310
RNA polymerase II holoenzyme and transcriptional regulation
Jack Greenblatt
RNA polymerase
mammalian
II holoenzymes
isolated from yeast and
cells are large, preassembled
complexes
contain some or all of the general transcription
factors and many other polypeptides.
suggest that these holoenzymes
that
initiation
Recent experiments
may mediate alterations
in chromatin structure and play a key role in regulatory
mechanisms
elongation,
that influence transcriptional
RNA processing
initiation,
and transcription
RNA chain
termination.
Addresses
Banting and Best Department of Medical Research, and
Department of Molecular and Medical Genetics, University of
Toronto, 112 College Street, Toronto, Ontario, Canada M5G 1 L6;
e-mail: [email protected]
Current Opinion in Cell Biology 1997, 9:31 O-31 9
http://biomednet.com/elecref/0955067400900310
0 Current Biology Ltd ISSN 0955-0674
Abbreviations
CREB-binding protein
CBP
CDK
cyclin-dependent kinase
cleavage and polyadenylation specificity factor
CPSF
CREB
CAMP response element binding protein
carboxy-terminal heptapeptide repeat domain
CTD
FCPl
TFtlF-associated CTD phosphatase 1
general transcription factor
GTF
HAT
histone acetyltransferase
polll
RNA polymerase II
SNF
sucrose nonfermenting
suppressor of RNA polymerase B
Srb
SWI
switch
TAF
TBP-associated factor
TATA-binding protein
TBP
TFII
transcription factor II
Introduction
Although the general transcription
factors (GTFs) and
RNA polymerase
II (polI1) can be assembled
in vitro
onto a promoter in an ordered, stepwise pathway [l’],
transcriptional
initiation may normally involve promoter
binding by a preassembled
holoenzyme
complex containing ~0111 and many or all of the GTFs that are
essential
for initiation.
A ~0111 holoenzyme
was first
identified
by Richard Young and his colleagues [Z] in
the yeast Saccharomyces cerevisbae as a consequence
of a
hunt for suppressors of a mutation that shortened
the
carboxy-terminal
heptapeptide
repeat domain (CTD) of
the largest subunit of ~0111 and conferred a conditional
growth phenotype.
These suppressor of ma polymerase b
(srb) mutations occurred in genes SRB.2 and SRB4sRBZ1,
whose protein products were found in a holoenzyme
complex containing
TFIIB, TFIIF, TFIIH,
all the Srb
proteins and ~0111 [3,4].
Promoter-specific
transcription
in vitro with crude yeast
nuclear
extracts
requires
at least some of the Srb
proteins
[Z]. Moreover, temperature-shift
experiments
with temperature-sensitive
s&and
sr66 mutants indicated
that some Srb proteins
are needed
for most or all
transcription
by ~0111 in viva [SO]. As Srb proteins were
found exclusively in the holoenzyme complex [3,4], it was
concluded that most or all transcription in yeast necessarily
involves this preassembled
holoenzyme.
In this review, I
shall assume that any genetic or biochemical experiment
dealing with one or more of the Srb proteins necessarily
involves the ~0111 holoenzyme.
An important
role for the yeast ~0111 holoenzyme
in
transcriptional
regulation was first revealed when Roger
Kornberg and his colleagues
[6] purified a complex,
termed the ‘mediator’, that stimulated basal transcription
and was required for transcriptional
activation in vitro.
This mediator copurified with a subfraction of the ~0111,
the holoenzyme,
and could be separated from ~0111 by
passage of the holoenzyme
through a column containing
an immobilized antibody directed against the CTD. The
purified mediator fraction contained TFIIF, several of the
Srb proteins (Srbs 2, 4, 5, 6 and 7), and a number of other
proteins. It did not include Srbs 8, 9, 10, and 11 ([6];
R Kornberg, personal communication),
which are therefore
not required for activation in vitro and whose presence in
the holoenzyme
is still controversial.
Also missing were
TFIIA and the TBP (TATA-binding
protein)-associated
factors (TAF11s) of the TFIID complex which are essential
for activated transcription
in vitro in fractionated
human
and Drosophila systems [7] and in reactions containing
crude yeast nuclear extracts [8]. That at least some Srb
proteins are generally required for transcription
in viva
[So], while the yeast TAFl1s are not [9*,10*] and perhaps
play more specialized roles in transcription
(e.g. in the
transcription of specific genes involved in cell cycle control
[l l]), has emphasized the critical role of the holoenzyme
in transcriptional
regulation. Association of the mediator
with the CTD in the ~0111 holoenzyme
may explain
earlier observations linking the CTD to the efficiency of
transcriptional
activation [12].
Transcription by ~0111 in vjvo can be limited by chromatin
structure [13], recruitment
of the GTFs and ~0111 to
the promoter [11,14-16,17**,18], isomerizations
that occur
in the partly or fully assembled
preinitiation
complex
or
promoter
clearance
and
chain
elongation
[19,20’1,
[21,22]. Activator proteins probably stimulate
transcription by overcoming
one or more of these barriers to
transcription. Recent experiments have suggested that the
holoenzyme
mediates activating and repressing
signals
RNA polymerase II holoenryme
that regulate several stages of the transcription process. For
example, activators may recruit holoenzyme as a complex
to the promoter [17**,18]. The holoenzyme
may also
mediate alterations in chromatin structure that enable the
initiation complex to bind to a promoter [23”,24”].
It is
also likely that the holoenzyme transduces activating and
repressing signals that alter the state of phosphorylation
of the CTD
and consequently
stimulate
or inhibit
promoter
clearance
and chain elongation
downstream
of the initiation
site ([25*]; see below). Finally, the
holoenzyme
may contain a CTD phosphatase
[X-28]
that is important for transcription
termination
and other
proteins involved in RNA processing that are carried along
in association with the CTD in the elongation complex
[29”]. In this article, I shall review the many ways
in which the ~0111 holoenzyme
has been implicated in
transcriptional
regulation recently.
Which GTFs are present in the holoenzyme?
The GTFs present in various holoenzyme
preparations
have varied as a function of the source of the holoenzyme
and the method used in its purification. In the case of
S. cemvisiae, one preparation
contained
TFIIB,
TFIIF,
and TFIIH
[3] while another contained only TFIIF [6].
Moreover, an earlier version of the former preparation also
contained TBP, the TATA box binding subunit of TFIID
[2]. The mammalian holoenzymes have only been purified
during the past year [30*,31,32’]. One preparation, from
human HeLa cells [32*], contained stoichiometric amounts
of TFIIE and TFIIF and a small amount of TFIIH, as
well as the human homologs of Srb7, SrblO (the human
homolog is cyclin-dependent
kinase [CDK]8), and Srbll
(the human homolog is cyclin C). It also contained some
of the proteins that are involved in nucleotide
excision
repair and DNA double strand break repair. Assuming
that these proteins did not copurify fortuitously,
this
raises the interesting
possibility that there are different
holoenzymes
with specialized
functions
(e.g. a special
form of holoenzyme that is involved in initiation or chain
elongation
that contains DNA-repair
proteins). Another
preparation
from calf thymus contained Srb7, but only
substoichiometric
amounts of TFIIE and TFIIH [31].
In each of these cases, an extensive series of columns
was used for the purification.
In contrast, purification
by immunoprecipitation
with an antibody against CDK7,
one of the subunits
of TFIIH,
led to the isolation
from rat liver nuclear extract of a holoenzyme containing
all five of the essential GTFs, including the TBP and
TAFII
components
of the TFIID
complex
[30*]. If
indeed this enzyme was a single complex, which was
not demonstrated,
it was truly a holoenzyme
in the
sense that it was capable of promoter-specific
initiation
on its own. Curiously, only the mammalian, and not the
yeast, holoenzyme
preparations
contained
TFIIE.
One
possibility is that the most abundant form of eukaryotic
holoenzyme
contains all the essential GTFs, and that
various GTFs other than TFIIF, which binds directly and
and transcriptional regulation Greenblatt
311
tightly to ~0111 [33], are weakly bound and can be partially
or entirely lost upon exposure of the complex to elevated
ionic strengths
or various ion exchangers.
Consistent
with this idea, purification
of the human holoenzyme
by a different
single-step
procedure
involving
affinity
chromatography
on a column containing the immobilized
elongation
factor TFIIS also led to the isolation of a
holoenzyme
complex containing
all five essential GTFs
in near-stoichiometric
amounts
(G Pan, J Greenblatt,
unpublished
data). This large 60s complex was similar
in size to the major polII-containing
complex observed
when HeLa nuclear extract was directly sedimented
on a
sucrose density gradient [32*] and was substantially
larger
than the 40s complexes that were purified by multicolumn
procedures [31,32’]. The yield of holoenzyme suggests it
includes about lo-20% of the soluble ~0111 and GTFs
in yeast nuclear extracts [3,4] and at least 510%
of
the soluble ~0111 and GTFs found in HeLa extracts
(G Pan, J Greenblatt, unpublished
data). The existence of
holoenzyme complexes containing all the essential GTFs
raises the possibility
that ~0111 can be recruited to a
promoter in conjunction
with the promoter-recognizing
multisubunit
factor TFIID
and all the other essential
GTFs (Fig. la).
Binding of polymerase to the promoter:
evidence for the involvement of the
holoenzyme
There is strong evidence that recruitment
of the GTFs
and ~0111 to some promoters
in S. cerevisiae is ratelimiting for transcription in v&o [ 11,14-16,17**, 18,341. For
example, tethering TBP or TAFtI
to a yeast promoter
by artificially fusing it to a heterologous
DNA-binding
domain (such as that from the bacterial LexA protein)
leads to a high level of transcription
in the absence of
an activator [14-161. In most studies, neither TBP nor
TFIID is thought to be a component
of the holoenzyme
(3,6,31,32*], but high levels of transcription
approaching
those stimulated
by a strong activator have also been
achieved
by tethering
to a promoter
many proteins
that are more commonly
thought to be components
of the yeast holoenzyme,
including
TFIIB
[ll], Gal11
[17”,18], components
of the SWI/SNF
(switch/sucrose
nonfermenting)
complex [35], Sin4 [36], and Srb9, Srbll,
and Rox3 [37’]. Moreover, recruiting holoenzyme
to a
promoter is sufficient for activation: strong activation was
also achieved when thega/Zlp mutation in the holoenzyme
enabled it to artificially interact with a promoter-bound
portion
of the DNA-binding
domain
of yeast Gal4
[17”,18]. These observations are most simply explained
by assuming that TBP or TFIID can be recruited to the
promoter as a component
of the holoenzyme
(Fig. la).
They also imply that interaction
of an activator with
many, if not most, of the components
of the holoenzyme
could cause the holoenzyme to bind to the promoter and
iead to an elevated level of transcription
(see Fig. 1).
The mediator may be important
for activation in vitro
[6] because it aids recruitment
of the holoenzyme to the
312
Nucleus and gene expression
Figure 1
r‘ossible roles for the poll1 holoenzyme in transcriptional initiation and reinitiation. Parallel lines represent DNA. Gray strand represents elongating
mRNA. TATA, TATA box; +l, transcription initiation site; gray shape surrounding the letter ‘A’, TFlIA; circled letters, transcription factors. (a) An
activator may recruit to the promoter a holoenzyme containing all the essential GTFs, including TFIID. In yeast, this need not include the TAFtt
subunits of TFIID because
TAFtts are not required for most transcription
in S. cerevisiae
[9’,10’1.
An alternative pathway for initiation could
involve the recruitment first of TFHD by the activator (b) and then of the holoenzyme lacking TFtlD (c) 1401. This pathway cannot be essential
in yeast because activation in yeast does not require TAFtts but the recruitment of TFflD to a promoter always seems to require the TAFtts, at
least in vitro [7,38,391. Once poll1 initiates transcription and clears the promoter (d), TFIID would presumably be left behind at the promoter
and there may be two possible pathways for reinitiation. One pathway would involve the binding of holoenqme lacking TFtlD (e) and could
accommodate the observation that recruiting holoenzyme leads to high levels of transcription in yeast [17”,18].
The other pathway would involve
stepwise assembly of factors from the large pools of free poll1 and GTFs Q and would accommodate
higher eukaryotic systems in vitro [20’,41,45].
promoter. The acidic activation domain of the herpes
simplex virus activator protein VP16 binds the yeast
holoenzyme
[4], but the actual target(s) of VP16 in the
holoenzyme complex remains to be determined.
The idea that transcriptional
initiation involves recruitment of holoenzyme to the promoter seems inconsistent
with
evidence for this type of activation in
observations
made in &t-o in the human and
systems. These observations
show that an
activator can cause recruitment
to the promoter DNA
of TFIIA and the promoter-recognizing
factor TFIID,
a process in which the TAFrr subunits of TFIID
are
particularly
important
[38,39].-As
the TAFIIs are not
needed for transcription
activation in yeast [9*,10*], one
Drosophifa
RNA polymerase II holoenzyme and transcriptional regulation Greenblatt
possibility is that transcriptional
activation in yeast and
the higher eukaryotes is fundamentally
different. Another
possibility (Fig. lb,c) is that activation proceeds in two
steps, namely, the recruitment
of TFIID and TFIIA and
then the recruitment
of a holoenzyme
complex lacking
TFIID and TFIIA [40]. An equally plausible and simpler
alternative is that TFIID associates with holoenzyme
in
wivo [2,30*] and that recruitment
of TFIID in vitro reflects
interactions
with the TAFIIs in TFIID
[7] that may
sometimes be important for recruiting holoenzyme in viwo
(Fig. la).
Another important
observation
is that an activator can
cause recruitment
to the promoter of TFIIB, the other
GTFs, and ~0111 when the TFIID concentration
is not
limiting in the reaction [20*,41]. In that case, the activator
may alter the conformation
of promoter-bound
TFIID in
such a way as to facilitate the recruitment
of TFIIB and
the other factors [ 19,20*]. Apparently, proper interaction
of TBP with TFIIB is important for activation [42,43].
This sort of phenomenon
requires a different kind of
explanation, and one possibility is also illustrated in Figure
1. Once holoenzyme
has bound to the promoter and
~0111 has initiated transcription, TFIID and TFIIA would
presumably
remain bound at the promoter. Reassembly
of the preinitiation
complex could then proceed by one
of two possible pathways. One pathway (Fig. le) would
involve binding of a holoenzyme complex lacking TFIID;
the other pathway (Fig. If) would involve stepwise
recruitment of TFIIB, the other GTFs, and ~0111 from the
large pools of unassembled
factors. Both of these pathways
might require continued
interaction of the activator with
TFIID and/or TFIIA [44*]. According to this model, the
observed activator-induced
isomerization
of the TFIID
complex followed by stepwise recruitment
of GTFs and
~0111 in vitro [19,20’,41,45] could reflect one of two
possible pathways for reinitiation in viva.
The holoenzyme
and remodeling
of chromatin
Transcription
in viva is repressed by nucleosomes
[13].
This effect varies in a manner that can depend on nucleosome positioning and the ability of activator proteins and
TFIID to bind nucleosomal DNA. Chromatin structure is
altered when the core histones are acetylated by histone
acetyltransferases
(HATS) [46]. There are also several multisubunit ATP-utilizing
enzymes (e.g. SWI/SNE NURF
[nucleosome-remodeling
factor], and RSC [remodel the
structure
of chromatin])
that can remodel chromatin
structure in vitro [13], and this remodeling can facilitate
the binding of sequence-specific
DNA-binding
proteins to
nucleosomal DNA [47-49]. The SWI/SNF complex is not
essential for the growth of S. cemisiae, although mutational
loss of the complex causes reduced expression of a subset
of yeast genes [SO]. Conversely, the much more abundant
RSC complex is essential in S. cemisiae [Sl’].
How various
to individual
chromatin-modifying
enzymes
genes has been an interesting
are targeted
and largely
313
unresolved question. There is evidence that certain HATS,
like the yeast ADA/GCNS complex and the mammalian
CBP (CREB-binding
protein)/p300
family of coactivator
proteins and their associated cofactor (P/CAF), can be
targeted
to particular
promoters
by interacting
with
DNA-binding
activator proteins [52,.53*]. The discovery
that the TAF112.50 subunit of human TFIID
and the
corresponding Drosophda and yeast TAFIIs contain a HAT
activity [24”] suggests that recruitment
of TFIID to the
promoter on its own or as a component
of holoenzyme
could lead to histone acetylation in the vicinity of that
promoter and thus facilitate the binding of holoenzyme
(Fig. 2).
Figure 2
Chromatin-remodeling enzymes may be recruited, by the
activator, to the promoter together with poll1 holoenzyme. Human
TAFti
and its equivalents in other organisms have histone
acetyltransferase activities [24**]. There is also evidence that the
SWVSNF chromatin-remodeling enzyme [501 may be an associated
substoichiometric component of the holoenzyme [23*‘,51’]. Either
or both of these proteins may be involved in the disassembly of the
nucleosome array that occurs upon recruiting holoenzyme to the
yeast PHO5 promoter [55”1.
The ~0111 holoenzyme
of S. cemisiae may also contain
the chromatin-remodeling
SWI/SNF complex [23**]. Most
of the SWI/SNF complex has been reported to copurify
with the holoenzyme
and co coimmunoprecipitate
with
the holoenzyme
[23”], although holoenzyme purified by
a slightly different
procedure was separated from the
SWI/SNF complex [Sl’]. The presence of the SWI/SNF
complex in the holoenzyme
would be consistent
with
earlier observations
that histone gene suppressors
of
SWI/SNF defects also suppress the effects of CTD truncthere are several thousand
ation [.54]. Nevertheless,
holoenzyme molecules and possibly only 100-200 molecules
314
Nucleus and gene expression
of the SWI/SNF complex per yeast cell [3,4,50,51’]. If so,
SWI/SNF could only be a substoichiometric
component of
the holoenzyme.
Targeting
of holoenzyme
to a promoter via an artificial
interaction between the Gal4 DNA-binding
domain and
the mutant
holoenzyme
component
GalllP
leads to
elimination of positioned nucleosomes at the yeast PHO5
promoter even in the absence of a TATA box [SS”].
Therefore, recruitment
of holoenzyme
to a promoter by
an activator should lead automatically
to alterations
in
chromatin structure that may be necessary for transcription
(see Fig. 2). Curiously, however, perturbation
of the
PHO5 nucleosomes
by recruiting holoenzyme
does not
depend on the SWI/SNF complex unless the enhancer
(upstream activation sequence) is weakened [5.5**]. Therefore, critical chromatin-remodeling
components
of yeast
holoenzyme may yet remain to be identified.
Activating elongation by RNA polymerase
also involves the holoenzyme
II
Although recruiting holoenzyme
to certain promoters is
apparently
sufficient for high levels of transcription
in
S. cemisiae [17**,18], there is important evidence
that
promoter clearance and chain elongation by ~0111 can also
be rate-limiting
for transcription
in U~VOand that these
processes are strongly stimulated by particular activators
[21,22]. For example, there is a transcriptionally
engaged
~0111 molecule that is paused 25-35 nucleotides
downstream of the initiation sites on Drosophda h-p70 (heat-shock
protein 70) genes even in the absence of induction by heat
shock [56]. Loading of the initiation complex onto these
promoters depends on binding sites for the GAGA factor,
while further elongation by ~0111 depends on heat shock
and the heat-shock factor (HSF). In this case, HSF is
regulating promoter clearance by ~0111. Paused polymerase
molecules were also found downstream of the promoter
on many other genes when nuclear run-on experiments
were used to investigate this phenomenon
[22]. Similarly,
the transactivator protein Tat of HIV-l almost exclusively
stimulates chain elongation downstream of position +60 on
HIV-l DNA [57,58*]. Here, polymerase loading depends
on binding sites for the human activator protein Spl in
the HIV-l promoter. Although Spl, GAGA, and CTF
(CCAAT box transcription
factor) activate only initiation
and Tat stimulates only elongation, strong acidic activators
like herpes simplex virus VP16, human ~53 and E2F1,
and yeast Gal4 activate both initiation and elongation by
~0111 [25’,58’,59].
There is increasingly
strong evidence
that activation
of elongation
by ~0111 depends on phosphorylation
of
the CTD.
Firstly, the paused polymerase
molecules
downstream
of the hsp70 promoters have unphosphorylated CTDs, while the elongating polymerase molecules
examined after a heat shock have phosphorylated
CTDs
[60]. Secondly,
DRB (5,6-dichloro-1-B-D-ribofuranosylbenzimidazole),
which inhibits the CTD kinase activity of
the CDK7 subunit of TFIIH, inhibits chain elongation by
~0111, and so does a specific antibody against TFIIH in
microinjected
Xenopus oocytes [61]. Thirdly, a Drosophila
factor, P-TEFb (positive transcription elongation factor b),
which stimulates chain elongation by ~0111 in vitro, is also
a CTD kinase that is inhibited
by DRB [62]. Finally,
the acidic activation
domains and Tat, all of which
stimulate elongation,
all bind TFIIH
[58*,63], and Tat
was recently shown to directly stimulate phosphorylation
of the CTD by TFIIH [64”]. Perhaps phosphorylation
of
the CTD can prevent an RNA-release factor from acting
on ~0111 [65]. Alternatively, it may somehow enable ~0111
to transcribe through nucleosomes
that would otherwise
arrest RNA-chain elongation by ~0111 [66].
Involvement
of the CTD
in regulating chain elongation could implicate the holoenzyme,
whose mediator
subcomplex
is associated
with the CTD
[2-4,6], in
chain elongation.
Recent nuclear
run-on experiments
with S. cmvisiae by David Bentley and his colleagues
have suggested
this may indeed
be the case [25*].
At the GAL1 promoter and at a CYCZ promoter with
upstream Gal4-binding
sites, in the absence of induction
by galactose, there were ~0111 molecules
engaged in
transcription
downstream of the promoters, but only near
the 5’ ends of the genes and not near the 3’ ends.
Induction by galactose substant.ly
increased the number
of polymerase molecules engaged in transcription at only
the 3’ ends of the genes. Therefore, an effect of induction
in these experiments
was to increase the elongation
competency
of promoter-proximal
polymerase molecules.
As transcription
in the absence of inducer was limited
by paused polymerase molecules near the 5 ends of the
genes, these experiments
could not have measured the
effects of the Gal4 activator protein on the frequency of
initiation by ~0111.
Interestingly,
efficient elongation was prevented by CTD
truncation,
by a temperature-sensitive
mutation in the
KIN28 gene, which encodes the cyclin-dependent
CTD
kinase subunit of yeast TFIIH, and by mutations in the
SRBZ and SRBIO genes whose products are hallmarks
of the holoenzyme
[24,6]. These experiments
therefore
imply that efficient elongation in yeast depends on CTD
phosphorylation
and that holoenzyme-specific
polypeptides transduce the elongation-enhancing
signal from the
acidic activator Gal4 (see Fig. 3). The effect of deleting
the SRBZO gene on elongation is particularly interesting
as the SRBIO gene product, like that of the KIN28 gene,
is a cyclin-dependent
kinase that can phosphorylate
the
CTD [67*]. Mutations in the KIN28 and SRBZO genes
each globally reduce phosphorylation
of the CTD in
viva [67’,68]. Whether Gal4 directly stimulates
Kin28
and/or SrblO to phosphorylate
the CTD is not yet clear.
In any case, it would appear that Gal4 can activate
transcription by two different mechanisms, both of which
involve the holoenzyme. Firstly, Gal4 is likely to interact
with a component(s)
of the holoenzyme,
possibly TBP
RNA polymerase
or TFIIB
[69], in order to recruit it to the promoter
(Fig. la); secondly, it is likely to interact with a different
holoenzyme component(s),
probably TFIIH [58*,63&P*],
in order to stimulate CTD phosphorylation
and enhance
promoter clearance or elongation by polymerase (Fig. 3).
In this view, CTD phosphorylation
by SrblO might be
necessary for efficient elongation even if SrblO activity is
not directly regulated by the activator.
Many
other
holoenzyme
including
components,
Galll,
Sin4, Rgrl, Rox3, and Srbll,
are important
for efficient
activation
by Gal4 and other activators
[3,4,6,17”,18,37*,70,71*].
Whether these polypeptides
influence initiation and/or chain elongation is not yet known.
Consistent
with the idea that the mediator may control
elongation, as well as initiation, in V&O, it greatly enhances
phosphorylation
of the CTD by TFIIH in vitro [6].
Transcriptional
holoenzyme
A system
unrelated
repression: another link to the
that represses transcription
of a number
of
genes in S. cemisiae involves the Tupl and
II holoenzyme and transcriptional regulation Greenblatt
315
Ssn6 (Cyc8) proteins, which are targeted to particular
genes (e.g. HO, CYCP and SUCZ) by interacting
with
particular site-specific
DNA-binding
proteins like Migl
(which is involved in glucose repression) and ct2 (which
is involved in a-specific repression in rx cells) (see [37*]
and references
therein). Mutations
that partly prevent
repression in these systems have been identified in genes
that encode a large number of holoenzyme
components,
specifically SIN4, RCRl, GALII, ROX3, and SRBs 8, 9,
20, and 22 [71*]. Remarkably, most of these holoenzyme
components
are also needed for efficient activation by
Ga14. In view of the involvement
in repression of the
cyclin dependent
kinase complex encoded by the SRBlO
and SRBlZ genes [72-741, and the observation that SrblO
is needed for Gal4 to stimulate efficient elongation
itr
viva [ZS’], it seems probable that the Tupl/Ssn6
system
in S. cerevisiae acts, at least in part, to inhibit efficient
elongation by ~0111. As suggested in Figure 3, interaction
of Tupl with a holoenzyme
component
may lead to
inhibition
of SrblO activity, causing target genes to be
transcribed
by an underphosphorylated
polymerase that
elongates poorly [67*].
Figure 3
CID
,
kmse
-
f$
..-
_“_~.__
-_-
A model for the involvement of the poll1 holoenzyme (dark gray shapes plus associated
bound at the promoter)
constitute
in the control of RNA-chain
a multiprotein ‘signal transducer’
by DNA-bound
complexes
RNA processing,
that controls the efficiencies
white tombstone-shaped
and transcription
by Tupl
termination.
mediator-containing
Holoenzyme-specific
in vitroinvolves
polypeptides
complex
polypeptides
both of activation by various activators and of transcriptional
containing yeast Tupl . Activation of initiation (not shown) by holoenzyme
complex [6]. Activation (a) and repression
holoenzyme-associated
elongation,
repression
in the mediator
(b) of elongation in vivainvolve at least some of the mediator proteins and the additional
proteins Srb8, 9, 10 and 11 [5’,37’,71
‘I. This signaling complex containing the Srb proteins and various other proteins
in the tombstone-shaped
diagram controls phosphorylation of the poll1 CTD by SrblO (CDKE) and Kin28 (CDK7) [6.25’] (c) and, therefore,
controls elongation by poll1 (d) L25.1. The holoenzyme initiation complex may also contain CPSF [2g**l, which recognizes the polyadenylation
signal AAUAAA in the nascent RNA (e), and FCPl, a component of a CTD phosphatase (f) that is stimulated by TFllF (arrow from F) and
inhibited earlier in the RNA-elongation process by TFIIB (bar from 8) ([271; J Archambault et al, unpublished data). This model supposes that
CPSF and FCPl are also present in the elongation complex, at least until the RNA sequence AAUAAA is synthesized. Dephosphorylation of
the CTD should lead to termination of transcription further downstream (not shown). The presence in poll1 holoenzyme of CPSF and FCPl
provides a biochemical link between initiation events at the promoter and the control of RNA processing and transcription termination. Ladder
represents DNA; gray line represents mRNA; circled Ps represent phosphorylation. The elongation complex (represented by the two right-hand
polll-containing complexes) is thought not to contain most of the polypeptides that are in the initiator complex (left-hand complex to which the
mediator complex is attached).
316
Nucleus
and gene expression
Does the holoenzyme also integrate initiation
with RNA processing and transcription
termination?
Recent experiments have functionally linked the CTD of
~0111 to RNA splicing and 3’-end formation (i.e. cleavage
and polyadenylation).
Firstly, CTD pepcides and an
anti-CTD antibody inhibit splicing in W&TO[75]. Secondly,
hyperphosphorylated
~0111 was found to be associated
with splicing complexes
assembled
ift vitro and was
partly localized to nuclear ‘speckle’ domains enriched in
splicing factors [76]. Finally, splicing and 3’-end formation
were very inefficient in transfected cells when the RNA
was synthesized by polymerase molecules with truncated
CTDs
[29**]. As the mediator complex is associated
with the CTD [6], involvement
of the CTD in RNA
processing raises the possibility
that particular splicing
and/or cleavage/polyadenylation
factors might interact
either with a form of holoenzyme
that is involved in
elongation or else with the holoenzyme complex at the
promoter and then be carried along in association with the
phosphorylated
CTD during chain elongation (Fig. 3). For
the cleavage and polyadenylation
specificity factors CPSF
and CstF (cleavage stimulatory factor), two observations
are consistent
with either possibility. First, CPSF and
CstF bind to columns containing immobilized unphosphorylated or phosphorylated
CTDs [29**]. Second, human
~0111 holoenzyme
purified by TFIIS-affinity
chromatography also contained some associated CPSF and CstF
[29”]. Although subsequent
experiments
have revealed
that the CPSF and CstF are substoichiometric
(E Wilson,
J Greenblatt, unpublished
data), the holoenzyme may well
integrate transcription
initiation
with RNA processing,
and future experiments
should reveal whether mutations
in genetically
defined subunits of the holoenzyme
can
influence RNA processing in viva.
If CTD phosphorylation
is critical for efficient elongation
[25’,58’,60-63,64**], then dephosphorylation
of the CTD
by a CTD phosphatase may be necessary for termination
of transcription downstream of the polyadenylation
signal.
Transcription
termination
is linked
to the initiation
complex and, by inference, to the holoenzyme
for two
reasons: firstly, the nature of 3’-end formation depends
on the type of promoter utilized by ~0111 [77]; and
secondly, downstream
termination
is functionally
linked
to cleavage and polyadenylation
(which are linked to the
holoenzyme via CPSF) because it fails if the polyadenylation signal is mutated [78] or if 3’-end formation is
inhibited by CTD truncation [29**]. CTD phosphatases
have been purified from HeLa and yeast cell extracts
[Z-B],
and recent experiments
have revealed that an
essential component
of the human CTD phosphatase,
FCPl (TFIIF-associated
CTD phosphatase
l), interacts
with TFIIF
and is present in the human holoenzyme
(J Archambault eta/., unpublished
data). This may explain
why human RNA polymerase II initiates transcription with
an unphosphorylated
CTD. In the elongation complex,
the phosphatase
may be activated to dephosphorylate
the CTD only when the polyadenylation
signal in the
nascent RNA is synthesized and associated with the CPSF
in the transcription
complex to activate cleavage and
polyadenylation
(see Fig. 3).
Conclusions
This has been an exciting year for research that relates
the ~0111 holoenzyme to transcriptional
regulation. Mammalian holoenzymes
were described for the first time
[30*,31,32’], and one of them contained
TFIID
[30*].
New discoveries that TAF11250 has a HAT activity [24”]
and that the chromatin-remodeling
SWI/SNF
complex
may be a substoichiometric
component
of the yeast
holoenzyme [23”] were followed by experiments showing
that recruitment of holoenzyme to the promoter can alter
chromatin structure [SS”]. Initial observations
that the
mediator complex in the yeast holoenzyme was important
for gene regulation [6] were extended by new experiments
suggesting
that the mediator integrates
activating
and
repressing signals [37*,67*,71*] that stimulate or inhibit
both promoter binding
[17”,18] and chain elongation
[25*]. Also, recent experiments have implicated the CTD,
and possibly the holoenzyme,
in the RNA-processing
events that accompany transcription
[29”]. In the year to
come, we may learn precisely which chromatin-remodeling
enzymes
are recruited
with the holoenzyme
to the
promoter and which polypeptides
in the holoenzyme are
important for initiation and chain elongation. We should
also begin to learn how the holoenzyme integrates the activating and repressing signals that it receives from various
regulatory proteins, and better appreciate the relationships
among the holoenzyme, RNA processing, and transcription
termination.
As so many yeast holoenzyme
components
have been identified and genetically characterized
in a
preliminary way, and because of the genetic power of that
system, it is likely that most critical observations on the
holoenzyme
will continue
to emerge from experiments
with S. cemvisiae.
Acknowledgements
I have been supported by the hledical Research
as an hIRC Distinguished Scientist and by grants
the Howard Hughes hledical Institute, and the
of Canada with funds from the Canadian Cancer
for his cornmenu
on this review.
References
and recommended
Council (AIRC) of Canada
from the AIRC of Canada,
National Cancer Instirute
Society. I thank Jim Ingles
reading
Papers of particular interest, published within the annual period of review,
have been highlighted as:
.
l
1.
.
*
of special interest
of outstanding interest
Orphanides G, Lagrange T, Reinberg D: The general
transcription factors of RNA polymerase II. Genes Dev 1996,
10:265?-2663.
This definitive review on the GTFs summarizes the properties of many of the
GTF polypeptides that are components of poll1 holoenzyme.
RNA polymerase II holoenzyme and transcriptional regulation Greenblatt
2.
Thompson CM, Koleske AJ, Chao DM, Young RA: A multisubunit
complex associated withthe RNA polymerase II CTD and
TATA-binding protein in yeast Cell 1993, 73:1361-l 375.
3.
Koleske AJ, Young RA: An RNA polyrnerase II holoenzyme
responsive to activators. Nature 1994, 366~466-469.
4.
Hengartner CJ, Thompson CM, Zhang J, Chao DM, Liao SM,
Koleske AM, Okamura S, Young RA: Association of an activator
with an RNA polymerase II holoenzyme. Genes Dev 1995,
9:897-910.
5.
Thomoson CJ. Youno RA: General reauirement for RNA
poly&erase Ii holo&xymes in vivo. broc Nat/ Acad Sci USA
1995.92:4587-4590.
Temperature-shift experiments with temperature-sensitive mutations in the
SR84 and SRB6 genes show that the products of these’genes are needed
for bulk poly (A)+ mRNA synthesis in S. cerevisiae. This implies that the
mediator in the holoenzyme is generally required for transcription by poll1 in
yeast.
.
6.
Kim Y-J, Bjorklund S, Li Y, Sayre MH, Kornberg RD: A
multfprotein mediator of transcriptional activation and
its interaction with the C-terminal repeat domain of RNA
polymerase II. Cell 1994, 77:599-608.
7.
Goodrich JA, Tjian R: TBP-TAF complexes: selectivity factors
for eukaryotic transcription. Curr Opin Cell Biol 1994,
61403-409.
6.
Reese JC, Apone L, Walker SS, Griffin LA, Green MR: Yeast
TAFIIS in a multisubunit complex required for activated
transcription. Nature 1994, 371:523-527.
9.
Walker SS, Reese JC, Apone LM, Green MR: Transcription
activation in cells lacking TAF&. Nature 1996, 363:185-l
88.
ieast cells with temperature-sensitive TAFjls or expressing particular TAFlls
under the control of the GAL7 promoter are depleted of those TAFlls when
the cells are shifted to high temperature or transferred to growth medium
containing glucose. In these circumstances, transcriptional activation is still
normal for most genes. This observation rationalizes why transcription by the
yeast holoenzyme in vitro can be activated in the absence of TAFlp [6].
10.
.
Moqtaderi Z, Bai Y, Poon D, Wail PA, Struhl K: TBP-associated
factors are not generally required for transcriptional activation
in yeast Nature 1996, 383:188-l
91.
Individual yeast TAFlls under the control of the copper-repressible AN67
promoter and tagged with an amino-terminal recognition signal for ubiquitindependent degradation are destroyed when the amino-terminal recognition
protein is induced with copper. Remarkably, many yeast activators still function normally in the absence of TAFils. TAFlls seem to be more important for
transcription from TATA-less promoters.
11.
12.
Apone LM, Virbasius CA, Reese JC, Green MR: Yeast TAFIISO
is required for cell-cycle progression through Gs/M but
not for general transcription activation. Genes Dev 1996,
10:2368-2380.
Allison LA, lngles CJ: Mutations in RNA polymerase II enhance
or suppress mutations in GAL4. Proc Nat/ Acad Sci USA 1989,
66:2794-2790.
13.
Svaren J, Horz W: Regulation of gene expression by
nucleosomes. Curr Opin Genet Dev 1996, 6:164-l 70.
14.
Klages N, Strubin M: Stimulation of RNA polymerase II
transcription initiation by recruitment of TBP in viva. Nature
1995, 374:822-823.
15.
16.
Chatterjee S, Struhl K: Connecting a promoter-bound protein to
TBP bypasses the need for a transcriptional activation domain.
Nature 1995, 374:820-622.
Xiao H, Friesen JD, Lis JT: Recruiting TATA-binding
protein to
a promoter: transcriptional activation without an upstream
activator. MO/ Ceil Biol 1995,
1 7.
..
16:5757-5761.
Barberis AJ, Pearberg J, Simkovich N, Farrell S, Reinagel P,
Bamdad C, Sigal G, Ptashne M: Contact with a component of
the polymerase II holoenxyme suffices for gene activation. Cell
1995, 61:359-366.
The GALIlP
mutation in the holoenzyme component Gall 1 enables the
mutant protein to bind to amino acids 58-97 of yeast Gai4. In cells with the
GAL7 7P mutation, this portion of Gal4, when tethered to DNA, can function
as a strong ‘activation domain! This argues that recruiting holoenzyme to a
promoter suffices for high levels of transcription in S. cerevisiae. Almost any
holoenzyme component could be the target of a natural activator.
18.
Farrel S, Simkovich N, Wu Y, Barberis A, Ptashne M: Gene
activation by recruitment of the RNA polymerase II
holoenzyme. Genes Dev 1996,10:2359-2367.
19.
317
Horikoshi M, Carey MF, Kakidani H, Roeder RG: Mechanism of
action of a yeast activator: direct effect of GAL4 derivatives
on mammalian TFIID-promoter
interactions. Cell 1988,
64:665-669.
20.
.
Chi T, Carey M: Assembly of the isomerized TFIIA-TFIID-TATA
ternary complex is necessary and sufficient for gene activation.
Genes Dev 1996, 10:2540-2550.
Transient interaction of the Epstein-Barr virus ZEBRA transactivator with
promoter-bound human TFlIA-TFIID complex is shown to stably convert
the ternary complex to an ‘activated form’ that can bind TFIIB, the other
GTFs, and polll. Would recruitment of holoenzyme to a promoter, as in yeast
[17”,161, bypass this isomerization requirement for activation, or does activation in a higher eukaryote require interaction of an activator with TFIID
and/or TFIIA?
21.
Greenblatt J, Nodwell J, Mason S: Transcriptional
Nature 1993, 364:401-406.
22.
Bentley DL: Regulation of transcriptional elongation by RNA
polymerase II. Curr Opin Genet Dev 1995, 6:21 O-21 6.
antitermination.
23.
..
Wilson CJ, Chao DM, lmbalrano AN, Schnitzler GR, Kingston RE,
Young RA: RNA polymerare II holoenzyme contains SWVSNF
regulators involved in chromatin remodeling. Cell 1996,
041235-244.
Copurification and coimmunoprecipitation experiments are used to show that
the SWllSNF chromatin-remodeling enzyme can associate with yeast poll1
holoenzyme, but the claim that SWllSNF is a stoichiometric component of
the holoenzyme or a stably bound component of the mediator has been
contested [51’1. Cooperative interaction of SWllSNF with holoenzyme and
chromatin could lead to chromatin remodeling and DNA binding by TBP
when holoenzyme is recruited to the promoter.
24.
..
Mizzen CA, Yang X-J, Kokubo T, Brownell JE, Bannister Al, OwenHughes T, Workman J, Wang L, Berger SL, Kouzarides T et a/.:
The TAF,,260 subunit of TFIID has histone acetyltransferase
activity. Cell 1996, 87:1261-l
270.
A portion of TAF11250, expressed as a recombinant protein, has HAT activity.
Histone acetylation in nucleosomes around the promoter may be induced
when TFIID is recruited to the promoter on its own as a component of holoenzyme or when it interacts with holoenzyme in the vicinity of the promoter.
25.
Akhtar A. Fave G. Bentlev DL: Distinct activated and
nonactiv&eh RNA poly&erase II complexes in yeast EM50 I
1996.1614654-4664.
Nuclear run-& experiments with yeast are used to show that poll1 that is
bound and transcriptionally engaged at the CYC7 or GAL7 promoter under
the influence of Gal4 does not usually reach the 3’ end of the gene if the
CTD is truncated or in the absence of Srb2 or the CTD kinase activities of
Kin28 or Srbl 0. This implies that some holoenzyme-specific polypeptides
transduce an activating signal that leads to CTD phosphorylation and enables poll1 to efficiently escape the promoter.
.
26.
Chambers RS, Dahmus ME: Purification and characterization of
a phosphatase from HeLa cells which dephosphorylates
the
C-terminal domain of RNA polymerase II. J Biol Chem 1994,
269:26243-26248.
27.
Chambers RS, Wang BQ, Burton ZF, Dahmus ME: The activity
of a COOH-terminal domain phosphatase is regulated
by a docking sita on RNA polymerase II and by the
general transcription factors IIB and IIF. J Biol Chem 1995,
270:14962-l
28.
4969.
Chambers RS, Kane CM: purification and characterization of an
RNA polymerase II phosphatase from yeast J Biol Chem 1996,
271124496-24504.
29.
..
McCracken S, Fong N, Yankulov K, Ballantyne S, Pan G,
Greenblatt J, Patterson SD, Wickens M, Bentley DL: The Cterminal domain of RNA polymerase II couples mRNA
processing to transcription. Nature 1997, 386:357-361.
Transcription of a gene in transfected cells by a poll1 that lacks most of
its CTD leads to defects in mRNA cleavage and polyadenylation, splicing,
and transcription termination. These observations functionally link mRNA processing to the poll1 CTD. Two cleavage and polyadenylation factors, CPSF
and CstF, bind to the CTD. The presence of some CPSF in the poll1 holoenzyme may explain why 3’send formation is linked to initiation [77].
30.
.
O&pow V, Tassan JP, Nigg EA, Schibler U: A mammalian
RNA polymerase II holoenryme containing all components
required for promoter-specific
transcription initiation. Cell
1995, 63:137-l 46.
A mammalian poll1 holoenzyme preparation is described for the first time. It
contains all the essential GTFs, including TFIID, and is capable of promoterspecific transcription in vitro. As the puriiication method involves immunoprecipitation with antiCDK7, the authon are unable to demonstrate that their
rat liver holoenzyme is a single complex.
318
49.
lmbalzano AN, Kwon H, Green MR, Kingston RE: Facilitated
binding of TATA-binding protein to nucleosomal DNA. Nature
1994,370:481-485.
50.
Peterson CL: Multiple SWltches to turn on chromatin?
Opin Genet Dev 1996, 6:171-l 75.
Curr
51.
Nature 1996, 381:86-69.
A purified human poll1 complex or set of similar complexes contains TFIIE,
TFIIF, SRB proteins, a small amount of TFIIH, and many DNA-repair proteins.
This suggests that holoenzyme-specific polypeptides-may link.tran&cription
to DNA repair.
33.
Burton ZF, Killeen, M, Sopta M, Ortolan LG, Greenblatt J:
RAP30/74: a general initiation factor that binds to RNA
polymerase II. MO/ Cell Biol 1988, 8:1602-l 613.
34.
Klein C, Struhl K: Increased recruitment of TATA-binding protein
to the promoter by transcriptional activation domains in vivo.
Science 1994, 288:280-282.
35.
Laurent BC, Treitel MA, Carlson M: Functional interdependence
of the yeast SNF2, SNFS. and SNF8 proteins in transcriptional
activation. Proc Nat/ Acad Sci USA 1991, 88:2687-2691.
36.
Jiang YW, Dohrmann PR, Stillman DJ: Involvement of the SIN4
global transcriptional regulator in the chromatin structure of
Saccharomyces cerevisiae. Mol Cell Biol 1992, 1214503-4514.
37.
Sona W. Treich I. Qian N. Kuchin S. Carlson M: SSN aenes that
affe& transcriptional redression in Saccharomyces cerevisiae
encode SIN4, ROX3, and SRB proteins associated with RNA
polymerase II. MO/ Cell Biol 1996, 16:115-l 20.
Four yeast genes that contribute to glucose repression are shown to encode the holoenzyme components Srb8, Srb9, Sin4 and Rox3. Previous
studies showed that mutations in the SRBf 0 and SRB7 7 genes had similar effects on glucose repression. A strong implication is that holoenzymespecific polypeptides transduce the repressing signal emanating from the
Tupl repressor of this control system.
.
36.
Chi T, Lieberman P, Ellwood K, Carey M: A general mechanism
for transcriptional synergy by eukaryotic activators. Nature
1995, 377:254-257.
39.
Kobayashi N, Boyer TG, Berk Al: A class of activation
domains interacts directly with TFIIA and stimulates
TFIIA-TFIID-promoter
complex assembly. MO/ Cell Biol 1995,
156465-6473.
40.
Stargell LA, Struhl K: Mechanisms of transcriptional activation in
vivo: two steps forward. Fends Gener 1996, 12:31 l-31 5.
41.
Choy B, Green MR: Eukaryotic activators function during
multiple steps of preinitiation complex assembly. Nature 1993,
366531-536.
42.
Kim TK, Hashmoto S, Kelleher RJ, Flanagan PM, Kornberg RD,
Horikoshi M, Roeder RG: Effects of activation-defective TBP
mutations on transcription initiation in yeast Nature 1994,
389:252-255.
43.
Bryant GO, Mattel LS, Burley SK, Berk Al: Radical mutations
reveal TATA-box binding protein surfaces required for
activated transcription in vivo. Genes Dev 1996, IO:2491 -2504.
Cairns BR. Larch Y. Li Y. Zhana M. Lacomis L. ErdiumentBromage l?, Tempsi P, du J, L&e&
B, Kornberg RD: RSC, an
essential, abundant chromatin-remodeling
complex. Cell 1996,
87:1249-l
260.
A new yeast chromatin-remodeling enzyme complex is described that is
about 10 times more abundant than the SWllSNF complex and which, unlike
the SWllSNF complex, is encoded by essential genes. It does not copurify
with poll1 holoenzyme.
.
52.
53.
.
Ogryzko W, Schiltz RL, Russanova V, Howard BH, Nakatani Y:
The transcriptional coactivators ~300 and CBP are histone
acetylbansferases.
Ce// 1996, 87:953-959.
Two related coactivators, ~300 and CBP, that interact with various activator
proteins are shown to have HAT activities. The activators that recruit ~300
and CBP should provoke acetylation by ~300 and CBP of the core histones in the neighboring chromatin. This may facilitate promoter binding by
holoenzyme.
54.
45.
46.
Horikoshi M, Hai T, Lin YS, Green MR, Roeder RG: Transcription
factor ATF interacts with the TATA factor to facilitate
establishment of a preinitiation complex. Ce// 1968,
54:1033-l
042.
Brownell JE, Allis CD: Special HATS for special occasions:
linking histone acetylation to chromatin assembly and gene
activation. Curr Opin Genet Dev 1996, 6:176-l 84.
47.
Cote J, Quinn J, Workman JL, Peterson CL: Stimulation of GAL4
derivative binding to nucleosomal DNA by the yeast SWI/SNF
complex. Science 1994, 265:53-60.
48.
Kwon H, lmbalzano AN, Khavari PA, Kingston RE, Green MR:
Nucleosome disruption and enhancement of activator binding
by a human SWI/SNF complex. Nature 1994, 370:477-461.
Kruger W, Peterson CL, Sil A, Coburn C, Arents G,
Moudrianakis EN, Herskowitz I: Amino acid substitutions in the
structured domains of histones H3 and H4 partially relieve the
requirement of the yeast SWI/SNF complex for transcription.
Genes Dev 1995,9:2770-2779.
55.
..
Gaudreau L, Schmid A, Blaschke D, Ptashne M, Horz W: RNA
polymerase II holoenzyme recruitment is sufficient to remodel
chromatin at the yeast PHOS promoter. Cell 1997, in press.
A fortuitous interaction between the Gall 1 P mutant component of the
holoenzyme and the DNA-binding domain of Gal4 [17**] is used to activate
transcription of the yeast PHO5 gene in the absence of a true activation
domain. This causes elimination of positioned nucleosomes at the PHO5
promoter even when there is no TATA box, indicating that at least one
chromatin-remodeling enzyme must associate with poll1 holoenzyme.
56.
Lis J, Wu C: Protein traffic on the heat shock promoter:
parking, stalling and trucking along. Ce// 1993, 74:1-4.
57.
Kao SY, Calman AF, Luciw PA, Peterlin BM: Anti-termination of
transcription within the long terminal repeat of HIV-1 by tat
gene product Nature 1987, 330:489-493.
58.
.
Blau J, Xiao H, McCracken S, O’Hare P, Greenblatt J, Bentley D:
Three functional classes of transcriptional activation domains.
Mol Cell Bioll996,
16:2044-2055.
Nuclear run-on and nuclease protection experiments are used to show that
Spl and CTF (CCAAT box transcription factor) stimulate only transcription
initiation, HIV Tat stimulates onlv elonaation. and the acidic activation domains ii herpes simplex virus \iP16 &d himan p53 and E2Fl stimulate
both initiation and elongation. Activators that stimulate elongation bind TFIIH.
They may recruit TFIIH to the promoter or stimulate TfllH to phosphorylate
the poll1 CTD.
59.
Yankulov K, Blau J, Purton T, Roberts S, Bentley D:
Transcriptional elongation by RNA polymerase II is stimulated
by transactivatorr
Cell 1994, 77:749-759.
60.
O’Brien T, Hardin S, Greenleaf A, Lis JT: Phosphorylation of
RNA polymerase II C-terminal domain and transcriptional
elongation. Nature 1994, 370~75-77.
61.
Yankulov KY, Pandes M, McCracken S, Bouchard D. Bentley DL:
TFIIH functions in regulating transcriptional elongation by
RNA polymerase II in Xenopus oocytes. MO/ Cell Biol 1996,
16:3291-3299.
62.
Marshall NF, Peng J, Xie Z, Price DH: Control of RNA polymerare
II elongation potential by a novel carboxyl-terminal domain
kinase. J Biol Chem 1996,271:27176-27183.
63.
Xiao H, Pearson A, Coulombe B, Truant R, Zh%g S, Regier J,
Triezenberg S, Reinberg D, Flores 0, lngles C, Greenblatt J:
Binding of basal transcription factor TFIIH to the acidic
activation domains of VP1 6 and ~53. Mel Cell Bioll994,
14:7013-7024.
64.
..
Parada CA, Roeder RG: Enhanced processivity of RNA
polymerase II triggered by Tat-induced phosphorylatton
carboxy-terminal domain. Nature 1996, 384:375-378.
44.
.
Ho SN, Biggar SR, Spencer DM, Schreiber SL, Crabree GR:
Dimeric ligands define a role for transcriptional activation
domains in reinitiation. Nature 1996, 382:622-626.
A lipid-soluble binary ligand molecule capable of mediating the association
of a DNA-binding domain and an activation domain is used for activating
transcription. Use of a monomeric ligand to disrupt the interaction shows that
continued function of the activation domain is needed for continued activated
transcription in yeast and mammalian cells.
Berger SL, Pina B, Silverman N, Marcus GA, Agapite J, Regier JL,
Triezenberg SJ, Guarente L: Genetic isolation of ADA2: a
potential transcriptional adaptor required for function of certain
acidic activation domains. Cell 1992, 70:251-265.
of its
RNA polymerase
HIV Tat stimulates purified TFIIH to phosphorylate the
Dfosophiia.
271 :11043-l
66.
and transcriptional
regulation
Greenblatt
319
The yeast mediator complex is shown to contain Sin4 and Rgrl proteins,
which were implicated previously in activation and repression. Sin4 and Rgrl
are present in a subcomplex containing Gall 1 and a 50 kDa polypeptide.
Therefore, the mediator is a ‘signal transducer’ that mediates responses to
both activators and repressors.
j Biol Chem 1996,
72.
Kuchin S, Yeghiayan P, Carlson M: Cyclin-dependent
protein
kinase and cyclin homotogs SSN3 and SSN8 contribute to
transcriptional control in yeast Proc Nat/ Acad Sci USA 1995.
92:4006-4010.
73.
Wahi M, Johnson AD: Identification of genes required for alpha
2 repression in S8ccharo1nyces cerevisiee. Genetics 1995,
140:79-90.
74.
Balciunas D, Ronne H: Three subunits of the RNA polymerase II
mediator complex are involved in glucose repression. Nucleic
Acids Res 1995, 23:4421-4425.
75.
Yuryev A, Patturajan M, Litingtung Y, Joshi RV, Gentile C,
Gebara M, Corden JL: The C-terminal domain of the largest
subunit of RNA polymarase II interacts with a novel set of
serinejarginine-rich
proteins. Proc Nat/ Acad Sci USA 1996,
93:6975-6960.
76.
Mortillaro MJ, Blencowe BJ, Wei X, Nakayasu H, Du L, Warren SL,
Sharp PA, Berezney R: A hyperphosphorylated
form of the
large subunit of RNA polyrnerase II is associated with splicing
complexes and the nuclear matrix. Proc Nat/ Acad SC; USA
1996,93:6253-8257.
77.
Hemandez N, Weiner AM: Formation of the 3’ end of Ul snRNA
requires compatible snRNA promoter elements. Cell 1966.
47~249-250.
70.
Whitelaw E, Proudfoot N: Alpha-thelassaemia
caused by a poly
(A) site mutation reveals that transcriptional termination is
linked to 3’ end processing in the human alpha 2 globin gene.
EMBO J 1966, 5:2915-2922.
1046.
lzban MG, Luse DS: Transcription on nucleosomal templates
by RNA polymerase II in vitro: inhibition of elongation with
enhancement of sequence-specific
pausing. Genes Dev 1991,
5:663-696.
67.
.
Liao SM, Zhang J, Jeffery DA, Koleske Al, Thompson CM,
Chao DM, Viljoen M, Van Vuuren HJ, Young RA: A kinase-cyclin
pair in the RNA polymerase II holoenzyme. Nature 1995,
374:193-l 96.
Two polypeptides that copurii with the yeast poll1 holoenzyme constitute a
kinase-cyclin pair. These polypeptides are Srbl 0 (CDKB) and Srbl 1 (cyclin
C) and they can phosphorylate the poll1 CTD in vifro. Cells lacking SrblO
have defects in activation and repression and contain poll1 with an underphosphorylated CTD.
66.
II holoenzyme
Valay J-G, Simon M, Dubois M-F, Bensaude 0, Facca C, Faya G:
The KIN28 gene is required for RNA polymerase II-mediated
transcription and phosphorylation of the Rpblp CTD. J MO/ Biol
1996, 24:535-544.
69.
Wu Y, Reece RJ, Ptashne M: Quantitatton of putative
activator-target
affinities predicts transcriptional activating
potentials. EMBO J 1996, 153951-3963.
70.
Bjorklund S, Kim Y-J: Mediator of transcriptional
Trends Biochem Sci 1996,21:335-337.
71.
.
Li Y, Bjorklund S, Jiang YW, Kim YJ, Lane WS, Stillman DJ,
Kornberg RD: Yeast global transcriptional regulators Sin4 and
Rgrl are components of mediator complex/RNA polymerase II
holoenzyme. Proc Nat/ Acad Sci USA 1995, 92:10664-l
0666.
regulation.