The hydrophobic patch of ubiquitin is required to
protect transactivator–promoter complexes from
destabilization by the proteasomal ATPases
Chase T. Archer1and Thomas Kodadek2,*
1Division of Translation Research and Department of Internal Medicine, University of Texas Southwestern Medical
Center, Dallas, TX 75390-9185 and2Departments of Chemistry and Cancer Biology, Scripps Research Institute,
Scripps Florida, 130 Scripps Way, No. 3A2, Jupiter, FL 33458, USA
Received March 6, 2009; Revised October 30, 2009; Accepted November 2, 2009
Mono-ubiquitylation of a transactivator is known
to promote transcriptional activation of certain
DNA–transactivator complex by the ATPases of the
26S proteasome. This inhibition of destabilization
depends on the arrangement of ubiquitin; a chain
of ubiquitin tetramers linked through lysine 48 did
mono-ubiquitin. This led to an investigation into
the properties of ubiquitin that may be responsible
for this difference in activity between the different
forms. We demonstrate the ubiquitin tetramers
proteasomal-mediated destabilization. In addition,
we show that the mutating the isoleucine residue
at position 44 interferes with proteasomal interac-
tion in vitro and will abolish the protective activity
hydrophobic patch of ubiquitin as required to
the proteasomal ATPases.
Many short-lived and damaged proteins are degraded by
the ubiquitin-proteasome system (UPS) (1). The UPS is
comprised of the 26S proteasome, the small protein
ubiquitin, and the protein machinery used to attach
ubiquitin to target proteins. Most proteins are targeted
for degradation by attachment of a poly-ubiquitin chain
containing multiple ubiquitin monomers linked together
through lysine 48. Poly-ubiquitin chains containing four
or more monomers of ubiquitin are efficiently recognized
by proteins in the regulatory particle (RP) of proteasome.
The RP then removes the ubiquitin chain and the six
proteasomal ATPases (Rpt 1–6) assist in translocation of
the target protein into the interior of the barrel shaped
core particle (CP) where the proteolytic active sites are
located. The CP can be capped on either end by the RP.
The proteolytic activity of the UPS is intimately
involved in RNA polymerase II transcription at many
levels. It has long been known that the UPS can negatively
regulate transcription by proteolysis of activators, thus
keeping their level too low to drive gene transcription
proteolysis has been found to have a stimulatory effect
on the transcription of many genes, for example,
through the degradation of repressor proteins such as
IkB (5). It has also been shown that proteasome-mediated
promoter-bound transcription factors are essential for
the expression of some genes, though the mechanistic
basis of this phenomenon is not clear. Finally, the
proteasome is involved in the efficient termination of
transcription and clearance of the RNAP II from sites
of DNA damage (6).
TheUPS also affects
cipitation followed by microarrays (ChIP-chip protocol)
have revealed that proteasomal proteins are associated
with DNA throughout the yeast genome (7,8). This
suggests that the proteasomal proteins played a role in
nucleic acid metabolism and, in agreement with this
view, several roles of the proteasome have been found at
different stages in transcriptional regulation. These roles
include chromatin modification (9,10) and transcriptional
elongation (11–14), both of which occur independent of
proteolytic activity. Studies of the GAL and heat shock
genes in yeast have shown that the proteasomal ATPases,
but not the 20S CP, are required for efficient elongation
in vitro and in vivo (11,12,15). It was shown that the
transactivator Gal4 binds directly to two of the Rpt
*To whom correspondence should be addressed. Tel: +1 214 648 1239; Fax: +1 214 648 4156; Email: email@example.com
Published online 25 November 2009Nucleic Acids Research, 2010, Vol. 38, No. 3789–796
? The Author(s) 2009. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/
by-nc/2.5), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
proteins (Rpt4 and Rpt6) and acts to recruit a fragment of
the proteasome that includes the six ATPases (Rpts 1–6),
Rpn1 and Rpn2, and perhaps other proteins, but excludes
the 20CP as well as the 19S RP lid sub-complex to GAL
promoters in vivo (16,17). The mechanism by which this
sub-complex of the 19S RP stimulates elongation is
unclear, but it has been speculated to be involved in the
complexes and in the partial disassembly of nucleosomes
in the pathway of the elongating polymerase.
More recently, a second non-proteolytic activity of the
proteasomal ATPase complex was discovered, which is the
complexes (18). This destabilization activity requires
physical contact between the activation domain (AD) of
the Gal4 transactivator and two of the proteasomal
proteins, Rpt4 and Rpt6, and probably involves the
unfolding of the activator by the proteasomal ATPases,
though this has not been shown conclusively (19). This
potent activity can strongly repress GAL transcription
by preventing stable association of the activator with
the promoter in vivo.
Interestingly, however, this activity is manifest only in
the context of certain Gal4 mutants, whereas the wild-type
protein is immune to this activity in vivo. Recent investi-
gations have revealed that the mutations that render Gal4
sensitive to this ‘stripping’ activity also prevent it from
domain, suggesting that mono-ubiquitylation of the
activator serves to protect it from the proteasomal
APTase complex (20). This provides a potential explana-
tion for the stimulatory effect that mono-ubiquitylation
has been shown to have on some activators [also see ref.
(20) for another possible mechanism].
In agreement with this idea, it was found that high levels
of soluble Ub blocks the destabilization of Gal4–DNA
complexes in vitro, arguing that Ub contacts with either
the activator or proteasomal ATPases down-regulate the
stripping reaction and that these interactions can be driven
in trans by high levels of Ub (19). There are several known
ubiquitin receptors present in the proteasome. Rpn10
is a known poly-ubiquitin chain receptor, but deletion of
the protein doesn’tgrossly
poly-ubiquitylated substrates in vivo (21,22). The ATPase
Rpt5 is knownto interact
tetra-ubiquitin chains using a cross-linking strategy, but
monomeric ubiquitin did not demonstrate any detectable
interaction by the same cross-linking methodology (23).
Finally, Rpn13 has recently been demonstrated to bind
to both monomeric and lysine 48-linked ubiquitin chains
(24,25). To determine the identity of the ubiquitin receptor
in the context of the Gal4 system, a novel chemical
cross-linking and label transfer strategy was used (26).
Ubiquitin was found to bind directly to Rpn1 and Rpt1
in the proteasomal ATPase complex and that these inter-
actions disrupt the AD-Rpt6/Rpt4 interactions, causing
complex and terminating the stripping reaction.
One of the interesting observations from the previous
study was that while mono-ubiquitylation added in trans
would prevent destabilization of the transactivators from
within the DNA-binding
DNA, lysine-48-linked tetra-ubiquitin did not display a
protective effect even at much higher concentrations.
This difference between monomeric and chain forms of
ubiquitin was also seen for interaction of ubiquitin to
the ATPase Rpt5 shown by Pickart and co-workers (23).
The explanation for this difference between forms of
ubiquitin was not apparent in either of the prior studies.
In the current study, we set out to determine what
surfaces of ubiquitin were important for its protective
function. Proteasomal-mediated destabilization assays
were used as a tool to isolate surfaces of ubiquitin for
versions of ubiquitin, but not versions that form higher
order packed quantanary structures, would effectively
Linkages of poly-ubiquitin chains that bury a hydro-
phobic patch, centered around isoleucine 44, do not
effectively prevent destabilization. Further, mutation of
the hydrophobic patch abrogates the ability of ubiquitin
to interact with the proteasomal subunit Rpt1. In vivo,
ubiquitin with the I44A mutation is no longer able to
fused to a mutant form of Gal4. We conclude that
the hydrophobic patch of ubiquitin is important for the
inhibition of destabilization of activator–DNA complexes.
MATERIALS AND METHODS
General methods and strains used
For chromatin immunoprecipitation (ChIP) experiments,
yeast strain Sc726 (SUG1 gal4::HIS3) was used (27).
plasmids (derived from pSB32) expressing either wild
type Gal4 or a mutant form of the protein expressed
from the native Gal4 promoter. In each case the encoded
proteins were tagged at their N-termini with three tandem
copies of the T7 epitope tag (Novagen). Genetic fusion of
ubiquitin to T7 tagged Gal4D in the pSB32 vector was
done by removing Ub from GST-Ub-Gap71-VP16 (18)
using a NcoI digest and inserting Ub into the NcoI site
at the start codon of the T7 tag.
The steady-state levels of the Ub-Gal4D fusions were
monitored using expression from a multi-copy plasmid
with the native GAL4 promoter. These proteins were
extracted from the pSB32 vector with the restriction
enzymes BamHI and EcoRI and inserted into the
YEp351 multi-copy vector. The constructs were then
transformed into Sc726. a-Galactosidase assays were
done using Sc244 (strain 21) (a gal4-2, Gal80, ura3-52,
leu2-3 112, ade1, MEL1) transformed with the pSB32
plasmids mentioned above.
26S proteasomes were purified using a FLAG affinity
tag as described (28) with modifications (11). The
hexahistidine tagged Rpt1 protein in the pET-28a vector
was kindly provided by George DeMartino (UTSW).
Hexahistidine tagged ubiquitin and CCPGCC-tagged
tagged proteins were purified using standard IMAC
790Nucleic Acids Research, 2010,Vol. 38,No. 3
procedures (Qiagen). Ubiquitin chains were purchased
from Boston Biochem.
Destabilization of activator–DNA complexes
by the proteasome
The destabilization assay has been described previously
(18) with changes (19). Destabilization in the presence of
mono-Ub or tetra-Ub chains was done by adding protein
to the reaction mix immediately before addition of the
Competition of ubiquitin/Rpt1 interaction using
ubiquitin and Rpt1 have been described previously (19).
Competition reactions using ubiquitin mutants were per-
formed by adding a 3-fold excess of purified ubiquitin to
the Rpt1 protein lysate 15min before adding the
cross-linkable form of ubiquitin and then performing
the reaction as described.
and label transfer reactionsbetween
Expression level of the ubiquitin-Gal4D proteins
A ?Gal4 strain (Sc726) was transformed with YEp351
multi-copy plasmid containing the indicated Ub-Gal4D
fusion protein. A 500ml culture of these strains was
grown to mid-log phase in complete media lacking
leucine with raffinose as the carbon source. Galactose
was added for 2h and the cells were collected by
centrifugation. A 1ml aliquot was saved to measure the
cell density at an OD of 600nm and an equal amount of
cells were resuspended in 50ml of water and 1?SDS
loading buffer. The cells were subjected to three cycles of
freeze/thaw and cell debris was spun down at 14K for
10min. The lysate was loaded on gel and subject to
SDS–PAGE and western blotting with an anti-Gal4
Chromatin immunoprecipitation (ChIP) assay
ChIP assays were performed according to the protocol
described (17). Cells were treated grown in raffinose con-
taining medium, with galactose added to 2% to induce the
GAL genes. Induction was carried out for 30min prior to
addition of formaldehyde. Immunoprecipitations were
carried out using anti-Gal4 N-terminal antibody as
previously described (17).
Quantitative PCR of precipitated chromatin was
performed using an iCycler Thermal Cycler and the IQ
SYBR Green Supermix, 2?mix containing 100mM
KCl, 40mM Tris–HCl, pH 8.4, 0.4mM each dNTP,
50 U/ml iTaq DNA polymerase, 6mM MgCl2, SYBR
Green I, 20nM fluorescein, stabilizers (Biorad, Hercules,
CA). Relative fold enrichment of specific DNA was
calculated by comparing Ctvalues derived from primers
against the GAL7 promoter from samples precipitated
with the specific antibody compared to an unspecific
control antibody relative Ct values from to the total,
unprecipitated DNA from each sample. Primers used for
analysis have been described (20). In the graphed figures
the Gal4D sample was arbitrarily set as 1.
Quantification of mRNA transcripts
Total RNA was isolated from 10ml of cells OD6000.6–0.8
after addition of galactose. Cells were centrifuged for
5min at 3000g in a Sorvall RT7 centrifuge with a
RTH-750 swing bucket rotor. Cells were washed with
PBS and centrifuged as before. Cell pellets were frozen
in liquid nitrogen and stored at ?80?C. Cell pellets were
resuspended in 400ml water and 400ml water-saturated
phenol was added and vortexed 1min. The mixture was
incubated at 65?C for 45min. The aqueous layer was
removed and extracted with water-saturated phenol
followed by chloroform. RNA was treated with RQ1
DNase (Promega corp., Madison, WI) for 1h. RNA was
(25:24:1) followedby chloroform.
precipitated by adding 40ml 3M NaOAc pH 5.3 and
1ml 95% EtOH. RNA quantity was measured by measur-
ing OD260. Total RNA of 1mg was used to make cDNA
using the Stratascript first strand cDNA synthesis kit
(Stratagene, La Jolla, CA) and oligo dT. CDNA was
measured by quantitative PCR as above using GAL1
and ACTI primers. The Ctvalue from each sample was
used to calculate the ratio of GAL1/ACT1. Replicated
from three samples were averaged and graphed. The
primers used for analysis have been described (20).
a-Galactosidase assays have been described (29). Briefly,
Sc244 with the pSB32 vectors containing Gal4 and Gal4
mutants were grown in synthetic complete lacking leucine
with raffinose as the carbon source. At an OD 600nm of
0.6 galactose was added to 2% final. After a 45min induc-
tion, the cells were pelleted and lysed by bead disruption.
Total protein of 50mg was used for each assay and the nM
min?1mg?1of p-Nitrophenyl produced was determined.
The mean and standard error of the mean was graphed
with the activity from Gal4 set as 100%.
Identification of Ub residues required for inhibition
We had previously reported an assay to monitor the ability
of the proteasomal ATPases to destabilize transactivator–
(see Supplementary Figure 1) (18). Briefly, a small,
biotinylated piece of DNA containing 5 Gal4-binding
sites (UASG) is tethered to a StreptAvidin-conjugated
bead and the binding
transactivator. After washing, the transactivator–DNA
complex is exposed to purified 26S or purified 19S
proteasome in the presence of ATP and an excess of
non-biotinylated DNA containing the UASGsites, which
is required to see the full stripping activity (18). As the
transactivator is removed
proteasome it is ‘trapped’ by the excess of non-
biotinylated DNA. Western blots are used to monitor
the amount of the transactivator remaining on the
tethered DNA in the presence of the proteasome.
sites are saturatedwith
from the DNAby the
Nucleic AcidsResearch, 2010, Vol.38,No. 3791
Using this assay, the ability of the proteasome to
destabilize transactivator–DNA complexes can be moni-
tored. Addition of monomeric ubiquitin would inhibit the
ability of the proteasome to destabilize transactivator–
DNA interactions, but addition of lysine-48-linked tetra
ubiquitin chains failed to inhibit the destabilization
activity of the proteasome (19).
A hypothesis consistent with these observations is that a
surface of Ub exposed in the monomer, but hidden in
lysine-48-linked polymers, might be critical for the protec-
tive activity. Structural studies have suggested that part of
surface of the lysine-48-linked ubiquitin is buried at the
interface of tetramer in the ‘closed’ confirmation (30,31).
This ‘closed’ confirmation might result in the necessary
buriedand prevent the
destabilization. Thus the lysine-48-linked tetra ubiquitin
would not inhibit destabilization when added to the
assay in trans. Indeed, the buried surface patch of
ubiquitin in the lysine 48-linked chains includes the
‘hydrophobic patch’ of ubiquitin, implicated as important
in many interactions and functions of ubiquitin (32,33).
The failure of the small-ubiquitin modifier, SUMO, to
inhibit destabilization also supports this hypothesis.
Although SUMO is structurally similar to ubiquitin,
it lacks a clear hydrophobic patch that is present on
ubiquitin monomers (34,35).
To test the importance of the hydrophobic patch,
lysine-63-linked tetra ubiquitin was included in the
assay. Ubiquitin chains that contain the lysine-63-linkage
display a more open structure and the hydrophobic patch
is solvent exposed and not buried (36). This allowed us to
use the biochemical assay as a tool to quickly isolate a
region of the ubiquitin that is required for inhibition of
destabilization. The lysine-63-linked ubiquitin was just as
effective at preventing destabilization of the Gal4-VP16
protein as monomeric ubiquitin (Figure 1A). As seen
previously, the lysine 48-linked ubiquitin had no activity
in preventing proteasomal-mediated destabilization of
transactivator–DNA complexes. This result was also
true when the assay was repeated with the mCla version
of Gal4 (Figure 1B). The mCla Gal4 protein contains both
the native DNA binding domain and AD of Gal4 but with
the middle region removed to make a protein more soluble
for in vitro studies. The mCla Gal4 protein induces
transcription of the GAL genes and responds to the
all of the endogenous signals in the same manner as
full-length Gal4 (37). The ability of mono-ubiquitin to
prevent destabilization of activator–DNA complexes
depends on a solvent exposed hydrophobic patch.
Mutation of the hydrophobic patch of ubiquitin reduces
tetra-ubiquitin to inhibit destabilization but the inability
of SUMO or lysine-48-linked tetra-ubiquitin suggests
that the hydrophobic patch of ubiquitin is required.
Mono-ubiquitin is known to interact with the Rpt1 and
Rpn1 subunits in a manner that prevents the interaction of
Gal4 with the Rpt4/Rpt6 subunits (19). These mutually
exclusive proteasomal interactions are thought to be the
reason behind the ability of mono-ubiquitin to inhibit
proteasome interaction is inhibited by mutation of the
hydrophobic patch this would strengthen the claim that
destabilization of transactivators by the proteasome.
To test this idea, a point mutation was made to disrupt
the hydrophobic patch of ubiquitin at isoleucine-44. This
mutation (or the control mutation, D58A) allowed for
a competition experiment to test if the mutated form of
ubiquitin would affect the ubiquitin/Rpt1 interaction.
Previously, our lab had described a novel cross-linking
and label transfer assay (26). This assay relies on a small
tag on the protein of interest (in this case ubiquitin), which
serves as a docking point for the cross-linking reagent
containing a biotin tag. After cross-linking and label
abilityof mono-ubiquitinor lysine-63-linked
Figure 1. Ability of ubiquitin forms to inhibit proteasomal-mediated
destabilization. (A) Destabilization of transactivator–DNA complexes
by 26S proteasome in the presence of ubiquitin forms. Controls without
(?) or with (+) proteasome are indicated. Reactions performed in the
in the presence of 10mM mono-Ub (lane 4), 10mM K48 linked Ub4
(lane 5), or 10mM K63 linked Ub4 (lane 6) are also indicated.
The average amount and SEM of Gal4-VP16 retained by DNA was
measured in three experiments by western blot and normalized with
total protein input (lane 1) being set as 100%. (B) Same as (A), but
the Gal4-VP16 protein was replaced with mCla Gal4 protein.
792 Nucleic Acids Research, 2010,Vol. 38,No. 3
transfer, only direct-binding partners of the protein of
interest become biotinylated. This assay was used to dem-
onstrate that mono-ubiquitin interacted with the Rpt1 and
Rpn1 subunits in the context of the 26S proteasome and
was also able to interact with the bacterially expressed
Rpt1 subunit (19).
To test the mutated ubiquitin forms, the cross-linkable
form of ubiquitin (CCPGCC-Ub) was mixed with bacte-
rial lysate containing a hexahistidine tagged Rpt1 protein.
After cross-linking and label transfer the samples were
subject to IMAC to enrich for the Rpt1 protein and
biotinylation was probed by blotting (Figure 2). Without
addition of any competitor, the biotinylation of the
Rpt1 protein was easily detected (Figure 2A, No Comp,
graphed in Figure 2B) demonstrating the interaction
detected. Not surprisingly, addition of an excess of
wild-type ubiquitin reduced the biotinylation of Rpt1 to
almost undetectable levels. This effect was also seen when
the competition cross-linking experiment was performed
with Ub D58A. Both wild-type and the D58A versions of
ubiquitin are able to out compete a tagged version of
ubiquitin when added in excess. However, competition
using Ub I44A did not decrease the interaction between
biotinylated. Analysis of multiple assays demonstrated
that adding UB I44A was the same as having no compet-
itor form of ubiquitin (Figure 2B). Ub I44A does not
prevent the interaction of Rpt1 and wild-type, tagged
ubiquitin most likely because its ability to interact with
Rpt1 is disrupted by the mutation. This demonstrates
the importance of the hydrophobic path of ubiquitin for
Rpt1 interaction and further demonstrates the importance
of this patchfor ability
and Rpt1, asRpt1 was strongly
of ubiquitinto inhibit
Mutation of I44 of ubiquitin abolishes the protective effect
The biochemical assays above demonstrate the impor-
tance of the hydrophobic path for interaction with
the proteasomal subunit Rpt1 and in inhibition of
proteasomal-mediated destabilization of transactivators.
These biochemical assays served as a useful tool to impli-
cate this region of ubiquitin. Further confirmation of
the functional effects of these mutations requires in vivo
experiments. To test the role of the hydrophobic patch of
ubiquitin in a cellular environment, the same two point
mutations were made in ubiquitin in the context of the
ubiquitin–Gal4D fusion proteins. The Gal4D version of
Gal4 is missing a small portion of the AD. The Gal4D
protein is not effectively mono-ubiquitylated and this
leads to a defect in DNA occupancy due to susceptibility
to proteasomal-mediated destabilization (20). However,
the portion of the AD required for transcriptional
activation remains. This partial truncation of Gal4
provides a useful system of to study the effect of
mono-ubiquitylation. Fusion of ubiquitin to Gal4D was
previously shown to partially restore activity and DNA
occupancy of the Gal4D protein. Would the hydrophobic
patch mutation prevent the rescue of Gal4D by ubiquitin?
DNA occupancy of the fusion protein was checked by
ChIP using antibodies directed against the N-terminus of
the Gal4 protein. Occupancy of the fusion protein on
the promoter of the Gal4 responsive genes GAL1-10
(Supplementary Figure S2) and GAL7 (Figure 3A) were
measured. On both promoters, the isoleucine-44 to alanine
mutation resulted in a loss of the ability of ubiquitin to
promote occupancy of Gal4D, as measured by qPCR fol-
lowing ChIP. Mutation of another region on the surface
of ubiquitin (D58A) did not cause a decrease in occupancy
and resultedin slightly
hydrophobic patch of ubiquitin is required for the rescue
of DNA occupancy of a partial truncation of Gal4.
The expression level of Gal4D is known to alter the
activity of the construct. Massive over expression in the
protein will result in a partial recovery of the activity of
Gal4D presumably due to forced occupancy of the
promoter driven by the over expression (38). To be
certain that the changes seen in DNA occupancy were
not due to expression level, the steady-state level of
the ubiqutin–Gal4D fusions were compared to Gal4D
(Figure 3B). There was little difference in protein levels
seen in the cellular lysate of the different constructs.
Thus, the difference in occupancy measured in panel A
Figure 2. Mutation of the hydrophobic patch of ubiquitin reduces
Rpt1 interaction. (A) Biotinylated Rpt1 product (NA-HRP) present
after cross-linking and label transfer between CCPGCC-ubiquitin and
Rpt1 is shown. Competitor ubiquitin is indicated above each lane.
Total His6-Rpt1 in the assay is shown by the a-His blot. (B) The
average ratio and SEM of three experiments of cross-linked Rpt1 to
total Rpt1 is graphed for each condition.
Nucleic AcidsResearch, 2010, Vol.38,No. 3793
cannot be solely due to differences in expression levels and
must be a result of the ubiquitin mutations.
Mutation of I44 of ubiquitin reduces the activity
The activity of the ubiquitin–Gal4D mutant constructs
was monitored to determine if the transcriptional output
correlated with the occupancy levels seen above. If true,
then the mutations of ubiquitin are most likely only affect-
ing DNA occupancy because of changes in inhibition of
proteasomal-mediated destabilization and not causing
changes in transcription activity to the Gal4 protein.
An enzyme assay was used to monitor the production
a-galactosidase, the gene product of the Gal4 responsive
MEL1 gene (Figure 4A). Adding ubiquitin to the
N-terminus of Gal4D (Ub Gal4D) increased enzymatic
activity by more than 2-fold compared with Gal4D.
Mutation of the hydrophobic patch (I44A) decreased
activity levels of the Ub-Gal4D constructs to levels
similar to Gal4D lacking the ubiquitin fusion. Mutation
of other residues (D58A) did not decrease the activity of
the Ub-Gal4D constructs.
As a second measure of activity, the mRNA produced
from the GAL1 gene was monitored by quantitative PCR
(Figure 4B and Supplementary Figure S3). The ability of
ubiquitin to promote activity of the Gal4D protein was
decreased by mutation of the hydrophobic patch of
ubiquitin (I44A). Mutation of other residues of ubiquitin
had little effect on the activity of the Ub-Gal4D fusion
protein (D58A). Both assays used to measure activity
demonstrated that the hydrophobic patch of ubiquitin
is required to promote transcriptional activation of
galactose responsive genes.
In this study we set out to find a possible explanation
for the differencesbetween
lysine-48-linked poly-ubiquitin chains to inhibit the
destabilization activity of the proteasome. Monomeric or
completely solvent exposed forms of ubiquitin chains
Figure 4. The I44A mutation abolishes the ability of ubiquitin to
promote Gal4D activity. (A) An a-galactosidase assay measuring the
MEL1gene product produced
The average and SEM of three experiments are shown normalized to
the levels produced with wild-type Gal4. (B) The relative ratio of GAL1
and ACT1 mRNA is graphed based on qPCR quantization of changes
in Ct values from strains expressing the indicated Gal4 form.
The average and SEM of three experiments are shown normalized to
the levels of GAL1 produced with wild-type Gal4.
fromthe indicated constructs.
Figure 3. Mutation of the hydrophobic patch of ubiquitin prevents
rescue of Gal4D. (A) A ChIP assay was used to monitor the DNA
occupancy of the different Gal4 constructs. The relative fold enrich-
ment of GAL7 promoter DNA is indicated for the Gal4D construct
indicated below each bar with the Gal4D construct set as 1. (B)
Expression levels of Gal4D constructs. The steady-state level of the
indicated protein was measured by western blot with an antibody
raised against Gal4.
794 Nucleic Acids Research, 2010,Vol. 38,No. 3
(lysine-63-linked) protected activator–DNA complexes
from destabilization, but tightly packed or ‘closed’ forms
(lysine-48-linked) did not (Figure 1). This biochemical
assay suggested that the hydrophobic patch, buried in
the closed form, might be important for the ubiquitin/
proteasome interaction. We then used mutational studies
of the hydrophobic patch of mono-ubiquitin in vitro
and in vivo as a more physiological relevant look at the
importance of the hydrophobic patch.
Mutation of the hydrophobic patch produced a form of
ubiquitin that was no longer able to bind to the Rpt1
protein (Figure 2). In vivo, mutation of the hydrophobic
patch of ubiquitin prevented partial rescue of the occu-
pancy of the Gal4D protein on the promoters of galactose
responsive genes (Figure 3), which correlated with the
inability of a mutated hydrophobic patch to restore
activity to the protein (Figure 4). Mutation of other
regions on the surface of ubiquitin did not cause any
decrease in the activity of ubiquitin. We conclude that
the hydrophobic patch of ubiquitin is required to inhibit
the destabilization activity of the proteasomal ATPases
on transactivator-DNA complexes.
The hydrophobic patch, made of up L8, I44 and V70
residues, has long been known to be an important region
of the surface to ubiquitin. Mutation of these residues is
known to disrupt proteolysis and endocytosis of target
proteins (32,33). There are now several structural studies
that suggest that many of the ubiquitin binding proteins
interact directly with the hydrophobic patch of ubiquitin
(39,40). The ability of point mutations to the hydrophobic
patch to alter the activity of ubiquitin suggests that these
structural studies are physiological relevant and the
hydrophobic patch is an interaction region for ubiquitin
Mutation of this patch resulted in a form of ubiquitin
that cannot compete for Rpt1 binding in vitro. In addition,
mutation of the hydrophobic patch, but not other regions
on the surface of ubiquitin, would also disrupt the protec-
tive effect of a monomeric ubiquitin in vivo. The fact that
mutation would disrupt the activity of monomeric forms
of ubiquitin suggests that the effect requires direct
interaction with an intact hydrophobic patch.
However, a final verdict on the role of the hydrophobic
patchof ubiquitinin proteasomal-mediated
bilization will require more in-depth structural studies.
Our previous cross-linking strategies can provide informa-
tion about potential interacting partners within the RP of
the proteasome and implicated Rpn1 and Rpt1 as binding
partners. The cross-linking studies provided no informa-
tion about potential surfaces or residues on the surface
of ubiquitin that were important for this interaction.
The comparisons between different forms of ubiquitin
chains in vitro and mutational data from in vivo studies
seen here clearly demonstrate that the hydrophobic
patch of ubiquitin is important for the inhibition of
This study also further defines the connection between
two non-proteolytic roles of the proteasomal-ATPases,
proteasomal-mediated destabilization of transactivators
(18,20) and promoting efficient transition into an elonga-
tion complex (11,12). Both activities required an exposed
transactivator is not protected by mono-ubiquitylation
or, as this study shows lacks a hydrophobic patch, then
the activator is stripped from the DNA. However, the
presence of the hydrophobic patch of mono-ubiquitin
allows interaction with the proteasome through Rpt1
and Rpn1. This interaction protects the transactivator
and allows the proteasomal ATPases to transition
into their role of promoting transcription elongation.
The exposed hydrophobic patch acts as an important
non-proteolytic activities of the proteasomal ATPase will
to recruitthe proteasomalATPases.If the
Supplementary Data are available at NAR Online.
The authors thank Prof George DeMartino (UTSW)
for kindly providing the bacterially expressed Rpt1
National Institutes of Health [GM 087283]. Funding for
open access charge: National Institutes of Health [GM
Conflicts of interest statement. None declared.
1. Baumeister,W., Walz,J., Zuhl,F. and Seemuller,E. (1998) The
proteasome: paradigm of a self-compartmentalizing protease.
Cell, 92, 367–380.
2. Treier,M., Staszewski,L.M. and Bohmann,D. (1994)
Ubiquitin-dependent c-Jun degradation in vivo is mediated by the
delta domain. Cell, 78, 787–798.
3. Chowdary,D.R., Dermody,J.J., Jha,K.K. and Ozer,H.L. (1994)
Accumulation of p53 in a mutant cell line defective in the ubiquitin
pathway. Mol. Cell. Biol., 14, 1997–2003.
4. Salghetti,S.E., Kim,S.Y. and Tansey,W.P. (1999) Destruction of
Myc by ubiquitin-mediated proteolysis: cancer-associated and
transforming mutations stabilize Myc. EMBO J., 18, 717–726.
5. Palombella,V.J., Rando,O.J., Goldberg,A.L. and Maniatis,T. (1994)
The ubiquitin-proteasome pathway is required for processing the
NF-kappa B1 precursor protein and the activation of NF-kappa B.
Cell, 78, 773–785.
6. Gillette,T.G., Gonzalez,F., Delahodde,A., Johnston,S.A. and
Kodadek,T. (2004) Physical and functional association of RNA
polymerase II and the proteasome. Proc. Natl Acad. Sci. USA, 101,
7. Auld,K.L., Brown,C.R., Casolari,J.M., Komili,S. and Silver,P.A.
(2006) Genomic association of the proteasome demonstrates
overlapping gene regulatory activity with transcription factor
substrates. Mol. Cell, 21, 861–871.
8. Sikder,D., Johnston,S.A. and Kodadek,T. (2006) Widespread, but
non-identical, association of proteasomal 19 and 20S proteins with
yeast chromatin. J. Biol. Chem., 281, 27346–27355.
9. Lee,D., Ezhkova,E., Li,B., Pattenden,S.G., Tansey,W.P. and
Workman,J.L. (2005) The proteasome regulatory particle alters the
SAGA coactivator to enhance its interactions with transcriptional
activators. Cell, 123, 423–436.
Nucleic AcidsResearch, 2010, Vol.38,No. 3 795
10. Ezhkova,E. and Tansey,W.P. (2004) Proteasomal ATPases link Download full-text
ubiquitylation of histone H2B to methylation of histone H3.
Mol. Cell, 13, 435–442.
11. Ferdous,A., Gonzalez,F., Sun,L., Kodadek,T. and Johnston,S.A.
(2001) The 19S regulatroy particle of the proteasome is required for
efficient transcriiption elongation by RNZ Polymerase II. Mol. Cell,
12. Ferdous,A., Kodadek,T. and Johnston,S.A. (2002) A
nonproteolytic function of the 19S regulatory subunit of the 26S
proteasome is required for efficient activated transcription by
human RNA Polymerase II. Biochemistry, 41, 12798–12805.
13. Lassot,I., Latreille,D., Rousset,E., Sourisseau,M., Linares,L.K.,
Chable-Bessia,C., Coux,O., Benkirane,M. and Kiernan,R.E. (2007)
The proteasome regulates HIV-1 transcription by both proteolytic
and nonproteolytic mechanisms. Mol. Cell, 25, 369–383.
14. Bhat,K.P., Turner,J.D., Myers,S.E., Cape,A.D., Ting,J.P. and
Greer,S.F. (2008) The 19S proteasome ATPase Sug1 plays a critical
role in regulating MHC class II transcription. Mol. Immunol., 45,
15. Sulahian,R., Sikder,D., Johnston,S.A. and Kodadek,T. (2006) The
proteasomal ATPase complex is required for stress-induced
transcription in yeast. Nucleic Acids Res., 34, 1351–1357.
16. Archer,C.T., Burdine,L. and Kodadek,T. (2005) Identification of
Gal4 activation domain-binding proteins in the 26S proteasome by
periodate-triggered cross-linking. Mol. Biosyst., 1, 366–372.
17. Fernando Gonzalez,A.D., Thomas,K. and Stephen,A.J. (2002)
Recruitment of a 19S proteasome subcomplex to an activated
promoter. Science, 296, 548–550.
18. Ferdous,A., Sikder,D., Gillette,T., Nalley,K., Kodadek,T. and
Johnston,S.A. (2007) The role of the proteasomal ATPases and
activator monoubiquitylation in regulating Gal4 binding to
promoters. Genes Dev., 21, 112–123.
19. Archer,C.T., Burdine,L., Liu,B., Ferdous,A., Johnston,S.A. and
Kodadek,T. (2008) Physical and functional interactions of
monoubiquitylated transactivators with the proteasome. J. Biol.
Chem., 283, 21789–21798.
20. Archer,C.T., Delahodde,A., Gonzalez,F., Johnston,S.A. and
Kodadek,T. (2008) Activation domain-dependent
monoubiquitylation of Gal4 protein is essential for promoter
binding in vivo. J. Biol. Chem., 283, 12614–12623.
21. Deveraux,Q., van Nocker,S., Mahaffey,D., Vierstra,R. and
Rechsteiner,M. (1995) Inhibition of ubiquitin-mediated proteolysis
by the Arabidopsis 26S protease subunit S5a. J. Biol. Chem., 270,
22. van Nocker,S., Sadis,S., Rubin,D.M., Glickman,M., Fu,H.,
Coux,O., Wefes,I., Finley,D. and Vierstra,R.D. (1996) The
multiubiquitin-chain-binding protein Mcb1 is a component of the
26S proteasome in Saccharomyces cerevisiae and plays a
nonessential, substrate-specific role in protein turnover. Mol. Cell.
Biol., 16, 6020–6028.
23. Lam,Y.A., Lawson,T.G., Velayutham,M., Zweier,J.L. and
Pickart,C.M. (2002) A proteasomal ATPase subunit recognizes the
polyubiquitin degradation signal. Nature, 416, 763–767.
24. Schreiner,P., Chen,X., Husnjak,K., Randles,L., Zhang,N.,
Elsasser,S., Finley,D., Dikic,I., Walters,K.J. and Groll,M. (2008)
Ubiquitin docking at the proteasome through a novel
pleckstrin-homology domain interaction. Nature, 453, 548–552.
25. Husnjak,K., Elsasser,S., Zhang,N., Chen,X., Randles,L., Shi,Y.,
Hofmann,K., Walters,K.J., Finley,D. and Dikic,I. (2008)
Proteasome subunit Rpn13 is a novel ubiquitin receptor. Nature,
26. Liu,B., Archer,C.T., Burdine,L., Gillette,T.G. and Kodadek,T.
(2007) Label transfer chemistry for the characterization of
protein-protein interactions. J. Amer. Chem. Soc., 129,
27. Russell,S.J. and Johnston,S.A. (2001) Evidence that proteolysis of
Gal4 cannot explain the transcriptional effects of proteasome
ATPase mutations. J. Biol. Chem., 276, 9825–9831.
28. Verma,R., Oania,R., Graumann,J. and Deshaies,R.J. (2004)
Multiubiquitin chain receptors define a layer of substrate selectivity
in the ubiquitin-proteasome system. Cell, 118, 99–110.
29. Kew,O.M. and Douglas,H.C. (1976) Genetic co-regulation of
galactose and melibiose utilization in Saccharomyces. J. Bacteriol.,
30. Cook,W.J., Jeffrey,L.C., Kasperek,E. and Pickart,C.M. (1994)
Structure of tetraubiquitin shows how multiubiquitin chains can
be formed. J. Mol. Biol., 236, 601–609.
31. Varadan,R., Walker,O., Pickart,C. and Fushman,D. (2002)
Structural properties of polyubiquitin chains in solution. J. Mol.
Biol., 324, 637–647.
32. Beal,R.E., Toscano-Cantaffa,D., Young,P., Rechsteiner,M. and
Pickart,C.M. (1998) The hydrophobic effect contributes to
polyubiquitin chain recognition. Biochemistry, 37, 2925–2934.
33. Sloper-Mould,K.E., Jemc,J.C., Pickart,C.M. and Hicke,L. (2001)
Distinct functional surface regions on ubiquitin. J. Biol. Chem., 276,
34. Bayer,P., Arndt,A., Metzger,S., Mahajan,R., Melchior,F.,
Jaenicke,R. and Becker,J. (1998) Structure determination of the
small ubiquitin-related modifier SUMO-1. J. Mol. Biol., 280,
35. Vijay-Kumar,S., Bugg,C.E. and Cook,W.J. (1987) Structure of
ubiquitin refined at 1.8A resolution. J. Mol. Biol., 194, 531–544.
36. Varadan,R., Assfalg,M., Haririnia,A., Raasi,S., Pickart,C. and
Fushman,D. (2004) Solution conformation of Lys63-linked
di-ubiquitin chain provides clues to functional diversity of
polyubiquitin signaling. J. Biol. Chem., 279, 7055–7063.
37. Ding,W.V. and Johnston,S.A. (1997) The DNA binding and
activation domains of Gal4p are sufficient for conveying its
regulatory signals. Mol. Cell. Biol., 17, 2538–2549.
38. Matsumoto,K., Adachi,Y., Toh-e,A. and Oshima,Y. (1980)
Function of positive regulatory gene gal4 in the synthesis of
galactose pathway enzymes in Saccharomyces cerevisiae: evidence
that the GAL81 region codes for part of the gal4 protein.
J. Bacteriol., 141, 508–527.
39. Swanson,K.A., Kang,R.S., Stamenova,S.D., Hicke,L. and
Radhakrishnan,I. (2003) Solution structure of Vps27
UIM-ubiquitin complex important for endosomal sorting and
receptor downregulation. EMBO J., 22, 4597–4606.
40. Lee,S., Tsai,Y.C., Mattera,R., Smith,W.J., Kostelansky,M.S.,
Weissman,A.M., Bonifacino,J.S. and Hurley,J.H. (2006) Structural
basis for ubiquitin recognition and autoubiquitination by Rabex-5.
Nature Struct. Mol. Biol., 13, 264–271.
796Nucleic Acids Research, 2010,Vol. 38,No. 3