Labelling Kit (Boehringer Mannheim)
according to the manufacturer’s protocol.
Tissues were vacuum infiltrated and fixed for 1 h in a solution
containing 0.1 M sodium phosphate (pH 7.2), 4% paraformaldehyde, 0.25% glutaraldehyde, 0.1% Tween-20 and 10% DMSO.
After fixation, tissues were dehydrated stepwise by a graded
ethanol series (10–100%), embedded in paraffin and sectioned.
Sections (7 µm thick) were transferred onto polylysine-coated
slides and dried at 42°C overnight. Prior to hybridization, sections
were deparaffinized with xylene and rehydrated by a graded (100–
0%) ethanol series. Hybridization was performed in a solution
containing 50% formamide, 0.3 M NaCl, 0.01 M Tris–HCl, 0.001 M
EDTA, 10% dextrane sulphate, 500 µg ml–1 polyadenylic acid,
150 µg ml–1 yeast tRNA and 100 ng riboprobe, at 42°C for 18 h.
After hybridization, non-hybridized RNA was removed by RNase
treatment (50 µg ml–1, at 37°C for 30 min). Prior to colour detection,
slides were washed with 2 3 SSC for 30 min at room temperature.
Colour detection was performed according to the manufacturer’s
protocol by using a DIG Nucleic Acid Detection Kit (Boehringer
Mannheim).
Alternatively, we used the same probes as described above and
the so-called ’whole mount’ method to determine the expression
pattern of these genes in root tips. This procedure was carried
out as described by de Almeide Engler et al. (1994).
Acknowledgements
We thank E.D. Schmidt and S.C. de Vries for help with the in situ
hybridization assays. The authors also thank D.W. Kirk, R. Nagy
and A. Redai for excellent technical help and C. Kolar and F. Turck
for critical reading of the manuscript. Work performed in Hungary
was supported by Howard Hughes grant HHMI 75195–542401 and
OTKA 16/167 grant to F. Nagy.
References
Ach, R.A. and Gruissem, W. (1994) A small nuclear GTP-binding
protein from tomato suppresses a Schizosaccharomyces pombe
cell-cycle mutant. Proc. Natl Acad. Sci. USA, 91, 5863–5867.
Aebi, M., Clark, M.W., Vijayraghavan, U. and Abelson, J. (1990)
A yeast mutant, PRP20, altered in mRNA metabolism and
maintenance of the nuclear structure, is defective in a gene
homologous to the human gene RCC1 which is involved in the
control of chromosome condensation. Mol. Gen. Genet. 224,
72–80.
de Almeide Engler, J., Van Montagu, M. and Engler, G. (1994)
Hybridization in situ of whole mount messenger RNA in plants.
Plant Mol. Biol. Rep. 12, 321–331.
Amberg, D.C., Fleischmann, M., Stagljar, I., Cole, C.N. and Aebi, M.
(1993) Nuclear PRP20 protein is required for mRNA export.
EMBO J. 12, 233–241.
Bartel, P.L. and Zhu, L. (1993) Rapid identification of genes
encoding interacting proteins with the MATCHMAKER™ twohybrid system. CLONTECHniques, 8, 1–5.
Ran-binding proteins in Arabidopsis 103
Beddow, A.L., Richards, S.A., Orem, N.R. and Macara, I.G. (1995)
The Ran/TC4 GTPase-binding domain: identification by
expression cloning and characterization of a conserved
sequence motif. Proc. Natl Acad. Sci. USA, 92, 3328–3332.
Belhumeur, P., Lee, A., Tam, R., DiPaolo, T., Fortin, N. and
Clark, M.W. (1993) GSP1 and GSP2, genetic suppressors of
the prp20–1 mutant in Saccharomyces cerevisae: GTP-binding
proteins involved in the maintenance of nuclear organization.
Mol. Cell. Biol. 13, 2152–2161.
Bischoff, F.R. and Ponstingl, H. (1991a) Mitotic regulator protein
RCC1 is complexed with a nuclear ras-related polypeptide. Proc.
Natl Acad. Sci. USA, 88, 10 830–10 834.
Bischoff, F.R. and Ponstingl, H. (1991b) Catalysis of guanine
nucleotide exchange on Ran by the mitotic regulator RCC1.
Nature, 354, 80–82.
Bischoff, F.R., Klebe, C., Kretschmer, J., Wittinghofer, A. and
Ponstingl, H. (1994) RanGAP1 induces GTPase activity of nuclear
Ras-related Ran. Proc. Natl Acad. Sci. USA, 91, 2587–2591.
Bischoff, F.R., Krebber, H., Kempf, T., Hermes, I. and Ponstingl, H.
(1995a) Human RanGTPase-activating protein RanGAP1 is a
homologue of yeast Rna1p involved in mRNA processing and
transport. Proc. Natl Acad. Sci. USA, 92, 1749–1753.
Bischoff, F.R., Krebber, H., Smirnova, E., Dong., W. and
Ponstingl, H. (1995b) Co-activation of RanGTPase and inhibition
of GTP dissociation by Ran-GTP binding protein RabBP1. EMBO
J. 14, 705–715.
Clark, K.L. and Sprague, G.F. Jr (1991) Yeast pheromone response
parthway: characterization of a suppressor that restores mating
to receptorless mutants. Mol. Cell. Biol. 9, 2682–2694.
Coutavas, E., Ren, M., Oppenheim, J.D., D’Eustachio, P. and Rush,
M.G. (1993) Characterization of proteins that interact with the
cell-cycle regulatory protein Ran/TC4. Nature, 366, 585–587.
Dallman, G., Sticher, L., Marshallsay, C. and Nagy, F. (1992)
Molecular characterization of tobacco cDNA clones encoding
two small GTP-binding proteins. Plant Mol. Biol. 19, 847–857.
Dingwall, C., Kandel-Lewis, S. and Seraphin, B. (1995) A family of
Ran-binding proteins that includes nucleoporins. Proc. Natl
Acad. Sci. USA, 92, 7525–7529.
Drivas, G., Shih, A., Coutavas, E., Rush, M.G. and D’Eustachio, P.
(1990) Characterization of four novel RAS-related genes
expressed in a human teratocarcinoma cell line. Mol. Cell. Biol.
10, 1793–1798.
Fleischmann, M., Clark, M.W., Forrester, W., Wickens, M.,
Nishimoto, T. and Aebi, M. (1991) Analysis of yeast prp20
mutations and functional complementation by the human
homologue RCC1, a protein involved in the control of
chromosome condensation. Mol. Gen. Genet. 227, 417–423.
Haizel, T., Merkle, T., Turck, F. and Nagy, F. (1995) Characterization
of membrane-bound small GTP-binding proteins from Nicotiana
tabacum. Plant Physiol. 108, 59–67.
Kadowaki, T., Goldfarb, D., Spitz, L.M., Tartakoff, A.M. and
Ohno, M. (1993) Regulation of RNA processing and transport
by a guanine nucleotide release protein and members of the
Ras superfamily. EMBO J. 12, 2929–2937.
Kornbluth, S., Dasso, M. and Newport, J. (1994) Evidence for a
dual role for TC4 protein in regulating nuclear structure and
cell cycle progression. J. Cell Biol. 125, 705–719.
Matsamuto, T. and Beach, D. (1991) Premature initiation of mitosis
in yeast lacking RCC1 or an interacting GTPase. Cell, 66, 347–360.
Merkle, T., Haizel, T., Matsamuto, T., Harter, K., Dallmann, G.
and Nagy, F. (1994) Phenotype of the fission yeast cell cycle
regulatory mutant pim1–46 is suppressed by a tobacco cDNA
encoding a small, Ran-like GTP-binding protein. Plant J. 6,
555–565.
Merkle, T., Leclerc, D., Marshallsay, C. and Nagy, F. (1996) A plant
in vitro system for the nuclear import of proteins. Plant J. 10,
1177–1186.
Moore, M.S. and Blobel, G. (1993) The GTP-binding protein Ran/
TC4 is required for protein import into the nucleus. Nature, 365,
661–663.
Nagy, F., Kay, S.A. and Chua, N.-H. (1988) The analysis of gene
expression in transgenic plants. In Plant Gene Research Manual
(Gelvin, S.B. and Schilperoort, R.A., eds). Dordrecht: Kluwer
Academic Press, pp. 1–29.
Nishimoto, T., Eilen, E. and Basilico, C. (1978) Premature
chromosome condensation in a ts DNA– mutant of BHK cells.
Cell, 15, 475–483.
Nishitani, H., Ohtsubo, M., Yamashita, K., Iida, H., Pines, J.,
Yasudo, H., Shibata, Y., Hunter, T. and Nishimoto, T. (1991)
Loss of RCC1, a nuclear DNA-binding protein, uncouples the
completion of DNA replication from the activation of cdc2
protein kinase and mitosis. EMBO J. 10, 1555–1564.
Ohtsubo, M., Yoshida, T., Seino, H., Nishitani, H., Clark, K.L.,
Sprague, G.F.J., Fracsh, M. and Nishimoto, T. (1991) Mutation
of the hamster cell cycle gene RCC1 is complemented by the
homologous genes of Drosophila and S. cerevisae. EMBO J.
10, 1265–1273.
Rush, M.G., Drivas, G. and D’Eustachio P. (1995) The small nuclear
GTPase Ran: how much does it run? BioEssays, 18, 103–112.
Saalbach, G. and Christov, V. (1994) Sequence of a plant cDNA
from Vicia faba encoding a novel Ran-related GTP-binding
protein. Plant Mol. Biol. 24, 969–972.
Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular
Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold
Spring Harbor Laboratory Press.
Sazer, S. (1995) The search for the primary function of the Ran
GTPase continues. Trends Cell Biol. 6, 81–85.
Terryn, N., Van Montagu, M. and Inze, D. (1993) GTP-binding
proteins in plants. Plant Mol. Biol. 22, 143–152.
Yang, Z. and Watson, J.C. (1993) Molecular cloning and
characterization of rho, a ras-related small-GTP-binding protein
from garden pea. Proc. Natl Acad. Sci. USA, 90, 8732–8736.
EMBL data library accession numbers X97377 (A. thaliana mRNA for RBP1a protein), X97378 (A. thaliana mRNA for RBP1b
protein), X97379 (A. thaliana mRNA for Ran1 protein), X97380 (A. thaliana mRNA for Ran2 protein), X97381 (A. thaliana
mRNA for Ran3 protein), X97382 (A. thaliana ran1 gene), X97383 (A. thaliana ran2 gene), and X97384 (A. thaliana ran3 gene).