town->5. . .h

 

zoo.

PHD

 

! DANISH!

P

11
‘1

 

 

 

 

, 33$?

"3

59"

.‘I
:%

 

#3153

m. L

“tun-‘1: . , ‘
w... ... .g.‘

3

 

LIBRARY

2 ’ Michigan State
2m: University
,9/3 39 *I 5

This is to certify that the
dissertation entitled

SHUTTLING OF GALECTlN-3 BETWEEN THE NUCLEUS
AND CYTOPLASM

presented by
PETER JOSEPH DAVIDSON

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Cell and Molecular Biology

9r“ ”bx

LMajor Professor’s Sidnature

 

09/24/04

Date

MSU is an Affirmative Action/Equal Opportunity institution

 

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.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 c:/CIRC/DateDue.indd-p.15

——_-__—‘_— .____ - ,_.._.....___..

SHUTTLING OF GALECTIN-3 BETWEEN THE NUCLEUS
AND CYTOPLASM

By

Peter Joseph Davidson

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Cell and Molecular Biology Program

2004

ABSTRACT

SHUTTLING OF GALECTIN-3 BETWEEN THE NUCLEUS AND CYTOPLASM
By

Peter Joseph Davidson

Galectin-3 (Ga13, Mr ~ 30 kDa) is a B—galactoside-specific lectin and a pre-
mRNA splicing factor. To test if 6313 might shuttle between the nucleus and cytoplasm,
human fibroblasts (LG-l) were fused with mouse fibroblasts (3T3). The antibody NCL-
GAL3, which specifically recognizes human Gal3, was used to monitor human Gal3
localization in heterodikaryons. Human Ga13 localized to both nuclei of a large
percentage of heterodikaryons. Addition of leptomycin B decreased the percentage of
heterodikaryons showing human Ga13 in both nuclei. In parallel, mouse 3T3 fibroblasts,
expressing Ga13 were fused with fibroblasts from a Ga13-null mouse. The results from
both assays suggested that galectin-3 can shuttle from one nucleus into another.

We engineered a vector that expressed a fusion protein containing Green
Fluorescent Protein (GFP); bacterial maltose-binding protein (MalE); and Gal3. Analysis
of fluorescence in mouse 3T3 fibroblasts transfected with this construct showed that the
GFP-MalE-Gal3(l-263) fusion protein localized predominantly in the nucleus.
Carboxyl-terminal truncations of the Ga13 polypeptide upstream of residue 259 showed
loss of nuclear localization. Amino-terminal truncations of the same construct retained
nuclear localization, and residues 228-263 of the Ga13 sequence were sufficient to direct
the fusion protein into the nucleus. These results suggest that residues 228-258 of the

Ga13 polypeptide are important for nuclear localization.

Incubation of fibroblasts with leptomycin B resulted in nuclear accumulation of
6313, suggesting that nuclear export of 6313 was mediated by the CRMl receptor. A
candidate leucine-rich nuclear export signal (NES) can be found between residues 240
and 255 of the murine Ga13 sequence. This sequence was engineered into the pRev(l .4)-
GFP reporter system. Residues 240-255 of the Ga13 polypeptide exhibited nuclear export
activity when tested in this system, and nuclear export of the fusion protein was sensitive
to leptomycin B. Site-directed mutagenesis of Leu247 and Ile249 in the Ga13 NES
decreased nuclear export activity, consistent with the notion that these two positions
correspond to critical residues identified in a prototype leucine-rich NES. These results
indicate that residues 240-255 of the galectin-3 polypeptide are important for nuclear

export.

To my parents, to Yu-Wen,
and to all my fi‘iends and family.
Their support has made this possible.

iv

ACKNOWLEDGEMENTS

I must first thank Dr. John Wang for his excellence as a mentor. Through him I
learned a great deal about science. Moreover, Dr. Wang is a true intellectual, and this
quality shined through in lab meetings and lunch discussions. These discussions were not
only entertaining in their own right, but they proved to be excellent opportunities to
refine the core elements of a scientific mind: critical reasoning, synthesis, and spirited
debate.

I also wish to thank the members of my guidance committee for their advice and
suggestions, particularly with respect to this thesis: Drs. Sue Conrad, Donna Koslowsky,
and Ronald Patterson. Thanks also to Dr. Mel Schindler and Sharon Grabski for
assistance and training on the InSight microscope.

Heartfelt thanks are also due to the other members of Dr. Wang’s laboratory, both
past and present. Dr. Eric Amoys provided guidance and advice in the early stages of my
project, but as time went on our relationship became more one of camaraderie and
friendship. Eric has been an excellent example of a young faculty member for a graduate
student to emulate. Thanks also to Patty Voss, who has provided advice and
encouragement on many levels throughout my tour in Dr. Wang’s laboratory. I also owe
a great debt of thanks to Richard Gray, who has been a far better companion than I could
have asked for.

The greatest thanks are due to my parents and to Yu-Wen, who have supported

and encouraged me through this long process. Without them, I’d have quit long ago.

TABLE OF CONTENTS

LIST OF TABLES ...................................................... . ............................ x
LIST OF FIGURES ................................................................................ xi
LIST OF ABBREVIATIONS ........................................................................................ xiii
CHAPTER 1. LITERATURE REVIEW ...................................................... 1
I. Galectins and Galectin-3 .......................................................................... 2
II. Nuclear Transport ................................................................................. 8
Overview ..................................................................................... 8
Nuclear Import .............................................................................. 13
Nuclear Export .............................................................................. 20
Proteins ............................................................................. 20

U snRNAs .......................................................... . ................ 27

tRNA ................................................................................ 27

Ribosomal subunits ............................................................... 28

mRN A .............................................................................. 28

Conclusions ................................................................................. 30
References ................................................................................... 31

CHAPTER 2. SHUTTLING OF GALECTIN-3

BETWEEN THE NUCLEUS AND CYOTPLASM ........................................ 36
Title Page ............................................................................................. 37
Abstract ............................................................................................... 38
Introduction .......................................................................................... 38

vi

Results ................................................................................................ 40

The monoclonal antibody NCL-GAL3 recognizes
human galectin-3 but not the mouse homologue.............................. ................4O

Bead-tagging distinguishes heterodikaryons from homodikaryons .................. 48
Localization of human galectin-3 to both nuclei
of human-mouse heterodikaryons was partially

dependent upon de novo protein synthesis .............................................. 51

Localization of human galectin-3 to both nuclei
of human-mouse heterodikaryons was dependent

upon nuclear export ........................................................................ 54
Mouse galectin-3 shuttles in 3T3-MEF Gal-3 -/- heterodikaryons .................. 55
Discussion ............................................................................................ 63
Materials and Methods .............................................................................. 66
Cell Culture and reagents .................................................................. 66
Bead-tagging and polyethylene glycol-mediated cell fusion ......................... 67
Immunostaining and fluorescence microscopy ......................... - ................ 68
Preparation of cell lysates and immunoblotting ........................................ 69
Acknowledgements ................................................................................. 71
References ............................................................................................ 72

CHAPTER 3. TRANSPORT OF GALECTIN-3
BETWEEN THE NUCLEUS AND CYTOPLASM I.

CONDITIONS AND SIGNALS FOR NUCLEAR IMPORT ............................. 74
Title Page ............................................................................................. 75
Abstract ............................................................................................... 76
Introduction ........................................................................................... 76

vii

Experimental Procedures ........................................................................... 78

Preparation of the pEGFP-cl vector for expression

of the fusion protein GFP-MalE-Gal3 ................................................... 78
Cell culture and transfection ................................... ' ........................... 82
Fluorescence Microscopy ................................................................. 83
SDS-PAGE and immunoblotting ......................................................... 84
Results ................................................................................................ 85
A GFP reporter construct for the localization of Gal3 ................................. 85
Comparison of the subcellular distribution of
GFP-MalE-GalB(l-263) and endogenous galectin-3 .................................. 91
Effects of truncation from the carboxyl terminus
on the localization of the GFP-MalE-Gal3 fusion protein ............................ 94
Effects of truncation of the amino-terminal domain ................................... 97
Discussion .......................................................................................... 100
References .......................................................................................... 107

CHAPTER 4. TRANSPORT OF GALECTIN—3
BETWEEN THE NUCLEUS AND CYTOPLASM II.

IDENTIFICATION OF THE SIGNAL FOR NUCLEAR EXPORT .................. 109
Title Page ........................................................................................... 110
Abstract ............................................................................................. 111
Introduction ......................................................................................... l 12
Experimental Procedures ......................................................................... 1 12

Site-directed mutagenesis of the putative NES sequence

in the GFP-MalE-Gal3 fusion protein .................................................. 112
The pRev(1.4)-GFP vector and variants ............................................... 113
Cell culture and transfection ............................................................ 116

viii

Fluorescence Microscopy ............................................................... 117

Statistical Analysis ........................................................................ 1 17
SDS-PAGE and immunoblotting............................. ........................ 117
Results ............................................................................................... 118
An attempt to identify the NES using the GFP-MalE-Gal3 reporter ............... 118
The Rev(l .4)-GFP vector for the analysis of a functional NES .................... 121
Analysis of the Gal NES in the Rev(l.4)-GFP vector ............................... 127
The effect of LMB on the fluorescence distribution ................................. 128
Site-directed mutagenesis of the Ga13 NES ........................................... 129
Discussion .......................................................................................... 133
References .......................................................................................... 138
CHAPTER 5. CONCLUDING STATEMENTS ........................................................ 140

Concluding Statements ........................................................................... 141

ix

LIST OF TABLES

CHAPTER 1.
Table 1. Nuclear Import: Receptors & Signals ......................................... 16
Table 2. Nuclear Export: Receptors and Signals ....................................... 23
CHAPTER 2.

Table 1. Percent of heterodikaryons showing
staining for galectin-3 in both nuclei .................................................... 60

LIST OF FIGURES

CHAPTER 1.
Figure l. Polypeptide architecture of the galectins ...................................... 4

Figure 2. Schematic diagram depicting the
architecture of the nuclear pore complex ..................................... 11

CHAPTER 2.
Figure 1. Schematic illustrating the use of heterokaryons to study shuttling. . ....42

Figure 2. Western blotting for galectin-3 in lysates of human HeLa cells,
human LG-l fibroblasts, and mouse 3T3 fibroblasts ...................... 45

Figure 3. Immunofluorescence staining for galectin-3 in fixed and
permeabilized human HeLa cells and mouse 3T3 fibroblasts ............ 47

Figure 4. Bead-tagged homodikaryons and heterodikaryons
immunostained for human galectin-3 ........................ . ............... 50

Figure 5. Effect of cycloheximide and leptomycin B on localization of
human galectin-3 in human-mouse heterodikaryons ...................... 53

Figure 6. (A) Western blotting for galectin-3 in lysates of MEF Gal-3 -/-
fibroblasts, MEF Gal-1 -/- fibroblasts, MEF WT fibroblasts,
and mouse 3T3 fibroblasts. (B) Immunofluorescence staining
for galectin-3 in fixed and permeabilized mouse 3T3 fibroblasts
and MEF Gal-3 -/- fibroblasts ................................................ 57

Figure 7. Effect of cycloheximide and leptomycin B on localization
of galectin-3 in mouse 3T3-MEF Gal-3 -/- heterodikaryons ............ 62

CHAPTER 3.
Figure 1. Schematic diagram illustrating the construction of the vector
for the expression of the fusion protein GFP-MalE-Ga13 in

mammalian cells ............................................................... 80

Figure 2. Analysis of the fusion proteins expressed from the GFP
reporter vector by Western blotting .......................................... 88

xi

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

CHAPTER 4.

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Representative fluorescence micrographs illustrating the
N, N>C, N~C, N<C, and C labeling patterns .............................. 90

Comparison of the histograms of fluorescence patterns
obtained with GFP-MalE-Gal3(1-263) with endogenous
Ga13 in 3T3 fibroblasts ........................................................ 93

Comparison of the histograms of fluorescence distribution for
GFP-MalE-Gal3(1-263) and mutants that are truncated from the
carboxyl terminus ............................................................... 96

Comparison of the histograms of fluorescence distributions for
GFP-MalE-Gal3(l-263) and mutants that are truncated from the
amino terminus .................................................................. 99

(A) The amino acid sequence of the carboxyl-terminal 40 residues

of the Ga13 polypeptide. (B) Ribbon diagram showing the three-
dimensional structure of the polypeptide backbone of the

CRD of G313 .................................................................. 102

The Rev(l.4)-GF P vector ................................................... 115

Comparison of the properties of GFP-MalE-Gal3 (1-263) and
GFP-MalE-Gal3(l-263; L247A; 1249A) .................................. 120

Representative fluorescence micrographs illustrating the GFP
localization patterns obtained with various test NES sequences
in the pRev(1.4)-GF P vector ................................................ 124

Histograms showing the distribution of the percent of cells with
fluorescence patterns N, N>C, N~C, N<C, and C.. ............................ 126

A comparison of the histogram distributions of the percent of cells
with various fluorescence patterns for Ga13 NES (WT), Ga13 NES
(1244A; L247A), and Ga13 NES (L247A; I249A) ............................... 131

xii

ActD
CD
CHX
CRD
DME
FITC
Gal 1
Ga13
GFP
GST
hnRNP
IBB
IgG
L-rich NES
LMB
MEF
mRNA
ND
NES
NLS
NPC
PBS
PCR
PI

PKI
RNA
RNP
rRNA
SMN
snRNP
TTF- 1
T-TBS
tRNA
WT

LIST OF ABBREVIATIONS

actinomycin D

Carboxy-terminal domain
cycloheximide

Carbohydrate Recognition Domain
Dulbecco's modified Eagle's medium
fluorescein isothiocyanate
galectin-l

galectin-3

green fluorescent protein
glutathione S-transferase
heterogeneous ribonucleoprotein
importin-beta binding
immunoglobulin G

leucine-rich nuclear export signal
leptomycin B

mouse embryo fibroblasts
messenger ribonucleic acid
Amino-terminal domain

nuclear export signal

nuclear localization signal

nuclear pore complex
phosphate-buffered saline
polymerase chain reaction
propidium iodide

inhibitor of CAMP-dependent protein kinase
ribonucleic acid

ribonucleoprotein complex
ribosomal ribonucleic acid

survival of motor neuron

small nuclear ribonucleoprotein
thyroid-specific transcription factor
Tris-buffered saline containing 0.05% Tween 20
transfer ribonucleic acid

wild-type

xiii

Chapter 1

Literature Review

I. Galectins

The galectin family of proteins, currently containing 14 members, is characterized
by: (a) an affinity for B-galactosides; and (b) the presence of at least one conserved
carbohydrate recognition domain (CRD), which is responsible for ligand binding (1; 2).
The galectins are widely distributed phylogenetically, and analysis of genomic DNA
sequence banks has yielded additional candidates for inclusion in the mammalian galectin
family, as well as galectin candidates in plants and viruses (3). The galectins have been
classified into three subfamilies based on the number and organization of their
carbohydrate binding domains. The prototype subfamily contains one CRD, the chimera
subfamily contains a unique proline- and glycine-rich domain fused to the amino
terminus of the CRD, and the tandem repeat subfamily contains two CRDs fused in
tandem (Figure 1) (4).

Galectin-3 (Mr ~30 kDa) is the sole member of the chimera subfamily (4), and
consists of two domains. The amino-terminal domain (ND) of galectin-3 is unique, and
contains multiple repeats of a nine-residue consensus sequence P-G-A-Y-P-G-X-X-X (1;
5). The carboxy-terminal domain (CD) of galectin-3 contains the conserved CRD.
Physical-chemical studies of the galectin-3 polypeptide, as well as the two domains
individually, have indicated that the ND and CD are structurally and firnctionally distinct
(5). Differential scanning calorimetry revealed that the ND possesses a lower transition
temperature than the CD (6). Accordingly, the ND exists in an unfolded, random coil

state (7), while the CD exists as five-stranded and six-stranded B-sheets arranged

Figure 1. Polypeptide architecture of the galectins.
Members of the galectin family are organized into three subfamilies based upon their
domain organization. Residues which are critical for carbohydrate binding are indicated

in bold.

Subfamily

Prototype

Tandem Repeat

Chimera

Figure 1

Domain Organization

 

 

H-NPR~V-N~WG-E-R~F-G~R

 

 

 

H-NPR~V-N~WG-E-R~F-G~R

 

H-NPR~V-N~WG-E-R~F-G~R

 

 

(PGAYPGXXX),

 

 

H-NPR~V-N~WG-E-R~F-G~R

 

 

 

 

in a [El-sandwich (8). Residues critical for carbohydrate binding lie in three strands of the
six-stranded sheet (8).

Although the galectin family was so named because the minimal monosaccharide
ligand is galactose, most members of the family, including galectin-3, have a much
higher affinity for lactose and N-acetyllactosamine disaccharides. Galectin-3 also binds
larger, multivalent oligosaccharides such as poly-N-acetyl-lactosaminoglycan, a
saccharide polymer common in extracellular matrix (1; 8). Indeed, while free galectin-3
is typically monomeric (7; 9), binding to multivalent ligands induces galectin-3 to form
oligomers (10). Self-association of individual ND and CRD fragments was also
observed, indicating that oligomerization of galectin-3 may occur through either ND-ND
or CRD-CRD association (7).

Galectin-3 is primarily intracellular, and has been documented in the nuclear and
cytoplasmic compartments (11; 12; 13; 14). The subcellular distribution of galectin-3 in
cultured fibroblasts is dependent upon the proliferative status of the culture. Galectin-3
in quiescent cultures is predominantly cytoplasmic, but proliferating cultures fiom the
same cell type show predominantly nuclear localization (15). Moreover, human diploid
fibroblasts have a finite replicative capacity in culture. Galectin-3 in low passage-
number cultures (high replicative capacity) exhibited nuclear and cytoplasmic
localization, but cultures monitored many passages later (at a low replicative capacity)
exhibited primarily cytoplasmic galectin-3 (16; 17).

Several lines of evidence indicate that nuclear galectin-3 is associated with
ribonucleoprotein (RNP) complexes (12). Treatment of unfixed, permeabilized

fibroblasts with ribonuclease A eliminated galectin-3 staining in the nucleus, but

treatment in parallel with deoxyribonuclease I produced no such effect (12; 13).
Moreover, separation of nucleoplasm by cesium sulfate centrifugation yielded galectin-3
in fractions with densities matching those reported for heterogeneous nuclear RNPs
(hnRNPs) and small nuclear ribonucleoproteins (snRNPs). This observation prompted
the use of depletion and reconstitution experiments to document that galectin-3 was a
required factor for splicing of pre-mRNA (18).

The yeast two-hybrid system was employed to search for ligands of galectin-1, a
relative of galectin-3. This assay revealed an interaction between galectin-1 and the
carboxy-terminal 50 amino acids of the protein Gemin4 [Gemin4(C50)]. This interaction
was confirmed using purified GST-Gemin4(C50), and in parallel galectin-3 was also
shown to interact directly with GST-Gemin4(C50) (19). Gemin4 is a component of a
macromolecular complex designated as the SMN (survival of motor neuron) complex
(20). The SMN complex functions in both the nucleus and the cytoplasm. In the
cytoplasm, the SMN complex mediates assembly of snRNP particles; in the nucleus, it
delivers the snRNPs to the pre-mRNA during the early stages of spliceosome formation
(21; 22). The association of galectin-3 with the SMN complex raises the possibility that
galectin-3 might perform related functions in both the nucleus and the cytoplasm in the
context of its role as a factor in pre-mRNA splicing.

In addition to members of the SMN complex, galectin-3 interacts with a variety of
other ligands in both the nuclear and cytoplasmic compartments (5; references therein).
In the nucleus, galectin-3 interacts directly with the homeodomain of the thyroid-specific
transcription factor (TTF-l), stimulating the DNA-binding activity of the latter protein.

This in turn increases the transcriptional activity of TTF- l , leading to increased

proliferation of thyroid cells (23). A second nuclear ligand for galectin-3 is CBP70, a
lectin with affinity for glucose/N-acetylglucosamine (GlcNAc) saccharides (24).
Galectin-3 was co-purified with CBP7O from HL6O cell nuclei. Interaction of the two
proteins was inhibited by lactose, suggesting that lactose induces a conformational
change in galectin-3 that disrupts the interaction with CBP7O (24; 25).

Galectin-3 interacts in the cytoplasm with Bel-2 (26; 27) and synexin (28), in both
cases producing an anti-apoptotic effect. Galectin-3 and Bel-2 share two aspects of
sequence similarity: the amino-terminal regions of both proteins are rich in proline,
glycine, and alanine residues; and both proteins contain a motif of NWGR residues in
their carboxyl termini. Mutation of the NWGR motif in galectin-3 abolished its anti-
apoptotic activity. Moreover, galectin-3 binds to Bel-2 in vitro, mimicking the
heterodimerization of Bel-2 family members (26; 27). Studies of apoptosis in human
breast carcinoma induced by cisplatin revealed that galectin-3 translocation to the
mitochondrial membranes prevented mitochondrial damage and apoptosis (28).
Translocation of galectin-3 to the mitochondrial membranes was mediated by synexin,
and reduction of synexin levels abrogated galectin-3 translocation and anti-apoptotic
activity.

The studies reviewed here illustrate a variety of intracellular ligands and fimctions
for galectin-3, both in the nuclear and cytoplasmic compartments. Concomitantly, they
also suggest several important questions for firture research on galectin-3. The
interactions of galectin-3 with many of its intracellular ligands occur through protein-
protein interactions rather than carbohydrate recognition (1). Therefore, one key topic for

future work is to determine the significance of galectin—3 carbohydrate-binding activity

with respect to its intracellular function. Are there intracellular functions of galectin-3
which can be definitively linked to its carbohydrate-binding activity? In a related vein, it
will be important to distinguish between activities which are either facilitated or inhibited
by carbohydrate-binding per se, and activities which are facilitated or inhibited through
the course of a conformational change in galectin-3 arising from carbohydrate-binding.
A second and more general point for future research is to rationalize how
galectin-3 is involved in such a number and variety of activities. Is galectin-3 so
multifunctional as to be involved, for example, in RNA processing, transcriptional
regulation, and anti-apoptotic signaling? Altemately, might galectin-3 participate in one
critical underlying process (e.g. RNA processing), through which other effects are
mediated, either directly or indirectly (29)? Answers to such questions are important not
only to the study of galectin-3, but in the broader context of understanding how multiple

(and apparently disparate) processes in a cell might be coordinated.

11. Nuclear Transport

II.A. Overview

A defining feature of eukaryotic cells is the isolation of the nucleus as a distinct
subcellular compartment by the nuclear envelope, a double-membrane system that is
contiguous with the endoplasmic reticulum (30; 31). This separation necessitates a
mechanism of regulated communication and transport between the nuclear and
cytoplasmic compartments in order to coordinate fundamental cellular processes such as

signal transduction, transcription and translation, and cell division (32; 33; 34).

Substrates for transport into (nuclear import) or out of (nuclear export) the nucleus
include proteins, tRN A molecules, U snRN A molecules, and mRNA molecules (the latter
two being transported as ribonucleoprotein complexes or RNPs).(30; 31; 33; 35).

Traffic between the nucleus and cytoplasm occurs through the nuclear pore complex
(NPC), a large multi-protein complex which spans the nuclear envelope and provides an
aqueous channel between the nuclear and cytoplasmic compartments (30; 33; 35; 36).
The vertebrate NPC is composed of approximately 30 distinct proteins termed
nucleoporins (36; 37). The nucleoporins are present and arranged in groups of eight, and
structural analysis has revealed that the NPC exhibits eight-fold symmetry (34; 36; 37).
As depicted in Figure 2 (34), the central feature of the NPC is the cylindrical central
framework. The central framework is composed of eight spoke-like structures which
form the aqueous channel through which cargo-receptor complexes traffic (33; 34; 37).
The central framework is sandwiched between the cytoplasmic and nuclear ring
structures, each of which contains eight protruding filaments. The filaments of the
cytoplasmic ring extend into the cytoplasm, whereas the filaments of the nuclear ring
extend into the nucleoplasm but are joined at a distal ring (33; 34; 37).

Recent evidence indicates that interactions between nucleoporins and transport
receptors play a significant role in moving cargo through the NPC (33; 37; 38). These
interactions occur predominantly between phenylalanine and glycine residues (so-called
FG-repeats) of the nucleoporins and hydrophobic residues of transport receptors (37).
Nucleoporins in the cytoplasmic fibrils bind importin-cargo complexes, possibly serving

as “loading platforms” for entry to the NPC (37).

Figure 2. Schematic diagram depicting the architecture of the nuclear pore complex.
The central framework is depicted as a complex of eight cylindrical subunits located in
the center of the nuclear pore. The cytoplasmic ring is depicted as a thick ring above the
central framework, and cytoplasmic fibers are indicated as wavy structures. The
nucleoplasmic ring is depicted as a thick ring below the central framework.
Nucleoplasmic filaments are depicted as rod-like structures extending conically from the
nucleoplasmic ring, and are joined in a smaller ring at the distal end to form the nuclear

basket. This figure is reproduced from reference 34.

10

Figure 2

Cytoplasmic
Fibers

Cytoplasmic
Ring

Central
8-spoke
Cylinder

 

Nucleoplasmic
Ring

Nuclear
Basket

 

Moreover, the import receptor importin-B (see below) exhibits progressively increasing
affinities for nucleoporins in the cytoplasmic fibrils, the central core, and the nuclear
basket, indicating that its interaction with these nucleoporins may augment the
directionality of transport imposed by Ran. Conversely, the export receptor CRMl (see
below) has high affinity for a nucleoporin in the cytoplasmic fibril (33).

Most cargo destined for nuclear import or export bear signals specifying their
fate. These have been dubbed nuclear localization signals (NLS) and nuclear export
signals (NBS), specifying nuclear import and nuclear export, respectively (34; 35).
Cargo destined for transport through the NPC is first bound by at least one soluble
receptor which recognizes the presence of a NLS or NES in the cargo, and mediates
transport of the cargo through the NPC. Transport receptors that recognize the NES of
nuclear cargo and mediate export to the cytoplasm have been termed exportins, while
receptors that bind the NLS of cytoplasmic cargo and mediate import to the nucleus have
been termed importins (32; 34; 39).

Implicit in this system is the principle of directionality: cargo is transported in the
proper direction and released in the proper compartment. Directionality of nuclear
transport is imposed by the GTPase Ran (30; 32; 34; 35; 39; 40), and is achieved by Ran
mediating the interaction between cargo and receptor relative to its GTP/GDP state. In
its GTP-bound state, Ran favors exportin-cargo interaction, but is antagonistic to
importin-cargo interaction. Conversely, Ran in the GDP-bound state favors importin-
cargo interaction, but is antagonistic to exportin-cargo interaction (41). A gradient of
Ran-GTP is maintained across the nuclear envelope by the concerted action of several

Ran-associated proteins. At the cytoplasmic face of the NPC, the GTPase activator

12

RanGAP and the coactivator Ran-binding protein 1 (RanBPl) promote GTP hydrolysis
by Ran. Conversely, at the nuclear face of the NPC, exchange of GDP for GTP by Ran is
promoted by the guanine exchange factor RanGEFl (3 5). Through the coordinated
activity of Ran, RanGAP/RanBPl, and RanGEFl , the nuclear compartment is enriched
for RanGTP, while the cytoplasmic face of the NPC is enriched for RanGDP. Thus the
formation of RanGDP-importin-cargo complexes is promoted at the cytoplasmic face of
the NPC, while the formation of RanGTP-exportin-cargo complexes is favored at the
nuclear face of the NPC (30; 35; 36; 40).

11.3 Nuclear Import

Substrates destined for nuclear import are identified by the presence of a nuclear
localization signal, or NLS, and include proteins and U snRNPs. In proteins, the NLS is
a region of amino acids within the protein which is recognized by at least one receptor,
most often referred to as importins or karyopherins (30; 31; 35; 39). Among the earliest
identified NLSs (so-called “classical” NLSs) were those of the SV40 virus large T
antigen (SV40 NLS) and nucleoplasmin (30). The SV40 NLS consists of the sequence
PKKKRKV (42), and the nucleoplasmin NLS consists of the sequence
KRPAATKKAGQAKKKK (43), where residues in bold are critical for nuclear import.
Because the SV40 NLS contains one continuous stretch of basic residues, it has been
dubbed a “simple” or “monopartite” NLS, while the nucleoplasmin NLS has been dubbed
a “bipartite” NLS owing to the presence of two separated clusters of basic residues (35;
39).

Owing to their early identification, these “classical” motifs dominated early

notions of what constituted an NLS. However, most proteins in which the NLS has been

13

identified do not bear a “classical” NLS, but rather variations on the theme, or even NLSs
without apparent similarity to the “classical” examples (32; 35). As illustrated in Table 1,
there is considerable variety among the characterized NLS motifs. For example, the HIV
Rev protein NLS consists of an approximately lZ-residue stretch rich in arginines, while
the SR protein NLS is characterized by the presence of clustered serine and arginine
residues. More divergent is the RGG domain (a 120-residue stretch of RGG repeats) of
the yeast Npl3p, Nab2p, and Hrplp proteins. Perhaps the most atypical NLSs
characterized to date are the trimethyl-guanosine cap employed by the U snRNPs (see
below), and the M9 signal (approximately 40 residues) identified in the hnRNP proteins
A1, A2, and F, as well as the TAP protein (32; 35; references therein). The M9 signal
deserves special mention because (a) it is almost entirely uncharged, and (b) it also acts
as a nuclear export signal (see below). It is important to note that not only the apparent
chemistry of these NLS motifs varies, but also the length of the NLS. Compare, for
example, the short “classical” c-myc NLS (only three basic residues) with the 120-
residue RG-rich NLS of Npl3p. Finally, many proteins which undergo nuclear import
have no characterized NLS to date.

NLSs are recognized by soluble transport receptors known as importins (30; 32;
35). The prototype importins were identified as factors required for nuclear import of
cargo bearing the SV40 NLS, and have been dubbed importin-a and importin-B (31; 44;
45; 46; 47; 48; 49; 50; 51). These studies revealed that importin-a binds directly to the
NLS and acts as an adaptor protein to coordinate binding of importin-B, while importin—B

mediates docking with and translocation through the NPC. In this scenario, importin-or

and importin-B were deemed to function as a heterodimer (31; 39).

14

Table 1. Nuclear import receptors and the signals they recognize.
Nuclear import receptors are listed in the left column (Receptors), cargo is listed in the
middle column (Cargo), and the signal recognized by each receptor is indicated in the
right column (Signal Recognized). N/K, not known. Compiled from references 30, 32,

35 and references therein, except: I52; 253.

Table 1

Nuclear Import: Receptors & Signals

 

 

R_ecentor(s) Cargo Sim] Recognized
Importin-B/ SV40 T Ag _ PKKKRKV
Importin-or nucleoplasmin KRPAATKKAGQAKKKKLD
c-myc PAAKRVKLD
Importin-B ribosomal proteins extended R-rich domains

Transportin 1
(aka Importin-BZ)

Transportin SR
(aka Importin-B3)
Importin-B3

Importin-S

Importin-7

Importin-9l
Importin-l l

Importin-B/
RIPa

Importin-B/
Snurportin

Importin-B/
Importin-7

NTF2

HIV Rev
HTLV Rex
HIV Tat
histone core proteins

hnRNPAl, A2, F
Nup153

histone core proteins

SR proteins

ribosomal proteins

ribosomal proteins
histone core proteins

ribosomal proteins
histone core proteins

histone core, esp. HZB

chM2

replication protein A

RQARRNRRRRWR
R-rich domain
R-rich domain
BIB domainI
M9 (bi-directional signal)
BIB domain

S-, R- rich domains

N/K

N/K
BIB domain

extended R-rich domains
BIB domain

BIB domain
N/K

N/K

U snRNPs m3G cap, Sm core proteins
histone H1 extended basic-rich domain
Ran salt bridge between NTF2 and Ran2

l6

Soon after these seminal discoveries, a variety of importin-or (54) and importin-B
(32; 35) relatives were characterized. Most importin-a relatives were characterized in
humans and were identified based upon the presence of a short amino-terminal basic
domain responsible for binding to importin-B (so-called importin-B binding or IBB
domain), and a large domain containing Arm repeats (named for homology to the
Drosophila protein Armadillo) responsible for NLS-binding (55; 56). All importin-a
relatives function as adaptors to bridge “classical” NLS motifs with importin-B.
However, nuclear import assays revealed that among the importin-or relatives, NLSs of
different proteins were bound with varying efficiency, suggesting the possibility of NLS
preference among the importin-a relatives (54).

Importin-B family members have been identified based upon the presence of an
amino-terminal binding domain for Ran, and of 15 — 20 repeats of the HEAT motif (32;
33). In contrast to the prototype scenario of an importin-or/B heterodimer, most of the
newly identified importin-B relatives mediate nuclear import without importin—or or other
adaptor(s), and exhibit a wide range of cargo recognition (32; 35). Several cases of direct
interaction between importin-B relatives and cargo are indicated in Table 1. Importin-B
recognizes several cargos directly, including the HIV Rev and Tat proteins, ribosomal
proteins, and the HTLV Rex protein, all of which appear to interact with importin-B via
extended R-rich motifs. Transportin l/importin-BZ recognizes the M9 signal in hnRNP
proteins to mediate their nuclear import, and Transportin SR/importin-B3 recognizes

serine- and arginine-rich domains in the SR proteins.

17

Import of histone core proteins by importin-B relatives (52) deserves special
mention for two reasons. First, many importin-B relatives import histone core proteins in
addition to other cargo. Importin—B, for example, imports a large variety of cargo.
Second, interaction with the histone core proteins occurs through their BIB domain (52),
an exceptionally basic motif distinct from the IBB domain of importin-or (57). The
histone core proteins therefore highlight a trend among cargos that interact directly with
importin-B family members: interaction occurs through motifs (NLS or BIB) of
exceeding basic chemistry. Moreover, importin-B relatives possess different binding sites
to accommodate different NLSs (52; 57), explaining how importin-B relatives (and
importin-B in particular) can accommodate such a variety of cargo and signals. The
question of whether importin-B relatives import multiple cargos simultaneously or if
import occurs singly is still open, but evidence to date suggests that import is most
efficient when cargo is handled singly (52; 57).

Two exceptions to the trend of direct interaction between importin-B relatives and
cargo deserve mention. First, there are cases in which importin-B interacts with cargo
through an adaptor distinct from importin-or. These include the importin-B/RIPor
pathway for import of replication protein A, and the importin-B/snurportin pathway for
import of U snRNPs. RIPOL was identified as a required cofactor for nuclear import of the
replication protein A, and is unrelated to importin-or (58). Similar to the histone core
proteins, RIPa bears a domain rich in basic residues, which is likely to mediate
interaction with importin-B. Snurportin was identified as a required cofactor for nuclear

import of the U snRNPs (59; 60). While snurportin contains an IBB domain reminiscent

18

of importin-a, the two are dissimilar in all other respects. Snurportin recognizes the m3G
cap structure of mature U snRNPs and represents the only case to date of an RNA
structure being identified as an NLS.

The second variation is the importin-B/importin-7-mediated import of ‘histone H1
(61). In this case, importin-B employs importin-7 as an adaptor to interact with H1. The
exact mechanism of interaction is uncertain, but depends upon importin-B or importin-7
recognizing a BIB domain in H1. The likeliest scenario (61) is that importin-B and
importin-7 form a loose heterodimer in the cytoplasm, which is greatly stabilized by the
binding of H1, and the resulting trimeric complex is imported to the nucleus. The
example of H1 import is particularly fascinating because importin-7 and importin-B
function independently to import other cargo. Their cooperation represents the only case
to date of two importin-B relatives acting together, with one in the role of adaptor.

As described above, directionality of nuclear transport is mediated by Ran. Ran
at the cytoplasmic face of the NPC is GDP-bound, promoting importin-cargo association.
However, Ran-GDP does not travel through the NPC with the importin-cargo complex,
but remains associated with nucleoporins at the cytoplasmic side of the NPC (41; 53).
Conversely, Ran at the nuclear face of the NPC is GTP-bound, promoting importin-cargo
dissociation and subsequent exportin-cargo association. Critically, Ran-GTP associates
with exportin-cargo complexes during export (see below), and upon reaching the
cytoplasmic face of the NPC is acted upon by RanGAP. This results in GTP hydrolysis
by Ran, leading to dissociation of the exportin-cargo complex. Since Ran-GDP is not
carried back into the nucleus with importin-cargo complexes, an alternate mechanism

must facilitate import of Ran back into the nucleus lest repeated rounds of nuclear import

l9

and export deplete the nuclear pool of Ran and eventually “stall” nuclear transport.
Indeed, the protein NTF2 (nuclear transport factor 2) binds cytoplasmic Ran-GDP and
carries it through the NPC. At the nuclear face of the NPC, Ran-GDP is acted upon by
RanGEF, effecting a replacement by Ran of GDP for GTP and resulting in dissociation of
Ran-GTP from NTF2 (53). NTF2 is a small protein of approximately 15 kDa, highly
conserved, and is unrelated to any of the importin family members (62; 63).

II. C Nuclear Export

Nuclear export proceeds in a manner that is analogous to import, but in many
cases nuclear export is a much more complex process (30; 39). Substrates destined for
nuclear export are recognized by receptors known as exportins, which together with Ran-
GTP, mediate translocation of the exportin-cargo-Ran-GTP complex through the NPC.
Cargos destined for nuclear export include proteins, immature U snRNAs, tRNA,
ribosomal subunits, and mRNA (30; 39). Each of these cargo families proceeds through
at least one unique export pathway, resulting in considerable diversity and complexity of
nuclear export machinery. For the sake of simplicity, the pathway(s) responsible for
export of each cargo type will be discussed individually.

II. C. 1 Proteins

Most proteins identify themselves for export by the presence of a conserved
nuclear export signal rich in leucine or isoleucine residues (30; 35; 39), dubbed the
leucine-rich (L-rich) NES. The prototype examples of the L-rich NES were identified in
the HIV Rev protein (64; 65) and the protein kinase inhibitor PKI (65). Since then, L-
rich NESs have been identified in a wide variety of proteins (30; 66), and the consensus

sequence of L-X(2/3)-Z-X(2/3)-L-X-L/I (where X represents any amino acid; Z represents

20

L, I, F, V, or M) has been accepted as typical of known L-rich NESs (39). The exportin
responsible for conveying L-rich NBS-bearing proteins out of the nucleus is CRMl (also
known as exportin-1) (67; 68), a relative of importin-B. Formation of a stable, “export
competent” CRMl-mediated complex requires the presence of cargo bearing an L-rich
NES, CRMI, and Ran-GTP. CRMl is specifically inhibited by the fungal antibiotic
leptomycin B, a property which figured prominently in early characterizations of CRMl
(67; 68), as well as subsequent identification of cargo exported by CRMl (30).

As highlighted in Table 2, many proteins are exported by CRMl, and they cover a
wide range of functions. One prime example is the HIV Rev protein, which functions as
an adaptor to funnel HIV RNA transcripts bearing the Rev response element (RRE) into
the CRMl export pathway for transport to the cytoplasm (64). This represents a general
trend of viruses in which cellular machinery is co-opted for the purpose of viral
propagation. In this case, HIV benefits from “hijacking” a cellular export pathway to
process its own transcripts.

Another prime example is PKI, the inhibitor of cyclic AMP-dependent protein
kinase (PKA) (65). Upon stimulation by increased CAMP, the PKA holoenzyrne
dissociates, allowing the catalytic subunits of PKA to enter the nucleus and
phosphorylate the cAMP response element binding (CREB) protein, thereby increasing
transcription of genes bearing the CAMP response element (CRE) (69). To negate this
effect, PKI enters the nucleus (apparently through simple diffirsion, 70; 71), binds the
catalytic subunits, and then firnnels them into the CRMl export pathway. In this case,

the end result is a change in gene regulation by removal of a signaling kinase.

21

Table 2. Nuclear export receptors and the signals they recognize.
Nuclear export receptors are listed in the left column (Receptors), cargo is listed in the
middle column (Cargo), and the signal recognized by each receptor is indicated in the
right column (Signal Recognized). N/K, not known. Compiled from references 30, 32,
35, and references therein except: 172; 267; 373; 474; 5 75; 676; 777; 360; 966; 1078; 1179;

1280; 1381;“‘82; ‘583.

22

Table 2:

Nuclear Export: Receptors and Signals

Signal Recognized

 

 

Proteins
Receptor“) Cargo
CRMl proteins containing L-rich NES

(aka exportin 1)

CAS
(aka exportin 2)

Exportin 4
Transportin 2
Exportin 6

N/K

CRMl

Exportin t

TAP/NXF 1
(aka Mex67p)
Transportin 2

Exportin 5

Rev; PKI; Icha; STATI; c-Abl
Snurportin

Importin-or

eIF-Sa
HuR
actin

proteins bearing M9 signal

RNA
immature U snRNA

43S rRNA
6OS rRNA

ARE-containing mRNA,

via HuR assoc. w/pp32 and April

tRNA

bulk mRNA
via Aly/REF, SR proteins

ARE-containing mRNA, via HuR

miRNA precursor

23

L-X(2/3)-Z-X(2/3)-L-X-L/I

~200 aa, regions similar
to L-NEs"2

~l40 aa, acidic“4

N/K5
HNS (M9-like)”
N/K7

N/K

m7G cap binds CBC;

CBC binds PHAX (L-NES)8
N/K9

subunit protein RpllO bound

by de3p (L-NES)'°’”

pp32/April bear L-NES

Acceptor and T‘I’C arms,
Mature 5’ and 3’ termini

N/K9,12,13

HuR bears HNS6

N/K; proper end
processing reqdws

The theme of regulating gene expression by controlling localization of signaling
proteins or transcription factors has several other examples. Regulation of the NFKB/Rel
family of transcription factors (84), as well as the STATl transcription factor (85) is
accomplished through nuclear export. The inhibitor of K B-a (Ichor) protein (84) binds
NFch/Rel proteins in the nucleus and acts as an adaptor for CRMl-mediated nuclear
export. In the case of STATl, binding to CRM] is direct and occurs through an L-rich
NES in STATl (85). Finally, activity of the c-Abl tyrosine kinase is also regulated by
CRMl-mediated nuclear export. c-Abl contains an L-rich NES, and is exported from the
nucleus in response to attachment of cells to fibronectin, indicating a shift in function of
c—ABL from gene regulation in the nucleus to transduction of adhesion signals in the
cytoplasm (86).

The snurportin protein, which functions as an adaptor for importin-B-mediated
nuclear import of U snRNPs, is also exported fi'om the nucleus by CRMl (67; 72). The
interaction between snurportin and CRMl is unique because it does not occur through a
typical L-rich NES, but rather through a region of approximately 200 amino acids in
snurportin which contain regions of similarity to the L-rich NES. Moreover, despite its
unusual mode, the interaction between snurportin and CRMl is much stronger than that
with cargos bearing the L-rich NES (72). The basis for the avidity of snurportin-CRM]
interaction compared to L-rich NES-CRMl interaction is unclear, but one possibility is
that several of the “imperfect” L-rich NES regions in snurportin might jointly interact
with CRMl, resulting in an interaction that is stronger than a single “perfect” NES (72).
This issue will likely be resolved only through improved structural knowledge of the

interaction between snurportin and CRM 1.

24

Importin-a is a prototype example of proteins that are exported by receptors other
than CRMl (30; 32; 35). Importin-a is the sole substrate of the CAS receptor, which
behaves similarly to CRMl and is also an importin-B relative (73; 74). Like CRMl, CAS
requires Ran-GTP to mediate nuclear export of importin-0L, but the mode of interaction
between CAS and importin-0t is unique and depends upon an acidic region of
approximately 140 amino acids near the carboxyl terminus of importin-or. Intriguingly,
importin-a bears a sequence of amino acids with homology to the L-rich NES, yet failed
to interact with CRMl either in vitro or in vivo (73). This highlights the critical point that
“predicting” the presence of an L-rich NES in a protein based on sequence alone is error-
prone (30). Indeed, both importin-a and importin-B bear sequences similar to the
consensus L-rich NES, yet neither is exported by CRM]. Importin-B exits the nucleus in
association with Ran-GTP and accesses the NPC directly, migrating through the pore by

interacting with nucleoporins (30; references therein).

The eukaryotic translation initiation factor 5A (eIF-SA) is exported by exportin 4
(75), an importin-[3 relative. Like CRM] , exportin 4 requires Ran-GTP to form an
export-competent complex with eIF-SA, but eIF-SA does not possess a typical L-rich
NES. Instead, a unique modification of a lysine residue to hypusine (75; references
therein) dramatically enhances the interaction between exportin 4 and eIF-5A. Moreover,
the site of this modification is located in a highly exposed and flexible region of eIF—SA,
suggesting that the hypusine modification makes direct contact with exportin 4. The
biological significance of the exportin 4 pathway and the eIF-SA protein are not
understood, but Lipowsky et al. have suggested that eIF-SA might function as an export

adaptor or chaperone for RNA (75).

25

Nuclear export of actin is mediated by exportin 6 (77), an importin-B relative.
Actin nuclear export requires Ran-GTP, as well as an interaction between actin and
profilin, which binds actin monomers and suppresses spontaneous actin polymerization.
Exportin 6 interacts with the actin-profilin complex much more avidly than it does with
either component alone, suggesting that profilin acts as an adaptor between actin
monomers and exportin 6. Entry of actin monomers into the nucleus is likely to occur
following cell division, when they are packaged into daughter nuclei during regeneration
of the nuclear envelope. Were the actin monomers to achieve sufficient concentration
inside the nucleus, they would polymerize into cytotoxic aggregates. Exportin 6
therefore plays a key role in protecting cells from actin toxicity, and it is not surprising
that exportin 6-mediated nuclear export of actin is conserved between vertebrates and
insects (77).

Nuclear export of the protein HuR is more complex, in that it accesses two export
pathways: one mediated by CRMl, and the other mediated by transportin 2, a relative of
transportin l (76). HuR accesses these pathways by different mechanisms, and the
pathways appear to be mutually exclusive. Access to the CRM] export pathway occurs
through HuR interaction with the nuclear phosphoproteins pp32 and April, both of which
bear L-rich NESs (76). In this fashion, HuR “piggy-backs” into the CRMl export
pathway via its interactions with pp32 and April. Altemately, HuR can access the
transportin 2 export pathway directly via its HuR nuclear shuttling (HNS) region (76; 87),
a region which bears moderate sequence similarity to the M9 signal of hnRNPAl. As

discussed below, HuR recognizes AU-rich elements (ARES) in the mRNAs of early

26

response genes, and participates in exporting ARE-bearing mRNAs from the nucleus (see
below) (76).

II. C.2 U snRNAs

CRM] mediates nuclear export of immature U snRNAs. U snRNAs are
transcribed in the nucleus, and most U snRNAs require export to the cytoplasm where
they associate with Sm core proteins and complete their maturation (60). Immature U
transcripts are bound prior to export by the cap binding complex (CBC) which recognizes
the monomethyl m7G cap at the 5’ end of the transcript. The CBC is bound by the
protein PHAX (phosphorylated adaptor for RNA export), which bears an L-rich NES.
PHAX thus serves as an adaptor to facilitate interaction of the U-m7G-CBC complex
with CRMl (60).

II.C.3 tRNA

Export of tRN A is mediated by exportin-t, another relative of importin-B (88).
All tRNAs are transcribed in the nucleus, and prior to export undergo a series of
modifications including 5’ and 3’ end trimming. Exportin-t affinity for tRNA is vastly
improved by the presence of proper 5’ and 3’ ends, correct tRN A folding, and base
modifications (30; 35; 88), suggesting that Exportin-t acts not only as an export receptor,
but also as a final checkpoint for proper maturation of the tRNA prior to export.
Exportin-t is unique because it does not employ an adaptor to interact with tRNA,
whereas all other export receptors that manage RNA cargos employ at least one adaptor

(see Table 2).

27

II.C.4 Ribosomal subunits

Ribosomal subunits are assembled in the nucleolus, from which they are exported
to the cytoplasm and assembled into functional ribosomes (66; 78; 79). The mechanism
of export of ribosomal subunits to the cytoplasm remains incompletely understood, but
recent experiments have indicated a role for CRMl in export of the 608 subunit. 60$
rRNA transcripts are bound in the nucleolus by the ribosomal subunit protein RpllOp,
which is in turn bound by the protein de3p. de3p bears two L-rich NESs, and
apparently functions as an adaptor between the RpllOp/6OS transcript and CRMl (78;
79). Export of the 43S ribosomal subunit requires the presence of Ran-GTP and is
sensitive to mutations in CRMl, indicating that 43S subunits may be exported through
the CRMl pathway (66). How the 43S subunit might interact with CRMl remains to be
demonstrated.

11. C5 mRNA ,

Most poly-adenylated transcripts (bulk mRNA) are exported by the protein
TAP/NXF 1 (Mex67p in yeast) (80; 81; 89). TAP was originally identified as a cellular
factor responsible for binding the constitutive transport element in retroviral RNAs and
exporting them to the cytoplasm (81; references therein). TAP is a member of the
nuclear export factor (NXF) family of proteins, and is unrelated to importin-B (89). TAP
acts as a heterodimer with the cofactor p15, and TAP/p15 heterodimers interact directly
with nucleoporins in order to traverse the NPC. The interaction of TAP/p15 with mRN A
occurs through RNA-binding proteins which serve as adaptors between the mRN A and

TAP/p15 (80; 89).

28

Two examples of adaptor proteins for TAP/p15 have been characterized. The
REF protein (also known in metazoans as Aly) is deposited upon pre-mRNAs during
splicing as a member of a protein complex situated upstream of the junction between two
exons (termed the exon-exon junction complex, EJC) (89; 90; references therein). REF
can interact directly with TAP/p15, and has been proposed to act as an adaptor to link
spliced mRNAs into the TAP/p15 export pathway (reviewed in 89).

The other adaptors for TAP/p15 are the shuttling serine/arginine-rich splicing
factors (SR proteins) 9G8 and SRp20 (81). These SR proteins shuttle between the
nucleus and cytoplasm, and interact directly with TAP. 9G8 and SRp20 bind a consensus
22-nucleotide export element present in many mRNAs (91), and once bound to the export
element serve as adaptors to bridge their transcript to TAP/p15.

In contrast to TAP-mediated export of bulk mRN A, transcripts bearing AU-rich
elements (ARES) are bound by the protein HuR and funneled into two pathways distinct
from TAP/p15 (76). As described above, HuR bears the HNS signal, which is recognized
by the export receptor transportin 2. Thus, HuR can function as an adaptor to firnnel
ARE-bearing transcripts into the transportin 2 export pathway. Altemately, HuR also
interacts with the proteins pp32 and April, both of which bear L-rich NESS and are
ligands for CRM]. By interacting with pp32 and April, HuR can funnel ARE-containing
transcripts into the CRM] pathway. ARES are commonly found in transcripts of early
response genes, and access of HuR to two distinct export pathways might ensure rapid

expression of these genes despite a variety of cellular conditions.

29

[11. Conclusions

The field of nuclear transport has progressed considerably in terms of descriptive
and mechanistic studies. Descriptive studies have documented (nuclear import, export, or
shuttling of many substrates, and in many cases mechanistic studies have identified the
machinery responsible for these activities. However, the vast majority of studies to date
have examined the trafficking of substrates individually. Thus, there is no notion of the
“scale” of nuclear transport. How many substrates undergo nuclear import or export?
How many import and export receptors are there? Questions such as these represent the
next level of study for the field of nuclear transport: to understand the total number and
identity of substrates that traffic between the nucleus and cytoplasm. Such knowledge
will facilitate the placement of nuclear transport into the context of multiple biological
pathways, and will illuminate the functional significance of nuclear transport at the level

of a whole cell.

30

References

l. Krzeslak, A. and Lipinska, A. (2004) Cell. Mol. Biol. Lett. 9(2), 305-328.
2. Barondes, S.H. et al. (1994) Cell 76, 597-598.

3. Cooper, D.N.W., and Barondes, SH. (1999) Glycobiology 9, 979-984.

4. Hirabayashi, J ., et al. (2002) Biochim. Biophys. Acta. 1572, 232-254.

5. Wang, J .L., Gray, R.M., Haudek, KC, and Patterson, R]. (2004) Biochim. Biophys.
Acta. 1673, 75-93.

6. Agrwal, N., Sun, Q., Wang, S.-Y., and Wang, J .L. (1993) J. Biol. Chem. 268(20),
14932-14939.

7. Birdsall, B., Feeney, J ., Burdett, I.D.J., Bawumia, S., Barboni, E.A.M., and Hughes,
RC. (2001) Biochemistry 40, 4859-4866.

8. Seetharaman, J ., Kanigsberg, A., Slaaby, R., Leffler, H., Barondes, SH, and Rini,
J.M. (1998).]. Biol. Chem. 273(21), 13047-13052.

9. Morris, 8., Ahmad, N., Andre, S., Kaltner, H., Gabius, H.J., Brenowitz, M., and
Brewer, C.F. (2004) Glycobiology 14, 293-300.

10. Ahmad, N., et a1. (2004).]. Biol. Chem. 279, 10841-10847.

11. Moutsatsos, I.K., Davis, J.M., and Wang, J.L. (1986) J. Cell Biol. 102, 477-483.
12. Laing, J.G. and Wang, J .L. (1988) Biochemistry 27, 5329-5334.

13. Hubert, M. and Wang, S.Y. (1995) Exp. Cell Res. 220, 397-406.

14. Craig, S.S., Krishnaswamy, P., Irani, A.M., Kepley, C.L., Liu, F.-T., and Schwartz,
LB. (1995) Anat. Rec. 242, 211-219.

15. Moutsatsos, I.K., Wade, M., Schindler, M., and Wang, J .L. (1987) Proc. Natl. Acad.
Sci. USA. 84, 6452.

16. Hamman, K.K., Cowles, E.A., Wang, J .L., and Anderson, R.L. (1991) Exp. Cell Res.
196, 82-91.

17. Openo, K.P., Kadrofske, M.M., Patterson, R.J., and Wang, J.L. (2000) Exp. Cell Res.
255, 278-290.

31

18. Dagher, S.F., Wang, J .L., and Patterson, R.J. (1995) Proc. Natl. Acad. Sci. USA. 92,
1213-1217.

19. Park, J.W., Voss, P.G., Grabski, S., Wang, J .L., and Patterson, R.J. (2001) Nucleic
Acids Res. 29, 3595-3602.

20. Charroux, B., Pellizzoni, L., Perkinson, R.A., Yong, J., Shevchenko, A., Mann, M.,
and Dreyfuss, G. (2000) J. Cell Biol. 148, 1177-1186.

21. Fischer, U., Qing, L., and Dreyfuss, G. (1997) Cell 90, 1023-1029.
22. Pellizzoni, L., Naoyuki, K., Charroux, B., and Dreyfuss, G. (1998) Cell 95, 615-624.
23. Paron, I., et al. (2003) Biochem. Bi0phys. Res. Commun. 302, 545-553.

24. Seve, A.P., F elin, M., Doyennette-Moyne, M.A., Sahraoui, T., Aubery, M., and
Hubert, J. (1993) Glycobiology 3, 23-30.

25. Rousseau, C., Muriel, M.-P., Musset, M., Botti, J ., and Séve, A.-P. (2000) J. Cell.
Biochem. 78, 638-649.

26. Yang, R.-Y., Hsu, D.K., and Liu, F.-T. (1996) Proc. Natl. Acad. Sci. USA. 93, 6737-
6742.

27. Akahani, S., Nangia-Makker, P., Inohara, H., Kim, HR, and Raz, A. (1997) Cancer
Res. 57, 5272-5276.

28. Yu, F ., Finley, R.L., Raz, A., and Kim, H.-R.C. (2002) J. Biol. Chem. 277, 15819-
15827.

29. Liu, F.-T., Patterson, R.J., and Wang, J .L. (2002) Biochim. Biophys. Acta. 1572, 263-
273.

30. Gdrlich, D. and Kutay, U. (1999) Annu. Rev. Cell Dev. Biol. 15, 607-660.
31. Nigg, EA. (1997) Nature 386, 779-787.

32. Chook, Y.M. and Blobel, G. (2001) Curr. 0p. Struc. Biol. 11, 703-715.
33. Bednenko, J ., Cingolani, G., and Gerace, L. (2003) Trafiic 4, 127-135.
34. Bayliss, R., Corbett, A.H., and Stewart, M. (2000) T raflic 1, 448-456.

35. Nakielny, S. and Dreyfuss, G. (1999) Cell 99, 677-690.

36. Weis, K. (2002) Curr. 0p. Cell Biol. 14, 328-335.

32

37. Fahrenkrog, B., Koser, J., and Aebi, U. (2004) Trends Biochem. Sci. 29(4), 175-182.
38. Powers, MA. and Dasso M. (2004) Nature Cell Biol. 6(2), 82-86.

39. Izaurralde, E. and Adam, S. (1998) RNA 4, 351-364.

40. Adam, SA. (1999) Curr. 0p. Cell Biol. 11, 402-406.

41. Gtirlich, D., Panté, N., Kutay, U., Aebi, U., and Bischoff, F .R. (1996) EMBO J.
15(20), 5584-5594.

42. Kalderon, D., Roberts, B.L., Richardson, W.D., and Smith, AB. (1984) Cell 39, 499-
509.

43. Dingwall, C., Shamick, S.V., and Laskey, RA. (1982) Cell 30, 449-458.
44. Gdrlich, D., Prehn, S., Laskey, RA, and Hartmann, E. (1994) Cell 79, 767-778.

45. Gdrlich, D., Vogel, F ., Mills, A.D., Hartmann, E., and Laskey, RA. (1995) Nature
377, 246-248.

46. Radu, A., Blobel, G, and Moore, MS. (1995) Proc. Natl. Acad. Sci. USA. 92, 1769-
1773.

47. Imamoto, N. et al. (1995) EMBOJ. 14, 3617-3626.

48. Imamoto, N., Tachibana, T., Matsubae, M. and Yoneda, Y.A. (1995).]. Biol. Chem.
270, 8559-8565.

49. Gdrlich, D., et al. (1995) Curr. Biol. 5, 383-392.

50. Weis, K., Mattaj, I.W., and Lamond, A.I. (1995) Science 268, 1049-1053.

51. Chi, N.C., Adam, E.J., and Adam, SA. (1995) J. Cell Biol. 130, 265-274.

52. Mtihlhaiisser, P, Muller, E.C., Otta, A., and Kutay, U. (2001) EMBO Rep. 2, 690-696.

53. Ribbeck, K., Lipowsky, G., Kent, H.M., Stewart, M., and Gbrlich, D. (1998) EMBO
J. 17(22), 6587-6598.

54. Kohler, M. et al. (1999) Mol. Cell Biol. 19(11), 7782-7791.
55. Romanelli, MG. and Morandi, C. (2002) Eur. J. Biochem. 269, 2727-2734.

56. Dingwall, C. and Laskey, RA. (1998) Curr. Biol. 8, R922-R924.

33

57. Jikel, S. and Gbrlich, D. (1998) EMBO J. 17, 4491-4502.
58. 58, D., Gbrlich, D., Laemmli, UK, and Adachi, Y. (1999) EMBOJ. 18, 4348-4358.

59. Huber, J ., Cronshagen, U., Kadokura, M., Marshallsay, C., Wada, T., Sekine, M., and
Luhrmann, R. (1998) EMBO J. 17, 4114-4126.

60. Will, C.L. and Liihrmann, R. (2001) Curr. 0p. Cell Biol. 13, 290-301.

61. Jakel, S., Albig, W., Kutay, U., Bischoff, F.R., Schwamborn, K., Doenecke, D., and
Gdrlich, D. (1999) EMBO J. 18, 2411-2423.

62. Moore, MS. and Blobel, G. (1994) Proc. Natl. Acad. Sci. USA. 91, 10212-10216.
63. Paschal, B.M. and Gerace, L. (1995) J. Cell Biol. 129, 925-937.

64. Fischer, U., Huber, J., Boelens, W.C., Mattaj, I.W., and Li'rhrmann, R. (1995) Cell 82,
475-483.

65. Wen, W., Meinkoth, J .L., Tsien, R.Y., and Taylor, SS. (1995) Cell 82, 463-473.

66. Lei, E.P and Silver, PA. (2002) Dev. Cell 2, 261-272.

67. Fomerod, M., Ohno, M., Yoshida, M., and Mattaj, I.W. (1997) Cell 90, 1051-1060.
68. Ossareh-Nazari, B., Bachelerie, F., and Dargemont, C. (1997) Science 278, 141-144.

69. Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J .D. (1994)
Molecular Biology of the Cell. Garland Publishing, 740-741.

70. Fantozzi, D.A., Harootunian, A.T., Wen, W., Taylor, S.S., Feramisco, J .R., Tsien,
R.Y., and Meinkoth, J .L. (1994).]. Biol. Chem. 269(4), 2676-2686.

71. Wiley, J.C., Wailes, L.A., Idzerda, R.L., and McKnight, GS. (1999).]. Biol. Chem.
274(10), 6381-6387.

72. Paraskeva, E. et al. (1999) J. Cell Biol. 145(2), 255-264.

73. Herold, A., Truant, R., Wiegand, H., and Cullen, BR. (1998) J. Cell Biol. 143(2),
309-318.

74. Kutay, U., Bischoff, F.R., Kostka, S., Kraft, R., and Gdrlich, D. (1997) Cell 90, 1061-
1071.

75. Lipowsky, G. et al. (2000) EMBOJ. 19(16), 4362-4371.

34

76. Gallouzi, I.-E. and Steitz, J .A. (2001) Science 294, 1895-1901.
77. Stiiven, T., Hartmann, E., and Gbrlich, D. (2003) EMBOJ. 22(21), 5928-5940.

78. Ho, J.H.-N., Kallstrom, G., and Johnson, A.W. (2000).]. Cell Biol. 151(5), 1057-
1066.

79. Gadal, O., StrauB, D., Kessl, J., Trumpower, B., Tollervey, D., and Hurt, E. (2001)
Mol. Cell. Biol. 21(10), 3405-3415.

80. Herold, A., Teizeira, L., and Izaurralde, E. (2003) EMBOJ. 22(10), 2472-2483.
81. Huang, Y., Gattoni, R., Stévenin, J ., and Steitz, J .A. (2003) Mol. Cell 11, 837-843.

82. Lund, E., Giittinger, S., Calado, A., Dahlberg, J .E., and Kutay, U. (2004) Science 303,
95-98.

83. Yi, R., Qin, Y., Macara, LG, and Cullen, BR. (2003) Genes Dev. 17, 3011-3016.
84. Lee, S.-H. and Hannink, M. (2001) J. Biol. Chem. 276(26), 23599-23606.
85. McBride, K., McDonald, C., and Reich, NC. (2000) EMBO J. 19(22), 6196-6206.

86. Taagepera, S., McDonald, D., Loeb, J .E., Whitaker, L.L., McElroy, A.K., Wang,
J.Y.Y., and Hope, T.J. (1998) Proc. Natl. Acad. Sci. USA. 95, 7457-7462.

87. Fan, x.c. and Steitz, J.A. (1998) EMBOJ. 17, 3448-3460.

88. Kutay, U., Lipowsky, G., Izaurralde, E., Bischoff, F.R., Schwarzmaier, P., Hartmann,
E., and Gdrlich, D. (1998) Mol. Cell 1, 359-369.

89. Erkmann, J.A. and Kutay, U. (2004) Exp. Cell Res. 296, 12-20.
90. Reed, R. and Hurt, E. (2002) Cell 108, 523-531.

91. Huang, Y. and Steitz, J .A. (2001) Mol. Cell 7, 899-905.

35

Chapter 2

Shuttling of Galectin-3 between the Nucleus and Cytoplasm

36

Shuttling of Galectin-3 between the Nucleus and Cytoplasm

Peter J. Davidson‘, Michael J. Davisz, Ronald J. Pattersonz, .
Marie-Anne Ripoche3, Francoise Poirier", and John L. Wang5

1Cell and Molecular Biology Program, 2Department of Microbiology, 5Department of
Biochemistry, Michigan State University, East Lansing, MI 48824 USA;

3INSERM 257, ICGM, Paris, France, 4Institut Jacques Monod, 75251 Paris Cedex 05,
France.

Peter Davidson was responsible for all aspects of the experimental work reported in the
figures and tables of this chapter. This included the design of the experiments, data
collection, and interpretation. Michael Davis, an undergraduate at Michigan State,
carried out the experiments using cells derived from the galectin—null mutant mice. This
was done under the direct supervision of Peter Davidson. Drs. M.-A. Ripoche and F.
Poirier generated the original galectin-null mice and provided the fibroblasts used in the
present experiments. Drs. Ron Patterson and John Wang are faculty members in whose
laboratories the experiments were performed.

37

 

Abstract

In previous studies, we documented that galectin-3 (Mr ~ 30,000) is a pre-mRN A
splicing factor. Recently, galectin-3 was identified as a component of a nuclear and
cytoplasmic complex, the SMN complex, through its interaction with Gemin4. To test
the possibility that galectin-3 may shuttle between the nucleus and the cytOplasm, human
fibroblasts (LG-l) were fused with mouse fibroblasts (3T3). The monoclonal antibody
NCL-GAL3, which recognizes human galectin-3 but not the mouse homologue, was used
to monitor the localization of human galectin-3 in heterodikaryons. Human galectin-3
localized to both nuclei of a large percentage of heterodikaryons. Addition of the
antibiotic leptomycin B, which inhibits nuclear export of galectin-3, decreased the
percentage of heterodikaryons showing human galectin-3 in both nuclei. In a parallel
experiment, mouse 3T3 fibroblasts, which express galectin-3, were fused with fibroblasts
derived fiom a mouse in which the galectin-3 gene was inactivated. Mouse galectin-3
localized to both nuclei of a large percentage of heterodikaryons. Again, addition of
leptomycin B restricted the presence of galectin-3 to one nucleus of a heterodikaryons.
The results from both heterodikaryon assays suggest that galectin-3 can exit one nucleus,
travel through the cytoplasm, and enter the second nucleus, matching the definition of

shuttling.

Introduction
Galectin-3 is a member of a family of proteins, defined on the basis of structural
analysis and binding specificity studies, that: (a) bind B-galactosides; and (b) share

Significant sequence Similarity in the carbohydrate-binding site (1). On the basis of

38

 

immunofluorescence and immunoelectron microsc0py, as well as quantitative
immunoblotting of subcellular fractions, it has been documented that galectin-3 was
localized in the nucleus and cytoplasm of cells (2; 3; 4). Using a cell-free assay for the
splicing of pre-mRNA, we have previously shown that galectin-3 is a required splicing
factor (5). More recently, we have found that galectin-3 interacts with Gemin4 (6), a
component of a macromolecular complex designated as the SMN (survival of motor
neuron) complex (7). Moreover, co-immunoprecipitation experiments demonstrated that
galectin-3 is a bonafide member of the SMN complex. The SMN complex fimctions in
both the nucleus and the cytoplasm. In the cytoplasm, the SMN complex mediates
assembly of small nuclear ribonucleoprotein (snRNP) particles; in the nucleus, it delivers
the snRNPs to the pre-mRNA during the early stages of spliceosome formation (8; 9).
Nucleocytoplasmic shuttling is typically defined as the repeated bi-directional
movement of a protein across the nuclear membranes (10). The association of galectin-3
with the SMN complex raises the possibility that galectin-3 might perform related
functions in both the nucleus and the cytoplasm and that it might shuttle between the two
compartments. To test directly whether galectin-3 can Shuttle, we monitored the
localization of galectin-3 in two different types of heterokaryon systems: (a) human-
mouse heterodikaryons; and (b) mouse-mouse heterodikaryons, in which one of the cell
types contained a null mutation in the galectin-3 gene (11). In the human-mouse
heterodikaryons, we monitored the localization of human galectin-3 using a mouse
monoclonal antibody that recognized human galectin-3 but not the mouse homologue. In

the mouse-mouse heterodikaryons, we monitored the localization of mouse galectin-3

39

 

using a rat monoclonal antibody directed against galectin-3. The results obtained from

both systems suggest that galectin-3 does shuttle between the nucleus and the cytoplasm.

Results

The monoclonal antibody NCL-GAL3 recognizes human galectin-3 but not the mouse
homologue

The experimental strategy used to test for shuttling of galectin-3 is depicted in
Figure 1. Heterokaryons were generated through fusion of two cell types: human LG-l
fibroblasts (cell X) and mouse 3T3 fibroblasts (cell Y). The presence of human galectin-
3 in both nuclei of a heterodikaryon (path I) could arise due to nuclear import from three
sources: (a) human galectin-3 in the cytoplasm; (b) newly synthesized human galectin-3;
and (c) human galectin-3 from the human nucleus. The last of these three mechanisms
requires export of human galectin-3 from the human nucleus into the cytoplasm of the
heterodikaryon and subsequent import into the mouse nucleus. Therefore, if shuttling
makes a significant contribution to the percentage of heterodikaryons exhibiting human
galectin-3 staining in both nuclei, it should be sensitive to inhibition of nuclear export.
On this basis, we expected the percentage of heterodikaryons Showing human galectin-3
in both nuclei to be decreased and, correspondingly, the percentage of heterodikaryons
showing galectin-3 in only one nucleus (path 11) to be increased by inhibitors of nuclear

export. This experimental scheme has two key requirements. First, we needed to be able

4O

 

Figure 1. Schematic illustrating the use of heterokaryons to study shuttling.
Heterokaryons were generated through fusion of two cell types, for example: human LG-
1 (cell X) and mouse 3T3 fibroblasts (cell Y). The presence of human galectin-3 in both
nuclei of a heterodikaryon (path I) could be due to three mechanisms: (3) human galectin-
3 originally in the cytoplasm of LG-l cells; (b) newly synthesized human galectin-3; and
(c) galectin-3 originally from the human nucleus. Mechanism (c) requires that human
galectin-3 first gets exported from the human nucleus into the cytoplasm of the
heterodikaryon and subsequently gets imported into the mouse nucleus. Path 11 depicts a
heterodikaryon in which only one of the nuclei contains human galectin-3. This general
scheme was also applied to the second cell pair in our test of galectin-3 shuttling: mouse
3T3 fibroblasts expressing galectin-3 (cell X) and MEF Gal-3 -/- cells not expressing

galectin-3 (cell Y).

41

Figure 1

[62%
Y

Fuse

Incubate

42

to distinguish heterodikaryons (human-mouse cell hybrids) from homodikaryons (human-
human and mouse-mouse cell hybrids). This was accomplished by tagging the two cell
types with distinguishable microsphere beads prior to fusion (see below). Second, we
needed to be able to immunostain for human galectin-3 with an antibody that fails to
recognize mouse galectin-3.

When lysates of human cells were subjected to immunoblotting with the mouse
monoclonal antibody NCL-GAL3, a single polypeptide (Mr ~29,000) corresponding to
the mobility of human galectin-3 was observed (Figure 2, panel A, lanes 1 and 2). In
contrast, no polypeptide band could be detected when NCL-GAL3 was used to blot a
corresponding lysate derived from mouse 3T3 fibroblasts (Figure 2, panel A, lane 3).
The presence of galectin-3 in the mouse cell lysate was ascertained by blotting with anti-
Mac-2, a rat monoclonal antibody directed against galectin-3 (Figure 2, panel B, lane 3).
The appearance of mouse galectin-3 at a higher molecular weight than human galectin-3
(Figure 2, panel B, lanes 1 — 3) is consistent with the difference in the lengths of the
respective polypeptide chains, as verified by cDNA sequencing. These results suggest
that the mouse monoclonal antibody NCL-GAL3 recognizes human galectin-3 but not the
mouse homologue.

This conclusion was substantiated by immunofluorescence staining of fixed and
permeabilized cells. NCL-GAL3 yielded strong staining in HeLa cells but not in 3T3
cells (Figure 3, panels A and C). Again, the presence of galectin-3 in both cell types was
verified by staining with anti-Mac-2 (Figure 3, panels E and G). In these experiments,
we also ascertained that the secondary antibodies used for these and all subsequent

immunofluorescence studies, FITC-conjugated goat anti-mouse IgG and FITC-

43

Figure 2. Western blotting for galectin-3 in lysates of human HeLa cells, human LG-
1 fibroblasts, and mouse 3T3 fibroblasts.

Lane 1: human HeLa lysate. Lane 2: human LG-l lysate. Lane 3: mouse 3T3 lysate. 50

pg of total protein was loaded in each lane. Panel A: galectin-3 was detected using the

mouse monoclonal antibody NCL-GAL3 (21 ng/ml) plus HRP-goat anti-mouse

immunoglobulin. Panel B: galectin-3 was detected using the rat monoclonal antibody

anti-Mac-2 (125 ng/ml) plus HRP-goat anti-rat immunoglobulin. The mobilities of

molecular weight standards are indicated.

44

Figure 2

—116—
_52_
'—35—
_30_

—22—’

45

Figure 3. Immunofluorescence staining for galectin-3 in fixed and permeabilized
human HeLa cells and mouse 3T3 fibroblasts.
Cells were fixed with paraformaldehyde (4%, room temperature, 20 min) and
permeabilized with Triton X-100 (0.2%, 4°C, 5 min). Panels A, B, E, and F: human
HeLa cells. Panels C, D, G, and H: mouse 3T3 cells. Panels A and C: galectin-3 was
detected with NCL-GAL3 (2.1 ug/ml) plus FITC-goat anti-mouse immunoglobulin.
Panels B and D: F ITC-goat anti-mouse immunoglobulin. Panels E and G: galectin-3 was

detected with anti-Mac-2 (25 ug/ml) plus FITC-goat anti-rat immunoglobulin. Panels F

and H: FITC-goat anti-rat immunoglobulin. Bar, 10 um.

46

Figure 3

 

conjugated goat anti-rat IgG, yielded negligible staining in human and mouse cells when
used in the absence of any primary antibodies (Figure 3, panels B, D, F, and H).
Bead-tagging distinguishes heterodikaryons from homodikaryons

Fusion of human and mouse cells was expected to produce three types of cell
hybrids: human-human, mouse-mouse, and human-mouse. In order to distinguish the
three types and focus only on human-mouse fusions, we tagged our cells with
microsphere styrene beads (~ 2 pm in diameter) prior to firsion. Human LG-l fibroblasts
were tagged with non-fluorescent (black) beads, whereas mouse 3T3 fibroblasts were
tagged with green fluorescent beads. Tagged cells were then co-cultured and fused as
depicted in Figure 1. Several hours post fusion, the cells were fixed and immunostained
with NCL-GAL3 and FITC-goat anti-mouse IgG to determine the localization of human
galectin-3 and the nuclei were stained with PI.

Monokaryons were immediately distinguished from di- or polykaryons by the
number of nuclei as observed by P1 fluorescence (not shown). As shown in Figure 4,
panels A — C, human-human homodikaryons were indicated by the presence of only non-
fluorescent beads in the cytoplasm. Similarly, mouse-mouse homodikaryons were
indicated by the presence of only fluorescent beads in the cytoplasm (Figure 4, panels D
— F). In contrast, the presence of both fluorescent and non-fluorescent beads within a di-
nucleated cell indicated a human-mouse heterodikaryon (Figure 4, panels G — 1). Thus,
the bead-tagging method proved to be a reliable technique for recognizing

heterodikaryons.

48

Figure 4. Bead-tagged homodikaryons and heterodikaryons immunostained for
human galectin-3.

Human LG-l fibroblasts were tagged with non-fluorescent (black) beads, and mouse 3T3
fibroblasts were tagged with green fluorescent beads. Following bead-tagging, both cell
types were co-cultured and fused. Fused cells were then fixed, permeabilized, and
immunostained for human galectin-3 using NCL-GAL3 (2.1 ug/ml) and FITC-
conjugated goat anti-mouse immunoglobulin. Panels A — B: a human-human
homodikaryon. Note the presence of only non-fluorescent (black) beads in the
cytoplasm. A cluster of black beads is indicated by the arrow. Panels C — D: a mouse-
mouse homodikaryon. Note the presence of only green fluorescent beads in the
cytoplasm. A single green fluorescent bead is indicated by the arrow head. Panels E — F:
a human-mouse heterodikaryon. Note the presence of both black and green beads in the
cytoplasm. The arrow indicates a single black bead, while the arrow head indicates a
single green bead. Panels A, C, and E: FITC fluorescence showing localization of human
galectin-3. Note the prominent nuclear and lesser cytoplasmic staining of human
galectin-3 in panels A and E. Panels B, D, and F: PI fluorescence showing the location of

both nuclei in each fusion. Bar, 10 um.

49

Figure 4

 

Localization of human galectin-3 to both nuclei of human-mouse heterodikaryons was
partially dependent upon de novo protein synthesis

Figure 4 also serves to illustrate how an individual nucleus was scored for the
presence of human galectin-3. In the mouse 3T3 homodikaryon (Figure 4, panel B), both
nuclei yielded fluorescence intensity values of less than 250. Based on our quantitative
scoring criteria (detailed in Materials and Methods), these nuclei would be negative in
terms of reactivity with the NCL-GAL3 antibody (i.e. no human galectin-3). In the
human LG-l homodikaryon (Figure 4, panel B), the two nuclei yielded fluorescence
intensity values of 1700 and 1800, far surpassing our threshold value of 300 for scoring
positive in terms of human galectin-3. In the human-mouse heterodikaryon (Figure 4,
panel H), the two nuclei yielded values of 760 and 900. Thus, human galectin-3 was
observed in both nuclei of this heterodikaryon (Figure 4, panel H), indicating that it was
redistributed into the mouse nucleus.

To assess the contribution of newly synthesized human galectin-3 in supplying
the mouse nucleus, cycloheximide (CHX) was employed to block de novo protein
synthesis (10; 12). CHX appeared to have two effects on the human-mouse
heterodikaryons. First, it decreased the overall fluorescence intensity due to NCL-GAL3
staining, presumably because the drug inhibited de novo synthesis of human galectin-3.
For example, in human-mouse heterodikaryons treated with CHX (Figure 5, panel E), the
values of fluorescence intensity for the two nuclei were 450 and 530, reflecting a further
decrease in fluorescence beyond the dilution of human galectin-3 into the two nuclei of
the heterodikaryon (described for Figure 4, panel H above; Figure 5, panel B).

Nevertheless, both of the nuclei of this heterodikaryon (Figure 5, panel E) were above the

51

Figure 5. Effect of cycloheximide and leptomycin B on localization of human
galectin-3 in human-mouse heterodikaryons.

Heterodikaryons were generated and immunostained as described in the legend to Figure
4. Panels A — B: Human-mouse heterodikaryons incubated in the absence of CHX and
LMB. Panels C — D: Human-mouse heterodikaryons incubated in the presence of CHX
(10 ug/ml) prior to fusion. Panels E — F: Human-mouse heterodikaryons incubated in the
presence of LMB (2 ng/ml; 3.8 nM) and CHX (10 ug/ml) prior to fusion. Panels A, C,
and E: F ITC fluorescence showing localization of human galectin-3. Note the staining
for human galectin-3 in both nuclei in panels A and C. In contrast, note the absence of
human galectin-3 staining in one of the nuclei in panel H and the accentuated staining in

the other nucleus. Panels B, D, and F: PI fluorescence showing the location of both

nuclei in each heterodikaryon. Bar, 10 um.

52

Figure 5

 

threshold value of 300 and therefore, were scored positive for human galectin-3,
indicating that there was still transport into the mouse nucleus.

Second, CHX also decreased the percentage of heterodikaryons exhibiting human
galectin-3 in both nuclei. In the absence of CHX (Figure 5, panels A-C), the fraction of
heterodikaryons exhibiting human galectin-3 in both nuclei nine hours post fusion was
72% (58 out of 81 heterodikaryons counted). In the presence of CHX (Figure 5, panels
D-F), the corresponding value was 50% (59 out of 119 heterodikaryons counted). Thus,
although de novo synthesized human galectin-3 contributed to the localization of the
protein in both nuclei of human-mouse heterodikaryons, at least half of the
heterodikaryons still exhibited human galectin-3 in both nuclei even in the presence of
CHX. We interpret these results to indicate that newly synthesized human galectin-3 was
not the only source for the mouse nucleus in the heterodikaryon. Rather, it appears that
the majority of human galectin-3 supplying the mouse nucleus came from either the
human nuclear pool or the human cytoplasmic pool. 2
Localization of human galectin-3 to both nuclei of human-mouse heterodikaryons was
dependent upon nuclear export

Using digitonin-permeabilized 3T3 fibroblasts, we had previously shown that
galectin-3 is rapidly and selectively exported from the nucleus and that this export was
sensitive to inhibition by leptomycin B (LMB) (13). Therefore, to distinguish between
the contributions of the nuclear pool and the cytoplasmic pool of human galectin-3, LMB
was employed to block nuclear export of galectin-3. LMB binds and inactivates the

CRMl (chromosome region maintenance) nuclear export receptor (14), which recognizes

54

the leucine-rich nuclear export signal. In Figure 5, panels G and H Show a representative
human-mouse heterodikaryon treated with CHX and LMB.

Nine hours after fusion, approximately 50% of heterodikaryons treated with CHX
alone exhibited human galectin-3 in both nuclei (59 out of 119 heterodikaryons counted,
Table 1). In the presence of CHX and LMB, the corresponding value was 19% (23 out of
120 heterodikaryons counted, Table 1). These results suggest that the cytoplasmic pool
of human galectin-3 was not the only source of the protein for the mouse nucleus.
Rather, human galectin-3 in the human nucleus must also contribute to supplying the
mouse nucleus, since inhibition of export of human galectin-3 from the human nucleus
concomitantly reduced the proportion of heterodikaryons exhibiting human galectin-3 in
both nuclei. On this basis, we conclude that at least some of the human galectin-3
appearing in the mouse nucleus of a heterodikaryon had to first exit the human nucleus
into the common cytoplasm followed by entry into the mouse nucleus, thus fulfilling the
definition of shuttling.

Mouse galectin-3 shuttles in 3T 3-MEF Gal-3 -/- heterodikaryons

In previous studies (11), the generation of mice carrying a null mutation of the
galectin-3 gene was reported. This provided the opportunity to corroborate our test of
galectin-3 Shuttling in another system. We isolated fibroblasts from embryos of this
strain of mice, designated MEF Gal-3 -/-.

To confirm the absence of the galectin-3 polypeptide in MEF Gal-3 -/- cells, we
blotted cell lysate from MEF Gal-3 -/- cells with the anti-Mac-2 antibody. We did not
detect any polypeptide in the lane containing MEF Gal-3 -/- lysate (Figure 6A, lane 2).

In contrast, immunoblotting with anti-Mac-2 yielded a band of the same mobility as

55

Figure 6. Panel A: Western blotting for galectin-3 in lysates of MEF Gal-3 -/-
fibroblasts, MEF Gal-l -/- fibroblasts, MEF WT fibroblasts, and mouse 3T3
fibroblasts.

Galectin-3 was detected using anti-Mac-Z (125 ng/ml) and HRP-conjugated goat anti-rat
immunoglobulin. Lane 1: recombinant mouse galectin-3, 50 ng. Lane 2: MEF Gal-3 -/-

lysate, 6.5 pg total protein. Lane 3: MEF Gal-1 -/- lysate, 6.5 pg total protein. Lane 4:

MEF WT lysate, 6.5 pg total protein. Lane 5: mouse 3T3 lysate, 6.5 pg total protein.
Panel B: Immunofluorescence staining for galectin-3 in fixed and permeabilized

mouse 3T3 fibroblasts and MEF Gal-3 -/- fibroblasts.

Cells were prepared as described in the legend to Figure 3. Galectin-3 was detected with

anti-Mac-2 (25 pg/ml) and FITC-conjugated goat anti-rat immunoglobulin. Panels A and

B: mouse 3T3 cells. Panels C and D: bead-tagged MEF Gal-3 -/- cells. Panels A and C:

anti-Mac-2 and FITC-conjugated goat anti-rat immunoglobulin. Panels B and D: FITC-

conjugated goat anti-rat immunoglobulin. Bar, 10 pm.

56

Figure 6

—173
- 61

-36

-"

-13

12345

 

57

recombinant galectin-3 (Figure 6A, lane 1) in lysates from MEF Gal-1 -/- cells (MEF
cells containing a null mutation in the Gal-1 gene) (15), MEF WT cells (MEF “wild-
type” cells, without any mutations), and mouse 3T3 cells (Figure 6A, lanes 3 - 5).

These results indicated that the MEF Gal-3 -/- cells did indeed lack the galectin-3
polypeptide. This conclusion is substantiated by immunofluorescence staining of fixed
and permeabilized cells. Anti-Mac-2 stained mouse 3T3 cells but not MEF Gal-3 -/- cells
(Figure 6B, panels A and C). In these experiments, we also ascertained that the
secondary antibody FITC-conjugated goat anti-rat IgG did not react non-specifically with
fixed and permeabilized MEF Gal-3 -/- cells (Figure 6B, panel D) when used in the
absence of anti-Mac-Z.

The experimental scheme of Figure 1 was used to study galectin-3 shuttling in
3T3-MEF Gal-3 -/- heterodikaryons. In this case, 3T3 cells served as the source of
galectin-3 (cell X) and were tagged with non-fluorescent (black) beads. The MEF Gal-3
/- cells served as the recipient of galectin-3 (cell Y) and were tagged with fluorescent
green beads. The localization of galectin-3 in these heterodikaryons was monitored by
immunofluorescence using anti-Mac-2. The results of Shuttling assays conducted in 3T3-
MEF Gal-3 -/- heterodikaryons were generally similar to those obtained from the assays
in human-mouse heterodikaryons. Four hours after fusion, 97% (47 out of 48) of
heterodikaryons treated with CHX prior to fusion exhibited galectin-3 in both nuclei. In
the presence of CHX and LMB, the corresponding value was 18% (16 out of 89
heterodikaryons counted, Table 1). In Figure 7, panels A - C Show a typical 3T3-MEF
Gal-3 -/- heterodikaryon treated with CHX, while panels D - F show a typical 3T3-MEF

Gal-3 -/- heterodikaryon treated with CHX and LMB. These results suggest that the

58

Table 1. Percent of heterodikaryons showing staining for galectin-3 in both nuclei.
Experiment A: Human-mouse fusion. Mouse monoclonal antibody NCL-GAL3 used to
detect human galectin-3. Experiment B: 3T3-MEF Ga13 -/- fusion. Rat monoclonal

antibody anti-Mac-Z used to detect mouse galectin-3.

59

Table 1
Percent of heterodikaryons showing staining for galectin-3 in both nuclei.

Experiment Harvest Time CHX CHX _+ LMB
A 9 50% (59/119) 19% (23/ 120)
B 4 97% (47/48) 18% (16/89)

60

Figure 7. Effect of cycloheximide and leptomycin B on localization of galectin-3 in
mouse 3T3-MEF Gal-3 -/- heterodikaryons.

Mouse 3T3 fibroblasts were tagged with non-fluorescent (black) beads, and MEF Gal-3 -
/- fibroblasts were tagged with green fluorescent beads. Following bead-tagging, both
cell types were co-cultured and fused. Fused cells were then fixed, permeabilized, and
immunostained for galectin-3 using anti-Mac-2 (25 pg/ml) and FITC-conjugated goat
anti-rat immunoglobulin. Panels A — B: 3T3-MEF Gal-3 -/- heterodikaryons incubated in
the presence of CHX (10 pg/ml) prior to fusion. Panels C — D: 3T3-MEF Gal-3 -/-
heterodikaryons incubated in the presence of LMB (2 ng/ml; 3.8 nM) and CHX (10
pg/ml) prior to fusion. Panels A and C: F ITC fluorescence Showing localization of
galectin-3. Note galectin-3 staining in both nuclei in panel A. In contrast, note the
absence of galectin-3 staining in one of the nuclei in panel C. Panels B and D: PI

fluorescence showing the location of both nuclei in each heterodikaryon. Bar, 10 pm.

61

Figure 7

 

62

localization of galectin-3 to both nuclei of 3T3-MEF Gal-3 -/- heterodikaryons was

independent of protein synthesis, but dependent upon nuclear export of galectin-3.

Discussion

The key findings of this study include: (a) Human galectin-3 was observed in both
nuclei of human-mouse heterodikaryons, and mouse galectin-3 was observed in both
nuclei of 3T3-MEF Gal-3 -/- heterodikaryons. In both cases, galectin-3 had been
imported into recipient nuclei that previously lacked it. (b) The addition of CHX caused
a reduction of approximately 20% in the fraction of human-mouse heterodikaryons
showing human galectin-3 in both nuclei. However, 50% of heterodikaryons still showed
staining for human galectin-3 in both nuclei. (c) The addition of CHX and LMB caused a
further reduction of approximately 30% in the fi'action of human-mouse heterodikaryons
Showing human galectin-3 staining in both nuclei. In the 3T3-MEF Gal-3 -/-
heterodikaryons, the addition of CHX and LMB caused a reduction of approximately
79% in the fraction of heterodikaryons showing galectin-3 staining in both nuclei. We
conclude from these results that galectin-3 indeed undergoes nucleocytoplasmic shuttling.
Our previous documentation of nuclear export of galectin-3, carried out in a
permeabilized cell system, demonstrated that galectin-3 fulfilled a necessary condition
for shuttling (13). We have now directly demonstrated shuttling of galectin-3 in an in
vivo assay.

Interspecies heterokaryons have been used to demonstrate nucleocytoplasmic
shuttling of several other proteins, including nucleolin and B23/No38 (10), hnRNP A1

(16), the hdm2 oncoprotein (17), and the 2A7 antigen (18). However, these proteins

63

differ from galectin-3 in that they are typically most prominent in the nucleus, whereas
galectin-3 typically exhibits both nuclear and cytoplasmic localization. In the case of a
predominantly nuclear protein, two pools of protein could supply a recipient nucleus in a
heterokaryon assay: the nuclear pool and the newly synthesized pool. In contrast, when
the protein is also present in the cytoplasm, as is the case for galectin-3, then the
cytoplasmic pool represents a third pool that can also supply the recipient nucleus.

The presence of a cytoplasmic pool of galectin-3 complicated our analysis of
galectin-3 shuttling, in that we had to differentiate between the contributions of
cytoplasmic galectin-3 and galectin-3 that was exported from the nucleus. Therefore, we
employed the antibiotic LMB to block nuclear export of galectin-3. By conducting
shuttling assays in the presence of CHX (inhibiting protein synthesis), as well as in the
presence of CHX and LMB (inhibiting protein synthesis and nuclear export), we were
able to assess the contribution of each of the three pools of galectin-3 to the recipient
nuclei in our heterokaryons. Indeed, our data from the human-mouse heterodikaryon
assays suggest that all three pools did in fact contribute to supplying the mouse nucleus in
roughly equal proportions.

The effect of LMB on the nuclear versus cytoplasmic distribution of galectin-3 has
been documented for both of the cell types used in our human-mouse heterokaryon
fusions. Incubation of live, intact mouse 3T3 fibroblasts in the presence of LMB (3.8
nM) resulted in the accumulation of galectin-3 in the nucleus, as revealed by accentuation
of the nuclear staining (13). At a higher concentration of LMB (15.2 nM), galectin-3 was
found predominantly, if not exclusively, in the nucleus. Similar results were obtained for

human LG] cells; incubation with LMB (15.2 nM) resulted in the accumulation of

64

 

galectin-3 in the nucleus, as revealed by the almost exclusively nuclear staining (19).
These results indicate that the effect of LMB on decreasing the appearance of galectin-3
in both nuclei of heterokaryons must be due to its inhibitory effect on the export of
galectin-3 from the donor nucleus, rather than any secondary effects of the drug on
nuclear import.

Other shuttling proteins that appear to bear functional similarity to galectin-3 are
nucleolin and hnRNP Al. Both of these proteins are associated with RNA processing,
and both are believed to leave the nucleus in association with their substrate RNA
molecule. Nucleolin imports ribosomal proteins from the cytoplasm into the nucleus and
then coordinates binding of the ribosomal proteins to the nascent rRNA transcript (20;
21). The resulting complex may mediate processing of the transcript or stimulate
cleavage reactions, after which nucleolin exports assembled ribosomal subunits out of the
nucleus and deposits them in the cytoplasm (20). hnRNP A1 mediates nuclear export of
mRNA from the nucleus (16). Pre-mRNAS are bound by hnRNP A1 in the nucleus, and
upon completion of splicing, the nuclear export signal in hnRNP A1 directs the export of
the hnRNP Al-mRNA complex to the cytoplasm (22; 23). Once in the cytoplasm,
hnRNP A1 dissociates from the mRNA and returns to the nucleus to repeat the cycle.

The significance of trafficking of galectin-3 between the nuclear and cytoplasmic
compartments is not fully understood. We have recently identified galectin-3 as a
component of a macromolecular complex, designated as the SMN complex (6). Like
previous immunofluorescence and ultrastructural studies on galectin-3 (2; 3), the SMN
complex is found in both the cytoplasm and the nucleus (24; 25). In the cytoplasm, the

SMN complex is associated with the core proteins of snRNPs and is involved in the

65

 

biogenesis of the snRNP particles (8; 26). In the nucleus, the SMN complex is found in
discrete nuclear bodies called gems (gemini of coiled bodies) and it is required for
supplying snRNPs to the H-complex during spliceosome assembly (9; 27). This H-
complex juncture is also where galectin-3 appears to be required for splicing, as
demonstrated by the effect of galectin depletion on the accumulation of intermediates
during the assembly of splicing complexes (5).

We propose that galectin-3 might initially associate with Gemin4 in the
cytoplasm, possibly during the course of snRNP biogenesis. When the SMN complex is
imported into the nucleus, galectin-3 may be taken along by way of its interaction with
Gemin4. Once in the nucleus, galectin-3 may participate with the SMN complex in the
assembly of the spliceosome. Lastly, galectin-3 may be exported from the nucleus, in
association with Gemin4 and possibly other members of the SMN complex, to repeat the

cycle of snRNP biogenesis and delivery.

Materials and Methods

Cell Culture and reagents

NIH mouse 3T3 fibroblasts were obtained from the American Type Culture
Collection and cultured in MEM-ASP (minimal essential Eagle’s medium, supplemented
with 0.2 mM L-aspartic acid, 0.2 mM L-serine, 1 mM sodium pyruvate, 100 U/ml
penicillin, and 0.1 pg/ml streptomycin) plus 10 % calf serum at 37°C and 5% C02.

The human fibroblast strain, designated LG-l (28), was a gift from Drs. J. J.
McCormick and V. M. Maher at Michigan State University. LG-l cells were used

through passage 20, and were serially passaged at a split ratio of 1:4 as described

66

 

previously (19) in MEM-ASP supplemented with 10% fetal calf serum at 37°C and 5%
C02.

Primary mouse embryonic fibroblasts (MEF) were derived from 129 wild type
strain (MEF WT), galectin-1 null mutant strain (MEF Gal-1 -/-) (15) and galectin-3 null
mutant stain (MEF Gal-3 -/-) (11) using the procedure described by Hogan et al. (29).
MEF WT, MEF Gal-1 -/-, and MEF Gal-3 —/- cells were cultured in DME-HG
(Dulbecco’s Modified Eagle Medium with 4.5 g/l glucose, supplemented with 44 mM
sodium bicarbonate, 100 U/ml penicillin, 0.1pg/ml streptomycin, 50 pg/ml gentamicin
sulfate) plus 10% calf serum at 37°C and 5% C02. Cells were passaged serially at split
ratios of 1:5 or 1:10.

The rat monoclonal antibody anti-Mac-Z recognizes an epitope in the amino-
terminus of galectin-3 (30). The mouse monoclonal antibody NCL-GAL3 was purchased
from Novocastra Laboratories, Ltd., UK. Fluorescein isothiocyanate (F ITC)-conjugated
goat anti-rat IgG was obtained from Boehringer Mannheim, and FITC-conjugated goat
anti-mouse IgG was obtained from Santa Cruz Biotech. Microsphere styrene beads were
obtained from Polysciences.

Bead-tagging and polyethylene glycol-mediated cell fusion

Human LG-l and mouse 3T3 fibroblasts were seeded separately in 100 X 20 mm
tissue culture plates at 5 X 103 cells/cmz, and allowed to attach for approximately Six
hours. Microsphere styrene beads were then added to the culture medium at a
concentration of 250 beads/cell and incubated overnight. Plates were then rinsed with
Versene (140 mM NaCl, 2.68 mM KCl, 10 mM Na2HP04, 1.47 mM KH2P04, 0.68 mM

EDTA, 0.15% phenol red, pH 7.2), trypsinized, and resuspended separately. The

67

 

concentration of each cell suspension was determined, and both cell types were seeded
together, each at a density of 3.9 X 103 cells/cmz, into 35 mm dishes containing sterile
glass coverslips. Co-cultured cells were then incubated ovemight. Cycloheximide
(Sigma) was added at 10 pg/ml in MEM-ASP one hour prior to fusion, removed during
fusion, and returned after fusion. LMB was a gift from Dr. Minoru Yoshida (University
of Tokyo), and was added at 2 ng/ml (3.8 nM) in MEM-ASP ten hours prior to fusion,
removed during fusion, and returned after fusion.

Polyethylene glycol-1000 (F luka Chemical Corp.) was melted and diluted to 45%
(v/v) with warmed serum-free MEM-ASP media, then filtered through a 0.22 pm Mill—ex
syringe driven filter unit (Millipore). Co-cultured LG-l and NIH 3T 3 cells were rinsed in
warmed phosphate-buffered saline (PBS, 140 mM NaCl, 2.68 mM KCl, 10 mM
Na2HPO4, 1.47 mM KH2PO4, pH 7.4), and treated with 45% polyethylene glycol-1000
for 55 seconds. The medium was promptly removed and the cells were rinsed gently
three times in warmed PBS. Cells were then incubated with MEM-ASP containing 10%
fetal calf serum until immunostaining.
Immunostaining and fluorescence microscopy

Cells were prepared for fluorescence microscopy following fixation with 4%
paraformaldehyde and permeabilization with 0.2% Triton X-100 (13). Human-mouse
fusions were stained with NCL-GAL3 (2.1 pg/ml in PBS containing 0.2% gelatin) for
one hour at room temperature, then washed three times with T-TBS (10 mM Tris, pH 7.5,
0.5 M NaCl, 0.05% Tween 20) on an orbital shaker at room temperature. The coverslips
were then stained with FITC-goat anti-mouse IgG (in PBS containing 0.2% gelatin) for

30 minutes at room temperature. This and all subsequent steps were carried out in the

68

dimmest light possible. The coverslips were washed with T-TBS, stained with propidium
iodide (PI, 32 pg/ml in PBS containing 0.2% gelatin) for 30 minutes at room temperature,
and rinsed again in T-TBS. The coverslips were then inverted onto clean microscope
slides with a drop of Perma-Fluor (Lipshaw Immunon) and allowed to dry in a darkened
environment overnight at room temperature. 3T3-MEF Gal-3 -/- fusions were treated
similarly except they were stained with anti-Mac-Z (25 pg/ml), followed by FITC-
conjugated goat anti-rat IgG.

Fluorescent cells were viewed using an Insight Plus laser scanning confocal
microscope (Meridian Instruments). The maximum and minimum fluorescence
intensities of each nucleus were determined using the image analysis subroutine. By
viewing over one hundred images, including human and mouse monokaryons as well as
homodikaryons and heterodikaryons, we found that a nucleus devoid of human galectin-3
(and therefore, not stained by NCL-GAL3) yielded a maximum fluorescence of 250
(arbitrary units). In contrast, a nucleus containing human galectin-3 always yielded a
fluorescence intensity of 300 or greater. On this basis: (a) a nucleus whose maximum
fluorescence intensity was 250 or less was scored negative for the presence of human
galectin-3; and (b) a nucleus whose minimum fluorescence intensity was 300 or greater
was scored positive for the presence of human galectin-3.

Preparation of cell lysates and immunoblotting

Cells were grown to near confluence in 100 X 20 mm tissue culture plates, rinsed
once with ice-cold PBS, and scraped in four ml of PBS with a rubber policeman. Scraped
cells were pooled into 15 ml tubes, centrifuged at 1470 X g for 10 minutes at 4°C, and the

supernatant was decanted. The cells were resuspended in 1 ml ice-cold PBS, transferred

69

 

to Eppendorf tubes, and pelleted in a microfirge at 3406 X g for 2 minutes at 4°C. The
supernatant was aspirated and the cells were resuspended in 150 pl of 10 mM Tris, pH
7.4, then incubated on ice for ten minutes. The cell suspension was then sonicated eight
times (15 seconds each) and stored at —20°C.

The amount of protein in each sample was quantitated by the Bradford method
(31) using Coomassie protein assay reagent (Pierce). Equal amounts of protein from each
lysate were separated on a 12.5% SDS-polyacrylamide gel and then transferred
electrophoretically to a nitrocellulose membrane in a buffer containing 25 mM Tris, 193
mM glycine, and 10% methanol (pH 8.3). The membrane was blocked overnight at room
temperature in T-TBS containing 10% non-fat dehydrated milk, then rinsed three times
briefly in T-TBS, and blocked with a 1:1000 dilution of goat anti-rabbit IgG in T-TBS
containing 1% non-fat dehydrated milk. Following blocking, the membrane was rinsed
with T-TBS.

In immunoblots employing NCL-GAL3, the membrane was incubated with NCL-
GAL3 (21 ng/ml) in T-TBS containing 1% non-fat dehydrated milk for one hour at room
temperature. Following incubation with NCL-GAL3, the membrane was rinsed with T-
TBS, and incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG
in T-TBS containing 1% non-fat dehydrated milk for one hour at room temperature. The
membrane was then rinsed with T-TBS, and the proteins were visualized using the
Renaissance western blot chemiluminescence reagents (New England Nuclear Life

Science Products, Boston, MA).

70

 

Immunoblots employing anti-Mac-2 were canied out in a similar fashion except
that anti-Mac-2 was used at 125 ng/ml, and HRP-conjugated goat anti-rat IgG was used

as the secondary antibody.

Acknowledgements

This work was supported by a grant (GM-38740) from the United States National
Institutes of Health, a grant (MCB 97-23615) fi'om the United States National Science
Foundation, and a grant (number 5751) from Association de la Recherche contre le
Cancer.

We thank Kyle Openo for his initial characterization of the specificity of the
NCL-GAL3 antibody for human galectin-3. We thank Dr. Minoru Yoshida, University
of Tokyo, for his generous gift of leptomycin B, and Drs. J. J. McCormick and V. M.

Maher for their generous gift of LG-l cells.

71

 

References
1. Barondes, S.H., et al. (1994) Cell 76, 597-598.
2. Moutsatsos, I.K., Davis, J.M., and Wang, J .L. (1986).]. Cell. Biol. 102, 477-483.

3. Hubert, M., Wang, S.-Y., Wang, J .L., Séve, A.-P., and Hubert, J. (1995) Exp. Cell
Res. 220, 397-406.

4. Craig, S.S., Krishnaswamy, P., Irani, A.-M.A., Kepley, C.L., Liu, F.-T., and
Schwartz, LB. (1995) Anat. Rec. 242, 211-219.

5. Dagher, S.F., Wang, J .L., and Patterson, R.J. (1995) Proc. Natl. Acad. Sci. USA. 92,
1213-1217.

6. Park, J.W., Voss, P.G., Grabski, S., Wang, J .L., and Patterson, R.J. (2001) Nucleic
Acids Res. 27, 3595-3602.

7. Charroux, B., Pellizzoni, L., Perkinson, R.A., Yong, J ., Shevchenko, A., Mann, M.,
and Dreyfuss, G. (2000) J. Cell Biol. 148, 1177-1186.

8. Fischer, U., Qing, L., and Dreyfuss, G. (1997) Cell 90, 1023-1029.
9. Pellizzoni, L., Naoyuki, K., Charroux, B., and Dreyfuss, G. (1998) Cell 95, 615-624.

10. Borer, R.A., Lehner, C.F., Eppenberger, H.M., and Nigg, EA. (1989) Cell 56, 379-
390.

11. Colnot, C., Fowlis, D., Ripoche, M.-A., Bouchaert, I., and Poirier, F. (1998) Dev.
Dyn. 211, 306-313.

12. Agrwal, N., Wang, J.L., and Voss, P. (1989) J. Biol. Chem. 264, 17236-17242.

13. Tsay, Y.G., Lin, N.Y., Voss, P.G., Patterson, R.J., and Wang, J .L. (1999) Exp. Cell
Res. 252, 250-261.

14. Nishi, K., Yoshida, M., Fujiwara, D., Nishikawa, M., Horinuchi, S., and Beppu, T.
(1994).]. Biol. Chem. 269, 6320-6324.

15. Poirier, F. and Robertson, E]. (1993) Development 119, 1229-1236.
16. Piflol-Roma, S. and Dreyfuss, G. (1992) Nature 355, 730-732.

17. Roth, J ., Dobbelstein, M., Freedman, D.A., Shenk, T., and Levine, AJ. (1998) EMBO
J. 17, 554-564.

72

18. Levasseur-Paulin, M., and Julien, M. (1999) Exp. Cell Res. 250, 439-451.

19. Openo, K.P., Kadrofske, M.M., Patterson, R.J., and Wang, J .L. (2000) Exp. Cell. Res.
255, 278-290.

20. Ghisolfi-Nieto, L., Joseph, G., Dutilleul, F.-P., Amalric, F., and Bouvet, P. (1996).].
Mol. Biol. 260, 34-53.

21. Ginisty, H., Amalric, F., and Bouvet, P. (1998) EMBOJ. 17, 1476-1486.
22. Michael, W.M., Choi, M., and Dreyfuss, G. (1995) Cell 83, 415-422.

23. Pollard, V.W., Michael, W.M., Nakielny, S., Siomi, M.C., Wang, F., and Dreyfuss,
G. (1996) Cell 86, 985-994.

24. Dietz, H. (1998) Nature Genet. 20, 321-322.
25. Matera, A.G. and Frey, MR. (1998) Am. J. Hum. Genet. 63, 317-321.
26. Mattaj, I.W. (1998) Curr. Biol. 8, 93-95.

27. Meister, G., Bithler, D., Laggerbauer, B., Zobawa, M., Lottspeich, F., and Fischer, U.
(2000) Hum. Mol. Genet. 9, 1977-1986.

28. Morgan, TL. ,Yang, D. F,ry, D.G., Hurlin, P.F., Kohler, S. K. ,Maher, V..M ,and
McCormick, J. J. (1991)Exp. Cell Res 197, 125- 136.

29. Hogan, B., Beddington, R, Costantini, F. ,and Lacy, E. (1994) Manipulating the
mouse embryo. (2’1 ed.) Cold Spring Harbor Laboratory Press.

30. Ho, MK. and Springer, T.A. (1982) J. Immunol. 128, 1221-1228.

31. Bradford, M.M. (1976) Anal. Biochem. 72, 248-254.

73

Chapter 3

Transport of Galectin-3 between the Nucleus and Cytoplasm
1. Conditions and Signals for Nuclear Import

74

 

Transport of Galectin-3 between the Nucleus and Cytoplasm.
I. Conditions and Signals for Nuclear Import

Peter J. Davidsong, Su-Yin Li”, Rianna VandergaastA,
Ronald J. Pattersonw, John L. Wang”, and Eric J. Amoys”

Cell and Molecular Biology Program‘, Departments of Biochemistry”
and Microbiology”, Michigan State University, East Lansing, MI 48824, USA

Department of Chemistry and Biochemistry”, Calvin College
Grand Rapids, MI 49546, USA

Peter Davidson was responsible for collecting and interpreting the data reported in
Figures 2 — 7 of this chapter. Su-Yin Li, a student in the MS. program, designed and
engineered the expression vector shown in Figure 1, from which all of the mutants
reported in the present experiments were derived. This was done under the direct
supervision of Dr. Eric Amoys. Rianna Vandergaast, and undergraduate student with Dr.
Eric Amoys, carried out the site-directed mutagenesis to produce the mutants described
in Figures 5 and 6. Drs. Ron Patterson, John Wang, and Eric Amoys are faculty
members in whose laboratories the reported experiments were performed.

75

 

ABSTRACT

Galectin-3 (Ga13), a factor involved in the splicing of pre-mRNA, shuttles
between the nucleus and the cytoplasm. We have engineered a vector that expresses the
fusion protein containing: (a) Green Fluorescent Protein (GF P) as a reporter of
localization; (b) bacterial maltose-binding protein (MalE) to increase the size of the
reporter polypeptide; and (c) Ga13 whose sequence we wish to dissect in search of a
nuclear localization signal.

Analysis of the fluorescence in mouse 3T3 fibroblasts transfected with this
expression construct showed that the GFP-MalE-Gal3(1-263) fusion protein was
localized predominantly in the nucleus. Mutants of this construct, containing truncation
of the Ga13 polypeptide from the carboxyl terminus, showed loss of nuclear localization:
this effect was observed beginning with truncation at residue 259 and the firll effect is
seen with truncation at residue 232. Mutants of the same construct, containing truncation
of the Ga13 polypeptide from the amino terminus, retained nuclear localization; residues
228-263 of the Ga13 sequence was sufficient to direct the fusion protein into the nucleus.

These results suggest that the carboxyl-terminal region of the murine Ga13

polypeptide, residues 228-258 in particular, is important for nuclear localization.

INTRODUCTION
Galectin-3 (Ga13') is a member of a family of galactose-specific carbohydrate-
binding proteins found in a variety of cell types (1). It is predominantly an intracellular
protein, being found in both the cytoplasm and nucleus of cells (2). The nuclear

localization of Ga13 was sensitive to ribonuclease treatment of permeabilized cells, prior

76

to their fixation for analysis by immunofiuorescence and immunoelectron microscopy (3,
4). Moreover, sedimentation of nucleoplasm over cesium sulfate density gradients
identified Ga13 in fractions with densities corresponding to those reported for
heterogeneous nuclear ribonucleoprotein complex (hnRNP) and small nuclear RNPS
(snRNP). Because these RNPS play important roles in the nuclear processing of pre-
mRNA, the possibility was raised that Ga13 was a Splicing factor as well.

Indeed, using a cell-free assay, depletion and reconstitution experiments showed
that Ga13 and another member of the galectin family, galectin-1 (Gall), were redundant
but required factors in the splicing of pre-mRNA (5, 6). More recently, it was
documented (7) that Gal3, as well as Gall, interacts with Gemin4, which has been
characterized as one component of a macromolecular complex, designated as the SMN
complex (8). The functional Significance of this interaction appears to be in the early
steps of spliceosome assembly. The addition of the NH2-terminal domain of Gal3 or the
COOH-terminal 50 amino acids of Gemin4 to Splicing competent nuclear extracts
inhibited splicing and blocked the conversion of early (H/E) complexes to active
spliceosomes (7).

The SMN complex is found in both the nucleus and in the cytoplasm. In the
cytoplasm, the SMN complex is involved in the biogenesis of snRNPs (9), prior to their
entry into the nucleus to firnction as required components in the Splicing of pre-mRNA.
In the nucleus, the SMN complex is localized in discrete bodies called Gems (10) and
appears to play a role in the "rejuvenation or recycling" of the snRNPs for supplying
them to the intermediates (H/E complex) in spliceosome assembly (11). The association

of Ga13 with the SMN complex raises the possibility that the protein might also perform

77

functions in both the nucleus and the cytoplasm and that it might shuttle between the two
compartments. Indeed, analysis of Ga13 localization in both nuclei of heterodikaryons
(e.g. derived from fusion of a Ga13-expressing cell with a Ga13-null cell) provided
definitive evidence for nucleocytoplasmic shuttling (12).

In this chapter, we document the conditions and sequences required for nuclear

import of the protein.

EXPERIMENTAL PROCEDURES

Mation of the pEGFP-cl vector for expression of the figion protein GFP-MalE-Ga13

In this study, the analysis of nuclear versus cytoplasmic localization was carried
out using a fusion protein containing Gal3, Green Fluorescent Protein (GFP) and
bacterial maltose-binding protein (MalE). The construction of the vector pGMG3 for
expression of the firsion protein GFP-MalE-Gal3 in mammalian cells is summarized in
Figure 1. In stage 1, the cDNA for Ga13 was excised from plasmid pWJ 31 (13) by EcoRI
digestion and then inserted into the corresponding restriction site of the bacterial
expression vector pmal-C2x (New England Biolabs, Beverly, MA). In stage II, this
plasmid encoding the fusion protein MalE-Gal3 was used as the template for polymerase
chain reaction (PCR) amplification, using primers: 5'-GGGGGTACCATGAAAATCGA
AGAAGGTAAAC-B' (which generates the KpnI restriction site not on the template); 5'-
AGGTCGACTCTAGAGGATC-3' (which reproduces the BamHI site on the template).
In stage 111, this PCR product was ligated into the mammalian expression vector, pEGFP-
cl (Clontech, San Jose, CA). The expression of the fusion protein GFP-MalE-Gal3 from

pGMG3 in transfected cells is driven by a cytomegalovirus

78

Figure 1. Schematic diagram illustrating the construction of the vector for the
expression of the fusion protein GFP-MalE-Gal3 in mammalian cells.
In stage I, the cDNA for Ga13 was inserted into the EcoRI restriction site of the bacterial
expression vector pmal-C2x. After ascertaining that this vector expressed the desired
fusion protein MalE-Gal3, it was used, in stage II, as template for PCR amplification of
the fragment coding for MalE-Gal3, with a KpnI Site at the 5'-end and BamHI site at the
3'-end. In stage 111, this fragment was ligated into the mammalian expression vector,
pEGFP-cl. The expression of the firsion protein GFP-MalE-Gal3 from the pGMG3
vector in mammalian cells is driven by a cytomegalovirus promoter. The relative

positions of the three EcoRI restriction Sites in the pGMG3 vector are indicated.

79

Figure l

 

 

 

 

-—>.
Ptac malE
EcoR I ECOR 1 pmal-c2x
-1111111111111[LII ECORI
Gal3 cDNA
EcoRI l I EcoR I

 

stagel l ligation
' KpnI

 

 

 

80

promoter (Fig. 1). The vector for the production of GFP-Ga13 was prepared from the
respective cDNAs in a similar fashion, using the same primers and taking advantage of
the same restriction sites.

The strategy for generating mutants in which the Ga13 sequence was truncated
from the carboxyl terminus was to introduce stop codons at specific positions in the
pGMG3 plasmid (Fig. 1). Using the QuikChange Site-Directed Mutagenesis Kit
(Stratagene, La Jolla, CA), this was carried out at amino acid residues 262, 261, 260, 259,
258, 253, and 232. For example, insertion of a stop codon at position 259 results in the
fusion protein GFP-MalE-Gal3(1-258).

The pGMG3 plasmid contains three EcoRI restriction sites (Fig. 1): (a) between
GF P and MalE; (b) at the start of the Ga13 coding sequence; and (c) at the end of the Ga13
sequence. Site-directed mutagenesis (5'-CATCCCGGACTTCGGATCCACC-3' and 5'-G
GTGGATCCGAAGTCCGGGATG-B') was carried out to remove the last of these EcoRI
sites. The resulting plasmid was used as the template to remove the first EcoRI site,
between the GFP and MalE sequences (5'-CGAGCTCAAGCTTCGACTTCTGCAGTC
GACGG-3' and 5’-CCGTCGACTGCAGAAGTCGAAGCTTGAGCTCG-3'). This
provided the starting material for the generation of mutants in which the Ga13 sequence
was truncated from the amino terminus. Site-directed mutagenesis was carried out to
insert EcoRI sites into specific positions of the Ga13 sequence. After digestion with the
restriction enzyme, the isolated DNA was religated with T4 DNA ligase. The following
forward and reverse primers were used to obtain the respective GFP-MalE-Gal3 mutants:
(a) (74-263), 5’-CCTAGTGCCTACCCCGAATTCACTACTGCCCCTGGAGC-3' and

5'-GCTCCAGGGGCAGTGAATTCGGGGTAGGCACTAGG-3'; (b) (121-263), 5'-GC

81

TATCCTGCTGCTGGCGAATTCGGTGTCCCCGCTGGACC-B' and 5'-GGTCCAGC
GGGGACACCGAATTCGCCCGAAGCAGGATAGC-3'; and (c) (131-263), 5'-GGT
GTCCCCGCTGGAGAATTCACGGTGCCCTATGAC-3' and 5’-GTCATAGGGCAC

CGTGAATTCTCCAGCGGGGACACC-3'.

Cell culture and transfection

NIH mouse 3T3 fibroblasts were obtained from the American Type Culture
Collection (Rockville, MD). The cells were grown as monolayers in Dulbecco's
modified Eagle's medium (DME) containing 10% calf serum, 100 U/ml penicillin, and
100 pg/ml streptomycin at 37 °C in a humidified atmosphere of 10% CO2 (14). Cells
were transfected with vectors expressing fusion proteins containing the GF P reporter
group described above. For transfections to be analyzed by immunoblotting, the cells
were cultured and transfected in 60 mm plates (29 cm2 growth surface). For transfections
to be analyzed by fluorescence, the cells were cultured on glass coverslips in 35 mm
plates (10 cm2 grth surface). The following describes the protocol used for
transfection of a single 35 mm plate; for transfection of 60 mm plates, the amounts of
reagents used are increased 3-fold.

Cells were seeded at a density of l x 104 cells/cm2 and cultured overnight. A 100
pl solution of serum-free DME containing 1 pg of the DNA construct was mixed with
100 pl of serum-free DME containing 3 pl of lipofectamine (Invitrogen, Carlsbad, CA; 2
mg/ml). The cells in the culture plate were washed with serum-free DME and the 200 pl
mixture containing DNA and lipofectamine was added, along with 0.8 ml of serum-free

DME. The plate was placed in the C02 incubator for three hours, at which time 1 ml of

82

DME containing 20% calf serum was added. The plate was incubated for another six
hours. The medium in the plate was then replaced with 2 ml of fresh DME containing
10% calf serum. In some experiments, cycloheximide (CHX; Boehringer Mannheim,
Indianapolis, IN; 10 pg/ml final concentration) and leptomycin B (LMB; LC
Laboratories, Wobum, MA; 5.4 ng/ml (10 nM) final concentration) were included during
this medium change. The cells were incubated for an additional 5 hours, at which time
they were processed for fluorescence analysis (see below). For immunoblotting analysis,
the cells in 60 mm plates were incubated for 24 hours after the medium change before

harvesting for preparation of lysates.

Fluorescence Microscopy

For examination of the transfected cells by fluorescence microscopy, the
coverslips were first washed three times with ice-cold phosphate-buffered saline (PBS).
The cells were then fixed by treating for 20 minutes in 2 ml of 4% paraformaldehyde in
PBS at room temperature. The cells were washed twice (10 minutes each, 3 ml PBS) at
room temperature. Finally, the coverslips were mounted on glass microscope slides using
Perma-Fluor (Thermo Shandon, Pittsburgh, PA). In some early experiments, we had also
transfected cells cultured in Lab-Tek Chamber Slides (N alge Nunc International,
Naperville, IL) using the same conditions as described for the transfection of the cells
cultured on coverslips in 35 mm plates. The results obtained with GFP-fluorescence in
the live cells in chamber slides and fixed cells on coverslips were essentially the same.

We also compared the localization of the GFP-MalE-Ga13 reporter against the

localization of endogenous Gal3 in the 3T3 fibroblasts. Endogenous Ga13 was detected

83

using the rat monoclonal antibody, anti-Mac2 (25 pg/ml in PBS containing 0.2% gelatin)
and fluorescein-conjugated goat anti-rat immunoglobulin (Sigma, St. Louis, MO; 1:500
dilution). The details of the indirect immunofluorescence protocol have been previously
described (15).

Fluorescent cells were examined using a Meridian Instruments (Okemos, MI)
Insight confocal laser scanning microscope. For each construct, we counted
approximately 100 cells, scoring the fluorescence labeling pattern in each cell in one of
five categories: (a) exclusively nuclear (N); (b) intensely nuclear over a cytoplasmic
background (N >C); (c) equal distribution between the nucleus and cytoplasm (N~C); (d)
less nuclear labeling than the cytoplasm (N<C); and (e) exclusively cytoplasmic (C).
These distributions of fluorescence are expressed in the form of histograms.
Representative cells were photographed at low (66X) magnification to Show a field

containing multiple cells and at high (200X) magnification to Show a single cell.

SDS-PAGE and immunoblotting

Proteins were resolved on SDS-PAGE (10% acrylamide) as described by
Laemmli (16). The procedures for immunoblotting after SDS-PAGE have also been
described (15). The antibodies used for immunoblotting and their sources were: (a)
polyclonal anti-GFP (Clontech); (b) anti-MalE (New England Biolabs); and (c)

polyclonal anti-Ga13 (#32 and #33, see reference 17).

84

RESULTS
A GF P reporter construct for the localization of Ga13

In order to define the NLS and NES of Gal3, we developed a reporter construct
expressing a fusion protein containing Ga13 (~33 kD) and GF P (~27 kD). This fusion
protein also contains bacterial maltose-binding protein (MalE; ~40 kD) to serve as a
"spacer" that increases the molecular weight of the reporter polypeptide. This was done
to insure that the size of the reporter polypeptide would exceed the exclusion limit of
nuclear pores (40-60 kD), even when the portion of the Ga13 polypeptide is decreased
through truncation.

The cDNA for Gal3 was digested with EcoRI and ligated into the corresponding
restriction site in the prokaryotic expression vector pMAL-C2x (Fig. 1). The success of
this step was indicated by: (a) the same MalE-Gal3 fusion protein (~74 kD) could be
detected in bacterial lysates by immunoblotting with either anti-MalE or anti-G313; (b)
the MalE-Gal3 fusion protein could be isolated on lactose affinity columns. This plasmid
then served as the template for PCR amplification of the coding region corresponding to
MalE-Gal3 and the product was ligated into the eukaryotic expression vector, pEGFP-cl
(Fig. 1). This initial construct and its variants were used to transfect mouse 3T3
fibroblasts. The authenticity of each of the constructs was confirmed by DNA
sequencing. In addition, extracts of transfected cultures were subjected to Western
blotting with antibodies directed against the three parts of the expressed fusion protein:
(a) anti-GFP; (b) anti-MalE; and (c) anti-Gal3. This was carried out to ascertain that the

predominant polypeptide bearing the GFP reporter group corresponded to the expected

85

molecular weight of the fusion protein. Thus, the observed fluorescence signal can be
ascribed to the localization of the construct under study.

Transfection of 3T3 cells with GFP resulted in the expression of a ~27 kD
polypeptide (Fig. 2, lane 1) and fluorescence in both the nuclear and cytoplasmic
compartments (data not shown). This observation is consistent both with the expectation
that a ~30 kD polypeptide would be able to diffuse across the nuclear pores, as well as
with previous reports on the localization of GP P (18, 19). Western blotting of cells
transfected with GFP-MalE-Gal3 (1-263) yielded an ~100 kD polypeptide (Fig. 2, lane
4), consistent with the sum of the molecular weights of the three parts of the fusion
protein. Some GFP-MalE-Gal3(1-263) transfected cells exhibit fluorescence exclusively
in the nucleus (N) (Fig. 3, panel A) while other cells Showed fluorescence intensely
nuclear over a cytoplasmic background (N>C) (Fig. 3, panel B).

Transfection with GFP-Ga13 (1-263) yielded the expected ~60 kD polypeptide
(Fig. 2, lane 2) and a predominantly N>C fluorescence pattern, similar to that shown in
Figure 3, panel B. The N~C pattern (equal distribution between the nucleus and
cytoplasm) was found in cells transfected with GFP-MalE (~67 kD) (Fig. 2, lane 3; Fig.
3, panel C). GFP-MalE-Gal3(l-252) (Fig. 2, lane 5), in which the Ga13 polypeptide is
truncated at residue 253, yielded the N<C (less nuclear labeling than the cytoplasm)
fluorescence pattern (Fig. 3, panel D). The exclusively cytoplasmic (C) labeling pattern
was found in cells transfected with GFP-MalE-Gal3(1-23 1) (Fig. 3, panel B), which

produced a ~96 kD polypeptide (Fig. 2, lane 6).

86

Figure 2. Analysis of the fusion proteins expressed from the GFP reporter vector by
Western blotting.

Lane 1: GF P; lane 2: GFP-Gal3(1-263); lane 3: GFP-MalE; lane 4: GFP-MalE-Gal3(l-

263); lane 5: GFP-MalE-Ga13(l-252); lane 6: GFP-MalE-Gal3(l-231). Mouse 3T3

fibroblasts were transfected with the various constructs. Extracts (~50 pg total protein)

derived from each of the transfected cultures were subjected to SDS-PAGE and

immunoblotting with antibodies directed against GFP.

87

Figure 2

28:28-9“on
anvcmaoaazdao
Gowzmaomazdao
mazdao
Gowsmaoéo

.EO

88

Figure 3. Representative fluorescence micrographs illustrating the N, N>C, N~C,
N<C, and C labeling patterns.
(A) cells exhibiting the N pattern (exclusively nuclear fluorescence); (B) cells exhibiting
the N>C pattern (intensely nuclear fluorescence over a cytoplasmic background); (C)
cells exhibiting the N~C pattern (approximately equal fluorescence distribution between
the nucleus and cytoplasm); (D) cells exhibiting the N<C pattern (weaker nuclear
fluorescence than the cytoplasm); (E) cells exhibiting the C pattern (exclusively

cytoplasmic fluorescence). Bar = 10 pm.

89

Figure 3

 

90

Manson of the subcellrdar distribution of GFP-MalE-Gal3l1-263) and endogengtg
galectin-3

To ascertain that the subcellular distribution of the GFP-MalE-Gal3(1-263)
reporter construct reflected that of the endogenous protein, we compared the GFP
fluorescence pattern against the localization of Ga13 in 3T3 fibroblasts as revealed by
indirect immunofluorescence using a rat monoclonal antibody directed against the
protein. This comparison was carried out on a quantitative basis by scoring each cell in
one of five categories: N, N>C, N~C, N<C, and C. The histograms of the fluorescence
patterns for GFP-MalE-Gal3(1-263) and endogenous Ga13 were very Similar, with ~10%
N, ~50% N>C, ~40% N~C, and <5% N<C (Figure 4).

In previous studies, we had documented that treatment of mouse and human
fibroblasts with LMB resulted in the accumulation of Ga13 in the nucleus, as revealed by
an accentuation of the nuclear staining pattern (15, 20). LMB inhibits the interaction
between the CRM] export receptor and a leucine-rich NES on the cargo (21). In the
present experiments, we included CHX (10 pg/ml) during the incubation with LMB (10
nM) to exclude complications that might be introduced by newly synthesized proteins
(e. g. due to different rates of synthesis of the endogenous protein versus the fusion
protein). Again, the histograms of the fluorescence patterns for GFP-MalE-Gal3(l-263)
and endogenous Ga13 were very similar, with ~30% N, ~60% N>C, and ~15% N~C (Fig.
4). More importantly, the data clearly showed that treatment with LMB Shifted the

fluorescence distribution "to the left," in favor of the nucleus.

91

Figure 4. Comparison of the histograms of fluorescence patterns obtained with
GFP-MalE-Gal3(l-263) with endogenous Ga13 in 3T3 fibroblasts.

The localization of GFP-MalE-Gal3(1-263) (Fusion) was monitored by GFP
fluorescence; the localization of endogenous Ga13 (Endo) was determined by indirect
immunofluorescence using a rat monoclonal antibody against Gal3, the anti-Mac2
antibody. The fluorescence distributions were compared both in the presence and
absence of a combination of the drugs LMB (5.4 ng/ml) and CHX (10 pg/ml). Data were
collected from three independent experiments in which at least 100 fluorescent cells were
scored for the localization of GFP fluorescence. The average percentages of cells
showing each localization was plotted. Error bars represent the standard deviation of the

data within each localization category.

92

Figure 4

 

 

 

 

 

 

 

 

 

100] No Drug 100 CHX / LMB
80 - 80 -
60 -
Endogenous
40 '
20-
0 - 1 .
N N>C N~C N< C N N>C N~C N<C C
100 ‘ 100
80 ' 80 ‘

60‘

Fusion
40 r

20‘

 

 

 

 

   

0
N N>C N~C N<C C N N>C N~C N<C C

93

Effects of truncation from the carboxyl terminus on the localization of the GFP-Mg];
_(_3_aj3 fu_sion protein

Transfection with the construct expressing the full-length Ga13 polypeptide, GFP-
MalE-Gal3(1-263), resulted in nuclear localization of the fusion protein (Fig. 3, panel A).
These qualitative observations of fluorescence patterns were confirmed by a more
quantitative analysis in terms of the histogram distribution (Fig. 5). Approximately 10%
of the transfected cells showed an exclusively nuclear localization while ~55% of the
cells showed fluorescence intensely nuclear over a cytoplasmic background (N>C). The
remaining 40% of the transfected cells exhibited fluorescence in both the nucleus and
cytoplasm (N~C). Using site-directed mutagenesis to insert stop codons, we found that
deletion of the last four amino acids of the Ga13 polypeptide did not alter the localization;
for example, the histogram of fluorescence patterns for GFP-MalE-Gal3(1-259) was very
Similar to that of the parent protein, GFP-MalE-Gal3(l-263) (data not shown).

Truncation of residue 259 (i.e. GFP-MalE-Gal3(l-258)), however, resulted in a
shift of the histogram "to the right," representing loss of nuclear localization in favor of
the cytoplasm (Fig. 5). This shift became more pronounced as the analysis is carried out
sequentially from residue 259 through residue 253. In GFP-MalE-Gal3(1-252) (~100
kD; Fig. 2, lane 5) transfected cells, ~50% of the cells exhibited an exclusively
cytoplasmic (C) localization while ~20% of the cells showed the N~C fluorescence

pattern.

94

Figure 5. Comparison of the histograms of fluorescence distribution for GFP-MalE-
Gal3(l-263) and mutants that are truncated from the carboxyl terminus.
For GFP-MalE-Gal3(1-263) and GFP-MalE-Gal3(1-23 1), the data represent the averages

of triplicate determinations with standard deviation.

95

Figure 5

 

 

  

 

 

N N>C N~C N<C c

 

100 r
80 .
GFP-
MalE- 60‘
Ga13
(1-263) 40-
20‘
o .
100
80‘
GFP-
MalE- 60‘
Gal3-
(1-231) 40-

20‘

 

0.

N N>C N~C N<C C

 

96

 

100

80 4

GFP-
MalE-
Ga13
(1-258) 40 i

60~

207

 

oz

 

 

N N>C N~C N<C C

 

100

80 ~
GFP-
MalE-
Ga13
(1457) 40 -

60-

20~

  

 

 

 

N N>C N~C N<C C

 

100

80~

GFP-
MalE- 60 ~
Gal3

(1-252) 40 ,

20-

 

0 _

 

N N>C N~C N<C C

 

When the truncation is made at residue 232 (GFP-MalE-Gal3(l-23 1, ~96 kD) (Fig. 2,
lane 6), more than 60% of the transfected cells had an exclusively cytoplasmic
localization, as evident by examining either the micrograph .(Fig. 3, panel B) or the
histogram (Fig. 5). These results strongly suggest that the carboxyl terminal portion of

the Ga13 polypeptide, upstream of residue 259, was important for nuclear localization.

Effects of truncation of the amino-terminal domaia

Using site-directed mutagenesis, EcoRI sites were inserted at various positions in
the G313 polypeptide. Mutants with truncation from the amino terminus of various
lengths were generated alter restriction enzyme digestion and religation, and were then
analyzed as GFP-MalE-Gal3 fusion proteins. The histogram of fluorescence patterns for
GFP-MalE-Gal3(121-263) was essentially the same as that of the full-length protein,
GFP-MalE-Gal3(1-263) (Fig. 6). Similar results were obtained with GFP-MalE-Gal3(74-
263) and GFP-MalE-Gal3(l3 1-263) (data not Shown). Together, the data indicated that
the amino-terminal domain of the Ga13 polypeptide is dispensable with respect to nuclear
localization.

Of particular significance is the mutant GFP-MalE-Gal3(228-263), which showed
that in practically all of the cells transfected with this construct, the fusion protein was
able to localize to the nucleus. The interpretation of this result may be complicated by
the fact that in transfections with GFP-MalE (lacking Ga13 sequence) we observed

nuclear localization of the fluorescent reporter in a majority of the cells. Thus,

97

Figure 6. Comparison of the histograms of fluorescence distributions for GFP-
MalE-Gal3(l-263) and mutants that are truncated from the amino terminus.

The data represent the averages of triplicate determinations with standard deviation.

98

Figure 6

 

 

 

N N>C N~C N<C C

 

1 00 1
80 ‘
GFP-
MalE- 60‘
Ga13
(1-263) 40 '
20 r
0 .
100
80 ‘
GFP- ,
MalE- 60
Ga13

(121-263) 40‘

20‘

0

 

 

N N>C N~C N<C C

 

100

80‘

GFP— 60 .

MalE-
Gal3

(228-263) 40 ‘

20:

0 .

 

 

N N>C N~C N<C C

99

 

 

 

the nuclear localization of GFP-MalE-Gal3(228-263) could simply reflect the behavior of
GFP-MalE. Two lines evidence argue against this notion. First, preliminary evidence
indicates that GFP-MalE-Gal3(228-252) yielded a cytoplasmic localization pattern,
consistent with the conclusion that amino acids upstream of residue 259 was important
for nuclear localization. Second, we also engineered the construct GFP-MalE-Gal3(1-
159; 226-263) and preliminary evidence indicates that this was also localized to the
nucleus. These results, coupled with the fact that over 90% of the cells expressing the
GFP-MalE-Gal3(1-231) construct showed cytoplasmic localization, suggests that the
amino acid sequence spanning the carboxyl- terminal 30 residues of the Ga13 polypeptide
contains information which allowed the protein to localize to the nuclear compartment.
This sequence is displayed in Figure 7A for the mouse protein as well as orthologs in

other species.

DISCUSSION
The murine Ga13 polypeptide contains 263 amino acids (22). Measurements of
the solution molecular weight of the purified protein by hydrodynamic (23) or by
thermodynamic methods (24) yielded a value of ~30,000, suggesting that the polypeptide
exists predominantly as a monomer. Although this is of a size that could be
accommodated by the aqueous channel of the nuclear pore complex (exclusion limit of
40-60 kD; for reviews, see 25, 26), two lines of evidence argue against the notion that

passive diffusion could account for the observed nuclear localization of Gal3.

100

Figure 7. (A) The amino acid sequence of the carboxyl-terminal 40 residues of the
Ga13 polypeptide.
The murine polypeptide, subjected to mutagenesis in the present study, is shown at the
top. The sequences of orthologs in other species are also shown for comparison. Residue
numbers corresponding to the positions in the polypeptides of the respective species are
shown on the left.
(B) Ribbon diagram showing the three-dimensional structure of the polypeptide
backbone of the CRD of Gal3.

The thick ribbons highlight the strands of two B-pleated sheets that form the core of the
structure. One B-Sheet contains five strands (Fl-F5); the other B-sheet contains six
strands (SI-86). The dark, twisted arrow at the lower left highlights the position of the
solvent exposed helix (His 236 through Ser 245) that connects the two B-Sheets. The

balls at the right represent the position of the bound carbohydrate ligand.

101

Figure 7

A

MOUSE 224 VAVNDAH , ’ GDTTS-lLA
HAMerR 206 VAVNDAH M

HUMAN 202

RAT 223

RABBIT 194

DOG 257

CHICKEN 223 o F GDTTSILVL

 

 

102

First, analysis of Ga13 in cell fractions suggests that the protein is associated with
high molecular weight complexes inside cells: (a) in ~40S particle when nuclear extracts
are fractionated on a sucrose gradient (3); and (b) in a ~670_ kD complex when the
transported fraction of a nuclear export assay in permeabilized cells is fractionated by gel
filtration ( 15). Second, there are reports in which an exclusively cytoplasmic localization
of Ga13 in one cell type can be altered to yield a predominantly nuclear phenotype and
vice versa. For example, the protein is cytoplasmic in quiescent cultures of 3T3
fibroblasts but it relocates to the nuclear compartment upon serum stimulation (27). In
the LGl strain of human diploid fibroblasts, Ga13 can be found in both the nucleus and
the cytoplasm of young, proliferating cells; in contrast, the protein was predominantly
cytoplasmic in senescent LG] cells that have lost replicative competence through in vitro
culture (20). In heterodikaryons derived fi'om fusion of young and senescent LGl cells,
the predominant phenotype was galectin-3 in both nuclei, suggesting that the senescent
cells might lack a factor(s) specifically required for Ga13 nuclear import. Conversely,
Gal3 has been reported to be concentrated in the nuclei of differentiated colonic epithelial
cells. The progression from normal mucosa to adenoma to carcinoma is Characterized by
a distinct absence of Ga13 in the nuclei of adenoma and carcinoma cells (28). Thus, the
finding that Ga13 can Show a uniquely nuclear localization under one set of conditions
and an almost exclusively cytoplasmic localization under other conditions suggests
specific and regulated mechanisms of balance between cytoplasmic anchorage, nuclear
import, nuclear retention, and nuclear export.

The results of the present study indicate that the carboxyl terminal region of the

murine Ga13 polypeptide is important for nuclear localization, as assayed using a GFP-

103

MalE-Gal3 reporter system. The critical residues appear to lie, at the least, within a
region spanning His 230 and Ala 258. Mutants truncated from the amino-terminal end
retained nuclear localization activity; in fact, GFP-MalE-Ga13(228-263) bearing the
carboxyl-terminal 35 amino acids could still localize to the nucleus. Truncation from the
carboxyl-terminal end, however, resulted in the loss of nuclear localization activity.
Although truncation at residue 258 resulted in a shift from the nuclear to the cytoplasmic
localization, the full effect is seen by truncation at residue 231, in which more than 60%
of the cells exhibited an exclusively cytoplasmic localization.

Our results need to be compared with those reported by two other laboratories that
have attempted to define the region of the Ga13 polypeptide important for nuclear
localization. Gong et al. (29) reported that deletion of the amino-terminal 11 amino acids
of human Ga13 resulted in a mutant exhibiting cytoplasmic (and no nuclear) localization
in BT-549, a human breast carcinoma cell line. Moreover, when the sequence
corresponding to the first 11 amino acids was firsed to GFP, a predominantly nuclear
distribution of the reporter was observed. 0n the basis of these and other data, the
authors concluded that the 11 residues at the amino terminus are involved in Ga13
translocation to the nucleus (29). The nuclear localization of our mutants truncated from
the amino-terminal end (e. g. GFP-MalE-Gal3 (121-263)) and the cytoplasmic
localization of our mutants truncated from the carboxyl-terminal end (e. g. GFP-MalE-
Gal3(l-231)) are clearly inconsistent with the results and conclusion of Gong et al. (29).

In contrast to the results of Gong et al. (29), Gaudin et al. (30) transfected Cos-7
(SV40 virus-transformed monkey kidney) cells and Rb-l (rabbit smooth muscle) cells

with cDNAs encoding mutants of hamster Ga13 containing amino-terminal or internal

104

deletions and showed that nuclear localization does not require the first 103 amino acid
residues. Further deletion of residues 104-110 drastically reduced nuclear localization
but the specific sequences between residues 104-110 were not obligatory (these residues
could be substituted by unrelated sequences). Gaudin et al. (30) concluded that nuclear
localization of Ga13 does not require determinants in the amino-terminal domain; rather,
the carbohydrate-binding carboxyl-terminal domain was sufficient to allow nuclear
localization. In this respect, our present results are at least consistent with their results
and conclusions. The requirement noted by Gaudin et al. (30) that the carboxyl-terminal
domain be extended by additional (and unrelated) peptide sequences could, in effect, be
satisfied in our system by the GFP-MalE portion of the fusion protein.

The three-dimensional structure of the carbohydrate-binding carboxyl terminal
domain of Ga13 has been determined by X-ray crystallography (31). Like the
carbohydrate recognition domains of other members of the galectin family, the Ga13
structure is composed of two B-pleated sheets associating in a sandwiCh-like arrangement
(see Fig. 7B). One of the 13 sheets contains five anti-parallel strands while the other sheet
contains six anti-parallel strands. An a-helix (highlighted in Fig. 7B) connects the five-
stranded Sheet with the six-stranded sheet. The stretch of the Ga13 sequence identified in
the present study to be important for nuclear localization (230-25 8) is found in this region
of the three-dimensional structure. In particular, His 236 through Ser 245 forms the
solvent exposed connector helix between the two B-pleated sheets. Within these ten
residues (see Fig. 7A), there are three positively-Charged residues (two arginines and one

lysine), as well as a histidine residue which could potentially also bear a positive charge.

105

This appears to be similar to the NLS of the C-myc polypeptide (PAAKRVKLD), in
which a nine-residue sequence contains one arginine and two lysines (32).

It should be emphasized that we do not know if the region of Ga13 important for
nuclear localization represents a binding site for a nuclear import receptor (importin-[3 or
relative). A direct interaction between Ga13 and a known importin remains to be
documented. Once identified, the binding between the particular importin and a
candidate NLS peptide (e. g. residues 236-245 of the Ga13 sequence) will also need to be
tested. Alternatively, the region of Ga13 important for nuclear localization could simply
represent a binding site to another nuclear protein which carries a bonafide NLS

mediated by a well-characterized importin.

106

7.

10.

ll.

12.

13.

14.

15.

16.

17.

REFERENCES

. Barondes, S.H., Castronovo, V., Cooper, D.N.W., Cummings, R.D., Drickamer, K.,

Feizi, T., Gitt, M.A., Hirabayashi, J ., Hughes, 0, Kasai, K., Leffler, H., Liu, F .-T.,
Lotan, R., Mercurio, A.M., Monsigny, M., Pillai, S., Poirier, F ., Raz, A., Rigby,
P.W.J., Rini, J .M., and Wang, J.L. ( 1994) Cell 76: 597-598.

. Liu, F .-T, Patterson, R.J., and Wang, J .L. (2002) Biochim. Biophys. Acta 1572: 263-

273.

. Laing, J.G., and Wang, J.L. (1988) Biochemistry 27: 5329-5334.

. Hubert, M., Wang, S.-Y., Wang, J .L., Seve, A.-P., and Hubert, J. (1995) Exp. Cell

Res. 220: 397-406.

. Dagher, S.F., Wang, J .L., and Patterson, R.J. (1995) Proc. Natl. Acad. Sci, USA. 92:

1213-1217.

. Vyakamam, A., Dagher, S.F., Wang, J .L., and Patterson, R.J. (1997) Mol. Cell. Biol.

17: 4730-4737.

Park, J .W., Voss, P.G., Grabski, S., Wang, J .L., and Patterson, R.J. (2001) Nucleic
Acids Res. 27 : 3595-3602.

. Meister, G., Eggert, C., and Fischer, U. (2002) Trends Cell Biol. 12: 472-478.

. Fisher, U., Liu, 0., and Dreyfirss, G. (1997) Cell 90: 1023-1029.

Liu, 0., and Dreyfirss, G. (1996) EMBOJ. 15: 3555-3565.
Pellizzoni, L., Kataoka, N., Charroux, B., and Dreyfirss, G. (1998) Cell 95: 615-624.

Davidson, P.J., Davis, M.J., Patterson, R.J., Ripoche, M.-A., Poirier, F., and Wang,
J.L. (2002) Glycobiology 12: 329-337.

Agrwal, N., Wang, J .L., and Voss, PG. (1989) J. Biol. Chem. 264: 17236-17242.
Steck, P.A., Voss, P.G., and Wang, J.L. (1979) J. Cell Biol. 83: 562-575.

Tsay, Y.-G., Lin, N.Y., Voss, P.G., Patterson, R.J., and Wang, J .L. (1999) Exp. Cell
Res. 252: 250-261.

Laemmli, UK. (1970) Nature 277: 680-685.

Agrwal, N., Sun, Q., Wang, S.-Y., and Wang, J.L. (1993) J. Biol. Chem. 268: 14932-
14939.

107

18. Sherman, M.P., de Noronha, C.M.C., Heusch, M.I., Greene, S., and Greene, WC.
(2001).]. Virol. 75: 1522-1532.

19. Alefantis, T., Barmak, K., Harhaj, E.W., Grant, 0, and Wigdahl, B. (2003).]. Biol.
Chem. 278: 21814-21822.

20. Openo, K.P., Kadrofske, M.M., Patterson, R.J., and Wang, J.L. (2000) Exp. Cell Res.
255: 278-290.

21. Ossareh-Nazari, B., Bachelerie, F., and Dargemont, C. (1997) Science 278: 141-144.
22. Swiss-Prot: P161 10 (http://us.expasy.org/cgi-bin/niceprot.pl?P161 10)
23. Roff, C.F. and Wang, J.L. (1983).]. Biol. Chem. 258: 10657-10663.

24. Morris, S., Ahmad, N., Andre, S., Kaltner, H., Gabius, H.-J., Brenowitz, M., Brewer,
C.F. (2004) Glycobiology 14: 293-300.

25. Nigg, EA. (1997) Nature 386: 779-787.

26. Gorlich, D. and Kutay, U. (1999) Ann. Rev. Cell Dev. Biol. 15: 607-660.

27. Moutsatsos, I.K.,Wade, M.,Schindler, M.,and Wang, J .L. (1987) Proc. Natl. Acad.
Sci. USA. 84: 6452-6456.

28. Lotz, M.M., Andrews, C.W., Korzelius, C.A., Lee, E.C., Steele, G.D., Clarke, A. and
Mercurio, A.M. (1993) Proc. Natl. Acad. Sci. USA. 90: 3466-3470. .

29. Gong, H.C., Honjo, Y., Nangia-Makker, P., Hogan, V., Mazurak, N., Bresalier, RS,
and Raz, A. (1999) Cancer Res. 59: 6239-6245.

30. Gaudin, J.-C., Mehul, B., and Hughes, RC. (2000) Biol. Cell 92: 49-58.

31. Seetharaman, J ., Kanigsberg, A., Slaaby, R., Leffler, H., Barondes, SH, and Rini,
J.M. (1998).]. Biol. Chem. 273: 13047-13052.

32. Dang, CV. and Lee, W.H. (1988) Mol. Cell. Biol. 8: 4048-4054.

108

Chapter 4

Transport of Galectin-3 between the Nucleus and Cytoplasm
11. Identification of the Signal for Nuclear Export

109

Transport of Galectin-3 between the Nucleus and Cytoplasm.
11. Identification of the Signal for Nuclear Export

Su—Yin Li#, Peter J. Davidsong, Beric R. Henderson”,
Ronald J. Patterson‘y, John L. Wang”, and Eric J. AmoysA'

Departments of Biochemistry” and Microbiology”, the Cell and Molecular
Biology Programg, Michigan State University, East Lansing, MI 48824, USA

Department of Chemistry and BiochemistryA, Calvin College
Grand Rapids, MI 49546, USA

Westmead Institute for Cancer Research”, Westmead Hospital,
University of Sydney, Westmead, New South Wales 2145, Australia

Dr. Beric Henderson engineered the original expression vector for testing nuclear export
signals, as described in Figure 1. He provided this vector for our study. Su-Yin Li, a
student in the MS. program, generated the expression vectors containing the nuclear
export signal of galectin-3, as well as the site-directed mutants. She also carried out most
of the transfection experiments in testing the galectin-3 nuclear export signal (Figures 4
and 6). Peter Davidson was responsible for Figures 2, 3, and 5, as well as the analysis of
the data shown in Figures 4 and 6. DrS Ron Patterson, John Wang, and Eric Amoys are
faculty members in whose laboratories the reported experiments were performed.

110

ABSTRACT

Galectin-3 (Ga13), a factor involved in the splicing of pro-mRNA, shuttles
between the nucleus and the cytoplasm. Previous studies had shown that incubation of
fibroblasts with leptomycin B resulted in the accumulation of Gal3 in the nucleus,
suggesting that the export of Ga13 from the nucleus may be mediated by the CRMl
receptor. A candidate nuclear export signal (NES), containing the requisite spacing of
leucine/isoleucine residues fitting the consensus sequence recognized by the CRMl
exportin, can be found between residues 240 and 255 of the murine Ga13 sequence. This
sequence was engineered into the pRev(1.4)-GFP (Green Fluorescent Protein) reporter
system, in which candidate sequences can be tested for nuclear export activity in terms of
counteracting the nuclear localization signal present in the Rev(l.4) protein. We found
that the sequence spanning residues 240-255 of the Ga13 polypeptide did exhibit nuclear
export activity when tested in this system. The previously characterizedNES sequences
of PKI (inhibitor of CAMP-dependent protein kinase) and Ich-Cr were also tested in the
pRev(1.4)-GFP system as examples of "strong" and "weak" nuclear export activities,
respectively. Compared in these terms, the NES of Ga13 was found to be "weak."
Nevertheless, the nuclear export of Rev(l .4)-Gal3 NBS-GFP firsion protein was sensitive
to inhibition by leptomycin B, accumulating in the nucleus of drug—treated cells. Site-
directed mutagenesis of Leu247 and Ile249 in the Ga13 NES decreased nuclear export
activity, consistent with the notion that these two positions correspond to the critical
residues identified on the NES of PKI. These results indicate that residues 240-255 of the
Ga13 polypeptide contains a leucine-rich NES which overlaps with the region (residues

228-258) identified to be important for nuclear localization.

111

INTRODUCTION

Galectin-3 is a pre-mRNA splicing factor found in both the cytoplasm and in the
nucleus of cells (1, 2). The protein shuttles between the two subcellular compartments
(3). Using digitonin-permeabilized mouse 3T3 fibroblasts, we had previously
documented that Ga13 is selectively exported from the nucleus (4). This export was
temperature dependent and could be blocked by the addition of wheat germ agglutinin,
which binds to the nuclear pore protein p62. In addition, we also showed that incubation
of mouse and human fibroblasts with leptomycin B (LMB) resulted in the accumulation
of Ga13 in the nuclei of the treated cells (4, 5). LMB binds to and inhibits the activity of
the nuclear export receptor, CRM] , which recognizes a leucine-rich nuclear export signal
(NES) on the cargo (6, 7). These results suggest that Ga13 is exported through the
nuclear pore complex by a receptor-mediated pathway involving CRMl. Thus, the
nuclear versus cytoplasmic distribution of the protein must represent some balance
between nuclear import versus export as well as mechanism(s) of retention in either one
of the compartments, cytoplasmic anchorage or binding to a nuclear component.

In this Chapter we present evidence for a leucine-rich NES at the COOH-terminal

portion of the protein.

EXPERIMENTAL PROCEDURES

Site-directed mutagenesis of the putative NES sequence in the GFP-MalE-CLal3 firsion

protein

The construction of the vector for expression of the fusion protein GFP-MalE-

Gal3 was described in Chapter 3. To generate the L247A and I249A mutant of GFP-

112

MalE-Gal3, site-directed mutagenesis was carried out with the QuikChange Site-Directed
Mutagenesis Kit (Stratagene, La Jolla, CA), using the vector for GFP-MalE-Gal3 as
template and primers: 5'-CGGGAAATCAGCCAAGCGGGGGCCAGTGGTGACATA

ACC-3'; 5’-GGTTATGTCACCACTGGCCCCCGCTTGGCTGATTTCCCG-3'.

The pRev(1.4l-GFP vector and man—ts

The pRev(1.4)-GFP vector (Fig. 1A) and its application for testing potential NES
sequences were developed by Henderson and Eleitheriou (8). Two previously identified
NES sequences were also used in our study as controls: (a) the NES of the inhibitor of
CAMP-dependent protein kinase (PKI) was Characterized as a strong NES (Fig. 1C, line
3); and (b) the NES of IKB-O. represented weak NES activity (Fig. 1C, line 4). Each of
these sequences was cloned as short fragments between the BamHI and the AgeI sites of
pRev(1.4)-GFP, sandwiched between the Rev and the GFP coding sequences.

The vector pRev(1.4)-GFP containing the putative NES sequence of Ga13
(residues 240-255) (Fig. 1C, line 5) was derived from the pRev(1.4)-GFP vector
containing the NES of PKI by site-directed mutagenesis in five steps, each of which
changed multiple amino acids. Finally, two mutants of the Ga13 NES, designated Gal3
NES (1244A; L247A) and G313 NES (L247A; 1249A) were derived from the wild-type
Gal3 NES in the pRev(1.4)-GFP vector by site-directed mutagenesis (Fig. 1C, lines 6 and
7). For Ga13 NES (I244A, L247A), the forward and reverse primers were, respectively:
5’-CCAAACCTTCGAGAGGCATCTCAGGCAGGTATCAGTGGG-3' and 5’-CCCACT

GATACCTGCCTGAGATGCCTCTCGAAGGTTTGG-3'.

113

Figure l. The Rev(l.4)-GFP vector.
(A) Schematic diagram of the pRev(1.4)-GFP vector for testing potential nuclear
export signals (N ES). A potential NBS is Cloned into this vector for the expression of a
fusion protein containing Rev( 1 .4)-test NBS-GF P. Normal Rev has both a NLS and a
NBS; the Rev(l.4) variant is NBS deficient. The expressed fusion protein uses the NLS
of Rev to import the reporter into the nucleus. The activity of the potential NBS to export
the reporter is measured in terms of the nuclear vs. cytoplasmic distribution of the GFP
fluorescence.
(B) Schematic diagram illustrating the basis of the assay to determine the nuclear
export activity of a potential NES. The Rev NLS is highlighted by bold letters in the
cytoplasm; Rev NLS-mediated import is reduced by treatment of cells with actinomycin
D (ActD). The test NBS is highlighted by bold letters in the nucleus; nuclear export
mediated by leucine-rich NBSs is reduced by leptomycin B (LMB).
(C) Summary of the contents of the fusion proteins expressed by various constructs.
The construct designated as GF P is simply the mammalian expression vector for the
production of GFP. The construct designated as Rev] .4 expresses a NBS-deficient
variant of Rev as a fusion protein with GFP. The construct designated as PKI NBS
expresses the Rev(l .4)-GFP fusion protein containing the specific NBS sequence shown,
derived from PKI. Similarly, the other constructs express the Rev(l.4)-GFP firsion

protein containing the specific test NBS sequence shown.

114

A) pRev(1.4)-GFP Revl.4 GFP

 

 

 

 

Figure 1

—>

 

 

 

 

 

 

\

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Test NES sequence
B)
Nucleus Cytoplasm
NLS NLS
LMB ActD
Rev +> ‘ARV— Rev
export Import
NES NBS
C)
Construct Test NES seguence comment
GFP GFP-N1(No NLS; No NES)
Rev1.4 NONE NLS but No NES
PKINES 33-NSNBLALKLAGLDINKTE residues33-500fPKI
[KB-a NES 26l-PSTRIQQQLGQLTLBNLQ residue5261-278oflch-a
Gal3 NES 240- N L R B I S Q L G l S G D I T L residues 240-255 of Ga13
Gal3 NES (1244A; L247A) A A residue 244: mutate I to A
residue 247: mutate L to A
Ga13 NES (L247A; 1249A) A A residue 247: mutate L to A

115

residue 249: mutate I to A

For Gal3 NBS (L247A, 1249A), the primers were: 5'-CGAGAGATATCTCAGGCAGG
TGCCAGTGGGGACATCACAC-3' and 5'-GTGTGATGTCCCCACTGGCACCTGCC
TGAGATATCTCTCG-3'. All of these experiments used the QuikChange Site-Directed

Mutagenesis Kit of Stratagene.

Cell culture and transfection

The conditions for the culture and transfection of NIH mouse 3T3 fibroblasts are
detailed in Chapter 3. In the present experiments, the effects of various drugs on the
nuclear versus cytoplasmic distribution of the reporter proteins were tested. At 9 hours
post transfection, either actinomycin D (ActD) and cycloheximide (CHX) or leptomycin
B (LMB) and CHX were added to the samples. The samples receiving no drugs served
as controls. After 5 hours of treatment (14 hours post transfection), the cells were
observed by fluorescence microscopy. ActD was purchased from Sigma (St. Louis, MO)
and wasdissolved in H2O as a 1 mg/ml stock solution and stored at -20 °C. It was added
to cultures at a final concentration of 5 pg/ml. CHX (Boehringer Mannheim,
Indianapolis, IN) was dissolved directly in culture medium at a concentration of 200
pg/ml and was added to cultures at a final concentration of 10 pg/ml. LMB was
purchased from LC Laboratories (Wobum, MA) as a 5.4 pg/ml stock solution in ethanol
and was stored at -20 °C. It was diluted in culture medium and then added to cultures at a

final concentration of 5.4 ng/ml (10 nM).

116

Fluorescence Microscopy

Transfected cells were examined by fluorescence microscopy as described in
Chapter 3, using a Meridian Instruments (Okemos, MI) Insight (confocal laser scanning
microscope. Approximately 100 cells were scored for GFP localization: (a) N,
fluorescence exclusively in the nucleus; (b) N>C, fluorescence intensely nuclear over a
cytoplasmic background; (0) N~C, fluorescence in both the nucleus and cytoplasm; (d)
N<C, less nuclear labeling than the cytoplasm; and (e) C, fluorescence exclusively in the
cytoplasm. Representative cells were photographed at low (66X) magnification to Show

a field containing multiple cells and at high (200X) magnification to Show a single cell.

S_tatistical Argrlvsis

The number of cells scored into each localization category was tabulated from
triplicate experiments. Chi-square analyses were carried out using the statistical analysis
program StatView, version 5.0.1 (SAS Institute, Inc., Cary, NC 27513 USA). Analyses
were performed using the “Contingency Table” firnction, selecting “Coded summary

data” and deselecting “Fisher’s Exact Test”.

SDS-PAGB and immunoblotting

Proteins were resolved on SDS-PAGB (10% acrylamide) as described by
Laemmli (9). The procedures for immunoblotting after SDS-PAGE have also been
described (4). Polyclonal anti-GFP antibodies were obtained from Clontech (San Jose,
CA); anti-MalB antibodies were from New England Biolabs (Beverly, MA); and

polyclonal rabbit anti-Ga13 (#32 and #33) have been described previously (10).

117

RESULTS

An attempt to identify the NES using the GFP-MalE-Gal3 reporter

Previous studies had documented that treatment of mouse and human fibroblasts
with LMB resulted in the accumulation of Ga13 in the nucleus, as revealed by
accentuation of the nuclear staining (4, 5). Moreover, in the accompanying manuscript,
we also showed that the GFP-MalB-Gal3(1-263) construct behaved in the same way as
endogenous galectin-3 in terms of its response to LMB by accumulating in the nucleus.
These results all suggest that LMB inhibited the export of Ga13 from the nucleus and that
the latter process was mediated by the CRM] exportin, which recognizes the leucine-rich
NBS (6). Indeed, a putative leucine-rich NBS, with requisite spacing of
leucine/isoleucine residues, can be identified between residues 241 and 249 of the murine
Ga13 sequence (see Fig. 1C, line 5). Moreover, this putative NBS motif appears to be
conserved in the Ga13 homologs of various species.

On this basis, we wanted to make use of the availability of the GFP-MalE-Gal3(l-
263) construct by testing the effect of mutating two key residues in the putative NBS:
leucine 247 to alanine (L247A) and isoleucine 249 to alanine (1249A) These two residues
were Chosen for mutagenesis because they occupy corresponding positions that had been
shown to be critical for the functioning of the leucine-rich NBS in PKI (11). If the
putative NBS was indeed functional in CRMl-mediated nuclear export, we would expect
the fusion protein expressed by the mutant construct (GFP-MalB-Gal3(l-263; L247A;
1249A) to exhibit a nuclear localization. Transfection of 3T3 cells with the mutant
construct resulted, however, in a predominantly cytoplasmic fluorescence pattern (Fig.

2B). Whereas ~60% of the cells transfected with the wild-type construct (GFP- MalE-

118

Figure 2. Comparison of the properties of GFP-MalE—Gal3 (1-263) and GFP-MalE-
Gal3(1-263; L247A; 1249A).

(A) Western blots of lysates (~50 pg total protein) derived from cells transfected with

the indicated constructs using antibodies directed against GF P. (B) Representative

fluorescence micrographs illustrating the GF P localization patterns. Bar = 10 pm. (C)

Histograms showing the distribution of the percent of cells with the indicated

fluorescence patterns.

119

Figure 2

WTMT

114—!‘
Q
B
lllilllllllllllllllll |||li|||||||||||||||

. LJ

N N>CN~CN<c C

 

 

 

N N>CN~CN<c c

120

Ga13 (1-263)) Showed nuclear localization, most of the cells transfected with the mutant
showed cytoplasmic fluorescence (Fig. 2C).

DNA sequence analysis confirmed that the mutations had been correctly carried
out. Lysates derived from the transfected cultures were subjected to SDS-PAGE and
immunoblotting. Antibodies directed against all three parts of the fusion protein (GFP,
MalB, and Ga13) yielded the same results. The most prominent band was observed at
~100 kD, corresponding to the expected molecular weights of the wild-type and mutant
polypeptides (Fig. 2A). We interpret the results to indicate that this stretch of the Ga13
sequence, containing the putative NBS, was also important for a functional NLS and that
our mutagenesis on residues 247 and 249 inactivated the nuclear import signal. This
notion is consistent with the results of our analysis on the NLS, which implicated the

carboxyl-terminal 30 amino acids as necessary for nuclear import (see Chapter 3).

The Rev( 1.4)-GF P vector for thwsis of a functirnral NBS

The apparent overlap of the NLS and NES in the Ga13 polypeptide precluded us
from using the GFP-MalB-Gal3 reporter construct to define the NES. Mutations intended
to inactivate the putative NES also inactivated the NLS and therefore, the properties of
the NES cannot be studied in a polypeptide that fails to enter the nucleus. To circumvent
these difficulties, we have taken advantage of the pRev(1.4)-GFP vector (8). Although
the HIV-1 Rev protein normally contains both a NLS as well as a NBS, the Rev(l .4)
variant is NBS-deficient. Instead, test sequences representing a putative NBS can be
cloned in frame between the Rev(l.4) segment and the GFP reporter (see schematic in

Fig. 1A). The fusion protein expressed by this vector contains the NLS of Rev, whose

12]

nuclear import activity could be decreased by treatment of cells with ActD (12, 13). This
allows the activity of very weak NBSS to be detected. Thus, the relative activity of
different NBSS can be distinguished by their ability to shift the. fusion protein to the
cytoplasm, both in the presence and absence of active nuclear import (i.e., in the absence
and presence of ActD, respectively). If the test NBS is recognized by the CRMl
exportin, then nuclear export is expected to be sensitive to LMB inhibition (6, 7).

On this basis, ActD and LMB will play critical roles in our dissection of the NLS-
based nuclear import and the CRMl-mediated nuclear export of fusion proteins derived
from the Rev(l.4)-GFP vector (Fig. 18). In addition, in order to obviate any
complications arising from newly synthesized proteins bearing the GFP reporter, the
ActD and LMB experiments were always carried out in the presence of CHX. Therefore,
we first tested the effects of the three drugs on the nuclear versus cytoplasmic distribution
of endogenous galectin-3 as revealed by staining with anti-galectin-3 antibodies. Neither
CHX (10 pg/ml) nor ActD (5 pg/ml), alone or in combination, altered the nuclear and
cytoplasmic localization of galectin-3 in 3T3 cells. On the other hand, treatment of the
same cells with LMB at a concentration as low as 5.4 ng/ml (10 nM) resulted in the
accumulation of galectin-3 in the nucleus, as was reported previously (4). This effect of
LMB was also observed in the presence of CHX (see Chapter 3) or a combination of
CHX and ActD (data not shown).

Cultures of 3T3 cells were transfected with the construct expressing the Rev(l.4)-
GFP fusion protein (~40 kD). In the absence of drugs, ~70% of the fluorescent cells
exhibited the N labeling pattern (Fig. 3, panel A; Fig. 4). Addition of CHX and ActD

shifted the histogram of fluorescence distributions to the right, at the expense of the N

122

 

Figure 3. Representative fluorescence micrographs illustrating the GFP localization
patterns obtained with various test NES sequences in the pRev(1.4)-GFP vector.
(A) Cells expressing Rev( 1 .4)-GP P protein in the absence of drugs, ~70% of which
yielded the N fluorescence pattern. (B) Cells expressing Rev( 1 .4)-GFP containing the
NES of PKI in the absence of drugs, which yielded the C fluorescence pattern. (C) Cells
expressing Rev( 1 .4 )-GF P containing the NES of [KB-a in the absence of drugs, which
yielded the N>C and N~C fluorescence patterns. (D) Cells expressing Rev(l.4)-GFP

containing the NES of [KB-a and incubated in the presence of CHX (10 pg/ml) and ActD

(5 pg/ml), many of which yielded the C fluorescence pattern. Bar = 50 pm.

123

Figure 3

 

124

Figure 4. Histograms showing the distribution of the percent of cells with
fluorescence patterns N, N>C, N~C, N<C, and C.
The constructs used for transfection are indicated on the left-hand side of each row.
Transfected cells were incubated in the absence of drugs (No Drug), the presence of CHX
(10 pg/ml) and ActD (5 pg/ml) (CHX / ActD), or the presence of CHX (10 pg/ml) and
LMB (5.4 ng/ml) (CHX / LMB). The data represent the averages of triplicate

determinations with standard deviation.

125

No Drug

1 00

80

60
Rev1.4

40

20

 

 

N N>C N~CN<C C A

 

100
80‘

Ga13 60
NES
(wt) 40 ‘

20

 

 

N N>CN~CN<C c N

 

100
80 ‘

PKI 60‘

20‘

 

 

N N$CN-'-c Néc c

 

80‘

 

 

 

0. .
N N>C N~C N<C C

Figure 4

CHX / ActD

 

100
80 r
60‘
40‘

20‘

 

 

.80

 

N N>CN~CN<C C A

 

100
80'

 

 

N N>C N~C N<C C

 

 

 

 

 

CHX I LMB

100

60
40

20

 

 

N N>CN~CN<C C

 

1 00
80
60
40

20

 

B

N N>CN~C N<C C

 

 

 

 

 

100 100
80: an '

60*

40*

204

N N10 N-C N20 c N N>CN~C N<C C

10 *

80‘ 80

60‘ 60

40‘ 40

20« 20

o a - 0 »

 

 

N N>C N~C N<C C

126

 

 

 

 

N N>C N~CN<C C

labeling pattern (Fig. 4). This is consistent with the notion that the Rev(l.4)-GF P
polypeptide contains a NLS whose activity could be decreased by ActD (Fig. 1B) (12,
13).

We also tested the effect of inserting test NES sequences whose strengths had
been previously Characterized (8). In 100% of the cells transfected with the Rev(l .4)-
GF P vector containing the NES of PKI as the test sequence, the fluorescence labeling
pattern was exclusively C, cytoplasmic (Fig. 3, panel B). This was true irrespective
whether CHX and ActD were included in the cultures (Fig. 4). Thus, the PKI NBS was
sufficiently strong to overcome an active NLS (Fig. 1C, line3). The NBS of Ich-a (Fig.
1C, line 4) could neutralize an active NLS, resulting in a majority of the cells exhibiting
the N~C labeling pattern in the absence of drugs (Fig. 3 panel C; Fig. 4). In the presence
of CHX and ActD, the histogram shifts to the right, with ~50% of the cells showing
either a N<C or an exclusively C fluorescence pattern (Fig. 3, panel D; Fig. 4) since the

NLS activity has been decreased.

flIIVSiS of the Gal NES in the Rev( 1 .4)-GF P vector

When the putative NES of Ga13 was inserted into the Rev(l.4)-GFP vector as the
test sequence (Fig. 1C, line 5), about 40% of the transfected cells showed the N labeling
pattern while ~55% of the cells showed the N~C fluorescence pattern (Fig. 4). This 40%-
60% distribution between the N versus N~C labeling patterns was reproducible from
experiment to experiment. This distribution should be compared to the corresponding

distribution obtained in the transfection with the Rev(l .4)-GP P vector (Fig. 4). There

127

 

was a higher percentage of cells showing the N~C labeling pattern with the Ga13 NBS
than with no NBS. Nevertheless, the activity of the Ga13 NES appeared

weaker than that of the IKB-(I. NES, neutralizing the effect of the active nuclear import in
only half of the cells.

When nuclear import was reduced by addition of CHX and ActD, the majority of
the cells transfected with Ga13 NBS yielded either the N>C or the N~C fluorescence
pattern (Fig. 4), and the fraction of cells with exclusively nuclear (N) fluorescence pattern
dropped below 10%. Some 2% of the transfected cells Showed a shift of the GFP
fluorescence to the cytoplasm. Thus, the NES activity of the Ga13 sequence becomes

more apparent when nuclear import is reduced.

The effect of LMB on the fluorescence distribution

The Ga13 NBS activity, as reported by the pRev(1.4)-GFP vector, should be
sensitive to LMB inhibition, as had been documented for endogenous Ga13 of mouse and
human fibroblasts (4, 5). Indeed, incubation of transfected cells with CHX and LMB
shifts the distribution in favor of the nucleus, with more than 90% of the cells showing an
exclusively N or the N>C fluorescence pattern (Fig. 4). Although about 60% of the cells
yielded the N labeling pattern, there was nevertheless ~10% that showed N~C (Fig. 4).
When 3T3 cells were incubated with CHX and LMB, they accumulated endogenous Ga13
in the nucleus, as reflected by an accentuation of the nuclear staining (4); however, there
was always some cytoplasmic fluorescence in these LMB-treated cells. This was also the
case when the effect of LMB was studied using the GFP-MalB-Gal3(1-263) reporter (see

Chapter 3).

128

Addition of CHX and LMB also affected the fluorescence of other test NBS
sequences, shifting the distribution in favor of the exclusively nuclear (N) pattern. In the
presence of CHX and LMB, about 50% of the PKI NBS showed the N labeling pattern;
this should be compared with the exclusively cytoplasmic (C) pattern obtained in the
absence of the export inhibitor (Fig. 4). Similarly, CHX and LMB Shifted the
fluorescence distribution for the Ich-or test NES sequence, from a predominantly N~C
labeling patterns to ~50% exclusively N pattern (Fig. 4). In both cases, the effect of
CHX and LMB was partial; not all of the cells showed an exclusively N labeling pattern.

Finally, the Rev( 1 .4)-GFP construct contains no NBS; therefore, it should not be
sensitive to LMB. Consistent with this notion, CHX and LMB did not shift the
fluorescence distribution of the Rev(l .4)-GF P fusion protein in favor of the N labeling

pattern (Fig. 4).

Site-directed mutagenesis of the Ga13 NBS

Site-directed mutagenesis was carried out to generate Gal3 NBS (L247A; I249A),
the two positions corresponding to critical residues in the leucine-rich NBS of PKI (11).
In parallel, the Ga13 NBS (1244A; L247A) mutant was also generated. The fluorescence
distributions of Ga13 NBS (1244A; L247A) and Ga13 NES (WT) were very Similar (Fig.
5), particularly in terms of the effects of CHX and ActD (p=0.1183), and CHX and LMB
(p=0.0505). Addition of CHX and ActD shifted the distribution histogram to the right, at
the expense of exclusively nuclear (N) pattern. In the presence of CHX and LMB, the

predominant labeling pattern was exclusively nuclear (N).

129

 

Figure 5. A comparison of the histogram distributions of the percent of cells with
various fluorescence patterns for Ga13 NES (WT), Ga13 NES (1244A; L247A), and
Ga13 NES (L247A; I249A).

The constructs used for each transfection are indicated at the left-hand Side of each row.
Transfected cells were incubated in the absence of drugs (N 0 Drug), the presence of CHX
(10 pg/ml) and ActD (5 pg/ml) (CHX / ActD), or the presence of CHX (10 pg/ml) and
LMB (5.4 ng/ml) (CHX / LMB). The data represent the averages of triplicate

determinations with standard deviation.

130

1o

8

6

Rev1.4
4

 

No drug

 

N N>CN~C N<C C

 

100

80‘

 

Ga13 60’
NES
(wt) 40
20
0 .

 

N N>C N~C N<C C

Figure 5

 

 

 

 

10 CHX/ActD
80
60-
4o
20
0' N>C N~C N<C C
100
80-

 

 

 

JL

N N>C N~C N<C C

 

 

 

10 10
80 80-
20 20
0 I *

N N>C N~C N<C C

 

 

 

 

 

N N>C N~C N<C C

 

 

10

80-

Ga13 60

NES
(1244A; 40f
L247A)

20

 

In

 

10

80‘

60

40

 

20

CHX/LMB

 

 

 

 

N N>C N~CN<C C

 

10
80-

 

 

 

N N>C N~C N<C C V

 

100

 

 

 

N N>CN~CN<C C

 

1 00
80‘

 

 

 

 

 

 

N N>C N~CN<C C

N N>C N~C N<C C

131

1.1L

 

N N>C N~CN<C C

 

The fluorescence distributions Shown in Figure 5 were compared using a Chi-
squared test. Comparisons were made across all rows (e.g. Ga13 NBS (WT) No Drug vs.
Gal3 NES (WT) CHX/ActD, p<0.0001; Ga13 NBS (WT) No Drug vs. Ga13 NBS (WT)
CHX/LMB), p<0.0001; Ga13 NBS (WT) CHX/ActD vs. Gal3 NES (WT) CHX/LMB,
p<0.0001) and across all columns (e.g. Rev(l.4) CHX/ActD vs. Ga13 NES (WT)
CHX/ActD, p<0.0001; Rev(l.4) CHX/ActD vs. Ga13 NBS (L247A;1249A) CHX/ActD,
p=0.0157; Rev(l.4) CHX/ActD vs. Ga13 NBS (1244A;L247A) CHX/ActD, p<0.0001;

Gal3 NES (WT) CHX/ActD vs. Ga13 NES (L247A;1249A) CHX/ActD, p<0.0001; Ga13

’ . o\. l‘. A. I) -

NES (WT) CHX/ActD vs. Ga13 NBS (1244A;L249A) CHX/ActD, p=0.1183; Ga13 NBS
(L247A;1249A) CHX/ActD vs. Ga13 NBS (1244A;L247A) CHX/ActD, p=0.001).

The histograms of Ga13 NBS (L247; 1249A) generally resembled those of
Rev(l.4), which does not carry a NBS (Fig. 5). In the absence of drugs, the distributions
of Ga13 NES (L247A;1249A) and Rev(l.4) were not statistically similar (p<0.0001),
although both exhibited predominantly N fluorescence distribution, compared to the
40%-60% split between N and N~C labeling pattern observed in Ga13 NBS (WT). In
contrast, Rev( 1 .4) and Ga13 NBS (L247A;1249A) responded very similarly to treatment
with CHX and ActD (p=0.0157), as well as CHX and LMB (p=0.003). If the leucine-rich
NBS were disrupted by the L247A and 1249A mutations, one would not expect a shift
toward more cytoplasmic fluorescence upon reduction of nuclear import by ActD, nor
would one expect a Shift toward more nuclear fluorescence upon LMB inhibition of

CRM].

132

DISCUSSION

The nuclear versus cytoplasmic distribution of Ga13 depends on the cell type. In
BHK (baby hamster kidney) and MDCK (Madin-Darby canine kidney) cells, the protein
is cytoplasmic (14, 15). This cytoplasmic localization was observed both for the
endogenous protein as well as for the protein over-expressed in the same cells transfected
with a cDNA construct encoding the hamster polypeptide. Over-expression of the same
cDNA in mouse 3T3 fibroblasts, however, resulted in a predominantly nuclear
localization (15). The distribution of Ga13 between the nuclear and cytoplasmic
compartments in a Single cell type is also dependent on the proliferative state of the
culture under analysis. For example, quiescent cultures of fibroblasts (serum starved or
density inhibited) exhibit a predominantly cytoplasmic localization whereas proliferative
cultures of the same cells (serum stimulated or low density cultures) show nuclear
accumulation (16).

It should be noted that observations of an exclusively cytoplasmic localization of
a protein in cells do not necessarily mean that it does not enter the nucleus. It may
simply reflect a dynamic situation in which the rate of nuclear export far exceeds nuclear
import such that in any steady state observation, the protein is apparently found only in
the cytoplasm. For example, the ubiquitin-protein ligase (B3), hRPFl/Nedd4, is a key
component of the ubiquitin proteasome pathway. Immunofluorescence analysis showed
a cytoplasmic localization, due to the presence of a Rev-like "strong" NBS. Incubation of
cells in the presence of LMB, however, showed nuclear localization of the same protein
and a nuclear speckle protein, hPRTB (human proline-rich transcript, brain-expressed),

has been identified as a substrate of hRPF 1/Nedd4 in the nucleus (17). Similar

133

 

observations regarding LMB-induced conversion from cytoplasmic to nuclear
localization have also been demonstrated for the mitogen activated protein kinase
interacting kinase Mnkl, as well as several protein translation factors, which shuttle
between the nucleus and the cytoplasm (18, 19). Incubation of cells with LMB resulted
in shifting Mnkl from an exclusively cytoplasmic to nuclear localization.

Indeed, the intracellular traffic and distribution of Ga13 are similar to those of
Mnkl. Ga13 shuttles between the nucleus and cytoplasm (3) and it has also been
documented that incubation of human and mouse fibroblasts with LMB accumulates the
protein in the nuclear compartment (4, 5). Thus, there must be specific and regulated
balance between the four key parameters that govern nuclear versus cytoplasmic
distribution of Ga13: (a) cytoplasmic anchorage; (b) nuclear import; (C) nuclear retention;
and (d) nuclear export.

Henderson and Eleftheriou (8) developed the Rev(l.4)-GFP reporter system for
testing potential NBS sequences. Bach putative NBS sequence is challenged to overcome
the active NLS of the HIV -1 Rev protein such that the fusion protein localizes to the
cytoplasm. Such an NBS, Classified as "strong," was found in proteins such as PKI, the
mitogen activated protein kinase kinase (MAPKK), and the C-Abl oncogene (8). Some
test NBS sequences display "weak" nuclear export activity. These can partially neutralize
the NLS of the Rev(l .4)-GFP reporter, resulting in nuclear and cytoplasmic localization
of the fusion protein. In the presence of ActD, which decreases the NLS activity in the
Rev(l.4)-GFP reporter, the fusion containing a "weak" NBS shifts further to the
cyt0plasm in the majority of the cells. "Very weak" NBSS cannot normally overcome the

rate of Rev NLS-mediated nuclear import in the absence of ActD but are able to shift the

134

 

GF P fluorescence partially to the cytoplasm in 20-50% of the cells when import is
decreased by ActD. The tumor suppressor p53 and its hdm2 regulator each has an NBS
that fits this latter category.

This notion of ranking of NBS strengths has received strong support fiom an
entirely independent line of investigation. Using microinjection of defined recombinant
export substrates, Heger et al. (20) showed that different leucine-rich NBSS varied
dramatically in determining the kinetics of export in intact cells. Thus, the NES of PKI,
Classified as a "strong" NBS by the pRev(1.4)-GFP assay (8), was found to export its
protein very "fast" (5-10 minutes) (20). On the other hand, p53 was found to contain a
"very weak" NBS by the Rev(l.4)-GFP reporter assay (8) and, indeed, it turned out to be
"very slow" (> 10 hours) in the kinetic assay of Heger et al. (20). More interestingly, the
latter study also reported that co-transfection experiments revealed that proteins
containing a "fast" NBS inhibited the export and biological activity in vivo of proteins
harboring a "slower" NBS (20). Thus, the export of a protein harboring a leucine-rich
NES could also depend what other export substrates are present in competition for
transport receptors/cofactors.

By the criteria established in the development of the Rev( 1 .4)-GF P test vector (8),
the NES of Ga13 (residues 240-255 tested) would fall in the "weak” category. This weak
NBS activity may be important for the nuclear function of the protein. A strong NBS
might result in futile shuttling of Ga13 between the nucleus and cytoplasm while a weak
NBS would allow longer residence in the nucleus so that the protein can accumulate to
sufficient concentrations to assemble into the SMN complex for pre-mRNA splicing.

This notion was first advanced to explain the very low affinity observed between Ich-or

135

and the CRM] exportin (21), which appears to be consistent with our own observation
that its NBS exhibits "weak" nuclear export activity in the Rev(l .4)-GFP assay system.

In this connection, it may be useful to note that, in the study of Lee and Hannink
(21), the addition of LMB only shifted the cytoplasmic localization of Ich-ol to a nuclear
and cytoplasmic (N +C) pattern, rather than the exclusively nuclear (N) pattern. This
corresponds well with our present results, in which there were appreciable percentages of
cells showing the N~C pattern in the presence of LMB for all NBSS tested in the
Rev(l.4)-GF P fusion constructs (Fig. 4). Similarly, the nuclear accumulation of GFP-
Ich-or (22), GFP-MalE-Gal3(l-263) (see Chapter 3), and endogenous Ga13 (4, 5) was
increased by LMB addition but there were still appreciable levels of the respective
proteins remaining in the cytoplasm.

We had generated the double mutant, L247A and 1249A, in the GFP-MalB-Gal3
system to test whether the putative leucine-rich NBS of Ga13 was functional. The
exclusively cytoplasmic localization of GFP-MalB-Gal3(l-263; L247A, 1249A)
indicated, however, that the mutations may have disrupted the NLS. This, in turn,
implies that the NLS and NES overlap in this segment of the Ga13 polypeptide. The M9,
KN S, and HNS sequences represent other examples of overlapping signals, in which the
same stretch of amino acid sequence is capable of mediating both nuclear import and
nuclear export (23-25). The M9 signal, a stretch of ~38 amino acids with critical glycine
and proline residues, was identified on hnRNP A1 protein, responsible for its Shuttling
property between the nucleus and the cytoplasm. The 39-residue KNS shuttling signal
was identified on hnRNP K protein. For nuclear export, the critical residues include

negatively charge acidic amino acids. Finally, Fan and Steitz (25) identified a 33-residue

136

 

 

sequence, designated HNS, responsible for the shuttling activity of HuR, an RNA-
binding protein that can stabilize labile mRNAs containing AU-rich elements in their 3'-
untranslated regions.

The purpose of exporting Ga13 from the nucleus and, more generally, of shuttling
the protein between the nucleus and the cytoplasm remains to be elucidated. In studying
nuclear export of Gal3 using a permeabilized cell system, it was found that in the
transported fraction, Ga13 is associated with high molecular weight complexes of ~650
kD (4). On the basis of our previous documentation that Ga13 is involved in pre-mRNA
splicing (26, 27) and that its detection in the nucleus is sensitive to ribonuclease (28, 29),
the possibility is raised that Ga13 is exported from the nucleus in the form of a
ribonucleoprotein complex (RNP) along with the processed mRNA. The intriguing
question then is whether Ga13 plays a role in the determining the stability of the mRN A

or in targeting it to ribosomes for translation.

137

REFERENCES
1. Patterson, R.J., Wang, W., and Wang, J .L. (2004) Glycoconjugate J. 19: 499-506.

2. Wang, J .L., Gray, R.M., Haudek, KC, and Patterson, R.J. (2004) Biochim. Biophys.
Acta 1673: 75-93. ‘

3. Davidson, P.J., Davis, M.J., Patterson, R.J., Ripoche, M.-A., Poirier, F., and Wang,
J.L. (2002) Glycobiology 12: 329-337.

4. Tsay, Y.-G., Lin, N.Y., Voss, P.G., Patterson, R.J., and Wang, J .L. (1999) Exp. Cell
Res. 252: 250-261.

5. Openo, K.P., Kadrofske, M.M., Patterson, R.J., and Wang, J .L. (2000) Exp. Cell Res.
255: 278-290.

6. Ossareh-Nazari, B., Bachelerie, F., and Dargemont, C. (1997) Science 278: 141-144.

7. Kudo, N., Wolff, B., Sekimoto, T., Schreiner, E.P., Yoneda, Y., Yanagida, M.,
Horinouchi, S., and Yoshida, M. (1998) Exp. Cell Res. 242: 540-547.

8. Henderson, BR, and Eleftheriou, A. (2000) Exp. Cell Res. 256: 213-224.
9. Laemmli, UK. (1970) Nature 277: 680-685.

10. Agrwal, N., Sun, 0., Wang, S.-Y., and Wang, J .L. (1993).]. Biol. Chem. 268: 14931-
14939.

11. Wen, W., Meinkoth, J .L., Tsien, R.Y., and Taylor, SS. (1995) Cell 82: 463-473.
12. Meyer, BB. and Malim, M.H. (1994) Genes Dev. 8: 1538-1547.

13. Kalland, K.H., Szilvay, A.M., Brokstad, K.A., Seatrevik, W., and Haukenes, G.
(1994) Mol. Cell. Biol. 14: 7436-7444.

14. Sato, S., Burdett, J., and Hughes, RC. (1993) Exp. Cell Res. 207: 8-18.
15. Gaudin, J .-C., Mehul, B., and Hughes, RC. (2000) Biol. Cell 92: 49-58.

16. Moutsatsos, I.K., Wade, M., Schindler, M., and Wang, J .L. (1987) Proc. Natl. Acad.
Sci. USA. 84: 6452-6456.

17. Hamilton, M.H., Tcherepanova, I., Huibregtse, J .M., and McDonnell, DP. (2001).].
Biol. Chem. 276: 26324-26331.

138

 

18. Parra-Palau, J .L., Scheper, G.C., Wilson, M.L., and Proud, CG. (2003).]. Biol. Chem.

278: 44197-44204.

19. Bohnsack, M.T., Regener, K., Schwappach, B., Saffiich, R., Paraskeva, E.,
Hartmann, B., and Gorlich, D. (2002) EMBO J. 21: 6205-6215.

20. Heger, P., Lohmaier, J ., Schneider, G., Schweimer, K., and Stauber, RH. (2001)
Traffic 2: 544-555.

21. Lee, S.-H. and Hannink, M., J. Biol. Chem. 276: 23599-23606 (2001).

22. Huang, T.T., Kudo, N, Yoshida, M., and Miyamoto, S. (2000) Proc. Natl. Acad. Sci.
USA. 97: 1914-1019.

23. Michael, W.M., Choi, M., and Dreyfuss, G. (1995) Cell 83: 415-422.
24. Michael, W.M., Eder, PS, and Dreyfuss, G. (1997) EMBO J. 16: 3587-3598.
25. Fan, X.C. and Steitz, J.A. (1998) Proc. Natl. Acad. Sci. USA. 95: 15293-15298.

26. Dagher, S.F., Wang, J .L., and Patterson, R.J. (1995) Proc. Natl. Acad. Sci. USA. 92:
1213-1217.

27. Park, J .W., Voss, P.G., Grabski, S., Wang, J .L., and Patterson, R.J. (2001) Nucleic
Acids Res. 27: 3595-3602.

28. Laing, J.G., and Wang, J .L. (1988) Biochemistry 27 : 5329-5334.

29. Hubert, M., Wang, S.-Y., Wang, J .L., Seve, A.-P., and Hubert, J. (1995) Exp. Cell
Res. 220: 397-406.

139

 

Luna
‘«

 

Chapter 5

Concluding Statements

140

Concluding Statements

Prior to this study, localization of galectin-3 to the nuclear and cytoplasmic
compartments had been reported in a variety of cell types. Moreover, the localization of
galectin-3 was also known to be sensitive to a variety of cellular conditions or cues,
including malignant transformation, serum starvation or stimulation, and proliferative
status. Several lines of evidence suggested two key points: (a) galectin-3 might traffic
between the nucleus and cytoplasm; and (b) galectin-3 could be exported from the
nucleus via a putative leucine-rich nuclear export signal. However, the subcellular
trafficking of galectin-3, as well as the mechanism of its movement between the nuclear
and cytoplasmic compartments, had not been rigorously examined.

This study focused on three points: (a) documenting the trafficking of galectin-3
between the nuclear and cytoplasmic compartments (nucleocytoplasmic Shuttling); (b)
identifying the signal(s) required for nuclear import; and (0) testing the activity of the
putative leucine-rich nuclear export Signal in galectin-3. These projects, and the data
derived from them, are descriptive in nature. Subcellular trafficking assays carried out in
heterodikaryons revealed that galectin-3 shuttles between the nuclear and cytoplasmic
compartments, a region of galectin-3 which is sufficient to specify nuclear import was
identified, and the activity of the leucine-rich nuclear export signal was verified.

These descriptive studies now serve as a foundation for more detailed,
mechanistic studies of the subcellular trafficking of galectin-3. At the biochemical level,
for example, the receptor(s) that mediate nuclear import of galectin-3 can be identified,
possibly by using the putative nuclear localization signal to identify interacting partners.

In addition, the activity of the leucine-rich export signal in galectin-3 suggests that the

141

protein might be exported by CRMl, and interaction between the two proteins should be
tested for.

The description of galectin-3 trafficking also opens avenues for biophysical
investigations of galectin-3 dynamics within different subcellular compartments, as well
as between compartments. For example, photobleaching studies could be employed to
determine the diffusion constant of galectin-3 in the cytoplasm and nucleoplasm. A
difference in mobility might indicate constraint of galectin-3 in one compartment,
indicating an interaction (or multiple interactions) that impede mobility. Alternatively,
mobility studies between compartments could provide valuable insight into the rate of
nuclear import and export of galectin-3. For example, a lower rate of galectin-3 nuclear
import than export might indicate a less avid association of galectin-3 with the nuclear
import machinery, a larger complex of machinery to be translocated through the nuclear

pore, or retention of galectin-3 in the cytoplasm via some other interaction.

142

 

A2i73ufi 217