Molecular Cell, Vol. 7, 1013–1023, May, 2001, Copyright 2001 by Cell Press
Evolutionarily Conserved Interaction
between CstF-64 and PC4 Links Transcription,
Polyadenylation, and Termination
G/U-rich element (Takagaki et al., 1992; MacDonald et
al., 1994; Takagaki and Manley, 1997). CstF-50, which
is characterized by seven WD-40 repeats (Takagaki and
Manley, 1992), has been shown to interact with the
RNAP II CTD (McCracken et al., 1997) and the tumor
suppressor-related protein BARD1 (Kleiman and Man-
ley, 1999). The CstF equivalent in yeast (CFIA) is also a
multimeric complex containing four subunits: Rna14p
(related to CstF-77), Rna15p (related to CstF-64), Pcf11,
and an unidentified 50 kDa polypeptide (Minvielle-
Sebastia et al., 1994; Kessler et al., 1996; Amrani et al.,
1997). A yeast polyadenylation factor with similarity to
CstF-50, Pfs2, has also been described (Ohnacker et
al., 2000). Homologs of the human CstF complex have
been identified in many other organisms (e.g., Hatton et
al., 2000), suggesting that this complex plays an evolu-
tionary conserved role in pre-mRNA polyadenylation.
CstF-64 binds RNA specifically and with high affinity
through its N-terminal RNP-type RNA binding domain
(RBD; Takagaki and Manley, 1997), which is connected
to an unusual C-terminal domain by a so-called hinge
region. The ?300 residue C terminus consists of a long
proline/glycine-rich region (?40%) in which there are 12
tandem copies of the pentapeptide MEARA/G in the
mouse and human proteins and 11 repeats of LEPRG
in chicken CstF-64 (Takagaki et al., 1992, 1996). This
it has the potential to form a long ?-helical structure
within the presumably unstructured proline/glycine-rich
the C-terminal region is unknown. Rna15p, the apparent
yeast homolog of CstF-64, contains an RBD at its N
terminus that closely resembles that of CstF-64 (47%
identity; Takagaki and Manley, 1994). Rna15p has a
stretch of glutamines and asparagines but lacks the
hinge and proline/glycine-rich regions as well as the
idues of Rna15p are evolutionary conserved, displaying
?33% identity between yeast and human proteins.
In the last several years, numerous studies have indi-
quent processing (i.e., capping, splicing, and polyade-
nylation) are intimately connected (reviewed by Hirose
and Manley, 2000; Proudfoot, 2000). In all cases, the
RNAP II CTD plays an important role. The interactions
between the transcription and polyadenylation machin-
eries seem especially extensive. In mammals, it was
found that a fraction of the cell’s CPSF can be isolated
associated with the general transcription factor TFIID,
and in reconstituted transcription reactions, CPSF is
transferred to RNAP II concomitant with transcription
initiation (Dantonel et al., 1997). It has also been shown
that mRNA precursors transcribed by a CTD-truncated
RNAP II are not efficiently polyadenylated in transiently
ciate with RNAP II and bind to the CTD (McCracken et
associate with RNAP II early during transcription, per-
haps at the preinitiation step, and remain attached dur-
ing elongation until 3? processing signals are reached.
Olga Calvo and James L. Manley1
Department of Biological Sciences
New York, New York 10027
Tight connections exist between transcription and
subsequent processing of mRNA precursors, and in-
tion machineries seem especially extensive. Using a
yeast two-hybrid screen to identify factors that inter-
PC4. Both human proteins have yeast homologs,
Rna15p and Sub1p, respectively, and we show that
these two proteins also interact. Given evidence that
certain polyadenylation factors, including Rna15p, are
necessary for termination in yeast, we show that dele-
tion or overexpression of SUB1 suppresses or en-
hances, respectively, both growth and termination de-
fects detected in an rna15 mutant strain. Our findings
provide an additional, unexpected connection between
transcription and polyadenylation and suggest that
PC4/Sub1p, via its interaction with CstF-64/Rna15p,
possesses an evolutionarily conserved antitermina-
tion involving endonucleolytic cleavage of the pre-
mRNA and synthesis of the poly(A) tail. The reaction
requiresa complexmachinery thathas beenextensively
studied in vitro and in vivo. Most of the factors involved
have been identified, and the overall composition of the
complex is conserved from yeast to higher eukaryotes
(reviewed by Keller and Minvielle-Sebastia, 1997; Col-
gan and Manley, 1997; Zhao et al., 1999; Shatkin and
Manley, 2000). In mammalian cells, the components of
the polyadenylation complex include CPSF, cleavage-
binds to a GU-rich sequence downstream of the site of
cleavage; two cleavage factors (CFI and CFII); poly(A)
polymerase (PAP); and additionally, RNA polymerase II
(RNAP II), which participates in the cleavage step via
In yeast, the machinery is equally complex and includes
factors with significant similarity to CPSF, CstF, and
PAP. Allthe yeast componentsare essentialfor viability.
of 77, 64, and 50 kDa. CstF-77 bridges CstF-50 and
with both CPSF and PAP (Murthy and Manley, 1995).
CstF-64 is responsible for RNA binding, recognizing the
Hirose and Manley (1998) extended these results by
showing that the RNAP II CTD can be an essential com-
ponentof the3? processingmachinery inin vitroassays.
Together, this data suggests that polyadenylation is
tightly coupled to transcription, although important de-
tails of regulation and mechanism remain to be eluci-
Consistent with this link between transcription and 3?
end formation, evidence has existed for many years
suggesting that an intact polyadenylation signal is nec-
essary for subsequent transcription termination (re-
viewed in Proudfoot, 1989). This conclusion has been
based on the results of nuclear run-on experiments,
although recent EM visualization of nascent RNAs in
microinjected frog oocytes has provided strong support
for the poly(A) signal-termination linkage (Osheim et al.,
1999). Evidence that polyadenylation factors required
for 3? cleavage are necessary for termination has come
from experiments in yeast showing that cells containing
conditional mutations in severalof these factors, includ-
ing Rna15p, are defective in termination (Birse et al.,
1998). But the mechanism by which 3? cleavage signals
termination remains unknown.
In this paper, we describe experiments that provide
an additional, unexpected link between transcription,
polyadenylation, and termination. Using a yeast two-
hybrid screen to identify factors that interact with the C
terminus of CstF-64, we uncovered an interaction with
a well-studied transcription factor, positive cofactor 4
(PC4) (Ge and Roeder, 1994; Kretzschmar et al., 1994).
In vitro experiments have suggested a complex role for
PC4 in transcription, such that the protein is capable of
functioning as a phosphorylation-regulatable coactiva-
tor and, under some conditions, as a repressor of basal
transcription (Malik et al., 1998; Werten et al., 1998). We
provide evidence that PC4 and CstF-64 can interact in
vitro and, more importantly, are associated in vivo. To
assess the significance of this interaction, we took ad-
vantage of the fact that both proteins have homologs
also interact physically. Importantly, we then show that
hance both growth and termination defects detected in
an rna15 mutant strain. Our results together support a
model in which PC4 possesses an evolutionary con-
served antitermination function mediated by its interac-
tion with CstF-64.
Figure 1. CstF-64 Interacts with PC4 via Its C Terminus
(A) Diagram of plasmids used in the two-hybrid screen (see text for
(B) GST “pull-down” experiments. Recombinant GST-PC4 or GST
alone was bound to glutathione-Sepharose beads and incubated
with HeLa cell nuclear extracts (NE). After extensive washing, pro-
teins bound to the beads (PD), 10% of supernatants (SN), and the
same amount of NE were resolved in a 12% SDS–PAGE gel and
immunobloted with anti-CstF-64 antibody.
(C and D) CstF-64 and PC4 coimmunoprecipitate from HeLa cell
NE. Aliquots of NE (lane 7, [C], or lane 1, [D]) were immunoprecipi-
tated with anti-CstF-64 (lanes 1 and 2, [C]; lanes 2 and 3, [D]), anti-
PC4 (lanes 3 and 4, [C]; lanes 4 and 5, [D]), or anti-actin (lanes 5
and 6, [C]; lanes 6 and 7, [D]) antibodies. NE, immunoprecipitates
(IP), and 10% of supernatants (SN) were resolved by SDS–PAGE,
and proteins were detected by immunobloting with anti-CstF64 ([C])
or anti-PC4 antibody ([D]). Phosphorylated and unphosphorylated
forms of PC4 are indicated.
CstF-64 Interacts with PC4 via Its C Terminus
To isolate proteins that interact with the C terminus
of CstF-64, a yeast two-hybrid screen was performed,
using as bait a 0.8 kb CstF-64 fragment that encodes
1A). A human fetal brain cDNA library was screened,
and analysis of positive colonies revealed among the
strongest interactors PC4. PC4 was initially described
as a 15 kDa protein that enhances activator-dependent
transcription by RNAP II in vitro and appears to bridge
interactions between gene- specific activators and the
Kretzschmar et al., 1994). To identify the region of the
CstF-64 C terminus involved in the interaction with PC4,
we subcloned a fragment of approximately 0.5 kb lack-
ing the last 300 bp (corresponding to the 63 final amino
acids) and used this in conjunction with a PC4 clone
(Figure 1A). This C-terminal truncation failed to interact
with PC4, indicating that the 63 final residues of CstF-
64 are necessary for the interaction.
CstF-64 and PC4 Associate In Vitro and In Vivo
We next wished to determine whether the PC4/CstF-
64 interaction can be detected in vitro. To this end, a
glutathione S-transferase (GST)-PC4 fusion protein was
Transcription, Polyadenylation, and Termination
Ohnacker, M., Barabino, S.M., Preker, P.J., and Keller, W. (2000).
The WD-repeat protein pfs2p bridges two essential factors within
the yeast pre-mRNA 3?-end-processing complex. EMBO J. 19,
Osheim, Y.N., Proudfoot, N.J., and Beyer, A.L. (1999). EM visualiza-
tion of transcription by RNA polymerase II: downstream termination
requires a poly(A) signal but not transcript cleavage. Mol. Cell 3,
Price, D.H. (2000). P-TEFb, a cyclin-dependent kinase controlling
elongation by RNA polymerase II. Mol. Cell. Biol. 20, 2629–2634.
tion in higher eukaryotes. Trends Biochem. Sci. 14, 105–110.
Proudfoot, N. (2000). Connecting transcription to messenger RNA
processing. Trends Biochem. Sci. 25, 290–293.
Richardson, J.M., McMahon, K.W., MacDonald, C.C., and Makha-
tadze, G.I. (1999). MEARA sequence repeat of human CstF-64 poly-
adenylation factor is helical in solution. A spectroscopic and calori-
metric study. Biochemistry 38, 12869–12875.
Schang, L.M., Hwang, G.J., Dynlacht, B.D., Speicher, D.W., Bantly,
A., Schaffer, P.A., Shilatifard, A., Ge, H., and Shiekhattar, R. (2000).
Human PC4 is a substrate-specific inhibitor of RNA polymerase II
phosphorylation. J. Biol. Chem. 275, 6071–6074.
Shatkin, A.J.,and Manley,J.L. (2000).The endsof the affair:capping
and polyadenylation. Nat. Struct. Biol. 7, 838–842.
Sherman, F. (1991). Getting started with yeast. Methods Enzymol.
Takagaki, Y., and Manley, J.L. (1992). A human polyadenylation
factor is a G protein B-subunit homolog. J. Biol. Chem. 267, 23471–
Takagaki, Y., and Manley, J.L. (1994). A polyadenylation factor sub-
unit is the human homolog of the Drosophila suppressor of forked
protein. Nature 372, 471–474.
Takagaki, Y.,and Manley,J.L. (1997). RNArecognition bythe human
polyadenylation factor CstF. Mol. Cell. Biol. 17, 3907–3914.
Takagaki, Y., and Manley, J.L. (2000). Complex protein interactions
within the human polyadenylation machinery identify a novel com-
ponent. Mol. Cell. Biol. 20, 1515–1525.
Takagaki, Y., Manley, J.L., MacDonald, C.C., Wilusz, J., and Shenk,
T. (1990). A multisubunit factor, CstF, is required for polyadenylation
of mammalian pre-mRNAs. Genes Dev. 4, 2112–2120.
Takagaki, Y., MacDonald, C.C., Shenk, T., and Manley, J.L. (1992).
The human 64-kDa polyadenylylation factor contains a ribonucleo-
Natl. Acad. Sci. USA 89, 1403–1407.
Takagaki, Y., Seipelt, R.L., Peterson, M.L., and Manley, J.L. (1996).
Thepolyadenylation factorCstF-64regulates alternativeprocessing
of IgM heavy chain pre-mRNA during B cell differentiation. Cell 87,
Wang, Z., and Roeder, R.G. (1998). DNA topoisomerase I and PC4
and transcription reinitiation by RNA polymerase III. Mol. Cell. 1,
Werten,S., Stelzer,G., Goppelt,A.,Langen, F.M.,Gros, P.,Timmers,
H.T., Van der Vliet, P.C., and Meisterernst, M. (1998). Interaction of
PC4with meltedDNA inhibitstranscription. EMBOJ. 17,5103–5111.
Wu, W.H., Pinto, I., Chen, B.S., and Hampsey, M. (1999). Mutational
analysis of yeast TFIIB. A functional relationship between Ssu72
and Sub1/Tsp1 defined by allele-specific interactions with TFIIB.
Genetics 153, 643–652.
Yankulov, K., and Bentley, D. (1998). Transcriptional control: Tat
cofactors and transcriptional elongation. Curr. Biol. 8, R447–R449.
Zhao, J., Hyman, L., and Moore, C. (1999). Formation of mRNA 3?
ends in eukaryotes: mechanism, regulation, and interrelationships
with other steps in mRNA synthesis. Microbiol. Mol. Biol. Rev. 63,