TRRAP and GCN5 are used by c-Myc to activate
RNA polymerase III transcription
Niall S. Kenneth*, Ben A. Ramsbottom*, Natividad Gomez-Roman*, Lynne Marshall*†, Philip A. Cole‡,
and Robert J. White*†§
*Institute of Biomedical and Life Sciences, University of Glasgow, Glasgow G12 8QQ, United Kingdom;†Beatson Institute for Cancer Research, Garscube
Estate, Switchback Road, Bearsden, Glasgow G61 1BD, United Kingdom; and‡Department of Pharmacology and Molecular Sciences, Johns Hopkins
University School of Medicine, Baltimore, MD 21205
Edited by Robert N. Eisenman, Fred Hutchinson Cancer Research Center, Seattle, WA, and approved July 31, 2007 (received for review March 29, 2007)
Activation of RNA polymerase (pol) II transcription by c-Myc gen-
erally involves recruitment of histone acetyltransferases and acet-
ylation of histones H3 and H4. Here, we describe the mechanism
used by c-Myc to activate pol III transcription of tRNA and 5S rRNA
genes. Within 2 h of its induction, c-Myc appears at these genes
along with the histone acetyltransferase GCN5 and the cofactor
TRRAP. At the same time, occupancy of the pol III-specific factor
TFIIIB increases and histone H3 becomes hyperacetylated, but
increased histone H4 acetylation is not detected at these genes.
The rapid acetylation of histone H3 and promoter assembly of
TFIIIB, c-Myc, GCN5, and TRRAP are followed by recruitment of pol
III and transcriptional induction. The selective acetylation of his-
tone H3 distinguishes pol III activation by c-Myc from mechanisms
observed in other systems.
acetyltransferase ? chromatin ? histone ? TFIIIB ? TFIIIC
complex transcriptional programs involving large numbers of
target genes (1, 2). Many of these targets encode components of
the ribosome (3). Thus, c-Myc induces transcription by RNA
polymerase (pol) II of many ribosomal protein genes and can
also stimulate the synthesis of rRNA and tRNA, by directly
activating transcription by pols I and III (3). This striking ability
to activate transcription by three different pols allows c-Myc to
induce multiple components of the ribosome coordinately and
was unexpected because their promoters do not usually contain
good matches to the E-box DNA sequence that is characteristically
used by c-Myc to activate transcription by pols I and II. Neverthe-
less, a hydroxytamoxifen (OHT)-inducible Myc-estrogen receptor
(MycER) fusion will specifically induce pol III-transcribed genes in
human diploid fibroblasts and transgenic mice (4, 5). The rapidity
of this response suggests direct regulation, which is supported by
experiments in which ?-amanitin was used to exclude pol II-
mediated effects (4). Direct action was confirmed by ChIP, which
genes in vivo (4, 6). The pol III-specific transcription factor TFIIIB
coimmunoprecipitation and cofractionation experiments showed
that endogenous c-Myc associates stably with endogenous TFIIIB
(4). Furthermore, scanning ChIP assays suggest that c-Myc colo-
calizes with TFIIIB in the vicinity of the transcription start site
[supporting information (SI) Fig. 6]. These observations support
attracts c-Myc to pol III templates.
Most recent work on the mechanisms of activation by c-Myc has
focused on its ability to influence chromatin structure (7–9). The
transactivation domain of c-Myc binds to the cofactor TRRAP
(10), a 400-kDa member of the ATM family that forms com-
he protooncogene product c-Myc is a potent inducer of cell
growth and proliferation (1, 2), which it achieves through
plexes with the histone acetyltransferases (HATs) GCN5, p300/
CBP-associated factor (PCAF), and TIP60 (11, 12). These
interactions allow c-Myc to recruit HAT activity to the vicinity
of E-box sites, leading to acetylation of histones H3 and H4
(13–16). Binding has also been observed between c-Myc and the
INI1/hSNF5 subunit of the SWI/SNF chromatin remodeling
complex and the ATPase/helicases TIP48 and TIP49 (17, 18).
Indeed, c-Myc can influence nuclear condensation and histone
modification across substantial domains (19, 20). Such effects on
chromatin organization are likely to impact strongly on the
expression of many genes.
The ability to induce chromatin structures that are conducive
to transcription is considered a key feature of gene activation by
Myc. However, it may not be universal, and other mechanisms
have also been described. For example, activation of the cad
promoter by c-Myc does not require recruitment of TRRAP and
associated HATs and involves little change in histone acetylation
(21, 22). In this case, c-Myc recruits the kinase P-TEFb, which
may phosphorylate preassembled pol II and thereby stimulate
promoter clearance and transcript elongation (22–24). A notable
feature of this mechanism is that polymerase occupancy is
already high at the uninduced cad promoter and does not
increase further in response to c-Myc (22, 24). Activation of each
of five other c-Myc targets examined also involved no apparent
increase in the occupancy of pol II (24). In contrast, loading of
P-TEFb, TFIIH, and Mediator increased at these promoters in
response to c-Myc (24). Several earlier studies have shown
association of c-Myc with TATA box-binding protein (TBP)
through direct binding (25–28), providing another plausible
mechanism for c-Myc to regulate gene expression. Indeed,
substantial TBP recruitment by c-Myc is observed at pol I-tran-
scribed rRNA genes (29). However, activation of the cyclin D2
promoter by c-Myc was not accompanied by any increase in TBP
Here, we have characterized the molecular basis of pol III
transcriptional activation by c-Myc. In some respects it resembles
what has been previously described for pol II induction. For
example, c-Myc recruits the cofactors TRRAP and GCN5 to
tRNA and 5S rRNA genes and stimulates acetylation of histone
H3 at these loci. In contrast, TIP60 is not detected and there is
no evidence of increased histone H4 acetylation. Gene occu-
Author contributions: N.S.K. and R.J.W. designed research; N.S.K., B.A.R., N.G.-R., and L.M.
performed research; P.A.C. contributed new reagents/analytic tools; N.S.K. and R.J.W.
analyzed data; and R.J.W. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Abbreviations: ARPP, acidic ribosomal phosphoprotein; HAT, histone acetyltransferase;
pol, RNA polymerase; TBP, TATA box-binding protein; TSA, trichostatin A.
§To whom correspondence should be addressed. E-mail: firstname.lastname@example.org.
This article contains supporting information online at www.pnas.org/cgi/content/full/
© 2007 by The National Academy of Sciences of the USA
September 18, 2007 ?
vol. 104 ?
no. 38 ?
pancy by TFIIIB increases rapidly in response to c-Myc and is
followed by recruitment of pol III.
TRRAP Associates with TFIIIB and Activates Pol III Transcription. To
stimulate pol III transcription, c-Myc requires both its N-
terminal transactivation domain (4) and its C-terminal dimer-
ization and DNA-binding domain (SI Fig. 7). The pol III
response is abolished by deletion of residues 106–143 from
within the transactivation domain (4). This region includes the
conserved MBII motif (residues 129–143), which is essential for
cell transformation by c-Myc (30). Because MBII is the binding
site for the transactivation/transformation domain-associated
protein TRRAP (10), we used ChIP to test whether this cofactor
is present at pol III-transcribed genes in vivo. Indeed, endoge-
rRNA genes, as was the TFIIIB subunit Brf1, which was used as
a positive control (Fig. 1A). In contrast, little or no binding was
detected for the negative control TFIIB, which is pol II-specific.
Furthermore, none of these proteins were found within the
coding region of the gene encoding acidic ribosomal phospho-
protein (ARPP) P0. We conclude that TRRAP associates
specifically with tRNA and 5S rRNA genes in vivo.
To test whether TRRAP influences pol III transcription,
RNAi was used to reduce its expression. Relative to mock-
transfected controls, levels of TRRAP mRNA and protein were
specifically decreased when HeLa cells were transfected with
siRNA against the message encoding TRRAP (Fig. 1 B and C).
In contrast, the TRRAP siRNA did not diminish expression of
Oct1, actin proteins, or ARPP P0 mRNA. To evaluate the effect
of TRRAP knockdown on pol III activity, we carried out
RT-PCR using primers that recognize intron-containing pre-
tRNA primary transcripts; as these products are processed very
rapidly, their levels reflect the rate of transcription (31). Deple-
tion of TRRAP was found to decrease expression of pre-tRNA
(Fig. 1C), suggesting that pol III transcription is stimulated by
TRRAP. No change in pre-tRNA level was seen in response to
a control siRNA that efficiently depleted Oct-1. The response to
TRRAP depletion is unlikely to be caused by fortuitous cross-
reaction by the siRNA with an alternative target, as the same
result was obtained with a vector encoding a specific short
hairpin RNA against a different part of the TRRAP sequence
(SI Fig. 8). We conclude that TRRAP has a positive effect on
expression of pol III products.
Recruitment of TRRAP to Pol III-Transcribed Genes Is Promoted by
c-Myc. ChIP assays were used to compare tRNA gene occupancy
by TRRAP in wild type and c-Myc null fibroblasts. Expression
of pre-tRNA is markedly diminished in the c-Myc?/?cells, when
compared with the wild type (SI Fig. 9A). Consistent with this
finding, pol III occupancy of the corresponding genes is lower in
the knockouts, as assessed with antibody against the pol III
subunit RPC155 (Fig. 2A). TRRAP occupancy is also decreased
in the knockouts. This difference cannot be attributed to dif-
ferences in expression, as levels of TRRAP and RPC155 are
comparable in the two cell types (SI Fig. 9B). Instead, it appears
Fibroblasts expressing an inducible MycER fusion construct
were used to test whether c-Myc can recruit TRRAP to pol III
templates. These cells were first starved of serum to deplete
endogenous c-Myc. OHT triggers rapid accumulation within
nuclei of the MycER fusion protein, as shown by Western
blotting of nuclear extracts (SI Fig. 9C). No change was seen in
the levels of representative subunits of TFIIIB, TFIIIC, and pol
III. ChIP analysis shows that within 2 h of OHT addition MycER
occupancy increases at tRNA and 5S rRNA genes (Fig. 2B and
SI Fig. 9D). Similarly rapid recruitment is seen at the cyclin D2
promoter, a paradigm c-Myc target that contains an E-box DNA
sequence. Nevertheless, binding is specific, as it is not detected
1 t c
1 t c
I I F
1 f r
ChIP using Brf1, TRRAP, and TFIIB antibodies with HEK293 cells and primers to
the indicated genes. (B) Immunoblot for TRRAP, Oct1, and actin in extracts of
HeLa cells transfected with siRNAs against TRRAP or Oct1 mRNAs (lanes 1 and
2, respectively) or mock-transfected (lane 3). (C) RT-PCR of pre-tRNA and the
TRRAP and ARPP P0 mRNAs in HeLa cells transfected with siRNAs against
TRRAP mRNA (lanes 1 and 2), Oct1 mRNA (lanes 3 and 4), or mock-transfected
(lanes 5 and 6).
TRRAP binds tRNA genes and stimulates their expression in vivo. (A)
- + - + - + - + - +
0 2 4
0 2 4
0 2 4 :hrs
0 2 4
0 2 4
0 2 4
0 2 4 0 2 4 :hrs
I I F
I I I l o
using RPC155, TRRAP, and TFIIB antibodies with wild-type and c-Myc null
antibodies with MycER-expressing fibroblasts after 0, 2, or 4 h of OHT
TRRAP is recruited to pol III templates in response to c-Myc. (A) ChIP
www.pnas.org?cgi?doi?10.1073?pnas.0702909104Kenneth et al.
at the ARPP P0 gene. The data are clearly consistent with the
pol III-transcribed genes serving as direct targets for c-Myc.
Furthermore, TRRAP occupancy of these genes also increases
within 2 h, in parallel with that of the MycER protein (Fig. 2C
and SI Fig. 9E). Neither c-Myc nor TRRAP were recruited to 5S
rRNA genes after OHT treatment of fibroblasts carrying empty
vector instead of the MycER expression construct (data not
shown). We conclude that c-Myc facilitates recruitment of
TRRAP to pol III templates.
GCN5 Activates Pol III-Transcribed Genes and Is Recruited in Response
to c-Myc. TRRAP has been shown to mediate recruitment by
c-Myc of the HAT GCN5 (32). Indeed, OHT treatment of
MycER fibroblasts triggered the rapid appearance of GCN5 at
tRNA genes in vivo (Fig. 3A and SI Fig. 10A). We therefore used
RNAi to test whether GCN5 levels influence expression of pol
III transcripts. Efficient depletion of GCN5 was achieved with a
specific siRNA (Fig. 3B), which resulted in a marked reduction
when cells were transfected with control siRNA against Oct-1
(Fig. 3C). We conclude that GCN5 is recruited to tRNA genes
in response to c-Myc and has a stimulatory effect on their
In addition to its complex with GCN5, TRRAP is found in a
distinct complex containing TIP60 (12). However, we found no
Indeed, the ChIP signal obtained with TIP60 antibodies is close
to background on tRNA and 5S rRNA genes. This finding does
not reflect an inability to ChIP TIP60, as this cofactor is clearly
detected at the promoter of pol I-transcribed large rRNA genes
(SI Fig. 10B) as reported (33). We also tested a cell line that
overexpresses a tagged version of TIP60, but still found little or
no binding to tRNA or 5S genes (data not shown). It therefore
appears that only a subset of TRRAP-containing complexes is
recruited to these pol III templates.
Induction of c-Myc Triggers Selective Acetylation of Histone H3 at Pol
III-Transcribed Genes. Because the above results indicate that
c-Myc recruits GCN5 to tRNA genes, we tested whether HAT
activity contributes to pol III transcriptional activation. H3-
CoA-20-Tat is a cell-permeable peptide CoA conjugate that
specifically inhibits PCAF and its close relative GCN5 (34). This
treatment was found to compromise tRNA gene induction when
added to MycER fibroblasts before OHT addition (Fig. 4A). No
such effect was observed when cells were treated in parallel with
a related, but inactive, peptide control. These data suggest that
the stimulatory influence of GCN5 on pol III transcription
involves its HAT activity.
We therefore carried out ChIP assays to examine acetylation
of histones at these loci (Fig. 4B and SI Fig. 11). As positive
control, we used the cyclin D2 promoter, where activation by
c-Myc is known to involve histone acetylation (13). This site
became enriched for acetylated histones H3 and H4 within 2 h
of OHT addition to MycER-transduced fibroblasts. The re-
sponse was specific, as it was not observed at the p21 gene
promoter, which is not activated by c-Myc. As with the cyclin D2
promoter, acetylation of histone H3 was increased markedly at
tRNA and 5S rRNA genes 2 h after MycER induction. In
contrast, no increase in acetylated histone H4 was observed at
these genes, either 2 or 4 h after induction. This finding is
consistent with the failure of c-Myc to recruit TIP60 to these loci
(Fig. 3D), as TIP60 is thought to be primarily responsible for H4
acetylation in response to c-Myc. Histone acetylation was not
affected by OHT treatment of empty vector control cells (data
1 t c
1 t c
0 2 4
0 2 4
0 2 40 2 4
- +- +
- +- +
I I F
their expression. (A) ChIP using c-Myc, GCN5, and TAFI48 antibodies with
MycER-expressing fibroblasts after 0, 2, or 4 h of treatment with OHT. (B)
Western blot for GCN5 and actin in extracts of HeLa cells transfected with
siRNAs against GCN5 or Oct1 mRNAs (lanes 1 and 2, respectively) or mock-
transfected (lane 3). (C) RT-PCR to compare levels of pre-tRNA and ARPP P0
mRNAs in HeLa cells transfected with siRNAs against GCN5 mRNA (lane 1),
Oct1 mRNA (lane 2), or mock-transfected (lane 3). (D) ChIP using GCN5, TIP60,
and TFIIB antibodies with matched wild-type and c-Myc knockout fibroblasts.
GCN5 is recruited to tRNA genes in response to c-Myc and stimulates
ARPP P0 mRNA
+ + + + + +
l o r t n
d i t p
r o t i b i h
d i t p
d i t p
d i t p
0 2 4
0 2 4
0 2 40 2 40 2 4
H3 Ac H4 Ac TAFI48
genes. (A) RT-PCR of pre-tRNA and ARPP P0 mRNA in MycER-expressing
fibroblasts after 0 (lanes 1 and 2) or 4 h (lanes 3–8) of OHT treatment. Where
indicated, cells were pretreated for 24 h with 50 ?M H3-Ac-20-Tat control
peptide (lanes 5 and 6) or H3-CoA-20-Tat GCN5/PCAF inhibitor peptide (lanes
7 and 8). (B) ChIP using antibodies against TAFI48 and acetylated histones H3
and H4 with MycER-expressing fibroblasts after 0, 2, or 4 h of treatment with
OHT, as indicated. Immunoprecipitated DNA was PCR-amplified by using the
indicated gene primers.
Induction of c-Myc stimulates H3 acetylation at pol III-transcribed
Kenneth et al.
September 18, 2007 ?
vol. 104 ?
no. 38 ?
Recruitment of TFIIIB and Pol III Is Stimulated by c-Myc and Trichos-
tatin A (TSA). In addition to the modification of histones, c-Myc
stimulates recruitment of TFIIH, P-TEFb, and mediator to the
cyclin D2 promoter (13, 24). We therefore examined whether
c-Myc also promotes assembly of pol III transcription factors.
and antibodies specific for TFIIIB, TFIIIC, and pol III (Fig. 5A).
As expected, each of these proteins was detected at tRNA genes,
but not at the cyclin D2 promoter or within the ARPP P0 coding
region. As in Fig. 2A, pol III occupancy increased significantly
after MycER induction, as did acetylation of histone H3 but not
of histone H4. In addition, a marked increase in TFIIIB binding
was observed. In contrast, TFIIIC occupancy remained unal-
tered when tested in parallel. These data suggest that c-Myc can
genes in vivo. This conclusion is reinforced by the experiments
in Fig. 5B and SI Fig. 12, which show selective recruitment of
TFIIIB and pol III to tRNA and 5S rRNA genes, but not to the
ARPP P0 gene. Both the Brf1 and Bdp1 subunits of TFIIIB are
subunits examined shows any evidence of a change in occupancy.
For both Brf1 and Bdp1, the increase in binding has occurred
after 4 h. This increase coincides with the recruitment of c-Myc
and that of TRRAP and GCN5 and the increased acetylation of
histone H3. However, an increase in pol III assembly is not seen
at the 2-h time point and only becomes apparent after 4 h. We
conclude that activation of tRNA and 5S rRNA genes by c-Myc
in vivo involves sequential assembly of TFIIIB followed by pol
To test whether elevated protein acetylation is sufficient to
equilibria of acetylation reactions. Thus, Western blots show
highly elevated acetylation of histones H3 and H4 in TSA-
treated cells (SI Fig. 13A). ChIP assays confirm that this increase
in acetylation applies to histones associated with pol III-
transcribed genes (Fig. 5C). They also reveal a rapid increase in
occupancy by TFIIIB and pol III. In contrast, little or no increase
is seen in gene occupancy by TFIIIC. Therefore, TSA provokes
the same selective recruitment of pol III factors as occurs after
c-Myc induction. The response cannot be attributed to induced
expression of these proteins, as Western blots show no change in
the levels of representative subunits after TSA treatment (SI Fig.
13B). Instead, all of the data support a model in which HAT
activity recruited via c-Myc stimulates assembly of active tran-
scription complexes onto pol III-transcribed genes.
This study presents a mechanistic analysis of how c-Myc stimu-
lates tRNA and 5S rRNA synthesis in vivo. It identifies several
ways in which activation resembles what has been reported for
other classes of c-Myc targets. However, an unusual feature of
pol III transcriptional induction is the selective increase in
acetylation of histone H3, but not of histone H4. The clear
increase in polymerase occupancy also contrasts with what has
been observed at pol II-transcribed c-Myc targets.
TRRAP and GCN5 are detected at pol III-transcribed genes
in vivo. RNAi-mediated depletion of either decreases tRNA
gene expression. This result is unlikely to reflect off-target
that match different TRRAP sequences, and the response to a
GCN5 siRNA was independently confirmed by using a peptide
inhibitor. We therefore conclude that TRRAP and GCN5 are
both involved in pol III regulation. This role appears to largely
depend on c-Myc, as c-Myc null cells show markedly reduced
TRRAP occupancy of tRNA genes. However, a weak TRRAP
ChIP signal is retained in the knockout fibroblasts, albeit much
diminished compared with the wild type; this result raises the
possibility that TRRAP or its associated proteins make contacts
with the pol III machinery, although it is clear that any inter-
action is strongly promoted by c-Myc. The MBII region of c-Myc
is responsible for its binding to TRRAP and association with
GCN5. MBII deletion prevents pol III induction by c-Myc.
Furthermore, tRNA gene induction by c-Myc is attenuated when
cells are treated with an inhibitor of GCN5 HAT activity. The
data suggest a model in which c-Myc recruits a TRRAP/GCN5
complex to pol III templates to stimulate their transcription. The
positive effect of GCN5 on tRNA production is consistent with
its well documented role in amino acid synthesis.
TRRAP also forms a complex with the HAT PCAF, which is
not tested whether PCAF regulates pol III, because the siRNA
and antibody reagents we used are specific for GCN5. However,
the close similarity between these HATs makes it likely that
PCAF also influences pol III transcription.
I I I l o
I I I F
I I I F
0 4 0 4 0 4 0 4
0 2 4
0 40 4 0 4
0 2 40 2 4 0 2 4 0 2 4 0 2 4 0 2 40 2 4
1 f r
I I I F
I I I F
0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24 0 6 12 24
H3 AcH4 Ac
TFIIICTFIIIB Pol III TFIIB
acetylated histones H3 and H4 with MycER-expressing fibroblasts after 0 or 4 h of treatment with OHT. Immunoprecipitated DNA was PCR-amplified by using
0, 2, or 4 h of treatment with OHT. Immunoprecipitated DNA was PCR-amplified by using the indicated gene primers. (C) ChIP using 5S and tRNA gene primers
and antibodies against acetylated histones H3 and H4, TFIIIC (TFIIIC110), TFIIIB (Brf1), pol III (RPC155), and TFIIB with fibroblasts treated for 0, 6, 12, or 24 h with
Pol III and TFIIIB are recruited selectively in response to c-Myc and TSA. (A) ChIP assay using antibodies against TAFI48, RPC155, Brf1, TFIIIC220, and
www.pnas.org?cgi?doi?10.1073?pnas.0702909104 Kenneth et al.
In contrast, the GCN5 and TIP60 complexes do not share
common components, apart from TRRAP (12). TIP60 does not
seem to be recruited to 5S or tRNA genes after induction of
c-Myc. This observation is consistent with the absence of in-
creased H4 acetylation, as histone H4 is the preferred substrate
of TIP60. In both of these respects, the mechanism of pol III
activation by c-Myc differs from what is commonly seen for pol
also been detected at pol I-transcribed genes (33), as has H4
acetylation in response to c-Myc (29, 35). It is not clear why
TIP60 is excluded from pol III templates, especially given the
presence of TRRAP. Perhaps the TIP60-binding site on
TRRAP is obscured by TFIIIB. It remains possible that TIP60
is present but not detected in our ChIPs, perhaps because of
Our data indicate that HAT activity stimulates recruitment of
TFIIIB and pol III to their target genes, which may be explained by
elevated histone acetylation, allowing increased access to DNA
to remember that HATs can acetylate proteins other than histones
However, we cannot exclude this mechanism as potentially con-
tributing to the regulatory response.
Even in serum-starved fibroblasts expressing little or no
c-Myc, tRNA and 5S rRNA genes are marked by the presence
of TFIIIC and acetylated histone H4. The latter observation may
explain why HATs that target H4 may not be required for pol III
activation by c-Myc. Human TFIIIC has been reported to exhibit
HAT activity in vitro (38, 39), which might account for the
apparently constitutive H4 acetylation we see at pol III tem-
plates. However, TFIIIC was reported to acetylate histone H3
and H4 (38, 39). Perhaps its specificity is altered when bound to
genes in vivo. Alternatively, a different HAT might associate
with TFIIIC in cells. Whether or not TFIIIC has intrinsic HAT
activity, it is noteworthy that it remains bound to genes under
conditions where transcription is low. This finding is consistent
with several previous studies in which TFIIIC was found to
maintain high promoter occupancy, even in the face of unfa-
vorable chromatin conditions (40). Indeed, it has been suggested
that one function of TFIIIC is to resist nucleosomal silencing of
pol III templates (41).
Recruitment of TFIIIB appears to be the stage at which c-Myc
exerts its effect on pol III transcription. In this regard, it
resembles many other regulatory factors, such as the retinoblas-
toma protein (42). The increased gene occupancy by pol III in
response to c-Myc may simply be a consequence of the elevated
levels of promoter-bound TFIIIB, because TFIIIB is necessary
and sufficient to bring pol III to its templates (43). However, this
step might also be facilitated by c-Myc in some way. In contrast,
high levels of pol II were found at the uninduced cad promoter
irrespective of the presence of c-Myc (22). Indeed, no increase
in pol II occupancy was seen at any of the six direct c-Myc targets
promoters. TFIID and TFIIB occupancy was also not enhanced
(24). However, c-Myc induction did increase the occupancy of
these promoters by P-TEFb, TFIIH, and Mediator (22, 24). This
situation is reminiscent of the pol III system, where activation by
c-Myc also involves recruitment of a basal factor (TFIIIB) onto
promoters that are already marked by the presence of prebound
Elevated levels of tRNA and 5S rRNA can be expected to
impact substantially on the protein synthetic capacity of a cell.
Stimulation of pol III transcription may therefore make an
important contribution to the biological efficacy of c-Myc,
especially in terms of growth promotion. Both TRRAP and
GCN5 are required for c-Myc to promote cell growth and
transformation (10, 32), and we have shown here that they are
also involved in pol III regulation. It is noteworthy that a recently
discovered Myc target gene encodes Misu, an RNA methyltrans-
ferase that specifically methylates tRNA and is required for
growth and proliferation of keratinocytes (44). Misu levels are
abnormally elevated in a variety of human cancers, and RNAi-
mediated knockdown of Misu reduces the growth of carcinoma
xenografts in mice (44). Elevated production and processing of
tRNA may therefore make a very significant contribution to
Cell Culture. HeLa and HEK293T cells were grown in DMEM
(Cambrex Biosciences, East Rutherford, NJ) supplemented with
10% FBS (Sigma, St. Louis, MO), 2 mM L-glutamine, and 100
units/ml penicillin and streptomycin. Wild-type and c-Myc null
fibroblasts were cultured as described (29). BALB/c 3T3 A31
fibroblasts transduced with empty pBabe vector or vector ex-
pressing MycER (pB-MYC-ER) were a generous gift from Carla
Grandori (Fred Hutchinson Cancer Research Center, Seattle,
induction with 200 nM 4-OHT as described (29).
ChIP Assays. ChIP was performed as described (4). Immuno-
precipitated DNA was quantified by PCR using published
primers and amplification procedures (4). Antibodies were
N-262 against c-Myc, H-75 against GCN5, C-19 against Max,
M19 against TAFI48, FL-109 against TFIIA, C-18 against
TFIIB, H-93 against TIP60, H-300 against TRRAP (Santa
Cruz Biotechnologies, Santa Cruz, CA), 06-866 against acety-
lated histone H3, 06-599 against acetylated histone H4 (Up-
state Biotechnology, Lake Placid, NY), MTBP-6 against TBP,
128 against Brf1, 4286 against TFIIIC110, Ab4 against
TFIIIC220, 2663 against Bdp1, and 1900 against RPC155 (42,
45). Serial dilutions of chromatin were used to establish that
PCRs were within a linear range.
Cell Extracts and Immunoblotting. Whole-cell extracts were pre-
pared as described (46). For nuclear extracts, cells were washed
twice with ice-cold PBS and scraped into hypotonic buffer (20
mM Hepes, pH 7.0/10 mM KCl/1 mM DTT/0.1% Triton X-100/
20% glycerol/2 mM PMSF/5 ?g/ml aprotinin/5 ?g/ml leupeptin).
The cell suspension was subjected to 10 strokes in a Dounce
homogenizer and then centrifuged at 800 ? g for 5 min. The
supernatant (cytoplasmic fraction) was removed, and the pellet
was resuspended in extraction buffer (20 mM Hepes, pH 7.0/420
mM NaCl/10 mM KCl/1 mM DTT/0.1% Triton X-100/20%
glycerol/2 mM PMSF/5 ?g/ml aprotinin/5 ?g/ml leupeptin). The
suspension was incubated end over end for 20 min at 4°C and
then centrifuged at 16,000 ? g for 10 min to generate the nuclear
Immunoblotting was performed as described (47) using anti-
bodies F-7 against HA, C11 against actin, H-75 against GCN5,
9E10 against c-Myc, T-17 against TRRAP, C-20 against Oct-1
(Santa Cruz Biotechnologies); 06-866 and 06-599 against acety-
lated histone H3 or H4, respectively (Upstate Biotechnology);
MTBP-6 against TBP, 128 against Brf1, 2663 against Bdp1, 1900
and 3208 against TFIIIC110 (45, 46).
RT-PCR Analysis. RT-PCR of ARPP P0 mRNA, 5S rRNA, and
tRNA transcripts was performed as described (31, 48). RT-PCR
ofTRRAP mRNA used
GAAACGCTTAGG-3? and 5?-GTCTTCAGAAGGTTCT-
GAAGAC-3? to give a 252-bp product using the following
cycling parameters: 95°C for 3 min, 25 cycles of 95°C for 1 min,
58°C for 30 s, 72°C for 1 min, 72°C for 5 min.
Kenneth et al.
September 18, 2007 ?
vol. 104 ?
no. 38 ?
RNAi. HeLa cells were transfected at 50% confluence to a final Download full-text
concentration of 100 nM siRNA oligonucleotides, delivered
using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). Medium
was replaced after 5 h, and cells were harvested after an
additional 72 h. The siRNAs used were TRRAP siRNA (h)
sc-36746, GCN5 siRNA (h) sc-37946, Oct-1 siRNA (h) sc-36119,
all from Santa-Cruz Biotechnologies.
We thank Carla Grandori for virally transduced fibroblasts, Craig
Robson (The Medical School, University of Newcastle upon Tyne,
Newcastle upon Tyne, U.K.) for TIP60-expressing cells, and Yuhong
Shen and Arnie Berk (Jonsson Comprehensive Cancer Center, Univer-
sity of California, Los Angeles, CA) for Ab4. This work was supported
by Biotechnology and Biological Sciences Research Council and Cancer
Research UK grants (to R.J.W.).
1. Grandori C, Cowley SM, James LP, Eisenman RN (2000) Annu Rev Cell Dev
2. Adhikary S, Eilers M (2005) Nat Rev Mol Cell Biol 6:635–645.
3. Oskarsson T, Trumpp A (2005) Nat Cell Biol 7:215–217.
4. Gomez-Roman N, Grandori C, Eisenman RN, White RJ (2003) Nature
5. Goodfellow SJ, Innes F, Derblay LE, MacLellan WR, Scott PH, White RJ
(2006) EMBO J 25:1522–1533.
6. Felton-Edkins ZA, Kenneth NS, Brown TRP, Daly NL, Gomez-Roman N,
Grandori C, Eisenman RN, White RJ (2003) Cell Cycle 3:181–184.
7. Cole MD, McMahon SB (1999) Oncogene 18:2916–2924.
8. Amati B, Frank SR, Donjerkovic D, Taubert S (2001) Biochim Biophys Acta
9. Eisenman RN (2001) Genes Dev 15:2023–2030.
10. McMahon SB, Van Buskirk HA, Dugan KA, Copeland TD, Cole MD (1998)
11. Vassilev A, Yamauchi J, Kotani T, Prives C, Avantaggiati ML, Qin J, Nakatani
Y (1998) Mol Cell 2:869–875.
12. Ikura T, Ogryzko VV, Grigoriev M, Groisman R, Wang J, Horikoshi M, Scully
R, Qin J, Nakatani Y (2000) Cell 102:463–473.
13. Bouchard C, Dittrich O, Kiermaier A, Dohmann K, Menkel A, Eilers M,
Luscher B (2001) Genes Dev 15:2042–2047.
14. Frank SR, Schroeder M, Fernandez P, Taubert S, Amati B (2001) Genes Dev
15. Nikiforov MA, Chandriani S, Park J, Kotenko I, Matheos D, Johnsson A,
McMahon SB, Cole MD (2002) Mol Cell Biol 22:5054–5063.
16. Fernandez PC, Frank SR, Wang L, Schroeder M, Liu S, Greene J, Cocito A,
Amati B (2003) Genes Dev 17:1115–1129.
17. Cheng S-WG, Davies KP, Yung E, Beltran RJ, Yu J, Kalpana GV (1999) Nat
18. Wood MA, McMahon SB, Cole MD (2000) Mol Cell 5:321–330.
19. Goodliffe JM, Wieschaus E, Cole MD (2005) Genes Dev 19:2941–2946.
20. Knoepfler PS, Zhang X, Cheng PF, Gafken PR, McMahon SB, Eisenman RN
(2006) EMBO J 25:2723–2734.
21. Eberhardy SR, D’Cunha C, Farnham PJ (2000) J Biol Chem 275:33798–33805.
22. Eberhardy SR, Farnham PJ (2001) J Biol Chem 276:48562–48571.
23. Eberhardy SR, Farnham PJ (2002) J Biol Chem 277:40156–40162.
24. Bouchard C, Marquardt J, Bras A, Medema RH, Eilers M (2004) EMBO J
25. Hateboer G, Timmers HT, Rustgi AK, Billaud M, van’t Veer LJ, Bernards R
(1993) Proc Natl Acad Sci USA 90:8489–8493.
26. Maheswaran S, Lee H, Sonenshein GE (1994) Mol Cell Biol 14:1147–1152.
27. Hoang AT, Lutterbach B, Lewis BC, Yano T, Chou T-Y, Barrett JF, Raffeld
M, Hann SR, Dang CV (1995) Mol Cell Biol 15:4031–4042.
28. McEwan IJ, Dahlman-Wright K, Ford J, Wright APH (1996) Biochemistry
29. Grandori C, Gomez-Roman N, Felton-Edkins ZA, Ngouenet C, Galloway DA,
Eisenman RN, White RJ (2005) Nat Cell Biol 7:311–318.
30. Stone J, de Lange T, Ramsay G, Jakobovits E, Bishop JM, Varmus H, Lee W
(1987) Mol Cell Biol 7:1697–1709.
31. Winter AG, Sourvinos G, Allison SJ, Tosh K, Scott PH, Spandidos DA, White
RJ (2000) Proc Natl Acad Sci USA 97:12619–12624.
32. McMahon SB, Wood MA, Cole MD (2000) Mol Cell Biol 20:556–562.
33. Halkidou K, Logan IR, Cook S, Neal DE, Robson CN (2004) Nucleic Acids Res
34. Guidez F, Howell L, Isalan M, Cebrat M, Alani RM, Ivins S, Hormaeche I,
McConnell MJ, Pierce S, Cole PA, et al. (2005) Mol Cell Biol 25:5552–5566.
35. Arabi A, Wu S, Ridderstrale K, Bierhoff H, Shiue C, Fatyol K, Fahlen S,
Hydbring P, Soderberg O, Grummt I, et al. (2005) Nat Cell Biol 7:303–310.
36. Berger S (1999) Curr Opin Cell Biol 11:336–341.
37. Bannister AJ, Miska EA (2000) Cell Mol Life Sci 57:1184–1192.
38. Hsieh Y-J, Kundu TK, Wang Z, Kovelman R, Roeder RG (1999) Mol Cell Biol
39. Kundu TK, Wang Z, Roeder RG (1999) Mol Cell Biol 19:1605–1615.
40. White RJ (2002) RNA Polymerase III Transcription (Landes Bioscience, Austin,
41. Burnol A-F, Margottin F, Huet J, Almouzni G, Prioleau M-N, Mechali M,
Sentenac A (1993) Nature 362:475–477.
42. Sutcliffe JE, Brown TRP, Allison SJ, Scott PH, White RJ (2000) Mol Cell Biol
44. Frye M, Watt FM (2006) Curr Biol 16:971–981.
45. Fairley JA, Scott PH, White RJ (2003) EMBO J 22:5841–5850.
46. Innes F, Ramsbottom B, White RJ (2006) Nucleic Acids Res 34:3399–3407.
47. White RJ, Gottlieb TM, Downes CS, Jackson SP (1995) Mol Cell Biol
Spandidos DA, White RJ (2005) Oncogene 24:880–888.
www.pnas.org?cgi?doi?10.1073?pnas.0702909104Kenneth et al.