r d‘.. . ‘
Ami?

v
‘5.

may”.

 

r... 5:
.36}. .. ,
.I 7!:
x I

 

I
... f.
1.}- .l
Led

. i-

f .o3.¢l.

«ouch

aft #5?” ‘ . gamma

.u .T 3.1 .5 A... .2 ,

 

 

 

Luli—Jill‘lll

" h'r‘aw-r‘v 0*.»3 43
h’iiCa 35:11:! Qtaib

I Efl'fizr'; ‘\ r'(‘l‘:'-
I 151:4’5 h-’lt.

 

 

 

This is to certify that the
dissertation entitled

Characterization of galectin-a-snRNP complexes and
mechanism of galectin entry into the splicing pathway

presented by

Kevin C. Haudek

has been accepted towards fulfillment

of the requirements for the
PhD. degree in Biochemistry and Molecular
Biology

 

 

?mu STQW

 

Major Professor’s Signature

lo / r5107

 

Date

MSU is an affinnative-action. equal-opportunity employer

_-... _-—._-—.-.-.—-—.-a-.—-o—o-o-.-n-o---o-c--o-o--—o.

.---o--o-o--n--o-o-n—.--.-u--.—-_.—.-._.-_-._.—._._._--.—--—.-._a_ ...t

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDUE

DAIEDUE

DAIEDUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/C|RC/DateDue.indd-p.1

 

Characterization of galectin-3-snRNP complexes and
mechanism of galectin entry into the splicing pathway

By

Kevin C. Haudek

A DISSERTATION
Subnnfledto
Michigan State University
in partial fulfillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY
Department of Biochemistry and Molecular Biology

2007

Abstract

CHARACTERIZATION OF GALECTIN-S-SNRNP COMPLEXES AND
MECHANISM OF GALECTIN ENTRY INTO THE SPLICING PATHWAY

By
Kevin C. Haudek

In previous studies, galectin-l and galectin-3 have been found to be redundant
pre—mRNA splicing factors using a cell free splicing system. In addition,
immunoprecipitation experiments showed both galectin—l and -3 assembled onto the
spliceosome, from early to late stages of complex formation. Results presented here
focus on the role of galectin-3 outside of the spliceosome and the mechanism of its
entry into the pre-mRN A splicing pathway. Irnmunoprecipitation experiments of
HeLa nuclear extract showed the association of galectin-3 with 5 snRNAs (U1, U2,
U4, U5 and U6) and associated snRNP proteins (Sm core polypeptides and U1
specific protein U1 -70K). In addition, galectin-3 was complexed with other RNA
processing proteins including PSF (PTB-associated splicing factor), general
transcription factor TFII-I and the SMN (survival of motor neuron) protein.
Fractionation of HeLa nuclear extract on glycerol gradients showed a co-
sedimentation of galectin-3 with multiple snRNP complexes. In particular, one
complex (~IOS) showed an RNase A sensitive association between U1 snRNP and
galectin-3. Because U1 snRNP is known to bind the pre-mRNA first at the 5' splice
site in a step-wise assembly of spliceosomes, we tested whether this isolated galectin-
3-U1 complex was sufficient to recognize and load galectin-3 onto pre-mRNA. Our
results suggest that galectin-3 enters the pre-mRN A splicing pathway with U1 under

conditions that allow early splicing complexes to form.

We then tested whether galectin—3 is a bonafide component of the early
splicing complex (B complex). To test this, we used a galectin-independent means of
spliceosome selection. A novel pre-mRNA containing three recognition sites for the
M82 bacteriophage protein has been engineered and used to select specific splicing
complexes. Using this system and differing incubation conditions, we investigated
whether we could isolate unique splicing complexes formed on the pre-mRN A. Of
particular interest was the isolation of the early spliceosome complex containing U1
snRNP. Once we had confirmed the specificity of complex formation and isolation,
we analyzed proteins selected on the pre-mRNA formed under conditions to allow the
formation of B complex. Galectin-3 was identified as a protein member of the B
complex. Our results from RNA selection methods validate galectin-3 as a member

of early splicing complexes containing U1 snRNP.

Acknowledgements

I would like to thank my mentor, Dr. Ron Patterson, for his invaluable support
and guidance in my graduate education. I appreciate his help and understanding
through all situations during my graduate career.

I would also like to thank Dr. John Wang for his assistance in developing me
as a scientist. I also appreciate his help in organizing and finalizing this thesis.

Finally, I would like to thank Penny and my family and friends for their
encouragement and support during this endeavor.

iv

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... vi

LIST OF FIGURES ............................................................................. vii

CHAPTER 1: Literature Review
I. Galectins ................................................................................ 1
A. Family overview ..................................................................... 1
B. Intracellular galectin-l .............................................................. 4
C. Intracellular galectin—3 .............................................................. 6
II. SnRNPs ................................................................................. 9
A. Biogenesis and assembly ............................................................ 9
B. Nuclear multi-snRNP complexes ................................................ 12
C. Other nuclear multi-snRNP complexes .......................................... 14
III. Pre-mRNA splicing ................................................................ 18
A. Spliceosome ........................................................................ 18
B. Assembly ............................................................................ 20
C. Proteomics ........................................................................... 23
IV. Galectins and pre-mRNA splicing ............................................... 26
References ............................................................................... 28

CHAPTER 2: Physico—chemical characterization of galectin-3-snRNP complexes
and role in pre-mRNA binding

Abstract .................................................................................. 44
Introduction .............................................................................. 44
Materials and Methods ................................................................. 46
Results .................................................................................... 52
Discussion ............................................................................... 67
References ............................................................................... 77

CHAPTER 3: Identification of galectin-3 in aptamer-selected early splicing

complexes
Abstract .................................................................................. 80
Introduction .............................................................................. 80
Materials and Methods ................................................................. 82
Results .................................................................................... 88
Discussion ............................................................................... 95
References .............................................................................. 101

CHAPTER 4: Concluding statements ................................................... 104

LIST OF TABLES
Chapter 2

Table 1. Listing of selected RNPs involved in pre-mRNA splicing and reported S
values .................................................................................... 63

vi

LIST OF FIGURES

Chapter 1
Figure 1.1. Schematic diagram illustrating the domain structure of the galectin family
subgroups ................................................................................. 3
Chapter 2
Figure 2.1. Analysis of nuclear RNA and proteins immunoprecipitated by anti-
Ga13 ....................................................................................... 54
Figure 2.2 Analysis of proteins immunoprecipitated by anti-TMG ....................... 57
Figure 2.3. Analysis of nuclear RNA and proteins separated on a 12-32% glycerol
gradient .................................................................................... 60
Figure 2.4. Analysis of RNA and proteins immunoprecipitated by anti-Gal3 from
glycerol gradient fractions ............................................................ 66

Figure 2.5. Effect of RNase A treatment of Gal3 association with U1 snRNP. . . . . ....69
Figure 2.6. Immunoprecipitation of radio-labeled pre—mRNA from glycerol gradient

fractions by anti-G313 .................................................................. 72
Figure 2.7. Diagram showing the association of Gal3 with snRNPs and the pre-
mRNA splicing substrate ............................................................... 75
Chapter 3
Figure 3.1. Schematic diagram illustrating the structure of AdML-M3 pre-mRNA and
its use in purifying spliceosomal components assembled on the RNA ......... 84
Figure 3.2. Specificity of selection by MBP-MSZ fusion protein ........................ 91
Figure 3.3. Analysis of protein and RNA components of early and active splicing
complexes selected via the AdML-M3 pre-mRN A ................................ 94
Figure 3.4. Analysis of proteins associated with early splicing complexes ............. 97

vii

Chapter 1
Literature Review

I. Galectins

A. Family overview

Members of the galectin family of proteins share two essential characteristics:
one, they have binding affinity for B-galactosides and two, they contain conserved
residues in the carbohydrate binding domain [1]. So far, fifteen mammalian proteins
have been classified as galectins. Other organisms, ranging from amphibians to fungi,
also have been found to contain galectins. However, other organisms vary in the number
and identity of galectins they contain [2].

The galectin family is divided into three sub-groups based on their domain
architecture (see Figure 1). The Prototype sub-group consists of members possessing a
single carbohydrate recognition domain (CRD) of approximately 130 amino acids. The
Tandem Repeat sub-group consists of members possessing two CRDs fused together by a
short linker region. The final sub-group, the Chimera, consists of a sole member,
galectin-3 (Gal3) which contains a CRD with a unique N-terminal domain
(approximately 130 amino acids). This N-terminal domain consists of a series of proline-
and glycine-rich repeats. Differential scanning calorimetry has determined the melting
temperature of the N-domain (~40°C) and the CRD (~5 5°C) of Gal3, which suggest the
domains fold independently of one another [3]. However, recent NMR evidence suggests
that portions of the N-domain may contact the CRD [4].

X-ray crystallography has been used to determine the three-dimensional structure

of the CRD from a range of galectin family members, including members of all three

Figure l. (A) Schematic diagram illustrating the domain structure of the
galectin family subgroups. Conserved amino acid residues in the CRD are
shown. The unique N-terminal domain of galectin—3 contains repeats rich
in proline and glycine residues. The single-letter amino acid code is used,
where X denotes any amino acid. (B) Illustration of the X-ray crystal
structure of human galectin-7. The overall structure of the CRD when
bound to ligand is shown by the polypeptide backbone (gray). Highlighted
and numbered residues are involved in binding of saccharide ligands. The
saccharide ligand is not shown in the figure.

HNR NWER

 

 

 

 

 

 

 

 

A. \\\ //
l ”I II I Prototype
CRD Galectin-l, —2, -5, -7,
PGAYPGXX 40, 41, ~13, 44, -15
l JHHH J Chimera
CRD
Galectin-3
L H Tandem Repeat
CRD CRD Galectin-4, -6,
-8, -9, -12
N-terminus C-terminus
B.

 

subgroups [5-7]. The overall structure contains two anti-parallel B-sheets in a sandwich-
like structure (see Figure 1B). The carbohydrate binding site consists of amino acids
contained on the side of the sandwich-like structure. The highlighted residues in Fig. 13
show the amino acids in human galectin-7 that are conserved among the galectin family
and make contact with the saccharide ligand [5]. It should be noted that the affinity of
galectins for the monosaccharide galactose is fairly weak [8]. Larger saccharides, such as
the disaccharide lactose or some oligosaccharides, bind to galectins with greater affinity.
This may indicate that these larger ligands can make additional contacts with the galectin
outside of the conserved residues in the CRD. However, these additional contacts are far
less conserved in the galectin polypeptide and different galectins exhibit different
affinities for these larger ligands [9].

Interestingly, many members of the galectin family exhibit dual-localization, that
is, they can be found both intra- and extracellularly. Inside the cell, galectins have been
found to localize both in the cytoplasm and the nucleus [9, 10]. The mechanism by
which the galectins are moved to the extracellular space is poorly understood as none of
the galectins contain an identifiable signal sequence for extemalization via the
endomembrane system [10, 11].

B. Intracellular galectin-1

Galectin-l (Gall) is a member of the Prototype sub-group and has been found in
the cytoplasm of various cell types [12-15]. It should be noted that the localization of
Gall can change on the basis of cell differentiation. For example, Cooper and Barondes
stained myoblasts for Gall and showed an intracellular distribution of the protein [16].

As myoblasts fused into myotubes, Gall was exported to the extracellular space via a

novel secretory pathway involving blebs of the plasma membrane. In studies using H-
Ras(l2V)-transformed Rat-1 (EJ) cells, Gall has been shown to interact with H-
Ras(12V) in the cytoplasm [17]. This interaction leads to membrane anchorage of H-Ras
and promotes cell transformation. The interaction between Gall and H-Ras is thought to
be direct due to several lines of evidence including UV cross-linking and reciprocal
immunoprecipitations [17, 18]. To prevent cell transformation by H-Ras( 1 2V), anti-
sense RNA for Gall was transfected into baby hamster kidney cells. These cells showed
a reduction in H-Ras(12V) clustering in non-rafi microdomains at the plasma membrane
[17, 18]. This is consistent with the fact that membrane anchorage of H-Ras is necessary
for cell transformation. Finally, a mutant form of Gall (L1 1A) shows normal
carbohydrate binding activity but mislocalizes H-Ras(12V) and prevents H-Ras(12V)
GTP-loading in COS-7 cells [19]. Transfections of Rat-1 cells with this Gall mutant
show reduced levels of cell transformation [19]. A recent report indicates that like Gall,
Gal3 may have a similar binding pocket to interact with Ras [20].

Other reports have found Gall localized in the nucleus of cells as well. Choi et al.
reported the association of Gal] with the nuclear matrix of rat calvarial osteoblasts [21].
This interaction is also dependent on cell development, as Gall was found in nuclear
matrix preparations only in differentiated osteoblasts. In other studies, HeLa cervical
carcinoma cells showed Gall in the nucleus and co-localized with known splicing factors
(such as the Sm core proteins of snRNPs, see below) by immunofluorescence [22].
Using the Gall protein as bait in a yeast-two-hybrid system and screening a HeLa cell
cDNA library identified the C—terminal 50 amino acids of gemin4 (Gemin4C50) as a

binding partner [23]. This was confirmed by direct binding assays of Gall and a fusion

protein containing Gemin4C50. Gemin4 can be found both in the cytoplasm and nucleus
of HeLa cells. However immunoprecipitation experiments of HeLa nuclear extracts (NE)
using antibodies against Gall co-immunoprecipitate other known gemin4 interacting
proteins such as the Survival of Motor Neuron protein (SMN, see below) and gemin2
[23]. This strongly suggests that Gall does interact with gemin4 in the nucleus, but does
not rule out the possibility of additional interactions in the cytoplasm. Gall also has been
shown to play a role in pre-mRNA splicing using HeLa NE [24]. Extracts depleted of
both Gall and Gal3 cannot catalyze the splicing reaction of an exogenous pre-mRNA.
Adding back recombinant Gall protein reconstitutes the ability of the extract to carry out
pre-mRNA splicing. As well, addition of known carbohydrate ligands of the galectins to
HeLa NE inhibits the splicing ability of that extract [25].
C. Intracellular galectin-3

Several reports place Gal3 in both the cytoplasm and nucleus of cells using a
variety of microscopy techniques [26-28]. This cellular distribution of Gal3 can change
according to cell proliferation state [29]. Phosphorylation of Gal3 also affects its cellular
distribution. In 3T3 cells, Gal3 exists in two variants, a non-phosphorylated form which
is entirely nuclear, and a phosphorylated form that can be found in both the nucleus and
cytoplasm [30]. It has also been shown that the phosphorylated form of nuclear Gal3 is
quickly exported to the cytoplasm as part of a large complex, indicating the
phosphorylation of Gal3 may play a key role in nuclear export [31]. Further studies of
the intracellular transport of Gal3 have determined it to be a shuttling protein, that is, it
moves back and forth between the nucleus and cytoplasm of cells [32]. This was

demonstrated by creating heterodikaryons of mouse and human cells and monitoring the

appearance of the human G313 polypeptide in the mouse nucleus [32]. Additional
research has been focused on the nuclear import and export signal of Gal3. It has been
found that the necessary nuclear import signal lies in a conserved sequence of amino
acids, ITLT, beginning at residue 253 [33]. Site-directed mutants of this sequence show
loss of nuclear accumulation of Gal3. An overlapping and neighboring leucine-rich
region, at residues 240-255, mediates the nuclear export of Gal3 [34]. Using this export
signal, a green fluorescent protein (GFP) construct, normally constrained to the nucleus,
is efficiently exported to the cytoplasm. GFP-Gal3 mutants of this export signal show
nuclear accumulation by fluorescent microscopy [34]. This leucine-rich export signal
indicates that Gal3 may be exported in a CRM-l dependant manner, consistent with
previous reports [32]. In contrast, a report by Nakahara et al. found that Ga13 is imported
into the nucleus via importin-alpha, using a nuclear localization signal at residues 223-
228[35]

Gal3 has been documented as having several intracellular binding partners.
. Known binding partners for Gal3 include Bel-2, Chrp, TTF-I, beta-catenin, CBP70,
cytokeratin and gemin4. The apoptosis repressor Bel-2 contains two regions of similarity
to Gal3; the N-terminus is proline and glycine rich and the C-terminal portion shares the
amino acid sequence NWGR [36]. Bel-2 binds directly to Gal3 in vitro and mutants of
Bel-2 in the NWGR sequence eliminate its anti-apoptotic activity [37]. Chrp was
originally identified as a Gal3 binding partner by a yeast-two-hybrid screen of a cDNA
library from mouse 3T3 cells [3 8]. This interaction was then confirmed by in vitro
binding assays and immunoprecipitations. It also was found that Chrp interacts at the

CRD of Gal3 and this interaction is not inhibitable by carbohydrate ligands of Gal3 [39].

It should also be noted Chrp may only be a cytosolic binding partner of Gal3, in that
immunofluorescence shows Chrp localized in the perinuclear space and cytoplasm of 3T3
cells [3 8]. The thyroid-specific transcription factor TTF-I has been identified as a nuclear
binding partner of Gal3 in papillary thyroid cancer cells [40]. Direct interaction between
the homeodomain of TTF-I and Gal3 was shown using GST-pull down assays. This
association with Gal3 enhances the ability of TTF-I to bind DNA in a gel-retardation
assay [40]. Gal3 was co-purified with another intracellular lectin, CBP70, from HL6O
nuclei when isolated on saccharide affinity beads [41]. Binding of Ga13 to lactose
disrupts the interaction between CBP70 and Gal3 [42]. Although some functions have
been assigned to CBP70, the significance of the Gal3-CBP7O interaction remains to be
uncovered. Gal-3 also has been reported to bind beta-catenin, determined by
immunoprecipitations and in vitro binding assays [43]. Localization studies of these two
proteins show a strong correlation and Gal-3 may activate beta-catenin stimulation of
cyclin and c-myc expression [28]. There is evidence to suggest an in vitro association of
cytokeratins with Gal3, due to novel glycosylated residues on the cytokeratins [44]. This
would implicate the cytokeratins as true intracellular carbohydrate ligands of Gal3 [10].
Like Gall, Gal3 has been described as a component of the nuclear matrix, however its
binding partner has not been identified [45]

Gal3, similar to Gall, also has been found to bind a fusion construct of
Gemin4C50 directly, using an in vitro binding assay [23]. Whether this indicates an
involvement of Gal3 (similar to Gall) with the SMN protein and complex remains to be
investigated. However, adding the purified N-domain of Gal3 to an in vitro splicing

assay inhibits the splicing reaction, suggesting that Gal3 makes important contacts

necessary for pre-mRNA splicing [23]. Equally interesting, GST-Gemin4C50 also acts

in a dominant negative manner when added to splicing extracts.
II. SnRNPS

A. Biogenesis and assembly

Five small nuclear RNAs (snRNAs) participate directly in pre-mRNA splicing;
U1, U2, U4, U5 and U6. These five snRNAs are closely conserved among vertebrates
and contain a large percentage of uracil bases and are commonly referred to as U
snRN As. These snRN As are commonly found complexed with proteins in the nucleus,
creating a small nuclear ribonucleoprotein particle (snRNP) [46]. Each snRNP contains
an unique snRNA, a common set of Sm core proteins and snRNP-specific proteins unique
only to that type of snRNP when isolated under stringent conditions [47]. The snRNAs
undergo an extensive biogenesis pathway before they can participate in pre-mRN A
splicing [48].

Transcription of most of the U snRNAs is carried out by RNA Polymerase H (Pol
11). U6 snRNA, which is a RNA Polymerase 111 (P01 HI) transcript, is the lone exception
(see below). The Pol H transcribed U snRNAs are made as long primary transcripts (pre-
snRNAs) with a monomethyl-guanosine cap structure [49]. Trimming of the pre-
snRNAs requires both phosphorylation of the C-terminal domain of Pol H and a cis-
acting signal in the transcript [50]. The snRNAs are then packaged for export with
proteins including the cap-binding complex (CBC); PHAX, which is found
phosphorylated in the nucleus; and CRMl/RanGTP. After export from the nucleus,
hydrolysis of Ran-bound GTP and dephosphorylation of PHAX induces disassembly of

the snRNA export complex [51, 52].

Assembly of the snRNAs into snRNPs occurs in the cytoplasm. The key factor in
the cytoplasmic assembly and biogenesis of snRNPs is the SMN protein complex [53,
54]. This complex is responsible for loading the full complement of Sm proteins (B/B',
D1-3, E, F, G) onto the consensus snRNA Sm site to form the core snRNP [54, 55]. The
Sm proteins are known to form a heptamer ring when bound to the snRNA. However,
the Sm proteins are loaded onto the snRNA in discrete sub-complexes, not in a fully
assembled heptamer ring [56, 57]. Some evidence suggests that the SMN protein, using
its Tudor domain, binds the Sm proteins, by recognizing the symmetrical dimethylated
arginines contained in the Sm proteins [5 8-60]. Another member of the SMN complex,
Gemin5, is responsible for binding the snRNAs during loading of the Sm core proteins
[61, 62]. The binding of the Sm proteins to the snRNA allows further modification of the
snRNP including hypennethylation of the cap structure. The monomethyl guanosine cap
is modified to become a trimethyl guanosine cap by Tgsl , a methyltransferase, which has
also been shown to interact with the SMN complex [63, 64].

After the snRN A has been loaded with Sm proteins and had its cap
hypermethylated, it is ready to be imported back to the nucleus. The snRNPs are
imported into the nucleus via a series of adapter proteins [65]. The snurportin-l protein
(SPNl) is responsible for recognizing and binding the hypermethylated cap of the
snRNPs [66]. SPNl binds the snRNP cap most likely near the nuclear pore complex
(NPC) and has been found to interact with a NPC protein, Nup214 [67]. SPNl also has a
domain for interacting with importin-B, a nuclear import factor, and at some point SPNl
binds importin-B and releases Nup214 [65, 67]. Importin-B then mediates another

interaction with the NPC to allow translocation of the snRNP into the nucleus [67]. In

10

other importin—B systems, importin-B binds RanGTP in the nucleus to dissociate its
cargo; however, this does not seem to be the case for snRNP import [68]. The
mechanism of dissociation of the snRNPs after import remains under investigation.
Some evidence suggests that the cytoplasmic SMN complex (or sub-complexes thereof)
is transported along with snRNPs into the nucleus through its affinity for the Sm proteins
[69, 70]. Other evidence suggests a second import pathway for snRNPs which
recognizes the assembled Sm core on the snRNP as opposed to the hypermethylated cap
structure [71].

After import into the nucleus, the snRNPs are trafficked to the Caj a1 bodies.
Coilin is a key protein component of the Caj al body and can, in most cell types, directly
bind the SMN protein in Cajal bodies [72]. It is hypothesized that the SMN binding
event to coilin is what carries the imported snRNPs to the Caj al bodies [73]. Once at the
Cajal body the imported snRNPs bind any necessary snRNP-specific protein, which enter
the nucleus independent of the snRNP. This completes assembly of the protein
components with the snRNAs.

Another important step in snRNP biogenesis at the Caj a] body is modification of
the snRNA. SnRNAs can be either 2'-O-methylated or pseudouridylated by small Cajal
body-specific RNAs (scaRNA) [74]. These scaRNAs resemble small nucleolar RNAs
(responsible for modifying ribosomal RNA) in their sequence and structure [75]. The
modifications performed on snRNPs are essential for their fimction in pre-mRN A
splicing [76]. After the snRNPs have been assembled and modified they are moved to
nuclear speckles, which are thought of as storage sites for various splicing factors [73,

77].

11

In contrast to the other U snRNAs, biogenesis of U6 snRNA is confined to the
nucleus and transcribed by Pol III. The transcribed U6 snRNA is given a unique y-
monomethyl cap structure, which it keeps throughout its lifecycle [48]. Transcription of
the U6 gene is terminated by a short stretch of uridines. There is evidence to suggest the
La protein binds this 3' U-stretch to give stability to the U6 snRN A and allow the core
snRNP to assemble [78, 79]. Another unique aspect of U6 is that it binds a set of proteins
called the Lsm (Like-Sm) proteins, Lsm2-8, in place of the Sm proteins [80]. These Lsm
proteins form a heptamer ring around the 3' U—stretch in the U6 snRN A and share
homology with the Sm proteins [81, 82]. There is some evidence that links the Lsm
proteins to recycling or regenerating snRNPs between rounds of pre-mRNA splicing [83].
Before being transported to nuclear speckles with the other U snRNAs, U6 also visits the
Caj a1 bodies and nucleolus [84, 85]. Binding of the Lsm proteins to U6 is required to
target the U6 snRNA to the nucleolus to undergo base modifications directed by
snoRNPs [85, 86].

B. Nuclear multi-snRNP complexes

SnRNPs are found in other nuclear complexes in addition to those described
during their biogenesis and role in the active spliceosome. In fact, snRNPs often
associate with one another in the nucleus in discrete complexes outside of the
spliceosome. These multi-snRNP complexes are loaded as a large particle when used in
pre-mRNA splicing, in particular the U4/U6.U5 particle.

1. U4/U 6
The U4/U6 di-snRNP was the first multi-snRNP complex to be identified [87].

The importance of this particle became obvious when it was found that disruption of the

12

basepairing between U4 and U6 or cleavage of the individual snRNAs halted pre-mRN A
splicing [88]. The basepaired U4/U6 exists as a 12S particle and contains specific
polypeptides, termed U4/U6-specific proteins, not found on either snRNP alone [89]. A
U4-specific protein, 15.5K, must be bound to U4 before loading of the U4fU6-specific
proteins, suggesting it may be responsible for additional protein loading [90]. p110
(SART3) has been identified as the protein responsible for annealing free U6 snRNP with
U4 to form the di-snRNP complex [91]. This is an important process to create functional
U4/U6 particles between rounds of pre-mRNA splicing. The formation of these U4/U 6
particles occurs in Caj a1 bodies, a conclusion based on the results of microscopy and
mutation studies [92, 93].
2. U4fU6.U5 tri-snRNP

A U4/U6.U5 particle was originally isolated using immuoaffinity and
sedimentation centrifugation [94]. This tri-snRNP particle was found to be about 258
and could not be reconstituted by mixing free US particle with the 12S U4/U 6 particle.
This is because the tri-snRNP contains proteins only associated with the U4/U6.U5
particle and not with any individual snRNP alone [94, 95]. It is thought that assembly of
this U4/U6.U5 tri-snRNP occurs in the Cajal bodies. Using RNAi to knock down
specific U4/U 6 or U5 proteins prevents the assembly of the tri-snRNP, while
accumulating assembled U4/U 6 particles in the Cajal bodies [96]. Recent reports have
described a tri-snRNP assembly factor associated with US [97]. Interestingly, this protein
is associated with the free US particle, but is not found associated with the 25S tri-
snRNP. It is suggested that this protein participates in loading U5 and the tri-snRNP

specific proteins onto U4/U 6 then is released from the complex. Other studies have

13

elucidated a complete protein-protein interaction map of the tri-snRNP by doing a series
of yeast-two-hybrid screens [98]. Evidence suggests that the U5 interaction with U4/U 6
is based solely on protein-protein contacts, as no RNA-RNA interactions could be
detected.
3. Pseudospliceosome

A large complex containing four of the snRNPs, U2, U4, U5 and U6, has been
isolated from HeLa cell nuclear extract [99, 100]. This complex, termed the
pseudospliceosome is formed in the presence of ATP and incubation at 30°C, conditions
similar to those used for in vitro splicing assays. Other research suggests the
pseudosplicesome forms due to the presence of a 5' splice site [99]. However, U1 is
dispensable for the formation of the U2/U4/U S/U 6 complex. The pseudospliceosome,
although similar to the B splicing complex (see below), is unique from bonafide splicing
complexes based on native gel mobility shifts and salt conditions used for isolation [100].
C. Other nuclear snRN P complexes

Using less stringent conditions to isolate snRNPs and their associated proteins
allows the characterization of several other nuclear complexes. These particles show the
association of multiple snRNAs or snRNP-proteins with non-snRNP proteins in large
complexes.
1. SMN

The SMN protein is responsible for assembling snRNP particles in the cytoplasm
and mutant forms of the SMN protein are the causative agent of spinal muscular atrophy
(SMA) [54, 101, 102]. A complex involving the full length SMN polypeptide and

several other proteins was originally identified in the nucleus, in a novel nuclear structure

14

termed a gem (for "gemini of Cajal bodies") [103]. As this nuclear SMN complex was
characterized it was found that the complex consisted of SMN, gemins 2-8, Sm core
proteins and several of the hnRNP proteins [104-106]. Other interacting proteins in an
SMN complex can be detected using less stringent salt conditions for isolation [23, 107].

In the nucleus, the SMN protein can be found in at least three distinct complexes
distinguished in sedimentation and chromatography experiments: an 18S complex, a 208
complex and a complex > 208 [108]. The protein composition of the 208 complex
contains SMN, geminsZ-4, and a subset of the Sm core proteins. It is now thought that
the >208 SMN complex represents the SMN complex in its entirety, including the other
gemin and Sm core proteins not found in the 208 complex. The full composition of the
18S SMN complex has yet to be investigated [108].

It is of some contention whether the firll nuclear SMN complex contains the U
snRNAs. Some reports using immunoprecipitation assays at high salt failed to find the U
snRNAs with a nuclear SMN complex [103, 108]. Xu et a1. however, showed that a
GST-SMN fusion protein did pull-down all 5 U snRNAs used in pre-mRN A splicing as
well as U7 snRNA from nuclear extract [72]. It was also demonstrated that SMN
interacts with coilin, which associates with the snRN As [72]. There is strong evidence
that the SMN interaction with snRNAs may be sensitive to the high salt conditions of the
earlier reports. It also should be noted that proteins present in the SMN complex,
gemin3 and gemin4, have been found in complexes with unidentified small nuclear
RNAs of several hundred nucleotides or U snRNAs [109, 110]. Whether these snRNA
complexes are truly exclusive of the nuclear SMN complex or are only the result of using

less stringent isolation conditions remains to be investigated.

15

Although the role of SMN and the SMN complex in the cytoplasm has been well
documented in assembling nascent U snRNAs into functional snRNP particles, their role
in the nucleus is open to speculation. Antibodies against the SMN protein inhibited pre-
mRNA splicing only when pre-incubated with nuclear extract using an in vitro splicing
assay [111]. Similar antibody inhibition experiments showed that antibodies against
SMN prevented active splicing complexes from forming [108]. These results suggested
that the SMN protein was responsible for either assembling or delivering snRNPs to the
spliceosome to form active splicing complexes. This is further backed by the association
of the SMN complex with several required splicing factors [106, 112]. There is also
evidence to suggest that SMN interacts with coilin in the Caj al bodies, possibly indicating
a role in modifying or assembling nascent snRNPs [113]. In the nucleus, SMN has taken
on the moniker of "master assembler" due to its interaction with many different types of
small, nuclear RNPs [72, 114-1 18].

2. Penta-snRN P

Stevens et a1. introduced a new model for spliceosome assembly by identifying
and characterizing a complex termed a penta—snRNP [119]. Using S. cerevisiae, it was
found that a 458 complex could be isolated containing all 5 snRNPs, with the snRNAs in
a stoichiometric ratio, that is equal amounts of each of the 5 snRNAs were found in the
458 complex. Furthermore, it was found that when the both the protein and snRNA
components of snRNPs were labeled, the penta-snRNP complex did not exchange a
snRNP in the penta-snRNP for free snRNPs in the extract. This penta-snRNP complex
was able to excise an intron from an exogenous pre-mRNA in a cell-free system when

complemented with nuclease treated yeast extract. The penta-snRNP complex

16

contained the Sm core proteins and the snRNP specific proteins. Interestingly, the penta-
snRNP also contained other splicing factors, including all 8 members of the Prp19
complex, a complex important for formation of an active spliceosome from early splicing
complexes [120, 121]. Using these findings, it was suggested that the penta-snRNP
represented a pre-formed spliceosome that could assemble immediately onto a pre-

mRN A when encountered.

Other studies in Schizosaccharomyces pombe found a similar penta—snRNP
complex [122]. Yeast extracts were subjected to sedimentation on glycerol gradients and
fractions containing 5 snRN As were immunoprecipitated by antibodies against two
splicing factors, Prpl and Prp 31. These two proteins were shown to be in a pre-
spliceosomal complex with 5 snRNAs. Interestingly, several studies indicate there may
be a similar mammalian penta—snRNP complex [123, 124]. Using HeLa NE, it was found
that U1 snRNA paired at the 5' splice site in a large complex containing all 5 snRNAs,
and that this association was abolished when a mutated splice site was used or U1 was
degraded [123]. Using a co-transcriptional/splicing assay in human A431 cells,
Listerman et al. found that both U1 and U5 snRNP specific proteins could be present on a
nascent pre-mRNA after only the 5' splice site in the pre-mRN A had been transcribed.
[124]. This transcripition/splicing assay relied on chromatin immunoprecipitation
techniques. Antibodies against known splicing factors and/or snRNP specific proteins
were used to immunoprecipitate a nascent mRNA still attached to the DNA template by
RNA Pol H. By using a series of reverse transcription primers, the researchers could
determine where on the gene the RNA polymerase was located. The key result from

these experiments was that antibodies against US (and U1) were able to precipitate the

17

nascent RNA, even though the RNA polymerase had not transcribed the branchpoint or 3'
splice site fo the gene. The association of U1 and US with the 5' splice site before
transcription of the 3' splice site suggests a multi-snRNP mammalian complex for
spliceosome assembly.
3. PCC

Further evidence for a mammalian penta-snRNP was provided by the isolation of
a complex containing all 5 snRNAs from HeLa nuclear extract using low salt conditions
[110]. This complex is called the PCC (PSF-gontaining gomplex) and contains many
RNA processing factors in addition to the snRNAs, snRNP proteins and PSF (PTB-
associated splicing factor). Proteomics uncovered the presence of many hnRNP proteins,
SMN, gemin4, RNA helicases, pre-mRNA splicing factors, several transcription factors
and most important for our studies, galectin-3. PCC is formed in the absence of pre-
mRNA and under incubation conditions that do not allow active spliceosomes to form.
Using velocity sedimentation, Peng et al [110] found the size of the PCC to be near 608,
similar to the size of a spliceosome [125]. It is suggested that the PCC may represent a
pre-formed pre-spliceosome containing all 5 snRNAs. Since the PCC contains
components of both the above described SMN complex and the penta-snRNP, it is
possible that the PCC represents an assembly of the penta—snRNP with or by the SMN
complex. However, there is currently no functional evidence to suggest how PCC, penta-

snRNP and the nuclear SMN complex relate to each other.
III. Pre-mRN A splicing

A. Spliceosome

18

Pre-mRNA splicing is an essential process for eukaryotic cells. Newly
transcribed pre-mRNA undergoes splicing to remove intervening sequences (introns).
Pre-mRNA splicing is carried out by a large complex consisting of both RNA and many
proteins, termed the spliceosome [126]. This process uses two transesterification
reactions to excise an intron and join two exons. The pre-mRNA contains important
information regarding the splice sites in its sequence. The most critical sequence motifs
are absolutely conserved: a 5' splice site, a 3' splice site and a branchpoint adenine [127-
129]. Other pre-mRNA sequences, such as the polypyrimidine tract and splicing
silencers and enhancers, also play a role in identifying the correct splice sites, but these
can vary between gene transcripts.

The 5 U snRNAs are members of the spliceosome as snRNPs and serve an
important role for basepairing the pre-mRNA at the correct sites [130]. Attempts at
cataloging the proteins in or associated with the spliceosome have shown over 300
distinct polypeptides involved in this large complex [131]. The numerous protein
components and 5 described snRNPs involved with the spliceosome suggest a large
assembly. Active spliceosomes have been sedimented to 60$, confirming the magnitude
of this complex [125]. There is some evidence that spliceosomes may oligomerize
forming even larger supraspliceosomes of 2008 [123, 132] .

The process of pre-mRNA splicing is ATP-dependent. This is likely due to the
numerous RNA helicases involved in pre-mRNA splicing, as the actual chemical
reactions necessary for intron removal are energetically favorable [133].

Another important finding is that the spliceosome may be considered another

example of a ribozyme. Using protein-free RNA fragments of U2, U6 and branchpoint

19

RNA the products of the first transesterification reaction can be detected, suggesting only
the properly assembled RNA components of the spliceosome are needed for the reaction
[134, 135]. Protein components of the spliceosome may serve such roles as a scaffold for
loading the snRNPs correctly, unwinding RNA to assure proper base-pairing or recruiting
other factors necessary for mRNA processing following pre-mRNA splicing. In addition,
protein components of the spliceosome play important roles in alternative splicing by
helping choose the correct splice site [136, 137].

B. Assembly

The canonical model of spliceosome assembly is termed the step-wise assembly
model. It entails the addition of U snRNPs in an ordered, sequential fashion [138].
Because of this ordered addition of factors, distinct complexes with varying snRNA and
protein components can be isolated. The common feature in all these complexes is that
they are formed on a pre-mRNA. These distinct complexes were most often found using
a cell-free splicing system and were originally described by their mobility in native gel
electrophoresis [1 39].

The first snRNA containing complex in the step-wise assembly model is the E, or
early, complex. In yeast, it is commonly referred to as the commitment complex, CC
[128]. This complex forms in the absence of ATP and at temperatures as low as 4°C.

The key feature of this complex is the base-pairing of U1 snRNP to the 5' splice site
[140]. In addition to the base-pairing of the U1 snRNA to the 5' splice site of the pre-
mRNA, many important protein-pre-mRN A contacts are detected as well [141]. There is
good evidence that the specificity of the U1 interaction at the 5' splice site may be due in

part to protein-RNA interactions in addition to the U1 -pre-mRNA base-pairing [142]. In

20

addition, the U2 snRNP may be present in the early E complex; however, it has not yet
base-paired with the pre-mRNA [143, 144]. It is currently not known if specific protein
contacts on the pre-mRN A are necessary before the loading of U1 in the B complex.
Kent et al. have data to suggest an even earlier complex in the spliceosome assembly
pathway, termed the E' complex, which may represent a precursor to the E complex
[145]. This E' complex contains the U1 snRNA bound to the pre-mRNA at the 5' splice
site but lacks U2AF, a protein component of the E complex.

The spliceosome then proceeds to the ATP-dependent A complex. In this
complex, U2 snRNA binds to the branchpoint sequence [146, 147]. The ATP-
dependence is due to UAP56, a member of the DExD/H-box helicase family of proteins.
UAP56 is required for the association of U2 with the pre-mRNA branchpoint [148].
There is some evidence to suggest that UAP56 is actually removing other proteins from
the branchpoint sequence, allowing U2 to bind there, as opposed to changing the RNA
structure of U2 [149]. Other proteins also add at this complex, including some that may
allow U1 and U2 to associate through a protein bridge [150, 151]. Recent studies have
also shown that the ends of the U1 and U2 snRNAs are in proximity with one another at
this stage of assembly, which may be another mechanism of communication between
snRNPs in the spliceosome [152].

The next complex to form on the pre-mRNA is the B complex. This occurs with
the addition of the tri-snRNP (U4/U6.U5) [153]. The spliceosome now contains all 5
snRNAs; however it is not catalytically active. To become catalytically active, U4/U 6
must unwind its basepairing and U1 must be displaced from the 5' splice site. The 5'

splice site is then free to pair with U6. There is also evidence to suggest that US makes

21

ATP-dependent contacts at the 5' splice site and 5' exon, even before U1 leaves the
complex [154, 155]. However, it is the switch of basepairing from U1 to U6 at the 5'
splice site that is thought to be catalytically important [133]. Afterwards, U1 and U4 are
both released to form the active spliceosome, termed the C complex.

For the unwinding of U4 and U6 to happen in the B complex more RNA helicases
are required [156]. It is also thought that this unwinding requires ATP. Without the
unwinding of U4/U 6, splicing cannot proceed presumably because U6 cannot switch for
U1 at the 5' splice site [157]. It is also at the B complex where the Prp19 complex, also
identified in the penta-snRNP, adds to the spliceosome [121, 158]. The switch to the
active catalytic spliceosome is thought to occur when U6, now at the 5' splice site, pairs
with U2. With U6 positioned at the 5' splice site and U2 at the branch point, the pairing
of U2 and U6 brings the participants in the first transesterification reaction in proximity.
The U2/U 6 complex adopts a structure of a four-helix junction, involving U2-U6
intermolecular helices I, II, III and an U6 intrastemloop (ISL) [159]. Interestingly, this
structure bears great similarity to the structure of a group II self-splicing intron [160].
The U6 intra-stem loop contains the base critical for binding the necessary Mg2+ involved
in splicing catalysis [161]. The structure of the pairing of U2 and U6 at the 5' splice site
also suggests that catalysis of pre-mRNA splicing is due to RNA, as experimental
evidence has suggested [134, 160]. Conserved bases on U6 are essential for both
transesterification reactions [162, 163]. U5 seems to serve a more structural role, by
contacting both exons and keeping them in proximity for the second transesterification
reaction to occur [155, 164]. This idea is supported by making deletion or insertion

mutants in an invariable loop of U5 and testing these mutants by in vitro splicing assays.

22

These U5 mutants led to defects in splicing, due to misalignment of the exons in the
splicing reaction [165]. The U5 particle also undergoes remodeling with respect to its
protein components during spliceosome activation, although the significance of this is not
yet understood [166].

An alternative model of spliceosome assembly using the previously described
penta-snRNP has been proposed [119]. The penta-snRNP could load onto a pre-mRNA
at once, containing all 5 snRNAs and many necessary splicing factors. This penta-
snRNP addition would resemble the B complex of the step-wise addition, with respect to
snRNA composition. For the penta-snRNP assembled on the pre-mRNA, all that would
be necessary is snRNA unwinding and correct base-pairing (plus additional necessary
protein contacts, if any) to proceed to the catalytic spliceosome [167, 168]. These two
models of spliceosome assembly are not necessarily mutually exclusive [169]. It may be
the case that either is capable of carrying out pre-mRNA splicing and both mechanisms
of assembling spliceosomes may be used in vivo [110].

C. Proteomics

Recently, there have been several attempts to characterize the protein components
of spliceosomes using tandem mass spectrometry. Rappsilber et al. used two different
biotinylated pre-mRNA substrates and gel filtration to analyze a mixture of spliceosomal
complexes [131]. Similarly, Zhou et al. used two different pre-mRNA substrates each
containing an affinity sequence for the M82 viral coat protein [170]. These selected
complexes were than separated by gel filtration chromatography. These two large
proteomic studies used conditions that allowed the pre-mRN A to be associated in many

spliceosomal complexes. Subj ecting the isolated complexes to tandem mass

23

spectrometry, these studies identified approximately 300 [131] and 150 [170] unique
proteins associated with the spliceosome, respectively. The most striking result however,
was the association of many proteins not before linked to pre-mRNA splicing [171].
Included in this group are transcription factors, translation factors, histone acetyl-
transferases and many ribosomal proteins.

Other proteomic studies have attempted to isolate a single spliceosomal complex
by stopping spliceosome assembly at a distinct step. Makarov et al. assembled B
complexes and immunoselected them with an antibody against a specific B complex
factor [166]. Further purification of this complex and subsequent tandem mass
spectrometric analysis identified approximately 100 proteins associated with the
spliceosomal B complex. Fewer novel proteins were identified than previous studies.
From the identified proteins, the researchers concluded that they indeed had isolated a
pure B complex due to the lack of identified C complex proteins. A second major
discovery was a significant rearrangement of the 358 US particle upon addition to the
spliceosome. This included dissociation of numerous proteins from the U5 particle after
US had bound the spliceosome, indicating a subset of snRNP proteins interacts with U5
outside of the spliceosome, but not in the spliceosome [166]. A second complex isolation
was performed by Jurica et al of the catalytically active C complex [172]. These
complexes were isolated by affinity selection and size exclusion chromatography.
Important in this study was the finding of unique C complex proteins. These proteins
associated with the spliceosome only in the active C complex, as opposed to ATP-

independent early complexes.

24

Finally, recent attempts have been made to characterize spliceosomes under
"physiological conditions" [173, 174]. Isolation procedures that lack heparin and use
lower salt concentrations in buffers were performed. These are unique in that previous
studies employed both heparin and salt concentrations ranging from 100 to 250 mM NaCl
[131, 166, 170]. The B complex was isolated and characterized under conditions of
lower salt and lacking heparin [173]. Similarly, the pre-spliceosomal A complex was
purified and submitted to tandem mass spectrometry [174]. In addition to finding many
of the same proteins identified previously, these studies did find heretofore unidentified
proteins associated with the spliceosome. This leads one to believe that by changing
isolation conditions, unique proteins may be found that were missed during the early
studies. However, the true significance is somewhat uncertain as the authors argue that
some of the identified proteins are probably contaminants due to the low stringency
conditions [173] . A new in vivo method of isolating spliceosomes from chicken and
human cells has recently been used [175]. This technique involves adding a tandem-
affinity purification (TAP) tag to the SmD3 protein at the native genomic location [176].
This tagged native protein and its associated proteins and RNA were then isolated using
the developed affinity tag and the resulting complexes were subjected to glycerol gradient
sedimentation followed by tandem mass spectrometry. This method allowed the
identification of splicing complexes assembled in vivo. The most striking result fi'om this
study was the identification of numerous nuclear proteins not isolated by the cell-free
splicing system applied earlier [131, 170]. However, many of these newly identified
proteins have functions outside of pre-mRNA splicing (such as nuclear transport and 3'

mRNA processing) and probably reflect the full life-cycle of the mRNA in the nucleus

25

[175]. Relatively few new splicing factors were identified in comparison to previous
proteomic studies. In conclusion, the results serve as a reminder of the closely coupled
nature of RNA biogenesis and processing in vivo. Other reports also indicate a close
linking of pre-mRNA splicing, transcription and RNA processing [124, 177, 178].
IV. Galectins and pre-mRN A splicing

Numerous lines of evidence have supported the idea that Gall and Ga13 are
redundant pre-mRNA splicing factors. For example, experimentation has shown co—
localization of the galectins with known splicing factors, sedimentation of Gal3 with
RNP complexes, depletion and reconstitution of cell free splicing assays and association
of galectins with known splicing factors [22-25, 179]. Recently, antisera against Gall
and Gal3 have been used to immunoprecipitate complexes in an in vitro splicing reaction
[180]. Co-precipitated with the respective galectin proteins were all the mRNA species
expected in a splicing reaction; pre-mRNA, mRNA, excised intron-lariat and
intermediates. As well as the RNA components, splicing factors such as the Sm core
proteins and Slu7 also were co-precipitated, suggesting that the galectins were interacting
with the assembled spliceosome. Other important findings were that galectins do not
bind RNA directly, the interaction between the galectins and the spliceosome was
sensitive to salt conditions and Gall and Gal3 existed in mutually exclusive spliceosomes
[1 80].

In addition to the association of galectins with the spliceosome, there are
indications that Gall and Gal3 may interact with splicing factors or the splicing
machinery outside the assembled spliceosome. Key points of evidence include the

galectin association with the SMN protein in nuclear extracts and the sedimentation of

26

Gal3 in cesium sulfate gradients with RNP complexes [23, 179]. Both these studies lack
the exogenous pre-mRNA scaffold necessary for cell-free splicing. Finally, galectin
association with the spliceosome appears to be at a very early step in spliceosome
formation based on two observations; one, when extracts are depleted of galectins, active
splicing complexes cannot form and two, antibodies against the galectins can
immunoprecipitate pre-mRNA fi'om early complexes ahnost immediately after adding the
pre-mRNA scaffold [25, 180]. Taken as a whole, these data suggest the possibility that
galectins may interact with splicing factors outside the spliceosome and load onto the

pre-mRNA with other necessary early factors.

27

References

1.

10.

11.

Barondes, S.H., Castronovo, V., Cooper, D.N., Cummings, R.D., Drickarner, K.,
Feizi, T., Gitt, M.A., Hirabayashi, J ., Hughes, C., Kasai, K., and et al. (1994).
Galectins: a family of animal beta-galactoside-binding lectins. Cell 76, 597-598.

Cooper, D.N., and Barondes, SH. (1999). God must love galectins; he made so
many of them. Glycobiology 9, 979-984.

Agrwal, N., Sun, Q., Wang, S.Y., and Wang, J .L. (1993). Carbohydrate-binding
protein 35. I. Properties of the recombinant polypeptide and the individuality of
the domains. J Biol Chem 268, 14932-14939.

Birdsall, B., Feeney, J ., Burdett, I.D., Bawurnia, S., Barboni, EA, and Hughes,
RC. (2001). NMR solution studies of hamster galectin-3 and electron
microscopic visualization of surface-adsorbed complexes: evidence for

interactions between the N- and C-terrninal domains. Biochemistry 40, 4859-
4866.

Leonidas, D.D., Vatzaki, E.H., Vorum, H., Celis, J .E., Madsen, P., and Acharya,
KR. (1998). Structural basis for the recognition of carbohydrates by human
galectin-7. Biochemistry 3 7, 13930-13940.

Nagae, M., Nishi, N., Murata, T., Usui, T., Nakarnura, T., Wakatsuki, S., and
Kate, R. (2006). Crystal structure of the galectin-9 N-terminal carbohydrate
recognition domain from Mus musculus reveals the basic mechanism of
carbohydrate recognition. J Biol Chem 281, 35884-35893.

Seetharaman, J ., Kanigsberg, A., Slaaby, R., Leffler, H., Barondes, SH, and Rini,
J .M. (1998). X-ray crystal structure of the human galectin-3 carbohydrate
recognition domain at 2.1-A resolution. J Biol Chem 273, 13047-13052.

Leffler, H., and Barondes, SH. (1986). Specificity of binding of three soluble rat
lung lectins to substituted and unsubstituted mammalian beta-galactosides. J Biol
Chem 261,10119-10126.

Wang, J .L., Gray, R.M., Haudek, KC, and Patterson, R.J. (2004).
Nucleocytoplasmic lectins. Biochim Biophys Acta 16 73, 75-93.

Liu, F.T., Patterson, R.J., and Wang, J .L. (2002). Intracellular functions of
galectins. Biochim Biophys Acta 15 72, 263-273.

Menon, RP, and Hughes, RC. (1999). Determinants in the N-terminal domains

of galectin-3 for secretion by a novel pathway circumventing the endoplasmic
reticulum-Golgi complex. Eur J Biochem 264, 569-576.

28

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Cho, M., and Cummings, RD. (1995). Galectin-l, a beta-galactoside-binding
lectin in Chinese hamster ovary cells. II. Localization and biosynthesis. J Biol
Chem 270, 5207-5212.

Cho, M., and Cummings, RD. (1995). Galectin-l, a beta-galactoside-binding
lectin in Chinese hamster ovary cells. 1. Physical and chemical characterization. J
Biol Chem 270, 5198-5206.

Savin, S.B., Cvejic, D.S., and Jankovic, M.M. (2003). Expression of galectin-1
and galectin-3 in human fetal thyroid gland. J Histochem Cytochem 51, 479-483.

Shimonishi, T., Miyazaki, K., Kono, N., Sabit, H., Tuneyama, K., Harada, K.,
Hirabayashi, J ., Kasai, K., and Nakanuma, Y. (2001). Expression of endogenous

galectin-1 and galectin-3 in intrahepatic cholangiocarcinoma. Hum Pathol 32,
302-310.

Cooper, D.N., and Barondes, SH. (1990). Evidence for export of a muscle lectin
from cytosol to extracellular matrix and for a novel secretory mechanism. J Cell
Biol 110, 1681-1691.

Paz, A., Haklai, R., Elad-Sfadia, G., Ballan, E., and Kloog, Y. (2001). Galectin-l
binds oncogenic H-Ras to mediate Ras membrane anchorage and cell
transformation. Oncogene 20, 7486-7493.

Elad-Sfadia, G., Haklai, R., Ballan, E., Gabius, H.J., and Kloog, Y. (2002).
Galectin-l augments Ras activation and diverts Ras signals to Raf-l at the
expense of phosphoinositide 3-kinase. J Biol Chem 277, 37169-37175.

Rotblat, B., Niv, H., Andre, S., Kaltner, H., Gabius, H.J., and Kloog, Y. (2004).
Galectin-1(L11A) predicted from a computed galectin-1 famesyl-binding pocket
selectively inhibits Ras-GTP. Cancer Res 64, 3112-3118.

Ashery, U., Yizhar, O., Rotblat, B., Elad-Sfadia, G., Barkan, B., Haklai, R., and
Kloog, Y. (2006). Spatiotemporal organization of Ras signaling: rasosomes and
the galectin switch. Cell Mol Neurobiol 26, 471-495.

Choi, J .Y., van Wijnen, A.J., Aslam, F., Leszyk, J .D., Stein, J.L., Stein, G.S.,
Lian, J .B., and Penman, S. (1998). Developmental association of the beta-
galactoside-binding protein galectin-1 with the nuclear matrix of rat calvarial
osteoblasts. J Cell Sci 111 (Pt 20), 3035-3043.

Vyakarnam, A., Lenneman, A.J., Lakkides, K.M., Patterson, R.J., and Wang, J .L.

(1998). A comparative nuclear localization study of galectin-1 with other splicing
components. Exp Cell Res 242, 419-428.

29

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

Park, J .W., Voss, P.G., Grabski, S., Wang, J .L., and Patterson, R.J. (2001).
Association of galectin-1 and galectin-3 with Gemin4 in complexes containing the
SMN protein. Nucleic Acids Res 29, 3595-3602.

Vyakarnam, A., Dagher, S.F., Wang, J .L., and Patterson, R.J. (1997). Evidence
for a role for galectin-1 in pre-mRNA splicing. Mol Cell Biol 1 7, 4730-4737.

Dagher, S.F., Wang, J .L., and Patterson, R.J. (1995). Identification of galectin-3
as a factor in pre-mRNA splicing. Proc Natl Acad Sci U S A 92, 1213-1217.

Hubert, M., Wang, S.Y., Wang, J .L., Seve, AP, and Hubert, J. (1995).
Intranuclear distribution of galectin-3 in mouse 3T3 fibroblasts: comparative

analyses by immunofluorescence and immunoelectron microscopy. Exp Cell Res
220, 397-406.

Moutsatsos, I.K., Davis, J .M., and Wang, J .L. (1986). Endogenous lectins from
cultured cells: subcellular localization of carbohydrate-binding protein 35 in 3T3
fibroblasts. J Cell Biol 102, 477-483.

Weinberger, P.M., Adam, B.L., Gourin, C.G., Moretz, W.H., 3rd, Bollag, R.J.,
Wang, B.Y., Liu, Z., Lee, J .R., and Terris, DJ. (2007). Association of nuclear,

cytoplasmic expression of galectin-3 with beta-catenin/Wnt-pathway activation in
thyroid carcinoma. Arch Otolaryngol Head Neck Surg 133, 503-510.

Moutsatsos, I.K., Wade, M., Schindler, M., and Wang, J .L. (1987). Endogenous
lectins from cultured cells: nuclear localization of carbohydrate-binding protein
35 in proliferating 3T3 fibroblasts. Proc Natl Acad Sci U S A 84, 6452-6456.

Cowles, E.A., Agrwal, N., Anderson, R.L., and Wang, J .L. (1990). Carbohydrate-
binding protein 35. Isoelectric points of the polypeptide and a phosphorylated
derivative. J Biol Chem 265, 17706-17712.

Tsay, Y.G., Lin, N.Y., Voss, P.G., Patterson, R.J., and Wang, J .L. (1999). Export
of galectin-3 from nuclei of digitonin-permeabilized mouse 3T3 fibroblasts. Exp
Cell Res 252, 250-261.

Davidson, P.J., Davis, M.J., Patterson, R.J., Ripoche, M.A., Poirier, F., and Wang,
J .L. (2002). Shuttling of galectin-3 between the nucleus and cytoplasm.
Glycobiology 12, 329-337.

Davidson, P.J., Li, S.Y., Lohse, A.G., Vandergaast, R., Verde, E., Pearson, A.,
Patterson, R.J., Wang, J .L., and Amoys, EL (2006). Transport of galectin-3

between the nucleus and cytoplasm. 1. Conditions and signals for nuclear import.
Glycobiology 16, 602-611.

30

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

45.

Li, S.Y., Davidson, P.J., Lin, N.Y., Patterson, R.J., Wang, J .L., and Amoys, E.J.
(2006). Transport of galectin-3 between the nucleus and cytoplasm. II.
Identification of the signal for nuclear export. Glycobiology 16, 612-622.

Nakahara, S., Hogan, V., Inohara, H., and Raz, A. (2006). Importin-mediated
nuclear translocation of galectin-3. J Biol Chem 281, 39649-39659.

Akahani, S., Nangia-Makker, P., Inohara, H., Kim, HR, and Raz, A. (1997).
Galectin-3: a novel antiapoptotic molecule with a functional BHl (NW GR)
domain of Bel-2 family. Cancer Res 5 7, 5272-5276.

Yang, R.Y., Hsu, D.K., and Liu, RT. (1996). Expression of galectin-3 modulates
T-cell growth and apoptosis. Proc Natl Acad Sci U S A 93, 6737-6742.

Menon, R.P., Strom, M., and Hughes, RC. (2000). Interaction of a novel cysteine
and histidine-rich cytoplasmic protein with galectin-3 in a carbohydrate-
independent manner. FEBS Lett 4 70, 227-231.

Bawumia, S., Barboni, E.A., Menon, RP, and Hughes, RC. (2003). Specificity
of interactions of galectin-3 with Chrp, a cysteine- and histidine-rich cytoplasmic
protein. Biochimie 85, 189-194.

Paron, I., Scaloni, A., Pines, A., Bachi, A., Liu, F .T., Puppin, C., Pandolfi, M.,
Ledda, L., Di Loreto, C., Damante, G., and Tell, G. (2003). Nuclear localization
of Galectin-3 in transformed thyroid cells: a role in transcriptional regulation.
Biochem Biophys Res Commun 302, 545-553.

Seve, A.P., F elin, M., Doyennette-Moyne, M.A., Sahraoui, T., Aubery, M., and
Hubert, J. (1993). Evidence for a lactose-mediated association between two
nuclear carbohydrate-binding proteins. Glycobiology 3, 23-30.

Seve, A.P., Hadj-Sahraoui, Y., Felin, M., Doyennette-Moyne, M.A., Aubery, M.,
and Hubert, J. (1994). Evidence that lactose binding to CBP35 disrupts its
interaction with CBP70 in isolated HL60 cell nuclei. Exp Cell Res 213, 191-197.

Shimura, T., Takenaka, Y., Tsutsumi, S., Hogan, V., Kikuchi, A., and Raz, A.
(2004). Galectin-3, a novel binding partner of beta-catenin. Cancer Res 64, 6363-
6367.

Goletz, S., Hanisch, F.G., and Karsten, U. (1997). Novel alphaGalNAc containing
glycans on cytokeratins are recognized invitro by galectins with type II
carbohydrate recognition domains. J Cell Sci 110 (Pt 14), 1585-1596.

Wang, L., Inohara, H., Pienta, KL, and Raz, A. (1995). Galectin-3 is a nuclear
matrix protein which binds RNA. Biochem Biophys Res Commun 21 7, 292-303.

31

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

Zieve, G.W., and Sauterer, RA. (1990). Cell biology of the snRNP particles. Crit
Rev Biochem Mol Biol 25, 1-46.

Hinterberger, M., Pettersson, I., and Steitz, J .A. (1983). Isolation of small nuclear
ribonucleoproteins containing U1, U2, U4, U5, and U6 RNAs. J Biol Chem 258,
2604-2613.

Will, CL, and Luhrrnann, R. (2001). Spliceosomal UsnRNP biogenesis, structure
and function. Curr Opin Cell Biol 13, 290-301.

Kiss, T. (2004). Biogenesis of small nuclear RNPs. J Cell Sci 11 7, 5949-5951.

Medlin, J .E., Uguen, P., Taylor, A., Bentley, D.L., and Murphy, S. (2003). The C-
terrninal domain of pol II and a DRB-sensitive kinase are required for 3'
processing of U2 snRNA. Embo J 22, 925-934.

Segref, A., Mattaj, I.W., and Ohno, M. (2001). The evolutionarily conserved
region of the U snRNA export mediator PHAX is a novel RNA-binding domain
that is essential for U snRNA export. Rna 7, 351-360.

Ohno, M., Segref, A., Bachi, A., Wihn, M., and Mattaj, I.W. (2000). PHAX, a
mediator of U snRN A nuclear export whose activity is regulated by
phosphorylation. Cell I 01, 187-198.

Yong, J ., Golembe, T.J., Battle, D.J., Pellizzoni, L., and Dreyfuss, G. (2004).
snRNAs contain specific SMN-binding domains that are essential for snRNP
assembly. Mol Cell Biol 24, 2747-2756.

Pellizzoni, L., Yong, J ., and Dreyfuss, G. (2002). Essential role for the SMN
complex in the specificity of snRNP assembly. Science 298, 1775-1779.

Urlaub, H., Raker, V.A., Kostka, S., and Luhnnann, R. (2001). Sm protein-Sm
site RNA interactions within the inner ring of the spliceosomal snRNP core
structure. Embo J 20, 187-196.

Raker, V.A., Plessel, G., and Luhrrnann, R. ( 1996). The snRNP core assembly
pathway: identification of stable core protein heteromeric complexes and an
snRNP subcore particle in vitro. Embo J 15, 2256-2269.

Karnbach, C., Walke, 8., Young, R., Avis, J .M., de la Fortelle, E., Raker, V.A.,
Luhrmann, R., Li, J ., and Nagai, K. (1999). Crystal structures of two Sm protein

complexes and their implications for the assembly of the spliceosomal snRNPs.
Cell 96, 375-387.

Brahms, H., Raymackers, J ., Union, A., de Keyser, F., Meheus, L., and
Luhrrnann, R. (2000). The C-terminal RG dipeptide repeats of the spliceosomal

32

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

Sm proteins D1 and D3 contain symmetrical dimethylarginines, which form a
major B-cell epitope for anti-Sm autoantibodies. J Biol Chem 275, 17122-17129.

Brahms, H., Meheus, L., de Brabandere, V., Fischer, U., and Luhrmann, R.
(2001). Symmetrical dimethylation of arginine residues in spliceosomal Sm
protein B/B' and the Sm-like protein LSm4, and their interaction with the SMN
protein. Rna 7, 1531-1542.

Cote, J ., and Richard, S. (2005). Tudor domains bind symmetrical dimethylated
arginines. J Biol Chem 280, 28476-28483.

Battle, D.J., Lau, C.K., Wan, L., Deng, H., Lotti, F ., and Dreyfirss, G. (2006). The
Gemin5 protein of the SMN complex identifies snRNAs. Mol Cell 23, 273-279.

Golembe, T.J., Yong, J ., and Dreyfuss, G. (2005). Specific sequence features,
recognized by the SMN complex, identify snRNAs and determine their fate as
snRNPs. Mol Cell Biol 25, 10989-11004.

Mouaikel, J ., Narayanan, U., Verheggen, C., Matera, A.G., Bertrand, E., Tazi, J.,
and Bordonne, R. (2003). Interaction between the small-nuclear-RNA cap

hypermethylase and the spinal muscular atrophy protein, survival of motor
neuron. EMBO Rep 4, 616-622.

Hausmann, S., and Shuman, S. (2005). Specificity and mechanism of RNA cap
guanine-N2 methyltransferase (Tgsl). J Biol Chem 280, 4021-4024.

Rollenhagen, C., and Pante, N. (2006). Nuclear import of spliceosomal snRNPs.
Can J Physiol Pharmacol 84, 367-376.

Huber, J ., Cronshagen, U., Kadokura, M., Marshallsay, C., Wada, T., Sekine, M.,
and Luhrmann, R. (1998). Snurportinl, an m3G-cap-specific nuclear import
receptor with a novel domain structure. Embo J 1 7, 4114-4126.

Rollenhagen, C., Muhlhausser, P., Kutay, U., and Pante, N. (2003). Irnportin beta-
depending nuclear import pathways: role of the adapter proteins in the docking
and releasing steps. Mol Biol Cell 14, 2104-2115.

Huber, J ., Dickmanns, A., and Luhrmann, R. (2002). The importin-beta binding
domain of snurportinl is responsible for the Ran- and energy-independent nuclear
import of spliceosomal U snRNPs in vitro. J Cell Biol 156, 467-479.

Narayanan, U., Achsel, T., Luhrmann, R., and Matera, AG. (2004). Coupled in

vitro import of U snRNPs and SMN, the spinal muscular atrophy protein. Mol
Cell 16, 223-234.

33

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

Narayanan, U., Ospina, J.K., Frey, M.R., Hebert, M.D., and Matera, A.G. (2002).
SMN, the spinal muscular atrophy protein, forms a pre-import snRNP complex
with snurportinl and importin beta. Hum Mol Genet 11, 1785-1795.

Marshallsay, C., and Luhrmann, R. (1994). In vitro nuclear import of snRNPs:
cytosolic factors mediate m3G-cap dependence of U1 and U2 snRNP transport.
Embo J 13, 222-231.

Xu, H., Pillai, R.S., Azzouz, T.N., Shpargel, K.B., Kambach, C., Hebert, M.D.,
Schumperli, D., and Matera, A.G. (2005). The C-tenninal domain of coilin
interacts with Sm proteins and U snRNPs. Chromosoma 114, 155-166.

Stanek, D., and Neugebauer, KM. (2006). The Cajal body: a meeting place for
spliceosomal snRNPs in the nuclear maze. Chromosoma 115, 343-354.

Jady, B.E., Darzacq, X., Tucker, K.E., Matera, A.G., Bertrand, E., and Kiss, T.
(2003). Modification of Sm small nuclear RNAs occurs in the nucleoplasmic
Cajal body following import from the cytoplasm. Embo J 22, 1878-1888.

J ady, BE, and Kiss, T. (2001). A small nucleolar guide RNA functions both in
2'-O-ribose methylation and pseudouridylation of the U5 spliceosomal RNA.
Embo J 20, 541-551.

Yu, Y.T., Shu, M.D., and Steitz, J .A. (1998). Modifications of U2 snRNA are
required for snRNP assembly and pre-mRNA splicing. Embo J 1 7, 5783-5795.

Lamond, AI, and Spector, D.L. (2003). Nuclear speckles: a model for nuclear
organelles. Nat Rev Mol Cell Biol 4, 605-612.

Xue, D., Rubinson, D.A., Pannone, B.K., Yoo, OJ, and Wolin, S.L. (2000). U
snRNP assembly in yeast involves the La protein. Embo J 19, 1650-1660.

Wolin, S.L., and Cedervall, T. (2002). The La protein. Annu Rev Biochem 71,
375-403.

Mayes, A.E., Verdone, L., Legrain, P., and Beggs, J .D. ( 1999). Characterization
of Sm-like proteins in yeast and their association with U6 snRN A. Embo J 18,
4321-433 1 .

Karaduman, R., Fabrizio, P., Hartrnuth, K., Urlaub, H., and Luhrmann, R. (2006).
RNA structure and RN A-protein interactions in purified yeast U6 snRNPs. J Mol
Biol 356, 1248-1262.

Achsel, T., Brahms, H., Kastner, B., Bachi, A., Wihn, M., and Luhrmann, R.
(1999). A doughnut-shaped heteromer of human Sm-like proteins binds to the 3'-

34

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

end of U6 snRNA, thereby facilitating U4/U6 duplex formation in vitro. Embo J
18, 5789-5802.

Verdone, L., Galardi, S., Page, D., and Beggs, J .D. (2004). Lsm proteins promote
regeneration of pre-mRNA splicing activity. Curr Biol 14, 1487-1491.

Lange, TS, and Gerbi, SA. (2000). Transient nucleolar localization Of U6 small
nuclear RNA in Xenopus Laevis oocytes. Mol Biol Cell 11, 2419-2428.

Ganot, P., Jady, B.E., Bortolin, M.L., Darzacq, X., and Kiss, T. (1999). Nucleolar
factors direct the 2'-O-ribose methylation and pseudouridylation of U6
spliceosomal RNA. Mol Cell Biol 19, 6906-6917.

Spiller, M.P., Boon, K.L., Reijns, M.A., and Beggs, J .D. (2007). The Lsm2-8
complex determines nuclear localization of the spliceosomal U6 snRNA. Nucleic
Acids Res 35, 923-929.

Hashimoto, C., and Steitz, J .A. (1984). U4 and U6 RNAs coexist in a single small
nuclear ribonucleoprotein particle. Nucleic Acids Res 12, 3283-3293.

Black, D.L., and Steitz, J .A. (1986). Pre-mRNA splicing in vitro requires intact
U4/U6 small nuclear ribonucleoprotein. Cell 46, 697-704.

Gozani, O., Patton, LG, and Reed, R. (1994). A novel set of spliceosome-
associated proteins and the essential splicing factor PSF bind stably to pre-mRN A
prior to catalytic step 11 of the splicing reaction. Embo J 13, 3356-3367.

Nottrott, S., Urlaub, H., and Luhrmann, R. (2002). Hierarchical, clustered protein
interactions with U4/U 6 snRN A: a biochemical role for U4/U 6 proteins. Embo J
21, 5527-5538.

Bell, M., Schreiner, S., Darnianov, A., Reddy, R., and Bindereif, A. (2002). p110,
a novel human U6 snRNP protein and U4/U 6 snRNP recycling factor. Embo J 21,
2724-2735.

Stanek, D., and Neugebauer, KM. (2004). Detection of snRNP assembly
intermediates in Caj a1 bodies by fluorescence resonance energy transfer. J Cell
Biol 166, 1015-1025.

Stanek, D., Rader, S.D., Klingauf, M., and Neugebauer, KM. (2003). Targeting
of U4/U6 small nuclear RNP assembly factor SART3/p110 to Cajal bodies. J Cell
Biol 160, 505-516.

Behrens, SE, and Luhrmann, R. (1991). Immunoaffinity purification of a
[U4/U6.U5] tri-snRNP from human cells. Genes Dev 5, 1439-1452.

35

95.

96.

97.

98.

99.

100.

101.

102.

103.

104.

105.

Fetzer, S., Lauber, J ., Will, CL, and Luhrmann, R. (1997). The [U4/U6.U5] tri-
snRNP-specific 27K protein is a novel SR protein that can be phosphorylated by
the snRNP-associated protein kinase. Rna 3, 344-355.

Schaffert, N., Hossbach, M., Heintzmann, R., Achsel, T., and Luhrmann, R.
(2004). RNAi knockdown of hPrp31 leads to an accumulation of U4/U 6 di-
snRNPs in Cajal bodies. Embo J 23, 3000-3009.

Laggerbauer, B., Liu, S., Makarov, E., Vomlocher, H.P., Makarova, O.,
Ingelfinger, D., Achsel, T., and Luhrmann, R. (2005). The human U5 snRNP 52K
protein (CD2BP2) interacts with U5-102K (hPrp6), a U4/U6.U5 tri-snRNP
bridging protein, but dissociates upon tri-snRNP formation. Rna 11, 598-608.

Liu, S., Rauhut, R., Vomlocher, HP, and Luhrmann, R. (2006). The network of
protein-protein interactions within the human U4/U6.U5 tri-snRNP. Rna 12,
141 8-1430.

Hall, KB, and Konarska, M.M. (1992). The 5' splice site consensus RNA
oligonucleotide induces assembly of U2/U4/U S/U 6 small nuclear
ribonucleoprotein complexes. Proc Natl Acad Sci U S A 89, 10969-10973.

Konarska, M.M., and Sharp, RA. (1988). Association of U2, U4, U5, and U6
small nuclear ribonucleoproteins in a spliceosome-type complex in absence of
precursor RNA. Proc Natl Acad Sci U S A 85, 5459-5462.

Lefebvre, S., Burlet, P., Liu, Q., Bertrandy, S., Clermont, O., Munnich, A.,
Dreyfuss, G., and Melki, J. (1997). Correlation between severity and SMN protein
level in spinal muscular atrophy. Nat Genet 16, 265-269.

Lefebvre, S., Burglen, L., Reboullet, S., Clermont, O., Burlet, P., Viollet, L.,
Benichou, B., Cruaud, C., Millasseau, P., Zeviani, M., and et al. (1995).

Identification and characterization of a spinal muscular atrophy-determining gene.
Cell 80, 155-165.

Liu, Q., and Dreyfuss, G. (1996). A novel nuclear structure containing the
survival of motor neurons protein. Embo J 15, 3555-3565.

Carissimi, C., Saieva, L., Baccon, J ., Chiarella, P., Maiolica, A., Sawyer, A.,
Rappsilber, J ., and Pellizzoni, L. (2006). Gemin8 is a novel component of the
survival motor neuron complex and firnctions in small nuclear ribonucleoprotein
assembly. J Biol Chem 281, 8126-8134.

Liu, Q., Fischer, U., Wang, F., and Dreyfuss, G. ( 1997). The spinal muscular

atrophy disease gene product, SMN, and its associated protein SIPl are in a
complex with spliceosomal snRNP proteins. Cell 90, 1013-1021.

36

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

Mourelatos, Z., Abel, L., Yong, J ., Kataoka, N., and Dreyfuss, G. (2001). SMN
interacts with a novel family of hnRNP and spliceosomal proteins. Embo J 20,
5443-5452.

Grimmler, M., Otter, 8., Peter, C., Muller, F ., Chari, A., and Fischer, U. (2005).
Unrip, a factor implicated in cap-independent translation, associates with the
cytosolic SMN complex and influences its intracellular localization. Hum Mol
Genet 14, 3099-3111.

Meister, G., Buhler, D., Laggerbauer, B., Zobawa, M., Lottspeich, F., and Fischer,
U. (2000). Characterization of a nuclear 20S complex containing the survival of

motor neurons (SMN) protein and a specific subset of spliceosomal Sm proteins.
Hum Mol Genet 9, 1977-1986.

Mourelatos, Z., Dostie, J ., Paushkin, S., Shanna, A., Charroux, B., Abel, L.,
Rappsilber, J ., Mann, M., and Dreyfuss, G. (2002). miRNPs: a novel class of
ribonucleoproteins containing numerous microRNAs. Genes Dev 1 6, 720-728.

Peng, R., Hawkins, 1., Link, A.J., and Patton, J .G. (2006). The splicing factor PSF
is part of a large complex that assembles in the absence of pre-mRNA and
contains all five snRNPs. RNA Biol 3, 69-76.

Pellizzoni, L., Kataoka, N., Charroux, B., and Dreyfuss, G. (1998). A novel
function for SMN, the spinal muscular atrophy disease gene product, in pre-
mRNA splicing. Cell 95, 615-624.

Meister, G., Hannus, S., Plottner, O., Baars, T., Hartmann, E., Fakan, S.,
Laggerbauer, B., and Fischer, U. (2001). SMNrp is an essential pre-mRNA
splicing factor required for the formation of the mature spliceosome. Embo J 20,

2304-2314.

Boisvert, F.M., Cote, J ., Boulanger, M.C., Cleroux, P., Bachand, F., Autexier, C.,
and Richard, S. (2002). Symmetrical dimethylarginine methylation is required for
the localization of SMN in Cajal bodies and pre-mRNA splicing. J Cell Biol 159,
957-969.

Tems, MP, and Tems, RM. (2001). Macromolecular complexes: SMN--the
master assembler. Curr Biol 11, R862-864.

Friesen, W.J., and Dreyfuss, G. (2000). Specific sequences of the Sm and Sm-like
(Lsm) proteins mediate their interaction with the spinal muscular atrophy disease

gene product (SMN). J Biol Chem 275, 26370-263 75.

Jones, K.W., Gorzynski, K., Hales, C.M., Fischer, U., Badbanchi, F., Tems, R.M.,
and Tems, MP. (2001). Direct interaction of the spinal muscular atrophy disease

37

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

protein SMN with the small nucleolar RNA-associated protein fibrillarin. J Biol
Chem 276, 38645-38651.

Pellizzoni, L., Charroux, B., Rappsilber, J ., Mann, M., and Dreyfuss, G. (2001). A
functional interaction between the survival motor neuron complex and RNA
polymerase II. J Cell Biol 152, 75-85.

Pillai, R.S., Grimmler, M., Meister, G., Will, C.L., Luhrmann, R., Fischer, U., and
Schumperli, D. (2003). Unique Sm core structure of U7 snRNPs: assembly by a
specialized SMN complex and the role of a new component, Lsml 1, in histone
RNA processing. Genes Dev 1 7, 2321-2333.

Stevens, S.W., Ryan, D.E., Ge, H.Y., Moore, RE, Young, M.K., Lee, TD, and
Abelson, J. (2002). Composition and functional characterization of the yeast
spliceosomal penta-snRNP. Mol Cell 9, 3 1-44.

Tarn, W.Y., Lee, K.R., and Cheng, SC. (1993). Yeast precursor mRNA
processing protein PRP19 associates with the spliceosome concomitant with or
just after dissociation of U4 small nuclear RNA. Proc Natl Acad Sci U S A 90,
10821-10825.

Chan, S.P., Kao, D.I., Tsai, W.Y., and Cheng, SC. (2003). The Prp19p-associated
complex in spliceosome activation. Science 302, 279-282.

Bottner, C.A., Schmidt, H., Vogel, S., Michele, M., and Kaufer, NP. (2005).
Multiple genetic and biochemical interactions of Brr2, Prp8, Prp31, Prpl and Prp4
kinase suggest a function in the control of the activation of spliceosomes in
Schizosaccharomyces pombe. Curr Genet 48, 151-161.

Malca, H., Shomron, N., and Ast, G. (2003). The U1 snRNP base pairs with the 5'
splice site within a penta-snRNP complex. Mol Cell Biol 23, 3442-3455.

Listerman, I., Sapra, AK, and Neugebauer, KM. (2006). Cotranscriptional
coupling of splicing factor recruitment and precursor messenger RNA splicing in
mammalian cells. Nat Struct Mol Biol 13, 815-822.

Grabowski, P.J., Seiler, SR, and Sharp, RA. (1985). A multicomponent complex
is involved in the splicing of messenger RNA precursors. Cell 42, 345-353.

Abmayr, S.M., Reed, R., and Maniatis, T. (1988). Identification of a functional

mammalian spliceosome containing unspliced pre-mRNA. Proc Natl Acad Sci U
S A 85, 7216-7220.

Reed, R., and Maniatis, T. (1988). The role of the mammalian branchpoint
sequence in pre-mRNA splicing. Genes Dev 2, 1268-1276.

38

128.

129.

130.

131.

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

Rosbash, M., and Seraphin, B. (1991). Who's on first? The U1 snRNP-5' splice
site interaction and splicing. Trends Biochem Sci 16, 187-190.

Ruskin, B., and Green, MR. (1985). Role of the 3' splice site consensus sequence
in mammalian pre-mRNA splicing. Nature 31 7, 732-734.

Maniatis, T., and Reed, R. (1987). The role of small nuclear ribonucleoprotein
particles in pre-mRNA splicing. Nature 325, 673-678.

Rappsilber, J ., Ryder, U., Lamond, AI, and Mann, M. (2002). Large-scale
proteomic analysis of the human spliceosome. Genome Res 12, 1231-1245.

Azubel, M., Habib, N., Sperling, R., and Sperling, J. (2006). Native spliceosomes
assemble with pre-mRNA to form supraspliceosomes. J Mol Biol 356, 955-966.

Brow, DA. (2002). Allosteric cascade of spliceosome activation. Annu Rev
Genet 36, 333-360.

Valadkhan, S., and Manley, J .L. (2001). Splicing-related catalysis by protein-free
snRNAs. Nature 413, 701-707.

Valadkhan, S., and Manley, J .L. (2003). Characterization of the catalytic activity
of U2 and U6 snRNAs. Rna 9, 892-904.

Shen, H., and Green, MR. (2006). RS domains contact splicing signals and
promote splicing by a common mechanism in yeast through humans. Genes Dev
20, 1755-1765.

Shen, H., Kan, J .L., and Green, MR. (2004). Arginine-serine-rich domains bound
at splicing enhancers contact the branchpoint to promote prespliceosome
assembly. Mol Cell 13, 367-376.

Bindereif, A., and Green, MR. (1987). An ordered pathway of snRNP binding
during mammalian pre-mRNA splicing complex assembly. Embo J 6, 2415-2424.

Das, K, and Reed, R. (1999). Resolution of the mammalian B complex and the
ATP-dependent spliceosomal complexes on native agarose mini-gels. Rna 5,
1 504-1 508.

Ruby, S.W., and Abelson, J. (1988). An early hierarchic role of U1 small nuclear
ribonucleoprotein in spliceosome assembly. Science 242, 1028-1035.

Puig, O., Gottschalk, A., Fabrizio, P., and Seraphin, B. (1999). Interaction of the

U1 snRNP with nonconserved intronic sequences affects 5' splice site selection.
Genes Dev 13, 569-580.

39

142.

143.

144.

145.

146.

147.

148.

149.

150.

151.

152.

153.

Stutz, F ., Liao, X.C., and Rosbash, M. (1993). U1 small nuclear ribonucleoprotein
particle-protein interactions are revealed in Saccharomyces cerevisiae by in vivo
competition assays. Mol Cell Biol 13, 2126-2133.

Das, R., Zhou, Z., and Reed, R. (2000). Functional association of U2 snRNP with
the ATP-independent spliceosomal complex E. Mol Cell 5, 779-787.

Hong, W., Bennett, M., Xiao, Y., F eld Kramer, R., Wang, C., and Reed, R.
(1997). Association of U2 snRNP with the spliceosomal complex B. Nucleic
Acids Res 25 , 354-361.

Kent, O.A., Ritchie, DB, and Macmillan, AM. (2005). Characterization of a
U2AF-independent commitment complex (E') in the mammalian spliceosome
assembly pathway. Mol Cell Biol 25, 233-240.

Newnham, C.M., and Query, CC. (2001). The ATP requirement for U2 snRNP
addition is linked to the pre-mRNA region 5' to the branch site. Rna 7, 1298-1309.

Sashital, D.G., Venditti, V., Angers, C.G., Comilescu, G., and Butcher, SE.
(2007). Structure and thermodynamics of a conserved U2 snRNA domain from
yeast and human. Rna 13, 328-338.

Fleckner, J ., Zhang, M., Valcarcel, J ., and Green, MR. (1997). U2AF65 recruits a
novel human DEAD box protein required for the U2 snRNP-branchpoint
interaction. Genes Dev 11, 1864-1872.

Zhang, M., and Green, MR. (2001). Identification and characterization of
yUAP/Sub2p, a yeast homolog of the essential human pre-mRNA splicing factor
hUAP56. Genes Dev 15, 30-35.

Xu, Y.Z., Newnham, C.M., Kameoka, S., Huang, T., Konarska, M.M., and Query,
CC. (2004). Prp5 bridges U1 and U2 snRNPs and enables stable U2 snRNP
association with intron RNA. Embo J 23, 376-385.

Dybkov, 0., Will, C.L., Deckert, J ., Behzadnia, N., Hartmuth, K., and Luhrmann,
R. (2006). U2 snRNA-protein contacts in purified human 17S U2 snRNPs and in
spliceosomal A and B complexes. Mol Cell Biol 26, 2803-2816.

Donmez, G., Hartmuth, K., Kastner, 8., Will, C.L., and Luhrmann, R. (2007). The
5' end of U2 snRNA is in close proximity to U1 and functional sites of the pre-
mRNA in early spliceosomal complexes. Mol Cell 25 , 399-411.

Weidenharnmer, E.M., Ruiz-Noriega, M., and Woolford, J .L., Jr. ( 1997). Prp31p

promotes the association of the U4fU6 X US tri-snRNP with prespliceosomes to
form spliceosomes in Saccharomyces cerevisiae. Mol Cell Biol 1 7, 3580-3588.

40

154.

155.

156.

157.

158.

159.

160.

161.

162.

163.

164.

165.

166.

Alvi, R.K., Lund, M., and Okeefe, RT. (2001). ATP-dependent interaction of
yeast U5 snRNA loop 1 with the 5' splice site. Rna 7, 1013-1023.

Newman, A.J., Teigelkamp, S., and Beggs, J .D. (1995). snRNA interactions at 5'
and 3' splice sites monitored by photoactivated crosslinking in yeast
spliceosomes. Rna 1, 968-980.

Kuhn, A.N., Li, Z., and Brow, D.A. ( 1999). Splicing factor Prp8 governs U4/U6
RNA unwinding during activation of the spliceosome. Mol Cell 3, 65-75.

Raghunathan, PL, 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.

Chen, C.H., Yu, W.C., Tsao, T.Y., Wang, L.Y., Chen, H.R., Lin, J .Y., Tsai, W.Y.,
and Cheng, SC. (2002). Functional and physical interactions between
components of the Prp19p-associated complex. Nucleic Acids Res 30, 1029-103 7.

Sashital, D.G., Comilescu, G., McManus, C.J., Brow, DA, and Butcher, SE.
(2004). U2-U6 RNA folding reveals a group II intron-like domain and a four-
helix junction. Nat Struct Mol Biol 11, 1237-1242.

Butcher, SE, and Brow, DA. (2005). Towards understanding the catalytic core
structure of the spliceosome. Biochem Soc Trans 33, 447-449.

Yean, S.L., Wuenschell, G., Termini, J ., and Lin, R.J. (2000). Metal-ion
coordination by U6 small nuclear RNA contributes to catalysis in the
spliceosome. Nature 408, 881-884.

Umen, J .G., and Guthrie, C. (1995). The second catalytic step of pre-mRNA
splicing. Rna 1, 869-885.

Datta, B., and Weiner, AM. (1993). The phylogenetically invariant ACAGAGA
and AGC sequences of U6 small nuclear RNA are more tolerant of mutation in
human cells than in Saccharomyces cerevisiae. Mol Cell Biol 13, 5377-5382.

Sontheimer, E.J., and Steitz, J .A. (1993). The U5 and U6 small nuclear RNAs as
active site components of the spliceosome. Science 262, 1989-1996.

O'Keefe, RT, and Newman, A.J. (1998). Functional analysis of the U5 snRNA
loop 1 in the second catalytic step of yeast pre-mRNA splicing. Embo J 1 7, 565-
574.

Makarov, E.M., Makarova, O.V., Urlaub, H., Gentzel, M., Will, C.L., Wihn, M.,

and Luhrmann, R. (2002). Small nuclear ribonucleoprotein remodeling during
catalytic activation of the spliceosome. Science 298, 2205-2208.

41

167.

168.

169.

170.

171.

172.

173.

174.

175.

176.

177.

178.

Nilsen, T.W. (2002). The spliceosome: no assembly required? Mol Cell 9, 8-9.

Tardiff, DE, and Rosbash, M. (2006). Arrested yeast splicing complexes indicate
stepwise snRNP recruitment during in vivo spliceosome assembly. Rna 12, 968-
979.

Behzadnia, N., Hartmuth, K., Will, C.L., and Luhrmann, R. (2006). Functional
spliceosomal A complexes can be assembled in vitro in the absence of a penta-
snRNP. Rna 12, 1738-1746.

Zhou, Z., Licklider, L.J., Gygi, SP, and Reed, R. (2002). Comprehensive
proteomic analysis of the human spliceosome. Nature 419, 182-185.

Jurica, MS, and Moore, M.J. (2003). Pre-mRNA splicing: awash in a sea of
proteins. Mol Cell 12, 5-14.

Jurica, M.S., Licklider, L.J., Gygi, S.R., Grigorieff, N., and Moore, M.J. (2002).
Purification and characterization of native spliceosomes suitable for three-
dimensional structural analysis. Rna 8, 426-439.

Deckert, J ., Hartmuth, K., Boehringer, D., Behzadnia, N., Will, C.L., Kastner, B.,
Stark, H., Urlaub, H., and Luhrmann, R. (2006). Protein composition and electron
microscopy structure of affinity-purified human spliceosomal B complexes
isolated under physiological conditions. Mol Cell Biol 26, 5528-5543.

Behzadnia, N., Golas, M.M., Hartmuth, K., Sander, B., Kastner, B., Deckert, J .,
Dube, P., Will, C.L., Urlaub, H., Stark, H., and Luhrmann, R. (2007).
Composition and three-dimensional EM structure of double affinity-purified,
human prespliceosomal A complexes. Embo J 26, 1737-1748.

Chen, Y.I., Moore, RE, Ge, H.Y., Young, M.K., Lee, TD, and Stevens, SW.
(2007). Proteomic analysis of in vivo-assembled pre-mRN A splicing complexes
expands the catalog of participating factors. Nucleic Acids Res.

Chen, Y.I., Maika, SD, and Stevens, SW. (2006). Epitope tagging of proteins at
the native chromosomal loci of genes in mice and in cultured vertebrate cells. J
Mol Biol 361, 412-419.

Das, R., Dufu, K., Romney, B., Feldt, M., Elenko, M., and Reed, R. (2006).
Functional coupling of RNAP II transcription to spliceosome assembly. Genes
Dev 20, 1100-1109.

Bird, G., Zorio, DA, and Bentley, D.L. (2004). RNA polymerase H carboxy-

tenninal domain phosphorylation is required for cotranscriptional pre-mRN A
splicing and 3'-end formation. Mol Cell Biol 24, 8963-8969.

42

179. Laing, J .G., and Wang, J .L. (1988). Identification of carbohydrate binding protein
35 in heterogeneous nuclear ribonucleoprotein complex. Biochemistry 27, 5329-
5334.

180. Wang, W., Park, J .W., Wang, J .L., and Patterson, R.J. (2006).

Immunoprecipitation of spliceosomal RNAs by antisera to galectin-1 and
galectin-3. Nucleic Acids Res 34, 5166-5174.

43

Chapter 2
Physico-chemical characterization of galectin-3-snRNP complexes and a

role in pre-mRNA binding

ABSTRACT

Previously, we showed that galectin-1 and galectin—3 are redundant pre-mRNA
splicing factors associated with the spliceosome throughout the splicing pathway. Here
we present evidence for the association of galectin-3 with snRNPs outside of the
spliceosome. Immunoprecipitation of HeLa nuclear extract using buffers optimal for in
vitro pre-mRN A splicing with anti-galectin-3 resulted in the coprecipitation of the 5
snRNAs needed for pre-mRN A splicing, snRNP proteins and other RNA processing
factors including the SMN protein and PSF. Galectin-3 also co-sedimented with snRNP
complexes when nuclear extract was fractionated on glycerol gradients. This co-
sedimentation represents bona fide galectin-3-snRNP complexes as immunoprecipitation
of gradient fractions with anti-galectin-3 yielded both snRN A and associated proteins. In
particular, one fraction at approximately 10S showed an association of galectin-3 with U1
snRNP that was sensitive to treatment with ribonuclease A. We tested the ability of this
U1 particle to recognize an exogenous pre-mRNA substrate. We found this isolated
galectin-3-U1 snRNP particle was sufficient to load galectin-3 onto the pre-mRNA
substrate under conditions that allow early splicing complexes to form. These data
suggest a mechanism for the entry of galectin-3 into the pre-mRNA splicing pathway and
describe a new class of polypeptides associated with spliceosomal snRNPs.

INTRODUCTION

44

Pre-mRNA splicing involves nearly 300 proteins and 5 snRNAs [1-4]. These
components are assembled into the machinery used to perform the splicing chemistry of
intron removal and exon ligation. The canonical model for the assembly process involves
the stepwise addition of the snRNPs into early, commitment and active complexes. U1
snRNP assembles onto the pre-mRNA at the 5’ splice site in the absence of ATP.
Addition of ATP allows U2 snRNP to recognize U2AF at the branchpoint and form a
stable commitment complex. Finally the U4/U6.U5 tri-snRNP particle binds at the 3’
splice site resulting in the active spliceosome [5, 6]. In addition, various protein
cofactors are differentially incorporated into the complexes and then disassemble once an
intron is removed and exons are ligated. In this step-wise assembly model, individual U1
and U2 snRNPs recycle directly while U4, U5 and U6 must reassemble into the tri-
snRNP for reutilization in subsequent rounds of splicing.

Recently another model of spliceosome assembly has been described. In this
model, a large complex containing all 5 snRNPs and many splicing proteins assembles in
the absence of a splicing substrate scaffold [7-9]. This penta-snRNP complex then
assembles onto the splicing substrate. A series of remodeling events with addition and
release of other proteins ensues to generate the catalytically active spliceosome.
Following splicing chemistry and release of products, the unit either remodels to form an
active penta-snRNP complex or disassembles before reforming another penta-snRNP
particle.

We have shown previously that galectin-1 (Gall) and galectin-3 (Gal3) were
required and redundant splicing factors using a cell free splicing assay. The key findings

were: (a) depletion of both galectins from HeLa nuclear extracts (NE) abolished splicing

45

activity and blocked spliceosome formation at an early complex; (b) both splicing activity
and spliceosome formation were restored by addition of recombinant Gall or Gal3; (c)
each galectin was a component of early and active splicing complexes as determined by
co-immunoprecipitation of splicing substrate at early times and all mRNA species as they
appeared in active complexes by galectin-specific antisera; ((1) each galectin was
incorporated into spliceosomes in a mutually exclusive manner and (e) neither galectin
bound directly to the pre-mRNA substrate [10-12].

How are galectins assembled into early splicing complexes? In this manuscript
we report that in the absence of splicing substrate, Gal3 is associated with several snRNP
particles and, in particular, the U1 snRNP under conditions of the in vitro splicing assay.
We present evidence that one mechanism of Gal3 incorporation into the splicing pathway
is mediated by the U1 snRNP particle.

MATERIALS AND METHODS
Antibodies

Polyclonal rabbit antibodies directed against Gal3 [13] have been previously
described. These antibodies were covalently cross-linked to protein A-Sepharose CL-4B
beads (Amersharn Biosciences) as previously described [11] using a 2:1 ratio of
antiserum to protein-A beads. A mouse monoclonal antibody against trimethylguanosine
(TMG; K121) was purchased as an agarose bead conjugate (Calbiochem). These
antibodies were used for immunoprecipitation experiments.

For immunoblotting, polyclonal rabbit antibodies against Gall have been
described previously [12]. The following antibodies were purchased for immunoblotting:

Affinity purified rabbit anti-PSF (PTB-associated splicing factor), goat anti-Slu7 (Santa

46

Cruz Biotechnology); rabbit anti-TFII-I, rabbit anti-Gemin4 (Bethyl Laboratories);
mouse monoclonals anti-SMN (survival of motor neurons protein) and anti-Ran (BD
Transduction Laboratories); mouse monoclonal anti-U1-70K (Synaptic Systems); and
human autoimmune sera ENA anti-Sm (The Binding Site). Rabbit anti-RAP30 was a
kind gift from Dr. Zach Burton (Michigan State University). Rabbit anti-Gal3, rat
monoclonal anti-Mac2 [14], or mouse monoclonal NCLGAL3 (Vector Laboratories)
were used for blotting Gal3 protein. Primary mouse monoclonal antibodies were detected
by goat anti-mouse IgG light chain specific-HRP (horseradish peroxidase) conjugates
(Jackson ImmunoResearch Laboratories). When blotting the bound fractions of
immunoprecipitation experiments, primary polyclonal rabbit blotting antibodies were
probed with secondary monoclonal mouse anti-rabbit IgG light chain specific-HRP
conjugates (Jackson ImmunoResearch Laboratories). All other secondary antibody-HRP
conjugates (Pierce Biotechnology) were directed against both the heavy and light chains
of the primary blotting antibodies.
Polyacrylamide gel electrophoresis and western blotting

For protein analysis, samples were subjected to 10% or 12% SDS-PAGE as
described by Laemmli [15]. Proteins were electrophoretically transferred from the gel
onto Hybond nitrocellulose membrane (Amersham Biosciences) in transfer buffer (25
mM Tris, 193 mM glycine and 20% methanol, pH 8.3). Following transfer, membranes
were blocked ovenright in 10% nonfat dry milk in Tris-buffered saline containing Tween-
20 (10 mM Tris, pH 7.5, 0.5 M NaCl, 0.05% Tween 20, T-TBS). Primary antibodies for
immunoblotting were diluted in 1% milk-T-TBS and incubated on the membrane for 1

hour at room temperature. After washing four times, 15 minutes each, in T-TBS, the

47

appropriate secondary antibody conjugated to HRP was added in 1% milk-T-TBS for 1
hour. Following four T-TBS washes as above, proteins were visualized using the
Western Lightning Chemiluminescence System (Perkin Elmer Life Sciences).

For RNA samples, the RNA was extracted as described below and precipitated
with 3 volumes of ethanol at -80°C. The precipitated RNA was dissolved in 10 pl of
sample buffer (9: 1/ fonnamidezbromophenol blue) and subjected to electrophoresis
through 13% polyacrylamide (bisacrylamide—acrylamide 1.9250 [wt/wt])-8.3 M urea gels,
run in 1X TBE (90 mM Tris base, 90 mM boric acid and 2.5 mM EDTA, pH 8.0). The
radioactive RNA species were revealed by autoradiography. For northem analysis the
RNA was transferred via wicking in 20X SSC (3 M NaCl, 0.3 M sodium citrate)
overnight onto a nylon membrane (Hybond-N, Amersham Biosciences) and cross-linked
by exposure to UV light (1200 uJoules).

Immunoprecipitation

Nuclear extract was prepared as described by Dignam et al. [16] from HeLa S3
cells obtained fiom the National Cell Culture Center (Minneapolis, MN) and was
dialyzed into buffer D (20 mM HEPES, pH7.9, 100 mM KCl, 0.2 mM EDTA, 0.5 mM
DTT, 20% glycerol). HeLa NE (30 pl) was incubated with 3 mM MgC12, 0.5 mM ATP
and 20 mM creatine phosphate in a 50 111 reaction at 30° C for 30 minutes before use in
immunoprecipitations [17]. After incubation, the reaction was diluted in 0.5 ml 60%
buffer D (D60) with 0.05% Triton-X 100 (TX) and incubated with 15 - 20 ul of antibody
coupled protein-A beads for 1 hour at 4°C with rocking. The beads were then washed
with 1 ml D60+0.05% TX three times and eluted twice with 20 [41 SDS-PAGE sample

buffer. The elution was divided into two aliquots for analysis of proteins and RNA.

48

RNA was extracted by treating the sample with proteinase K (4 mg/ml final
concentration) at 37°C for 20 minutes and diluting to 100 pl with 125 mM Tris (pH 8), 1
mM EDTA, 300 mM sodium acetate. RNA was extracted by mixing with 200 pl of
phenol—chlorofonn (50:50 [vol/vol]), followed by 100 pl of chloroform.

For immunoprecipitation with anti-TMG, 20 pl antibody beads were mixed with
40 pl of pre-incubated NE for 1 hour at 4°C with rocking. Beads were washed 3 times
with 0.5 ml D60+0.05% TX and eluted with 20 pl of 2X SDS sample buffer. Protein-G
agarose beads without antibody were used as a negative control. Protein analysis was
canied out with 15 pl of the eluted material as described above. For the sequential
immunoprecipitation, 25 pl of the coupled anti-TMG beads were incubated with 50 pl of
pre-incubated NE and treated as above. The beads were eluted twice with 15 pl 5 mM
cap structure analog (m7G(5')ppp(5')G; New England Biolabs) in D60. The elutions were
combined and mixed with 20 pl of anti-Gal3 coupled beads and immunoprecipitation was
carried out as above. The bound material was eluted with 40 pl 2X SDS sample buffer
and proteins were analyzed as described above.
Northern analysis

The following oligonucleotides were synthesized at Research Technology Support
Facility at Michigan State University: U1: TCCCCTGCCAGGTAAGTATC, U2:
TTAGCCAAAAGGCCGAGAAGCGAT, U4: GGGGTATTGGGAAAAGTTTTC, U5:
GATTTATGCGATCTGAAGAGAAACC, U6: TTCTCTGTATCGTTCCAAT, SS
rRNA: TTCAGGGTGGTATGGCCGTAGAC. The oligonucleotide probes were labeled
with 32P04 using T4 Polynucleotide Kinase (Invitrogen). The cross-linked nylon

membrane was pre-hybridized in 20 ml of hybridization solution containing 20%

49

deionized forrnamide, 3X SSC, 5X Denhardt's, 50 mM NazHPOr/NaHzPO4, pH 6.8, 0.1%
SDS and 0.1 mg/ml heat denatured herring sperm DNA for about 2 hours at room
temperature. After pre-hybridization, the appropriate 32P-labeled probes (~2 pmol) were
added to the hybridization solution and the membrane was hybridized overnight (~16
hours) at room temperature. The membrane was then washed once with 30 ml 2X SSC +
0.1% SDS at room temperature for 20 minutes. Northern hybridization was quantitated
by phosphorimage analysis (Molecular Dynamics).
Glycerol gradient

For glycerol gradient sedimentation, a 250 pl reaction containing 150 pl HeLa
NE, 3 mM MgC12, 0.5 mM ATP and 20 mM creatine phosphate was incubated for 30
minutes at 30°C and loaded onto a 5 ml 12-32% glycerol gradient in D60 (minus
glycerol). The gradient was centrifuged in a Beckman SW50.1 rotor at 44,000 rpm, at
4°C for 3.5 to 4 hours. Gradient fractions were collected manually as 250 pl aliquots
from the top. For RNA and protein analysis, 10% of each fraction was subjected to gel
electrophoresis and analyzed as described above. Size markers separated on parallel
gradients include immunoglobulin G (78), thyroglobulin (19S) and post-nuclear
supernatant for ribosomal subunits (408 and 608).
Gradient complexes

For immunoprecipitation experiments from gradient fractions, 200 pl of the
indicated fractions were pooled and the total volume brought up to 600 pl with D60. The

total reaction was incubated with 20 pl antibody coupled beads at 4°C for 1 hour with

rocking, washed 5 times with 0.5 ml D60+0.05% TX and eluted with 40 pl 2X SDS

50

sample buffer. The elutions were split into 20 pl aliquots and analyzed for protein and
RNA as described above.

To test ribonuclease (RN ase) sensitivity of gradient complexes, 50 p1 of the
pooled indicated fraction were treated with 2 pg RNase A for 30 minutes at 30°C, after
which it was mixed with 15 pl anti-Gal3 coupled beads for 1 hour at 4°C with rocking in
0.5 ml D60+0.05% TX. The beads were washed 5 times with 0.5 m1 binding buffer and
eluted with 30 pl 2X SDS sample buffer.

To test the association of Gal3 with pre-mRNA substrate the indicated pooled
fractions were subjected to digestion by micrococcal nuclease. Each reaction assembled
included 45 pl of pooled fractions 3 and 4, 3 mM MgC12, 4 mM CaClz and 4000 gel units
of micrococcal nuclease (New England Biolabs). Control reactions without micrococcal
nuclease were supplemented by equal volumes of D60. Reactions were incubated at
37°C for 30 minutes. Micrococcal nuclease activity was stopped by adding EGTA to a
final concentration of 8 mM. MINX pre-mRNA has been described previously and the
plasmid containing this construct was a gift from Dr. Sue Berget (Baylor College of
Medicine, Houston, TX) [18]. The MINX pre-mRNA was labeled with [32P]GTP and a
monomethyl cap added during in vitro transcription by SP6 RNA polymerase [19]. After
stopping the digestion reaction, 32P-labeled MINX pre-mRNA was added to each tube,
incubated for 15 minutes at 30°C and incubated with 15 pl antibody-coupled beads in 0.5
ml D60+0.05% TX for 1 hour at 4°C with rocking. The beads were washed five times
with 0.5 ml binding buffer and eluted with 40 pl 2X SDS sample buffer. Bound MINX
pre-mRNA was detected by subjecting an aliquot of the bound material to

electrophoresis, as described above, drying the gel and phosphorimage analysis.

51

RESULTS
Galectin-3 is associated with snRNPs in the absence of splicing substrate

NE was incubated without splicing substrate at 30 °C with ATP for 30 minutes to
disassemble endogenous splicing complexes [17]. This NE was then subjected to
immunoprecipitation with anti-Gal3 serum. The bound material was analyzed for
specific RNA species by northern hybridization (Figure 1A). First, anti-Gal3 co-
precipitated U1, U2, U4, U5 and U6 snRN As whereas pre-immune serum did not.
Second, SS rRNA, a prominent RNA species in NE, was not observed at all in the anti-
Gal3 precipitate. Phosphorimage quantitation revealed that, at the most, 0.1% of the
nuclear 5S rRNA was found in the anti—Gal3 precipitated fraction. In contrast,
approximately 2-10% of each of the U-rich snRNAs was co-precipitated by anti-Gal3.
Finally, since the RNA probes for the hybridization were of the same specific
radioactivity, direct comparisons of the U-rich snRNA could be made between any two
lanes of Figure 1A. Such a comparison indicated that the ratios of the snRNAs in the
anti-Gal3 precipitate were different from the corresponding ratios in the NE subjected to
precipitation. For example, the ratio of U4 to U5 snRNA in the input is 2.3, but in the
ant-Gal3 bound material, this ratio changes to 4.8 (see lanes 1 and 3; Fig. 1A). All of
these results suggest that the association of snRNAs with the anti-Ga13 precipitate was
specific.

A similar conclusion regarding the specificity of the anti-Gal3 precipitation was
obtained through analysis of protein components by western blotting (Fig. 1B). In

addition to its cognate antigen, anti-Gal3 co-precipitated several key proteins that were

52

Figure 1. Analysis of nuclear RNA and proteins immunoprecipitated by anti-Gal3.
NE was pre-incubated as described in Materials and Methods. The reaction was then
incubated with antibody-coupled beads, either Gal3 (chal3) or pre-immune serum (PI).
Panel A shows a northern blot of 25% of the bound material probed with 32P-labelled
oligonucleotides complementary to the RNA species indicated on the left. Lane 1: NE
representing 12% of the amount subjected to immunoprecipitation. Lane 2: RNA
species bound by pre-immune serum; lane 3: RNA species bound by anti-Gal3. Panel B
shows western blots of the bound material. Antibodies used to detect the proteins are
indicated at right.

53

D>
NE
PI
orGal3

 

 

 

__Gal3

 

\ Ran

Sm B/B,

 

 

3453:2922: . Sm D

 

Gall

54

not found in the corresponding fraction of pre-immune serum. These include: (a) the Sm
core polypeptides B/B' and D of snRNPs and the 70K protein specific to U1 snRNP; (b)
the splicing factor PSF [8] and the general transcription factor TFII-I, both of which have
been recently identified to be in complexes containing Gal3 (P. Voss, J .L. Wang,
unpublished observations); and (c) the Survival of Motor Neuron (SMFD protein. In
contrast, Slu7, a second-step splicing factor, was not detected in the anti-Gal3 precipitate,
ruling out the possibility that anti-Gal3 precipitated any endogenous active spliceosomes
present in the NE. Similarly, the nuclear transport factor Ran was not found in the anti-
Gal3 bound fraction. Finally, the anti-Gal3 precipitate did not contain Gall, a finding
consistent with the mutually exclusive association of either galectin with spliceosomes
that we previously described [12].

To provide additional evidence for the association of Gal3 with nuclear snRNPs
independent of Gal3 immunoselection, snRNPs were affinity purified with antiserum
specific for the trimethyl guanosine cap of U1 , U2, U4 and U5. This procedure
precipitated all five snRNAs, as determined by ethidium bromide staining of the gel and
by northern hybridization (data not shown). The core Sm B/B’ and D proteins, as well as
Gal3 and Gall, were detected in the anti-TMG bound material (Fig. 2A, lane 3). In
contrast, RAP30, a subunit of the general transcription factor TFII-F, was not detected in
the anti-TMG precipitate (Fig. 2A, lane 3). Neither galectin could be detected in the
bound material from control beads (Fig. 2A, lane 2). Lastly, the anti-TMG selected
snRNPs were eluted with soluble cap analogue and then subjected to further precipitation
by anti-Gal3. Panel B of Figure 2 shows that Sm B/B’ proteins are co-precipitated by

anti-Gal3 antibodies in the sequential immunoprecipitation protocol. Thus, these

55

Figure 2. Analysis of proteins immunoprecipitated by anti-TMG. NE was pre-
incubated for 30 minutes at 30°C, then passed over anti-TMG-coupled beads (orTMG) or
naked protein-G agarose beads (Control). Panel A: Western blots of the bound material
for the proteins indicated at right; lane 1; NE represents 20% of the amount subjected to
immunoprecipitation; lane 2, proteins bound by the control beads; lane 3, proteins bound
by anti-TMG beads. Panel B shows the results of a double immunoselection. The pre-
incubated NE was first incubated with anti-TMG beads. The aTMG bound material was
eluted with cap structure analog and then incubated with anti-Gal3 beads. Material from
both the unbound (lane 1) and bound (lane 2) fractions of the anti-Gal3 column was
western blotted for the proteins indicated at right. Approximately 85% of both the
unbound and bound material is shown in Panel B.

56

 

   

 

'5 (D
‘2 E B.
L; o % Cap structure elution
D from orTMG
Fl SIT]. BB, 2
(6 M
‘3 a
“'3 a
5 "o
.8 5
r: o
D m
Sm D _‘ Gal3
Ga13
I“ Sm B/B’
1 2
new Gall
- Rap30
1 2 3

57

reciprocal co-immunoprecipitation experiments suggest an association of Gal3 with
snRNP complexes, in the absence of any pre-mRNA scaffold.
Galectin-3 is associated with multiple snRNP complexes

Multiple snRNP complexes exist in the nucleus including the U4/U 6 di-snRNP
and the U4/U6.U5 tri-snRNP as well as the mono-snRNPs U1 and U2. Recently,
complexes containing all five snRNPs have been described [7, 8, 20]. In light of the
different ratios of snRNAs immunoprecipitated from NE with anti-Gal3, we hypothesized
Gal3 is associated with complexes containing multiple snRNPs. To test this, pre-
incubated NE was fractioned by glycerol gradient sedimentation and the distribution of
snRNAs and several RNA processing proteins were determined. Both the northern
blotting of U1 snRNA (Fig. 3A) and the western blotting of U1-70K protein (Fig. 3B)
indicate that fiactions 3-5 contain the mono U1 snRNP predominantly, consistent with
previous findings that U1 snRNP sediments at about 10S (see Table I) [21]. Fractions at
higher molecular weight (~19S and greater) contained various combinations of the
snRNAs (Fig. 3A).

Several key points should be highlighted regarding the distribution profile of
Gal3, our protein of interest. First, fraction 1 yielded the most intense Gal3 staining, at
the top of the gradient (Fig. 3B). Second, the protein could be detected through fi'action
10 (~30S). Finally, although Gal3 was not detected by direct western blotting of an
aliquot of individual fractions beyond fraction 10, we did find the protein at least through
fraction 18 (>608) after enrichment by anti-Gal3 immunoprecipitation (see below). A

similar observation was made for the Sm core proteins and U1-7OK. The distribution of

58

Figure 3. Analysis of nuclear RNA and proteins separated on a 12-32% glycerol
gradient. NE was treated as in Figure 1. Reactions were then loaded onto gradients and
centrifuged for 3.5 hours at 44,000 rpm at 4°C. The gradients were fractionated by
collecting 250 1“] aliqouts from the top. Panel A shows a northern blot of 10% of each
fraction using 2P-labelled oligonucleotides specific for each snRN A. Size markers run
in parallel are indicated at the top. Panel B shows western blots of 10% of each fraction
for the proteins indicated at left.

59

 

wow

it. ill! III-2‘21!!!

1

 

m9. mg. my

990
_-__n_._.

2.2m

van- _. D

D Em
.mE Em

w:
v m:
\ 3
/ 5

/S

.m

60

the SMN protein across the gradient suggests there is little uncomplexed SMN in NE, an
observation made previously [22].

Individual or pooled fractions throughout the gradient were immunoprecipitated
with anti-Ga13 antibodies to determine the distribution of Gal3-containing complexes
(Fig. 4). Panel A shows northern analysis of the immunoprecipitated material using
probes for all five snRN As; panel B shows the western blotting results for SMN, U1-
70K, Sm B/B’ and the cognate antigen Gal3. The anti-Gal3 precipitate of pooled
fractions 3 and 4 yielded a single snRNA, U1, and associated proteins Sm B/B' and U1
70K (lane 3, panels A and B, Fig. 4). This suggests that at least some of the Gal3 in
fractions 3 and 4 was associated with the mono U1 snRNP. On the other hand, the
absence of U1 70K-protein (Fig. 3B) and the miniscule amount of U1 snRNA in fi'action
1, relative to the much more intense U1 bands in fractions 3 and 4 (Fig. 3A), both suggest
that the majority of Gal3 is not associated with a snRNP complex. Consistent with this
notion, fraction 1 yielded no snRNA or other proteins in the anti-Ga13 precipitate (lane 2,
panels A and B, Fig. 4).

The anti-Gal 3 precipitate of fractions 6 and 7 (~19S; lane 4), fractions 9-11 (lane
5), fractions 13 and 14 (>40S; lane 6) and fractions 16-18 at ~6OS (lane 7) yielded
multiple snRNAs in various proportions along with Sm B/B’ and U1 70K polypeptides
(Fig. 4). Gal3, the cognate antigen, was found in all immunoprecipitated material. On
the other hand, the SS rRNA was not detected in the anti-Gal3 precipitated material (data
not shown). All of these results suggest that Gal3 is associated with numerous and
distinct endogenous snRNP particles in NE. In very high molecular weight fractions

(fractions 10 and beyond), we also found the SMN protein co-precipitated by anti-Gal3

61

Table 1. Listing of selected RNPs involved in pre-mRNA splicing and reported S values.

62

 

 

 

 

 

 

 

 

 

 

RN P Reported S value Reference
U1 snRNP 108 [21]
U4/U6 di-snRNP 12S [23]
U2 snRNP 17S [24]

U5 snRNP 20S, 358 [25, 26]
U4/U6.U5 tri-snRNP 25S [25]
S. cerevisiae penta-snRNP 458 [7]
PCC 50 — 60$ [8]
Spliceosome (various complexes) 40 — 608 [3, 26, 27]

 

63

(Fig. 4B). Although these precipitates also contain various snRNAs, we do not know
whether all three components are in the same complex or whether there are separate
complexes.

The U1 snRNP contains the core Srn proteins and the U1 -specific proteins 70K,
A1 and C1 assembled onto the U1 snRNA. Our hypothesis was that the association of
Gal3 with U1 snRNP would be dependent on the integrity of the U1 snRN A. To test this,
pooled material from fi'actions 3 and 4 was immunoprecipitated with anti-Gal3 before
and after RN ase A treatment. Fig. 5 shows that co-precipitation of the U1-7OK protein
by anti-Ga13 is abolished following degradation of U1 snRNA (compare lane 3 to lane 2).
In contrast, the co-precipitation of the SMN protein from fractions containing large
snRNA complexes was unaffected by RNase A treatment (Fig. 5, lanes 4 and 5). As
expected, the immunoprecipitation of Gal3 itself is unaffected by RNase A treatment
(data not shown).
The U1 snRNP mediates the loading of Gal3 onto pre-mRNA

The binding of U1 snRNP to the pre-mRNA substrate at the 5' splice site results
in the formation of the early (E) complex in spliceosome assembly [18, 28, 29]. We had
previously documented that either Gall or Gal3 is associated with E-complexes [12].
Thus, our present findings that at least some Gal3 is bound to U1 snRNP prompted the
question whether such Gal3 -containing U1 snRNPs in fiactions 3 and 4 could bind to 32P-
labeled pre-mRNA substrate under conditions that lead to B complex formation. Fig. 6
shows that, in the presence of fractions 3 and 4, anti-Gal3 co-precipitated exogenously
added MINX pre-mRNA substrate (lane 4) whereas pre-immune serum failed to yield the

same result (lane 3). That this interaction is RNA dependent is shown in lane 5 in which

64

Figure 4. Analysis of RNA and proteins immunoprecipitated by anti-Gal3 from
glycerol gradient fractions. Indicated fractions (pooled, if necessary) from glycerol
gradient fractionation of NE (see Fig. 3) were subjected to immunoprecipitation by anti-
Gal3 coupled beads. Bound material was collected and analyzed for RNA (Panel A) or
proteins (Panel B). Panel A shows a northern blot of the bound material (~50%) of the
anti-Gal3 beads from each immunoprecipitation. 32P-labeled oligonucleotides specific
for each snRNA were used to detect the RNA species indicated at right. Panel B shows a
western blot for proteins immunoprecipitated from the indicated fractions by anti-Gal3
beads. Material subjected to western blotting represents 50% of the bound fraction; lane
2, immunoprecipitate from fraction 1; lane 3, immunoprecipitate of fractions 3 & 4; lane
4, immunoprecipitate of fractions 6 & 7; lane 5, immunoprecipitate of fractions 9, 10, 11;
lane 6, immunoprecipitate of fractions 13 & 14 and lane 7, immunoprecipitate of
fractions 16, 17, 18. NE (lane 1) is included to verify the identity and migration of the
proteins in western blotting. Respective antibodies against the proteins indicated at right
were used in western blotting.

65

 

 

NE 1 3+4 6+7 9-11 l3+1416-18 Fraction

H . W U1-70K

Vii. M SmB/B’

 

 

66

fractions 3 and 4 were treated with micrococcal nuclease (whose activity was
subsequently inhibited by the addition of EGTA) prior to incubation with MINX. After
nuclease digestion, the level of MINX precipitated by anti-Gal3 bound was reduced to
levels bound to pre-immune serum. However, residual micrococcal nuclease activity did
not degrade the MINX substrate as approximately equal amounts of MINX were used for
the anti-Gal3 precipitation (Fig. 6A, lanes 1 and 2). Northern blotting analysis showed
that the U1 snRNA in fractions 3 and 4 was indeed degraded by the micrococcal nuclease
treatment (Fig. 6B, lanes 1 and 2). Examination of the precipitated material shows the
presence of U1 snRN A in the anti-G313 precipitate, but not in the pro-immune control or
in fractions first treated with micrococcal nuclease and precipitated with anti-Ga13 (Fig.
6B, lanes 3-5). Finally, blotting for the Gal3 polypeptide in the bound material shows
no significant change in precipitation of the protein after nuclease treatment (Fig. 6C,
lanes 4 and 5). These results suggest that Gal3 is first assembled onto the pro-mRNA
substrate via its interaction with the U1 snRNP particle.
DISCUSSION

Previous experiments had documented that Gall and Gal3 are factors involved in
pre-mRNA splicing assayed in a cell-free system [10, 11]. Depletion of the galectins
resulted in an arrest of spliceosome assembly at an early step, corresponding to the H-/E-
complex. Given that neither Gall nor Gal3 interacts directly with pre-mRN A [12], it was
of some interest to define how and at what step either protein is brought into the
spliceosome. The most important conclusion derived from the present series of studies
is, therefore, that the U1 snRNP can mediate the loading of Gal3 onto the pre-mRNA

substrate during spliceosome assembly (see Figure 7). This association of U1 snRNP

67

Figure 5. Effect of RNase A treatment on Gal3 association with U1 snRNP.
Fractions 3 and 4 or fraction 15 of NE separated on a glycerol gradient (see Fig. 3) were
subjected to treatment with RNase A or a control mock treatment at 30°C for 30 minutes.
After treatment, the reactions were incubated with anti-Gal3 coupled beads and the bound
material analyzed for protein. Bound material from the mock treated (lanes 2 and 4) or
RN Ase treated (lane 3 and 5) fractions was western blotted for the proteins indicated on
the right.

68

Fraction 3+4 Fraction 15

 

+ _ + RNase A
U1-70K

   

69

with Ga13 also exists outside of the pre-mRNA splicing pathway, i.e. in HeLa NE without
any pre-mRNA scaffold. In addition to associating with U1 snRNP outside of the
splicesome, we have found that Gal3 exists in larger complexes containing multiple
snRN As and other RNA processing factors (Figure 7). The involvement of Gal3 in the
pre-mRNA splicing pathway, and in particular, with the spliceosome has been shown
previously [12].

The conclusion that U1 snRNP facilitates the loading of Gal3 onto the pre-mRNA
scaffold is based on several key considerations. First, Gal3 is associated with multiple
snRNP complexes in the absence of splicing substrate as deduced from
immunoprecipitation analysis of glycerol gradient fractions derived from NE. In
particular, the position of sedimentation (~10S; see Table I) and the RNA and protein
composition of fractions 3 and 4 indicate these fiactions contain, predominantly, the
mono U1 snRNP. Irnportantly, the anti-Gal3 precipitate of fractions 3 and 4 contained
U1 snRNA and the Ul-specific polypeptide U1-70K, suggesting that some of the Gal3 in
these fractions was associated with the mono U1 snRNP.

Second, purified U1 snRNP binds selectively to the 5' splice site [29] and in the
canonical step-wise scheme of spliceosome assembly, the U1 snRNP participates early
during assembly and is required for the stable association of other snRNPs with pre-
mRNA [18, 30]. A discrete ATP-independent complex containing U1 snRNP, designated
the E-complex, is the first specific complex detected by gel filtration [31] and it
constitutes the functional precursor to the ATP-dependent A complex, committing the

pre-mRNA to the spliceosome assembly pathway [6].

70

Figure 6. Immunoprecipitation by anti-Gal3 of radio-labeled pre-mRN A after
incubation with glycerol gradient fractions. Pooled fractions 3 and 4 were either
treated with micrococcal nuclease or mock treated at 37°C for 30 minutes and then
nuclease activity stopped by addition of EGTA. 32P-labeled MINX pre-mRNA was
added to the reactions and incubated for 15 minutes at 30°C. The reactions were
incubated with antibody coupled beads, either anti-Gal3 (orGal3) or pre-immune serum
(PI). Panel A shows the MINX pre-mRNA by autoradiography. Panel B shows a
northern blot for the U1 snRNA. Panel C shows a western blot for the Ga13 protein.
Input represents the fi'action material and 32P-MINX subjected to immunoprecipitation in
reactions either mock treated (lane 1) or treated with micrococcal nuclease (lane 2).
Bound material from immunoprecipitation of mock-treated fractions by pre-immune
serum (lane 3) and anti-Gal3 (lane 4) is shown. Bound material from
immunoprecipitation by anti-Gal3 beads using fractions first treated with micrococcal
nuclease is shown in lane 5.

71

Bound

 

 

A. Input
_ +
C.
1 2

Pl orGal3 orGal3 1P Antibody

- - + Micrococcal
Nuclease

MINX

 

U1

. - Gal3

72

Finally, we have taken advantage of this initial, ATP-independent recognition of
the 5' splice site by U1 snRNP as a functional test of the Gal3-containing U1 snRNP
complex in fractions 3 and 4. Under conditions for E-complex formation (30° C in the
absence of ATP), incubation of splicing substrates with fractions 3 and 4 should result in
the association of Gal3 with the pre-mRNA. Indeed, we were able to demonstrate this by
immunoprecipitating fractions 3 and 4 with antibodies against Gal3 and detecting
radioactive MIN X pre-mRNA as well as U1 snRNA.

A key caveat to our conclusion is that previous studies on the protein composition
of purified U1 snRNP all describe the same core Sm proteins as well as Ul-specific
polypeptides: (a) U1 70K (~70 kD); (b) Ul-specific A (~34 kD); (c) Sm B/B' (~28 kD);
(d)U1-specific C (~22 kD); (e) Sm D (~16 kD); (f) Sm E (~12 kD); (g) Sm F (~11 kD);
and (h) Sm G (~9 kD). The SDS gels of these studies did not reveal any band (~30 kD)
that might correspond to G313 [32-34]. To reconcile this apparent discrepancy with our
present observations, we note that the protocols used for the purification of U1 snRNP all
involved the use of buffers of high ionic strength (150 mM NaCl or 175 mM NH4C1).
We had shown, however, that antibodies directed against Gall or Gal3 can
immunoprecipitate radioactive spliceosomal RNA species under conditions of the
splicing assay (60 mM KCl) but that higher salt concentrations (130 mM or greater)
release the galectins from spliceosomal complexes [12]. In the context of the present
study, we have also observed that the co-precipitation of snRNAs by anti-G313 was
sensitive to disruption by high ionic strength (K.C. Haudek and R.J. Patterson,
unpublished observations). It seems reasonable to expect, therefore, that snRNPs and

spliceosomes isolated under splicing conditions (60 mM salt) would contain Gal3.

73

Figure 7. Diagram showing the association of Gal3 with snRNPs and the pre-mRNA
splicing substrate. The pre-mRNA is shown as two rectangular exons joined by a single
line intron. Newly identified associations of Gal3 with snRNPs outside the splicesome
are indicated on the right. Gal3 can enter the splicing pathway at an early assembly stage
via its association with U1 snRNP. In addition, Gal3 does associate with multiple
snRNPs in larger complexes outside of the spliceosome, shown here as the U4/U6.U5 tri-
snRNP. Although U2 exists as a single snRNP particle in the nucleus, it is unknown
whether Gal3 can interact with this single snRNP alone, illustrated by a semi-hatched
Gal3. Whether Gal3 interacts with snRNPs being recycled between rounds of splicing or
nascent snRNPs in the nucleus is unknown. Solid arrows indicate steps supported by
data shown previously or contained in this report. Dashed arrows represent hypothetical
galectin involvement in assemblies or mechanisms of snRNP loading onto pre-mRNA.
Unlabeled shaded ovals indicated previously identified and novel proteins associated with
snRNPs isolated under conditions optimal for in vitro splicing activity

74

Pre-mRNA splicing Galectin-snRNP complexes

 

75

Peng et al. recently reported the isolation, under conditions of the in vitro splicing
assay (60 mM salt), of a macromolecular complex that assembles in the absence of pre-
mRN A substrate and contains all five uracil-rich snRNAs [8]. Mass spectrometry
analysis of this complex, designated as the PCC (PSF-containing complex; see Table 1),
revealed a protein composition similar to, but nevertheless distinct from, the
corresponding composition of fully assembled active spliceosomes. Ga13 was found as a
component of the PCC. Our present findings, that all five snRNAs are co-precipitated by
anti-Gal3, either from complete NE or from selected gradient fractions, raises the
possibility that we also have isolated a mammalian penta-snRNP, possibly similar if not
identical to the PCC. In fact, multiple components (PSF, TFII-I, SMN) identified in the
PCC were detected with Ga13 in the anti-Ga13 precipitate from NE. Together with the
discovery of a functional, preassembled penta-snRNP in yeast [7], these results suggest
that, under certain conditions, spliceosome assembly may be facilitated through the
association of large ribonucleoprotein complexes that are already preformed in the
absence of a pre-mRNA scaffold. In addition, the identification of novel proteins
complexed with snRNAs under less stringent isolation conditions, characterized in this
report and previously [7, 8], raises the possibility that the network of snRNA-interacting
proteins outside of the spliceosome contains more members than originally described.

Although we have been able to demonstrate a function for the Ga13-U1 snRNP
complex in terms of 5' splice recognition and E-complex formation, the role(s) of the
higher molecular weight multi-snRNP complexes containing Ga13 remains as a challenge

for future studies.

76

REFERENCES

l.

10.

11.

12.

Rappsilber, J ., Ryder, U., Lamond, AL, and Mann, M. (2002). Large-scale
proteomic analysis of the human spliceosome. Genome Res 12, 1231-1245.

Zhou, Z., Licklider, L.J., Gygi, SR, and Reed, R. (2002). Comprehensive
proteomic analysis of the human spliceosome. Nature 419, 182-185.

Jurica, M.S., Licklider, L.J., Gygi, S.R., Grigorieff, N., and Moore, M.J. (2002).
Purification and characterization of native spliceosomes suitable for three-
dimensional structural analysis. Rna 8, 426-439.

Hartmuth, K., Urlaub, H., Vomlocher, H.P., Will, C.L., Gentzel, M., Wilm, M.,
and Luhrmann, R. (2002). Protein composition of human prespliceosomes
isolated by a tobramycin affinity-selection method. Proc Natl Acad Sci U S A 99,
16719-16724.

Brow, DA. (2002). Allosteric cascade of spliceosome activation. Annu Rev
Genet 36, 333-360.

Michaud, S., and Reed, R. (1991). An ATP-independent complex commits pre-
mRNA to the mammalian spliceosome assembly pathway. Genes Dev 5, 2534-
2546.

Stevens, S.W., Ryan, D.E., Ge, H.Y., Moore, RE, Young, M.K., Lee, TD, and
Abelson, J. (2002). Composition and firnctional characterization of the yeast
spliceosomal penta-snRNP. Mol Cell 9, 31-44.

Peng, R., Hawkins, 1., Link, A.J., and Patton, J .G. (2006). The splicing factor PSF
is part of a large complex that assembles in the absence of pre-mRNA and
contains all five snRNPs. RNA Biol 3, 69-76.

Malca, H., Shomron, N., and Ast, G. (2003). The U1 snRNP base pairs with the 5'
splice site within a penta-snRNP complex. Mol Cell Biol 23, 3442-3455.

Dagher, S.F., Wang, J .L., and Patterson, R.J. ( 1995). Identification of galectin-3
as a factor in pre-mRNA splicing. Proc Natl Acad Sci U S A 92, 1213-1217.

Vyakarnam, A., Dagher, S.F., Wang, J .L., and Patterson, R.J. (1997). Evidence
for a role for galectin-1 in pre-mRNA splicing. Mol Cell Biol 1 7, 4730-4737.

Wang, W., Park, J .W., Wang, J .L., and Patterson, R.J. (2006).

Immunoprecipitation of spliceosomal RNAs by antisera to galectin-1 and
galectin-3. Nucleic Acids Res 34, 5166-5174.

77

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Agrwal, N., Sun, Q., Wang, S.Y., and Wang, J .L. (1993). Carbohydrate-binding
protein 35. 1. Properties of the recombinant polypeptide and the individuality of
the domains. J Biol Chem 268, 14932-14939.

Ho, M.K., and Springer, TA. (1982). Mac-2, a novel 32,000 Mr mouse
macrophage subpopulation-specific antigen defined by monoclonal antibodies. J
Immunol 128, 1221-1228.

Laemmli, UK. (1970). Cleavage of structural proteins during the assembly of the
head of bacteriophage T4. Nature 22 7, 680-685.

Dignam, J .D., Lebovitz, R.M., and Roeder, KG. (1983). Accurate transcription
initiation by RNA polymerase II in a soluble extract from isolated mammalian
nuclei. Nucleic Acids Res 11, 1475-1489.

Conway, G.C., Krainer, A.R., Spector, D.L., and Roberts, R.J. (1989). Multiple
splicing factors are released from endogenous complexes during in vitro pre-
mRNA splicing. Mol Cell Biol 9, 5273-5280.

Zillmann, M., Zapp, ML, and Berget, SM. (1988). Gel electrophoretic isolation
of splicing complexes containing U1 small nuclear ribonucleoprotein particles.
Mol Cell Biol 8, 814-821.

Park, J .W., Voss, P.G., Grabski, S., Wang, J .L., and Patterson, R.J. (2001).
Association of galectin-1 and galectin-3 with Gemin4 in complexes containing the
SMN protein. Nucleic Acids Res 29, 3595-3602.

Bottner, C.A., Schmidt, H., Vogel, S., Michele, M., and Kaufer, NP. (2005).
Multiple genetic and biochemical interactions of Brr2, Prp8, Prp31, Prpl and Prp4
kinase suggest a function in the control of the activation of spliceosomes in
Schizosaccharomyces pombe. Curr Genet 48, 151-161.

Patton, J .R., Patterson, R.J., and Pederson, T. (1987). Reconstitution of the U1
small nuclear ribonucleoprotein particle. Mol Cell Biol 7, 4030-4037.

Meister, G., Buhler, D., Laggerbauer, B., Zobawa, M., Lottspeich, F., and Fischer,
U. (2000). Characterization of a nuclear 20$ complex containing the survival of

motor neurons (SMN) protein and a specific subset of spliceosomal Sm proteins.
Hum Mol Genet 9, 1977-1986.

Gozani, O., Patton, J .G., and Reed, R. (1994). A novel set of spliceosome-
associated proteins and the essential splicing factor PSF bind stably to pre-mRNA
prior to catalytic step II of the splicing reaction. Embo J 13, 3356-3367.

Behrens, S.E., Tyc, K., Kastner, B., Reichelt, J ., and Luhrmann, R. (1993). Small
nuclear ribonucleoprotein (RNP) U2 contains numerous additional proteins and

78

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

has a bipartite RNP structure under splicing conditions. Mol Cell Biol 13, 307-
319.

Behrens, SE, and Luhrmann, R. (1991). Irnmunoaffinity purification of a
[U4/U6.U5] tri-snRNP from human cells. Genes Dev 5, 1439-1452.

Makarov, E.M., Makarova, O.V., Urlaub, H., Gentzel, M., Will, C.L., Wilm, M.,
and Luhrmann, R. (2002). Small nuclear ribonucleoprotein remodeling during
catalytic activation of the spliceosome. Science 298, 2205-2208.

Grabowski, P.J., Seiler, SR, and Sharp, RA. (1985). A multicomponent complex
is involved in the splicing of messenger RNA precursors. Cell 42, 345-353.

Bindereif, A., and Green, MR. (1987). An ordered pathway of snRNP binding
during mammalian pre-mRNA splicing complex assembly. Embo J 6, 2415-2424.

Mount, S.M., Pettersson, I., Hinterberger, M., Karmas, A., and Steitz, J .A. (1983).
The U1 small nuclear RNA-protein complex selectively binds a 5' splice site in
vitro. Cell 33, 509-518.

Zillmann, M., Rose, SD, and Berget, SM. (1987). U1 small nuclear
ribonucleoproteins are required early during spliceosome assembly. Mol Cell Biol
7, 2877-2883.

Reed, R. (1990). Protein composition of mammalian spliceosomes assembled in
vitro. Proc Natl Acad Sci U S A 87, 8031-8035.

Hinterberger, M., Pettersson, I., and Steitz, J .A. (1983). Isolation of small nuclear
ribonucleoproteins containing U1, U2, U4, U5, and U6 RNAs. J Biol Chem 258,
2604-2613.

Pettersson, I., Hinterberger, M., Mimori, T., Gottlieb, E., and Steitz, J .A. ( 1984).
The structure of mammalian small nuclear ribonucleoproteins. Identification of

multiple protein components reactive with anti-(U1)ribonuc1eoprotein and anti-
Sm autoantibodies. J Biol Chem 259, 5907-5914.

Bringrnann, P., and Luhrmann, R. (1986). Purification of the individual snRNPs

U1 , U2, U5 and U4/U 6 from HeLa cells and characterization of their protein
constituents. Embo J 5, 3509-3516.

79

Chapter 3

Identification of galectin-3 in aptamer-selected early splicing complexes

ABSTRACT

We have shown that galectin-1 and galectin-3 are associated with the spliceosome
from early complex formation (loading of U1 snRNP) through active complexes by
immunoprecipitations using galectin-specific antisera. Here we use a galectin-
independent method to document an association of galectin-3 with the early spliceosome
(B complex). Using a pre-mRNA engineered to contain three recognition sites for the
MS2 bacteriophage protein, assembled splicing complexes can be isolated and analyzed.
We show that our isolation system selected only a single pro-mRNA containing our
aptamer and that bonafide splicing factors could recognize and assemble on this pre-
mRN A. Furthermore, the protein constituents assembled on the pre-mRNA varied, as
expected, with incubation conditions to allow the assembly of unique splicing complexes.
Characterization of selected complexes showed that isolated E complexes contained U1
snRN A, but not U6 snRNA. Subsequent analysis of proteins showed galectin-3
associated with the E complex. These results verify the association of galectin-3 with the
spliceosome using an isolation method independent of galectin selection.
INTRODUCTION

Splicing cofactors containing the U snRNAs (snRNPs) exist either as single
entities (the U1 and U2 snRNPs) or as larger complexes, such as with 3 snRNAs (the
U4/U6.U5 tri-snRNP) [1, 2]. These snRNPs assemble onto a pre-mRNA scaffold to
carry out pre-mRNA splicing. During spliceosome assembly, the first snRNP to bind the

pre-mRNA is the U1 snRNP [3, 4]. This U1 binding event is the marker for early

80

complex formation (E-complex) and can be detected by gel mobility shifts [5, 6]. The
binding of U1 to pre-mRNA is ATP independent and allows the step-wise assembly of
higher-order splicing complexes (A/B/C) with the addition of ATP and incubation [7].
Each of these active complexes can be distinguished based on gel mobility shifts and on
protein and RNA constituents.

We recently have identified galectin-3 (Gal3) as a snRNP-associated protein in
HeLa nuclear extract (NE) (see Chapter 2). Immunoprecipitation of NE incubated in the
absence of a splicing substrate with antiserum specific for Ga13 revealed the 5 splicing
snRNAs along with several snRNA core proteins. Using glycerol gradient sedimentation
to separate the endogenous nuclear snRNPs, we could detect a Ga13-containing mono-
snRNP (U1) at ~10S. Using an in vitro binding assay, we showed that the Gal3-
containing U1 snRNP formed a Gal3-immunoprecipitable complex on an exogenous pre-
mRN A. Pretreatment of the Ga13-containing U1 snRNP with micrococcal nuclease
abolished pre-mRNA binding activity. This complex was reminiscent of early splicing
complexes that incorporated Ga13 (or galectin-1 (Gall)) when splicing substrate is
incubated with pre-mRNA in a complete NE [8]. Both types of early complexes relied on
galectin-specific antiserum in co-immunoprecipitation protocols.

To confirm these data, a method independent of galectin selection was explored to
show that Ga13 was indeed associated with U1 snRNP and pre-mRNA in an early
splicing complex. Methods to select nucleic acids and associated proteins fi‘om complex
mixtures have been evaluated. One powerful approach is to engineer an aptamer
sequence into the nucleic acid of interest and subsequently select the aptamer-containing

nucleic acid via interaction with its specific ligand. To purify specific RNP complexes

81

involved in splicing, Zhou et al. [9] developed the selection system schematically
illustrated in Figure 1. First, a pre-mRNA substrate derived from the adenovirus major
late gene (AdML-M3) was engineered such that the 3'-end contained three hairpin loops
that bind to the bacteriophage MS2 protein with high affinity. In parallel, a fusion
protein containing MS2 and maltose-binding protein (MBP) was expressed and purified
on the basis of its binding to amylose beads and specific elution using soluble maltose.
The purified MSZ-MBP fusion protein, bound to fresh amylose beads, is then used to
select for RNP complexes formed by incubation of the AdML-M3 pre-mRNA substrate
with NEs under various conditions. The isolated RNP complexes can be similarly
released from the beads using soluble maltose. Using this (or similar) aptamer selection
protocol, other groups have isolated and purified various splicing complexes which were
subjected to protein identification by mass spectrometry [10-13]. These latter studies
have revealed that the spliceosome proteome consists of more than 300 proteins.

Using this MS2 aptamer selection protocol, we now show that early splicing
complexes assembled in a complete NE contain Ga13 in addition to the Sm core
polypeptide Sm B/B’ and Sm D and the U1 snRNA.

MATERIALS AND METHODS
Antibodies

Polyclonal rabbit antibodies directed against Ga13 [14] were described previously.
The following antibodies were purchased for immunoblotting: rabbit anti-hnRNP C1/C2,
goat anti-Slu7 (Santa Cruz Biotechnology); rabbit anti-MBP (Chemicon Intenrational);
monoclonal mouse anti-SMN (BD Transduction Lab) and human autoimmune sera ENA

anti-Sm (The Binding Site). Rabbit anti-Ga13, described previously [14], was used for

82

Figure 1. Schematic diagram illustrating the structure of AdML-M3 pre-mRNA and
its use in purifying spliceosomal components assembled on the RNA. (A) The
AdML-M3 pre-mRN A consists of two exons (rectangles) joined by a single intron
(dark .line) and three hairpin structures recognized by the MS2 protein. The M82
recognition hairpins are situated at the 3’ end of exon 2. (B) The MBP-MS2 fusion
protein can bind to amylose beads via the maltose binding protein (MBP) and to
AdML-M3 through MS2 recognition of the hairpins. Components assembled on the
AdML-M3 pre-mRNA can thus be affinity selected on the amylose beads and
subsequently eluted with soluble maltose.

83

 

A A A A
C-G C-G C-G

Ac-o Ac-s AC-G
c-a c-e c-c
A-u A-u A-u
u-A u-A u-A

a
s-ccceAucccAUAucc'GAscurtccccrtuaec 'GACUAGUAGAUCIP'GGAAthnUAGAccs-

\Wfil/

 

 

 

 

AdML-M3 pre-mRNA

B. (53+EO

bind MBP-MSZ
fusion protein to
amylase bead
0mm 0 (.230
/ .
’AdML-MZB O
incubate loaded
beads with extract

 

l elute with maltose

M

84

blotting Ga13 protein. Primary mouse monoclonal antibodies were detected by goat anti-
mouse IgG light chain specific-HRP conjugates (Jackson ImmunoResearch
Laboratories). All other secondary antibody-HRP conjugates (Pierce Biotechnology)
were directed against both the heavy and light chains of the primary blotting antibodies.
Plasmids and in vitro transcription

Plasmids encoding AdML-M3 pre-mRNA and MBP-MSZ protein were kindly
provided by Dr. Robin Reed (Harvard University) [9, 12]. Plasmids containing anti-
sense constructs for U1 and U6 snRNA for use as northern blot probes were gifts from
Dr. Jeff Patton (University of South Carolina). The plasmid containing MINX pre-
mRNA [6] and in vitro transcription reactions have been described previously [15]. For
32P-labeled AdML-M3, the plasmid was digested with Xba I and in vitro transcribed
using T7 RNA polymerase. For anti—sense probes against AdML-M3, the plasmid was
digested with Cla I and in vitro transcribed using SP6 RNA polymerase. For U1 and U6
anti-sense probes, plasmids were digested with Nae I and in vitro transcribed with 32P-
GTP using SP6 RNA polymerase for U6 anti-sense and T7 RNA polymerase for U1 anti-
sense. For large quantities of non-radioactive AdML-M3, transcription using
AmpliScribe T7 High Yield Transcription kit was used (Epicentre). Amylose-conjugated
agarose beads were purchased from New England Biolabs and the MBP-MSZ firsion
protein was isolated according to the manufacturer's protocol and dialyzed into 60%
buffer D (buffer D is 20 mM HEPES, pH 7.9, 100 mM KCl, 0.2 mM EDTA, 0.5 mM
DTT, 20% glycerol).

Polyacrylamide gel electrophoresis, immunoblotting and northern blotting

85

For protein analysis, samples were subjected to 12% SDS-PAGE as described by
Laemmli [16]. Proteins were electrophoretically transferred from the gel onto Hybond
nitrocellulose membrane (Amersham Biosciences) or Biotrace PVDF membrane (Pall) in
transfer buffer (25 mM Tris, 193 mM glycine and 20% methanol, pH 8.3). Following
transfer, membranes were blocked ovemight in 10% nonfat dry milk in Tris-buffered
saline containing Tween-20 (10 mM Tris, pH 7.5, 0.5 M NaCl, 0.05% Tween 20, T-
TBS). Primary antibodies for immunoblotting were diluted in 1% milk-T-TBS and
incubated on the membrane for 1 hour at room temperature. After washing four times, 15
minutes each, in T-TBS, the appropriate secondary antibody conjugated to horseradish
peroxidase was added in 1% milk—T-TBS for 1 hour. Following four T-TBS washes as
above, proteins were visualized using the Western Lightning Chemiluminescence System
(Perkin Elmer Life Sciences).

For RNA samples, the RNA was extracted as described below and precipitated
with 3 volumes of ethanol at -80°C. The precipitated RNA was dissolved in 10 p1 of
sample buffer (9:1/formamidezbromophenol blue) and subjected to electrophoresis
through 13% polyacrylamide (bisacrylamide —acrylamide 1.9:50 [wt/wt])-8.3 M urea
gels, run in 1X TBE (90 mM Tris base, 90 mM boric acid and 2.5 mM EDTA, pH 8.0).
The radioactive RNA species were revealed by autoradiography.

For Northern analysis the RNA was transferred via wicking in 2X SSC (0.3 M
NaCl, 0.03 M sodium citrate) ovemight onto a nylon membrane (Hybond-N, Amersham
Biosciences) and cross-linked by exposure to UV. After cross-linking, the nylon
membrane was pre-hybridized in 20 ml of hybridization solution containing 20%

deionized fonnamide, 3X SSC, 5X Denhardt's, 50 mM NazHP04/NaHzPO4, pH 6.8, 0.1%

86

SDS and 0.1 mg/ml hening sperm DNA for about 4 hours at 42°C. After pre-
hybridization, 32P-labeled anti-sense probes were added to the hybridization solution and
the membrane was hybridized overnight (~16 hours) at 42°C . The membrane was then
washed once with 30 ml 2X SSC + 0.1% SDS at 50°C for 30 minutes. Northern
hybridization was imaged by phosphorimage analysis (Molecular Dynamics).
Selection of AdML-M3

Nuclear extract (NE) from HeLa S3 cells (National Cell Culture Center,
Minneapolis, MN) was prepared as described in Dignam et al [17]. To test the selectivity
of the MS2 system, 10 pg of MBP-MS2 were loaded onto 20 pl of amylose beads in 0.5
ml binding buffer (20 mM HEPES, pH 7 .9, 60 mM NaCl, 1 mM DTT) at 4°C for 1 hour.
Then 0.8 pmol of in vitro transcribed MINX and AdML-M3 pre-mRNA were mixed and
incubated in an 80 p1 reaction in binding buffer with 10 pl HeLa NE for 20 minutes at
4°C and incubated with the loaded beads. The beads were washed three times with 0.5
ml binding buffer containing 0.05% Triton-X 100 (TX) and eluted with 20 mM maltose
in binding buffer. Eluted material was split into aliquots and analyzed for protein and
RNA, as described above. RNA was extracted by adding proteinase K (4 mg/ml final
concentration) to the sample and incubating at 37°C for 30 minutes, then diluting to 100
pl with 125 mM Tris (pH 8), 1 mM EDTA, 300 mM sodium acetate. RNA was extracted
by mixing with 200 pl of phenol—chloroform (50:50 [vol/vol]), followed by 100 pl of
chloroform.
Pre-mRNA complex isolation

For early and active complex selection, 1 pg of in vitro transcribed AdML-M3

was incubated with roughly 5 pg of MBP-MSZ and 20 pl of amylose beads for one hour

87

at 4°C with rocking. The beads were washed once with binding buffer. A 150 pl
reaction consisting of 90 pl HeLa NE, 3 mM MgC12, 40 U RNasin (Promega) was
incubated with the AdML-M3 -loaded amylose beads at 30°C for 30 minutes to form B
complex. Similarly, active complexes were formed with an identical reaction, also
containing 0.5 mM ATP and 20 mM creatine phosphate. Active complexes were formed
by incubating the reaction with the AdML-M3 -1oaded amylose beads for 30 minutes at
30°C. Control beads had the same amount of MBP-MSZ loaded onto the amylose beads
and were incubated with NE under active complex conditions. After incubation, the
volume was brought up to 0.5 ml with binding buffer and mixed for one hour at 4°C.
After binding the complexes, beads were washed 3 times in 1 ml of binding buffer
containing 0.05% TX and eluted with 20 mM maltose in binding buffer.
E complex analysis

For analysis of the E complex, 1 pg of in vitro transcribed AdML-M3 was
incubated with roughly 4 pg of MBP-MSZ in a total of 20 pl of binding buffer on ice for
30 minutes. After incubation, the protein and AdML-M3 were added to a 150 pl reaction

containing 100 pl NE, 3 mM MgC12 and 80 U RNasin and incubated for 20 minutes at
30°C. The whole reaction was then incubated with 30 pl of amylose beads for one hour
at 4°C with rocking, with the total volume being brought up to 0.5 ml with binding
buffer. A control reaction lacking only the AdML-M3 pre-mRNA was performed. After
binding the beads were washed 3 times with 1 ml of binding buffer + 0.05% TX and
eluted with 20 mM maltose in binding buffer.

RESULTS

Selection of RNA species through MS2 recognition of its cognate RNA hairpin

88

We have adopted the AdML-M3 selection system as developed by Zhou et al. and
outlined in Figure 1 [9]. We tested this selection system by comparing AdML-M3
against MINX, another derivative of the adenovirus major late gene but lacking the M82
binding sites. The two RNAs, both in vitro transcribed and 32P-labeled, were added in an
equimolar ratio (Figure 2, lane 1) to HeLa nuclear extract, incubated and passed over
amylose beads loaded with MBP-M82 fusion protein. After washing, the bound fraction
was eluted and subjected to gel electrophoresis. The bound fraction (Figure 2, lane 2)
showed a strong selection for AdML-M3 over MINX pre-mRNA. There was nearly a
15-fold enrichment for the AdML-M3 substrate over the starting ratio of the mRNAs in
the input. These results indicate that our experimental set-up, including the firsion
protein and AdML-M3 transcript, is functional and that this M82 selection allows a high
degree of specificity in RNA purification.

Isolation of splicing complexes assembled on AdML—M3

To test if we could isolate distinct splicing complexes, AdML-M3 was incubated
with HeLa NE, with and without added ATP, in order to form two unique splicing
complexes: the ATP-independent early (E-) complex and catalytically active
spliceosomes that require ATP [18]. Splicing complexes containing AdML-M3 were
selected by passing the incubated extract over amylose beads loaded with MBP-M82
fusion protein. A control reaction lacking the AdML-M3 substrate was carried along
under conditions to form active splicing complexes. The bound and eluted material was
western blotted for the 8m proteins to confirm the selection of splicing complexes
(Figure 3A, lanes 2 and 3). Under both incubation conditions, early and active, we found

the association of Sm proteins with AdML-M3, suggesting that snRNPs had been isolated

89

Figure 2. Specificity of selection by MBP-MS2 fusion protein. Two different 32P-
labeled pre-mRNAs, AdML-M3, containing three M82 recognition sites and
MINX, lacking any M82 recognition sites, were incubated together with NE and
passed over amylose beads loaded with MBP-MS2. The bound material was eluted
with maltose. Input (lane 1) shows the starting amount of AdML-M3 and MINX in
the reaction. Material eluted off the MBP-MS2 column is shown as bound (lane 2).

90

28m .

39:

 

m.
L
m
A

91

with the AdML-M3 substrate. To test whether we could distinguish between early and
active complexes we probed for Slu7, a required second-step splicing factor that has only
been found to associate with catalytically active spliceosomes [10, 19]. The AdML-M3
incubated under active splicing conditions did indeed pull out Slu7 (Figure 3A, lane 3).
However, conditions that only allow B complex formation did not assemble Slu7 onto the
AdML-M3 substrate (Figure 3A, lane 2). These results confirm that we can distinguish
between active and early splicing complexes by AdML-M3 and that incubation
conditions can control the assembly of splicing complexes.

Blotting for both Slu7 and 8m proteins failed to reveal any of these proteins in the
eluted material from the control column (Figure 3A, lane 4). A final control was
performed by blotting for the SMN protein. SMN is a nuclear protein that interacts with
a variety of RNPs but has failed to be identified as directly interacting with the
spliceosome. Immunoblotting for the SMN protein showed that it is not selected by
AdML-M3 under any condition despite there being plenty of SMN present in the NE
(data not shown). Blotting with antibodies against MBP confirm an equal amount of
fusion protein in all samples.

Because U1 snRNA binding to the 5' splice site of a pre-mRNA is the hallmark of
early complex formation, we examined the eluted material for the presence of snRNA.
Northern blots revealed the presence of U1 snRNA in both the AdML-M3 assembled E
complex and active complex, but U6 snRNA only in the active complex elution (Figure
3B). The U6 snRNP is necessary for a catalytically active spliceosome and its

incorporation onto the spliceosome is ATP-dependent. Neither U1 nor U6 could be

92

Figure 3. Analysis of protein and RNA components of early and active splicing
complexes selected via the AdML-M3 pre-mRN A. Amylose beads, loaded with
AdML-M3 pre-mRN A and MBP-M82 protein, were mixed with NE diluted to
splicing conditions and incubated to form early or active splicing complexes. A
control column lacking AdML was incubated under conditions that give rise to active
complexes. The beads were washed and eluted with maltose. Panel A shows western
blots of the bound material from the early complex (lane 2), active complex (lane 3)
and control column (lane 4) for the proteins indicated at right. Panel B shows a
northern blot of the eluted material from early complex (lane 1), active complexes
(lane 2) and control column (lane 3) using anti-sense RNA probes against AdML-M3,
U1 and U6 snRNAs.

93

35:00
5388 o>uo<

onEoo m

m2

Slu7

IV:-‘

M.

8m B/B’

 

D
m
S

 

Hill's and: ” MBP-M82

35:00
5388 o>flo<

xoafico m

m
M
A

 

U6

94

detected eluted from control beads (Figure 3B, lane 3). All of these results indicate the
selection of complexes by AdML-M3 to be specific and that the association of AdML-
M3 complexes can be controlled by incubation conditions. In particular, selection of
complexes under E conditions allow the selection of U1 snRN A, an important early
complex marker, but not Slu7 protein or U6 snRNA, both components of active splicing
complexes.
E-complexes assembled on AdML-M3 contain Ga13

Based on our previous data showing the association of Ga13 with U1 snRNP (see
Chapter 2), our hypothesis was that Ga13 is a bonafide component of early splicing
complexes, in particular the B complex. To test this hypothesis, we performed AdML-
M3 selection of E complexes and the bound material was blotted for known B complex
factors. The Sm proteins along with hnRNP C1/C2 were detected in our AdML-M3
selected material but not under our control conditions (Figure 4, lanes 2 and 3), allowing
us to conclude we had successfully pulled out an B complex. We then blotted for the
Ga13 polypeptide, which revealed the protein in our B complex specific pull-out but not
in the control experiment (Figure. 4, lanes 2 and 3). As a final control, blotting with
antibodies against MBP showed an equal amount of firsion protein loaded in both the
experimental and control conditions. The results of this experiment confirmed our
hypothesis that Ga13 is a protein component of a splicing E complex.
DISCUSSION

The high affinity binding (Kd ~3 nM) of the coat protein of bacteriophage R17 to
a hairpin loop on its genomic RNA has been extensively studied by Uhlenbeck and co-

workers [20, 21]. Bardwell and Wickens took advantage of this system in developing

95

Figure 4. Analysis of proteins associated with early splicing complexes.
AdML-M3 pre-mRNA was bound to MBP-M82 then added to a splicing
reaction and incubated under conditions to give early complexes. The
reaction was then passed over amylose beads and the bound material eluted
with maltose. A reaction lacking AdML-M3 pre-mRN A was used as a
control. Bound material from both the early complex (lane 2) and control
reaction (lane 3) was western blotted for the indicated proteins at right.

96

 

X
2
E“ E
a 8 5
LIJ U
i ' hnRNPCl/C2
”'" . Ga13
- .- Sm B/B’
r- 4... SI‘HD
W MBP-M82
1 2 3

97

a purification scheme for RNA and RNA complexes but noted that the binding of an
RNA containing non-R17 sequences required two recognition sites in tandem [22]. The
system was further refined in the AdML-M3 pre-mRNA construct to contain three
hairpin loops and in the use of MBP-M82 firsion protein and amylose beads for the
selection of assembled spliceosomal complexes that were subsequently visualized by
electron microscopy [9] and analyzed by proteomics [10, 12]. For these latter studies on
spliceosome assembly as well as for our present work, it is of particular importance to
note that neither the presence of the hairpin loops nor the binding of the fusion protein
affected either splicing complex formation or the splicing reaction [9, 23].

Our present application of this aptamer selection system has provided key
confirmatory evidence to two previous experiments. First, we showed that either anti-
Gall or anti-Ga13 can immunoprecipitate 32P-labeled MINX pre-mRNA under conditions
(incubation with NE at 30°C in the absence of ATP) that would lead to the formation of
only early complexes [8]. Second, we have shown that an isolated Ga13-U1 snRNP
complex is sufficient to load Ga13 onto a pre-mRNA under conditions that allow the
formation of early splicing complexes (see Chapter 2) and basepairing of U1 at the 5'
splice site [24, 25]. Using aptamer recognition to select spliceosomal complexes
assembled on the AdML-M3 pro-mRNA under the same conditions, we have now
documented the detection of Ga13 in these early spliceosomal complexes containing the
U1 snRNA.

These results implicate that Ga13 enters the splicing reaction early in the
spliceosome assembly process. This notion is consistent with three additional lines of

evidence. First, we had reported that NEs depleted of the galectins by affinity adsorption

98

on lactose-agarose beads failed to form active spliceosomal complexes and gel mobility
shift assays of 32P-labeled pre-mRNA revealed only bands migrating in the H-/E-complex
region [26]. The activities of the galectin-depleted extract, in forming active splicing
complexes and in performing the in vitro splicing reaction, were reconstituted by the
addition of recombinant Ga13 with similar dose-response curves.

Second, we have expressed and purified a fragment of the murine Ga13
polypeptide containing residues 1-137, corresponding to the arnino-tenninal domain
(ND) bearing multiple repeats of the nine-residue motif, PGAYPGXXX [14]. When the
splicing assay was carried out in the presence exogenously added Ga13 ND, we observed
a dose-dependent inhibition of product formation [15]. This apparent dominant negative
effect of the ND was associated with the arrest of spliceosome assembly at the H-/E-
complex. In both the splicing reaction and in the assembly of active spliceosomes,
parallel additions of either the full-length Ga13 polypeptide or the carboxyl terminal
carbohydrate recognition domain failed to yield the same effect. Finally, we have
recently found that addition of the NCL-GAL3 monoclonal antibody directed against
Ga13 to a splicing competent NE inhibited the splicing reaction [27]. Again, native gel
electrophoresis showed that NCL-GAL3 exerted its effect early in the spliceosome
assembly process, blocking the progression of H-lE-complexes into active spliceosomes.

A key question posed by our studies is whether there is any difference between
two complexes formed on the pre-mRNA substrate: (a) the classical E-complex
assembled upon incubation of NE with pre-mRN A at 30°C in the absence of ATP and
selected by the M82 aptamer; and (b) the complex resulting from the binding of Ga13-U1

snRNP (in fractions 3 and 4 derived from glycerol gradient fractionation of NE; see

99

Chapter 2) to the pre-mRNA selected by antibodies against Gal3. The aptamer selection
procedure documented in the present studies, coupled with additional gel filtration (size
fractionation) steps, could provide the requisite extent of purification such that the two

complexes can be compared in terms of protein composition by proteomic analysis.

100

REFERENCES

1.

10.

11.

Behrens, SE, and Luhrmann, R. (1991). Irnmunoaffinity purification of a
[U4/U6.U5] tri-snRNP from human cells. Genes Dev 5, 1439-1452.

Bringrnann, P., and Luhrmann, R. (1986). Purification of the individual snRNPs
U1, U2, US and U4/U6 from HeLa cells and characterization of their protein
constituents. EMBO J 5, 3509-3516.

Zillmann, M., Rose, SD, and Berget, 8M. (1987). U1 small nuclear
ribonucleoproteins are required early during spliceosome assembly. Mol Cell Biol
7, 2877—2883.

Abmayr, S.M., Reed, R., and Maniatis, T. (1988). Identification of a functional
mammalian spliceosome containing unspliced pre-mRNA. Proc Natl Acad Sci U
S A 85, 7216-7220.

Ruby, S.W., and Abelson, J. ( 1988). An early hierarchic role of U1 small nuclear
ribonucleoprotein in spliceosome assembly. Science 242, 1028-1035.

Zillmann, M., Zapp, ML, and Berget, SM. (1988). Gel electrophoretic isolation
of splicing complexes containing U1 small nuclear ribonucleoprotein particles.
Mol Cell Biol 8, 814-821.

Bindereif, A., and Green, MR. (1987). An ordered pathway of snRNP binding
during mammalian pre-mRNA splicing complex assembly. EMBO J 6, 2415-
2424.

Wang, W., Park, J.W., Wang, J .L., and Patterson, R.J. (2006).
Immunoprecipitation of spliceosomal RNAs by antisera to galectin-1 and
galectin-3. Nucleic Acids Res 34, 5166-5174.

Zhou, 2., Sim, J ., Griffith, J ., and Reed, R. (2002). Purification and electron
microscopic visualization of functional human spliceosomes. Proc Natl Acad Sci
U S A 99, 12203-12207.

Jurica, M.8., Licklider, L.J., Gygi, S.R., Grigorieff, N., and Moore, M.J. (2002).
Purification and characterization of native spliceosomes suitable for three-
dimensional structural analysis. RNA 8, 426-439.

Hartmuth, K., Urlaub, H., Vomlocher, H.P., Will, C.L., Gentzel, M., Wilm, M.,
and Luhrmann, R. (2002). Protein composition of human prespliceosomes
isolated by a tobramycin affinity-selection method. Proc Natl Acad Sci U S A 99,
16719-16724.

101

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Zhou, Z., Licklider, L.J., Gygi, SP, and Reed, R. (2002). Comprehensive
proteomic analysis of the human spliceosome. Nature 419, 182-185.

Rappsilber, J ., Ryder, U., Lamond, AL, and Mann, M. (2002). Large-scale
proteomic analysis of the human spliceosome. Genome Res 12, 1231-1245.

Agrwal, N., Sun, Q., Wang, S.Y., and Wang, J .L. (1993). Carbohydrate-binding
protein 35. I. Properties of the recombinant polypeptide and the individuality of
the domains. J Biol Chem 268, 14932-14939.

Park, J .W., Voss, P.G., Grabski, 8., Wang, J .L., and Patterson, R.J. (2001).
Association of galectin-1 and galectin-3 with Gemin4 in complexes containing the
SMN protein. Nucleic Acids Res 29, 3595-3602.

Laemmli, UK. (1970). Cleavage of structural proteins during the assembly of the
head of bacteriophage T4. Nature 22 7, 680-685.

Dignam, J .D., Lebovitz, R.M., and Roeder, R.G. (1983). Accurate transcription
initiation by RNA polymerase II in a soluble extract from isolated mammalian
nuclei. Nucleic Acids Res 11, 1475-1489.

Brow, D.A. (2002). Allosteric cascade of spliceosome activation. Annu Rev
Genet 36, 333-360.

Chua, K., and Reed, R. (1999). Human step 11 splicing factor hSlu7 functions in
restructuring the spliceosome between the catalytic steps of splicing. Genes Dev
13, 841-850.

Romarriuk, P.J., Lowary, P., Wu, H.N., Stormo, G., and Uhlenbeck, DC. (1987).
RNA binding site of R17 coat protein. Biochemistry 26, 1563-1568.

Carey, J ., Cameron, V., dc Haseth, PL, and Uhlenbeck, CC. (1983). Sequence-
specific interaction of R1 7 coat protein with its ribonucleic acid binding site.
Biochemistry 22, 2601 -261 0.

Bardwell, V.J., and Wickens, M. (1990). Purification of RNA and RNA-protein
complexes by an R17 coat protein affinity method. Nucleic Acids Res 18, 6587-
6594.

Das, R., Zhou, Z., and Reed, R. (2000). Functional association of U2 snRNP with
the ATP-independent spliceosomal complex B. Mol Cell 5, 779-787.

Mount, S.M., Pettersson, I., Hinterberger, M., Karmas, A., and Steitz, J .A. (1983).

The U1 small nuclear RNA-protein complex selectively binds a 5' splice site in
vitro. Cell 33, 509-518.

102

25.

26.

27.

Michaud, 8., and Reed, R. (1991). An ATP-independent complex commits pre-
mRN A to the mammalian spliceosome assembly pathway. Genes Dev 5, 2534-
2546.

Dagher, S.F., Wang, J .L., and Patterson, R.J. (1995). Identification of galectin-3
as a factor in pre-mRNA splicing. Proc Natl Acad Sci U S A 92, 1213-1217.

Gray, R.M., Davis, M.D., Ruby, K.M., Voss, P.G., Patterson, R.J., and Wang, J .L.
(2007). Galectin Domains and pre-mRNA Splicing. II. Epitope for the Mac-2
Monoclonal Antibody and Its Functional Significance in the Splicing Activity of
Galectin-3. RNA Biol submitted for publication.

103

Chapter 4
Concluding statements

Previous work in our laboratory had focused on galectin-1 (Gall) and galectin-3
(Gal3) as essential pre-mRNA splicing factors. In addition, recent work has shown both
Gall and Ga13 to be assembled on the spliceosome.

My work presented here moves our understanding of nuclear galectins forward in
several significant ways: first, Ga13 is a member of numerous nuclear complexes
containing both proteins and RNA; second, nuclear Ga13 interacts with snRNPs outside
of the spliceosome; third, Ga13 interaction with snRNPs leads to the entry of Ga13 into
the pre-mRNA splicing pathway. These new data fit well with the existing evidence of
galectins as pre-mRNA splicing factors. However, more experiments must be performed
to fully understand the role of galectins in snRNP biogenesis and their firnction in pre-
mRN A splicing.

Firstly, it is of great interest to determine the binding partner of Ga13 on the
snRNPs. Does Ga13 interact directly with the 8m core polypeptides, a snRNP-specific
protein or does another non-snRNP protein mediate this interaction? A series of in vitro
binding assays may reveal the binding partner of galectins. One likely protein candidate
to mediate the snRNP interaction may be the general transcription factor TFII-I, as recent
work shows this protein is associated (i) with Ga13 in a GST-binding assay (TFII-I can
also interact with Gall) and (ii) with the snRNP-containing complex, PCC. This
interaction must be tested to see if the Gal3-TFII-I interaction is direct and if so, whether
TFII-I can bind snRNPs directly.

Another interesting finding is the association of Ga13 with snRNPs in large

complexes. What exactly are these large multi-snRNP complexes? Are they a

104

mammalian penta-snRNP complex? These complexes should be tested for functional
pre-mRNA binding and splicing activity. Gradient fractions containing this penta-snRNP
may be sufficient to compliment an extract depleted of firnctional snRNPs in an in vitro
splicing assay. The identity of several splicing factors with the yeast penta-snRNP and
mammalian PCC points to the idea that these large complexes are involved in
spliceosome formation at some level.

The presence of the SMN protein in these large complexes is intriguing in what it
suggests about the assembly of these complexes. Is the nuclear function of the SMN
protein to assemble pre-fonned spliceosomes? Or is SMN in these complexes due to its
association with another snRNP related protein, such as coilin? It would be interesting to
test whether these large multi-snRNP complexes are involved in performing the snRNA
modifications that are carried out at the Cajal body. Methods of assessing this include
searching the snRNAs for modified bases, blotting for the presence of small Cajal body
RNAs (scaRNAs) and immunoblotting for coilin, the protein marker of Caj a1 bodies and
other known snRNA modifying enzymes.

A mechanism for galectin entry into the splicing pathway is a meaningful step
forward to determine the role of galectins in splicing. Although numerous lines of
evidence show galectin-inhibited splicing reactions halt at an early complex stage, the
exact mechanism for this inhibition is not known. Armed with these new data and the
binding assay demonstrated in this thesis, perhaps the association of galectins and/or Ul
loading onto a pre-mRN A in an inhibited nuclear extract and/or gradient fraction can be

assayed. Could this be the step at which pre-mRN A splicing is blocked?

105

 

Another important question to ask is how the Gal3-U1 snRNP complex relates to
the early splicing complex (B) assembled on the pre-mRNA. Proteomic analysis using
mass spectrometry may allow the identification of unique factors in both complexes. The
Ga13-U1 snRNP complex also should be tested for functional assembly of active
spliceosomes. This may be assayed by either gel mobility shift assays or
complementation assays of U1 depleted extracts.

Recent work has shown that Gall and Ga13 are mutually exclusive in
spliceosomes. That is, a spliceosome may contain either Gall or Gal3, but not both
galectins simultaneously. A logical extension of this mutually exclusive finding is that
the loading of the galectins onto the spliceosome should show this exclusivity. _
Therefore, I believe it is important to document which snRNPs Gall is associated with.
If Gall and Ga13 show similar snRNP binding and a similar Gall-U1 snRNP is found, it
too should be tested for pre-mRNA binding and the presence of Gal3. My hypothesis is
that the exclusivity seen in the spliceosome is actually conferred at the galectin binding of
the snRNP before entry into the spliceosome. For this to be the case, a snRNP or snRNP
complex would have to contain only a single galectin binding site, recognizable by both
Gall and G313.

The foundation for numerous and interesting studies further investigating these
questions has been laid by the data presented in this thesis. In the future, the role of

nuclear galectins in snRNP and spliceosome assembly can be elucidated.

106

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

1|llflillilllllfllllll][][]]lll][[l]ll