See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/23960687

Ubiquitin ligase Rsp5p is involved in the

gene expression changes during nutrient
limitation in Saccharomyces cerevisiae
ARTICLE in YEAST JANUARY 2009
Impact Factor: 1.63 DOI: 10.1002/yea.1645 Source: PubMed

CITATIONS

READS

21

3 AUTHORS:
Fernando Cardona

Agustn Aranda

Spanish National Research Council

Spanish National Research Council

15 PUBLICATIONS 74 CITATIONS

36 PUBLICATIONS 854 CITATIONS

SEE PROFILE

SEE PROFILE

Marcell del Olmo

University of Valencia
14 PUBLICATIONS 449 CITATIONS
SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate,

letting you access and read them immediately.

Available from: Fernando Cardona

Retrieved on: 19 January 2016

Yeast
Yeast 2009; 26: 115.
Published online in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/yea.1645

Research Article

Ubiquitin ligase Rsp5p is involved in the gene

expression changes during nutrient limitation
in Saccharomyces cerevisiae
F. Cardona1,2# , A. Aranda1,2 *and M. del Olmo1
1 Department
2 Department

of Biochemistry and Molecular Biology, University of Valencia, Spain

of Biotechnology, IATA (CSIC), Valencia, Spain

*Correspondence to:
A. Aranda, Departament de
Bioqumica i Biologia Molecular,
Universitat de Val`encia, Apartado
73, Burjassot, Valencia
46100, Spain.
E-mail: agustin.aranda@uv.es
# Present address: Unitat de
Gen`etica Molecular, Institut de
Biomedicina de Val`encia (CSIC),
Valencia, Spain.

Received: 5 June 2008

Accepted: 21 October 2008

Abstract
Rsp5p is an essential ubiquitin ligase involved in many different cellular events,
including amino acid transporters degradation, transcription initiation and mRNA
export. It plays important role in both stress resistance and adaptation to the change
of nutrients. We have found that ubiquitination machinery is necessary for the correct
induction of the stress response SPI1 gene at the entry of the stationary phase. SPI1
is a gene whose expression is regulated by the nutritional status of the cell and whose
deletion causes hypersensitivity to various stresses, such as heat shock, alkaline stress
and oxidative stress. Its regulation is mastered by Rsp5p, as mutations in this gene
lead to a lower SPI1 expression. In this process, Rsp5p is helped by several proteins,
such as Rsp5p-interacting proteins Bul1p/2p, the ubiquitin conjugating protein Ubc1p
and ubiquitin proteases Ubp4p and Ubp16p. Moreover, a mutation in the RSP5 gene
has a global effect at the gene expression level when cells enter the stationary phase.
Rsp5p particularly controls the levels of the ribosomal proteins mRNAs at this stage.
Rsp5p is also necessary for a correct induction of p-bodies under stress conditions,
indicating that this protein plays an important role in the post-transcriptional fate of
mRNA under nutrient starvation. Copyright 2009 John Wiley & Sons, Ltd.
Keywords:

RSP5; SPI1 ; stress response; stationary phase; p-bodies

Introduction
Nutrient depletion in Saccharomyces cerevisiae
induces a plethora of physiological, biochemical
and morphological changes. A yeast culture grown
on glucose-based medium changes from a fermenting metabolism to a respiratory one when glucose
becomes limited. This phenomenon is called the
diauxic shift. When the carbon source is exhausted,
the culture enters the so-called stationary phase
(Gray et al., 2004; Herman, 2002). This state
allows cell survival over long periods of time without added nutrients. Cells in stationary phase cultures accumulate glycogen and trehalose, develop a
thickened cell wall and become resistant to stresses
such as increased temperature and oxidative stress.
Copyright 2009 John Wiley & Sons, Ltd.

Entry into the stationary phase has been described

under laboratory conditions (Gray et al., 2004; Herman, 2002). The target of rapamycin kinase (TOR)
and protein kinase A (PKA) transduction signal
pathways coordinately respond to nutrient availability, thus favouring growth and cellular division
and repressing the stress response (Jorgensen et al.,
2004; Martin and Hall, 2005). One of the gene families involved in cell growth that is regulated by
these pathways and depends on nutrient availability is the ribosomal proteins (RP) (Jorgensen et al.,
2004; Powers, 2004). On the other hand, these
pathways under nutrient depletion block the cell
cycle and allow the activation of stress response
mechanisms (Hohman and Mager, 2003, and references therein). Several reports have described

extensive transcriptomic changes under stationary

phase conditions (DeRisi et al., 1997; Gasch et al.,
2000). Recently the existence of a large number of
extraction-resistant mRNAs in the stationary phase
has been described (Aragon et al., 2006). These
transcripts are protein-bound, as they are extracted
by protease action and are rapidly released after a
stressful condition such as oxidative shock. Under
stress conditions, such as nutrient deprivation, this
phenomenon has been linked to the induction of
mRNA relocalization to sites of degradation or
storage of RNA called processing bodies (p-bodies)
(Parker and Sheth, 2007). RNAs accumulated in
these protein-bound structures can return to translation when the stress disappears (Brengues et al.,
2005). Therefore, they are a post-transcriptional
way to regulate the gene expression.
Ubiquitination is a complex protein modification
process that involves a cascade of E1 (ubiquitinactivating), E2 (ubiquitin-conjugating) and E3
(ubiquitin ligases) enzymes which regulate ubiquitination specificity (Pickart, 2004). Polyubiquitin
chains are built through the K29 and K48 residues
of ubiquitin target proteins for proteasomal degradation by the 26S proteasome (Hochstrasser, 1996),
whereas polyubiquitin K63 chains are associated
with stress resistance, DNA repair, signal transduction, and degradation via the vacuole (Horak, 2003;
Sigismund et al., 2004). Mono-ubiquitination at
K63 controls many processes, including membrane trafficking, meiosis and chromatin modelling
(Hicke, 2001). Ubiquitinated proteins undergo deubiquitination by proteases called de-ubiquitinating
enzymes (DUBs) prior to degradation (Pickart,
2004). The yeast HECT-domain ubiquitin-ligase
Rsp5p, which adds mono- and polyubiquitin chains
linked through K63, has been involved in a
wide variety of physiological processes, such as
minichromosome and actin cytoskeleton maintenance, mitochondrial inheritance, drug resistance,
regulation of intracellular pH, biosynthesis of fatty
acids, cell wall organization and protein sorting
(Horak, 2003; Kaminska et al., 2005; Kus et al.,
2005). Recent studies have linked Rsp5p to all
the levels of gene expression. The activity of transcription factor Spt23p is specifically activated by
Rsp5p-mediated proteolysis (Hoppe et al., 2000).
At the general transcription machinery level, Rsp5p
interacts with RNA polymerase II CTD and stimulates its phosphorylation (Max et al., 2007) and
ubiquitination, as well as the destruction of RNA
Copyright 2009 John Wiley & Sons, Ltd.

F. Cardona, A. Aranda and M. del Olmo

pol II in response to DNA damage (Somesh et al.,

2005). The export of all three kinds of RNA
requires Rsp5p (Neumann et al., 2003) and also
Hpr1p, a member of the THO/TREX (transcription/export) complex, which is regulated by Rsp5mediated ubiquitination (Gwizdek et al., 2005).
Finally, recent data indicate an important role of the
Rsp5p-mediated ubiquitination in translation accuracy (Kwapisz et al., 2006).
There are evidences that links Rsp5p to the
stress response in S. cerevisiae. Rsp5p binds to
two homologous proteins, Bul1p and Bul2p, and
the double disruptant bul1bul2 is sensitive to various stresses, including a growth defect on a
non-fermentable carbon source (Yashiroda et al.,
1998). The Rsp5pBul1pBul2p complex is necessary for heat shock element (HSE)-mediated transcription (Kaida et al. 2003). This ubiquitin ligase has been described to regulate the expression
of stress proteins via post-translational modification of the stress response transcription factors
Hsf1p and Msn4p (Haitani et al., 2006). Recently
it has been shown that Rsp5p is required for the
nuclear export of HSF1 and MSN2/4 mRNAs
under stress conditions (Haitani and Takagi, 2008).
Besides, cells overexpressing Rsp5p or ubiquitinconjugation enzymes (E2, Ubc1Ubc13) display
enhanced tolerance to several forms of stress (heat
shock, osmotic stress, oxidative stress and ethanol)
(Hiraishi et al., 2006).
We are interested in the transcriptional response
at postdiauxic and stationary phases and we have
used SPI1, one of the yeast genes induced under
these conditions during industrial (Puig and PerezOrtn, 2000) and laboratory conditions (Gasch
et al., 2000), as a model. Its transcription also rises
under several stress conditions (Gasch et al., 2000).
It encodes a protein which is important for the
structure and biogenesis of the cell wall (Horie
and Isono, 2001). In this study we analysed the
relevance of SPI1 under several stress conditions
and we particularly focused on its response during
entrance into the stationary phase. We showed
that the expression of this gene is affected by
not only the ubiquitin-ligase Rsp5p but also some
interacting proteins, and we provide evidence of
this proteins involvement in the tolerance to this
and to other forms of stress. Furthermore, we have
found that Rsp5p represses the expression levels
of ribosomal proteins (RPs) at the start of the
Yeast 2009; 26: 115.
DOI: 10.1002/yea

Rsp5p controls gene expression at the entry of stationary phase

Table 1. Yeast strains used in this work

Strain
BY4742
BY4742 spi1
YPH499
FAY18A
FAY29E
FAJ72
yYDcG
yHDcG
yBDcG
yYDhG
yHDhG
yBDhG
BY4741
BY4741 ubi4
BY4741 ubc7
BY4741 ubp4
BY4741 ubp6
BY4741 ubp14
BY4741 ubc2
BY4741 ubc1
BY4741 ubc4
BY4741 ubc5
W303-1a
2344c (WT)
27038a (npi1)
OS2718(bul1bul2)
MLY40 (WT)
XPY80 (sok2)

Description

Origin

MAT, his3-1, leu2-0, lys2-0, ura3-0

BY4742 spi1::KanMX4
MATa, his3-200, leu2-1, lys2-801, trp1-1, ade2-101, ura3-52
YPH499 HPG1-1 (Pro514>Thr)
YPH499 HPG1-4 (Ala799>Thr)
YPH499 bul1::URA3,bul2::HIS3
YPH499 DCP2-GFP (kanMX)
FAY18A DCP2-GFP (kanMX)
FAJ72 DCP2-GFP (kanMX)
YPH499 DHH1-GFP (kanMX)
FAY18A DHH1-GFP (kanMX)
FAJ72 DHH1-GFP (kanMX)
MATa, his3-1, leu2-0, met17-0, ura3-0
BY4741 ubi4::KanMX4
BY4741 ubc7::KanMX4
BY4741 ubp4::KanMX4
BY4741 ubp6::KanMX4
BY4741 ubp14::KanMX4
BY4741 ubc2::KanMX4
BY4741 ubc1:: KanMX4
BY4741 ubc4:: KanMX4
BY4741 ubc5:: KanMX4
MATa ade2-1, ura3-1, leu2-3, his3-1, trp1-1
MAT ura3
MAT npi1 ura3
MAT bul1::KanMX4bul2::KanMX4 ura3
MAT ura3-52
MAT sok2::hygB ura3-52

EUROSCARF
EUROSCARF
F. Abe
F. Abe
F. Abe
F. Abe
This work
This work
This work
This work
This work
This work
EUROSCARF
EUROSCARF
EUROSCARF
EUROSCARF
EUROSCARF
EUROSCARF
EUROSCARF
EUROSCARF
EUROSCARF
EUROSCARF
R. Rothstein
B. Andre
B. Andre
B. Andre
J. Heitman
J. Heitman

stationary phase and that it has a positive role in pbody formation in the glucose-depleted condition.

Materials and methods

Yeast strains and growth conditions
The yeast strains used in this work are listed
in Table 1. For yeast growth, YPD medium (1%
w/v yeast extract, 2% w/v bactopeptone, 2% w/v
glucose), SD or SC medium (0.17% w/v yeast
nitrogen base without amino acids and ammonium
sulphate, 0.5% w/v ammonium sulphate, 2% w/v
glucose) supplemented with the required amino
acids, were used. Cultures were incubated at 30 C
with shaking. Solid plates contained 2% agar
and the specific plates contained 0.008% SDS,
0.1 g/ml rapamycin and cycloheximide.
To analyse stress resistance, the cells from exponential cultures in the YPD medium were affected
by the following adverse conditions: ethanol addition up to 12% v/v final concentration and incubation for 1 h; heat shock at 45 C for 1 h; shift
Copyright 2009 John Wiley & Sons, Ltd.

to YPD containing 1 M KCl and incubation for

1 h (osmotic stress); and YPD buffered at pH 2
or 10 for acid or basic stress, respectively. In all
cases, cell viability was determined by counting the
number of colonies growing in YPD plates from
appropriate dilutions of cultures. To analyse resistance to oxidative stress, the diameter of the growth
inhibition region produced by a paper disc with
10 l 33% H2 O2 on stationary phase cultures on
YPD plates was measured (Stephen et al., 1995).
For microscopy studies, the cells in the exponential phase were transferred from YPD to YP for
10 min and were then washed with SC. Observations were made using a Nikon PCM 2000 microscope, using a 100 objective with a 3 zoom.

Construction of yeast strains

In order to analyse the transcriptional regulation of
the SPI1 gene, a fusion of the promoter of this
gene with the lacZ reporter gene was constructed
in the Yep357 plasmid. For this purpose, we cloned
the intergenic region of the SPI1PEA2 intergenic
Yeast 2009; 26: 115.
DOI: 10.1002/yea

region by PCR and added the XbaI and EcoRI

restriction sites to the oligonucleotides used (see
Supporting information, Table S1).
Dcp2p and Dhh1p proteins were C-terminal
tagged with GFP from plasmid pFA6GFP(S65T)
KanMX6, following the PCR-based gene modification method described by Longtine et al. (1998),
using the appropriate oligonulceotides (see Supporting information, Table S1). The appropriate
strains were transformed.

-galactosidase analysis
The SPI1placZ fusion expression was determined
as -galactosidase activity via the method of permeabilized cells, using ONPG as a substrate, as
described by Adams et al. (1997). To detect blue
colonies, plates containing SC ura + Xgal
0.4 mg/ml + phosphate buffer, pH 7, were used.
The white colonies defective in SPI1placZ fusion
activity were transformed with a library constructed
in the episomal plasmid YEp32 (a gift from J. C.
Igual). The transformation was replica-plated on Xgal plates.

RNA analysis
RNA isolation, quantification and analysis were
carried out as previously described (Carrasco et al.,
2003). The expression of the SPI1 gene on several
strains and conditions was followed by Northern
blot analysis, using a specific probe for this gene
obtained with the oligonucleotides SPI1-F/G (see
Supporting information, Table S1).
For microarray analyses, cDNA preparation,
labelling and hybridization were carried out as
described by Fazzio et al. (2001). The combinations for hybridization Cy3Cy5 were: WTa
HPG11a, HPG11bWTb, where the a and b
samples were obtained from two independent cultures of each strain. The intensity obtained in
each channel for each pair of microarrays was
normalized by Lowess (Yang et al., 2002). The
overrepresentation of categories containing functionally related genes in each strain was statistically analysed using the Function Associate
tool (http://llama.med.harvard.edu/cgi/func/func
associate). The analysis of the transcription factors involved in the expression of differentially
expressed genes was carried out with the YEASTRACT package (http://www.yeastracts.com).
Copyright 2009 John Wiley & Sons, Ltd.

F. Cardona, A. Aranda and M. del Olmo

The expression of some genes with the differential expression in the wild-type (WT) and HPG1-1
strains in the microarray analysis was also analysed
by semiquantitative RTPCR. cDNA preparation
was carried out as previously described (JimenezMart et al., 2007). For the amplification of specific
genes in the resulting cDNA, it was diluted 20-fold
and 5 l were used for a PCR reaction carried out in
a final volume of 1525 l, containing the primers
(0.5 mM of each), dNTPs (0.2 mM of each), MgCl2
(3 mM), buffer and 1 U DNA polymerase Biotaq
(Bioline) from Thermus aquaticus YT-1. The reaction conditions were: 1 cycle at 94 C for 3 min,
2025 cycles at 94 C for 1 min, 1 min at the optimal primer hybridization temperature in each case,
1 min at 72 C and, finally, a 10 min cycle at 72 C.
The ACT1 gene was used to normalize the data.
The specific oligonucleotides used for the analysis
of each gene considered are shown in the Supporting information.

Western blotting
To prepare protein extracts, 2.5 OD unit cells
were collected by centrifugation and were resuspended in 100 l water, then 100 l NaOH 0.2
M were added and the preparation was incubated
for 5 min. After centrifuging at 12 000 r.p.m.
for 1 min, the pellet was resuspended in 50 l 2
loading buffer (TrisHCl 150 mM, pH 6.8, DTT
300 mM, SDS 6%, bromophenol blue 0.3%, glycerol 30%) and incubated at 95 C for 5 min. After
cooling and centrifuging (3000 r.p.m. for 10 min),
the supernatant was collected. The total protein
concentration was determined by the Bradford
method (BioRad), and 50 g protein was used for
each analysis. After SDSPAGE electrophoresis
in 7.5% polyacrylamide, proteins were transferred
to a 0.45 m nitrocellulose membrane (Trans-Blot,
1620113, Bio-Rad) by electrotransference. Immunodetection was carried out using a GFP antibody
(Sigma). This primary antibody and the secondary
antibody (peroxidase anti-rabbit) were diluted at
1/10 000 and 1/50 000, respectively. Blocking and
incubation with the antibodies were carried out
in TBST buffer (20 mM TrisHCl, pH 8, 0.5 M
NaCl, 0.05% v/v Tween 20) with 5% w/v non-fat
dried milk. Washes were done in TBST. ECF plus
the Western Blotting detection system (RPN2132,
Amersham) were used for detection, following the
manufacturers instructions.
Yeast 2009; 26: 115.
DOI: 10.1002/yea

Rsp5p controls gene expression at the entry of stationary phase

Spi1p is involved in stress tolerance and is

expressed in the late growth phases
We have previously used the SPI1 promoter as
a biotechnological tool to increase stress tolerance by overproducing the general stress response
transcription factor Msn2p (Cardona et al., 2007).
We aim now to obtain a better understanding
of not only its role in stress resistance but also
its regulation. First, we tested the effect of the
spi1 deletion on viability after exposure to several adverse conditions. Table 2 shows the viability
of the wild-type strain and the spi1  mutant after
heat shock, ethanol stress and high and low pH.
The strain carrying the deletion is hypersensitive
to heat shock and alkaline stress, while ethanol
and acidic stress determine minor, but clear, differences between the wt and mutant strains. There is
no influence of the spi1 deletion regarding viability under other stress conditions, e.g. acetaldehyde
or osmotic stress (data not shown). Oxidative stress
was measured as the diameter of the halo of inhibition caused by hydrogen peroxide. The mutant
strain also shows increased sensitivity to oxidative insult (Table 2). To assess the influence of
Spi1p in the late phases of growth, the survival
viability of a culture of the deletion strain in YPD
was measured. The number of viable cells in the
spi1  deletion mutant compared to the wild-type
was lower during the postdiauxic and early stationary phases (Figure 1). Therefore, Spi1p is a protein
that is relevant to many stress conditions, including
starvation.
In order to study the expression of the SPI1
gene, we constructed a SPI1 promoter-controlled
version of the lacZ reporter gene on a plasmid.
To test this reporter gene, we measured the galactosidase activity along the growth curve in a
minimal medium on strain YPH499 (Figure 2A).
The results indicated that the maximal activity of

N/No

Results

160
140
120
100
80
60
40
20
0

BY4742
spi1

10

15

20 25
t (days)

30

35

40

45

Figure 1. SPI1 deletion causes decreased cell viability along

with growth. Exponentially growing cells of wt (BY4742)
and spi1 deletion mutant on YPD at OD600 = 0.3 were
taken as the starting point. Samples of cells were taken at
different times, diluted and plated on YPD plates to obtain
the cfu number (N), which is expressed as the ratio to the
initial cfu value (N0 )

this gene during growth is around 7 h after the culture started, with a dilution of an overnight culture
at OD600 = 0.5. Similar results were observed in
the rich medium YPD using microarrays (Gasch
et al., 2000), where the maximum of expression is
reached between 612 h (depending on individual
experiments) from a starting point of OD = 0.3,
and its expression remains high throughout the
stationary phase. However, in our -galactosidase
assay, the enzyme activity decreases at later stages
of growth, probably reflecting the general decrease
in translation that happens in the stationary phase
(Gray et al., 2004). Therefore, the SPI1placZ
construct is a useful tool to study the gene expression changes previous to the entry into the stationary phase.

Expression of the SPI1 gene is affected by RSP5

We used the SPI1placZ fusion to identify the
genes involved in SPI1 gene regulation. We carried out an EMS mutagenesis and isolated colonies
that were white on SD plates with low sugar concentration (0.5% glucose) and X-gal, instead of
the wild-type blue colour. One of those white

Table 2. Stress tolerance in the wild-type and spi1 mutant. The percentage of colony-forming units after the stress
condition is shown

wt
spi1

Ethanol

Heat shock

Low pH

High pH

Oxidative (mm)

22 1.5%
17 1.2%

0.88 0.001%
0.42 0.0001%

67 7%
56 2%

25 2%
10 1%

3.08 0.15
3.9 0.3

For oxidative stress, the diameter (mm) of the inhibition halo is shown. All experiments were carried out in triplicate. SD is shown.
Copyright 2009 John Wiley & Sons, Ltd.

Yeast 2009; 26: 115.

DOI: 10.1002/yea

-galactosidase (a.u)

F. Cardona, A. Aranda and M. del Olmo

35

WT
HPG1-1

30
25
20
15
10
5
0
0

12

16

20

24

time (hours)
B

m4

pTRP1

YPH499
E PD

pTAT2

HPG1-1
E PD
SPI1
rRNA

Figure 2. SPI1 expression depends on the growth phase

levels and Rsp5p. (A) The SPI1 expression measured as
-galactosidase activity, wild-type (YPH499 strain) and
the rsp5 mutant (HPG1-1) throughout its growth in
minimal medium using the SPI1 promoterlacZ fusion.
-Galactosidase activity was expressed as arbitrary units
(a.u.) according to Adams et al. (1997). The experiments
were carried out by triplicate and standard deviation (SD)
is shown. (B) -Galactosidase assays on plates with low
glucose (0.5%) in a mutant unable to express the SPI1placZ
fusion, transformed with either an empty plasmid (m4) or a
plasmid containing genes implied in tryptophan metabolism
(pTRP1) or transport (pTAT2). (C) SPI1 mRNA levels in
both wild-type and HPG1-1 mutant in either the exponential
(E) or postdiauxic (PD) phase

mutants, named m4 (Figure 2B), was transformed

with a multicopy library in order to isolate regulators of SPI1 expression. We found two kinds
of plasmids that cause a reversion to the blue
colour (Figure 2B). One contains the high-affinity
tryptophan transporter TAT2, and the other contains the TRP1 gene that codes for phosphoribosylanthranilate isomerase, which catalyses the
third step in tryptophan biosynthesis. The strain
used in the screening, W303-1a, is a trp1-1 strain.
We failed to fully complement the m4 mutation with a centromeric plasmid containing neither
the TRP1 marker nor the TAT2 gene (data not
shown). So far, we have been unable to identify
Copyright 2009 John Wiley & Sons, Ltd.

the location of mutation m4, but it seems clear that

tryptophan metabolism is apparently important for
SPI1 expression.
It is known that the activity of the tryptophan
permeases Tat1p and Tat2p is controlled by the
ubiquitin ligase Rsp5p (Abe and Iida, 2003) and
that Rsp5p also controls the stress response (Haitani et al., 2006). Therefore, we tested the role of a
mutant of the RSP5 gene called HPG1-1 (Abe and
Ida, 2003) in SPI1lacZ fusion (Figure 2A). The
mutant had lower levels of -galactosidase activity at the entry of the stationary phase, indicating a
positive role of Rsp5p in SPI1 expression at the
late phases of cell growth. To confirm that this
effect is indeed located at the transcriptional level,
a Northern blot analysis of SPI1 was carried out
using the same mutant (Figure 2C), and showed a
defect in the transcriptional induction of SPI1 in
the postdiauxic phase.

Role of the ubiquitination machinery in SPI1

expression
We next attempted to gain a better understanding
of the role that ubiquitination plays in SPI1 gene
expression, using different mutations in the genes
related to this process. RSP5 is an essential gene
and HPG1-1 is one of the mutants isolated from
this gene. In this case, it carries a mutation of
the Pro 514 to Thr in the catalytic centre of the
enzyme, which causes a semi-dominant mutation
that confers resistance to high pressures, given its
inability to ubiquitinate and to lower the levels of
tryptophan permease Tat2p (Abe and Ida, 2003).
To test whether other mutations in the RSP5
gene show the same phenotype, we transformed
one called npi1, which plays a clear role in the
degradation of the amino acid transporters Gap1p
and Fur4p (Hein et al., 1995) with the SPI1placZ
fusion. A decreased expression was also observed
along the culture curve (Figure 3A). Bul1p and
Bul2p are two functional homologous proteins that
interact with Rsp5p and are sometimes referred to
as E4 proteins (Hoppe, 2005). The deletion of both
genes also leads to a reduction of SPI1 expression
at the entry of the stationary phase (Figure 3B),
which also indicates a positive contribution of these
two proteins to SPI1 expression.
We then tested the effect of deleting other components from the ubiquitination machinery, using
a single measurement of -galactosidase activity
Yeast 2009; 26: 115.
DOI: 10.1002/yea

Rsp5p controls gene expression at the entry of stationary phase

-galactosidase (a.u.)

30

WT
npi1

25
20
15
10
5
0

12

16

20

24

12
16
time (hours)

20

24

time (hours)

-galactosidase (a.u.)

-galactosidase (a.u.)

40
35
30
25
20
15
10
5
0

WT
bul1bul2

2
1.5
1
0.5

sok2

ubp14

ubp6

ubp4

ubc7

ubc5

ubc4

ubc2

ubc1

ubi4

WT

Figure 3. The ubiquitination machinery controls the SPI1

expression. (A) -galactosidase levels of the SPI1lacZ
fusion expression of 2344c strain (WT) and 27038a npi1
mutant strain throughout its growth in SC-URA medium.
(B) Same as (A) for TB50a (WT) and JC60-4b (bul1bul2)
strains. (C) The SPI1lacZ fusion expression in several
mutants of the ubiquitination pathway after 8 h of growth
from an OD = 0.3 starting point. In this case the activity is
referred to as wt activity = 1. All experiments were carried
out in triplicate and SD is shown

in the postdiauxic phase (Figure 3C). S. cerevisiae ubiquitin is a 76-amino acid protein encoded
by four structural genes, UBI1, UBI2, UBI3 and
UBI4. While each of these genes is reported to
be expressed during yeast exponential growth, only
Copyright 2009 John Wiley & Sons, Ltd.

UBI4 is shown to be strongly inducible by stresses

such as starvation, heat and DNA-damaging agents
(Simon et al., 1999). However, UBI4 deletion has
no effect on SPI1 expression (Figure 3C). Therefore, this stress-regulated version of ubiquitin is not
essential to the regulation of the stress-responsive
gene SPI1. There are around 13 genes which
encode for the ubiquitin-conjugating (E2 or UBC)
enzymes in yeast (Hiraishi et al., 2006). Ubc4p
and Ubc5p E2s have been named stress ubiquitinconjugating enzymes because of their role in stress
resistance (Arnason and Ellison, 1994). However
their deletion has no effect on SPI1 expression. The
same result was obtained with the UBC2 /RAD6
mutant, an E2 involved in the N-end rule degradation (Dohmen et al., 1991). UBC1 deletion has
a negative effect, as it decreases lacZ activity, so it
is apparently necessary to achieve full SPI1 expression. Ubc1p acts with Rsp5p to signal Gal2p for an
effective internalization by endocytosis and subsequent proteolysis in the vacuole (Horak and Wolf,
2001). Ubc7p physically and genetically interacts
with Rsp5p to achieve proper chromatin formation,
and it has been proposed that it is a bona fide E2
for Rsp5p (Arnason et al., 2005). UBC7 deletion
shows an increase of SPI1 expression (Figure 3C),
which either indicates that it may compete with
other E2 enzymes, such as Ubc1p for Rsp5p, or acts
on an opposite pathway, maybe at the chromatin
level, causing a more relaxed chromatin on the
SPI1 promoter. Therefore, different E2 enzymes
could regulate Rsp5p activity in order to properly
control its function as a gene expression regulator.
On the other hand, there are 17 ubiquitin-specific
protease genes (UBP ) in yeast. Ubp4p (Doa4p),
Ubp6p and Ubp14p are involved in the degradation of Tat2p mediated by Rsp5p (Miura and Abe,
2004). We tested the effect of the deletion of these
genes in SPI1 expression (Figure 3C). The deletion of UBP4 did not decrease the SPI1placZ
levels and even had an slightly positive effect.
Nonetheless, the deletion of both UBP6 and
UBP14 dramatically decreased the SPI1lacZ levels (Figure 3C), indicating that these proteins play
an important role in this particular event of the
gene expression, by triggering the degradation of
an unknown factor that probably acts negatively
on SPI1 expression. In conclusion, the ubiquitination machinery shows a complex pattern in terms
of the SPI1 expression during starvation, with both
positive and negative regulatory effects.
Yeast 2009; 26: 115.
DOI: 10.1002/yea

F. Cardona, A. Aranda and M. del Olmo

Transcriptomic analysis of the rsp5 mutant

indicates an induction of the genes involved in
protein biosynthesis

Table 3. Genes with a differential expression in HPG1-1

mutant vs. the wild-type strain

In order to understand the role of Rsp5p in yeast

gene expression at the beginning of the stationary
phase, a transcriptomic analysis was carried out
with the mutant HPG1-1 and the corresponding
wild-type (YPH499) strain by microarrays. For this
purpose samples were obtained from both strains at
the entry of the stationary phase after 8 h growth
in YPD from a starting point of OD600 = 0.3.
It is worth mentioning that no differences in the
growth phase of the culture were found under the
conditions used (data not shown).
In the HPG1-1 mutant, the expression of 88 nondubious genes was induced by a factor of 3, 118
genes by a factor of 2 and 255 genes by a factor of 1.5. Among the repressed genes, 14 were
repressed by a factor of 3, 54 by a factor of 2
and 192 by 1.5. We used the Funcassociate tool
to detect the functional categories overrepresented
among the genes that had a differential expression
of 2 or more. Among the genes induced (Table 3A),
most belonged to the overlapping categories related
to translation (97/118 genes; p value = 5.6e76),
particularly to the cytosolic ribosome (96/118;
p value = 3.6e144). Some of these proteins were
also involved in other events, such as ribosome
assembly (16 genes) and telomere maintenance (18
genes). Of the 22 genes not related to the ribosome,
the most significant category was the cell wall
components (6/22; p value = 1.5e6). The Yeastract
tool, which groups genes by the transcriptional factor regulating them (Teixeira et al. 2006), showed
that 83.9% of these genes were controlled by the
Rap1p transcription factor, while 80.5% were controlled by Ifh1p, two well-known regulators of ribosome biogenesis (Wade et al., 2004).
The categories among the genes repressed in
the HPG1-1 mutant are more diverse but are all
related to metabolic processes, such as carboxylic
acid metabolism (13/54; p value = 1.16e-06), the
amino acid metabolic process (10 genes; p value =
5e-06) and proline transporters (3 genes; p value =
2.2e-06) (Table 3B). Using the Yeastract tool,
61.1% of the genes were controlled by the cell
cycle regulator Sok2p (although the SOK2 mRNA
itself is not changed in the HPG1 mutant). SPI1
was also downregulated in the microarray analysis, together with a few stress-response genes. To

(A) Induced

Copyright 2009 John Wiley & Sons, Ltd.

Categories

Genes

Translation

RPL9A (8.19), RPS5 (7.42), RPS22 (7.35),

RPL32 (6.82), RPL24A (6.71), RPS15 (6.56),
RPL8B (6.46), RPS4B (6,19), RPS0A (6.15),
RPS9B (6.13), RPS9A (6.12), RPS24B (6.12),
RPS1B (5.9), RPL22A (5.84), RPS24A (5.77),
RPL8A (5.75), RPP2A (5.7), RPL37A (5.65),
RPS4A (5.63), RPS16B (5.52), RPL6A (5.45),
RPS17B (5.4), RPL25 (5.34), RPP0 (5.34),
RPS17A (5.33), RPL9B (5.32), RPL12A (5.27),
RPL26B (5.24), RPS10A (5.23), RPL6B (5.23),
RPL42A (5.19), RPL13B (5.11), RPS1B (5.11),
RPS26B (5.06), RPL20A (5.02), RPL19A
(4.98), RPL5 (4.95), RPS6A (4.94), RPS31
(4,90), RPL12B (4.79), RPL21A (4.76), RPS1A
(4.76), RPL21B (4.75), RPL42B (4.72), RPL2A
(4.70), RPS13 (4.63), RPL20B (4.62), RPL34A
(4.54), RPS14A (4.53), RPL19B (4.49),
RPL43A (4.48), RPS3 (4.47), RPS2 (4.44),
RPL39 (4.43), RPS12 (4.42), RPS21B (4.33),
RPL23A (4.29), RPL27B (4.20), RPL13A
(4.17), RPL34B (4.15), RPL33A (4.15), RPL30
(4.14), RPL33B (4.06), RPS27B (4.03), RPL29
(4.03), RPS14B (3.97), RPL15A (3.81),
RPL36B (3.78), RPS29B (3.71), RPS29A
(3.67), RPL35B (3.62), RPL15B (3.56),
RPL43B (3.57), RPL37B (3.57), RPL36A
(3.55), RPL1A (3.50), RPL35A (3.48), RPS25B
(3.44), RPL11B (3.39), RPS20 (3.37), RPL38
(3.23), RPL11A (3.22), RPS25A (3.18), RPL1B
(3.13), RPL22B (3.13), RPP2B (2.91), STM1
(2.88), RPL4B (2.70), RPS30A (2.62), RPL41A
(2.55), RPS30B (2.54), RPL41B (2.47),
RPS28B (2.41), RPP1A (2.33), RPL4A (2.32),
TEF1 (2.21), RPS28A (2.12)

Cell wall

GAS3 (3.05), YLR042c (3.00), PLB2 (2.65),

SCW10 (2.33), SRL1 (2.13), MCD4 (2.00)

Other

CHA1 (19.62), RNR1 (2.74), NMR1 (2.67),

POL30 (2.57), BTN2 (2.48), MIS1 (2.34),
HXK1 (2.30), ACM1 (2.25), ARO10 (2.18),
RNH203 (2.15), SCP160 (2.10), TOS4
(2.08), SPO16 (2.08), PMI40 (2.04), GND1
(2.02)

(B) Repressed
Amino acid
metabolism

SER3 (3.11), AGX1 (3.09), MET3 (2.90),

YAT2 (2.70), LEU2 (2.57), MET17 (2.53),
IDP2 (2.11), GDH1 (2.11), MET2 (2.10),
MLS1 (2.07)

Proline transport

GAP1 (3.19), PUT4 (3.10), AGP1 (2.25)

Stress

HSP12 (11.94), HSP26 (4.11), MEI5 (2.55),

GRE1 (2.48), SPL2 (2.40), DDR2 (2.29), SPI1
(2.25), TSA2 (2.25), NCE103 (2.13)

Yeast 2009; 26: 115.

DOI: 10.1002/yea

Rsp5p controls gene expression at the entry of stationary phase

Table 3. Continued
Categories
Other

Unknown

Genes
PHO89 (4.73), FMP45 (4.36), FMP16 (4.11),
NDE2 (3.73), NQM1 (3.63), SPS100, (3.57),
PHM6 (2.95), YCT1 (2.93), GND2 (2.88),
MSC1 (2.78), AQY1 (2.71), PUR5 (2.67),
NCE102 (2.60), FRM2 (2.58), GMP2 (2.40),
LAP4 (2.30), BOP2 (2.26), PHM7 (2.20),
PDH1 (2.17), ALD4 (2.15), FLR1 (2.06),
ZRT1 (2.04), RMD6 (2.03)
YMR107w (6.93), YDL218w (5.04),
YKL071w (2.82), YHR033w (2.60),
YHR140w (2.48), YBR285w (2.32), YAL061w
(2.24), YGR154c (2.16), YDL223c (2.10)

Those genes that are induced or repressed by >2 are shown, and
their induction or repression values are shown in brackets.

HSP26, the amino acid biosynthetic genes SER3

and MET3 and the amino acid permease GAP1
were also confirmed (Figure 4).

Stress resistance in HPG1 and bul1/2 mutants

We went on to attempt to relate the expression data
to the role of RSP5 mutants and bul1/2 deletion in
stress resistance. In addition to the HPG1-1 mutant,
we used a different allele of RSP5, HPG1-4, whose
mutation is placed in a different region of the
RSP5-coding region, providing a thermosensitive
phenotype (Abe and Ida, 2003). Some cell wallrelated genes were induced in the HPG1-1 mutant
(Table 3A), a fact that may indicate an imbalance
in cell wall assembly or a response to cell wall
damage. We grew all these mutants on plates
containing 0.008% SDS to detect a cell wall defect.
Indeed, we found this to be the case, particularly
for the HPG1-1 and bul1/2 mutants (Figure 5A).
The identification of genes involved in amino
acid metabolism in our microarray analysis suggests a putative involvement of Rsp5p in the
response to nutrient limitation. To address this possibility, our mutant cells were grown in the presence of rapamycin, which inhibits the conserved

WT

HPG1-1
RPL9a

CHA1

GAS3

SPI1

HSP26

SER3

MET3

GAP1

ACT1

test whether Sok2p is a positive regulator of SPI1

expression, we transformed a sok2 deletion strain
with the SPI1placZ fusion. We found that Sok2p
is indeed necessary to fully activate the fusion
under starvation conditions (Figure 3C).
To confirm the array data, a semi-quantitative
RTPCR analysis of selected genes was carried
out, using the actin gene ACT1 as a control
(Figure 4). RPL9 was used as a representative of
the ribosomal protein cluster. It was clearly overexpressed in the mutant strain. CHA1, a gene involved
in serine catabolism, was the most overexpressed
gene in the HPG1-1 mutant, according to the array
data (19.6-fold; see Table 3). Its induction was confirmed by RTPCR. GAS3 encodes for a 1,3-glucanosyltransferase representative of the cell wall
cluster that also showed the expected transcription
pattern. Among the repressed genes, examples of
each functional family were tested, and the repressions of SPI1 itself, the stress-responsive gene

Figure 4. Gene expresson analysis by semi-quantitative

RTPCR of several genes genes that were either identified
as being induced or repressed in the HPG1-1 mutant at the
post-diauxic shift. ACT1 was used as a control

Figure 5. Effects of toxic agents on the growth of ubiquitination mutants. The HPG1-1, HPG1-4 and bul1bul2 mutants
were replica-plated on (A) SDS 0.008% or rapamycin 0.1 g/ml and (B) cycloheximide 0.1 g/ml
Copyright 2009 John Wiley & Sons, Ltd.

Yeast 2009; 26: 115.

DOI: 10.1002/yea

10

F. Cardona, A. Aranda and M. del Olmo

bul1bul2 Dcp2p-GFP

HPG1-1 Dcp2p-GFP

A WT Dcp2p-GFP

YP 15

1M KCl 15

HPG1-1 Dhh1p-GFP

WT Dhh1p-GFP

bul1bul2 Dhh1p-GFP

YP 15

1M KCl 15

bul1bul2

HPG1-1

YP / 15

WT

bul1bul2

HPG1-1

KCl 1M / 15

WT

Dcp2p-GFP
Ponceau
Dhh1p-GFP
Ponceau

Figure 6. The RSP5 mutant is defective in p-body formation. Microscopy after 15 min without glucose (YP) or with osmotic
stress (1 M KCl) in WT, HPG1-1 and bul1bul2 strains transformed with (A) Dcp2pGFP fusion and (B) Dhh1pGFP
fusion. (C) Western blot of the Dcp2pGFP and Dhh1pGFP protein levels of these strains under the same stress
conditions, using an anti-GFP antibody

Copyright 2009 John Wiley & Sons, Ltd.

Yeast 2009; 26: 115.

DOI: 10.1002/yea

Rsp5p controls gene expression at the entry of stationary phase

protein kinase that links nutrient status and cell proliferation, target of rapamycin (TOR). Our mutations were hypersensitive to rapamycin, indicating
that ubiquitination machinery contributes to the
response to nutrient starvation.
A relationship between Rsp5p and translation
accuracy has been recently reported (Kwapisz
et al., 2006), according to which a S. cerevisiae
strain carrying an rsp5-13 mutation shows altered
sensitivity to antibiotics and a slower rate of translation. The results obtained in the microarray and
in the semi-quantitative RTPCR analysis indicate that there is a higher level of transcripts
related to protein biosynthesis in the mutant HPG11 under stationary phase conditions. In order to
confirm this effect on the translation of the rsp5
mutant alleles that we used and the role of the
BUL1/2 role in translation, experiments of resistance to the translation inhibitor cycloheximide
were carried out (Figure 5B). The HPG1-1 allele
presented a significant defect on translation. However, the HPG1-4 allele did not have such a significant growth defect. Therefore, the cycloheximide sensitivity showed some allele specificity.
The bul1/2 deletion also revealed an increased
cycloheximide sensitivity. These data confirm the
role of Rsp5pBul1pBul2p in translation and
indicate that this role may be related to the alteration of the translation machinery expression.

RSP5 is involved in p-body formation under

stress conditions
According to our microarray data an increased
expression of the ribosomal proteins was found
in the rsp5 mutant. To a lower extent, there was
also an increase in the levels of several factors
involved in translation elongation, such as the EF1 component TEF1 (2.2-fold induction), EF-1
component EFB1 (1.87-fold), EF-2 components
EFT2 (1.96-fold) and EFT1 (1.62), and the EF-3
component YEF3 (1.77-fold). In higher eukaryotes,
ribosomal and translational elongation factors are
co-regulated at the translation initiation level in
a TOR-dependent manner during nutrient deprivation, given the presence of a 5 -terminal oligopyrimidine tract (TOP) at their 5 -UTR (Hamilton
et al., 2006). This regulation system is not present
in S. cerevisiae, but the co-regulation of these genes
in the HPG1-1 mutant suggests that a common regulation mechanism may be involved in the expression of this family of functionally related genes.
Copyright 2009 John Wiley & Sons, Ltd.

11

Rsp5p plays a role in gene expression at the posttranscriptional level (see Introduction). Therefore
we tested the effect of our ubiquitination mutants in
another post-transcriptional event related to stress
and nutrient starvation, that of the formation of processing bodies (p-bodies).
To analyse this event, we fused the green fluorescent protein GFP to two proteins implicated
in mRNA degradation that are well established pbody markers, Dcp2p and Dhh1p in the HPG1-1
and the bul1/bul2 mutants. Under glucose deprivation stress (15 min in YP) and osmotic shock
(15 min in 1 M KCl), the p-bodies were clearly
formed in the wild-type cells, both when Dcp2p
(Figure 6A) and Dhh1p (Figure 6B) were fused to
GFP. In the RSP5 mutant tested, HPG1-1, the number and intensity of the p-bodies decreased for both
GFP fusions (Figure 6) in both conditions, particularly in the starvation condition. This effect was not
seen in the bul1/bul2 mutant, where there is still
a high number of cells containing p-bodies after
the stress. Therefore, the Rsp5p has an important
role in the formation of p-bodies under different
stress conditions without the help of Bul1/2p in
this particular event.
In order to rule out a defect in the fusion protein amount, we carried out a Western blot analysis using an anti-GFP antibody (Figure 6C). The
levels of GFP-fused Dcp2p and Dhh1p proteins
were similar in the HPG1-1 mutant compared to
the wild-type. Therefore, Rsp5 seems to affect the
assembly of the p-bodies during these environmental changes, not their protein levels.

Discussion
In this work, we have used the SPI1 gene as a
marker of the entry into the stationary phase. Spi1p
is a GPI-anchored cell wall protein that plays a
positive role in stress response. Its deletion caused
an increased sensitivity to various insults, such as
heat shock, high ethanol concentration, extreme pH
values, oxidative stress and starvation (Table 2).
As a typical stress-regulated gene, SPI1 expression is induced under many adverse conditions
(Gasch et al., 2000) and is highly expressed in the
late growth phases under both laboratory (Gasch
et al., 2000) and industrial conditions (Puig and
Perez-Ortn, 2000). We have constructed a fusion
of the SPI1 promoter to lacZ to identify new
Yeast 2009; 26: 115.
DOI: 10.1002/yea

12

regulators of its nutrient-controlled expression. It

is known that increasing tryptophan concentration
or overexpressing tryptophan transporters in tryptophan auxotrophs increases tolerance to weak acids
(Bauer et al., 2003) and to ethanol (Hirasawa et al.,
2007). We found that multicopy plasmids containing the tryptophan biosynthetic gene TRP1 and the
high-affinity tryptophan transporter TAT2 suppress
a mutation that causes low SPI1 expression. Therefore, the effects of tryptophan metabolism on stress
are also applied to SPI1 gene expression. Similarly,
we have found that TAT2 overexpression regulates
a similar fusion of another starvation-induced gene,
YGP1 (data not shown).
A known link between stress resistance and tryptophan transport is the ubiquitin ligase Rsp5p. Its
deletion leads to sensitivities to various environmental challenges and it plays a role in Tat2p
degradation under starvation. We used a strain containing the HPG1-1 allele of RSP5 to prove the
role of this gene in SPI1 expression (Figure 2).
This mutant was isolated by Abe and Ida (2003)
as a semi-dominant mutant that allows growth at
high pressures, due to its incapacity to degrade
Tat2p. We have further studied the contribution of
the ubiquitination machinery to SPI1 expression
(Figure 3). The analysis of the double mutant of
the Rsp5p interacting proteins Bul1p and Bul2p
have established a clear function of this complex
in SPI1 gene activation at entry to the stationary phase. Of the potential E2 of Rsp5p tested,
we found that Ubc1p and Ubc7p have a positive
and negative effect, respectively, on the control
of the SPI1 expression. Therefore, it seems that
Ubc1p is necessary for the ubiquitination reaction that leads to SPI1 activation, and that Ubc7
competes with it, or activates an antagonistic pathway. Ubp6 and Ubp14 may contribute to degrading whatever target may lead to the SPI1 induction, because the deletion of both proteins leads to
a defect in SPI1 expression. Therefore, it seems
that Rsp5pBul1/2p/Ubc1/Ubp6/14p may degrade
a factor that negatively affects the SPI1 gene induction at the entry to stationary phase. There is no
evidence in the global analysis of Rsp5p-mediated
ubiquitination of Spi1p as a direct target of Rsp5p
activity (Kus et al., 2005).
We carried out a global analysis to further study
the role of Rsp5p in gene expression at the onset
of the stationary phase. A small number of genes
show lower levels in the HPG1-1 mutant, together
Copyright 2009 John Wiley & Sons, Ltd.

F. Cardona, A. Aranda and M. del Olmo

with SPI1. The presence of the genes involved

in proline transport and amino acid metabolism is
significant, which provides a clue that links Rsp5p
to the response to the nutritional status. Sok2p
transcription factor controls 61% of the genes
repressed in the mutant HPG1-1 (Teixeira et al.,
2006). We confirmed that the SPI1lacZ fusion
activation is indeed Sok2p-dependent. SOK2 acts
downstream of a main nutrient control pathway,
PKA, to regulate the expression of those genes
that are important for growth and development
(Ward et al., 1995). Rsp5p may somehow regulate
Sok2p activity when the stationary phase begins.
There is ongoing controversy as to whether Rsp5p
controls the activity of the transcription factor
Gln3p, involved in nitrogen catabolite repression
(Crespo et al., 2004; Feller et al., 2006). Our
arrays do not suggest a global control of the
nitrogen catabolite repression regulon by Rsp5p
under the conditions tested. It is intriguing that
the gene which displays the greatest induction in
the HPG1-1 mutant (19.6 fold) is CHA1 (Table 3,
Figure 4), a gene involved in serine catabolism,
while the genes in the serine biosynthetic pathway,
such as SER3 (3.1-fold repression), AGX1 (3.1fold), MET17 (2.5-fold) and SER33 (1.9-fold)
are repressed. This indicates a particular role of
Rsp5p in the control of the serine and threonine
metabolism by a mechanism that is unknown to
date. The antagonistic regulation of CHA1 and
SER3 depends on the Cha4p transcription factor
that can act as both an activator and repressor
in response to serine (Martens et al., 2005). It
is temping to suggest that Rsp5p may somehow
regulate serine metabolism via this transcription
factor or their interacting proteins, although an
indirect effect due to the control of amino acid
permease by ubiquitination cannot be ruled out.
A larger number of genes are repressed by Rsp5p
at the entry of the stationary phase. They overwhelmingly belong to the category of ribosomal
proteins (RP), along with some translation elongation factors (see Table 3). Rsp5p is involved in
translation accuracy and mutations on it cause sensitivity to cycloheximide (Zwapisz et al., 2002).
Cycloheximide is an antibiotic that targets ribosomal proteins, which leads to a defect on translation. The rsp5 allele we were using, HPG1-1, is
also cycloheximide-sensitive (Figure 5B), despite
the larger amount of ribosomal protein expression,
which either suggests additional effects at the postYeast 2009; 26: 115.
DOI: 10.1002/yea

Rsp5p controls gene expression at the entry of stationary phase

transcriptional level or an unbalanced situation that

leads to a translation deficiency. The translation
accuracy defect detected in the rsp5-13 allele is
partially suppressed by an additional copy of the
translation elongation factor eIF1A TEF2 (Zwapisz
et al., 2002). Our rsp5 mutation, HPG1-1, activates
the levels of the other coding gene for eIF1A, TEF1
(Table 3), and also other transcription elongation
factors, thus reinforcing the role of Rsp5p in translation accuracy. Rsp5p has an antagonistic effect
on the TOR pathway at the RP expression level.
TOR stimulates RP transcription when nutrients are
abundant (Powers, 2004), while Rsp5p represses
RP levels when nutrients become scarce. A similar antagonistic role between Rsp5p and TOR
regarding amino acid transporter levels has been
proposed. For instance, under starvation and upon
Rsp5p/Bul11/2p-dependent ubiquitination, amino
acid transporters are internalized, transported to the
vacuole and degraded by vacuolar hydrolases (Abe
and Ida, 2003). TOR inhibits this turnover when
nutrients are high in the medium via Npr1p kinase
(Schmidt et al., 1998). A similar molecular mechanism has been proposed to explain the role of
Rsp5p in the regulation of the nitrogen metabolism
controller Gln3p (Crespo et al., 2004). TOR activates Npr1p kinase when nutrients are high, and in
turn inhibits Rsp5p activity. Under nitrogen limitation, Rsp5p would be active and would stimulate
the Gln3p shift to the nucleus. However, this effect
seems to be ammonia-specific (Tate et al., 2006).
The HPG1-1 mutant is rapamycin-sensitive, which
indicates that Rsp5p works in the same pathway as
TOR, even though they have opposite effects on
the RP levels.
Outside the RP genes, the transcriptomic effect
of rsp5 mutation does not match that caused by
either TOR deletion or rapamycin treatment under
different conditions (Hardwick et al., 1999; Shamji
et al., 2000; Chen and Powers, 2006). This suggests
that the Rsp5p mechanism may be either different
or located at another level of the gene expression
other than transcription. We have found a role of
the ubiquitination machinery on p-body formation
for the very first time. Under several stresses,
such as sudden glucose deprivation, proteinRNA
aggregates (p-bodies) are formed. It is known that
the mRNA can exit p-bodies and resume translation
when glucose is added back. The mutation of
Rsp5p results in a decrease of p-body formation
under sudden starvation conditions (see Figure 6).
Copyright 2009 John Wiley & Sons, Ltd.

13

It has been postulated that the mRNAs in these

aggregates formed during the stationary phase are
more resistant to extraction (Aragon et al., 2006).
We postulate that the increase in the mRNA levels
that we observed in the transcriptomic analysis of
the HPG1-1 mutants may be related to the fact
that a deficiency in p-body formation increases
the pool of protein-free mRNA, which is easily
extracted by the usual phenol RNA purification.
This would imply that Rsp5p has a specific ability
to sequestrate the mRNAs of RP proteins on pbodies under starvation conditions. A potential link
between ubiquitination and p-body formation may
be the fact that a global screening of Rsp5p targets
has identified the p-body component Dhh1p as a
target of this ubiquitin ligase (the 34th place in the
most likely targets; Kus et al., 2005). As a matter
of fact, we have observed a slower growth when
Dhh1p is fused to GFP in the HPG1-1 background
(data not shown), an observation that may indicate
a functional relationship between the two proteins.
Additional global studies using proteases to release
protein-bound mRNAs would be necessary to fully
assess this hypothesis.

Acknowledgements
We are indebted to F. Estruch for supplying us with the
GFP antibody and J. C. Igual for the multicopy library.
We also thank F. Abe, B. Andre and J. Heitman for the
strains provided. This work was supported by Grant No.
AGL2005-00508 from the Ministerio de Educacion y Ciencia and Grant No. GRUPOS03/012 from the Generalitat
Valenciana to M.O. and A.A. F.C. was an FPU Fellow of
the Ministerio de Educacion y Ciencia.

Supporting Information
Supporting information may be found in the online
version of this article.

References
Abe F, Iida H. 2003. Pressure-induced differential regulation of
the two tryptophan permeases Tat1 and Tat2 by ubiquitin ligase
Rsp5 and its binding proteins, Bul1 and Bul2. Mol Cell Biol 23:
75667584.
Adams A, Gottschlings DE, Kaiser CA, Stearns T. 1997. Methods
in Yeast Genetics. Cold Spring Harbor Laboratory Press: Cold
Spring harbor, NY.
Aragon AD, Quinones GA, Thomas EV, et al. 2006. Release of
extraction-resistant mRNA in stationary phase Saccharomyces

Yeast 2009; 26: 115.

DOI: 10.1002/yea

14

cerevisiae produces a massive increase in transcript abundance

in response to stress. Genome Biol 7: R9.
Arnason T, Ellison MJ. 1994. Stress resistance in Saccharomyces
cerevisiae is strongly correlated with assembly of a novel type
of multiubiquitin chain. Mol Cell Biol 14: 78767883.
Arnason TG, Pisclevich MG, Dash MD, et al. 2005. Novel
interaction between Apc5p and Rsp5p in an intracellular
signalling pathway in Saccharomyces cerevisiae. Eukaryot Cell
4: 134146.
Bauer BE, Rossington D, Mollapour M, et al. 2003. Weak organic
acid stress inhibits aromatic amino acid uptake by yeast,
causing a strong influence of amino acid auxotrophies on the
phenotypes of membrane transporter mutants. Eur J Biochem
270: 31893195.
Brengues M, Teixeira D, Parker R. 2005. Movement of eukaryotic
mRNAs between polysomes and cytoplasmic processing bodies.
Science 310: 486489.
Cardona F, Carrasco P, Perez-Ortn JE, et al. 2007. A novel
approach for the improvement of stress resistance in wine yeasts.
Int J Food Microbiol 114: 8391.
Carrasco P, Perez-Ortn JE, del Olmo M. 2003. Arginase activity
is a useful marker of nitrogen limitation during alcoholic
fermentations. System Appl Microbiol 26: 471479.
Chen JCY, Powers T. 2006. Coordinate regulation of multiple and
distinct biosynthetic pathways by TOR and PKA kinases in S.
cerevisiae. Curr Genet 49: 281293.
Crespo JL, Helliwell SB, Wiederkehr C, et al. 2004. NPR1 kinase
and RSP5BUL1 /2 ubiquitin ligase control GLN3-dependent
transcription in Saccharomyces cerevisiae. J Biol Chem 279:
3751237517.
DeRisi JL, Iyer VR, Brown PO. 1997. Exploring the metabolic
and genetic control of gene expression on a genomic scale.
Science 278: 680686.
Dohmen RJ, Madura K, Bartel B, Varshavsky A. 1991. The N-end
rule is mediated by the UBC2 (RAD6 ) ubiquitin-conjugating
enzyme. Proc Natl Acad Sci USA 88: 73517355.
Fazzio TG, Kooperberg C, Goldmark JP, et al. 2001. Widespread
collaboration of Isw2 and Sin3-Rpd3 chromatin remodeling
complexes in transcriptional repression. Mol Cell Biol 21:
64506460.
Feller A, Boeckstaens M, Marini AM, et al. 2006. Transduction
of the nitrogen signal activating Gln3-mediated transcription is
independent of Npr1 kinase and Rsp5-Bul1/2 ubiquitin ligase in
Saccharomyces cerevisiae. J Biol Chem 281: 2854628554.
Gasch AP, Spellman PT, Kao CM, et al. 2000. Genomic
expression programs in the response of yeast cells to
environmental changes. Mol Biol Cell 11: 42414257.
Gray JV, Petsko GA, Johnston GC, et al. 2004. Sleeping beauty:
quiescence in Saccharomyces cerevisiae. Microbiol Mol Biol Rev
68: 187206.
Gwizdek C, Hobeika M, Kus B, et al. 2005. The mRNA nuclear
export factor Hpr1 is regulated by Rsp5-mediated ubiquitylation.
J Biol Chem 280: 1340113405.
Haitani Y, Shimoi H, Takagi H. 2006. Rsp5 regulates expression
of stress proteins via post-translational modification of Hsf1 and
Msn4 in Saccharomyces cerevisiae. FEBS Lett 580: 34333438.
Haitani Y, Takagi H. 2008. Rsp5 is required for the nuclear export
of mRNA of HSF1 and MSN2 /4 under stress conditions in
Saccharomyces cerevisiae. Genes Cells 13: 105116.
Hardwick JS, Kuruvilla FG, Tong JK, et al. 1999. Rapamycinmodulated transcription defines the subset of nutrient-sensitive
Copyright 2009 John Wiley & Sons, Ltd.

F. Cardona, A. Aranda and M. del Olmo

signaling pathways directly controlled by the Tor proteins. Proc

Natl Acad Sci USA 96: 1486614870.
Hein C, Springael JY, Volland C, et al. 1995. NPl1, an essential
yeast gene involved in induced degradation of Gap1 and Fur4
permeases, encodes the Rsp5 ubiquitinprotein ligase. Mol
Microbiol 18: 7787.
Herman PK. 2002. Stationary phase in yeast. Curr Opin Microbiol
5: 692607.
Hicke L. 2001. Protein regulation by monoubiquitin. Nat Rev Mol
Cell Biol 2: 195201.
Hiraishi H, Mochizuki M, Takagi H. 2006. Enhancement of stress
tolerance in Saccharomyces cerevisiae by overexpression of
ubiquitin ligase Rsp5 and ubiquitin-conjugating enzymes. Biosci
Biotechnol Biochem 70: 27622765.
Hirasawa T, Yoshikawa K, Nakakura Y, et al. 2007. Identification
of target genes conferring ethanol stress tolerance to
Saccharomyces cerevisiae based on DNA microarray data
analysis. J Biotechnol 131: 3444.
Hochstrasser M. 1996. Ubiquitin-dependent protein degradation.
Annu Rev Genet 30: 405439.
Hohmann S, Mager WH. 2003. Yeast Stress Responses. SpringerVerlag: Berlin.
Horak J. 2003. The role of ubiquitin in down-regulation and
intracellular sorting of membrane proteins: insights from yeast.
Biochim Biophys Acta 1614: 13955.
Horak J, Wolf DH. 2001. Glucose-induced monoubiquitination of
the Saccharomyces cerevisiae galactose transporter is sufficient
to signal its internalisation. J Bacteriol 183: 30833088.
Hoppe T. 2005. Multiubiquitylation by E4 enzymes: one size
doesnt fit all. Trends Biochem Sci 30: 183187.
Hoppe T, Matuschewski K, Rape M, et al. 2000. Activation
of a membrane-bound transcription factor by regulated
ubiquitin/proteasome-dependent processing. Cell 102: 577586.
Horie T, Isono K. 2001. Cooperative functions of the
mannoprotein-encoding genes in the biogenesis and maintenance of the cell wall in Saccharomyces cerevisiae. Yeast 18:
1493503.
Jimenez-Mart E, Aranda A, Mendes-Ferreira A, et al. 2007. The
nature of nitrogen source added to nitrogen depleted vinifications
conducted by a Saccharomyces cerevisiae strain in synthetic
medium affects gene expression and the levels of several volatile
compounds. Antonie Van Leeuwenhoek 92(61): 75.
Jorgensen P, Rupes I, Sharom JR, et al. 2004. A dynamic
transcriptional network communicates growth potential to
ribosome synthesis and critical cell size. Genes Dev 18:
24912505.
Kaida D, Toh-e A, Kikuchi Y. 2003. Rsp5Bul1/2 complex is
necessary for the HSE-mediated gene expression in budding
yeast. Biochem Biophys Res Commun 306: 10371041.
Kaminska J, Kwapisz M, Grabilnska K, et al. 2005. Rsp5p
ubiquitin ligase affects isoprenoid pathway and cell wall
organization in S. cerevisiae. Acta Biochim Pol 52: 207220.
Kus B, Gajadhar A, Stanger K, et al. 2005. A high throughput
screen to identify substrates for the ubiquitin ligase Rsp5. J
Biol Chem 33: 2947029478.
Kwapisz M, Cholbinski P, Hopper AK, et al. 2006. Rsp5 ubiquitin
ligase modulates translation accuracy in yeast Saccharomyces
cerevisiae. RNA 11: 17101718.
Longtine MS, McKenzie A III, Demarini DJ, et al. 1998.
Additional modules for versatile and economical PCR-based

Yeast 2009; 26: 115.

DOI: 10.1002/yea

Rsp5p controls gene expression at the entry of stationary phase

gene deletion and modification in Saccharomyces cerevisiae.

Yeast 14: 953961.
Martens JA, Wu PYJ, Winston F. 2005. Regulation of an
intergenic transcript controls adjacent gene transcription in
Saccharomyces cerevisiae. Genes Dev 19: 26952704.
Martin DE, Hall MN. 2005. The expanding TOR signaling
network. Curr Opin Cell Biol 17: 158166.
Max T, Sgaard M, Svejstrup JQ. 2007. Hyperphosphorylation of
the C-terminal repeat domain of RNA polymerase II facilitates
dissociation of its complex with mediator. J Biol Chem 282:
14113141120.
Miura Y, Abe F. 2004. Multiple ubiquitin-specific protease genes
are involved in degradation of yeast tryptophan permease Tat2
at high pressure. FEMS Microbiol Lett 239: 171179.
Neumann S, Petfalski E, Brugger B, et al. 2003. Formation and
nuclear export of tRNA, rRNA and mRNA is regulated by the
ubiquitin ligase Rsp5p. EMBO Rep 4: 11561162.
Parker R, Sheth U. 2007. P bodies and the control of mRNA
translation and degradation. Mol Cell 25: 635646.
Pickart CM. 2004. Back to the future with ubiquitin. Cell 116:
181190.
Powers T. 2004. Ribosome biogenesis: giant steps for a giant
problem. Cell 119: 901902.
Puig S, Perez-Ortn JE. 2000. Stress response and expression
patterns in wine fermentations of yeast genes induced at the
diauxic shift. Yeast 16: 139148.
Schmidt A, Beck T, Koller T, et al. 1998. The TOR nutrient
signalling pathway phosphorylates NPR1 and inhibits turnover
of the tryptophan permease. EMBO J 17: 69246931.
Simon JR, Treger JM, McEntee K. 1999. Multiple independent
regulatory pathways control UBI4 expression after heat shock
in Saccharomyces cerevisiae. Mol Microbiol 31: 823832.

Copyright 2009 John Wiley & Sons, Ltd.

15

Sigismund S, Polo S, Di Fiore PP. 2004. Signaling through

monoubiquitination. Curr Top Microbiol Immunol 286: 149185.
Shamji AF, Kuruvilla FG, Schreiber SL. 2000. Partitioning the
transcriptional program induced by rapamycin among the
effectors of the Tor proteins. Curr Biol 10: 15741581.
Somesh BP, Reid J, Liu WF, et al. 2005. Multiple mechanisms
confining RNA polymerase II ubiquitylation to polymerases
undergoing transcriptional arrest. Cell 121: 913923.
Stephen DW, Rivers SL, Jamieson DJ. 1995. The role of the YAP1
and YAP2 genes in the regulation of the adaptative oxidative
stress responses of Saccharomyces cerevisiae. Mol Microbiol
16: 415423.
Tate JJ, Rai R, Cooper TG. 2006. Ammonia-specific regulation of
G1n3 localization in Saccharomyces cerevisiae by protein kinase
Npr1. J Biol Chem 281: 2846028469.
Teixeira MC, Monteiro P, Jain P, et al. 2006. The YEASTRACT
database: a tool for the analysis of transcription regulatory
associations in Saccharomyces cerevisiae. Nucleic Acids Res 34:
446451.
Wade JT, Hall DB, Struhl K. 2004. The transcription factor Ifh1
is a key regulator of yeast ribosomal protein. Nature 432:
10541058.
Ward MP, Gimeno CJ, Fink GR, Garrett S. 1995. SOK2 may regulate cyclic AMP-dependent protein kinase-stimulated growth
and pseudohyphal development by repressing transcription. Mol
Cell Biol 15: 68546863.
Yang IV, Chen E, Hasseman JP, et al. 2002. Within the fold:
assessing differential expression measures and reproducibility in
microarray assays. Genome Biol 3: research 0062.10062.12.
Yashiroda H, Kaida D, Toh-e A, Kikuchi Y. 1998. The PY-motif
of Bul1 protein is essential for growth of Saccharomyces
cerevisiae under various stress conditions. Gene 225: 3946.

Yeast 2009; 26: 115.

DOI: 10.1002/yea