Department of Biology, Washington University, St. Louis, Missouri
63130
 |
INTRODUCTION |
Sessile organisms such as plants and
fungi have developed sophisticated responses to environmental stresses
that allow them to tolerate adverse conditions (Hohmann and Willem,
1997
; Hoekstra et al., 2001). One approach to understanding
these responses is through the identification of genes that are
up-regulated during abiotic stress, followed by functional analyses of
the corresponding gene products. In plants, stress stimulates the
production of the phytohormone abscisic acid (ABA) (Zeevaart and
Creelmann, 1988; Bray, 2002), which induces the expression of a variety
of genes (Chandler and Robertson, 1994; Bray, 2002). Furthermore, elevated levels of ABA are correlated with late embryogenesis and the
onset and maintenance of seed dormancy (Zeevaart and Creelmann, 1988). Because the dehydration of the seed during late embryogenesis is
a normal part of the developmental program, these tissues provide a
valuable resource for the identification of genes that are involved in
desiccation tolerance.
In cereals, a metabolically active tissue, the aleurone layer,
surrounds the starchy endosperm of the seed. Upon germination, the
embryo produces the phytohormone GA, which induces the production of
hydrolytic enzymes by the aleurone tissue. These enzymes are secreted
into the endosperm, where they liberate sugars and amino acids for the
growing embryo. ABA blocks the production of these enzymes at the
transcriptional level (Lovegrove and Hooley, 2000). The gene
HVA22 was originally identified as a transcript that accumulates in barley (Hordeum vulgare) aleurone tissue upon
treatment with ABA, and was later found to be induced in vegetative
tissues exposed to ABA, drought, or cold stress (Shen et al., 1993;
Shen et al., 2001). Through RNA-blot analysis and study of promoter structure and activity, the regulation of HVA22 expression
has been well characterized (Shen et al., 1996, 2001). However, the role of the HVA22 protein in stress response is not yet understood. Database searches reveal homologs of HVA22 in diverse
eukaryotic organisms, including other plants, animals, fungi, and
protists (Shen et al., 2001). Arabidopsis has at least five homologs of HVA22, and analysis of their expression patterns shows that
they are differentially regulated by hormonal and developmental signals (Chen et al., 2002). It has also been shown that expression of the
yeast (Saccharomyces cerevisiae) homolog can be induced by salt stress (Shen et al., 2001).
The yeast homolog of HVA22 was named YIP2
(YPT-interacting protein) based on a physical interaction with the rab
GTPase Ypt1p (http://genome-www4.stanford.edu/cgi-bin/SGD/locus.pl?locus=YIP2;Yang et al., 1998). The Rab GTPases are present in all eukaryotes, and have
been established as important players in various aspects of vesicular
transport, with their function dependent on coordinated interaction
with upstream regulators and downstream effectors. These small proteins
cycle between GTP- and GDP-bound forms, and this cycling is controlled
by the upstream regulators. The GTP-bound form acts on downstream
effectors, recruiting them to the site of action (Segev,
2001a).
More recently, YIP2 has been named YOP1 (YIP one
partner) based on a physical interaction with Yip1p. Two-hybrid assays,
glutathione S-transferase pull downs, and subcellular
localization experiments all indicate that Yop1p and Yip1p interact in
vivo (Calero et al., 2001). YIP1 encodes an essential
membrane-bound protein that interacts with the rab GTPases Ypt1p and
Ypt31p, perhaps recruiting them to the Golgi membrane (Yang et al.,
1998).
The presence of a homolog of HVA22 in yeast provides the
opportunity to study this gene in a versatile model organism. Because deletion of the YOP1 coding sequence results in only a mild
phenotype, a synthetic enhancement screen was performed to identify
genetic interactions that may provide some insight into the function of Yop1p. Because of the convenient features of yeast genetics, we believe
that studying the action of Yop1p will enhance our understanding of the
function of HVA22. We report here the results of that screen, characterization of the isolated mutants, and discuss the potential function of the YOP1 and HVA22 proteins.
| |
RESULTS |
Barley HVA22, Yeast Yop1p, Human (Homo sapiens) Dp1,
and Arabidopsis AtHVA22d Share Sequence Homology and Similar Hydropathy
Profiles
Homologs of HVA22 are present in diverse eukaryotic organisms
(Fig. 1; Calero et al., 2001; Shen et
al., 2001). The region of highest homology is a short hydrophilic loop
flanked by two hydrophobic stretches. The C-terminal region shows the
highest degree of variability between species, although it is
hydrophilic in all cases examined (Fig. 1B). The yeast and human
homologs contain a 40- to 48-amino acid N-terminal region that is not
present in HVA22 and AtHVA22d.
View larger version (56K):
[in this window]
[in a new window]
|
Figure 1.
Comparison of yeast Yop1p, barley HVA22, human
Dp1, and Arabidopsis AtHVA22d. A, Amino acid sequence alignment of
Yop1p and homologs from barley, human, and Arabidopsis. Identical
residues are highlighted; similar residues are outlined. The entire
barley sequence is shown, and the others are truncated accordingly. B,
Hydropathy plot of Yop1p and homologs from barley, human, and
Arabidopsis, comparing the hydrophobic and hydrophilic regions in the
four proteins.
|
|
yop1
Has a Mild Temperature-Sensitive Growth
Defect
To begin our investigation of YOP1, the coding sequence
was replaced by a kanamycin resistance cassette by homologous
recombination. The resulting yop1 strain, ABY16,
was viable under various growth conditions. (See Table
I for yeast strains used in this study.) No difference in growth between wild type and the yop1
mutant was observed at room temperature under salt stress up to 1.5 M NaCl (data not shown). When grown at room
temperature in yeast peptone dextrose (YPD) liquid media, there
is no detectable difference in A600 between
wild type and the yop1 mutant (Fig.
2A). However, when cultured at 37°C in
YPD, the yop1 mutant reached an
A600 of approximately 75% that of the
wild-type strain (Fig. 2B). Viable cell counts were determined by
plating dilutions of these cultures. The ABY16 cultures were found to
have approximately 93% the cell count of the wild-type cultures. The
difference in cell count was the same whether the cultures were grown
in separate flasks or in competition (data not shown). The same
phenotype was observed in strain ABY14, in which the YOP1
coding sequence was replaced by an HIS3 cassette (data not
shown).
View larger version (12K):
[in this window]
[in a new window]
|
Figure 2.
The Yop1 mutant shows a
temperature-sensitive growth phenotype. Liquid YPD media was inoculated
with wild type or yop1 (deletion mutant) and shaken at
25°C or 37°C. The density of the cultures was quantified by
measuring A600. Values shown are
averages ± SE, n = 3. A, At
25°C, there is no significant difference between wild type (solid
line) and the yop1 mutant (dashed line). B, At 37°C,
the A600 of the mutant cultures are about
75% that of wild type after 55 h.
|
|
yop1 Shows a Synthetic Enhancement Defect with
sey1
Because the yop1 mutant did not show a severe
phenotype, a synthetic enhancement screen was performed in a
yop1 background to identify genetic interactions with
YOP1. Thirty-nine non-sectoring (sect
) mutants were
isolated, subjected to complementation group analysis, and tested for
YOP1 and URA3 dependence. A single
YOP1-dependent comple-mentation group, consisting of
four individually isolated mutants, was identified.
The mutant ABY51 was transformed with a genomic library, and sectoring
colonies recovered. Plasmids from these colonies were isolated and
characterized by end sequencing of the genomic inserts. As expected,
multiple YOP1 genomic clones were recovered. A 16-kb clone
containing eight open reading frames (ORFs) from chromosome 15 was recovered several times. In addition, an overlapping 12-kb genomic
clone containing four ORFs was recovered several times. Subcloning
fragments of this clone and testing for sect+ led to the identification
of ORF YOR165W as the complementing gene (Fig. 3). We named this ORF SEY1
(synthetic enhancement with YOP1). SEY1 is a
homolog of the Arabidopsis gene RHD3 (Root Hair
Defective 3). Both of these proteins contain the two predicted
GTP-binding motifs GXXXXGKS and DXXG near the N terminus (Wang et al.,
1997; Fig. 4), although neither has been
demonstrated to bind GTP. An LEU plasmid containing either a
YOP1 or SEY1 genomic clone restored sectoring in
the synthetic enhancement mutant, but an empty LEU plasmid did not
(Fig. 3).
View larger version (42K):
[in this window]
[in a new window]
|
Figure 3.
sey1 synthetically enhances
yop1. A synthetic enhancement mutant was transformed with
pRS315 containing no insert, a YOP1 genomic clone, a
SEY1 genomic clone, or an N-terminal deletion mutant of
YOP1 and streaked out on rich media. After 1 week, colonies
were examined for sectoring.
|
|
View larger version (53K):
[in this window]
[in a new window]
|
Figure 4.
Molecular nature of sey1 mutant alleles and
comparison with Arabidopsis RHD3. A, Alleles 1 through 3 were cloned by the gap repair procedure described in "Materials and
Methods." Allele 4 was cloned by PCR (see "Materials and Methods"
for details). Recovered clones were sequenced to identify the lesions.
B, Alignment of the N-terminal region of Sey1p and RHD3. Identical
residues are highlighted; similar residues are outlined. The location
of the sey1 lesions are marked by stars, and the GTP-binding
motifs are double underlined.
|
|
The N-Terminal Region of Yop1p Is Not Required for Its
Function
It has been demonstrated that the N-terminal 17 amino acids of
Yop1p are necessary and sufficient for its physical interaction with
Yip1p, as determined by the two-hybrid assay (Calero et al., 2001). We
wanted to determine if this interaction is necessary for the activity
of Yop1p. To test this, a truncated yop1 mutant lacking
coding sequence for amino acids 2 through 17 was constructed in a LEU
vector. This was transformed into the synthetic enhancement mutant
ABY51. Transformants were streaked onto rich media and scored for
sectoring. After 1 week of growth, transformants carrying the truncated
yop1 showed sectoring (Fig. 3D), whereas those carrying the
empty vector did not (Fig. 3A), indicating that the N-terminal region
of Yop1p is not necessary for its function.
Characterization of sey1 Alleles
The sey1 alleles were cloned from three of the mutants
by gap repair, and the fourth allele was recovered from two independent PCR amplifications. Molecular lesions were identified by sequencing (Fig. 4). Three of the alleles (sey1-1, sey1-2,
and sey1-3) result from 1-bp mutations that give rise to
amino acid changes: Ser-41-Leu, Gly-106-Asp, and Thr-186-Ile,
respectively. All three of these amino acids are conserved between
SEY1 and RHD3. The sey1-1 mutation is
near the first GTP-binding motif, whereas the sey1-2
mutation disrupts the second GTP-binding motif. Two of the alleles are conditional. The sey1-3 mutant is sect on YPD, but
sectoring on YPD supplemented with 25 mM
CaCl2. The sey1-4 mutant contains a G
to A base pair change, resulting in a stop codon in place of the
Trp-273 codon. Despite this severe truncation resulting in the loss of
more than 60% of the predicted protein, this mutant is sectoring at
37°C, but sect at room temperature (data not shown).
An sey1 mutant (BY2421) was obtained from Research Genetics
(Huntsville, AL), and its growth was compared with its parent wild-type strain (BY4741). When grown at room temperature or 37°C in
YPD liquid media, there is no detectable difference in
A600 between wild type and the sey1
mutant (Fig. 5).
View larger version (13K):
[in this window]
[in a new window]
|
Figure 5.
The Sey1 mutant shows no growth
phenotype at 25°C or 37°C. Liquid YPD media was inoculated with
wild type or sey1 (deletion mutant) and shaken at 25°C
or 37°C for 74 h. The density of the cultures was quantified by
measuring A600. Values shown are
averages ± SE, n = 3. A, At
25°C, there is no significant difference between wild type (solid
line) and the sey1 mutant (dashed line). B, At 25°C,
there is no significant difference between wild type (solid line) and
the sey1 mutant (dashed line).
|
|
A yop1/sey1 Double Mutant Is Impaired in
Vesicular Transport
YOP1 was previously named YIP2 (Ypt1p
interacting protein 2) based on a two-hybrid interaction with the rab
GTPase Ypt1p (http://genome-www4.stanford. edu/cgi-bin/SGD/locus.pl?locus=YIP2; Yang et al., 1998). This entry in
the database provided the first indication that Yop1p is involved in
vesicular transport. In addition, a BLAST search performed with the
Sey1p amino acid sequence against the yeast genome shows weak sequence
similarity to Uso1p (BLAST score 117, 23% identity, 40% similar), a
coiled coil protein believed to be a vesicle tethering factor and
required for vesicular transport (Nakajima et al., 1991; Barlowe,
1997), although Sey1p does not appear to have a coiled coil region
(data not shown). RHD3, an Arabidopsis homolog of
SEY1, had previously been identified in a screen for root
hair abnormalities. Close examination of this mutant shows an
accumulation of transport vesicles in the tip of expanding root hairs,
indicating a problem with vesicular transport (Galway et al., 1997).
These published results, taken together with the genetic interaction of
SEY1 and YOP1 reported here, also indicate that
YOP1 is involved in vesicular transport. To examine the
possibility that the yop1/sey1 synthetic enhancement
mutants have a defect in vesicular transport, transmission electron
microscopy (EM) was used to examine the internal morphology of the mutants.
The yop1/sey1-4 mutant was used to recover white colonies
at 37°C, which were grown in liquid media at 37°C, then shifted to
room temperature for 3 h. Transmission EM on these cells reveals an accumulation of vesicles and occasional ring-shaped structures known
as Berkeley bodies (Novick et al., 1980), which are not seen in
wild-type, yop1, or sey1 cells (Fig.
6).
View larger version (120K):
[in this window]
[in a new window]
|
Figure 6.
A yop1/sey1 double mutant
accumulates transport vesicles. Thin-section EM of wild-type yeast (A),
yop1 mutant (B), sey1 mutant (C), and a
yop1 mutant carrying sey1-4 (D). The
yop1/sey1-4 mutant was used to recover white colonies at
37°C, which were grown in liquid media at 37°C, then shifted to
room temperature for 3 h. n, Nucleus; V, vacuole; Bb,
Berkeley bodies.
|
|
The yop1/sey1 Mutant Is Defective in the
Secretory Process
To determine if the accumulation of vesicles correlates with a
secretion defect, secreted and total invertase activities of the double
mutant were assayed and compared with those of the wild type. Cells
were grown to log phase in media containing 5% (w/v) Glc. The
yop1/sey1-4 mutant was grown at 37°C, then shifted to
room temperature for 3 h. At log phase, invertase expression was
derepressed by transferring the cells to media containing 0.05%
(w/v) Glc. After 3 h, total and external invertase activity was measured. The mutant shows inefficient secretion of invertase, with
only 50% of its total invertase secreted, as compared with more than
80% for wild type (Fig. 7A).
Transformation of the mutant with either a YOP1 or
YOR165W genomic clone rescued the secretion defect (Fig.
7B).
View larger version (31K):
[in this window]
[in a new window]
|
Figure 7.
The synthetic enhancement mutant
has a secretion defect that can be rescued by YOP1
or SEY1. A, Wild type or yop1/sey1-4 mutant
yeast were grown in yeast peptone (YP) supplemented with 5%
(w/v) Glc at 37°C. During log phase, the cultures were washed
and resuspended in YP media supplemented with 0.05% (w/v) Glc
at 25°C. After 3 h, cells were collected and resuspended in 10 mM NaN3. Samples were
divided in two for external invertase assays and total invertase
assays. The values were used to calculate the percentage of invertase
secreted. Values shown are averages ± SE,
n = 3. B, The yop1/sey1-4 mutant was
transformed with pRS315 containing a YOP1 genomic clone, no
insert, or an SEY1 genomic clone. Cultures were
grown in synthetic complete-Leu media supplemented with 5%
(w/v) Glc at 37°C. During log phase, the cultures were washed
and resuspended in synthetic complete-Leu media supplemented with
0.05% (w/v) Glc at 25C. After 2.5 h, cells were collected
and resuspended in 10 mM
NaN3 and processed as in A. Values shown are
averages ± SE, n = 3.
|
|
Yop1p Interacts with Itself, But Not with Sey1p, in a Two-Hybrid
Assay
Synthetic enhancement indicates a functional relationship between
two gene products. In some cases, the gene products physically interact
(Rose, 1995). We wanted to test the possibility of a physical
interaction between Yop1p and Sey1p. To do this, both the
YOP1 cDNA and SEY1 coding sequence were cloned in
frame into both the pBD-GAL4 and pAD-GAL4 two-hybrid vectors. Yeast
strain YRG-2 (Stratagene) was transformed with one binding domain
fusion and one activation domain fusion. Transformants were streaked on
to plates lacking His, Leu, and Trp. After 5 d, growth was monitored. Neither combination of YOP1 and SEY1
fusion constructs allowed for growth, revealing no interaction between
the fusion proteins (data not shown). To investigate the possibility
that Yop1p or Sey1p form a homodimer, YRG-2 was transformed with both YOP1 fusion constructs or with both SEY1 fusion
constructs. Transformants carrying both SEY1 fusion
constructs showed no growth (data not shown). Transformants carrying
both YOP1 fusion constructs showed growth comparable with
the positive control, whereas transformants carrying one fusion
construct and one empty vector did not grow (Fig.
8A). Interestingly, when the fusion
constructs were made using the YOP1 genomic sequence, which
contains one intron, no interaction could be detected (data not shown).
Sequence analysis shows that the predicted Yop1p protein is Leu rich
and contains a potential Leu zipper (Fig. 8B), a motif characterized by
a Leu, Ile, or Val every seventh residue, and allows for dimer
formation.
View larger version (29K):
[in this window]
[in a new window]
|
Figure 8.
Yop1p interacts with itself in a
two-hybrid assay. A, Yeast host strain YRG-2 transformants carrying: i,
positive control plasmids pADWT and pBDWT; ii, GAL4AD-YOP1 and GAL4BD;
iii, GAL4AD and GAL4BD-YOP1; or iv, GAL4AD-YOP1 and GAL4BD-YOP1 were
streaked out on synthetic media lacking Leu, Trp, and His and grown at
30°C for 5 d. B, The amino acid sequence of Yop1p showing a
potential Leu zipper (highlighted).
|
|
| |
DISCUSSION |
Homologs of the plant stress-induced gene, HVA22, are
present in diverse eukaryotic organisms. We have shown previously that expression of the yeast HVA22 homolog, YOP1, is
also induced by NaCl treatment (Shen et al., 2001) Taking advantage of
the versatile genetic features of yeast, we have studied the yeast
YOP1 gene as a part of the effort in elucidating the
function of HVA22 in stress response. Because a
yop1 mutant does not have a severe phenotype, a synthetic
enhancement screen in a yop1 background was performed to
identify genetic interactions with YOP1. From this screen, a
single complementation group, composed of four alleles of YOR165W, was
identified. We have named this gene SEY1 (synthetic
enhancement with YOP1). It is homologous to the Arabidopsis gene RHD3, which encodes a protein containing two
GTP-binding motifs. These motifs are present in Sey1p as well (Fig. 4;
Wang et al., 1997). The interactions between YOP1,
SEY1, and several other yeast genes involved in vesicle
traffic are summarized in Figure
9.
View larger version (28K):
[in this window]
[in a new window]
|
Figure 9.
Summary of reported interactions involving
YOP1 and SEY1. Interactions involving
YOP1 and SEY1 are shown schematically and are
listed as follows: 1 and 2, two-hybrid interaction (Yang et al., 1998);
3, two-hybrid interaction and glutathione
S-transferase pull down (Calero et al., 2001); 4, synthetic enhancement (this study); 5, synthetic enhancement
(Sapperstein et al., 1996); 6 through 9, two-hybrid interaction (Ito et
al., 2001a); 10, sequence similarity (this study).
|
|
A synthetic enhancement interaction can represent several kinds of
relationships between gene products. One possibility is simple
redundancy, where the two proteins perform the same biochemical function (Rose, 1995). Yop1p and Sey1p do not show any kind of sequence
similarity, so it is unlikely that they have the same biochemical
activity. Another possible relationship is that of functioning in
parallel pathways that have a common endpoint. For example, a Leu
auxotroph, such as a leu2 mutant, will grow normally on
media that contains Leu. However, in a genetic background in which a
gene required for sensing or transport of extracellular Leu is
disrupted, the LEU2 gene is essential (Nigavekar and Cannon, 2002). A third possibility is that the two gene products physically interact in a complex. Elimination of one of those products may not be
enough to disrupt the complex, but elimination or mutation of two
components will (Rose, 1995). Two-hybrid analyses do not show an
interaction between Yop1p and Sey1p. Furthermore, there is no report of
Sey1p physically interacting with any other protein. In contrast, Yop1p
has been reported to physically interact with several proteins,
including Ypt1p, Yip1p, and Yif1p, as summarized in Figure 9 (Ito et
al., 2000, 2001a, 2001b). A synthetic enhancement interaction may also
represent two steps of a single pathway. In this case, disruption of
one protein may reduce the efficiency of a single step, but not enough
to result in a severe phenotype. However, if two steps in the pathway
are disrupted, the overall flux of the pathway may be greatly reduced
(Rose, 1995).
We propose that Yop1p and Sey1p function together in the process of
membrane fusion. There are three major steps involved in membrane
fusion. The first is tethering, which is dependent on YPT GTPases,
and is mediated by peripheral membrane proteins. The second step is
docking, or SNARE engagement, in which the integral membrane coiled
coil SNARE proteins, located on both vesicle and target membranes, form
very stable complexes, holding the vesicle in close proximity to the
target membrane. Finally, there is fusion of the lipid bilayers
(Pelham, 2001). Sequence homology searches indicate that many of the
molecular components of the secretory pathway are conserved between
plants and fungi. It may be that Yop1p and Sey1p act in the same or
sequential steps in this pathway, which would explain the synthetic
interaction between the two.
rhd3 Phenotype
In plants, root hairs serve to increase the surface area of roots
for increased water and mineral absorption from the soil. They are
extensions of single cells, and are an extreme example of polarized
cell expansion. This growth takes place at the very tip, and is made
possible by the delivery of cell wall components to the expanding zone
via membrane-bound vesicles (Gilroy and Jones, 2000). When
RHD3, the Arabidopsis homolog of SEY1, is
mutated, the root hairs become short and wavy (Galway et al., 1997).
Closer examination reveals an accumulation of vesicles in the root hair tip, indicating a problem with vesicle fusion (Galway et al., 1997). It
is interesting to note that although the Arabidopsis rhd3
mutant has a very noticeable phenotype, deletion of SEY1 in
yeast results in no noticeable morphological abnormalities (Fig. 6C) or
growth phenotype (Fig. 5).
SEY1 Alleles
Calcium has been established as an important player in both
regulated and constitutive membrane fusion, although its exact role
remains unclear. High levels of Ca2+ can rescue
the temperature-sensitive ypt1-1 mutant at the nonpermissive temperature (Schmitt et al., 1988). It may be that high levels of
calcium facilitate some step of membrane fusion, which could overcome
the inefficiency of fusion caused by the yop1/sey1-3 synthetic interaction. The yop1/sey1-3 mutant is
sectoring when grown on media containing additional calcium, suggesting
that the sey1-3 allele maintains some level of activity, and
that Sey1p acts in some process involving calcium.
The SEY1-4 allele is conditional because it has sufficient
activity at 37°C to allow sectoring, but not at room temperature. It
is somewhat surprising that this allele has any activity because the
mutation causes a truncation resulting in the loss of more than 60% of
the predicted protein. Because the other three recovered alleles have
lesions in the first quarter of the coding sequence, it is suggested
that the N-terminal region is the most important section of the
protein. This is where the putative GTP-binding motifs are located. In
fact, the sey1-1 lesion is very close to the first
GTP-binding motif, and the sey1-2 lesion disrupts the second
GTP-binding motif.
Yop1p N Terminus Is Not Required for Function
It has been previously shown that Yop1p and Yip1p interact
physically. Two-hybrid data show that the N-terminal 17 amino acids of
Yop1p are necessary and sufficient for this interaction (Calero et al.,
2001). Because YIP1 is essential for viability, whereas YOP1 is not, it is clear that the direct physical
interaction between these two proteins is not necessary for the
function of Yip1p. Because no biochemical function has been determined
for Yop1p, it is not possible to directly assay for activity of the protein. However, the yop1/sey1 sect mutants provide a
useful background for performing functional assays for Yop1p activity. The ability of the YOP1 N-terminal deletion mutant to
restore sectoring to the yop1/sey1 double mutants clearly
demonstrates the activity of the truncated protein. Based on this
observation, it can be inferred that the direct physical interaction
between Yop1p and Yip1p is not required for the function of Yop1p.
Large-scale two-hybrid analyses have identified proteins that interact
with both Yip1p and Yop1p, including Yif1p and YLR324W (Ito et al., 2000, 2001a, 2001b; summarized in Fig. 9). It is possible that these
four proteins form a complex, and the elimination of the direct
interaction between Yip1p and Yop1p is not sufficient to disrupt the
complex. Although no two-hybrid interactions involving Sey1p have been
reported, an interesting possibility is that Sey1p somehow acts to
stabilize this complex and that in a yop1 background, mutation of SEY1 may be sufficient to disrupt its formation.
Alternatively, it is possible that the association of Yip1p and Yop1p
is simply not functionally important.
Yop1p Interactions with Yip1p and GTPases
Calero et al. (2001) have recently demonstrated a physical
interaction between Yop1p and Yip1p (Ypt1p-interacting protein) and the
GTPases Sec4p, Ypt6p, and Ypt7p. In addition, Yop1p had previously been
in a two-hybrid screen with Ypt1p
(http://genome-www.stanford.edu/Saccharomyces/; Yang et al., 1998). The
YPT family of proteins includes rab-like GTPases that play important
roles in vesicular trafficking. Their function has been reviewed
recently (Segev, 2001a, 2001b). Ypt GTPases cycle between GTP- and
GDP-bound forms. The GTP to GDP switch is accomplished through GTP
hydrolysis, whereas the GDP to GTP switch is accomplished by nucleotide
exchange. Four classes of proteins regulate this cycling: GAP
(GTPase-activating protein), GEF (guanine nucleotide exchange factor),
GDF (guanine dissociation factor), and GDI (GDP dissociation
inhibitor). Vesicular fusion is mediated by the GTP-bound form of Ypt,
so this form is considered active, whereas the GDP-bound form is
considered inactive. As such, interacting factors involved in the
formation or maintenance of GTP-bound Ypt (GEF and GDF) are positive
regulators of Ypt function, whereas those that favor the GDP-bound form
(GAP and GDI) are negative regulators of Ypt. Because GAP and GDI are
important in the recycling of Ypt back to the membrane of the donor
compartment, they have a positive role as well (Segev, 2001a,
2001b).
Although it is clear that Yop1p is involved in the process of membrane
fusion, it is not clear why this protein is up-regulated by
environmental stress. It has been reported that yeast and plants change
the lipid composition of their plasma membrane in response to
environmental conditions (Alexandre et al., 1994; Sharma et al., 1996;
Smaoui and Cherif, 2000; Hamrouni et al., 2001). The newly synthesized
lipids are in large part delivered to the plasma membrane by transport
vesicles (Holthuis et al., 2001). It is possible that the cell will
regulate components of the vesicular transport machinery in response to
environmental stress to hasten the changes in membrane lipid composition.
Alternatively, Yop1p may be a negative regulator of Ypt proteins, which
is suggested by the observation that overexpression of YOP1
blocks secretion. This seems a likely function of HVA22, the barley
homolog of Yop1p, considering its regulation and the physiology of the
barley seed. The primary function of the barley aleurone layer is the
secretion of hydrolytic enzymes into the endosperm. Under favorable
conditions, the phytohormone GA is produced by the embryo of the seed,
and stimulates the production of hydrolytic enzymes by the aleurone
layer. These are secreted into the endosperm, releasing sugars and
amino acids to support the growth of the germinating embryo. If
conditions become unfavorable, the phytohormone ABA is produced and
antagonizes the action of GA (Lovegrove and Hooley, 2000).
HVA22 was initially identified as an ABA-induced transcript
in barley aleurone layers. If the antagonistic actions of ABA and GA
serve to regulate secretion, a reasonable possibility is that HVA22
acts as a negative regulator of secretion, and may block secretion at
elevated levels of expression. In vegetative tissues, GA promotes
growth and elongation of tissues, whereas ABA blocks that action. In
this case, ABA could be acting to block the expansion of cell membranes
and the delivery of cell wall components as a mechanism to stop growth.
Transgenic Arabidopsis containing T-DNA disruptions of the five known
HVA22 homologs (AtHVA22a-e) have been obtained recently. In
addition, RNAi lines designed to knock down expression of
AtHVA22a and AtHVA22d, as well as plants
expressing AtHVA22a and AtHVA22d under the
control of the 35S promoter, have been generated. These plants are all
viable, and more detailed phenotypic characterization is under way (N. Chen, personal communication).
| |
MATERIALS AND METHODS |
Sequence Analyses
Amino acid sequences were aligned by MegAlign (DNA Star Inc.,
Madison, WI). Identification of similar residues was determined by SeqVu version 1.0 (The Garvan Institute of Medical Research, Sydney) using 90% on the Goldman, Engelman, Steitz scale.
Kyte-Doolittle hydrophilicity plots (Kyte and Doolittle, 1982)
were generated by Protean (DNA Star Inc.) using a 9-amino acid window.
Strains, Growth Media, and Growth Conditions
Yeast (Saccharomyces cerevisiae) strains were
grown in 1% (w/v) yeast extract, 2% (w/v) peptone, 2% (w/v)
Glc (YPD), or synthetic media supplemented with the appropriate
nutrients. Growth curve experiments were performed by inoculating 50 mL
of media with 100 µL of stationary phase culture that had been grown
at room temperature. Three independent cultures for each strain were
inoculated, and cell density determined by reading the
A600 determined by making a dilution of the
culture sufficient to keep the spectrophotometer reading below
0.5. Yeast transformation was done by the lithium acetate method (Ito,
1983).
Deletion of the YOP1 Coding Sequence
The KANmx cassette was amplified with the primers
5'-TTGG-TAGTGAAAACAAATAAACAAAGACATAACCGCACTCCAATCCAGCT-GAAGCTTCGTACGC-3' and
5'-GAACAAAAACGAGAGTTTGATTTGA-GGATATAGGTGAGTTGCCTCGCATAGGCGACTAGTGGATCTG-3'. The product was transformed into w303a/
and transformants selected for on YPD plates containing 100 µg mL
1 geneticin
(Sigma, St. Louis). Several transformants were screened by PCR to
confirm the presence of the cassette at the YOP1 locus. One of the positive clones was designated ABY15. This strain was sporulated and dissected to create the strain ABY16.
DNA Constructs and Manipulations
Sequencing was done with the ABI Prism BigDye Terminator Cycle
Sequencing Ready Reaction Kit (Perkin-Elmer Applied Biosystems, Foster
City, CA) according to the manufacturer's instructions.
The YOP1 genomic clone was constructed by gap repair.
Plasmid pSW896 (ADE3 URA3 CEN; Murphy et al., 1996) was
cut with NotI, treated with Klenow, and religated to
destroy the NotI site, creating pABS28. The
YOP1 genomic region was amplified with the primers 5'-ATATATGGATCCTTGAAAAACTGTGGAGCC-3' and
5'-TATATAGGATCCTCTTCCAC-AATAAACGCC-3'. The PCR product was cloned
into the TA cloning vector pCRII (Invitrogen, Carlsbad, CA),
making pAB21. The resulting plasmid was cut with BstBI and
NdeI to drop out the YOP1 coding sequence
but leave behind the flanking genomic sequence. A linker fragment made
by annealing primers 5'-CGAATGCGGCCGCACA-3' and
5'-TATGTGCGGCCGC-ATT-3' was cloned into pABS28, which had been cut
with BstBI and NdeI to introduce an NotI
site, making pAB22. The BamHI insert from this plasmid
was ligated into pABS28 to make pABS29. This plasmid was cut with
NotI and transformed into yeast strain w303-1a to create
a high-fidelity YOP1 clone by gap repair. The resulting plasmid (pABS30) was recovered by plasmid rescue, transformed into
Escherichia coli, and checked by restriction digestion.
The Yop1p N-terminal deletion construct was made by introducing
restriction sites to the genomic clone by PCR. The promoter region was
amplified using the primers 5'-ATATATGGATCCTTGAAAAACTGTG-GAGCC-3' and 5'-CATGCCATGGTTGGAGTGCGGTTATG-3'. The second exon and
5'-untranslated region were amplified with the primers
5'-CATGCCATGGGTAAGTACTCTGGTAATAG-3' and
5'-TATATAGGATC-CTCTTCCACAATAAACGCC-3'. The resulting products were
cut with NcoI and BamHI for cloning into
the vector pRS315 (LEU2 CEN).
Yeast Mutagenesis and Screening for Synthetic Enhancement
Mutants
ABY16 was mated to YCH128, sporulated, and dissected to make the
strains ABY29 and ABY30. These were both transformed with plasmid
pABS30 (YOP1 URA3 ADE3 CEN) to create the strains ABY41 and ABY42. When grown on nonselective plates, these strains produce sectoring colonies. Overnight cultures of yeast strains ABY41 and ABY42
grown in synthetic complete minus uracil media were used to
inoculate 100 mL of synthetic complete minus uracil media. The cultures
were grown to early log phase. For each culture, 5 × 108 cells were harvested and washed twice with 10 mL of 50 mM potassium phosphate buffer, pH 7.0. The cells were
resuspended in 10 mL of potassium phosphate buffer. Thirty microliters
of ethyl methanesulfonate was added to 1 mL of each sample, then
incubated for 70, 80, 90, and 100 min. At the end of the incubation
time, 1 mL of 10% (v/v) filter-sterilized fresh sodium
thiosulfate was added to each tube. The samples were washed twice with
10 mL of water and resuspended in 1 mL of water. Dilutions were plated
on rich media to determine survival rate. Cells from the 80-min
treatment had an approximately 20% survival rate, and were used for
screening. Approximately 120,000 total colonies were screened for
sect. After sect colonies were streaked on rich media to confirm
the sect phenotype, sect strains were isolated and subjected to
complementation group analysis and tested for YOP1 and
URA3 dependence. For complementation group analysis, all
ABY41-derived strains were crossed to all ABY42-derived strains, and
scored for sectoring. Four mutants belonging to one complementation
group were found to become sect+ upon transformation with pABS33
(YOP1 LEU2), but not upon transformation with pABS32 (URA3 LEU2), indicating dependence upon
YOP1.
Cloning of YOR165W
sect mutant ABY51 was transformed with a genomic library
(LEU2 CEN), and transformants were screened for
sectoring. Sectoring colonies were picked and restreaked on C-LEU
plates. Individual colonies were used to grow overnight cultures for
plasmid recovery. The inserts of recovered plasmids were end sequenced
to determine the genomic region they contained. Subcloning and deletion
analysis was carried out to identify ORF YOR165W as the complementing gene.
Cloning of YOR165W Alleles
Plasmid pABS34 (SEY1 LEU2) was cut with
SnaBI, releasing an 8,928-bp fragment containing the
SEY1 locus. The plasmid was religated to make pABS45.
This plasmid was cut with SnaBI and transformed into the
mutants to clone the SEY1 locus by gap repair. Recovered plasmids were checked by restriction digest for the presence of SEY1, and sequenced. sey1-2 was resistant
to cloning by this method, and therefore was cloned by PCR using the
primers 5'-GAGTTTCGCTTGTACAGCATTAGAT-3' and
5'-TGAACA-ATTTTGGGAGACTGTATT-3'. Clones from two independent PCR
reactions were sequenced, and found to contain the same lesion.
EM
Yeast cultures were grown in YPD media to log phase. White
colonies derived from the conditional mutant ABY53 were isolated at
37°C. Liquid media was inoculated with a well-isolated white colony
and the culture grown at 37°C to log phase, then shifted to 25°C
for 2 h. Samples were washed with water, and then resuspended in
2.5% (v/v) gluteraldehyde in 40 mM phosphate
buffer, pH 6.5, and 0.5 mM MgCl2. Samples were
incubated at room temperature for 2 h, then washed twice with 5 mL
of 0.1 M PO4 citrate buffer, pH 5.8. To remove cell walls,
cells were treated with zymolyase for 2 h. Samples were then
washed twice with 5 mL of 1 M NaOAc pH 6.2. Post-fixation
was done with 2% (w/v) osmium tetroxide followed by 1%
(w/v) uranyl acetate and the samples were embedded in Spurr's resin.
Invertase Assays
Yeast cultures were grown in YP media supplemented with 5%
(w/v) Glc to log phase. 1 × 108 cells were
harvested, washed twice with water, and resuspended in YP media
containing 0.05% (w/v) Glc. After 3 h of shaking, NaN3 was added to 10 mM, the samples were
aliquoted into five 1-mL samples, washed with 10 mM
NaN3, then resuspended in 10 mM NaN3. Each sample was divided in two for total and external
invertase assays. External invertase activity was measured from intact
cells. For total activity, Triton X-100 was added to 0.2%
(v/v), and the samples were frozen and thawed twice on dry ice
to lyse the cells. For the assays, a 50-µL aliquot of cells was mixed
with 100 µL of 0.2 M sodium acetate, pH 5.1, at 37°C.
To start the reaction, 50 µL of 0.5 M Suc (Pfanstiehl
Laboratories, Waukegan, IL) was added to the samples. After 20 min, the
reaction was stopped by adding 300 µL of 0.2 M
K2HPO4. After mixing, 100 µL of this mix was
added to 400 µL of 0.2 M K2HPO4
in a glass tube and immediately boiled for 3 min. A 2-mL aliquot of a
solution containing 0.1 M potassium phosphate (pH 7.0), 20 µg mL1 Glc oxidase, 2.5 µg mL1
peroxidase, and 150 µg mL1 O-dianisidine
was added to each sample and incubated at 37°C. After 30 min,
the reaction was stopped by adding 2 mL of 6 M HCL and
absorption was read at 540 nm.
Two-Hybrid Assays
The YOP1 cDNA was amplified from total RNA by
reverse transcriptase-PCR using primers to add restriction
sites for cloning. For cloning into the pBD-GAL4 vector (Stratagene),
the primers 5'-CGGGTCGACTCAT-GTCCGAATATGCATCT-3' and
5'-TAACTGCAGTTAATGAACAGAAG-CACC-3' were used to add SalI
and PstI sites. For cloning into the pGAD424 vector
(CLONTECH Laboratories, Palo Alto, CA), the primers
5'-TAGGGATCCGTATGTCCGAATATGCATCT-3' and
5'-TAACTGCAGTTA-ATGAACAGAAGCACC-3' were used to add
BamHI and PstI sites. Constructs were transformed
into YRG-2. Transformants were streaked out on C-Leu,-Trp, and -His and
scored for growth after 5 d at 30°C.
The authors thank Susan Wente, Kathy Iovine, Mirella Bucci,
Albert Ho, and Kathy Ryan for protocols, reagents, and technical advice; Mike Veith for technical assistance with EM; Ken Blumer for use
of his tetrad dissection apparatus; and members of the Ho lab for
technical assistance, discussion, and comments on the manuscript.
Received April 26, 2002; returned for revision June 3, 2002; accepted July 19, 2002.
Article, publication date, and citation information can be found at
www.plantphysiol.org/cgi/doi/10.1104/pp.007716.
TOP
ABSTRACT
INTRODUCTION
