U01 _Oct_12_1630_AS2_YMCx

thingyoutstandingΒιοτεχνολογία

1 Οκτ 2013 (πριν από 4 χρόνια και 3 μήνες)

132 εμφανίσεις

Specific Aims


1 Page

Introductory statement…why we are doing this (< ½ page)

NCDIR, tech dev for structure determination of multicomponent complexes.

This is done by integrating many disparate data types




Specific

Aim 1
. The Nuclear Pore Complex
and Environs

MPR


Specific Aim
2
. Information Flow into the Nucleus.

YMC

This Aim proposes structural, biochemical, biophysical and cell biological analyses of import
-
Kapβs and their
interactions with cargoes. We will determine crystal structures of nine
human import
-
Kapβs that have not
studied structurally (Imp4, Imp5, Imp7, Imp8, Imp9, Imp11, Imp13, TrnSR1/2 and Exp4). We will also
determine structures of nine fungal import
-
Kapβs (Kap104, Kap108, Kap111, Kap114, Kap119, Kap120,
Kap121, Kap122 and Kap123
) since homologous human and fungal Kapβs do not import homologous cargoes
and are thus not necessarily functional orthologs. The structures will serve as baselines for future structural
studies of functional complexes with ligands such as nucleoporins an
d cargoes. They will also be critical for
comparative structural analysis to understand protein motion in these highly flexible, non
-
globular, solenoid
proteins. We will also solve crystal structures of at least one cargo complex for each of the nine hum
an and
nine fungal import
-
Kapβs. Cargoes have been identified for all known human and
S. cerevisiae

import
-
Kapβs.
However, nuclear localization signals (NLSs) recognized by most import
-
Kapβs have not been defined.
Structural analyses of the Kapβ
-
cargo c
omplexes will complement biochemical approaches such as
identification of additional import cargoes for each Kapβ, site
-
directed mutagenesis and NLS
-
mapping of
multiple cargoes to determine both the sequence patterns and structural requirements of NLSs for

recognition
by their Kapβs a
s well as

investigation of nuclear import activities in yeast and mammalian cells. This
information will be used in simple bioinformatics searches to identify new NLS
-
containing cargoes and to
define new classes of NLS recogni
zed by each of the import
-
Kapβs.


Specific Aim
3
.

From Gene to Nuclear Envelope


MPR and JDA

Why is it Important to study RNPs?

-

fundamental to the nuclear information pathway

-

new insights into under
-
studied gene regulatory control mechanism.

-

We have the tools to isolate these complexes

Specific Aim
4
. Usurping the Nuclear Information Pathw
ays

JDA

Focus on Influenza

-


Background and Significance (2
-
3 pages)

Central to the maintenance and behavior of any eukaryotic cell is
the
nuclear informati
on pathway. In this
pathway, c
ells decode external
cues

and transport the
resulting
signals

through the nuclear pore complex

(NPC)
to
the
DNA in the nucleus. There, the

DNA, packaged as silent chromatin, is activated for either replication or
transcription into RNA. The RNA is packaged with proteins into a ribonuclear protein (RNP) complex and spliced; it
is then exported
into the cytoplasm
via
the association
wit
h
export factors and exchange of RNA
-
binding proteins
before being translated into protein (Fig.
XX
).

All the metabolic and signaling pathways in a cell are depend
ent
on this flow of information
.

As a result
,

this pathway is also an Achilles heel for the
cell
. It can be

disrupted in
disease and usurped by viruses.

The word “virus” comes from the Latin, meaning “poison”. A virus is certainly
a poison for the genetic pathway in a cell; in a sense, each virus is a
natural cell biologist, which has
experimented through evolution with the genetic pathway, deleting some of the cell’s own programming and
replacing it with its own (Fig.
XX
).
To be successful
, the virus must interact not only with the g
enetic pathway,
but also to some extent with
many

cellular pathways. Technically, this makes the study of the viral life cycle in
cells a
problem
similar
to the study of the nuclear information pathway.

Proteins are the central players in
the
cellular information pathway
. Most eukaryotic cells express several
thousand proteins, and multicellular organisms commonly express tens of thousands of proteins, exponentially
increasing the probable total number of macromolecular interactions in thes
e organisms,
linking each other
throughout the cell. For example in yeast,
analyses suggest a range of 16,000

26,000 different protein
interaction
pairs
{Grigoriev, 2003 #2}, and this number
is

much higher if
interactions between proteins and
other macromolecules (e.g., DNA, RNA)

are included
.
Given such large numbers of interactions, filtering out
those
of interest to us

represents a formidable challenge, especially
when one considers the dynamic nature of
the majority of these interactions.

The desperate need for technologies that can quickly and reliably
meet this challenge
drives a current
collaboration betw
een us in the
National
Center for Dynamic Interactome Research (NCDIR
; http://ncdir.org
)
.

We
are creating techniques that
allow us to map protein interactions
at a timescale and spatial resolution
appropriate for the elucidation of how cellular processes work.

The primary

goal of
NCDIR

is to create robust,
innovative new tools to enable the visualization, isolation, and
analysis of macromolecular complexes to reveal their dynamic
behavior
. The tools are being
test
ed

on discrete
biological systems found at key steps along the
nuclear information
pathway
; these systems together
encapsulate many of the major problems that are faced in proteomic investigations of cellular processes.
The

rationale of
NCDIR

is to build on the current developments in proteomic technologies

and computational
b
iology
, including affinity isolation of macromolecular complexes, mass spectrometry, microarrays
,

next
generation DNA sequencing
,
and

integrative modeling of assembly structures

based on varied proteomic data
.

By judiciously
adopting
new technologies to address bottlenecks and limitations to
improve
resolution,
accuracy, applicability,
cost,
throughput,
and
speed
,

these technologies are bringing proteomic and
interactomic studies to a whole new level
,

allow
ing

us

to produce enlightening dynamic pictures of how
macromolecular assemblies form and function in the living cell.

Our
approach

currently
yield
s

low to medium
resolution
representations of
molecular architectures of assemblies
. The challenge before us is to extend
the
approach to gain dynamic representations of macromolecular structures at atomic resolution;
to achieve this
goal,

we must have the
atomic structures

that can only be garnered by X
-
ray crystallography and
NMR
sp
ectroscopy
. Moreover, the atomic resolution structures we need must be carefully targeted to maximize our
ability to leverage tools and data obtained thus far. Finally,

this integrated approach will enable us to develop
and disseminate approaches for incorporating atomic resolution structures into dynamic
processes involving
transient
macromolecular
complexes


a challenging but critical
step for
maximizing contributions of
structural
biology

to biology
.

We propose four specific
focus areas
that represent key aspects of t
he nuclear information pathway:
the
NPC
;
protein i
mport

through the NPC
; RN
A assembly and export

through the NPC
; and usur
pation

of the
nuclear
information pathway by influenza virus. The

corresponding
complementary aims build on the expertise and
ongoing research programs of the applicants. Studies on the NPC are ongoing in a collaboration between
MPR and AS as well as Brian Chait and members of the New York
SGX Research Center for Structural
Genomics
(NYSGXRC
; Stephen K. Burley, PI
)
.

Protein import is the focus of YMC’s laboratory
, and has been
an area of intense research in MPR and JDA’s laboratories.
URNA
assembly and export is being studied
from
the structural point of view by the groups of MPR, JDA and AS as part of the NCDIR and in collaboration with
Susan Wente’s laboratory

at Vanderbilt

University
. Finally, the NCDIR has developed technologies to study
virus
-
host interactions (HIV, CMV wi
th Mark Muesing
at Aaron Diamond

AIDS
Research
Center
and Tom
Shenk

at Princeton

University
) which are being extended and applied to Influenza through a collaboration
between JDA and Alan Aderem

at ISB
as part of a larger contr
act involving
drop names here…
As outlined in
the next section, all these research programs have made very significant progress; while each is at a different
level of maturity, the
y all benefit from strong leveraging
that
result
s

from a synergy in their approaches and a
focus on different aspects of the
nuclear information pathway. However, each research program has reached
a point where it can be considerably informed by the data obtained from a large
-
scale
production of
atomic
structur
es

in a way that would otherwise be impossible.

C.
Preliminary Studies

C.
1
The nuclear pore complex (NPC)

We started with a single paradigm system in y
east, the NPC
, in order to determine how
to

study macromolecular
interactions and to hone our proteomic methodologies and
computational
analyses.
NPCs are the sole mediators
of bi
-
directional traffic across the nuclear envelope (NE) between the nucleoplasm

and
cytoplasm. NPCs are
octagonally

symmetric

cylindrical assemblies some 100 nm across

and
50 MDa in mass, whose constituent
proteins are termed nucleoporins (nups).
Transport across the NPC is
rapid
, energy
-
dependent,

and
receptor
-
mediated, depending on the interplay between soluble t
ransport factors, transport cargoes

and
the NPC
(reviewed
in
refs.
{
Macara, 2001 #32; Rout, 2001 #33; Rout, 2003 #34; Suntharalingam, 2003 #35; Weis, 2003 #36}
)
.

C.1.1

A map of the NPC’s molecular organization.

T
he sheer size

and
flexibility of the NPC make it all but
impossible to solve its molecular architecture by conventional
atomic resolution
techniques
, such as
X
-
ray
crystallography and NMR spectroscopy
. We therefore took
an

orthogonal
multi
-
pronged
approach that is at the
h
e
art of the technology being developed within the
NCDIR
.
The most important aspect of
the

approach is its
potential to simultaneously use almost any conceivable type of information to determine
low
-
resolution

(
i
.
e
.
,
low precision)
assembly structures, including
information

not normally used for structure determination
. The
information can be obtained from traditional structural biology

(
i.e.,
X
-
ray crystallography, NMR spectroscopy
,
and
electron microscopy
)
, as well as diverse proteomic, biophysical, biochemical, and systems biology
approaches.
We
first developed methods that determined the composition and stoichiometry of the yeast
NPC
{Rout, 2000}, and then gained a large amount of information concerning t
he relative connectivity of its
component Nups by redundant immunoisolations of Nup subcomplexes
{Alber, 2007A; Alber, 2007B}. This was
complemented with data for each Nup on its approximate shape
from
s
equence comparisons and
ultracentrifugation as well a
s
an immunoEM map of its position within the NPC.
Integration of this data into a
structure is similar to the determination of protein structures by NMR spectroscopy, in which the folding of the
polypeptide chain is determined by satisfying distance re
straints between pairs of atoms
{Havel, 1985 #3465}
.
Here, atoms are replaced by proteins,

and
their positions

and
relative proximities are restrai
ned based on our
proteomic data
. In this way, we determined the position of
nup
s

with a
n average

precision of ~5 nm, sufficient
to resolve the molecular organization of the entire NPC (Fig.
XX
)
{Alber #547; Alber #546}
.

We also used
computational fold predictions

and
biochemical domain mapping to analyze the fold organization of every nup

{Devos, 2004

#985;Devos, 2006 #984
;Dokudovskaya Structure
}
.

C.1.2

Discoveries made using our map of the NPC: Evolution of eukaryotes.

We discovered that the NPC
is surprisingly simple, consisting of a few, highly repetitive fold types. This
indicates

that the bulk of its structure
has evolved through extensive gene duplication from a simple precursor set of only a few proteins. Indeed, we
find that each
octagonal symmetry unit
can be divided into two parallel columns, in which every nup in one
column
contains a similarly

positioned homolog in the adjacent column (Fig.
XX
). This pattern
can

be explained
if an ancient duplication event gave rise to the two columns comprising each
symmetry unit {Alber, 2007 #546}
.
The

NPC’s core scaffold
(roughly half the

mass of the whole NPC
; Fig. XX
)
is made of a set of cage
-
like
structures containing nups composed
almost entirely of either

-
propeller folds,

-
solenoid folds, or an N
-
terminal

-
propeller followed by a C
-
terminal

-
solenoid fold
{Devos, 2004 #985;Devos,

2006 #984;DeGrasse,
2009 #3572}
. Striking
ly, t
here are similarities
between

the structures of the
core NPC scaffold curving

around
the pore membrane and complexes such as clathrin/adaptin, COPI, and COPII (Fig.
XX
)

{Devos, 2004
#985;Devos, 2006 #984 #978}

that support

coat
-
mediated
vesicular
transport between

the plasma membrane
and endomembrane systems such as the Golgi and ER. Indeed
,

the similarities between
core scaffold Nups
and coating complexes
have been borne out in numerous crystallographic studie
s

{Debler, 2008 #3577;Hsia,
2007 #3579;Brohawn, 2008 #3573;Whittle, 2009 #3576;Seo, 2009 #3580;Leksa, 2009 #3575
}
.
These
similarities indicate that
early proto
-
eukaryotes
acquired

a membrane
-
curving protein module,
the “proto
-
coatomer” (
likely composed of
a simple

-
propeller /

-
solenoid protein
)
, that allowed
them
to mold their
plasma membranes into internal compartments. Modern eukaryotes diversified this module into many
specialized
membrane coating complexes
,
accounting
for the evolution of
their
inter
nal membrane systems
{Devos, 2004 #985;Devos, 2006 #984}
.

C.1.3

Discoveries made using our map of the NPC:
A

dynamic molecular understanding of nuclear
transport.

The framework of the NPC serves two key transport purposes: to form a barrier of defined
permeability within the pore, and to facilitate transport of selected macromolecules across it. Both processes
are dependent on the correct positioning of critical Nup
s in the NPC architecture
{Liu, 2005 #922;Radu, 1995
#365;Strawn, 2004 #1055}
. Attached to the inside face of the NPC core scaffold, facing the central tube’s
cavity, are groups of nucleoporins termed

linker nucleoporins

. Together with the inner ring, th
ese seem to
form most of the attachment sites
for

the

nucleoporins

termed “FG Nups

{Alber, 2007 #3240}
. These FG Nups
are the direct mediators of nucleocytoplasmic transport
, by binding reversibly to
transport factors

{Liu, 2005
#922;Radu, 1995 #365;Strawn, 2004 #1055}; their disposition in the architecture of the NPC has suggested a
model for the mechanism of nuclear transport, in which, b
y forming a dense meshwork in and around the
central tube, the FG repeats appear to set up a
n entropic
barrier

that excludes macromolecules from their
vicinity, while permitting the approach of small molecules such as water and nucleotides
{Lim, 2006 #1049}
.


C.1.4

Structure, function, origins and dynamics can be obtained from proteomic

studies.

The NPC
studies illustrate the tremendous potential of the combined proteomic and analytic approaches. In only a
handful of years, they have led us to a basic understanding of the fundamental structure of the NPC, its
evolutionary origin, and the mechanis
m of NPC
-
mediated transport
{Devos, 2004 #86; Rout, 2003 #34; Rout,
2000 #47; Zilman #550}
. T
he significant progress we have made in the functional elucidation of the NPC’s
macromolecular interactions augurs well for the expansion of these kinds of approac
hes to study other
subcellular systems.

C.1.5
The structures cannot be completed without atomic resolution information.
In essence, our
method works by determining which proteins are in a complex, which ones interact, where they interact, what
the shape
and relative position of each protein is in the complex, and how all these data change over time. The
more detailed this information, the higher the resolution of our dynamic maps. We have developed or adapted
methodologies to garner data on all these, up
to and to some degree including the atomic scale (see below).
W
e have not explicitly sought atomic scale data and currently have no infrastructure to do so, relying instead
on published data and
comparative protein structure
modeling. However, published structures do not
necessarily target the most
informative
structures needed to
determine

the

architectur
e

and function

of the
NPC

at high
-
resolution
.
Conversely, atomic structure
s on their own cannot give the “full picture”; we
continue to
develop technologies to integrate these structures into large
-
scale dynamic architectural maps, constituting an
ideal synergy of efforts
. Hence the current proposal seeks to fill this
gap, by commissioning atomic structures
of selected proteins, fragments and interacting domains. These structures will inform all our approaches,
enriching them considerably.

C.2
Nuclear Transport Pathways


Import and Export

While NPCs are freely permeab
le to small molecules and ions, transport of almost all macromolecules into and
out of the nucleus requires the assistance of soluble transport factors that effectively chaperone cargoes
through the NPC

by binding to FG
-
nups (
Section C.1
.xx
)
, most

of which belong to the Karyopherin or Kap
family of proteins. Some of these karyopherins, known as “importins”, specialize in transporting cargos to the
nucleus. Other karyopherins, known as “exportins”, ferry cargoes out
of the nucleus
(reviewed in

refs.
{Stewart,
2007, 17287812}
). Each cargo is targeted for nucleocytoplasmic transport by a short amino acid sequence
termed either a nuclear localization sequence (NLS, for import) or a nuclear export sequence (NES, for export).
Several experimental approa
ches led to a consensus three
-
step model of the b
asic transport mechanism

{Akey, 1989, 2475512}
. In this model, macromolecular NLS/NES
-
containing cargos, which are poised for
transport into or out of the nucleus, are recognized and bound by Kaps. This Kap
-
cargo complex
then docks to
the NPC and translocates through the NPC. Upon reaching its target compartment (either nucleoplasm or
cytoplasm), the complex dissociates. Because molecules lacking effective NLS/NES cannot bind transport
factors and thus are n
ot afforded passage through the NPC, this mechanism also serves as an effective barrier
function. For Kaps, the directionality of transport is enforced by the cellular distribution of the GTP
-

versus

GDP
-
bound states of Ran, a small GTPase
{reviewed in
refs.{
Macara, 2001, 11729264}
)
. RanGTP is
predominantly found in the nucleus and by binding to an importing Kap, drives the release of import cargo
inside the nuclear compartment. In the case of nuclear export, the affinity of an exporting Kap for NES
-
con
taining cargo is increased in the presence of RanGTP. This ferries export cargo to the other side of the
NPC where GTP hydrolysis induces the release of cargo in

the cytoplasm. The resulting Kap

and RanGDP are
recycled back into the transport pathway.

C.
2.1
A first glimpse at the mechanism of import

by
Kap

2
.

Our work on Kaps and transport should be discussed here


Yuh Min

YMC has taken a multidisciplinary approach to understand the mechanisms of
nuclear import by Kap

2 and its
S. cerevisiae

homolog Kap104. Kap

2 is a prototypical Kap, which binds import cargoes and nucleoporins
simultaneously to target
cargoe
s to the NPC. In the nucleus, Kap

2 binds RanGTP with high affinity and
cargo
is released
. Prior to YMC’s work,
~20 mRNA processing
proteins (including hnRNPs A1, D, F, M, HuR, DDX3,
Y
-
box binding protein 1 and TAP) had been identified as Kap

2 cargoes {Kawamura, 2002 #1530;Suzuki,
2005 #1529;Truant, 1999 #1442;Guttinger, 2004 #1527;Rebane, 2004 #1526;Fan, 1998 #13;Siomi, 1997
#644;Pol
lard, 1996 #590;Bonifaci, 1997 #387}. Kap

2 binds its best
-
characterized
cargo
, splicing factor
hnRNP A1, through the 38
-
residue M9 sequence {Pollard, 1996 #590;Bonifaci, 1997 #387;Siomi, 1995
#1444;Weighardt, 1995 #1537}. NLSs in HuR {Fan, 1998 #13}, TA
P
/NXF1

{Truant, 1999 #1442}, hnRNP D
and its homologs, the JKTPB proteins had previously been mapped but showed marginal
to

no sequence
homology to the hnRNP A1
-
NLS {Suzuki, 2005 #1529;Kawamura, 2002 #1530}. Sequence diversity of NLSs
recognized by Kap

2
had prevented prediction of new NLSs in this pathway.

C.2.1.1.
The mechanism of Ran
-
mediated cargo dissociation.

YMC solved a series of four Kap

2
structures to complete the structural snapshots of this nuclear import pathway {Chook, 1999
#1665;Cansizoglu, 2007 #1572;Cansizoglu, 2007 #1694;Lee, 2006 #1564}. The first Kap

2 crystal structure
was of full length Kap

2 bound to Ran (Fig. 2
A
) {Chook, 1999 #1665}, which provided rationale for the
specificity of Kap


for the GTP state of Ran, all
owed identification of
cargo

binding site, and provided insights
into higher order interactions with other regulators and effectors of Ran. The structure and additional
biochemical studies also suggested a mechanism for Ran
-
mediated cargo dissociation: Ra
n and cargo bind
spatially distinct sites but are thermodynamically coupled by the internal H8 loop that occupies the
cargo

site
and displaces cargo when Ran is bound {Chook, 1999 #1665; Chook, 2002 #1668}. The next structure
,
that of
Kap

2 bound to cargo

hnRNP A1
-
NLS (or M9 sequence), confirmed this structural mechanism and further
suggested a biophysical mechanism involving electrostatic effects of Ran binding driving the Kap

2 loop into
the cargo site (Fig. 2
A
)

{Lee, 2006 #1564}.

C.2.1.2. Discovery of a

new class of NLS, the PY
-
NLS.

More importantly, combined
structural and
biochemical analyses of the Kap

2
-
hnRNP A1
-
NLS complex revealed physical rules that described the
recognition of NLS by Kap

2. The first rule emerged from the extended conformation o
f the 26
-
residue bound
hnRNP A1
-
NLS, which suggested that an NLS recognized by Kap

2 should exist within a stretch of at least 30
residues that is structurally disordered in its native state

(Figure 2A)
. The second rule emerged from the
observation that
t
he
NLS

interface
of
Kap

2

is highly negatively charged and thus an NLS with overall positive
charge would be favored. The structure also guided biochemical studies to understand the distribution of
binding energy along the hnRNP A1
-
NLS. We measured
affinities of hnRNP A1
-
NLS mutants binding to
Kap

2 using isothermal titration calorimetry (ITC) and biochemically mapped NLSs of three previously
uncharacterized Kap

2 cargoes. Despite apparent diversity of the resulting seven NLS sequences,
mutagenesis
and sequence analysis identified two regions of conservation, allowing us to propose a set of
consensus sequences. Together, structural and biochemical analyses of Kap

2 bound to the NLS of hnRNP
A1 revealed three rules that define signal recognition: 1)
intrinsic structural disorder, 2) overall basic character
and 3) a set of weakly conserved sequence motifs composed of a loose N
-
terminal hydrophobic or basic motif
and a C
-
terminal RX
2
-
5
PY motif. The composition of N
-
terminal motifs divides PY
-
NLSs into
hydrophobic and
basic subclasses (hPY
-

and bPY
-
NLSs).

hPY
-
NLSs contain four consecutive predominantly hydrophobic
residues, while the equivalent region in bPY
-
NLSs is enriched in basic residues (Fig. 2
B
). These rules
described the recognition of numerous

signals that previously seemed unrelated and unified them into a new
NLS class named PY
-
NLS. The NLS rules were not strong filters individually, but together they were predictive
and revealed new NLSs in seven known Kap

2 cargoes and 81 new candidate car
goes when applied in a
bioinformatics effort. These new NLSs are complex signals, discovered using a collection of individually weak
rules rather than just a strongly restrictive sequence motif. This

multidisciplinary approach anchored in
structural and b
iophysical analyses successfully advanced biology by
defining a new NLS that could not be
predicted by primary sequence analysis alone
. Many uncharacterized NLSs/NESs are poorly defined in
sequence, and the multidisciplinary approach we used for
Kap

2

will likely be applicable to identify analogous
signals across the entire
Kap


family

(Aim 2)
.
More generally, this concept of signals as a collection of

Fig. 2.

Nu c l e a r i mp o r t b y
Kap

2.
A.

Ribbon diagrams of unliganded,
hnRNP A1
-
NLS
-
, hnRNP M
-
NLS
-

and Ran
-
bound Kapβ2s. Unliganded
Kapβ2 is blue, Kapβ2 bound to
cargoe
s are pink, Kapβ2 bound to Ran is
red and H8 loops are yellow.
Cargoe
s hnRNP A1
-
NLS is green, hnRNP M
-
NLS is magenta and Ran is drawn as surface representation in grey.
B
.

NLSs of hnRNPs M (magenta) and A1 (blue) upon superposition of Kap

2
residues 435
-
780. Regions of structural similarity are highlighted with yellow
ovals.
C.

Loss of
Kapβ2
-
binding energy in alanine mutants of hnRNP A1 and
hnRNP M.
D.

Affinities of
Kapβ2

binding to cargoes hnRNP A1, M,
RanGTP and designed

peptide

inhibitor M9M
.
E.

H i s t o g r a m s h o w s
p e r c e n t a g e s o f t r a n s f e c t e d c e l l s w i t h c y t o p l a s m i c
K a p β 2

c a r g o e s. H e L a
c e l l s w e r e t r a n s f e c t e d w i t h p l a s m i d s e n c o d i n g M y c
-
t a g g e d M B P o r M B P
-
M 9 M, a n d i m m u n o f l u o r e s c e n c e a n d d e c o n v o l u t i o n m i c r o s c o p y

p e r f o r m e d
u s i n g a n t i b o d i e s t o h n R N P A 1, M
a n d H u R.


p h y s i c a l r u l e s r a t h e r t h a n s p e c i f i c s e q u e n c e m o t i f s a l o n e, c o u l d a l s o b e e x p a n d e d a c r o s s o r g a n e l l e s
y s t e m s t o
s t u d y t h e n u m e r o u s o b s c u r e t a r g e t i n g s i g n a l s i n e u k a r y o t i c c e l l s.

C.2..3. S t r u c t u r e
-
b a s e d d e s i g n o f a n u c l e a r i m p o r t i n h i b i t o r.

Structural comparison of Kap

2
-
hnRNP A1
-
NLS (hPY
-
NLS) and Kap

2
-
hnRNP M
-
NLS (bPY
-
NLS) complexes (Fig

2B
) explained recognition of the
chemically diverse motifs {Cansizoglu,
2007 #157
2}. The hPY
-

and bPY
-
NLSs
converged structurally only at consensus
sequence motifs, confirming the
consensus designations and suggesting
multipartite interaction. Thermodynamic
analyses of hnRNP M
-
NLS and hnRNP
A1
-
NLS mutants revealed asymmetric
binding
hotspots in the two sequentially
diverse NLSs

(Figure 2C)
. This finding
allowed harness
ing of

the avidity effect to
design chimeric peptides with enhanced
Kap

2 binding affinities that compete with
natural
cargoes
, be resistant to Ran
-
mediated release in
the nucleus, and
thus may function as a nuclear import
inhibitor. A chimeric peptide that fuse
d

the N
-
terminal half of hnRNP A1
-
NLS to
the C
-
terminal half of hnRNP M
-
NLS
bound Kap

2 200
-
fold tighter than natural
ligands and specifically inhibited Kap

2
nu
clear import in cells

(Figs 2D and E)
.
This first pathway
-
specific nuclear import
inhibitor is usefulness for the biological
community to determine nuclear import
pathways of macromolecules. In
addition, the inhibitor is critical for
systems
-
scale
in vi
vo

validation of new
Kap

2
cargoes

and for mapping potential
redundancy in nuclear import networks.
Finally, the inhibitor works because it
binds Kap

2 so tightly that it out
competes natural
cargoe
s and Ran can’t
displace it. This inhibition mechanism
le
nds support to the concept that a Kap

-
NLS interaction should occur within a
range of affinity suitable for both binding
and release.
This

inhibitor has provided
a high affinity limit of the range for Kap

2.

C.2.1.4. Biophysical abd cellular
analyses of
import by Kap104.

Multipartite PY
-
NLSs are highly diverse in
sequence and structure. Their variability
is consistent with weak consensus motifs,
but such diversity potentially renders
comprehensive genome
-
scale searches int
ractable. Studies using Kap104

(
S. cerevisiae

Kap

2 homolog
{Aitchison, 1996 #192;Lee, 1999 #1414}) as a model system to understand the energetic organization of PY
-
NLS have informed on how Kap

2 recognizes
such diverse sequences. Kap104

cargoe
s
also contain PY
-
NLSs but Kap104

binds s
pecifically only to bPY
-

but not hPY
-
NLSs. Thermodynamic studies of Kap104p
-
NLS
revealed physical properties that govern PY
-
NLS binding affinity: 1) PY
-
NLSs contain three energetically
significant linear epitopes, 2) each epitope accommodates substantial
sequence diversity, within defined limits,
3) the epitopes are energetically quasi
-
independent and 4) a given linear epitope can contribute differently to
total binding energy in different PY
-
NLSs, amplifying signal diversity through combinatorial mixing o
f
energetically weak and strong motifs. The modular organization of PY
-
NLS coupled with its combinatorial
energetics lays a path to decode this diverse and evolvable signal for future comprehensive genome
-
scale
identification of nuclear import
cargoe
s.

T
he

PY
-
NLS predictive rules
also
identified

potential Kap104 cargoes
in a bioinformatics search.
YMC

confirmed that one of these putative cargoes, the yeast general transcription
factor Tfg2p (a subunit of TFIIF),indeed contains a functional PY
-
NLS

{Suel,
2009 #1731}
. Energetic
dissection of this signal supports previously identified physical properties and suggests additional constraints
for the consensus sequence of PY
-
NLS. More importantly,
the

steady state nuclear localization of Tfg2p is
governed by
two independent mechanisms: 1) import into the nucleus through recognition of its PY
-
NLS, and
2) nuclear retention through its DNA binding domain. These results emphasize
d

the often
-
overlooked concept
that the subcellular localization of a protein is not dictated solely by the presence or absence of a nuclear
import or export signal. Rather, localization is a sum of many factors including relative strengths of
import/expor
t signals and nuclear/cytoplasmic retention.

C.2.1.5. Conformational flexibility
in the
Kapβ2 pathway.

Finally, YMC solved the crystal structure of
unliganded Kapβ2, which completed the structural snapshots of the Kapβ2 nuclear import pathway. Previously
determined structures
had suggested that
Kaps

undergo large conformational changes upon binding differe
nt
ligands. However, there were few
systematic
studies of Kap conformational flexibility and most involve only
qualitative descriptions of regions of structural changes. Their unusual non
-
globular, non
-
modular and HEAT
-
repeat containing solenoid architec
ture, and the limited number of structures within each Kap series
complicated analyses of conformational flexibility.
Availability of four full length Kapβ2 structures allowed the
first comprehensive and quantitative analyses of conformational heterogenei
ty for a protein with non
-
globular
repeat architecture. These analyses involved
independent methods of domain motion analyses by rotation
vector clustering, B
-
factor groupings and structural superpositions.
Multiple molecules of unliganded Kapβ2 in
the c
rystallographic asymmetric unit allowed studies of its intrinsic flexibility. Conformational flexibility of
Kapβ2, both intrinsic and ligand
-
induced, is segmental, with movements occurring between three major
segments. Two segments are rigid and rotate a
bout a hinge, and one shows conformational changes
throughout its length.

S
imilar segmental architecture

of
conformational flexibility

was also observed with
Kapβ1
, suggesting generality across the nuclear import factors. More generally, this
approach

of
studying
protein flexibility could be expanded for all import
-
Kapβs as proposed for Aim 2.





include Aitchison and Rout work on Kaps and transport.


C.3
RNP
p
rocessing
and
e
xport

The communication of correct information along the
nuclear information pathway does not only depend on
proper conversion of a gene’s sequence into RNA, but also on the correct subsequent processing of these
RNAs. In recent years, it has become clear that RNAs play an even more significant role within cells

than
previously thought; they are not only the conveyer of genetic information in the form of messenger RNA
(mRNA), but can themselves be regulators of gene expression and other important cellular pathways
{Costa,
2007 #1060}.

All RNA
s

are

transcribed and
assembled into
ribonucleoprotein
(RNP) complexes, in order to
be

modified, processed and transported to their final destination within the cell. It

has been established that

these
processes occur

by different pathways defined by
specific processing

factors
. These factors

form discrete
subsets of proteins that associate with each specie
s

of RNA in a dynamic fashion
to define the order of
maturation events, allowing for precise transcript maturation{Fatica, 2002 #1334;Hopper, 2003
#1445;Vincig
uerra, 2004 #542;Costa, 2007 #1060}.
While some RNAs can then use a Kap mediated nuclear
export mechanism
{Hellmuth, 1998, 9774653}
, mRNA uses a Kap
-

and Ran
-
independent mechanism for
export, in which after being processed and packaged into mRNP (messenger
RNP
) particles, non
-
Kap TFs
such as Mex67 and Mtr2
associate which chaperone the mRNP out through the NPC
{Erk
mann, 2004,
15120988}
.

mRNP assembly and export involve

strict surveillance mechanisms to ensure that only fully mature and
functional RNPs are transpo
rted to the cytoplasm and

there are many different species of mRNA, each
potentially with its own particular maturation pathway
{Hieronymus, 2003 #1734}.
While many features of pre
-
rRNA processing steps themselves are relatively clear
{Venema, 1999 #1300}
,
we still have only a poor

understanding
of how these factors
interact dynamically to
facilitate
the
different processing steps.
There is
therefore a pressing need for development of technologies to follow and understand these dynamic changes,
in terms of the assembling structures
of RNPs.

C
.3.1
Proteomics studies of mRNP complexes

NCDIR
expends a
significant
effort on
studying mRNP assembly and maturation as part of the nuclear export
pathway and regulatory control. Through this program we have developed new methodologies for revealing
the protein and nucleic acid
composition
of large mRNP complexes. We
are currently acquiring various types
of low
-
resolution structural data on these complexes to inform structure determination through the methods
employed for the
NPC
.

W
e developed
cryolysis and rapid isolation techniques
optimized to maximize recovery
and purity of intact
RNP complexes

{Oeffinger, 2007 #1803}
,
sufficient to
routinely
recover >90% of the tagged
complex in every case studied (Fig
.

XX
)
.
These
procedures
ensure that the entire pool of RNPs associated
with our
tagged protein is captured
.



The versatility of the novel affinity purification approach was demonstrated by the isolation of two different
types of RNP complexes (Fig.
XX
)
.
The first survey included th
e pre
-
rRNA processing factors Nop15p, Rpf2p
and Nop
7p{Oeffinger, 2002 #1645;Oeffinger, 2003 #1129;Zhang, 2007 #1809;Oeffinger, 2007 #1803}
. The
second survey focused on
six proteins from different stages (early nuclear to cytoplasmic) of the mRNA
maturation pathway
:

Cbp80p, a component of the cap binding complex

{Rigaut, 1999 #1115}; Thp2p, a
component of the THO complex
that
is associated with early mRNPs{Jimeno, 2002 #751}; Yra1p, a
n

RNA
binding protein that interacts with Mex67p and is required for mRNA expo
rt{Strasser, 2000 #847} and Sac3p,
a
n

mRNP component of very low abundance that forms a complex with Thp1p and is NPC
-
associated{Fischer,
2002 #397}; the mRNA export factor Mex67p; and She2p, a well
-
studied component of the cytoplas
mic RNA
localization mac
hinery

(Fig XX)
.

Details of this study can be found in
{
REF
}
.

In all cases
,

we were able to
isolate factors that had not previously been found in affinity purified RNPs isolated using these tagged proteins
(Fig.
XX
)
, thus

demonstrating that this
rapid
method enables us to preserve far more interactions than
other
current

methods. Repetition of affinity purifications of RNPs associated with Mex67p and Nop15p using GFP
-
tagged version of these proteins isolated the same subset of factors for e
ither protein
as
their PrA
-
tagged
counterparts. This
result
shows that the efficiency of the method is not dependent on any particular protein tag
and that it
can
therefore
be readily adapted to other appropriately tagged constructs.

Beyond
identifying proteins present in RNP complexes, this approach is suitable for purifying and analyzing

the
RNAs associated with tagged

protein components by
using mic
roarrays{Oeffinger, 2007 #1803} (or next
generation sequencing methods).

Using custom

made
Agilent arrays
, RNAs co
-
isolating with Nop15p,
Mex67p, She2p, Sac3p, Yra1p, Cbp80p and Thp2p have been analyzed. These preliminary data showed high
degrees of specificity in the RNPs and provide a foundation for a deeper analysis of the composition and
st
ructure of RNPs along the nuclear export and assembly pathway.

Moreover, the purifications for different mRNPs show interactions that have previously been suggested for
these complexes based on separate isolations and
bioinformatics
prediction
s
, but

have never been shown in
its
entirety by affinity purification in a single pullout experiment.
Our newly i
solated pre
-
ribosomal particles
provided a more detailed picture into the dynamics of these particles, isolating a large
number of factors many
of which have never been found together in the same affinity purified isolate
,

but according to functional data
were expected to be present{Dez, 2004 #224}.

We were further able to
show
that the presence
of
these factors
was

not due to rearrangements taking place
in
the tagged complexes during cryo
-
lysis and sample
preparation. Therefore, all the interactions we have isolated
likely
represent

the
in vivo
situation.

We have now expanded the survey and selected 25
known mRNP maturation factors to get a deeper insight
into the pathway. The proteins were selected according to their involvement in different stages of the pathway,
from transcription, splicing, 3’
-
end formation to export and degradation of faulty complex
es. All the selected
proteins have been PrA
-
tagged and ideal purification conditions are being determined empirically. The isolated
complexes are being subjected to mass spectrometry for an analysis of complex components and co
-
purifying
mRNAs are being an
alyzed by microarray and Solexa analysis
to determine which different classes of mRNAs
are processed by pathways involving selected protein factors. Th
ese efforts
set us up to determine which key
mRNP complexes will be analyzed in Specific

Aim
3
.

We have

an outstanding opportunity to integrate
our
various
proteomics
data
with
atomic

structure
s
, resulting in

methods and approaches for elucidating the
structures
of
a
hierarchically assembling macromolecular complex.


C.4
Viruses As Tools To Investigate The Nuclear Information Pathway

The NCDIR has funded
two

projects
, on
human immunodeficiency virus
(
HIV
)

and
human cytomegalovirus
(HCMV)
,

aimed at discovering the interactions between viral proteins and host proteins. In this way,
integrative, synergistic methodologies of NCDIR have been used to visualize dynamic macromolecular
assemblies, using nuclear information pathway as a model syste
m, including “usurped” pathways.
In both
cases, we have (genomically) integrated epitope tags into the viral genome, and used cryolysis and ra
pid
affinity purification techniques to discover interactions between viral and host
proteins
.

C.4.1 HIV

In
the case of
HIV
-
1
,

t
he extraordinarily compact nature of the viral coding sequences has made the genomic
tagging of HIV extremely challenging. However, our collaboration with Mark Muesing and
David Ho

at the
Aaron Diamond
AIDS Research Center has allowed the

development of a technique to select for genomically
-
tagged, functional HIV variants.
Libraries (~10
5
) of random and independent HIV
-
1 clones were selected to find
those rare sites within the viral proteome that can accommodate a potent immunogenic tag, y
et remain fully
replication
-
competent through multiple rounds of infection.

These modifications allowed for the affinity isolation
of viral
-
host macromolecular complexes from cells infected with the viable tagged viruses.
We analyzed three
independently ta
gged derivatives in the integrase, viral infectivity factor (Vif) and envelope proteins, all
essential for viral replication. Mass spectrometry was used to identify

cellular complexes that interact with the
targeted viral machinery, using heavy isotope lab
eling to discriminate
in vivo

interactors from contaminants
.

Among the
unique

proteins recovered and identified from infected culture, several
have been previously
reported to associate with HIV
-
1 integrase (i.e., SNF5 or INI1, Ku80 and DNA
-
PK), whereas
others
,

such as
p85 MCM
,

ha
ve

not. This is also true for the Env
-
3xFLAG immunoprecipitation. Here, the HIV
-
1 receptor,
CD4, required for viral entry into the cell was
identified as well as BiP, a major ER
chaperone
. Similar
assignments could be made for
Vif (Fig.
xx
).


The

newly discovered components of the viral
-
host interactome
provide the impetus for subsequent investigation of those host factors required by the pathogen

some of
which may be
relatively refractory to the emergence of viral resistance
and thus good
targets for antiviral
intervention
.


C.4.
2

HCMV

In the case of
HCMV
,

the
genome encodes for 208 open reading frames. Attachment to the cell surface is
followed by penetration and
fusion of the viral envelope with cell membranes. Upon entry into the cytoplasm,
viral nucleocapsids make their way to NPCs and deliver viral DNA to the nucleus, dramatically reversing the
normal direction of flow of genetic information through the NPC. Af
ter its replication, HCMV DNA is packaged
into preformed nuc
leocapsids in the cell nucleus
. Rather than exiting the nucleus by nuclear transport, progeny
virions appear to use an alternative pathway by acquiring an envelope from the inner nuclear membrane
and
then moving into cytoplasmic vesicles, which are transported to the cell surface via the Golgi apparatus. To
date, however, little is known about how the maturing CMV interacts with cellular proteins
to subvert its host’s
genetic information p
athway.

A number of the proteins encoded by HCMV have been genomically tagged with GFP in the viral genome in
Thomas Shenk’s laboratory at Princeton University, in order to follow their normal patterns of expression,
localization and
roles
in vivo
.
Th
e Shenk

laboratory has collaborated with us to test two of these constructs as
potential immunoaffinity tags, by isolating them from cells infected with the tagged CMV strains. These pilot
studies revealed that we could successfully isolate specific
co
mplexes
in novel and functionally relevant
complexes, a significant achievement that opens the way to a comprehensive study of the dynamic CMV
-
host
cell interaction during infection.

For example
,
isolation of pUL38, identified
TSC1/2, which
integrates
stress
signals and regulates the mammalian target of
the
rapamycin complex 1 (mTORC1), a protein complex that
responds to stress by limiting protein synthesis and cell growth. We
also
showed that
through this interaction,
pUL38 supports virus replication a
t least in part by blocking cellular responses to stress.


pUL38
immunopurification also
identified

si
x
subunits

involved in

the
nucleosome remodeling and histone
deacetylation (NuRD)
(
Mi
-
2b, MTA1 and 2, HDAC1 and 2, and
RbAp48/46
)
,
which repress
es

transcription
.

This
finding
leads us to the hypothesis that

that HCMV antagonizes NuRD to optimize expression of its
genome
; the

hypothesis

is currently being investigated
, as are
numerous additional
viral
-
host protein
interactions
.

C.4.3 Influenza

We have recently embarked on a close collaboration with a large
inter
-
institutional consortium
dedicated to
using systems approaches to develop a detailed network model of the interaction between the influenza virus
and the host innate immune response. The program is focused on understanding how different influenza
strains lead to high and low patho
genicity infections in a murine influenza model. The long
-
term goal of the
program is to provide new approaches for the development of improved vaccines and immunotherapeutics
directed against the emerging threat of pandemic influenza. This study is part
icularly germane given the
paucity of information on the innate immune response to influenza, the recent insight
being
that an over
exuberant inflammatory response appears to be responsible for the devastating virulence of the 1918 influenza
pandemic
strain, and the emergence of 2009 H1N1 flu (swine flu).

The link between the NCDIR and the
Flu program
lies in proteomic approaches to identify the compendium of
host proteins that physically interact with influenza proteins. This information is being
integrated with the host
regulatory networks defined through systems biology approaches and will provide insights into the mechanism
by which virulence factors modify host responses. Specifically, we will study the interactions between viral and
host prot
eins, integrate the data from the systems approaches to the immune response, and prioritize structure
determination of viral
-
host interactions that are implicated in highly pathogenic infection. Such an approach
holds tremendous potential for gaining fund
amental insights into the nuclear information pathway itself, the
virus life cycle, how the virus usurps the nuclear information pathway, and ultimately rational drug design to
disrupt the ability of the virus to propagate within cells.

Influenza viruses

possess a segmented genome consisting of 8 single
-
stranded, negative sense RNA
molecules. Five of these segments code for a single polypeptide each:
the polymerase proteins PB2 and PA,
hemagglutinin (HA), neuraminidase (NA), and the nucleoprotein (NP).

The remaining three segments code for
two proteins each using alternative reading frames. One gene codes for the matrix protein (MP) and the
transmembrane proton channel M2. A second gene codes for two non
-
structural proteins NS1 and NS2, and a
third cod
es for a basic polymerase PB1 and a pro
-
apoptotic protein, PB1
-
F2. Influenza A viruses are
designated by the subtype of their HA (16 known subtypes) and NA genes (9 known subtypes). For example,
a virus with a subtype
-
5 HA and a subtype
-
1 NA would be desi
gnated H5N1.

Despite an intensive research focus on influenza infection, the mechanisms by which some strains of the virus
induce a highly

pathogenic phenotype remain unclear. Although several mechanisms have been described
based on studies of indivi
dual host and viral factors, a quantitative account of the relative contribution of these
factors to pathology and the interplay between them has not been developed.
The
Flu Glue

is a
coordinated
systems biology program that utilizes a combination of computational and experimental methodologies to
understand the overall complexity of the molecular events within the pathogen and the host
that

lead to severe
disease. This informati
on
promises to

reveal unique molecular signatures

and interactions

that may represent
targets for future intervention by aiding in the development of new drugs, vaccines, and diagnostics to combat
the emerging threat of a future influenza pandemic.

The int
rinsic properties leading to the high virulence of the 1918 H1N1 and the currently circulating avian H5N1
influenza viruses have begun to be identified. Multiple genes are required to confer high virulence for each of
these viruses, including the HA, the
polymerase genes PB1 and PB2, and NS1.

The influenza NS1 gene codes for a non
-
structural viral protein that is believed to be present only in the
infected cell and not the virion itself. NS1 has been shown to contribute to the rate of viral replication
by
blunting the host interferon response. NS1 sequesters viral RNA within the infected cell
{Garcia
-
Sastre, 1998
#63;Seo, 2002 #65;Seo, 2004 #64}

and forms an inhibitory complex with the host pattern recognition receptor
retinoic acid
-
inducible gene I (RI
G
-
I)
{Mibayashi, 2007 #66}
. The NS1 proteins from the 1918 pandemic virus
{Geiss, 2002 #67}

and the currently circulating HPAI H5N1 virus
{Seo, 2002 #65;Seo, 2004 #64}

maintain
these properties. Moreover, the multivalent properties of NS1 have been shown

to promote viral replication
through activation of the phosphatidylinositol
-
3
-
kinase signaling pathway
{Hale, 2006 #68}
, inhibition of host
mRNA processing
{Krug, 2003 #69;Nemeroff, 1998 #70}
, and inhibition of host mRNA nuclear export
{Satterly,
2007 #31
}
.

AITCHISON


PRELIMINARY RESULTS FROM NS1

C.5 Integrative
atomic resolution modeling of macromolecular assemblies



1.

Kawamura, H., Tomozoe, Y., et al. (2002). "Identification of the nucleocytoplasmic shuttling
sequence of heterogeneous nuclear ribonucleoprotein D
-
like protein JKTBP and its interaction with
mRNA."
J Biol Chem

277(4): 2732
-
9.

2.

Suzuki, M., Iijima, M., et al. (2005). "Two separate regions essential for nuclear import of the
hnRNP D nucleocytoplasmic shuttling sequence."
Febs J

272(15): 3975
-
87.

3.

Truant, R., Kang, Y. and Cullen, B. R. (1999). "The human tap nuclear RNA export fac
tor contains a
novel transportin
-
dependent nuclear localization signal that lacks nuclear export signal function."
J Biol
Chem

274(45): 32167
-
71.

4.

Guttinger, S., Muhlhausser, P., Koller
-
Eichhorn, R., Brennecke, J. and Kutay, U. (2004).
"Transportin2 func
tions as importin and mediates nuclear import of HuR."
Proc Natl Acad Sci U S A

101(9): 2918
-
23.

5.

Rebane, A., Aab, A. and Steitz, J. A. (2004). "Transportins 1 and 2 are redundant nuclear import
factors for hnRNP A1 and HuR."
Rna

10(4): 590
-
9.

6.

Fan, X.

C. and Steitz, J. A. (1998). "HNS, a nuclear
-
cytoplasmic shuttling sequence in HuR."
Proc
Natl Acad Sci U S A

95(26): 15293
-
8.

7.

Siomi, M. C., Eder, P. S., et al. (1997). "Transportin
-
mediated nuclear import of heterogeneous
nuclear RNP proteins."
J Cell

Biol

138(6): 1181
-
92.

8.

Pollard, V. W., Michael, W. M., et al. (1996). "A novel receptor
-
mediated nuclear protein import
pathway."
Cell

86(6): 985
-
94.

9.

Bonifaci, N., Moroianu, J., Radu, A. and Blobel, G. (1997). "Karyopherin beta2 mediates nuclear
impo
rt of a mRNA binding protein."
Proc Natl Acad Sci U S A

94(10): 5055
-
60.

10.

Siomi, H. and Dreyfuss, G. (1995). "A nuclear localization domain in the hnRNP A1 protein."
J Cell
Biol

129(3): 551
-
560.

11.

Weighardt, F., Biamonti, G. and Riva, S. (1995). "Nucl
eo
-
cytoplasmic distribution of human hnRNP
proteins: a search for the targeting domains in hnRNP A1."
J Cell Sci

108 ( Pt 2): 545
-
55.

12.

Chook, Y. M. and Blobel, G. (1999). "Structure of the nuclear transport complex karyopherin
-
beta2
-
Ran x GppNHp."
Natur
e

399(6733): 230
-
7.

13.

Cansizoglu, A. E., Lee, B. J., Zhang, Z. C., Fontoura, B. M. and Chook, Y. M. (2007). "Structure
-
based
design of a pathway
-
specific nuclear import inhibitor."
Nat Struct Mol Biol

14(5): 452
-
4.

14.

Cansizoglu, A. E. and Chook, Y. M.
(2007). "Conformational heterogeneity of karyopherin beta2 is
segmental."
Structure

15(11): 1431
-
41.

15.

Lee, B. J., Cansizoglu, A. E., et al. (2006). "Rules for nuclear localization sequence recognition by
karyopherin beta 2."
Cell

126(3): 543
-
58.

16.

Aitchison, J. D., Blobel, G. and Rout, M. P. (1996). "Kap104p: a karyopherin involved in the nuclear
transport of messenger RNA binding proteins."
Science

274(5287): 624
-
7.

17.

Lee, D. C. and Aitchison, J. D. (1999). "Kap104p
-
mediated nuclear import. Nucle
ar localization
signals in mRNA
-
binding proteins and the role of Ran and Rna."
J Biol Chem

274(41): 29031
-
7.