Tài liệu Báo cáo khoa học: Major phosphorylation of SF1 on adjacent Ser-Pro motifs enhances interaction with U2AF65 doc

Major phosphorylation of SF1 on adjacent Ser-Pro motifs
enhances interaction with U2AF
65
Vale
´
rie Manceau
1
, Matthew Swenson
2
, Jean-Pierre Le Caer
3
, Andre
´
Sobel
1
, Clara L. Kielkopf
2
and Alexandre Maucuer
1
1 INSERM, U706, UPMC, Institut du Fer a
`
Moulin, Paris, France
2 Department of Biochemistry and Molecular Biology, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD, USA
3 Ecole Polytechnique, Laboratoire de Chimie des Me
´
canismes Re
´
actionnels, Palaiseau, France
The expression of the genome requires the precise and
controlled removal of intervening sequences within pre-
messenger RNAs (pre-mRNA splicing). Assembly of

the active spliceosome entails successive rearrangements,
including the entry and exit of molecular partners
(reviewed in [1]). Phosphorylation events are probably
molecular switches to control these conformational
changes. Indeed, experiments with phosphatase inhibi-
tors, purified phosphatases and nonhydrolysable ATP
analogues have shown that multiple phosphorylation
and dephosphorylation events are required for spliceo-
some assembly and splicing [2–4]. Among the best-char-
acterized of the phosphorylated splicing factors are the
serine-arginine rich (SR) proteins (reviewed in [5]),
whose intranuclear distribution and activity are influ-
enced by phosphorylation by specific kinases inclu-
ding SRPK1, SRPK2 [6,7], and Clk ⁄ Sty [8].
SF3b155 ⁄ SAP155, an integral spliceosome component
and substrate of cyclin E ⁄ CDK2 [9], is a non-SR protein
whose phosphorylation state is also regulated during the
splicing process [10]. In addition, other factors that
regulate splicing in a phosphorylation-dependent man-
ner have been identified (reviewed in [11,12]).
Splicing factor 1 (SF1) was identified as necessary
for spliceosome assembly by in vitro reconstitution
assays with protein fractions from HeLa cell nuclear
extracts [13], and in a synthetic lethality screen with
Mud2p, the yeast homologue of the splicing factor U2
auxiliary factor large subunit (U2AF
65
) [14]. Moreover
SF1 was found to physically interact with U2AF
65

[14–16]. SF1 binds to the branch point pre-mRNA
consensus sequence (BPS) near the 3¢ splice site [17],
and facilitates binding of U2AF
65
to the adjacent poly-
pyrimidine tract [15]. Next, SF1 is displaced from the
Keywords
protein phosphorylation; RNA splicing; SF1;
kinase KIS; U2AF homology motif
Correspondence
A. Maucuer, INSERM U706, 17, rue du Fer
a
`
Moulin, F-75005 Paris, France
Fax: +33 14587 6132
Tel: +33 14587 6139
E-mail:
(Received 14 October 2005, revised 5
December 2005, accepted 6 December
2005)
doi:10.1111/j.1742-4658.2005.05091.x
Protein phosphorylation ensures the accurate and controlled expression of
the genome, for instance by regulating the activities of pre-mRNA splicing
factors. Here we report that splicing factor 1 (SF1), which is involved in an
early step of intronic sequence recognition, is highly phosphorylated in mam-
malian cells on two serines within an SPSP motif at the junction between its
U2AF
65
and RNA binding domains. We show that SF1 interacts in vitro with
the protein kinase KIS, which possesses a ‘U2AF homology motif’ (UHM)

domain. The UHM domain of KIS is required for KIS and SF1 to interact,
and for KIS to efficiently phosphorylate SF1 on the SPSP motif. Import-
antly, SPSP phosphorylation by KIS increases binding of SF1 to U2AF
65
,
and enhances formation of the ternary SF1–U2AF
65
–RNA complex. These
results further suggest that this phosphorylation event has an important role
for the function of SF1, and possibly for the structural rearrangements asso-
ciated with spliceosome assembly and function.
Abbreviations
BPS, branch point sequence; CIP, calf intestinal phosphatase; DTT, dithiothreitol; GST, glutathione-S-transferase; SF1, splicing factor 1;
siRNA, small interfering RNA; snRNP, small nuclear ribonucleoprotein particle; RRM, RNA recognition motif; U2AF, U2 auxiliary factor;
UHM, U2AF homology motif.
FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS 577
spliceosome by the ATP-dependent entry of the U2
small nuclear ribonucleoprotein particle (snRNP),
whose SF3b155 ⁄ SAP155 protein subunit interacts with
U2AF
65
[18] and whose RNA component (U2
snRNA) anneals with the BPS [19,20]. This first ATP-
dependent step of 3¢ splice site recognition represents a
critical juncture for regulation of pre-mRNA splicing.
Protein kinase PKG is a potential regulator of this
step, by inhibiting the SF1 ⁄ U2AF
65
interaction upon
phosphorylation of a conserved SF1 serine (Ser20)

within its U2AF
65
interaction domain [21].
Comparison of the structure of an RNA recognition
motif (RRM)-containing domain of the U2AF small
subunit (U2AF
35
) complexed to a U2AF
65
peptide
[22], with the structure of the C-terminal RRM
(RRM3) of U2AF
65
complexed to an N-terminal pep-
tide of SF1 [23] reveals that the unique RRM of
U2AF
35
and RRM3 of U2AF
65
belong to a subclass
of RRMs with specialized features for protein–protein
interactions [23,24]. Members of this RRM subclass
are called U2AF homology motifs (UHMs). Diverse
UHM-containing proteins have been identified that
contain critical sequence features for interaction with
peptide ligands, including the protein kinase KIS
[23,24]. The KIS polypeptide is organized into a kinase
core followed by a C-terminal UHM (Fig. 1B) [25,26],
and preferentially phosphorylates Ser-Pro sites in vitro
[27]. A likely role of KIS during the control of cell-

cycle division is supported by the phosphorylation of
the cyclin dependent kinase inhibitor p27
kip1
[28] and
the observation that KIS mRNA levels are misregu-
lated in neurological tumours [29].
We report here that SF1 is phosphorylated on two
major adjacent Ser-Pro motifs (hereafter called the
SPSP motif). We show that the protein kinase KIS can
interact with SF1 through its UHM domain and effi-
ciently phosphorylate SF1 on both serines of this SPSP
motif, and therefore is likely to participate in control-
ling the phosphorylation state of SF1. Finally, SF1
phosphorylation on its SPSP motif enhanced the inter-
action of SF1 with U2AF
65
, and also promoted forma-
tion of a ternary complex with a model 3¢ intronic
sequence, suggesting the importance of this major SF1
phosphorylation for recognition of the 3¢ splice site.
Results
Phosphorylation of SF1 on serines 80 and 82
in vitro and in vivo
The structural features and the mutated forms of SF1,
KIS and U2AF
65
used in this study are represented
in Fig. 1. The presence of an SPSP motif in a highly
Proline rich
SF1f

KIS
U2AF
KIS[54R]
KIS[1-313]
KIS[∆369-409]
∆RS
KIS[EDKK]
136-22811-25
25-63
85-112
134-229 242-333 374-465
54
341-342
KR
1
1
1
1
1 313
1
1
419
419
475
475
419
255
279-292 324-590
639
Zn

SF1
8280
ED KK
U2AF
U2AF
SF1
65
35
65
U2AF
65
KH-QUA2
A
B
C
23-304 316-415
1 419
Kinase UHM
RRM3/UHMRRM2RRM1RS
Fig. 1. Schematic representations of SF1 (A), KIS (B) and U2AF
65
(C). (A) Representation of SF1HL1, a major SF1 splice variant in
HeLa cells [32]. (The various SF1 splice variants differ by the length
and sequence of their proline rich C-terminal region [32,53]). The
SPSP motif is in a highly phylogenetically conserved region [16],
between the N-terminal U2AF
65
binding region and the KH-QUA2
domain that mediates recognition of the branchpoint sequence
[23,54]. The aligned sequences of SF1 are from the following

organisms: Hs, Homo sapiens; Ce, Caenorhabditis elegans; Dm,
Drosophila melanogaster;Sc,Saccharomyces cerevisiae;Sp,Schizo-
saccharomyces pombe. Zn, zinc knuckle motif. The SF1f truncated
form of SF1 used for in vitro experiments in this study retains its
U2AF
65
and RNA binding properties [16]. (B) KIS is formed by the
juxtaposition of a serine ⁄ threonine kinase domain and a C-terminal
UHM domain. Mutation of lysine 54 to arginine (KIS[54R]) impairs
the kinase activity of KIS [26]. KIS [1–313] entirely lacks the UHM
domain, whereas KIS[D369–409] is deleted within the UHM. Negat-
ively charged residues Glu341 and Asp342 are replaced with
positively charged lysines in the KIS[EDKK] variant. (C) U2AF
65
possesses an N-terminal arginine and serine rich (RS) domain and
three C-terminal RRM domains [39]. The C-terminal domain inter-
acts with SF1 and is part of a subclass of RRM-like domains called
UHMs [23,24]. U2AF is a heterodimer of U2AF
65
and U2AF
35
[38].
The domain of U2AF
65
interacting with the UHM domain of U2AF
35
is indicated [22]. U2AF
65
DRS lacks residues 25–63 [39].
SF1 SPSP motif phosphorylation V. Manceau et al.

578 FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS
conserved region of SF1 suggested that these serine
residues could be targets of the proline directed kinase
KIS. Further, we speculated that KIS could interact
with SF1 based upon sequence similarity of the KIS
and U2AF
65
UHM domains (42% pairwise sequence
identity between UHMs of human KIS, accession
code NP_787062 and human U2AF
65
, accession code
NP_009210). Pull-down experiments were used to
probe the physical interaction of KIS with SF1 in vitro
(Fig. 2). We observed that KIS efficiently binds to the
SF1f fragment of SF1 containing the SPSP motif and
the essential domains for BPS and U2AF
65
recognition
[16,17] (human SF1 residues 1–255) (Fig. 2, lane 7).
Interestingly, KIS bound SF1f as efficiently as
U2AF
65
DRS, a U2AF
65
construct lacking its N-ter-
minal arginine and serine rich (RS) domain (Fig. 2,
lane 12). In our conditions, binding of U2AF
65
DRS to

SF1f was about twofold less than that of intact
U2AF
65
(Fig. 2, lane 11), suggesting that the U2AF
65
RS domain contributes to SF1f binding. A possible
role for the U2AF
65
RS domain for binding to SF1 is
consistent with the established interaction between the
U2AF
65
RS domain and the BPS [30,31], which is also
targeted by SF1 during pre-mRNA splicing [17]. A
complete or a partial deletion of the UHM domain of
KIS severely reduced binding to SF1f (Fig. 2, lanes 8
and 9). Mutations of acidic residues Glu396 and
Glu397 to lysine within the UHM of U2AF
65
have
been shown to impair its interaction with SF1 [23]. We
observed that mutations of the corresponding KIS
acidic residues (Glu341 and Asp342) to lysine also
impaired KIS interaction with SF1 (Fig. 2, lane 10).
Finally the interactions were specific, as no significant
binding of the KIS or U2AF
65
variants to glutathione-
S-transferase (GST) was detected (Fig. 2, lanes 1–6).
Altogether, when compared with the established SF1

interaction partner U2AF
65
, KIS bound efficiently to
SF1f and this interaction required structural features
of its UHM that are shared with U2AF
65
.
Kinase assays showed that recombinant KIS effi-
ciently phosphorylated SF1f, as evidenced by
32
P phos-
phate incorporation and a marked shift of the SF1f
band on SDS ⁄ PAGE (Fig. 3A). The interaction of the
UHM domain of KIS with SF1 appears to be import-
ant for efficient phosphorylation, as deletions within
this domain prevented SF1f phosphorylation (Fig. 3B,
lanes 3 and 4). Under all conditions tested, phosphate
incorporation never exceeded an evaluated stoichio-
metry of two phosphates per one SF1f molecule, indi-
cating that phosphorylation occurs on two residues.
Phosphoamino acid analysis showed that phosphoryla-
tion occured exclusively on serine residues (data not
shown). Following phosphorylation of SF1 with KIS
to high stoichiometry (over 1.5 phosphates per SF1
8 9 10 11 121342567
GST-SF1fGST
8 9 10 11 121342567
GST-SF1fGST
U2AF
65

U2AF
65
∆RS
KIS / mutant
KIS[∆369-409]
KIS[1-313]
0
5
10
15
20
25
KIS
∆369-409
KIS[1-313]
KIS(EDKK)
U2AF
65
U2AF
65
∆RS
binding to SF1f
% of Input
35
S
GST
GST-SF1f
Coomassie
Fig. 2. KIS interaction with SF1 in vitro. Var-
ious forms of KIS and U2AF

65
were transla-
ted in vitro in the presence of
35
S-labelled
methionine and tested for their binding to
GST–SF1[1–255] (GST–SF1f) in a GST pull-
down assay (top right, lanes 7–12). Lane 7:
wild-type KIS; lane 8: KIS[D369–409] with a
deletion within the UHM domain of KIS;
lane 9: KIS[1–313] lacking the UHM domain;
lane 10: KIS[EDKK] with mutations of
Glu341 and Asp342 to lysine, lane 11:
full-length U2AF
65
and lane 12: U2AF
65
DRS
lacking the RS domain. Lanes 1–6: back-
ground binding on GST beads. The binding
of the different proteins to GST–SF1f was
quantified (bottom) as the fraction of the
input protein (% of input) that was bound to
the beads (the total input protein was deter-
mined by running 0.5% of starting material
in parallel (not shown)). Data are mean ± SD
of three experiments. The equivalent loading
of the beads with GST and GST–SF1f was
checked by Coomassie staining of the gel
(top left).

V. Manceau et al. SF1 SPSP motif phosphorylation
FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS 579
molecule) and digestion with trypsin, we identified, by
MS, masses corresponding to peptide 67–92 and pep-
tide 80–92 of human SF1, each with two phosphates.
Both of these SF1 peptides contained Ser80 and Ser82
of the putative KIS target SPSP motif. Phosphoryla-
tion efficiency decreased by approximately twofold
when using SF1f with either Ser80 or Ser82 mutated to
alanine, whereas the double alanine mutation (here-
after called 8082A) completely inhibited phosphoryla-
tion (Fig. 3C). Thus, KIS can phosphorylate both sites
with similar efficiency and no phosphorylation occurs
outside the SPSP motif. Altogether, KIS phosphorylated
SF1f on Ser80 and Ser82 with high efficiency that
depended on anchoring its UHM domain to SF1f.
To determine whether the SPSP motif is phosphoryl-
ated in vivo, we metabolically labelled HEK293 cells
overexpressing the major spliced form SF1HL1 [32]
with a myc tag (hereafter SF1myc), and immuno-
precipitated SF1myc using an antimyc antibody. As
shown in Fig. 4A, SF1myc phosphate incorporation
was dramatically reduced (by approximately fourfold)
when both Ser80 and Ser82 of SF1 were mutated to
alanine. Thus, the SPSP motif contains major phos-
phorylation sites of SF1 in vivo. The remaining incor-
porated phosphate most likely corresponds to
phosphorylation of other SF1 sites such as Ser20, a
known in vivo phosphorylation target [21].
To confirm the major phosphorylation of the SPSP

motif, we compared the tryptic phosphopeptide map of
SF1f phosphorylated in vitro by KIS to that of SF1myc
(min)
10
20
30
40
60
90
120
180
240
360
A
C
B
KIS
KIS[54R]
KIS[1-313]
KIS[∆369-409]
Coomassie
SF1f
SF1f
32
P incorporation (au)
32
P incorporation (au)
SF1f
0 100 200 300
32

P
32
P
0
2
4
6
8
GST-
KIS
GST-
SF1/
mutants
Coomassie
SF1f
80A
82A
13-255
8082A
SF1f
80A
82A
13-255
8082A
time (min)
SF1f
13-255
80A
82A
8082A

1.2
1.0
0.8
0.6
0.4
0.2
0
32
P
Fig. 3. KIS phosphorylates SF1 in vitro on serines 80 and 82. (A)
SF1 [1–255] (SF1f) was phosphorylated by KIS with 10 l
M
[c-
32
P]ATP for the indicated times. Reaction products were ana-
lysed by SDS ⁄ PAGE, Coomassie blue staining and phosphorimag-
ing. Phosphorylation of SF1f induced a shift of the SF1f band on
SDS ⁄ PAGE. (B) The ability of similar amounts of the indicated
forms of KIS to phosphorylate SF1f in vitro was compared, show-
ing that deletion of or within the UHM domain of KIS impaired SF1f
phosphorylation. (C) In vitro phosphorylation of different forms of
GST–SF1f as indicated. Reactions were stopped while phosphate
incorporation was still linear with time (see Experimental proce-
dures). Mutation of both Ser80 and Ser82 completely abolished
phosphate incorporation. Data are mean values of three experi-
ments.
SF1f - KIS
SF1
SF1 - KIS[54R]SF1 - KIS
b

a
1
2
3
4
c
d
e
b
a
c
d
e
c
d
e
b
a
c
d
e
SF1myc
SF1myc
SF1
~80kDa
~60kDa
Ig HC
Coomassie
A
B

C
1
vect
54R
KIS
vect
54R
KIS
3425671342567
SF1
8082A
vect
54R
KIS
vect
54R
KIS
SF1
8082A
anti-myc
anti-SF1
CIP(u)
t(hrs)
001 225
022 022
+
32
P
Fig. 4. SF1 SPSP motif phosphorylation in cells. (A) HEK293 cells
were transfected with SF1myc or SF1myc(8082 A) together with

either pCDNA3 (vector), pCDNA3-KIS(54R) (kinase defective) or
pCDNA3-KIS. After metabolic labelling of the cells with
32
P-labelled
inorganic phosphate for 4 h, overexpressed SF1myc was immuno-
precipitated with antimyc mAb 9E10 and analysed by SDS ⁄ PAGE,
Coomassie staining (left) and phosphorimaging (right). (Ig HC:
Immunoglobulin heavy chains). (B) Phosphorylated SF1 was ana-
lysed by tryptic phosphopeptide mapping. Map1: SF1f phosphoryl-
ated by KIS in vitro (with a moderate stoichiometry of about 0.1
phosphate per SF1 molecule). Maps 2, 3 and 4: phosphorylated
immunoprecipitated SF1myc from lanes 1, 3 and 2 of the gel in A.
SF1f (Map1) and SF1myc (Map2) present almost identical tryptic
maps showing that major phosphorylation occurs on the SPSP
motif in vivo. Coexpression of KIS leads to a major relative
decrease of the more basic a and b peptides (two experiments),
indicating an increase in the phosphorylation state of the SPSP
motif. (C) Five micrograms of protein extract of HEK293 cells over-
expressing SF1myc (top) or of untransfected HEK293 cells (bottom)
were analysed by phosphatase (CIP) treatment with the indicated
amounts (units from New Englands Biolabs) and Western blotting
with antimyc (top) or anti-SF1 antibody (bottom). In additional char-
acterization experiments we showed that the 80-kDa band migra-
ted just below SF1myc, and was nuclear and heat soluble, which
are characteristics of SF1 [13]. The 60-kDa band most probably
corresponds to an alternatively spliced form [32].
SF1 SPSP motif phosphorylation V. Manceau et al.
580 FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS
phosphorylated in cells (Fig. 4B, maps 1 and 2). These
peptide maps were highly similar. They displayed mul-

tiple spots corresponding to incomplete digestion by
trypsin, as previously observed by MS, and to the
presence of single and double phosphorylation of
the peptides as suggested by our analysis of SF1f
phosphorylated by KIS to different extents in vitro (data
not shown). Co-expression of KIS did not clearly
increase the incorporation of radioactive phosphate
(Fig. 4A, lane 3), but phosphopeptide mapping revealed
its effect to increase the relative amounts of more acidic
peptides (c, d, e) (Fig. 4B, compare maps 2 and 3),
which suggests that KIS increases the phosphorylation
level of SF1 on the SPSP motif. Coexpression with
KIS(54R), bearing a mutation within the kinase active
site that extensively suppresses activity [26], had an
intermediate effect (Fig. 4B, map 4). This agreed with
our unpublished observations that KIS(54R) retains low
levels of kinase activity towards SF1, due to the unusu-
ally high activity of KIS for this substrate.
The slight effect of KIS overexpression on phos-
phate incorporation in SF1myc suggested that SF1 is
already mostly in a phosphorylated state in prolifer-
ating HEK293 cells. This was also supported by the
observation that wild-type SF1myc expressed in
HEK293 cells migrates slower than the 8082A mutant
(compare lanes 1–3 with 4–6 in Fig. 4A). We tested
this possibility by treating nuclear extracts of SF1myc
overexpressing cells with alkaline phosphatase. As
shown in Fig. 4C, phosphatase treatment induced a
faster migration of SF1myc on SDS ⁄ PAGE showing
that SF1myc is, indeed, already highly phosphorylated

in proliferating cells. This was also the case for endo-
genous forms of SF1 in HEK293 (Fig. 4C, lower
panel) and HeLa cells (data not shown).
To summarize, these analyses show that the SPSP
motif contains major in vivo phosphorylation sites of
SF1. In agreement with this result, two recent inde-
pendent large-scale analyses of the phosphoproteome,
one in nuclear extracts of HeLa cells, the other in
WEHI-231 B lymphoma cells, lead to the identification
of the same 67–92 peptide of SF1 phosphorylated on
both Ser80 and Ser82 [33,34]. Altogether, this high
level of phosphorylation of its SPSP motif indicates a
particular involvement in SF1 function.
Modulation of SF1 binding properties upon SPSP
motif phosphorylation
The SPSP motif lies at the junction between the
U2AF
65
binding N-terminal region of SF1 and its
RNA binding domain, suggesting that phosphorylation
of these sites might regulate the formation of the
SF1-U2AF-RNA complex. We first tested the effect
of phosphorylation on the interaction of SF1 with
U2AF
65
by GST pull-down experiments. Using GST–
SF1f that was phosphorylated by KIS to high stoi-
chiometry (over 1.5 phosphate per SF1f molecule) we
observed a twofold increase in U2AF
65

and a threefold
increase of U2AF
65
DRS binding (Fig. 5). In contrast
no increase in binding could be observed with the
SF1f(8082A) mutant upon treatment with KIS, show-
ing that phosphorylation on the SPSP motif of SF1 is
responsible for U2AF
65
binding enhancement.
As SF1 and U2AF
65
have been shown to bind in a
cooperative manner to a model 3¢ splice site RNA, we
hypothesized that the modification of SF1 binding to
U2AF
65
upon phosphorylation on the SPSP motif
could in turn modulate the formation of the SF1–
U2AF
65
–RNA ternary complex. We analysed the
formation of this complex and the effect of SF1 phos-
phorylation by RNA gel-shift with a model 3¢ intronic
sequence previously used by Berglund and colleagues
to characterize the cooperative binding of SF1 and
U2AF
65
to RNA [15]. As shown in Fig. 6, the RNA
shifts induced by U2AF

65
, and SF1f plus U2AF
65
binding were clearly identified as previously described
[15]. Interestingly we observed an increase in ternary
complex formation when using SF1 previously phos-
phorylated by KIS (Fig. 6A, compare lane 3 with lanes
GST
GST-SF1f
(1%)
8082A
KIS
KIS
+ + +
+
+
+
U2AF
65
∆RS
Prey
35
S
0
10
20
30
40
50
60

70
80
90
GST SF1f 8082A
binding
(% of input)
U2AF
65
U2AF
65
∆RS
Fig. 5. SF1 SPSP motif phosphorylation enhances binding to
U2AF
65
. Pull-down experiments were performed using purified
GST, GST–SF1f or GST–SF1f(8082A) that were phosphorylated by
KIS (+) or mock-phosphorylated (–).We used as input a mixture of
in vitro translated [
35
S]methionine-labelled U2AF
65
and U2AF
65
DRS.
We checked that interaction was the same when these proteins
were alone or in the mixture (not shown). Interactions with GST–
SF1f phosphorylated by KIS vs. mock-phosphorylated were in dupli-
cate. One per cent of the input was loaded to allow quantification
by phosphorimaging of the fraction of the
35

S-labelled proteins that
was bound to the beads. The mean values of duplicates with stand-
ard deviation are represented. Representative results of two experi-
ments.
V. Manceau et al. SF1 SPSP motif phosphorylation
FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS 581
4 and 5). In contrast, in the absence of U2AF
65
,we
observed no major difference in RNA binding of
mock-phosphorylated SF1f and phosphorylated SF1
(Fig. 6B). Further analysis using increasing concentra-
tions of mock-phosphorylated SF1f or phosphorylated
SF1f with a constant concentration of U2AF
65
(1.2 lm) confirmed the enhancement of ternary
complex formation upon phosphorylation of SF1
(Fig. 6C). This was observed using three different pro-
ductions of phosphorylated SF1 vs. parallel produc-
tions of mock-phosphorylated SF1. The ratios of the
apparent equilibrium dissociation constants (K
d
) were
calculated by linear regression of the Hill plot for four
experiments, and showed an approximately threefold
decrease of apparent K
d
for SF1 binding to RNA
upon phosphorylation by KIS. To summarize, these
results show that phosphorylation of SF1 on the SPSP

motif enhances binding to U2AF
65
and formation of
the ternary SF1–U2AF
65
–RNA complex.
Discussion
Given that the kinase KIS contains a UHM domain
with significant similarity to the SF1-binding UHM
domain of the pre-mRNA splicing factor U2AF
65
,we
hypothesized that KIS likewise might be targeted to
pre-mRNA splicing factors by its UHM [26]. Accord-
ingly, the splicing factor SF1 was a very efficient sub-
strate for KIS in vitro when compared with other
candidate substrates that we have tested ([27] and
unpublished data). Using MS and mutagenesis, we
demonstrated that KIS phosphorylates SF1f (human
SF1 residues 1–255) exclusively on Ser80 and Ser82
within an SPSP motif. Interestingly, we observed that
KIS can interact with SF1f as efficiently as with
U2AF
65
DRS, and that its UHM domain is required
for binding to, and for the phosphorylation of SF1f.
This result further illustrates the molecular similarity
between the UHM domains of KIS and U2AF
65
. Alto-

gether, KIS is a likely candidate for controlling the
phosphorylation state of the SF1 SPSP motif in verte-
brates, where KIS is highly conserved (for example,
63% sequence identity between human KIS, accession
code NP_787062 and zebrafish KIS, accession code
XP_698499). Other proline directed kinases may phos-
phorylate the highly conserved SPSP motif of SF1 in
nonvertebrates (Fig. 1A). The relative importance of
KIS, compared with other kinases and phosphatases,
for regulating the phosphorylation state of the SF1
SPSP motif in vivo is an intriguing subject for further
investigation.
By comparing the phosphopeptide map of SF1f fol-
lowing in vitro phosphorylation by KIS with that of
C
B
U2AF
65
(1.2 µM)
pSF1f
SF1f
[SF1] µM
µM
0.27
0.53
0.80
0.27
0.53
0.80
A

pSF1f
SF1f
SF1f
0.46
0
1.4
4.6
0.46
1.4
4.6
Kd(SF1)
Kd(pSF1)
= 3.0 +/- 0.6 (n=4)
0
4
8
12
16
20
0 0,2 0,4 0,6 0,8 1
Fraction of RNA
bound to SF1f (%)
SF1f
pSF1f
54R
GST
KIS
KIS
treatment
U2AF

65
U2AF
65
free RNA
free RNA
SF1f/U2AF
65
U2AF
65
SF1f/U2AF
65
SF1f
0
1.2 µM
0 0.3µM
1 2345
free RNA
Fig. 6. SF1 SPSP motif phosphorylation enhances formation of an
SF1–U2AF
65
–RNA ternary complex. (A) We used a [
32
P]-labelled
RNA oligonucleotide corresponding to a model 3¢ intronic sequence
previously characterized in gel-shift experiments with SF1 and
U2AF
65
(see Experimental procedures) [15]. The RNA oligonucleo-
tide was incubated with different protein mixtures. A single shift
with U2AF

65
and supershift with U2AF
65
plus SF1f were clearly
identified (lanes 2 and 3). Formation of SF1f-U2AF
65
-RNA complex
(upper band) was enhanced by previous phosphorylation of SF1f by
KIS (compare lane 3 with lanes 4 and 5). (B) Increasing concentra-
tions of mock-phosphorylated SF1f or SF1f phosphorylated by KIS
(pSF1f) showed no major difference in RNA shift. (C) To quantify
the enhancement of SF1f–U2AF
65
–RNA complex formation upon
phosphorylation of the SPSP motif, we used increasing concentra-
tions of mock-phosphorylated SF1f or SF1f phosphorylated by KIS
(pSF1f) in the presence of 1.2 l
M U2AF
65
. The different bands
were quantified by phosphorimaging and the formation of the tern-
ary complex was plotted as a fraction of total RNA. The decrease
of the apparent K
d
of SF1 for RNA upon phosphorylation by KIS
was calculated by determining the y intercept of the Hill plot for
each experiment (data not shown). We used three different prepa-
rations of phosphorylated and mock-phosphorylated SF1f in four
experiments and found that phosphorylation induced an average
threefold decrease of the apparent K

d
of SF1 for RNA in the pres-
ence of 1.2 l
M U2AF
65
.
SF1 SPSP motif phosphorylation V. Manceau et al.
582 FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS
SF1myc immunoprecipitated from HEK293 cells, we
demonstrated that SF1myc is phosphorylated on the
SPSP motif in these cells. Accordingly, mutation of
Ser80 and Ser82 to alanine strongly reduced phos-
phate incorporation in SF1myc. The high level of
phosphorylation of overexpressed and endogenous
SF1 was demonstrated by treating cell extracts with
phosphatase, which resulted in a faster migration of
SF1 on SDS ⁄ PAGE, and showed that all detectable
SF1 is initially in a phosphorylated state in HEK293
cells. This was also the case in HeLa cells and mouse
brain (data not shown). The functional significance of
this major phosphorylation of SF1 was further sug-
gested by the induced modification of its interaction
with U2AF
65
. Remarkably, U2AF
65
recovery on
GST–SF1 beads in pull-down experiments was mark-
edly enhanced when GST–SF1 was first phosphorylated
by KIS (Fig. 5). Moreover, in gel-shift experiments

with a model 3¢ splice site RNA, phosphorylation of
the SF1 SPSP motif by KIS reproducibly enhanced
formation of the ternary SF1–U2AF
65
–RNA complex
(Fig. 6). Two a-helices are predicted to form on each
side of the proline-rich region containing the SPSP
motif [16]. Thus, local structural modifications of the
SPSP region by phosphorylation could be amplified
by these helical structures, thereby facilitating the
SF1 ⁄ U2AF
65
interaction. This hypothesis is supported
by the observation that phosphorylation within pro-
line rich regions often induces structural rearrange-
ments [35], which could propagate up to the U2AF
65
binding site in the case of SF1. However, it is also
possible that the phosphorylated SPSP motif directly
participates in the interaction with U2AF
65
. Of note,
the experiments with SF1 fragments showing that
regions C-terminal to residue 25 were dispensable for
interaction were performed with unphosphorylated
SF1 and thus do not exclude this possibility [23]. Our
data suggest that the phosphorylated SPSP motif of
SF1 at least does not interact with the basic RS
domain of U2AF
65

, because enhancement of binding
upon phosphorylation was also observed for the
U2AF
65
DRS fragment. Further investigations are nee-
ded to examine the structural basis for the effect of
SPSP motif phosphorylation on the interaction of
SF1 with U2AF
65
.
A comprehensive set of data supports the functional
importance of the interaction of SF1 with U2AF
65
.In
mammals, the interaction between U2AF
65
and SF1
was demonstrated by two-hybrid, pull-down and far-
western assays [14–16]. These interactions involve the
highly conserved N terminus of SF1 and the UHM
domain of U2AF
65
. In addition, SF1 and U2AF
65
were shown to be both present in the early spliceosomal
complex E [14,36]. SF1 was shown to bind preferen-
tially to the pre-mRNA branchpoint sequence [17,37]
while U2AF
65
interacts with the nearby poly-pyrimid-

ine tract and its RS domain contacts the branchpoint
[30,31,38,39]. Furthermore the interactions of U2AF
65
and SF1 with the branchpoint region were found to be
cooperative [15]. Finally structural bases for this inter-
action have been determined [23].
The SF1 ⁄ U2AF
65
interaction appears to be highly
conserved throughout evolution. In Schizosaccharomy-
ces pombe, SF1 and the U2AF subunit homologs
U2AF
59
and U2AF
23
form a stable complex [40]. In
Saccharomyces cerevisiae, the SF1 orthologue Msl5p is
highly conserved while the U2AF
65
orthologue Mud2p
shows a sequence conservation restricted to the UHM
domain involved in its interaction with SF1. Like their
mammalian counterparts, Msl5p and Mud2p interact
and are recruited early in spliceosome assembly
[41,42]. Msl5p has a marked preference for binding to
the consensus S. cerevisiae branchpoint sequence [17]
and Mud2p requires an intact branchpoint region for
binding [41,43]. Furthermore MSL5 was identified by
virtue of its genetic interaction with MUD2 [14].
Importantly, SF1 is thought to perform an essential

conserved function for cell viability. Actually, in HeLa
cells, small interfering RNA (siRNA) depletion of SF1
leads to cell death [44] and in S. cerevisiae MSL5 dis-
ruption is lethal [14]. In this context, our finding that
the SF1 ⁄ U2AF
65
interaction can be enhanced by phos-
phorylation is likely to be significant for understanding
SF1 function.
Of note, despite strong evidence that SF1 is neces-
sary for spliceosome assembly using in vitro reconstitu-
tion assays with protein fractions from HeLa nuclear
extracts [13], several reports suggest that the SF1–
U2AF
65
interaction is dispensable for splicing of at
least a subset of pre-mRNAs. Actually, thorough
immunodepletion of SF1 from a HeLa splicing extract
only modestly reduces the kinetics of spliceosome
assembly [45]. Similarly, no differences in the splicing
of three pre-mRNA substrates could be detected when
a U2AF
65
-depleted nuclear extract was complemented
with full-length U2AF
65
or U2AF
65
lacking its UHM
domain [46]. Moreover, SF1 knockdown in HeLa cells

using siRNA did not apparently affect splicing of sev-
eral pre-mRNAs, nor the levels of a variety of proteins
translated from spliced mRNA [44]. Finally, using
S. cerevisiae extracts that had been either depleted of
SF1 or prepared from temperature-sensitive mutants
grown at a nonpermissive temperature, no dramatic
defect in splicing of a pre-mRNA with consensus splice
sites was detected [42,47]. Nevertheless, splicing in vivo
of pre-mRNA with weakened splice site sequences was
V. Manceau et al. SF1 SPSP motif phosphorylation
FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS 583
reduced in temperature-sensitive mutants grown at a
nonpermissive temperature, suggesting that SF1 func-
tion is important for splicing efficiency [47]. It has been
proposed that a low level of SF1 remains in depleted
extracts or in siRNA treated cells, and is sufficient for
its transient catalytic action in pre-spliceosome assem-
bly on most pre-mRNAs [42,44,47]. This leaves open
the possibility that SF1 has an essential role for spli-
cing in vivo that is not evident under in vitro conditions
and in siRNA treated cells. In this context, the phos-
phorylation of the SPSP motif that enhances its bind-
ing to U2AF
65
could be important for splicing by
facilitating recognition of weak acceptor sequences.
The highly phosphorylated state we observe for SF1
in vivo suggests that SF1 dephosphorylation is transient,
for example to allow the recycling of SF1 after pre-
spliceosome assembly. Specifically, dephosphorylation

of SF1 may facilitate replacement by the U2 snRNP
component SF3b155 ⁄ SAP155 within the assembling
spliceosome [18], by decreasing the affinity of SF1 for
U2AF
65
. Accordingly, the fact that 54 amino acid resi-
dues separate the SPSP phosphorylation site from the
minimal U2AF
65
binding domain suggests that the
SPSP motif remains accessible to regulatory phospha-
tases in the early spliceosome complex.
Several proposals have been made to explain the SF1
requirement for cell viability. When SF1 levels are
decreased, it is possible that SF1 becomes rate limiting
for splicing of a subset of pre-mRNAs, some of which
encode essential peptides [47]. Alternatively, a cumula-
tive reduction in the kinetics of splicing of a larger sub-
set of pre-mRNAs may be lethal. Finally, functions of
SF1 in pathways other than pre-mRNA splicing have
been suggested to explain its requirement for cell viabil-
ity [42,44,47]. For example, SF1 was found to bind tran-
scription factors and to repress transcription in a
reporter assay in mammalian cells [48,49]. An inter-
action of SF1 with the transcription elongation factor
CA150 was also reported [50]. In addition, the UHM-
containing proteins PUF60, Tat-SF1 and CAPER that
may also associate with SF1 have been implicated in
transcription (see [23,24] and references herein). In
S. cerevisiae, MSL5 temperature-sensitive mutants at

a nonpermissive temperature presented a pre-mRNA
retention defect [47]. In this context, U2AF
65
also has
been shown to be a shuttling protein and its role during
RNA export has been documented [51,52]. Thus, the
observed effect of phosphorylation of the SPSP motif
on ternary SF1–U2AF
65
–RNA complex formation may
regulate any of several putative SF1 functions: pre-
mRNA splicing, transcription, or export. In conclusion,
given the strong evidence supporting the conserved
SF1 ⁄ U2AF
65
interaction, it is likely that the regulations
of the SF1 ⁄ U2AF
65
interaction by phosphorylation of
SF1 on the highly conserved SPSP motif may play a
major functional role. Our identification of this novel
phosphorylated form of SF1 provides a starting point to
evaluate the questions of whether and how pre-mRNA
splicing and ⁄ or the other potential functions of SF1 are
regulated by SPSP motif phosphorylation and by KIS.
Experimental procedures
Plasmids and mutagenesis
The plasmid for expression of the human SF1 fragment,
SF1f (residues 1–255) was constructed in the pGEX6P-1 vec-
tor (Amersham Biosciences, Uppsala, Sweden). SF1f con-

tains the U2AF
65
binding domain and KH-QUA2 domain
for BPS recognition. Prior to PCR, the SF1 template was
corrected for a Arg19Gly mutation inadvertently present
from previous published work [14,15]. Plasmids pSP64-
U2AF
65
and U2AF
65
DRS (lacking residues 25–63) [39] for
in vitro translation were kindly provided by J. Valcarcel. To
allow expression of rat KIS from the same vector, an NcoI–
XhoI blunt insert from plasmid BSKIS [26] was transferred
to pSP64-U2AF
65
NcoI–EcoRI blunt yielding pSP64-KIS.
For expression in mammalian cells, SF1HL1 cDNA was
amplified from cDNA of HeLa cells and subcloned in
pCDNA3 with a myc tag at the C terminus (SF1myc in the
text). Site-directed mutagenesis was performed using the
Quikchange protocol from Stratagene (La Jolla, CA, USA).
Protein expression and purification
GST–KIS and mutants were produced as previously des-
cribed [27]. GST–SF1f was purified by glutathione affinity
chromatography, followed by removal of GST using Preci-
sion Protease (Amersham Biosciences) and further purifi-
cation on SP-sepharose (Amersham Biosciences). Purified
fractions were frozen in 200 m m NaCl, 25 mm Hepes
pH 6.8, 20% v ⁄ v glycerol. Plasmid for U2AF

65
was a gift
of M.R. Green, and the histidine-tagged protein was
expressed in bacteria and purified using standard protocols.
GST pull-down
In vitro translations were performed using the TNT system
(Amersham Biosciences) and [
35
S]-methionine (NEN, Bos-
ton, MA, USA). For experiments in Fig. 2, 1 lLofin vitro
translation product was mixed with 250 lL of GST or GST–
SF1f extract corresponding to 5 mL of the BL21 cell cultures,
in GST sonication buffer (GSB; 25 mm Hepes pH 7.5,
100 mm KCl, 1 mm EDTA, 0.1% NP40, 10% glycerol) with
1mm dithiothreitol (DTT) and antiprotease mix from Roche
(Mannheim, Germany). For experiments in Fig. 5, in vitro
translation products were incubated with 1 lg purified GST
SF1 SPSP motif phosphorylation V. Manceau et al.
584 FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS
or GST–SF1 in the presence of 1 lgÆlL
)1
BSA as a nonspe-
cific competitor. After a 90-min incubation at 4 °C, 10 lLof
glutathione beads (Amersham Biosciences) were added for a
further 30 min, beads were washed rapidly five times with
GSB buffer, and proteins were analysed by SDS ⁄ PAGE,
Coomassie blue staining and phosphorimaging (Cyclone,
Packard Instrument Company, Meriden, CT, USA).
Phosphorylation reactions
Phosphorylation reactions were performed as described

previously [27]. Briefly, 20 lL reactions contained  20 ng
recombinant GST–KIS and 1 lg of substrate in 50 mm
Mes pH 8.0, 10 mm MgCl
2
,2mm DTT, 2 mm EDTA,
25% glycerol, 10 lm [ c -
32
P]ATP (5 nCiÆpmole
)1
; NEN).
For substrate and kinase mutant comparison we performed
30-min reactions and checked the linearity of phosphate
incorporation. For stoichiometric phosphorylation we per-
formed 6 h incubations with 0.4 mm ATP.
Mass spectrometry
After phosphorylation by KIS, SF1f was dialysed against
50 mm NH
3
HCO
3
pH 8.0, digested with trypsin, and
HPLC reverse phase fractions were analysed with a
FT-ICR mass spectrometer APEX III (Bruker Daltonics,
Bremen, Germany) equipped with a 7 Tesla supraconduct-
ing magnet and an infinity cell.
Metabolic labelling of cells
HEK293 cells were transfected using lipofectamine reagent
(Invitrogen, Carlsbad, CA, USA). After 24 h, the medium
was replaced with 1 mL fresh medium containing 1 mCi
[

32
P]-inorganic phosphate (NEN) for 4 h. Cells were lysed
in immunoprecipitation buffer containing 50 mm Tris
pH 7.5, 150 mm NaCl, 1% Nonidet P40, 1 mm EDTA,
1mm DTT and protease inhibitors. Immunoprecipitation
was performed with the monoclonal antimyc 9E10 antibody
(Santa Cruz Biotechnology, Santa Cruz, CA, USA).
Phosphopeptide mapping
Proteins were digested ‘in-gel’ overnight with trypsin at
30 °C as previously described [27]. Eluted peptides were
lyophilized and resuspended in electrophoresis buffer
(pH 3.5). After electrophoresis and chromatography, plates
were revealed by phosphorimaging.
Phosphatase treatment
Cell extracts (5 lg of proteins) were treated with the indica-
ted amount of calf intestinal phosphatase (CIP) (New Eng-
land Biolabs, Beverly, MA, USA) for 2 h. SF1myc was
analysed with the monoclonal antimyc 9E10 antibody
(Santa Cruz Biotechnology). Endogenous forms of SF1
were detected with a rabbit polyclonal antibody (Cemines,
Golden, CO, USA).
RNA gel-shift assays
RNA oligonucleotide with sequence derived from the
Adenovirus major late pre-mRNA (5¢-UUCGUGCU
GACCCUGUCCCUUUUUUUUCCACAGC-3¢) was syn-
thesized by Dharmacon (Lafayette, CO, USA). 5¢ labelling
was performed with [c-
32
P]ATP and T4 polynucleotide kin-
ase. Labelled RNA was purified on Nensorb20 column

(NEN). About 10000 cpm ⁄ 2 fmole oligonucleotide was
used with the indicated concentrations of purified proteins
and with tRNA (0.5 l g Æ lL
)1
) as nonspecific competitor.
Interactions were in 15 lL containing 25 mm Tris pH 7.5,
90 mm NaCl and 1 mm EDTA, for 1 h at room tempera-
ture. We checked that equilibrium was reached at this time
point. The mixture was loaded on 6% acrylamide gels in
0.5 · TBE buffer, and run at 4 °C for 3 h.
Acknowledgements
We thank colleagues from INSERM U706, D. Weil
and J. Chamot-Rooke for stimulating discussion and
support. We are grateful for the generous gifts of
U2AF
65
constructs from J. Valca
´
rcel and M.R. Green,
and SF1 cDNA from J.A. Berglund. We thank
S. Lindley for technical assistance. This work was
funded by the ‘Institut National de la Sante
´
et de la
Recherche Me
´
dicale’, the ‘Universite
´
Pierre et Marie
Curie’ the ‘Association Franc¸ aise contre les Myopa-

thies’ and the ‘Association pour la Recherche contre le
Cancer’ to A.S and A.M., and National Institutes of
Health grant (GM070503-01) to C.L.K.
References
1 Brow DA (2002) Allosteric cascade of spliceosome acti-
vation. Annu Rev Genet 36, 333–360.
2 Mermoud JE, Cohen P & Lamond AI (1992) Ser ⁄ Thr-
specific protein phosphatases are required for both cata-
lytic steps of pre-mRNA splicing. Nucl Acids Res 20,
5263–5269.
3 Mermoud JE, Cohen PT & Lamond AI (1994) Regula-
tion of mammalian spliceosome assembly by a protein
phosphorylation mechanism. EMBO J 13, 5679–5688.
4 Tazi J, Kornstadt U, Rossi F, Jeanteur P, Cathala G,
Brunel C & Luhrmann R (1993) Thiophosphorylation
of U1–70K protein inhibits pre-mRNA splicing. Nature
363, 283–286.
5 Huang Y & Steitz JA (2005) SRprises along a messen-
ger’s journey. Mol Cell 17, 613–615.
V. Manceau et al. SF1 SPSP motif phosphorylation
FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS 585
6 Gui JF, Tronchere H, Chandler SD & Fu XD (1994)
Purification and characterization of a kinase specific for
the serine- and arginine-rich pre-mRNA splicing factors.
Proc Natl Acad Sci USA 91, 10824–10828.
7 Wang HY, Lin W, Dyck JA, Yeakley JM, Songyang Z,
Cantley LC & Fu XD (1998) SRPK2: a differentially
expressed SR protein-specific kinase involved in mediat-
ing the interaction and localization of pre-mRNA splicing
factors in mammalian cells. J Cell Biol 140, 737–750.

8 Colwill K, Pawson T, Andrews B, Prasad J, Manley JL,
Bell JC & Duncan PI (1996) The Clk ⁄ Sty protein kinase
phosphorylates SR splicing factors and regulates their
intranuclear distribution. EMBO J 15, 265–275.
9 Seghezzi W, Chua K, Shanahan F, Gozani O, Reed R
& Lees E (1998) Cyclin E associates with components
of the pre-mRNA splicing machinery in mammalian
cells. Mol Cell Biol 18, 4526–4536.
10 Wang C, Chua K, Seghezzi W, Lees E, Gozani O &
Reed R (1998) Phosphorylation of spliceosomal protein
SAP 155 coupled with splicing catalysis. Genes Dev 12,
1409–1414.
11 Shin C & Manley JL (2004) Cell signalling and the con-
trol of pre-mRNA splicing. Nat Rev Mol Cell Biol 5,
727–738.
12 Loyer P, Trembley JH, Katona R, Kidd VJ & Lahti JM
(2005) Role of CDK ⁄ cyclin complexes in transcription
and RNA splicing. Cell Signal 17, 1033–1051.
13 Kramer A (1992) Purification of splicing factor SF1, a
heat-stable protein that functions in the assembly of a
presplicing complex. Mol Cell Biol 12, 4545–4552.
14 Abovich N & Rosbash M (1997) Cross–intron bridging
interactions in the yeast commitment complex are con-
served in mammals. Cell 89, 403–412.
15 Berglund JA, Abovich N & Rosbash M (1998) A coop-
erative interaction between U2AF65 and mBBP ⁄ SF1
facilitates branchpoint region recognition. Genes Dev 12,
858–867.
16 Rain JC, Rafi Z, Rhani Z, Legrain P & Kramer A
(1998) Conservation of functional domains involved in

RNA binding and protein–protein interactions in
human and Saccharomyces cerevisiae pre-mRNA spli-
cing factor SF1. RNA 4, 551–565.
17 Berglund JA, Chua K, Abovich N, Reed R & Rosbash
M (1997) The splicing factor BBP interacts specifically
with the pre-mRNA branchpoint sequence UACUAAC.
Cell 89, 781–787.
18 Gozani O, Potashkin J & Reed R (1998) A potential
role for U2AF–SAP 155 interactions in recruiting U2
snRNP to the branch site. Mol Cell Biol 18, 4752–4760.
19 Wu J & Manley JL (1989) Mammalian pre-mRNA
branch site selection by U2 snRNP involves base pair-
ing. Genes Dev 3, 1553–1561.
20 Zhuang Y & Weiner AM (1989) A compensatory base
change in human U2 snRNA can suppress a branch site
mutation. Genes Dev 3, 1545–1552.
21 Wang X, Bruderer S, Rafi Z, Xue J, Milburn PJ, Kra-
mer A & Robinson PJ (1999) Phosphorylation of spli-
cing factor SF1 on Ser20 by cGMP-dependent protein
kinase regulates spliceosome assembly. EMBO J 18,
4549–4559.
22 Kielkopf CL, Rodionova NA, Green MR & Burley SK
(2001) A novel peptide recognition mode revealed by
the X-ray structure of a core U2AF35 ⁄ U2AF65 het-
erodimer. Cell 106, 595–605.
23 Selenko P, Gregorovic G, Sprangers R, Stier G, Rhani
Z, Kramer A & Sattler M (2003) Structural basis for
the molecular recognition between human splicing fac-
tors U2AF65 and SF1 ⁄ mBBP. Mol Cell 11, 965–976.
24 Kielkopf CL, Lucke S & Green MR (2004) U2AF

homology motifs: protein recognition in the RRM
world. Genes Dev 18, 1513–1526.
25 Maucuer A, Camonis JH & Sobel A (1995) Stathmin
interaction with a putative kinase and coiled-coil-form-
ing protein domains. Proc Natl Acad Sci USA 92, 3100–
3104.
26 Maucuer A, Ozon S, Manceau V, Gavet O, Lawler S,
Curmi P & Sobel A (1997) KIS is a protein kinase with
an RNA recognition motif. J Biol Chem 272, 23151–
23156.
27 Maucuer A, Le Caer JP, Manceau V & Sobel A (2000)
Specific Ser-Pro phosphorylation by the RNA-recogni-
tion motif containing kinase KIS. Eur J Biochem 267,
4456–4464.
28 Boehm M, Yoshimoto T, Crook MF, Nallamshetty S,
True A, Nabel GJ & Nabel EG (2002) A growth factor-
dependent nuclear kinase phosphorylates p27 (Kip1)
and regulates cell cycle progression. EMBO J 21, 3390–
3401.
29 Bieche I, Manceau V, Curmi PA, Laurendeau I,
Lachkar S, Leroy K, Vidaud D, Sobel A &
Maucuer A (2003) Quantitative RT-PCR reveals a
ubiquitous but preferentially neural expression of the
KIS gene in rat and human. Brain Res Mol Brain
Res 114, 55–64.
30 Valcarcel J, Gaur RK, Singh R & Green MR (1996)
Interaction of U2AF65 RS region with pre-mRNA
branch point and promotion of base pairing with U2
snRNA [corrected]. Science 273, 1706–1709.
31 Shen H & Green MR (2004) A pathway of sequential

arginine-serine-rich domain–splicing signal interactions
during mammalian spliceosome assembly. Mol Cell 16,
363–373.
32 Arning S, Gruter P, Bilbe G & Kramer A (1996) Mam-
malian splicing factor SF1 is encoded by variant cDNAs
and binds to RNA. RNA 2, 794–810.
33 Beausoleil SA, Jedrychowski M, Schwartz D, Elias JE,
Villen J, Li J, Cohn MA, Cantley LC & Gygi SP (2004)
Large-scale characterization of HeLa cell nuclear phos-
phoproteins. Proc Natl Acad Sci USA 101, 12130–
12135.
SF1 SPSP motif phosphorylation V. Manceau et al.
586 FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS
34 Shu H, Chen S, Bi Q, Mumby M & Brekken DL
(2004) Identification of phosphoproteins and
their phosphorylation sites in the WEHI-231 B
lymphoma cell line. Mol Cell Proteomics 3, 279–
286.
35 Lu KP, Liou YC & Zhou XZ (2002) Pinning down pro-
line-directed phosphorylation signaling. Trends Cell Biol
12, 164–172.
36 Bennett M, Michaud S, Kingston J & Reed R (1992)
Protein components specifically associated with prespli-
ceosome and spliceosome complexes. Genes Dev 6,
1986–2000.
37 Peled-Zehavi H, Berglund JA, Rosbash M & Frankel
AD (2001) Recognition of RNA branch point sequences
by the KH domain of splicing factor 1 (mammalian
branch point binding protein) in a splicing factor com-
plex. Mol Cell Biol 21, 5232–5241.

38 Zamore PD & Green MR (1989) Identification, purifica-
tion, and biochemical characterization of U2 small
nuclear ribonucleoprotein auxiliary factor. Proc Natl
Acad Sci USA 86, 9243–9247.
39 Zamore PD, Patton JG & Green MR (1992) Cloning
and domain structure of the mammalian splicing factor
U2AF. Nature 355, 609–614.
40 Huang T, Vilardell J & Query CC (2002) Pre-spli-
ceosome formation in S. pombe requires a stable
complex of SF1-U2AF(59)-U2AF(23). EMBO J 21,
5516–5526.
41 Abovich N, Liao XC & Rosbash M (1994) The yeast
MUD2 protein: an interaction with PRP11 defines a
bridge between commitment complexes and U2 snRNP
addition. Genes Dev 8, 843–854.
42 Rutz B & Seraphin B (1999) Transient interaction of
BBP ⁄ ScSF1 and Mud2 with the splicing machinery
affects the kinetics of spliceosome assembly. RNA 5,
819–831.
43 Rain JC & Legrain P (1997) In vivo commitment to
splicing in yeast involves the nucleotide upstream from
the branch site conserved sequence and the Mud2 pro-
tein. EMBO J 16, 1759–1771.
44 Tanackovic G & Kramer A (2005) Human splicing
factor SF3a, but not SF1, is essential for pre-mRNA
splicing in vivo. Mol Biol Cell 16, 1366–1377.
45 Guth S & Valcarcel J (2000) Kinetic role for mamma-
lian SF1 ⁄ BBP in spliceosome assembly and function
after polypyrimidine tract recognition by U2AF. J Biol
Chem 275, 38059–38066.

46 Banerjee H, Rahn A, Gawande B, Guth S, Valcarcel J
& Singh R (2004) The conserved RNA recognition
motif 3 of U2 snRNA auxiliary factor (U2AF 65) is
essential in vivo but dispensable for activity in vitro.
RNA 10, 240–253.
47 Rutz B & Seraphin B (2000) A dual role for BBP ⁄
ScSF1 in nuclear pre-mRNA retention and splicing.
EMBO J 19, 1873–1886.
48 Zhang D & Childs G (1998) Human ZFM1 protein is a
transcriptional repressor that interacts with the tran-
scription activation domain of stage-specific activator
protein. J Biol Chem 273, 6868–6877.
49 Zhang D, Paley AJ & Childs G (1998) The transcrip-
tional repressor ZFM1 interacts with and modulates the
ability of EWS to activate transcription. J Biol Chem
273, 18086–18091.
50 Goldstrohm AC, Albrecht TR, Sune C, Bedford MT &
Garcia-Blanco MA (2001) The transcription elongation
factor CA150 interacts with RNA polymerase II and
the pre-mRNA splicing factor SF1. Mol Cell Biol 21,
7617–7628.
51 Gama-Carvalho M, Carvalho MP, Kehlenbach A, Val-
carcel J & Carmo-Fonseca M (2001) Nucleocytoplasmic
shuttling of heterodimeric splicing factor U2AF. J Biol
Chem 276, 13104–13112.
52 Blanchette M, Labourier E, Green RE, Brenner SE &
Rio DC (2004) Genome-wide analysis reveals an unex-
pected function for the Drosophila splicing factor
U2AF50 in the nuclear export of intronless mRNAs.
Mol Cell 14, 775–786.

53 Kramer A, Quentin M & Mulhauser F (1998) Diverse
modes of alternative splicing of human splicing factor
SF1 deduced from the exon-intron structure of the gene.
Gene 211, 29–37.
54 Liu Z, Luyten I, Bottomley MJ, Messias AC, Houngni-
nou-Molango S, Sprangers R, Zanier K, Kramer A &
Sattler M (2001) Structural basis for recognition of the
intron branch site RNA by splicing factor 1. Science
294, 1098–1102.
V. Manceau et al. SF1 SPSP motif phosphorylation
FEBS Journal 273 (2006) 577–587 ª 2006 The Authors Journal compilation ª 2006 FEBS 587