Distinct activities of the DExD/H-box
splicing factor hUAP56 facilitate
stepwise assembly of the spliceosome
Haihong Shen,1,5Xuexiu Zheng,1Jingping Shen,2Lingdi Zhang,2Rui Zhao,2and
Michael R. Green3,4
1Department of Life Science, Gwangju Institute of Science and Technology, Gwangju 500-712, Korea;2Department of
Biochemistry and Molecular Genetics, University of Colorado Denver, Anschutz Medical Campus, Aurora, Colorado 80045,
USA;3Howard Hughes Medical Institute and Programs in Gene Function and Expression and Molecular Medicine,
University of Massachusetts Medical School, Worcester, Massachusetts 01605, USA
The essential splicing factor human UAP56 (hUAP56) is a DExD/H-box protein known to promote
prespliceosome assembly. Here, using a series of hUAP56 mutants that are defective for ATP-binding, ATP
hydrolysis, or dsRNA unwindase/helicase activity, we assess the relative contributions of these biochemical
functions to pre-mRNA splicing. We show that prespliceosome assembly requires hUAP56’s ATP-binding and
ATPase activities, which, unexpectedly, are required for hUAP56 to interact with U2AF65and be recruited
into splicing complexes. Surprisingly, we find that hUAP56 is also required for mature spliceosome assembly,
which requires, in addition to the ATP-binding and ATPase activities, hUAP56’s dsRNA unwindase/helicase
activity. We demonstrate that hUAP56 directly contacts U4 and U6 snRNAs and can promote unwinding of
the U4/U6 duplex, and that both these activities are dependent on U2AF65. Our results indicate that hUAP56
first interacts with U2AF65in an ATP-dependent manner, and subsequently with U4/U6 snRNAs to facilitate
stepwise assembly of the spliceosome.
[Keywords: hUAP56; splicing; prespliceosome assembly; spliceosome assembly; ATP binding; ATPase; RNA
Received January 30, 2008; revised version accepted May 9, 2008.
Pre-mRNA splicing occurs in a ribonucleoprotein (RNP)
complex called the spliceosome, which comprises a large
number of proteins and several U small nuclear RNP
particles (snRNPs) termed U1, U2, U4, U5, and U6. As-
sembly of the spliceosome on the pre-mRNA substrate
proceeds in a stepwise manner and involves the recogni-
tion of intron-defining splice signals (for review, see Her-
tel and Graveley 2005). Spliceosome assembly is initi-
ated by the formation of complex E, in which U1 snRNP
is stably associated with the 5? splice site, SF1 is associ-
ated with the branchpoint, and U2 snRNP auxiliary fac-
tor (U2AF) subunits U2AF65and U2AF35are associated
with the polypyrimidine (Py)-tract and 3? splice site, re-
spectively. Subsequently, SF1 is replaced by U2 snRNP
at the branchpoint, leading to the formation of complex
A (the prespliceosome). Mature spliceosome assembly
occurs upon entry of the U4/U6 ? U5 tri-snRNP to form
complex B, followed by structural rearrangements to
form the catalytically active complex C, in which U2
and U6 snRNAs interact, and U6 replaces U1 at the 5?
Many steps in spliceosome assembly require ATP hy-
drolysis and are mediated by a series of splicing factors
that are members of the DExD/H-box protein family, the
founding member of which is eIF-4A, a known RNA he-
licase (for review, see Staley and Guthrie 1998; Silver-
man et al. 2003; Rocak and Linder 2004). Several of these
DExD/H-box splicing factors have been shown to pos-
sess an ATP-dependent RNA unwinding/helicase activ-
ity (Laggerbauer et al. 1998; Raghunathan and Guthrie
1998; Wagner et al. 1998; Wang et al. 1998) and are
thought to use ATP hydrolysis as a driving force to
modulate specific RNA structural rearrangements dur-
ing spliceosome assembly. DExD/H-box proteins typi-
cally have a series of conserved sequence motifs; struc-
tural, mutational, and biochemical analyses have sug-
gested roles for these motifs in ATP-binding, ATP
hydrolysis (ATPase), RNA-binding, and dsRNA unwind-
ing/helicase activities (for review, see Rocak and Linder
2004; Cordin et al. 2006; Linder 2006).
We originally identified a human 56-kDa DExD/H-box
protein, U2AF65-Associated Protein (hUAP56), in a two-
hybrid screen for proteins that interact with U2AF65
4E-MAIL email@example.com; FAX: (508) 856-5473.
5E-MAIL firstname.lastname@example.org; FAX: 82-62-970-2484.
Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.1657308.
1796 GENES & DEVELOPMENT 22:1796–1803 © 2008 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/08; www.genesdev.org
(Fleckner et al. 1997). Subsequent studies in Drosophila,
Xenopus, and yeast have shown that UAP56 is also in-
volved in other aspects of RNA metabolism, including
general mRNA export from the nucleus (Gatfield et al.
2001; Luo et al. 2001; Herold et al. 2003) and cytoplasmic
mRNA localization (Meignin and Davis 2008).
hUAP56 is an essential splicing factor that is required
for the U2 snRNP–branchpoint interaction during pre-
spliceosome assembly. Recruitment of hUAP56 to the
pre-mRNA is dependent on the Py-tract and U2AF65; the
interaction with U2AF65presumably directs hUAP56 to
the vicinity of the branchpoint. The detailed mechanism
by which hUAP56 facilitates splicing and complex as-
sembly, however, is not known. Here we study the role
of hUAP56 in spliceosome assembly through analysis of
hUAP56 derivatives bearing mutations that selectively
interfere with DExD/H-box biochemical functions. Our
results reveal a new role for ATP hydrolysis in the func-
tion of a DExD/H-box protein.
Requirement of the hUAP56 ATP-binding, ATPase,
and dsRNA unwindase/helicase activities for in vitro
To analyze the role of hUAP56 in spliceosome assembly,
we developed a biochemical complementation system.
We raised a polyclonal ?-hUAP56 antibody and used it to
immunodeplete hUAP56 from HeLa nuclear extract (NE;
?UAP56NE) (Fig. 1A). As expected, splicing of an Ad ML
pre-mRNA substrate did not occur in the ?UAP56NE
(Fig. 1B, lane 2); however, addition of purified, recombi-
nant wild-type hUAP56 to the ?UAP56NE restored
splicing to levels approaching that observed in HeLa NE
(Fig. 1B, cf. lanes 3 and 1). Similarly, splicing of another,
unrelated pre-mRNA substrate, human ?-globin, did not
occur in the ?UAP56NE, but was restored following ad-
dition of hUAP56 (Fig. 1C).
To study the role of hUAP56 in pre-mRNA splicing,
we analyzed a previously characterized series of hUAP56
mutant derivatives (Shen et al. 2007). The hUAP56 mu-
tants K95A and E197A are defective for ATPase activity
and, consequently, also for dsRNA unwinding/helicase
activity. Based on the structures of several DExD/H-box
proteins, the K95 residue is proposed to be important for
ATP-binding (Benz et al. 1999; Caruthers and McKay
2002; Shi et al. 2004; Sengoku et al. 2006), whereas E197
is thought to be a key catalytic residue involved in ATP
hydrolysis (Caruthers and McKay 2002; Cordin et al.
2006; Sengoku et al. 2006). The hUAP56 mutant D199A
retains ATPase activity but is defective for dsRNA un-
winding/helicase activity (Shen et al. 2007). For conve-
nience, we will refer to these derivatives as an ATP-
binding mutant (K95A), ATPase mutant (E197A), and
dsRNA unwinding/helicase mutant (D199A).
All three mutant proteins were expressed in and puri-
fied from Escherichia coli. The overall structures of the
three mutant proteins were intact, as evidenced by gel
filtration elution profiles (Fig. 2A) and thermal melting
curves (Fig. 2B) that were similar to those of the wild-
type hUAP56 protein. We assessed the ability of each
mutant to support splicing of the Ad ML pre-mRNA sub-
strate in the ?UAP56NE. The results of Figure 2C show
that following addition to the ?UAP56NE, all three
hUAP56 mutants failed to support splicing.
hUAP56 functions during both prespliceosome
and spliceosome assembly
We next analyzed the ability of the three hUAP56 mu-
tants to support splicing complex assembly. Figure 3A
shows that the ?UAP56NE failed to support formation of
the prespliceosome (complex A) or the mature spliceo-
some (complex B/C). This result is consistent with our
original finding that hUAP56 is required for formation of
the prespliceosome (Fleckner et al. 1997), a precursor to
the mature spliceosome. Figure 3A also shows that the
three hUAP56 mutants differentially affected spliceo-
some assembly. The ATP-binding (K95A) and ATPase
(E197A) mutants failed to support assembly of the pre-
spliceosome and mature spliceosome. In contrast, the
dsRNA unwinding/helicase mutant (D199A) supported
assembly of the prespliceosome but not the mature
spliceosome. As expected, formation of complex A in the
presence of the dsRNA unwinding/helicase mutant re-
quired ATP and did not occur following RNAse H-di-
rected cleavage of U2 snRNA (Fig. 3B). hUAP56 was not
required for early complex (complex E) assembly (Fig.
3C), consistent with the fact that formation of complex
system to analyze the role of hUAP56 in spliceosome assembly.
(A) Immunoblot analysis showing protein levels of hUAP56
and, as a specificity control, U2AF65, in mock-depleted HeLa
NE and in hUAP56-depleted HeLa NE (?UAP56NE). (B)
hUAP56 was analyzed for its ability to complement splicing of
the Ad ML pre-mRNA substrate in the ?UAP56NE. Splicing of
Ad ML in mock-depleted HeLa NE is shown as a control. The
signal corresponding to fully spliced pre-mRNA was quanti-
tated; for NE, the value was arbitrarily set to 100%, and for
?UAP56NE and ?UAP56NE+hUAP56 measured 2% and 64%,
respectively. (C) Splicing of the ?-globin pre-mRNA substrate
following addition of hUAP56 to the ?UAP56NE. Quantitation
of the fully spliced pre-mRNA signal was as follows: NE, 100%;
?UAP56NE, 3%; ?UAP56NE+hUAP56, 63%.
Development of a biochemical complementation
Role of hUAP56 in spliceosome assembly
GENES & DEVELOPMENT1797
E is an ATP-independent step (for review, see Hertel and
Graveley 2005). Collectively, these results demonstrate
that hUAP56 acts during assembly of both the prespli-
ceosome and the mature spliceosome.
Role of hUAP56 in prespliceosome assembly
We previously found that hUAP56 is recruited to the
pre-mRNA through an interaction with Py-tract-bound
U2AF65(Fleckner et al. 1997). Thus, a possible explana-
tion for the failure of the hUAP56 mutants to support
spliceosome assembly was an inability to interact with
U2AF65and thus be recruited into splicing complexes.
As a first test of this hypothesis, wild-type hUAP56 or
one of the three hUAP56 mutants were added to the
?UAP56NE in the presence of a biotinylated pre-mRNA
substrate, and splicing complexes were then affinity-pu-
rified and analyzed by immunoblotting for hUAP56. The
results of Figure 4A show that only wild-type hUAP56
and the hUAP56 dsRNA unwinding/helicase mutant
were recruited into splicing complexes. Thus, the inabil-
ity of the ATP-binding and ATPase mutants to support
prespliceosome assembly results from a failure to enter
into splicing complexes.
We next asked whether the failure of the hUAP56
ATP-binding and ATPase mutants to enter into splicing
complexes was due to an inability to interact with
U2AF65. Wild-type hUAP56 or an hUAP56 mutant was
incubated under splicing conditions with purified GST-
U2AF65, and, following purification on glutathione aga-
rose, bound hUAP56 was detected by immunoblotting.
The results of Figure 4B show, as predicted from our
previous study (Fleckner et al. 1997), that U2AF65and
hUAP56 directly interact. Significantly, neither the
U2AF65, explaining their failure to enter into splicing
The inability of the hUAP56 ATP-binding and ATPase
mutants to interact with U2AF65suggested that the
U2AF65–hUAP56 interaction was ATP-dependent. In
support of this prediction, the results of Figure 4B show
that hUAP56 failed to interact with GST-U2AF65in the
absence of ATP. Furthermore, Figure 4C shows that the
U2AF65–hUAP56 interaction did not occur when ATP
was replaced by either the non-hydrolyzable ATP analog
spliceosome assembly. (A) Wild-type hUAP56 and mutant de-
rivatives were analyzed for their ability to support assembly of
the prespliceosome (complex A) and mature spliceosome (com-
plexes B and C) in the ?UAP56NE. Also shown is assembly of
the nonspecific complex H. (B) The D199A mutant was ana-
lyzed for its ability rescue complex A formation in the presence
and absence of U2 snRNA and ATP. (C) Mutant hUAP56 de-
rivatives were analyzed for their ability to support complex E
assembly. U2AF-depleted and mock-depleted nuclear extracts
were analyzed as negative and positive controls, respectively.
hUAP56 functions during both prespliceosome and
ATPase, and dsRNA unwindase/helicase domains for in
vitro pre-mRNA splicing. (A) Gel filtration profiles of
wild-type hUAP56 or the ATP-binding (K95A), ATPase
(E197A), or dsRNA unwindase/helicase (D199A) mu-
tants. (mAU) milliabsorption unit. (B) Thermal melting
curves for wild-type hUAP56 and mutant derivatives, as
monitored by circular dichroism spectroscopy. The melt-
ing temperatures (Tm) for wild-type, K95A, E197A, and
D199A hUAP56 are 55°C, 56°C, 52°C, and 53°C, respec-
tively. (C) Wild-type hUAP56 or mutant derivatives were
analyzed for their ability to support splicing of the Ad ML
pre-mRNA substrate in the ?UAP56NE.
Requirement of the hUAP56 ATP-binding,
Shen et al.
1798 GENES & DEVELOPMENT
adenosine-5?-O-(3-thio)triphosphate (ATP-?S), ADP, or
GTP, confirming a requirement for ATP hydrolysis. Col-
lectively, these results indicate that ATP-binding and
hydrolysis are required for the U2AF65–hUAP56 interac-
tion and the subsequent recruitment of hUAP56 into
hUAP56 contacts U4 and U6 snRNAs
We next analyzed the role of hUAP56 in mature splice-
osome assembly. We considered the possibility that
hUAP56 promoted mature spliceosome assembly by di-
rectly contacting and recruiting a mature spliceosomal
component, such as U4/U6 or U5 snRNP. To test this
idea, we analyzed the potential association of various U
snRNAs with hUAP56 using a UV cross-linking/immu-
noprecipitation assay. Following UV irradiation of HeLa
NE to induce RNA–protein cross-links, hUAP56 was im-
munoprecipitated, and snRNAs in the immunoprecipi-
tate were purified and detected by primer-extension
analysis. The results of Figure 5A show that hUAP56
specifically contacted U4 and U6 snRNAs. We tested the
ability of the three hUAP56 mutants to interact with U4
and U6 snRNAs. The results of Figure 5B show that only
the dsRNA unwinding/helicase mutant (D199A) re-
tained the ability to contact U4 and U6 snRNAs.
The identical behaviors of each of the three hUAP56
mutants in the hUAP56–U2AF65(Fig. 4B) and hUAP56–
U4/U6 snRNA (Fig. 5B) interaction assays raised the pos-
sibility that U2AF65was required for hUAP56 to contact
U4/U6 snRNAs. Consistent with this idea, Figure 5C
shows that the interaction of hUAP56 with U4 and U6
snRNAs was ATP-dependent. To directly test the role of
hUAP56 and U2AF65were added to HeLa NEs that had
been depleted of either hUAP56, U2AF, or both hUAP56
and U2AF, and association of hUAP56 with U4 and U6
snRNAs was analyzed by the UV cross-linking/immu-
noprecipitation assay. The results of Figure 5C demon-
strate that the interaction of hUAP56 with U4 and U6
snRNAs required both hUAP56 and U2AF65. Notably,
the hUAP56–U4/U6 snRNA interaction occurred in the
absence of added pre-mRNA substrate (Fig. 5D), indicat-
ing that the requirement for U2AF65was independent of
its ability to recruit hUAP56 into spliceosomal com-
hUAP56 can unwind the U4/U6 duplex
Like other DExD/H-box proteins, hUAP56 can unwind a
short, artificial RNA duplex in an ATP-dependent man-
ner (Shen et al. 2007). The finding that hUAP56 con-
tacted U4 and U6 snRNAs raised the possibility that
hUAP56 could mediate unwinding of the natural U4/U6
duplex in U4/U6 snRNP. As an initial test of this idea,
we used a previously described native gel assay that mea-
Primer-extension analysis in untreated HeLa NE shows the po-
sitions of U snRNAs. (Right) Following UV irradiation, hUAP56
was immunoprecipitated, and snRNAs in the immunoprecipi-
tate were purified and detected by primer-extension analysis. (B)
Wild-type hUAP56 or mutant derivatives were added to the
?UAP56NE, and association with U4 and U6 snRNAs was ana-
lyzed by UV cross-linking/immunoprecipitation. (C) Recombi-
nant hUAP56, U2AF65, or both hUAP56 and U2AF65were
added to HeLa NEs that had been depleted of either hUAP56,
U2AF, or both hUAP56 and U2AF, and association of hUAP56
with U4 and U6 snRNAs was analyzed by UV cross-linking/
immunoprecipitation. (D) The ability of hUAP56 to bind U4
and U6 shRNAs was tested in the presence and absence of an
unlabeled Ad ML pre-mRNA.
hUAP56 contacts U4 and U6 snRNAs. U snRNA
hUAP56 recruitment experiments. Wild-type hUAP56 or mu-
tant derivatives were added to the ?UAP56NE, followed by ad-
dition of a biotinylated Ad ML pre-mRNA substrate. Splicing
complexes were affinity-purified and analyzed by immunoblot-
ting for hUAP56. (B) GST pull-down experiments. hUAP56 or
mutant derivatives were incubated with GST-U2AF65and, fol-
lowing GST pull-down, bound hUAP56 was detected by immu-
noblotting. Protein–protein interaction experiments were per-
formed in the presence or absence of ATP, as indicated. (C) GST
pull-down experiments were performed as described in B, ex-
cept that ATP was replaced by either the non-hydrolyzable ATP
analog ATP-?S, ADP, or GTP.
Role of hUAP56 in prespliceosome assembly. (A)
Role of hUAP56 in spliceosome assembly
GENES & DEVELOPMENT 1799
sures the ratio of free and duplex forms of U4 and U6
snRNAs (Laggerbauer et al. 1998). In brief, total RNA
was purified and fractionated on a native gel to separate
free U4 and U6 snRNAs from the U4/U6 duplex. The
results of Figure 6A show that following incubation of a
standard HeLa NE under splicing conditions, very little
U4/U6 duplex was detectable. When the experiment was
repeated in the ?UAP56NE, a substantially higher level
of U4/U6 duplex was observed. Addition of wild-type
hUAP56 to the ?UAP56NE markedly decreased the
amount of U4/U6 duplex, and this effect was ATP-de-
pendent. Significantly, none of the three hUAP56 mu-
tants was able to support the ATP-dependent decrease in
the level of U4/U6 duplex. Identical results were ob-
tained in the presence or absence of an unlabeled Ad ML
pre-mRNA substrate (Fig. 6A).
To verify that hUAP56 promoted unwinding of the
U4/U6 duplex, we also performed a series of psoralen
cross-linking experiments (Wassarman and Steitz 1992;
Eperon et al. 2000; Zhu et al. 2003). In brief, HeLa NEs
containing psoralen were irradiated with UV light to in-
duce RNA–RNA cross-links, and the RNA products
were purified and fractionated on a denaturing polyacryl-
amide gel. The identities of the U4/U6 snRNA cross-
links were confirmed by Northern blot analysis. Consis-
tent with the results of the native gel assay, the psoralen
cross-linking results of Figure 6B show that a substan-
tially higher level of U4/U6 duplex was observed in the
?UAP56NE compared to HeLa NE. Moreover, addition
of wild-type hUAP56, but not any of the three hUAP56
mutants, to the ?UAP56NE markedly decreased the
level of U4/U6 duplex. Collectively, the results of Fig-
ures 5 and 6 show that hUAP56 contacts U4 and U6
snRNAs, and promotes unwinding of the U4/U6 duplex,
which requires all three hUAP56 biochemical functions.
Finally, we asked whether U2AF65affected the ability
of hUAP56 to unwind the U4/U6 duplex. Recombinant
hUAP56, U2AF65, or both hUAP56 and U2AF65were
added to HeLa NEs that had been depleted of either
hUAP56, U2AF, or both hUAP56 and U2AF, and un-
winding of the U4/U6 duplex was analyzed by the pso-
ralen cross-linking assay. Consistent with the results of
the hUAP56–U4/U6 snRNA interaction assay (Fig. 5C),
Figure 6C shows that unwinding of the U4/U6 duplex by
hUAP56 requires U2AF65.
In this study, we show that hUAP56 has multiple and
surprisingly diverse roles in splicing complex assembly.
Our major conclusions are summarized in Figure 6D and
discussed below. We had previously found that hUAP56
was required for prespliceosome assembly (Fleckner et
al. 1997). Here we confirm this result and find, unexpect-
edly, that hUAP56 is also required for the conversion
of the prespliceosome to the mature spliceosome. Pre-
spliceosome assembly is dependent on the hUAP56
ATP-binding and ATPase activities, which are required
for interaction with U2AF65and recruitment into splic-
ing complexes. Mature spliceosome assembly requires,
presence or absence of Ad ML pre-mRNA, and fractionated on a native gel to separate free U4 and U6 snRNAs from the U4/U6 duplex.
Gels were probed with either a U4 (top) or U6 (bottom) probe. The ratio of U4/U6 duplex to free U4 or U6 was quantitated and is
shown. (B) Psoralen cross-linking analysis. Assays were performed in the ?UAP56NE following addition of wild-type hUAP56 or
mutant derivates. (C) hUAP56 was depleted from U2AF-depleted extracts, and psoralen cross-linking experiments were performed
following addition of hUAP56, U2AF65, or a combination of hUAP56 and U2AF65. (D) Schematic diagram of the hUAP56 protein,
showing the residues required for ATP-binding, ATPase, and dsRNA unwindase/helicase activities and their role in prespliceosome
and mature spliceosome assembly.
hUAP56 can unwind the U4/U6 duplex. (A) Native gel assay. Total RNA was purified from extracts, incubated in the
Shen et al.
1800 GENES & DEVELOPMENT
in addition to the ATP-binding and ATPase activities,
the dsRNA unwindase/helicase activity. We show that
hUAP56 directly contacts U4 and U6 snRNAs and pro-
motes unwinding of the U4/U6 duplex, and that both of
these activities require U2AF65. Thus, U2AF65is re-
quired to confer essential specificity to hUAP56 during
both prespliceosome assembly, in which it recruits
hUAP56 to the pre-mRNA branchpoint, and spliceo-
some assembly, in which it directs hUAP56 to contact
U4/U6 snRNAs. Collectively, our results indicate that
hUAP56 facilitates multiple steps of spliceosome assem-
bly through distinct mechanisms.
hUAP56 mediates unwinding of the U4/U6 duplex
We found that during mature spliceosome assembly
hUAP56 directly contacts U4 and U6 snRNAs, and pro-
vide evidence that hUAP56 can promote unwinding of
the U4/U6 duplex. We note that previous studies have
implicated other DExD/H-box proteins, in particular
Brr2 (Laggerbauer et al. 1998; Raghunathan and Guthrie
1998; Kim and Rossi 1999), as candidates for mediating
unwinding of the U4/U6 duplex. It is possible that dur-
ing splicing more than one DExD/H-box protein partici-
pates in unwinding of the U4/U6 duplex. In this regard,
U2AF65-mediated recruitment of hUAP56 to the branch-
point region may be particularly important for the rear-
rangement of U2–U6 snRNA interactions in the region
during the prespliceosome to spliceosome transition.
As described above, the ability of hUAP56 to contact
U4/U6 snRNAs and unwind the U4/U6 duplex is depen-
dent on U2AF65. We note that a previous study reported
that U2AF65did not affect the ability of purified, recom-
binant hUAP56 to mediate unwinding of an artificial,
short RNA duplex in the absence of other splicing factors
(Shen et al. 2007). There are several plausible explana-
tions for the differential requirement of U2AF65in the
two studies. In particular, U2AF65may be dispensable
for unwinding short, artificial dsRNA substrates but re-
quired for unwinding the natural U4/U6 snRNA duplex
in the context of the authentic U4/U6 snRNP. Alterna-
tively, the difference may be attributable to the presence
or absence of NE; for example, other splicing factors in
the NE may establish a requirement for U2AF65in U4/
U6 duplex unwinding. Relatedly, the unwinding activity
observed by Shen et al. (2007) required a high concentra-
tion of recombinant hUAP56, which likely far exceeds
that in HeLa NE. Finally, it may be relevant that Shen et
al. (2007) used a U2AF65derivative lacking 63 N-termi-
nal amino acids, whereas we used full-length U2AF65.
A new role for ATP hydrolysis in the function
of a DExD/H-box protein
It has been previously shown that DExD/H-box proteins
can use ATP hydrolysis to promote two reactions: RNA–
RNA unwinding (Laggerbauer et al. 1998; Raghunathan
and Guthrie 1998; Wagner et al. 1998; Wang et al. 1998)
and RNA–protein dissociation (Fairman et al. 2004).
Here we show that hUAP56 uses ATP hydrolysis to pro-
mote a protein–protein interaction, which to our knowl-
edge is unprecedented. For several DExD/H-box pro-
teins, including hUAP56 (Shen et al. 2007), ATPase
activity is stimulated by RNA. Thus, an attractive pos-
sibility is that during spliceosome assembly, contacts
between hUAP56 and an RNA would stimulate ATP hy-
drolysis, enhancing the hUAP5–U2AF65interaction and
thereby stabilizing the association of hUAP56 with the
splicing complex. The mechanistic basis for the ATP de-
pendence of the hUAP56–U2AF65interaction remains to
be determined. One possibility is that ATP hydrolysis
changes the conformation of hUAP56 to a form that can
interact with U2AF65. Although such a role for ATP hy-
drolysis has not been previously observed, we speculate
that it may represent a more general paradigm for regu-
lating interactions between DExD/H-box proteins and
their RNA or protein targets.
Materials and methods
Immunodepletion of hUAP56
hUAP56-depleted nuclear extract was prepared by incubating
150 µL of HeLa nuclear extract at 4°C with 50 pg of affinity-
purified rabbit polyclonal ?-hUAP56 antibody (generated by
AnaSpec against purified human UAP56 protein) cross-linked to
25 µL of protein A beads (Pierce). After 2 h of incubation, the
beads were separated from the extract by centrifugation. The
degree and specificity of hUAP56 depletion were monitored by
immunoblot analysis for hUAP56 and U2AF65. For immuno-
depletion of hUAP56 from U2AF-depleted extract, U2AF-de-
pleted nuclear extract was first generated by chromatography on
oligo(dT)-cellulose as described previously (Valcarcel et al.
hUAP56 expression and purification by gel filtration
hUAP56 was expressed as a GST fusion protein in E. coli strain
BL21(DE3) and purified as described previously (Shen et al.
2007). Briefly, GST-hUAP56 was first purified on glutathione
resin and, after thrombin-mediated cleavage of the GST tag,
hUAP56 was further purified using a Superdex 200 size exclu-
sion column. Mutant hUAP56 derivatives were generated by
the QuikChange Site Directed Mutagenesis Kit (Stratagene).
Circular dichroism spectroscopy
Circular dichroism (CD) spectroscopy was performed on a Jasco-
810 spectrometer using 0.6 mg/mL protein in 25 mM Tris-HCl
(pH 8.0) and 100 mM NaCl. Temperature-induced denaturation
was monitored by CD at a wavelength of 222 nm using 0.6
mg/mL protein from 20°C to 80°C. CD data points were taken
at 1°C intervals at a scan rate of 60°C per hour.
In vitro splicing assays and spliceosome assembly reactions
For splicing reactions, 2 µL of hUAP56-depleted or mock-de-
pleted nuclear extract was used per splicing reaction in a final
volume of 10 µL, and splicing reactions were performed essen-
tially as described previously (Fleckner et al. 1997) using a plas-
mid-encoded Minx (Ad ML) (Zillmann et al. 1988) or ?-globin
(Krainer et al. 1984) substrate and hUAP56 at a protein concen-
tration of 0.1 µM.
Role of hUAP56 in spliceosome assembly
GENES & DEVELOPMENT1801
Spliceosome assembly reactions were performed essentially
as described previously (Kan and Green 1999). Spliceosomal
complexes H, A, B, and C were resolved on nondenaturing 4%
(LMA) in 50 mM Tris base/50 mM glycine buffer (Wu and Green
1997); spliceosomal complexes H and E were separated on a
1.5% LMA gel in 0.5× TBE (Das and Reed 1999). Signals were
visualized by PhosphorImager (FujiFilm FLA-5000 imaging sys-
tem). U2AF-depleted nuclear extract was generated by chroma-
tography on oligo(dT)-cellulose as described previously (Valcar-
cel et al. 1997). Inactivation of U2 snRNA in the ?UAP56NE by
RNase H-directed cleavage was performed as described previ-
ously (Shen et al. 2004). To deplete ATP, the ?UAP56NE was
preincubated for 30 min at 30°C.
Affinity purification of splicing complexes
For affinity purification of RNA–protein complexes, a biotinyl-
ated Ad ML pre-mRNA substrate (100 fmol) was incubated with
HeLa nuclear extract (64 µL in a total volume of 100 µL) and 0.1
µM purified hUAP56 protein for 20 min under conditions that
allow spliceosome formation, after which heparin (200 ng/µL;
Sigma) was added to stop the splicing reaction. The reaction
mixture was then incubated with streptavidin-agarose beads
(Pierce) and washed five times (for 10 min each) in buffer R (20
mM Tris-HC1 at pH 7.8, 0.1% Triton X-100, 150 mM KC1, 2.5
mM EDTA). Affinity-purified complexes were then digested
with RNase A (Roche) to release the bound proteins, which
were analyzed for the presence of hUAP56 by immunoblotting
using a polyclonal ?-hUAP56 antibody (Fleckner et al. 1997).
GST pull-down assays
The GST-U2AF65fusion protein was expressed in E. coli strain
BL21 and purified as described previously (Valcarcel et al. 1996).
For the pull-down assay, 1 µg of GST-U2AF65fusion protein and
0.5 µg of purified hUAP56 protein were incubated in 100 µL of
Buffer D (25 mM HEPES at pH 7.5, 100 mM NaCl, 0.05% Noni-
det NP-40, 5 mM DTT, 10% glycerol, 50 µg/mL bovine serum
albumin) supplemented with 2 mM MgCl2for 20 min at 30°C in
the presence of 0.5 mM ATP (Promega), ATP-?S (Sigma), ADP
(Sigma), or GTP (Promega). Following incubation, 10 µL of glu-
tathione agarose (Pierce) was added to the reaction mixture, and
the beads were washed five times in Buffer D. Bound proteins
were eluted by boiling in protein loading buffer, separated by
SDS-PAGE, and analyzed by immunoblotting using a poly-
clonal ?-hUAP56 antibody (Fleckner et al. 1997).
UV cross-linking/immunoprecipitation assays
Purified hUAP56, hUAP56 mutant derivatives, or GST-U2AF65
were added to nuclear extract depleted of hUAP56 and/or U2AF
at a final concentration of 0.8 µM. Where indicated, Ad ML
pre-mRNA substrate was added to the NE, and ATP was de-
pleted by pre-incubating the nuclear extract for 30 min at 30°C.
The splicing reaction mixture was then irradiated with UV light
(254 nm) and incubated with 10 µL of ?-hUAP56 polyclonal
antibody (generated by Anaspec) for 2 h at 4°C, followed by
addition of 15 µL of anti-rabbit IgG agarose beads (Pierce) and
incubation for an additional 2 to 3 h at 4°C with continuous
mixing on a rotator device. The beads were washed four times
with high salt buffer (500 mM NaCl, 1% NP-40, 50 mM Tris-Cl
at pH 8.0), and, following Protease K treatment and phenol ex-
traction, RNA was precipitated by ethanol. The precipitated
RNAs were detected by primer extension analysis as described
previously (Valcarcel et al. 1996).
Native gel hybridization assays
Splicing reactions were performed as described above in the
presence or absence of Ad ML pre-mRNA, and total RNA was
purified from the reaction mixture by phenol extraction and
ethanol precipitation. To generate U4 and U6 probes, oligonu-
cleotides (U6, AAAATATAACTCTTCACGAATTTGCGTG;
U4, AGAGACTGTCTCAAAAATTGCCAA) were kinase-la-
beled with32P-ATP and gel purified before use. Native gel hy-
bridization assays were carried out as previously described (Kim
and Rossi 1999). Briefly, 2.5 µg of total RNA was incubated with
5 nM32P-labeled probe and vacuum-dried without heat. The
dried RNA and probe mixture was then dissolved in 6 µL of
hybridization buffer (150 mM NaCl, 50 mM Tris-HCl at pH 7.4,
and 1 mM EDTA) and incubated with total RNA for 15 min at
37°C. The annealing reactions were stopped by chilling in ice,
and then mixed with 6 µL of loading dye and electrophoresed on
a 9% nondenaturing polyacrylamide gel with 1× TBE buffer at
175 V in a cold room.
Psoralen cross-linking reactions were carried out in 40%
nuclear extract in the presence of unlabeled Ad ML pre-mRNA
under conditions that promote splicing as described previously
(Wassarman and Steitz 1992). At 20 min following the start of
the splicing reaction, 4?-aminomethyl-4,5?,8-trimethyl psoralen
(Sigma) was added to a final concentration of 20 µg/mL. The
reaction mixture was UV-irradiated (365 nm) for 10 min on ice
to generate RNA–RNA cross-links, and then deproteinized with
proteinase K treatment followed by phenol-chloroform (1:1) ex-
traction and ethanol precipitation to isolate RNA. To identify
the snRNA involved in potential RNA–RNA cross-links, iso-
lated RNA was analyzed on a 4% denaturing polyacrylamide gel
and subjected to Northern blot analysis using the U4 and U6
probes mentioned above.
We thank Woo Keun Song, Jang-Soo Chun, Hyon E. Choy, and
their laboratory members for providing laboratory space and
technical support, and Sara Evans for editorial assistance. This
work was supported in part by a Dasan Young Faculty Grant
from Gwangju Institute of Science and Technology, and the
Brain Korea 21 Project Research Foundation to H.S. and X.Z.; an
American Heart Association Scientist Development Grant and
an American Cancer Society Research Scholar Grant to R.Z.;
and a National Institutes of Health Grant to M.R.G. R.Z. is a V
Scholar and a Kimmel Scholar, and M.R.G. is an investigator of
the Howard Hughes Medical Institute.
Benz, J., Trachsel, H., and Baumann, U. 1999. Crystal structure
of the ATPase domain of translation initiation factor 4A
from Saccharomyces cerevisiae—The prototype of the
DEAD box protein family. Structure 7: 671–679.
Caruthers, J.M. and McKay, D.B. 2002. Helicase structure and
mechanism. Curr. Opin. Struct. Biol. 12: 123–133.
Cordin, O., Banroques, J., Tanner, N.K., and Linder, P. 2006. The
DEAD-box protein family of RNA helicases. Gene 367: 17–
Das, R. and Reed, R. 1999. Resolution of the mammalian E
complex and the ATP-dependent spliceosomal complexes on
native agarose mini-gels. RNA 5: 1504–1508.
Shen et al.
1802 GENES & DEVELOPMENT
Eperon, I.C., Makarova, O.V., Mayeda, A., Munroe, S.H., Cace-
res, J.F., Hayward, D.G., and Krainer, A.R. 2000. Selection of
alternative 5? splice sites: Role of U1 snRNP and models for
the antagonistic effects of SF2/ASF and hnRNP A1. Mol.
Cell. Biol. 20: 8303–8318.
Fairman, M.E., Maroney, P.A., Wang, W., Bowers, H.A., Goll-
nick, P., Nilsen, T.W., and Jankowsky, E. 2004. Protein dis-
placement by DExH/D “RNA helicases” without duplex un-
winding. Science 304: 730–734.
Fleckner, J., Zhang, M., Valcarcel, J., and Green, M.R. 1997.
U2AF65 recruits a novel human DEAD box protein required
for the U2 snRNP–branchpoint interaction. Genes & Dev.
Gatfield, D., Le Hir, H., Schmitt, C., Braun, I.C., Kocher, T.,
Wilm, M., and Izaurralde, E. 2001. The DExH/D box protein
HEL/UAP56 is essential for mRNA nuclear export in Dro-
sophila. Curr. Biol. 11: 1716–1721.
Herold, A., Teixeira, L., and Izaurralde, E. 2003. Genome-wide
analysis of nuclear mRNA export pathways in Drosophila.
EMBO J. 22: 2472–2483.
Hertel, K.J. and Graveley, B.R. 2005. RS domains contact the
pre-mRNA throughout spliceosome assembly. Trends Bio-
chem. Sci. 30: 115–118.
Kan, J.L. and Green, M.R. 1999. Pre-mRNA splicing of IgM ex-
ons M1 and M2 is directed by a juxtaposed splicing enhancer
and inhibitor. Genes & Dev. 13: 462–471.
Kim, D.H. and Rossi, J.J. 1999. The first ATPase domain of the
yeast 246-kDa protein is required for in vivo unwinding of
the U4/U6 duplex. RNA 5: 959–971.
Krainer, A.R., Maniatis, T., Ruskin, B., and Green, M.R. 1984.
Normal and mutant human ?-globin pre-mRNAs are faith-
fully and efficiently spliced in vitro. Cell 36: 993–1005.
Laggerbauer, B., Achsel, T., and Luhrmann, R. 1998. The human
U5-200kD DEXH-box protein unwinds U4/U6 RNA du-
plices in vitro. Proc. Natl. Acad. Sci. 95: 4188–4192.
Linder, P. 2006. Dead-box proteins: A family affair–active and
passive players in RNP-remodeling. Nucleic Acids Res. 34:
Luo, M.L., Zhou, Z., Magni, K., Christoforides, C., Rappsilber,
J., Mann, M., and Reed, R. 2001. Pre-mRNA splicing and
mRNA export linked by direct interactions between UAP56
and Aly. Nature 413: 644–647.
Meignin, C. and Davis, I. 2008. UAP56 RNA helicase is required
for axis specification and cytoplasmic mRNA localization in
Drosophila. Dev. Biol. 315: 89–98.
Raghunathan, P.L. and Guthrie, C. 1998. RNA unwinding in
U4/U6 snRNPs requires ATP hydrolysis and the DEIH-box
splicing factor Brr2. Curr. Biol. 8: 847–855.
Rocak, S. and Linder, P. 2004. DEAD-box proteins: The driving
forces behind RNA metabolism. Nat. Rev. Mol. Cell Biol. 5:
Sengoku, T., Nureki, O., Nakamura, A., Kobayashi, S., and Yo-
koyama, S. 2006. Structural basis for RNA unwinding by the
DEAD-box protein Drosophila Vasa. Cell 125: 287–300.
Shen, H., Kan, J.L., and Green, M.R. 2004. Arginine-serine-rich
domains bound at splicing enhancers contact the branch-
point to promote prespliceosome assembly. Mol. Cell 13:
Shen, J., Zhang, L., and Zhao, R. 2007. Biochemical character-
ization of the ATPase and helicase activity of UAP56, an
essential pre-mRNA splicing and mRNA export factor. J.
Biol. Chem. 282: 22544–22550.
Shi, H., Cordin, O., Minder, C.M., Linder, P., and Xu, R.M. 2004.
Crystal structure of the human ATP-dependent splicing and
export factor UAP56. Proc. Natl. Acad. Sci. 101: 17628–17633.
Silverman, E., Edwalds-Gilbert, G., and Lin, R.J. 2003. DExD/
H-box proteins and their partners: Helping RNA helicases
unwind. Gene 312: 1–16.
Staley, J.P. and Guthrie, C. 1998. Mechanical devices of the
spliceosome: Motors, clocks, springs, and things. Cell 92:
Valcarcel, J., Gaur, R.K., Singh, R., and Green, M.R. 1996. In-
teraction of U2AF65 RS region with pre-mRNA branch point
and promotion of base pairing with U2 snRNA. Science 273:
Valcarcel, J., Martinez, C., and Green, M.R. 1997. Functional
analysis of splicing factors and regulators. In mRNA forma-
tion and function (ed. J.D. Richter), pp. 31–53. Academic
Press, New York.
Wagner, J.D., Jankowsky, E., Company, M., Pyle, A.M., and
Abelson, J.N. 1998. The DEAH-box protein PRP22 is an
ATPase that mediates ATP-dependent mRNA release from
the spliceosome and unwinds RNA duplexes. EMBO J. 17:
Wang, Y., Wagner, J.D., and Guthrie, C. 1998. The DEAH-box
splicing factor Prp16 unwinds RNA duplexes in vitro. Curr.
Biol. 8: 441–451.
Wassarman, D.A. and Steitz, J.A. 1992. Interactions of small
nuclear RNA’s with precursor messenger RNA during in
vitro splicing. Science 257: 1918–1925.
Wu, S. and Green, M.R. 1997. Identification of a human protein
that recognizes the 3? splice site during the second step of
pre-mRNA splicing. EMBO J. 16: 4421–4432.
Zhu, H., Hasman, R.A., Young, K.M., Kedersha, N.L., and Lou,
H. 2003. U1 snRNP-dependent function of TIAR in the regu-
lation of alternative RNA processing of the human calcito-
nin/CGRP pre-mRNA. Mol. Cell. Biol. 23: 5959–5971.
Zillmann, M., Zapp, M.L., and Berget, S.M. 1988. Gel electro-
phoretic isolation of splicing complexes containing U1 small
nuclear ribonucleoprotein particles. Mol. Cell. Biol. 8: 814–
Role of hUAP56 in spliceosome assembly
GENES & DEVELOPMENT1803