ArticlePDF AvailableLiterature Review

RNA transport

February 1997
Annual Review of Neuroscience 20(1):269-301

DOI:10.1146/annurev.neuro.20.1.269

Source
PubMed

Authors:

University of Wuerzburg

RNA molecules synthesized in the nucleus are transported to their sites of function throughout the eukaryotic cell by specific transport pathways. This review focuses on transport of messenger RNA, small nuclear RNA, ribosomal RNA, and transfer RNA between the nucleus and the cytoplasm. The general molecular mechanisms involved in nucleocytoplasmic transport of RNA are only beginning to be understood. However, during the past few years, substantial progress has been made. A major theme that emerges from recent studies of RNA transport is that specific signals mediate the transport of each class of RNA, and these signals are provided largely by the specific proteins with which each RNA is associated.

Potential trans -acting factors for mRNA nuclear export. Messenger RNA is transcribed in the nucleus by RNA polymerase II as a precursor with an m 7 G cap, and, in the majority of higher eukaryotic transcripts, introns and a 3 poly(A) tail. Evidence indicates that shuttling hnRNP proteins mediate export of mRNA from the nucleus to the cytoplasm. Interaction with spliceosomes and nonshuttling hnRNP proteins prevents mRNA export. The roles of the cap-binding complex (CBC), the Ran GTPase cycle, and individual nuclear pore complex (NPC) proteins are less clear. See text for details.

…

RNA export from the nucleus into the cytoplasm is mediated by RNA-binding proteins. Proteins that specifically interact with each class of RNA carry the signals for RNA transport to the cytoplasm. Messenger RNA export is most likely mediated by shuttling hnRNP proteins; U snRNA export is mediated by the cap-binding complex (CBC); and 5S rRNA export is mediated by TFIIIA and L5. Factors involved in tRNA and rRNA transport remain to be identified.

…

Figures - uploaded by Utz Fischer

Content may be subject to copyright.

Content uploaded by Utz Fischer

Content may be subject to copyright.

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

Annu. Rev. Neurosci. 1997. 20:269–301



RNA TRANSPORT

Sara Nakielny, Utz Fischer, W. Matthew Michael,

and Gideon Dreyfuss

Howard Hughes Medical Institute, and Department of Biochemistry and Biophysics,

University of Pennsylvania School of Medicine, Philadelphia,

Pennsylvania 19104-6148

KEY WORDS: nucleocytoplasmic trafﬁcking, mRNA, U snRNA, rRNA, tRNA

ABSTRACT

RNA molecules synthesized in the nucleus are transported to their sites of func-

tion throughout the eukaryotic cell by speciﬁc transport pathways. This review

focuses on transport of messenger RNA, small nuclear RNA, ribosomal RNA,

and transfer RNA between the nucleus and the cytoplasm. The general molecular

mechanisms involved in nucleocytoplasmic transport of RNA are only beginning

to be understood. However, during the past few years, substantial progress has

been made. A major theme that emerges from recent studies of RNA transport is

that speciﬁc signals mediatethe transport of each class of RNA, and these signals

are provided largely by the speciﬁc proteins with which each RNA is associated.

INTRODUCTION

Most eukaryotic RNAs are produced in the nucleus by RNA polymerase I, II,

or III. The RNA molecules undergo a variety of posttranscriptional processing

events, after which they are transported to their sites of function throughout

the cell. Clearly, the subcellular locations of each type of RNA—namely mes-

senger RNA (mRNA), small nuclear RNA (snRNA), ribosomal RNA (rRNA),

and transfer RNA (tRNA)—are critical to the normal functions of the cell.

A few RNAs are retained in the nucleus, and may be targeted to subnuclear

domains, but the majority of RNAs are transported to the cytoplasm. Some

RNAs need to be reimported to the nucleus for their function, for example U

snRNA and 5S rRNA.

The exchangeof macromolecules between thenucleus and the cytoplasm oc-

curs mostly, if not exclusively, through nuclear pore complexes (NPCs). Each

269

0147-006X/97/0301-0269$08.00

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

270 NAKIELNY ET AL

NPC is a large structure comprised of upwards of 100 different proteins that

assemble to form a pore across the nuclear envelope. The NPC allows the free

diffusion of ions, metabolites, and small proteins, but large molecules, in con-

trast, traverse the NPC by an active, energy-dependent mechanism (Rout &

Wente 1994, Davis 1995). The best understood active transport process is pro-

tein import to the nucleus, which can be reproduced in vitro (Adam et al 1990).

Protein import is mediated by speciﬁc sequences in the protein, termed nuclear

localization sequences (NLSs) (Boulikas 1993). The NLS is recognized in the

cytoplasm by a speciﬁc receptor, termed importin, karyopherin, or NLS re-

ceptor. Translocation into the nucleus requires several other factors, including

a GTPase cycle (Sweet & Gerace 1995, G¨orlich & Mattaj 1996). While the

molecular mechanism of import of proteins to the nucleus has been character-

ized in some detail, our understanding of RNA transport across the NPC has

lagged behind considerably, because of the difﬁculty of analyzing RNA export

from the nucleus. Despite this limitation, some general principles of RNA ex-

port have been established. All cellular RNA species exit the nucleus through

NPCs by an active, mediated mechanism (Stevens & Swift 1966, Zasloff 1983,

Dworetzky & Feldherr 1988, Featherstone et al 1988, Khanna-Gupta & Ware

1989, Bataill´e et al1990, Guddat et al1990, Mehlin etal1992, Mehlin &Dane-

holt 1993, Jarmolowski et al 1994). Recently, it has become clear that RNA

transport through the NPC is mediated by proteins associated with the RNA

molecules, and that these proteins, like proteins that are actively imported to

the nucleus, utilize signal-dependent, receptor-mediated pathways to transport

RNA molecules. The export pathway for each class is unique: Class-speciﬁc

factors are used (Jarmolowski et al 1994, Pokrywka & Goldfarb 1995). How-

ever, the pathwayslikelymergeat some point, sotheymayalso utilize common

factors.

This review focuses on transport of RNA molecules between the nucleus and

thecytoplasm. Wediscusswhatisunderstood aboutthe transportmechanismof

each class of RNA, and then consider some common themes and perspectives.

Experimental Approaches

Unlike transcription, RNA splicing, and nuclear import, RNA export from the

nucleus to the cytoplasm has not been reproduced in an in vitro system. Most

information on the molecular mechanisms of RNA trafﬁcking between the nu-

cleusandthecytoplasmhas,sofar,beenobtainedfromstudiesinXenopuslaevis

oocytes. The large size of the oocyte (∼1 µl, or up to 200,000 times the size

of a HeLa cell) facilitates the microinjection of a large amount of radiolabeled

RNA into either the nucleus or the cytoplasm. The subcellular distribution of

the injected RNA can be directly visualized by manually dissecting the nucleus

from the cytoplasm and analyzing theRNA by gel electrophoresis (e.g. Zasloff

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 271

et al 1982a). Although the oocyte system offers several attractive features for

the analysis of RNA transport, the oocyte is a highly specialized cell type.

Therefore, results obtained from X. laevis oocyte studies may not always be

relevant to RNA transport in other cell types.

Some information on the molecular mechanisms of RNA trafﬁcking, in par-

ticular mRNA trafﬁcking, has been provided by the analysis of Saccharomyces

cerevisiae mutants. In this approach, yeast cells are usually randomly muta-

genized, and mutants displaying defects in mRNA localization are selected.

Poly(A)

RNA (i.e. mRNA) is detected by in situ hybridization to a ﬂuores-

cently tagged oligo (dT) probe. In wild-type cells, themajority of the poly(A)

RNA is in the cytoplasm. When mRNA export is defective, poly(A)

RNA

accumulates in the nucleus. Genes required for RNA export are identiﬁed

by screening temperature-sensitive mutant strains for those that accumulate

poly(A)

RNA in the nucleus at the restrictive temperature (Kadowaki et al

1992, 1994a; Amberg et al 1992).

In the following sections, we integrate the information obtained from both

of these approaches to the study of RNA transport.

TRANSPORT OF mRNA

Nuclear Export of mRNA

VISUALIZATION OF NUCLEAR EXPORT OF mRNA Balbiani ring transcripts in

Chironomus tentans salivary glands have been particularly useful for study-

ing mRNA export at the ultrastructural level. This RNA is very large, ∼40 kb,

and abundant in gland cells, which allows it to be easily identiﬁed by electron

microscopy. Moreover, Balbiani ring pre-mRNA undergoes minimal splicing,

allowing the transcripts to be followed from their site of synthesis, through

the nucleoplasm to the NPC (reviewed in Mehlin & Daneholt 1993). The

protein-coated transcript released from chromatin appears as a ribbon that is

bent into an asymmetric ring-like structure 50 nm in diameter. Of the four

domains of this structure that can be distinguished by electron microscopy,

one can be identiﬁed that contains the 5

end of the mRNA, and another that

contains the 3

end (Skoglund et al 1986). During translocation through the

NPC, the mRNA-protein complex undergoes a dramatic reorganization, which

appears to occur in ordered steps (Stevens & Swift 1966; Mehlin et al 1991,

1992). During the ﬁrst step of RNA-protein particle translocation, the complex

associates with ﬁbrous material extending from the rim of the NPC into the

nucleoplasm. It then orients itself in front of the NPC; the domain containing

the 5

end facesthe opening of the pore, and the domain containing the 3

end is

closely aposed to the rim of the pore. Finally, thering structure of the transcript

is gradually relaxed and translocated through the pore, with the 5

end in the

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

272 NAKIELNY ET AL

lead. Thus, several temporally and spatially coordinated events, all of which

appear to involve contacts between the RNA-protein complex and the NPC,

lead to the translocation of Balbiani ring mRNA through the NPC. Although

in the absence of other examples to draw on, it cannot yet be established that

these ordered structural rearrangements leading to translocation are typical for

all transcripts, Balbiani ring RNA currently serves as a useful model for the

analysis of mRNA export at the ultrastructural level.

PROCESSING EVENTS THAT INFLUENCE NUCLEAR EXPORT Most mRNAs are

transcribed as pre-mRNAs, which generally undergothree co- or posttranscrip-

tionalprocessingevents: 5

capping,i.e. additionofa 7-monomethylguanosine

G) cap structure to the 5

end of the transcript; removal of introns by splic-

ing; and polyadenylation at a deﬁned site within the 3

untranslated region (for

reviews, see Banerjee 1980, Moore et al 1993, Keller 1995, Shuman 1995,

Adams et al 1996). The relationship between processing and export has been

addressed in many studies. As discussed below, these studies have shown that

splicing signals and introns are, in general, cis-acting nuclear retention ele-

ments, whereas the 5

cap and 3

end appear to enhance the rate of mRNA

export.

Splicing signals and introns Pre-mRNA molecules bearing splice sites are

largely retained in the nucleus. The block in cytoplasmic accumulation of

pre-mRNA is overcome when splicing complex formation is disturbed by mu-

tating either the splicing signals or the trans-acting factors required early in

the formation of splicing complexes (e.g. U1 snRNA). These observations led

to the spliceosome retention hypothesis, which predicts that certain splicing

factors can function as negative trans-acting factors in mRNA export (Legrain

& Rosbash 1989, Chang & Sharp 1989, Hamm & Mattaj 1990). It is unclear

how retention of pre-mRNA is accomplished. Certain splicing factors may

interact with nuclear structures and thereby hold the RNA in the nucleus, or the

spliceosome may prevent interaction of the RNA with export factors.

Removal of introns by splicing does not, inmost cases, appearto be essential

for export of RNA; for example, overexpression of an intron-containing gene

in yeast results in the appearance of pre-mRNA in the cytoplasm. This sug-

gests that an interaction of the pre-mRNA with retention factors is saturable,

allowing the pre-mRNA to leave the nucleus (Legrain & Rosbash 1989). Fur-

thermore, some mechanism must exist to allow export of alternatively spliced

(i.e. intron-containing) transcripts. It is currently unclear whether alternatively

spliced mRNAs are exported by a pathway similar to that which exports fully

spliced transcripts, or whether a specialized mechanism ensures that alterna-

tively spliced mRNAs can leave the nucleus (see section on Viral Regulation

of mRNA Export, Rev and Rex Proteins).

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 273

Although splicing is believed to inﬂuence export indirectly by a retention

mechanism, the export of some mRNAs appears to be directly dependent on

splicing. For example, β-globin mRNA is not exported to the cytoplasm when

expressed from intronless transfected DNA (Collis et al 1990). The process

of splicing appears to convert the mRNA from an export-incompetent to an

export-competent form. It is unknown why splicingis apparently necessary for

efﬁcient export of these transcripts but not for those encoded by other intron-

containing genes, such as cellular thymidine kinase, or for those encoded by

natural intronless genes, such as α-interferon and c-jun (Nagata et al 1980,

Gross et al 1987, Hattori et al 1988).

Recently, two naturally occurring intronless genes, herpes simplex virus

thymidine kinase andhepatitis B virusS transcripts, were each foundto contain

an element that when fused to the intron-dependent β-globin gene, allowed

efﬁcient cytoplasmic accumulation of β-globin RNA in the absence of splicing

(Liu & Mertz 1995, Huang & Yen 1995). These viral RNA sequences may

function as cis-acting RNA export elements.

Cap structure and 3

end Mattaj and coworkers have demonstrated a corre-

lation between the export rate of a transcript and its 5

cap structure. Tran-

scripts synthesized in vitro with a trimethylguanosine m

2,2,7

G(m

G) cap or

an adenosine (A) cap (uncapped transcripts cannot be analyzed because they

are degraded in cells) were found to be exported from X. laevis oocyte nuclei

more slowly than an mRNA synthesized with an m

G cap (Hamm & Mattaj

1990, Jarmolowski et al 1994). The m

G cap structure can therefore enhance

the rate of mRNA export, but it does not appear to be essential. Several obser-

vations indicate that the m

G cap structure is neither necessary nor sufﬁcient

for mRNA export. For example, Caenorhabditis elegans mRNAs processed by

trans-splicing acquire a trimethylated cap and are still translated in, and there-

fore transported to, the cytoplasm (Liou & Blumenthal 1990, Van Doren &

Hirsh 1990), and transfected histone genes lacking their normal 3

end produce

an m

G-capped mRNA that does not leave the nucleus (Sun et al 1992).

The observation that the m

G cap inﬂuences the rate of export of some

mRNAssuggeststhatcap-bindingproteinsmaybeinvolvedinexport. Anuclear

Gcap-bindingcomplex(CBC)comprisingtwocap-bindingproteins, CBP80

and CBP20, has been characterized (Ohno et al 1990, Kataoka et al 1994,

Izaurralde et al 1994; see also the section on U snRNA Export). Although it

is not certain whether CBC mediates the effect of the cap on mRNA export,

Visa et al (1996b) haveshownby immunolocalization that CBP20 accompanies

Balbiani ring mRNA out of the nucleus.

Several studies suggest that the 3

poly(A) tail and the histone 3

end stem-

loop structure can stimulate mRNA export, but that they are not an absolute

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

274 NAKIELNY ET AL

requirement for transport (Eckner et al 1991, Sun et al 1992, Jarmalowski et al

1994). Neither is a 3

poly(A) tail alone sufﬁcient to export an mRNA, since

somepoly(A)

RNAmoleculesneverleavethenucleus(forexample,XistRNA

and omega-n RNA;Brockdorffet al 1992, Brownet al 1992, Hogan et al 1994).

In summary, although the m

G cap and 3

end may inﬂuence the rate of

transport, the role these structures play in mRNA export is unclear. Since

neither structure is essential, other RNA sequence elements yet to be identiﬁed

may be important for mRNA export.

TRANS-ACTING mRNA EXPORT FACTORS This section describes evidence for

the involvement of trans-acting protein factors in mRNA export, which is sum-

marized in Figure 1. These potential trans-acting factors include hnRNP pro-

teins, NPC proteins, components of a GTPase cycle, and various yeast proteins

identiﬁed in poly(A)

RNA export mutants.

hnRNPproteins From the moment pre-mRNAemergesfrom the transcription

complex and throughout its lifetime in the nucleus, pre-mRNA/mRNA proba-

bly never exists as free RNA, but rather is associated with proteins (reviewed

in Dreyfuss et al 1993; see the section on Visualization of Nuclear Export of

mRNA).Thesepre-mRNA/mRNA-bindingproteinscomprisetheabundanthet-

erogeneous nuclear RNA-bindingproteins (hnRNP proteins)and otherproteins

involved in transcription and pre-mRNA processing (Pi˜nol-Roma et al 1988;

reviewed in Dreyfuss et al 1993).

The hnRNP proteins, which have been most extensively characterized in

human cells, comprise a group of about 20 major proteins that associate with

nascent pre-mRNA, and participate in the processing reactions that generate

mature mRNA (Munroe & Dong 1992, Mayeda & Krainer 1992, Dreyfuss

et al 1993, Portman & Dreyfuss 1994, Yang et al 1994, C´aceres et al 1994).

All transcripts appear to be associated with most of the hnRNP proteins, but

the stoichiometry varies, so the protein constellation that assembles on each

transcript is probably unique (Pi˜nol-Roma et al 1989, Matunis et al 1993).

Several hnRNP proteins, for example, hnRNP A1 and hnRNP K, although

predominantly nuclear, shuttle continuously between the nucleus and the cy-

toplasm. While in the cytoplasm, hnRNP A1 is associated with mRNA, sug-

gesting that shuttling hnRNP proteins accompany mRNA from the point of

its emergence from the transcription machinery, through the nucleoplasm to

the NPC, and during translocation of the RNP complex through the NPC

(Pi˜nol-Roma & Dreyfuss 1992, 1993). This suggestion is now supported by

the recent observation that an insect A1-like hnRNP protein in Chironomus

tentans, Ct-hrp36, translocates through the NPC associated with RNA (Visa

et al 1996a). The question of whether mRNA is carrying, or is being carried

by, the shuttling hnRNP proteins has been addressed with the identiﬁcation of a

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 275

Figure 1 Potential trans-acting factors for mRNA nuclear export. Messenger RNA is transcribed

in the nucleus by RNA polymerase II as a precursor with an m

G cap, and, in the majority of

highereukaryotic transcripts, introns and a3

poly(A)tail. Evidenceindicates that shuttling hnRNP

proteinsmediate exportofmRNAfrom thenucleus tothe cytoplasm. Interactionwith spliceosomes

and nonshuttling hnRNP proteins prevents mRNA export. The roles of the cap-binding complex

(CBC), the Ran GTPase cycle, and individual nuclear pore complex (NPC) proteins are less clear.

See text for details.

proteinnuclear exportsignal (NES) withinhnRNPA1 (Siomi &Dreyfuss1995,

Michael et al 1995). The A1 NES (a 38–amino acid segment of the protein

termed M9) is not involved in RNA binding and, when fused to a protein that is

normallyrestrictedtothenucleus, is capable of promoting its export to the cyto-

plasm in a temperature-sensitive manner (Michael et al 1995). Together, these

observations support the model that hnRNP A1, and probably other shuttling

hnRNP proteins, mediate, via their NESs, nuclear export of cellular mRNA.

While the shuttlinghnRNP proteins appearto remain associatedwith mRNA

until it reaches the cytoplasm,nonshuttling hnRNPproteins, such ashnRNP C1

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

276 NAKIELNY ET AL

and hnRNP U, do not leave the nucleus. Recently, a nuclear retention sequence

(NRS) has been identiﬁed in the hnRNP C proteins (Nakielny & Dreyfuss

1996). The NRS is capable of overriding hnRNP protein NESs, suggestingthat

pre-mRNA/mRNA that is associated with both NES-bearing and NRS-bearing

hnRNP proteins cannot leave the nucleus. It is therefore an attractive possibil-

ity that nonshuttling NRS-containing hnRNP proteins prevent transcripts that

have not been fully processed from exiting the nucleus to the cytoplasm. Non-

shuttling hnRNP proteins are removed from mRNA prior to, or during, mRNA

export. This removal is likely to be critical to allow mRNA to be exported, and

may be an important regulatory step in mRNA transport.

After translocation through the NPC as components of the RNP complex,

shuttling hnRNP proteins must dissociate from the mRNA and be reimported

to the nucleus. Import of hnRNP A1 is mediated by the same region of the

protein (M9) that mediates A1 export (Siomi & Dreyfuss 1995). It is not

yet clear whether the import and export signals in A1 are one and the same,

although so far, mutations within M9 that abolish import also prevent export

(Michael et al 1995). Interestingly, nuclear import of A1, and of the majority of

other shuttling hnRNP proteins, is dependent upon ongoing RNA polymerase

II transcription, which suggests that import of these candidate trans-acting

mRNA export factors is coupled to the nuclear content of exportable mRNA

(Pi˜nol-Roma & Dreyfuss 1991, 1993; Michael et al 1996). Import of proteins

bearing classical basic NLSs does not depend on transcription, indicating that

the import mechanism of shuttling hnRNP proteins is fundamentally different

from that of basic NLS-bearing proteins.

Nuclear pore complex proteins At least nine yeast NPC proteins have been

identiﬁed that, when they bear a temperature-sensitive mutation or when they

are deleted, cause poly(A)

RNA to accumulate in the nucleus. These in-

clude Rat2p/Nup120p, Rat3p/Nup133p, Nup100p, Nup49p, Nup116p, Nup1p,

Rat10p/Nup145p, Nup82p, and Rat7p/Nup159p (reviewed in Schneiter et al

1995, Maquat 1996). Fragments of two of these proteins, Nup145p and

Nup116p, are capable of binding RNA homopolymers in vitro, leading to

the suggestion that these NPC proteins interact with RNA molecules being

exported, or with a putative RNA component of the NPC (Fabre et al 1994).

For most of these genes, disruption results in phenotypes additional to that

of nuclear poly(A)

RNA accumulation, for example, defective protein im-

port and abnormalities in NPC morphology (reviewed in Schneiter et al 1995,

Maquat1996). Also, ina numberofcases, only asmall proportionofthemutant

cells show poly(A)

RNA accumulation. Although it may well be that some

of these NPC proteins are not only involved in mRNA export, but also function

in other processes (such as protein import and NPC assembly), a direct role for

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 277

these yeast NPC proteins in mRNA export cannot yet be assigned. One possi-

ble exception is Rat7p/Nup159p. Identiﬁed in a screen for genes required for

mRNA export, a temperature-sensitive mutation causes a rapid accumulation

of poly(A)

RNA in the nucleus at thenonpermissive temperature (Gorsch et al

1995). No protein import defect couldbedetected, and although NPC morphol-

ogyis abnormal, thisphenotypeis apparentat thepermissive temperature, when

mRNA export appears normal. Further characterization of Rat7p/Nup159p

should establish whether it plays a direct role in mRNA transport.

GTPasecycle Importofproteinstothenucleushasbeenfoundtobedependent

upon a GTPase cycle comprising a GTPase (termed Ran in higher eukaryotes,

spi 1 in ﬁssion yeast, and GSP1 and 2/CNR1 and 2 in budding yeast), aguanine

nucleotideexchangefactor (termed RCC1 in higher eukaryotes, pim1in ﬁssion

yeast, and PRP20/MTR1/SRM1 in budding yeast), and a GTPase-activating

protein (termed RanGAP1 in higher eukaryotes and rna1 in ﬁssion and budding

yeasts). This GTPase cycle may also play a role in mRNA export (Dasso 1993,

Melchior et al 1993, Moore & Blobel 1993, Tachibana et al 1994, Schlenstedt

et al 1995, Sweet & Gerace 1995, Tartakoff & Schneiter 1995, Sazer 1996).

Mutation of the genes encoding the GTPase cycle components in yeast or

mammaliancellscausespoly(A)

RNAtoaccumulateinthenucleus. However,

mutation of these genes results in a variety of defects in addition to a block in

mRNA export, for example, abnormalities in nuclear morphology (Aebi et al

1990),mRNAtranscriptioninitiationand3

endformation(Forresteretal1992),

rRNA and tRNA maturation (Kadowaki et al 1993), and pre-mRNA splicing

(Vijayraghavan et al 1989). A direct role for the GTPase cycle components in

mRNA export remains to be demonstrated.

Other potential trans-acting factors In addition to NPC proteins and Ran

GTPase cycle proteins, yeast genetic screens have identiﬁed a number of other

genes, defects in which result in inhibition of mRNA export (Schneiter et al

1995, Maquat 1996). The proteins encoded include Rat1p, an exonuclease

(Amberg et al 1992), Mtr2p, a novel nuclear protein (Kadowaki et al 1994b);

Mas3p, a heat shock transcription factor (Kadowaki et al 1994a); Rpa190p, a

subunit of RNA polymerase I (Schneiter et al 1995); Rae1p, a novel protein

with β-transducin repeats (Brown et al 1995); Mtr3p, a novel nucleolar protein

(Kadowakiet al 1995); andMtr4p, anovelprotein with sequence motifs charac-

teristic of DEAD-box proteins (Liang et al 1996). For all the mutants that have

been analyzed, defects additional to that in mRNA export have been described.

These frequently include abnormal nucleolar morphology and deﬁciences in

rRNA processing. It is currently unclear whether the mRNA export defect is

primary or secondary to the other effects of the mutations, and/or whether the

gene products are multifunctional.

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

278 NAKIELNY ET AL

REGULATIONOFmRNAEXPORT:VIRALREGULATION Studiesofviralpre-mRNA

processing and transport have uncovered the ﬁrst clear examples of regulated

RNA export (reviewed in Izaurralde & Mattaj 1992, 1995; Elliott et al 1994;

Krug 1993). A number of viral proteins, or protein complexes, are currently

known to inﬂuence the nuclear export of viral and cellular RNA molecules.

The best understood of these are human immunodeﬁciency virus 1 (HIV-1)

Rev protein, human T cell leukemia virus 1 (HTLV-1) Rex protein, adenovirus

early region 1B 55-kDa protein/early region 4 34-kDa protein complex (E1B

55-kDa /E4 34-kDa), and M1 matrix protein and nonstructural protein 1 (NS1)

of the inﬂuenza virus.

Rev and Rex proteins Rev and Rex proteins induce the nuclear export of par-

tially spliced or unspliced transcripts encoding structural proteins of HIV-1 and

HTLV-1, respectively (reviewed in Cullen & Malim 1991; Cullen 1992, 1995).

In the absence of Rev or Rex, these transcripts are retained in the nucleus un-

til they are either spliced or degraded (Malim et al 1989b, Felber et al 1989,

Emerman et al 1989). Since Revand Rex are functional equivalents (Hope et al

1991, Weichselbraun et al 1992), the following discussion focuses on Rev, the

more extensively characterized of the two proteins.

ThemechanismofactionofRevinviralmRNAexporthasbeencontroversial.

It has been suggested that Rev promotes the export indirectly by dissociating

splicing factors, directly by activating transport, or by a combination of these

two mechanisms (Cullen & Malim 1991, Cullen 1992). As outlined below,

several lines of evidence have shown that Rev can directly promote export,

although it is unclear whether it also functions in part by dissociating splicing

factors (Chang & Sharp 1989, Lu et al 1990, Kjems et al 1991, Luo et al 1994,

Stutz & Rosbash 1994).

Rev, a small protein of 116 amino acids, binds via an arginine-rich re-

gion to a complex 234-nucleotide RNA stem-loop structure in the partially

spliced/unspliced viral transcripts, termed the Rev response element (RRE).

This Rev-RRE interaction is essential for Rev function (Zapp & Green 1989,

Daly et al1989). Revshuttles between thenucleus and thecytoplasm, an activ-

ity that requires another domain critical for Rev function, termed the activation

domain(Malimet al1989a, 1991; Mermeretal1990; Venkatesh& Chinnadurai

1990; Malim & Cullen 1991; Daly et al 1993; Meyer & Malim 1994; Kalland

et al 1994; Wolff et al 1995). The leucine-rich activation domain (sequence:

LPPLERLTL) is capable of promoting nuclear export of heterologous proteins

in a saturable and temperature-sensitive manner, and it can function in the

absence of the RRE (Wen et al 1995, Fischer et al 1995, Meyer et al 1996).

Evidence that Rev directly promotes RNA export came from microinjec-

tion studies in X. laevis oocytes. Rev protein coinjected with RRE-containing

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 279

pre-mRNA into oocyte nuclei induces export of the RNA independently of

splicing (Fischer et al 1994b). Rev is therefore capable of carrying an RRE-

containing RNA molecule from the nucleus to the cytoplasm.

These ﬁndings together showed that the leucine-rich Rev activation domain

is an NES and that an RNA can be transported to the cytoplasm by associating

with an NES-containing protein.

The effect of Rev on RNA export is evident in the absence of any other

viral protein, suggesting that Rev promotes export by interacting with cellular

factors. Several cellular proteins that interact with Rev and may mediate Rev

functionhaverecentlybeendescribed, includingeukaryoticinitiationfactor5A,

human Rev interacting protein or Rev/Rex activation domain binding protein

(hRIP/Rab) and yeast Rev interacting protein (Rip1p) (Fritz et al 1995, Bogerd

et al 1995, Stutz et al 1995, Bevec et al 1996). Rip1p and hRIP/Rab are

distantly related, and both proteins contain phenylalanine-glycine (FG) repeats

and other sequence repeats, all of which are characteristic of NPC proteins. It

is still unclear if either hRIP/Rab or Rip1p are bona ﬁde components of NPCs.

The Rev interaction domain of hRIP/Rab and Rip1p lies within the FG repeat-

containing region of the proteins (Fritz et al 1995, Stutz et al 1995). Rev can

interactwithasubsetofyeastNPCproteins,allofwhichcontainFGrepeats,and

one of which is Rat7p/Nup159p, that has been identiﬁed as a potential cellular

mRNA export factor (Stutz et al 1995, Gorsch et al 1995; see the section on

Trans-Acting mRNA Export Factors, NPC proteins). A model that arises from

this information is that translocation of Rev from the nucleus to the cytoplasm

is mediated by sequential interactions between Rev and FG repeat proteins in

the nucleoplasm and the NPC.

Saturation of the Rev export pathway, by nuclear injection of bovine serum

albumin (BSA) coupled to peptides comprising the Rev activation domain, has

no effect on the export of coinjected mRNA, and saturation of mRNA transport

does not interfere with Rev-mediated RNA export (Fischer et al 1995). Thus,

Revaccesses an exportpathway that ismechanistically differentfrom the cellu-

lar mRNAexport pathway, eventhough Rev exports RNAmolecules belonging

to the mRNA class. The reason for this redirection of RRE-containing mRNAs

by Rev is still unclear, but the Rev NES, in contrast to NESs that may medi-

ate cellular mRNA export, may direct the RNA to an export pathway that can

override nuclear retention of intron-containing pre-mRNA molecules.

The export of tRNA and rRNA (in the form of ribosomes) is also unaffected

underconditionsthatsaturate Rev-mediatedexport,indicatingthatthese RNAs,

like cellular mRNA, use export pathways that are different from that used by

Rev. Under the same conditions, however, export of U snRNA and 5S rRNA

is inhibited. These observations indicate that at least one limiting factor for

Rev-mediatedtransport is shared by the cellular5S rRNA and U snRNAexport

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

280 NAKIELNY ET AL

pathways (Fischer et al 1995). It has therefore been proposed that export of

5S rRNA and U snRNA is mediated by proteins that harbor an NES that is

functionally equivalent to the NES of Rev (see the sections on U snRNA and

5S rRNA Transport, below).

Adenovirus early region 1B 55-kDa protein/early region 4 34-kDa protein

complex A compex of two adenovirus proteins, E1B 55-kDa and E4 34-kDa,

facilitates the cytoplasmic accumulation of late viral mRNAs and blocks the

cytoplasmicaccumulation of hostcell mRNAs. In mutant viruses lacking func-

tional E1B or E4 protein, reduced levels of late viral mRNAs accumulate in

the cytoplasm, and cytoplasmic accumulation of host cell mRNAs is normal

(Babiss & Ginsberg 1984, Babiss et al 1985, Halbert et al 1985, Pilder et al

1986). Also, in the absence of the E1B protein, the nuclear distribution of

viral transcripts is altered (Pilder et al 1986, Leppard & Shenk 1989). The

molecular mechanisms that underlie the effects of the E1B/E4 protein complex

are not understood, but it has been suggested that the complex recruits a factor

that is essential for cellular mRNA export to viral replication/transcription cen-

ters, thereby enhancing viral mRNA transport and at the same time inhibiting

cellular mRNA export (Ornelles & Shenk 1991).

Inﬂuenza virus NS1 and M1 proteins Inﬂuenza virus encodes two proteins

that appear to regulate export of RNA molecules from the nucleus at differ-

ent stages during the production of viral particles (reviewed in Krug 1993,

Whittaker et al 1996). The NS1 protein affects cytoplasmic accumulation of

viral protein-coding transcripts, whereas M1 protein regulates transport of ma-

ture viral ribonucleoproteins (vRNPs).

Inﬂuenza virus proteins are synthesized in two phases, early and late, and the

switch between these two phases appears to be regulated at the level of mRNA

export from the nucleus (Shapiro et al 1987). The NS1 protein mediates this

regulation by inhibiting the export of late protein transcripts until the appropri-

ate time. The molecular mechanisms involved are unknown, although NS1 has

been shown to bind the poly(A) tails of all RNA molecules tested (Qiu & Krug

1994,Qianetal1994). NS1proteinfunction mustbe inactivatedtoallowexport

of late protein transcripts, and this appears to be mediated by post-translational

modiﬁcations, since protein kinase and methyltransferase inhibitors block ex-

port of late protein transcripts (Kurokawa et al 1990, Martin & Helenius 1991,

Vogel et al 1994). NS1 also indirectly inhibits transport of intron-containing

viral and cellular transcripts by inhibiting pre-mRNA splicing (Lu et al 1994,

Fortes et al 1994).

A prerequisite for the production of virus particles in the cytoplasm is ex-

port of vRNPs, which assemble in the nucleus. Export of vRNPs is dependent

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 281

on nuclear-localized M1 protein (Fraser 1967, Martin & Helenius 1991). The

mechanism by which M1 functions has not yet been characterized, although

studies of the transport properties of vRNPs of a virus bearing a temperature-

sensitive transport mutation in M1 have been informative (Whittaker etal 1995,

Yeetal1995). Atthenonpermissivetemperature, mostofthemutantM1protein

appearstoberestrictedtothenucleus, andthevRNPsareexportedwithoutbind-

ing detectable M1 protein. This suggests that M1 facilitates vRNP export not

by associating with them and accompanying them to the cytoplasm, but rather

by allowing their release from a nuclear retention mechanism. Viral RNP com-

ponents that mediate export of the released vRNPs have not yet been identiﬁed.

REGULATION OF mRNA EXPORT: CELLULAR REGULATION Clear examples of

regulated nuclear export of cellular mRNA have yet to be described at the

mechanistic level. However, several reports suggest that like viruses, cells reg-

ulate nucleocytoplasmic mRNA transport. For example, heat-shocked yeasts

accumulate mRNA in their nuclei, but can selectively export mRNA-encoding

heat shock proteins (Saavedra et al 1996), and overexpression of eukaryotic

initiation factor 4E appears to alter the nucleocytoplasmic distribution of cy-

clin D mRNA (Rousseau et al 1996). In addition, it is of interest that the

export of unspliced RNA of two simple retroviruses, Mason-Pﬁzer monkey

virus (MPMV) and Rous sarcoma virus (RSV), is mediated by cis-acting RNA

elements, which can function in the absence of any viral protein (Bray et al

1994, Ogert et al 1996). Furthermore, these RNA elements, termed constitu-

tive transport elements (CTEs), can substitute for RRE RNA/Rev protein to

induce nuclear export of unspliced HIV transcripts, and the MPMV CTE can

promote export of a cellular intron-containing RNA that is normally retained

in the nucleus (Bray et al 1994, Ogert et al 1996). These observations suggest

that a cellular protein(s) can mediate the effects of these viral CTEs. It will be

of considerable interest to identify this factor, since it may play some role in

regulating export of alternatively spliced cellular transcripts.

Nuclear Import of Viral RNA

Many RNA viruses, including HIV-1 and inﬂuenza virus, replicate their ge-

nomes in the nucleus of the infected cell. In nondividing cells, import of

these nucleic acids has been shown to require ATP and to be protein mediated.

HIV nucleic acid import is mediated by viral matrix protein (MA) and viral

protein R (Vpr). MA contains a classicalbasic NLS and facilitates viral nucleic

acid import by utilizing the cellular classical NLS protein import pathway

(Bukrinskyetal1993,vonSchwedleretal1994). Thesequenceelementswithin

Vpr required for nuclear localization have not been delineated (Heinzinger et al

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

282 NAKIELNY ET AL

1994). Inﬂuenza virus RNA import appears to be mediated by viral NP, which

like HIV MA, utilizes the cellular classical NLS import machinery (O’Neill

et al 1995, Whittaker et al 1996). In an in vitro import assay, viral RNA is

imported only when NP is present, indicating that this RNA is transported as a

speciﬁc RNP (O’Neill et al 1995).

TRANSPORT OF U snRNAs

Overview of U snRNA Transport

The snRNPs U1, U2, U4/6, and U5 are RNA-protein complexes and are part

of the spliceosome in which pre-mRNA processing takes place. Each snRNP

complex consists of one (U1, U2, and U5) or two (U4/6) snRNAs, a common

set of proteins (the Sm proteins B, B

, D1, D2, D3, E, F, and G), and proteins

that are speciﬁc to each U snRNP (L¨uhrmann et al 1990).

All spliceosomal U snRNAs, with the exception of U6, undergo bidirec-

tional transport across the nuclear envelope as part of their maturation pathway

from precursor transcripts to functional nuclear U snRNP complexes (Figure 2)

(DeRobertis1983, Mattaj1988,Izaurralde&Mattaj1992).ThesnRNAsU1-U5

are synthesized in the nucleus by RNA polymerase II and acquire a 5

-terminal

G cap cotranscriptionally (Reddy & Bush 1988). In contrast, the U snRNP

proteins are synthesized and stored in the cytoplasmand do not migrateon their

own into the nucleus (DeRobertis 1983, Mattaj & DeRobertis 1985). Instead,

the newly transcribed snRNAs are exported from the nucleus to the cytoplasm,

where they assemble with the Sm proteins to form the Sm core of U snRNP—a

common structure of these particles. Thereafter, the m

G cap is hypermethy-

lated to form the m

G cap structure, and the U snRNP is then imported into the

nucleus (Figure 2) (Mattaj 1986, 1988). When and where the speciﬁc proteins

are incorporated into the snRNP particles is unknown.

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→

Figure 2 Schematic drawing of the U snRNP biogenesis cycle in oocytes as exempliﬁed by U1

snRNP. The snRNA is transcribed in the nucleus by RNA polymerase II as a precursor with an

G cap and in some cases a 3

-terminal extension. The RNA is then transported to the cytoplasm

by virtue of binding to the nuclear CBC, and possibly by other factors that are so far unknown. In

the cytoplasm the snRNA associates with the Sm proteins B, B

, D1, D2, D3, E, F, and G, which

form, together with the RNA, the Sm core domain. The m

G cap is then hypermethylated to form

the m

G cap, and this is dependent on the prior formation of the Sm core domain. Eventually, the

U snRNP particle is transported to the nucleus, which requires interaction of the m

G cap and the

Sm core domain with a U snRNP-speciﬁc import receptor. The time and place of association of U

snRNP-speciﬁc proteins are in most cases unclear and are therefore not included in the ﬁgure.

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 283

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

284 NAKIELNY ET AL

Nuclear Export of U snRNA

ROLE OF CIS-ACTING STRUCTURES IN NUCLEAR EXPORT OF U snRNAs All spli-

ceosomal snRNAs appear to exit the nucleus via a common export pathway,

because nuclear injection of large amounts of any one particular snRNA inter-

feres with the export of all the others. This suggests that common structures in

all spliceosomal snRNAs contribute to their export (Jarmolowski et al 1994).

In search for such structures that may mediate snRNA nuclear export, Mattaj

and coworkers identiﬁed the m

G cap structure as one important signal element

(Hamm & Mattaj 1990). This observation led to the puriﬁcation of a nuclear

Gcap-bindingcomplex(CBC) (Ohno et al 1990; Izaurraldeet al 1992, 1994,

1995; Kataoka et al 1994, 1995). CBC consists of two proteins, CBP80 and

CBP20, whose properties suggest a role in export. Antibodies against CBP20

interfere with CBC binding to the cap, and speciﬁcally inhibit snRNA export,

thereby demonstrating a direct involvement of CBC in the nuclear export of

spliceosomal snRNA (Izaurralde et al 1995). The mechanism by which CBC

inducesUsnRNAnuclear exportisstillunclear. OnepossibilityisthattheCBC

contains an NES that allows access to a speciﬁc export pathway. Alternatively,

CBC may interact with cellular components that contain NESs.

The m

G cap alone, however, is insufﬁcient for mediating the export of

snRNAs. Other sequences that inﬂuencethe export of U1 snRNAhave been lo-

calizedto the 5

-terminal124 nucleotides as well as tothe 3

-terminalstem/loop

of the RNA (Terns et al 1993).

TRANS-ACTING U snRNA EXPORT FACTORS GTPase cycle Evidence suggests an

involvement of the Ran GTPase cycle in nuclear export of spliceosomal U

snRNAs(seealsothemRNAtrans-actingexportfactorssection). IntheChinese

hamster cell line tsBN2, whose mutant RCC1 is unstable at 40

◦

C, nuclear

export of spliceosomal snRNAs is inhibited (Cheng et al 1995). Under the

same conditions, U3 snoRNA targeting to the nucleoli is also inhibited. Since,

unlike spliceosomal U snRNAs, U3 snoRNA is not exported to the cytoplasm,

the Ran GTPase cycle may be generally required to enable the intranuclear

movement of certain classes of RNPs.

Nuclear Import of U snRNA

UsnRNAsexportedfrom the nucleus assemblein the cytoplasminto U snRNPs

before returning to the nucleus. Because neither the snRNA nor the snRNP

proteins alone are imported independently of each other, the functionally active

NLS is generated during assembly of the U snRNP particle.

NUCLEAR IMPORT OF U snRNA IN X. LAEVIS OOCYTES Microinjection studies in

X. laevis oocytes showed that the assembly of the common Sm proteins with

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 285

the snRNA is an essential step in the formation of the snRNP NLS. Mutant U

snRNAs lacking the Sm protein binding site (Sm site) fail to accumulate in the

oocyte nucleus, whereas binding sites for the speciﬁc proteins are dispensible

for import (Mattaj & DeRobertis 1985, Mattaj 1986).

Because cap hypermethylation is also dependent on the assembly of the

common proteins on the Sm site of the U snRNA, the possibility of whether

the m

G cap may also contribute to the formation of a functional NLS was

tested. U1 snRNA carrying either no cap or the artiﬁcial adenosine (A) cap

assembles in the cytoplasm with the common proteins but fails to localize to

the nucleus (Fischer & L¨uhrmann 1990, Hamm et al 1990). In addition, the

GpppG cap dinucleotide speciﬁcally inhibits nuclear import of U snRNAs

(Fischer & L¨uhrmann 1990, Fischer et al 1991, Michaud & Goldfarb 1991).

Thus, the m

G cap has an essential signaling role in nuclear import of these

RNAs. However, the m

G cap alone is not sufﬁcient for import of U snRNPs.

Another signal on the Sm core has been deﬁned that, together with the m

cap, is necessary and sufﬁcient for nuclear import (Fischer et al 1993). Not

all U snRNAs, however, have the same m

G cap requirement for their nuclear

import. For example, U4 and U5 snRNAs, or an artiﬁcial RNA that contains

only the Sm site ﬂanked by two stem loops, do not absolutely require the m

cap for import, since they can enter the nucleus with an A cap. However, in

these cases import is considerably slower than that of m

G-capped snRNAs

(Fischer et al 1991, 1993; Jarmolowski & Mattaj 1993).

The mechanism by which the m

G cap and the signal on the Sm core domain

cooperateto create the fully activecomposite NLS isnot yet clear. In particular,

it is unclear whether one import receptor recognizes both signals or whether

two different receptors are required that recognize the cap and the Sm core

separately. Another possibility is that the m

G cap induces a conformational

change in the U snRNP particle and thereby exposes a preexisting signal on the

Sm core domain.

Competitionexperimentshavebeencarriedout inordertodeterminewhether

thenuclearimportpathwaysofclassicalNLS-containingproteinsandUsnRNPs

sharecommon features. These experimentsindicated that protein and U snRNP

import are mediated by different rate-limiting factors (Michaud & Goldfarb

1991, 1992; Fischer et al 1991, 1993, 1994a; van Zee at al 1993). However, it

is unknown whether proteins and U snRNAs use entirely different import path-

ways or whether their pathways converge at some point and therefore share

some common factors.

NUCLEAR IMPORT OF U snRNAs IN SOMATIC CELLS In somatic cells, import of

microinjected U1 snRNPs is not strictly dependent on the m

G cap. However,

transport is acceleratedby the presence of an m

G cap. The Sm core domain, in

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

286 NAKIELNY ET AL

contrast,isnecessaryandsufﬁcientfornuclearimport. Thesamesignalrequire-

mentisobservedindifferentcelltypes, includingX.laevissomaticcells(Fischer

et al 1994a). The development of an in vitro transport system that faithfully

mimics in vivo snRNA import has provided further insights into mechanistic

aspects of U snRNA import. As in vivo, the Sm core domain is required for nu-

clear import of U1 and U2 snRNPs in vitro. However, the m

G cap dependence

varies, depending on the source of cytosol. When the assay is provided with X.

laevisoocyte cytosol, U1 andU2 snRNPnuclear importis strictly dependent on

the m

G cap. In contrast, somatic cell cytosol supports the import of ApppG-

capped U snRNPs. Therefore, cell type-speciﬁc rather than species-speciﬁc

differences in the transport machinery apparently account for the differential

G cap requirement for nuclear import of U1 snRNPs in X. laevis oocytes and

somatic cells (Fischer et al 1994a, Marshallsay & L¨uhrmann 1994).

TRANSPORT OF 5S rRNA

5S rRNA is a component of large ribosomal subunits. Synthesis of 5S rRNA

differsfromthat oftheother rRNAs(18S,28S,and5.8S)bytwogeneralcriteria.

First, it is transcribed by RNA polymerase III, as opposed to RNA polymerase

I.Second, the 5S rRNAgenes are not included in the rDNArepeats that produce

thenucleolar organizerregions; consequently 5S rRNAissynthesized at sites in

the nucleoplasm distinct from nucleoli (Warner 1990, Scheer & Weisenberger

1994).

Much of the analysis of 5S rRNA transport has been accomplished by using

X.laevisoocytesas a model system, butasexplainedbelow, the oocytepathway

is highly specialized and more complicated than the somatic cell pathway.

Therefore, we discuss the oocyte experiments ﬁrst, followed by what is known

about the pathway in somatic cells, and attempt to highlight general features

that relate to both cell types.

Overview of 5S rRNA Transport in X. laevis Oocytes

X. laevis contains two types of 5S rRNA genes, the oocyte and the somatic,

which are differentially expressed during development and differ by only six

nucleotide substitutions (Wolffe & Brown 1988). During oogenesis, oocyte-

type 5S rRNA is expressed before the other ribosomal components (Mairy &

Denis 1971) and migrates to the cytoplasm, where it is found in one of two

storage particles, the 7S and 42S RNPs. The 42S particle is a mixture of 5S

rRNA, tRNAs, and several proteins, while the 7S particle is composed of a 1:1

ratio of 5S rRNA and the transcription factor TFIIIA (reviewed in Tafuri &

Wolffe 1993). This composition remains the case until the onset of vitelloge-

nesis, when synthesis of the other ribosomal components commences. At this

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 287

stage the concentration of the storage particlesdecreases (Dixon & Ford 1982),

and 5S rRNA rapidly localizes to the nucleoli of the oocyte nucleus, where it

becomes incorporated into large ribosomal subunits(Allison et al 1991). These

subunits are then exported to their site of function in the cytoplasm. Thus, the

oocyte 5S rRNA pathway is unique in that two distinct nuclear export events

occur and can be summarized as follows: transcription in the nucleus, export

to the cytoplasm as a 7S or 42S storage particle, nuclear import, nucleolar

localization, and re-export in the context of the large ribosomal subunit. Per-

haps because of this complex, bidirectional transport pathway, 5S rRNA is the

only RNA examined for which nuclear export cannot be completely saturated

(Jarmolowski et al 1994).

NUCLEAR EXPORT OF 5S rRNA IN XENPOUS OOCYTES Trans-acting factors Three

proteinsare knowntospeciﬁcallybindto 5SrRNA:theLa protein, transcription

factor IIIA (TFIIIA), and ribosomal protein L5. La protein, which transiently

interacts with all pol III products, functions in pol III transcription termination

(Rinke & Steitz 1982, Gottlieb & Steitz 1989). TFIIIA binds both to the 5S

rRNA gene, where it facilitates transcription, and to 5S rRNA itself (Honda &

Roeder 1980). The other known 5S rRNA-binding protein is ribosomal protein

L5 (Chan et al 1987).

Pielerandcolleaguesﬁrstexaminedtherelationshipbetween5SrRNAexport

and protein factors by immunoprecipitationof deﬁned5S RNP complexesfrom

the nucleus and cytoplasm of fractionated oocytes (Guddat et al 1990). These

experiments revealed that the La-5S RNP is conﬁned to the nucleus at all times,

thus excluding a role for La in nuclear export of 5S rRNA, and that RNP

complexes containing either TFIIIA or L5 form initially in the nucleus and then

migrate to the cytoplasm. Additionally, using 5S rRNA mutants, they found

that as long as a given 5S mutant can interact with either TFIIIA or L5, then

export to the cytoplasm occurs. However, mutants that no longer bind either

protein are retained in the nucleus. These studies demonstrate that both TFIIIA

and L5 have the capacity to export 5S rRNA to the cytoplasm.

NUCLEARIMPORTOF5S rRNAINX.LAEVISOOCYTES Nuclearimportof5S rRNA

is similar to import of karyophilic proteins in that it is temperature- and ATP-

dependent and occurs through NPCs (Allison et al 1993). Analysis of 5S rRNA

nuclear import inoocytes reveals that L5 plays a signiﬁcantrole in this process.

Allison and colleagues haveshownthat during vitellogenesis, when production

of ribosomal proteins begins, 5S rRNA is exchanged from the cytoplasmic 7S

storage particle onto L5 and then enters the nucleus (Allison et al 1991, 1993).

Additionally, it has been shown that 5S rRNA mutants are only imported into

the nucleus if they retain L5-binding activity (Rudt & Pieler 1996).

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

288 NAKIELNY ET AL

A MODEL FOR 5S rRNA TRANSPORT IN OOCYTES The 5S rRNA transport path-

wayinX.laevisoocytesisquitecomplicated,principallybecauseoftheoocyte’s

strategy for stockpiling ribosomes within the egg for later use during embryo-

genesis. The picture that emerges en toto is that, in previtellogenic oocytes,

oocyte-type 5S rRNA is transcribed, transiently associates with La protein, and

is then bound by TFIIIA to form the 7S RNP. This RNP is then exported to

the cytoplasm, where it exists as a storage particle until synthesis of other ri-

bosomal components begins. Consistent with a role for TFIIIA in 5S rRNA

export is the fact that the protein contains an NES similar to the Rev/PKI-type

of NES and that the Rev NES conjugated to BSA efﬁciently competes for ex-

port of 5S rRNA (Fischer et al 1995, Fridell et al 1996). The storage particles

remain in the cytoplasm until L5 becomes available and displaces TFIIIA to

form the 5S RNP. This RNP is then rapidly imported into the nucleus, be-

comes concentrated within the nucleoli, and is assembled into large ribosomal

subunits that are then exported to the cytoplasm. This model, summarized in

Figure 3, can account for the fact that both TFIIIA and L5 mediate export of 5S

rRNA.

5S rRNA Transport in Somatic Cells

The 5S rRNA transport pathway is simpler in somatic cells than in X. laevis

oocytes. Using HeLa cells, Steitz and coworkers showed that after a transient,

posttranscriptional association with La protein, 5S rRNA immediately binds

L5 to form a particle that is a precursor to assembly of 5S rRNA into nascent

ribosomal subunits within the nucleoli (Steitz et al 1988). Therefore nascent

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→

Figure 3 Intracellular trafﬁcking of 5S rRNA. 5S rRNA is transcribed in the nucleoplasm by

RNA polymerase III and then brieﬂy associates with La antigen, which functions in transcription

termination. At this point the pathway as it is known in X. laevis oocytes appears to differ from

the pathway that is operational in somatic cells. The area shaded in gray depicts the X. laevis

oocyte-speciﬁc portion of the pathway: 5S rRNA is bound by TFIIIA, which also functions in

its transcription, to form a 7S RNP that is then rapidly exported to the cytoplasm, where it is

sequestered as the 7S storage particle. As the level of L5 accumulates in the oocyte, an exchange

reaction occurs in which 5S rRNA is transferred to L5 to form the 5S RNP. Free TFIIIA, which is

now competent for nuclear import, then presumably returns to the nucleus. The 5S RNP is rapidly

imported into the nucleus and is subsequently concentrated within nucleoli, where it is assembled

into nascent large ribosomal subunits. After complete assembly, the large ribosomal subunits

are exported to the cytoplasm, where they function in translation. In somatic cells, which do not

stockpile ribosomes for use later indevelopment,the pathway is less complex. After disassociation

from La, 5S rRNA immediately interacts with the L5to form the 5S rRNPin the nucleoplasm. The

RNP then localizes directly to the nucleolus, where it is incorporated into nascent large ribosomal

subunits and then exported to the cytoplasm.

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 289

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

290 NAKIELNY ET AL

5S rRNA does not shuttle out of and back into the nucleus in somatic cells but

instead accumulates directly in the nucleoli as a complex with L5. Consistent

with these ﬁndings is the fact that, despite considerable effort, the 5S rRNA-

containing storage particles have not been found in the cytoplasm of somatic

cells (e.g. see Honda & Roeder 1980).

What is common between the two pathways is the fact that transcription of

the 5S genes occurs in the nucleoplasm and that the L5-5S RNP forms prior to

ribosomalsubunitassemblyin thenucleolus, a pointwhich promptedSteitz etal

(1988) to suggest that L5 functions in part to deliver 5S rRNA to the nucleolus.

InX.laevisthisdeliverypathwaycanbegininthecytoplasm,whereasinsomatic

cells the pathway is conﬁned to within the nucleus (summarized in Figure 3).

Recent work on the domain structureof L5 supports this model, as it was found

that the 5S rRNA-binding and nucleolar localization domains are separable

within the protein. Furthermore, L5 mutants that maintain 5S rRNA-binding

activitycannotproperlylocalizetothenucleoluswhenthenucleolarlocalization

domainis interrupted, suggestingthatanintactnucleolar localization domain of

L5is required for efﬁcientaccumulationof the L5-5S RNP within thenucleolus

(Michael & Dreyfuss 1996). Once targeted to the nucleolus, the L5-5S RNP is

probably assembled into nascent ribosomal subunits that are then exported to

the cytoplasm. Therefore, the problem of 5S rRNA export in somatic cells can

be reduced to the problem of ribosomal subunit export.

RIBOSOMAL RNA TRANSPORT

All rRNAs except 5S rRNA—namely 18S, 28S, and 5.8S rRNAs—are tran-

scribed as a commonprecursor thatis completely processed and assembledinto

ribosomal subunits, all within the nucleolus (Warner 1990, Scheer & Weisen-

berger 1994, Shaw & Jordan 1995). Therefore, nuclear export of these rRNAs

occurs in the context of ribosomal subunits. Beyond the fact that ribosomal

subunit export is a unidirectional, saturable process that requires energy and

occurs through NPCs (Khanna-Gupta & Ware 1989, Bataill´e et al 1990), very

little is known about export of ribosomes.

TRANSPORT OF tRNA

Nuclear Export

PROCESSING EVENTS THAT INFLUENCE NUCLEAR EXPORT Transfer RNA is

transcribed in the nucleus by RNA polymerase III, and like RNA polymerase

II–transcribedmRNA,it is produced as a precursortRNA.Pre-tRNAmolecules

are processed in the nucleus before transport to their functional location in the

cytoplasm, although they are further modiﬁed in the cytoplasm. Nuclear tRNA

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 291

processing reactions comprise excision of 5

and 3

terminal sequences, addi-

tion of CCA to the 3

end, base modiﬁcations, and for some tRNA molecules,

removal of introns and base editing (reviewed in Deutscher 1995; Altman et al

1993, 1995; Westaway & Abelson 1995; Bj¨ork 1995).

OnlytRNAmoleculeswithmatureterminiandwithout intronsaredetectedin

the cytoplasm of X. laevis oocytes, suggesting that 5

and 3

terminal extensions

and intron sequences prevent pre-tRNA export (Melton et al 1980, Zasloff et al

1982a; reviewed in Westaway & Abelson 1995). It has been suggested that the

process of splicing is coupled to transport through the NPC, because in yeast,

components of the splicing process (an endonuclease and a ligase) appear to be

localized in close apposition to NPCs, and mutations in certain NPC proteins

result in accumulation of unspliced tRNA (Peebles et al 1983, Clark & Abelson

1987,Sharmaetal1996;reviewedinWestaway&Abelson1995). However,the

blockinpre-tRNAsplicinginNPCproteinmutantsmaybesecondarytoadefect

in the nuclear import of a tRNA splicing factor. In addition, microinjection of

large amounts of pre-tRNA into the nucleus of X. laevis oocytes results in the

appearance of unspliced pre-tRNA in the cytoplasm, suggesting that tRNA

splicing is not coupled to the export process (Boelens et al 1995).

CIS-ACTING AND TRANS-ACTING tRNA EXPORT FACTORS Zasloff and coworkers

deﬁned the critical regions of human tRNAmet for nuclear export in X. laevis

oocytes. Although point mutations in many regions of the molecule perturb

transport, the most drastic effects are caused by mutations in the most highly

conserved regions of tRNA, the D and T stem loops (Santos & Zasloff 1981,

Zasloffet al 1982a,b, Tobian et al 1985). The most critical regions for transport

alsoappear tobethe mostimportantregionsfor tRNAtranscription, processing,

and ribosome binding (see discussion of Tobian et al 1985). All transport

mutants described so far are also defective in processing, suggesting that a

processing factor also serves as an export factor or, more likely, that the same

structural features ofthe tRNA molecule are recognized by both processing and

transport machineries (Zasloff et al 1982a, Tobian et al 1985).

Different tRNA species competitively inhibit the export of each other when

microinjected into X. laevis oocyte nuclei, demonstrating that different tRNAs

share a common rate-limiting factor(s), which likely recognizes the highly

conserved stem loops D and T (Zasloff 1983). The glycolytic enzyme glycer-

aldehyde 3 phosphate dehydrogenase has been identiﬁed as a factor that binds

to wild-type tRNAs but not to transport-defective mutants, although a function

for this enzyme in tRNA export has not been described (Singh & Green 1993).

It has recently been suggestedthat sometransport-defective mutanttRNAsmay

be defective, not because they fail to interact with proteins of the transport ma-

chinery, but rather because they are retained in the nucleus by interaction with

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

292 NAKIELNY ET AL

nuclear components (Boelens et al 1995). This is unlikely to be the case for

all transport-defective tRNA mutants, so some of them should be of use in the

identiﬁcation of tRNA export factors.

THEMES AND PERSPECTIVES

A common theme that emerges from the mechanistic information on nucleocy-

toplasmic trafﬁcking of RNA is that transport of possibly all RNA classes, in

both directions across the NPC, is mediated largely by signals on proteins that

are associated with the RNA (Figure 4). Thus, mRNA export is most likely

mediated, at least in part, by NESs of the shuttling hnRNP proteins such as the

M9 NES of hnRNP A1; U snRNA export is mediated, at least in part, by CBC

(whose NES remains to be identiﬁed); and 5S rRNA export is mediated by L5

(whose NES remains to be identiﬁed) and TFIIIA (which bears a leucine-rich

NES similar to that of HIV-1 Rev protein) (Figure 4). The factors that mediate

rRNA and tRNA export are still obscure.

Nuclear import of U snRNAs, 5S rRNA (in oocytes), and viral RNAs is also

largelyproteinmediated. ThenuclearimportsignalofUsnRNAsisacomposite

signalconsistingofproteinsthatassociatewiththeRNAandthe5

terminalm

cap. This signal accesses a different import pathway than that used by classical

NLS-containing proteins. In contrast, import of 5S rRNA in oocytes appears

Figure 4 RNA export from the nucleus into the cytoplasm is mediated by RNA-binding proteins.

Proteins that speciﬁcally interact with each class of RNA carry the signals for RNA transport to

the cytoplasm. Messenger RNA export is most likely mediated by shuttling hnRNP proteins; U

snRNA export is mediated by the cap-binding complex (CBC); and 5S rRNA export is mediated

by TFIIIA and L5. Factors involved in tRNA and rRNA transport remain to be identiﬁed.

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 293

to be mediated exclusively by the L5 protein via the classical NLS pathway.

Viral RNA import is also, in some cases, mediated by proteins that utilize the

cellular classical NLS import pathway (HIV-1 MA and inﬂuenza virus NP) or

by proteins whose import pathway has not yet been characterized (HIV-1 Vpr).

In most cases, RNA-binding proteins dictate the transport pathway taken by

a particular RNA molecule. However, the sequence of the RNA molecule, in

turn, determines the complement of proteins with which it will interact, so the

pathway taken is ultimately speciﬁed by RNA sequence.

This review has focused on transport of RNAmolecules between the nucleus

and the cytoplasm. However, once exported to the cytoplasm, some mRNAs

(and probably other classes of RNA that function in the cytoplasm) are trans-

ported and anchored to very precise subcytoplasmic locations. This targeting

is essential in processes as diverse as development of the embryo and neuronal

cell function, and the molecular mechanisms appear to follow a principle sim-

ilar to that of nucleocytoplasmic transport of RNA molecules. Transport and

anchoring of mRNAs within the cytoplasm is also mediated by proteins that

interact with the RNA (reviewed in Ding & Lipshitz 1993, Wilhelm & Vale

1993, Singer 1993, Okita et al 1994, Stebbings et al 1995, St Johnston 1995a,b,

Micklem 1995, Steward 1995). This basic principle of RNA transport is likely

to be of considerable generality.

Our understanding of cellular RNA transport pathways is at an early and

exciting stage. For most RNA classes, known mediators/effectors of transport

are few, but for mRNA export, at least, potential mediators are plentiful, and

a number of these may become bona ﬁde export factors with more extensive

investigation. Immediate questions, such as the number and diversity of pro-

tein NESs and protein NLSs that mediate RNA export and import, seem quite

tractable. Transportofeach typeofRNAmayverylikelyinvolveboth pathway-

speciﬁc components and components that are shared with other pathways. The

identiﬁcation of points of pathway convergence should also be possible. In

addition, regulatory mechanisms of mRNA export, as exempliﬁed by viruses,

have yet to be characterized for cellular transcripts. However, with progress

in our understanding of viral regulation of mRNA export, cellular regulatory

mechanisms (for example, those required for export of alternatively spliced

transcripts) are expected to be uncovered.

Because the detailed molecular mechanisms by which RNA molecules move

from their site of synthesis to the NPCs, and by which they are translocated

through the NPCs, remain mysterious, we are currently left to speculation. We

doknowthat there areenergy-consumingsteps, which may includeintranuclear

transport of RNA and, most likely, translocation through the NPC. Maybe

translocation is motor protein driven, which would certainly account for part of

the energy requirement. If the Ran GTPase cycle is directly involved in RNA

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

294 NAKIELNY ET AL

export, this will also account for some of the energy consumption. It will be

fascinating to learn how directionality is imposed on RNA-protein complexes

as they move through the NPC, and how shuttling components of the transport

machineries that move with them through the NPC are recycled.

A major tenet of RNA transport, that it is mediated by proteins associated

with the RNA, has been ﬁrmly established. With this understanding, a more

detailed knowledge of these fundamental transport pathways is expected in the

near future.

CKNOWLEDGMENTS

The research in this laboratory is supported by a grant from the National

Institutes of Health and by the Howard Hughes Medical Institute. UF received

a grant from the Deutsche Forschungsgemeinschaft and an AIDS fellowship

from the Deutsches Krebsforschungszentrum.

Visit the Annual Reviews home page at

http://www.annurev.org.

Literature Cited

Adam SA, Marr RS, Gerace L. 1990. Nuclear

protein import in permeabilized mammalian

cells requires soluble cytoplasmic factors. J.

Cell Biol. 111:807–16

Adams MD, Rudner DZ, Rio DC. 1996. Bio-

chemistryandregulationof pre-mRNAsplic-

ing. Curr. Opin. Cell Biol. 8:331–39

Aebi M, Clark MW, Vijayraghavan U, Abelson

J. 1990. A yeast mutant, PRP20, altered in

mRNA metabolism and maintenance of the

nuclear structure, is defective in a gene ho-

mologous to the human gene RCC1 which is

involved in the control of chromosome con-

densation. Mol. Gen. Genet. 224:72–80

Allison LA, North MT, Murdoch KJ, Romaiuk

PJ, Deschamps S, Le Marie M. 1993. Struc-

tural requirements of 5S rRNA for nuclear

transport, 7S ribonucleoprotein particle as-

sembly, and 60S ribosomal subunit assem-

bly in Xenopus oocytes. Mol. Cell. Biol.

13:6819–31

Allison LA, Romaniuk PJ, Bakken, AH. 1991.

RNA-protein interactions of stored 5S RNA

with TFIIIA andribosomal protein L5 during

Xenopus oogenesis. Dev. Biol. 144:129–44

Altman S, Kirsebom L, Talbot S. 1993. Recent

studies of ribonuclease P. FASEB J. 7:7–14

Altman S, Kirsebom L, Talbot S. 1995. Recent

studies of RNase P. In tRNA Structure,

Biosynthesis, and Function, ed.DS¨oll,

UL RajBhandary, 6:67–78. Washington, DC:

ASM

Amberg DC, Goldstein AL, Cole CN. 1992.

Isolation and characterization of RAT1:an

essential gene of Saccharomyces cerevisiae

required for the efﬁcient nucleocytoplasmic

trafﬁcking of mRNA. Genes Dev. 6:1173–89

BabissLE,GinsbergHS.1984.Adenovirustype

5 early region 1b gene productis required for

efﬁcient shutoff of host protein synthesis. J.

Virol. 50:202–12

Babiss LE, Ginsberg HS, Darnell JE Jr. 1985.

Adenovirus E1B proteins are required forac-

cumulationoflateviralmRNAandforeffects

on cellular mRNA translation and transport.

Mol. Cell. Biol. 5:2552–58

Banerjee AK. 1980. 5

-terminal cap structure

in eukaryotic messenger ribonucleic acids.

Microbiol. Rev. 44:175–205

Bataill´eN, HelserT, FriedHM. 1990.Cytoplas-

mic transport of ribosomal subunits microin-

jectedintotheXenopuslaevisoocytenucleus:

ageneralized, facilitatedprocess.J.Cell Biol.

111:1571–82

Bevec D, Jaksche H, Oft M, W¨ohl T, Him-

melspach M, et al. 1996. Inhibition of HIV-1

replication in lymphocytes by mutants of the

Rev cofactor e1F-5A. Science 271:1858–60

Bj¨ork GR. 1995. Biosynthesis and function

of modiﬁed nucleosides. In tRNA Structure,

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 295

Biosynthesis, and Function, ed.DS¨oll, UL

RajBhandary, 11:165–205.Washington, DC:

ASM

Boelens WC, Palacios I, Mattaj IW. 1995. Nu-

clear retention of RNA as a mechanism for

localization. RNA 1:273–83

Bogerd HP, Fridell RA, Madore S, Cullen BR.

1995. Identiﬁcation ofa novel cellular cofac-

tor for the Rev/Rex class of retroviral regula-

tory proteins. Cell 82:485–94

Boulikas T. 1993. Nuclear localization signals

(NLS).Crit.Rev.Eukaryot. GeneExp. 3:193–

227

Bray M, Prasad S, Dubay JW, Hunter E, Jeang

K-T, et al. 1994. A small element from the

Mason-Pﬁzer monkey virus genome makes

human immunodeﬁciency virus type 1 ex-

pression and replication Rev-independent.

Proc. Natl. Acad. Sci. USA 91:1256–60

Brockdorff N, Ashworth A, Kay GF, McCabe

VM,NorrisDP,etal.1992.Theproductofthe

mouseXist geneis a 15kb inactiveX-speciﬁc

transcript containing no conserved ORF and

located in the nucleus. Cell 71:515–26

Brown CJ, Hendrich BD, Rupert JL, Lafreni`ere

RG, Xing Y, et al. 1992. The human XIST

gene: analysis of a 17 kb inactive X-speciﬁc

RNA that contains conserved repeats and

is highly localized within the nucleus. Cell

71:527–42

Brown JA, Bharathi A, Ghosh A, Whalen W,

Fitzgerald E, Dhar R. 1995. A mutation in

the Schizosaccharomyces pombe rae1 gene

causes defects in poly(A)

RNA export and

inthe cytoskeleton.J.Biol. Chem.270:7411–

Bukrinsky MI, Haggerty S, Dempsey MP,

Sharova N, Adzhubel A, et al. 1993. A nu-

clear localization signal within HIV1 matrix

proteinthatgovernsinfectionofnon-dividing

cells. Nature 365:666–69

C´aceres JF, Stamm S, Helfman DM, Krainer

AR. 1994. Regulation of alternative splic-

ing in vivo by overexpression of antagonistic

splicing factors. Science 265:1706–9

ChanYL,LinA,McNallyJ,WoolIG.1987.The

primarystructure ofratribosomal proteinL5:

a comparison of the sequence of amino acids

in the proteins that interact with 5S rRNA.

J. Biol. Chem. 262:12879–86

Chang DD, Sharp PA.1989. Regulation by HIV

Revdepends upon recognitionof splice sites.

Cell 59:789–95

Cheng Y, Dahlberg JE, Lund E. 1995. Di-

verse effects of the guanine nucleotide ex-

change factor RCC1 on RNA transport. Sci-

ence 267:1807–10

Clark MW, Abelson J. 1987. The subnuclear

localization of tRNA ligase in yeast. J. Cell

Biol. 105:1515–26

Collis P, Antoniou M, Grosveld F. 1990. Deﬁni-

tion of the minimal requirements within the

human β-globin gene and the dominant con-

trol region for high level expression. EMBO

J. 9:233–40

Cullen BR. 1992. Mechanism of action of reg-

ulatory proteins encoded by complex retro-

viruses. Microbiol. Rev. 56:375–94

Cullen BR. 1995. Regulation of HIV gene ex-

pression. AIDS 9(Suppl. A):S19–S32

Cullen BR, Malim MH. 1991. The HIV1 Rev

protein: prototype ofa novel class of eukary-

otic post-transcriptional regulators. Trends

Biochem. Sci. 16:346–50

Daly TJ, Cook KS, Gray GS, Maione TE,

Rusche JR. 1989. Speciﬁc binding of

HIV1 recombinant Rev protein to the Rev-

responsive element in vitro.Nature342:816–

Daly TJ, Rennert P, Lynch P, Barry JK, Dundas

M, et al. 1993. Perturbation of the carboxy

terminus of HIV1 Rev affects multimeriza-

tionontheRevresponsiveelement.Biochem-

istry 32:8945–54

DassoM.1993.RCC1inthecellcycle: Thereg-

ulator of chromosome condensation takes on

new roles. Trends Biochem. Sci. 18:96–101

Davis LI. 1995. The nuclear pore complex.

Annu. Rev. Biochem. 64:865–96

DeRobertis EM. 1983. Nucleoplasmic segrega-

tion of proteins and RNAs. Cell 32:1021–25

Deutscher MP. 1995. tRNA processing nucle-

ases. In tRNA Structure, Biosynthesis, and

Function, ed.DS¨oll, ULRajBhandary, 5:51–

65. Washington, DC: ASM Press

Ding D, Lipshitz HD. 1993. Localized RNAs

and their functions. BioEssays 15:651–58

Dixon LK, Ford PJ. 1982. Regulationof protein

synthesisand accumulation during oogenesis

in Xenopus laevis. Dev. Biol. 93:478–97

Dreyfuss G, Matunis MJ, Pi˜nol-Roma S, Burd

CG. 1993. hnRNP proteins and the biogene-

sis of mRNA. Annu. Rev. Biochem. 62:289–

321

Dworetzky SI, Feldherr CM. 1988. Transloca-

tion of RNA-coated gold particles through

the nuclear pores of oocytes. J. Cell Biol.

106:575–84

Eckner R, Ellmeier W, Birnstiel ML. 1991.Ma-

turemRNA3

endformation stimulates RNA

export from the nucleus. EMBO J. 10:3513–

ElliottDJ,StutzF,LescureA,RosbashM.1994.

mRNA nuclear export. Curr. Opin. Genet.

Dev. 4:305–9

Emerman M, Vazeux R, Peden K. 1989. The

rev gene product of the human immunode-

ﬁciency virus affects envelope-speciﬁc RNA

localization. Cell 57:1155–65

Fabre E, Boelens WC, Wimmer C, Mattaj IW,

Hurt ED. 1994. Nup145p is required for nu-

clear export of mRNA and binds homopoly-

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

296 NAKIELNY ET AL

meric RNA in vitro via a novel conserved

motif. Cell 78:275–89

Featherstone C, Darby MK, Gerace L. 1988. A

monoclonalantibodyagainstthenuclear pore

complex inhibits nucleocytoplasmic trans-

port of protein and RNA in vivo. J. Cell Biol.

107:1289–97

FelberBK, Hadzopoulou-Cladaras M, Cladaras

C, Copeland T, Pavlakis GN. 1989. Rev pro-

teinof human immunodeﬁciencyvirus type 1

affects the stability and transport of the viral

mRNA. Proc. Natl. Acad. Sci. USA 86:1495–

Fischer U, Darzynkiewicz E, Tahara SM,

Dathan NA, L¨uhrmann R, et al. 1991. Di-

versity in the signals required for nuclear ac-

cumulation of U snRNPs and variety in the

pathways of nuclear transport. J. Cell Biol.

113:705–14

Fischer U, Heinrich J, van Zee K, Fanning E,

L¨uhrmann R.1994a. Nuclear transport of U1

snRNP in somaticcells: differences in signal

requirement compared with Xenopus laevis

oocytes. J. Cell Biol. 125:971–80

Fischer U, Huber J, Boelens WC, Mattaj IW,

L¨uhrmann R. 1995. The HIV1 Rev activa-

tion domain is a nuclear export signal that

accesses an export pathway used by speciﬁc

cellular RNAs. Cell 82:475–83

Fischer U, L¨uhrmann R. 1990. Anessential sig-

nallingroleforthem

Gcapinthetransportof

U1 snRNP to the nucleus. Science 249:786–

Fischer U, Meyer S, Teufel M, Heckel C,

L¨uhrmann R, Rautmann G. 1994b. Evidence

that HIV1 Rev directly promotes the nuclear

exportofunspliced RNA.EMBO J.13:4105–

Fischer U, Sumpter V, Sekine M, Satoh

T, L¨uhrmann R. 1993. Nucleo-cytoplasmic

transport of U snRNPs: deﬁnition of a nu-

clear location signal in the Sm core domain

that binds a transport receptor independently

of the m

G cap. EMBO J. 12:573–83

Forrester W, Stutz F, Rosbash M, Wickens M.

1992. Defects in mRNA 3

-end formation,

transcription initiation, and mRNA transport

associated with the yeast mutation prp20:

possible coupling of mRNA processing and

chromatin structure. Genes Dev. 6:1914–26

FortesP,BelosoA,Ort´ınJ.1994.Inﬂuenzavirus

NS1 protein inhibits pre-mRNA splicing and

blocks mRNA nucleocytoplasmic transport.

EMBO J. 13:704–12

Fraser KB. 1967. Immunoﬂuorescence of

abortiveandcomplete infectionsbyinﬂuenza

A virus in hamster BHK21 cells and mouse

L cells. J. Gen. Virol. 1:1–12

Fridell RA, Fischer U, L¨uhrmann R, Meyer BE,

Meinkoth JL, et al. 1996. Amphibian tran-

scription factor IIIA proteins contain a se-

quenceelementfunctionallyequivalenttothe

nuclear export signal of human immunodeﬁ-

ciency virus type I Rev. Proc. Natl. Acad. Sci

USA 93:2936–40

Fritz CC, Zapp ML, Green MR.1995. A human

nucleoporin-like protein that speciﬁcally in-

teracts with HIV Rev. Nature 376:530–33

G¨orlichD,MattajIW.1996.Nucleocytoplasmic

transport. Science 271:1513–18

Gorsch LC, Dockendorff TC, Cole CN. 1995.

A conditional allele of the novel repeat-

containing yeast nucleoporin RAT7/NUP159

causes both rapid cessation of mRNA export

andreversibleclusteringofnuclearporecom-

plexes. J. Cell Biol. 129:939–55

Gottlieb E, Steitz JA. 1989. Function of the

mammalian La protein: evidence for its ac-

tion in transcription termination by RNA

polymerase III. EMBO J. 8:851–61

GrossMK,Kainz MS,MerrillGF.1987. Introns

are inconsequential to efﬁcient formation of

cellular thymidine kinase mRNA in mouse L

cells. Mol. Cell. Biol. 7:4576–81

Guddat U, Bakken AH, Pieler T. 1990. Protein-

mediated nuclear export of RNA: 5S rRNA

containing small RNPs in Xenopus oocytes.

Cell 60:619–28

Halbert DN, Cutt JR, Shenk T. 1985. Aden-

ovirus early region 4 encodes functions re-

quired for efﬁcient DNA replication, late

gene expression, and host cell shutoff. J. Vi-

rol. 56:250–57

Hamm J, Darzynkiewicz E, Tahara SM, Mattaj

IW. 1990. The trimethylguanosine cap struc-

ture of U1 snRNA is a component of a bipar-

tite nuclear targeting signal. Cell 62:569–77

HammJ,MattajIW.1990.Monomethylatedcap

structures facilitate RNA export from the nu-

cleus. Cell 63:109–18

Hattori K, Angel P, Le Beau MM, Karin M.

1988. Structure and chromosomal localiza-

tion of the functional intronless human JUN

protooncogene. Proc. Natl. Acad. Sci. USA

85:9148–52

Heinzinger NK, Bukrinsky MI, Haggerty SA,

Ragland AM, Kewalramani V, et al. 1994.

TheVpr proteinof humanimmunodeﬁciency

virustype 1 inﬂuencesnuclear localization of

viral nucleic acids in nondividing host cells.

Proc. Natl. Acad. Sci. USA 91:7311–15

Hogan NC, Traverse KL, Sullivan DE, Pardue

M-L. 1994. The nucleus-limited Hsr-omega-

n transcript is a polyadenylated RNA with a

regulated intranuclear turnover. J. Cell Biol.

125:21–30

Honda BM, Roeder RG. 1980. Association of

a 5S gene transcription factor with 5S RNA

andaltered levelsof the factorduring cell dif-

ferentiation. Cell 22:119–26

Hope TJ, Bond BL, McDonald D, Klein NP,

Parslow TG. 1991. Effector domains of

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 297

human immunodeﬁciency virus type 1 Rev

and human T-cell leukemia virus type I Rex

arefunctionally interchangeable and sharean

essential peptide motif. J. Virol. 65:6001–7

HuangZ-M, YenTSB. 1995.Roleof the hepati-

tis B virus posttranscriptional regulatory ele-

ment in export of intronless transcripts. Mol.

Cell. Biol. 15:3864–69

Izaurralde E, Lewis J, Gamberi C, Jarmolowski

A, McGulgan C, et al. 1995. A cap-binding

protein complex mediating U snRNA export.

Nature 376:709–12

Izaurralde E, Lewis J, McGuigan C, Jankowska

M, Darzynkiewicz E, Mattaj IW. 1994. Anu-

clear cap binding protein complex involved

in pre-mRNA splicing. Cell 78:657–68

Izaurralde E, Mattaj IW. 1992. Transport of

RNAbetweennucleusandcytoplasm.Semin.

Cell Biol. 3:279–88

IzaurraldeE,MattajIW.1995.RNAexport.Cell

81:153–59

Izaurralde E, Stepinski J, Darzynkiewicz E,

Mattaj IW. 1992. A cap binding protein that

may mediate nuclear export of RNA poly-

merase II-transcribed RNAs. J. Cell Biol.

118:1287–95

Jarmolowski A, Boelens WC, Izaurralde E,

Mattaj IW. 1994. Nuclear export of differ-

ent classes of RNA is mediated by speciﬁc

factors. J. Cell Biol. 124:627–35

Jarmolowski A, Mattaj IW. 1993. The deter-

minants for Sm protein binding to Xenopus

U1 and U5 snRNAs are complex and non-

identical. EMBO J. 12:223–32

Kadowaki T, Chen S, Hitomi M, Jacobs E,

Kumagai C, et al. 1994a. Isolation and char-

acterization of Saccharomyces cerevisiae

mRNA transport-defective (mtr) mutants. J.

Cell Biol. 126:649–59

Kadowaki T, Goldfarb D, Spitz LM, Tartakoff

AM, Ohno M. 1993. Regulation of RNApro-

cessing and transport by a nuclear guanine

nucleotide release protein and members of

the Ras superfamily. EMBO J. 12:2929–37

Kadowaki T, Hitomi M, Chen S, Tartakoff AM.

1994b. Nuclear mRNA accumulation causes

nucleolar fragmentation in yeast mtr2 mu-

tant. Mol. Biol. Cell 5:1253–63

Kadowaki T, Schneiter R, Hitomi M, Tartakoff

AM. 1995. Mutations in nucleolar proteins

lead to nucleolar accumulation of polyA

RNAinSaccharomycescerevisiae.Mol.Biol.

Cell 6:1103–10

Kadowaki T, Zhao Y, Tartakoff AM. 1992. A

conditional yeast mutant deﬁcient in mRNA

transport from nucleus to cytoplasm. Proc.

Natl. Acad. Sci. USA 89:2312–16

Kalland K-H, Szilvay AM, Brokstad KA,

Sætrevik W, Haukenes G. 1994. The human

immunodeﬁciency virus type 1 Rev protein

shuttles between the cytoplasm and nuclear

compartments. Mol. Cell. Biol. 14:7436–

Kataoka N, Ohno M, Kangawa K, Tokoro Y,

Shimura Y. 1994. Cloning of a complemen-

tary DNA encoding an 80 kilodalton nu-

clear cap binding protein. Nucleic Acids Res.

22:3861–65

Kataoka N, Ohno M, Moda I, Shimura Y. 1995.

Identiﬁcation of the factors that interact with

NCBP, an 80 kDa nuclear cap binding pro-

tein. Nucleic Acids Res. 23:3638–41

Keller W. 1995. No end yet to messenger RNA

processing! Cell 81:829–32

Khanna-Gupta A, Ware VC. 1989. Nucleocy-

toplasmic transport of ribosomes in a eu-

karyotic system: Is there a facilitated trans-

port process? Proc. Natl. Acad. Sci. USA

86:1791–95

Kjems J, Frankel AD, Sharp PA. 1991. Speciﬁc

regulation of mRNA splicing in vitro by a

peptide from HIV1 Rev. Cell 67:169–78

Krug RM. 1993. The regulation of export of

mRNA from nucleus to cytoplasm. Curr.

Opin. Cell Biol. 5:944– 49

Kurokawa M, Ochiai H, Nakajima K, Niwa-

yama S. 1990. Inhibitory effect of protein

kinase C inhibitor on the replication of in-

ﬂuenza type A virus. J. Gen. Virol. 71:2149–

Legrain P, Rosbash M. 1989. Some cis- and

trans-acting mutants for splicing target pre-

mRNA to the cytoplasm. Cell 57:573–83

Leppard KN, Shenk T. 1989. The adenovirus

E1B 55 kd protein inﬂuences mRNA trans-

port via an intranuclear effect on RNA

metabolism. EMBO J. 8:2329–36

Liang S, Hitomi M, Hu Y-H, Liu Y, Tartakoff

AM. 1996. A DEAD-box family protein is

required for nucleocytoplasmic transport of

yeast mRNA. Mol. Biol. Cell 16:5139–46

Liou R-F, Blumenthal T. 1990. trans-spliced

Caenorhabditis elegans mRNAs retain

trimethylguanosine caps. Mol. Cell. Biol.

10:1764–68

Liu X, Mertz JE. 1995. HnRNP L binds a cis-

acting RNA sequence element that enables

intron-independent gene expression. Genes

Dev. 9:1766–80

Lu X, Heimer J, Rekosh D, Hammarskj¨old

M-L. 1990. U1 small nuclear RNA plays a

direct role inthe formation of a rev-regulated

human immunodeﬁciency virus env mRNA

that remains unspliced. Proc.Natl. Acad. Sci.

USA 87:7598–602

Lu Y, Qian X-Y, Krug RM. 1994. The inﬂuenza

virus NS1 protein: a novel inhibitor of pre-

mRNA splicing. Genes Dev. 8:1817–28

L¨uhrmann R, Kastner B, Bach M. 1990. Struc-

ture of splicesomal snRNPs and their role in

pre-mRNA splicing. Biochim. Biophys. Acta

Gene Struct. Express. 1087:265–92

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

298 NAKIELNY ET AL

Luo Y, Yu H, Peterlin BM. 1994. Cellular pro-

teinmodulateseffectsofhumanimmunodeﬁ-

ciency virus type 1 Rev. J. Virol. 68:3850–56

Mairy M, Denis H. 1971. Recherches biochim-

iques sur l’oogenese. 1. Synthese et accumu-

lation du RNA pendent l’oogenese du cra-

paud sud-africain Xenopus laevis. Dev. Biol.

24:143–65

Malim MH, B¨ohnlein S, Hauber J, Cullen BR.

1989a. Functional dissection of the HIV1

Rev trans-activator: derivation of a trans-

dominant repressor of Rev function. Cell

58:205–14

Malim MH, Cullen BR. 1991. HIV1 structural

gene expression requires the binding of mul-

tiple Rev monomers to the viral RRE: impli-

cations for HIV1 latency. Cell 65:241–48

MalimMH,HauberJ,LeS-Y,MaizelJV,Cullen

BR.1989b.The HIV1revtrans-activatoracts

through a structured target sequence to acti-

vatenuclearexport ofunspliced viral mRNA.

Nature 338:254–57

Malim MH, McCarn DF, Tiley LS, Cullen BR.

1991. Mutational deﬁnition of the human im-

munodeﬁciency virus type 1 Rev activation

domain. J. Virol. 65:4248–54

Maquat LE. 1996. RNA transport from the

nucleus. In Modern Cell Biology, Post-

TranscriptionalGeneRegulation,ed.JBHar-

ford, DK Morris, New York: Wiley. In press

Marshallsay C, L¨uhrmann R. 1994. In vitro nu-

clear import of snRNPs: cytosolic factors

mediate m

G-cap dependence of U1 and U2

snRNP transport. EMBO J. 13:222–31

Martin K, Helenius A. 1991. Nuclear transport

of inﬂuenza virus ribonucleoproteins: the vi-

ral matrix protein (M1) promotes export and

inhibits import. Cell 67:117–30

Mattaj IW. 1986. Cap trimethylation of U

snRNA is cytoplasmic and dependent on U

snRNP protein binding. Cell 46:905–11

Mattaj IW. 1988.U snRNP assembly and trans-

port. In Structure and Function of Major

and Minor Small Nuclear Ribonucleopro-

tein Particles, ed. ML Birnstiel, pp. 100–

14. Berlin/Heidelberg/New York: Springer-

Verlag

Mattaj IW, DeRobertis EM. 1985. Nuclear seg-

regation of U2 snRNA requires binding of

speciﬁc snRNP proteins. Cell 40:111–18

Matunis EL, Matunis MJ, Dreyfuss G. 1993.

Association of individual hnRNP proteins

and snRNPs with nascent transcripts. J. Cell

Biol. 121:219–28

Mayeda A, Krainer AR. 1992 Regulation of al-

ternative pre-mRNA splicing by hnRNP A1

and splicing factor SF2. Cell 68:365–75

Mehlin H, Daneholt B. 1993. The Balbiani ring

particle: a model for the assembly and export

of RNPs from the nucleus? Trends Cell Biol.

3:443–47

Mehlin H, Daneholt B, Skoglund U. 1992.

Translocation of a speciﬁc premessenger ri-

bonucleoprotein particle through the nuclear

pore studied with electron microscope to-

mography. Cell 69:605–13

Mehlin H, Skoglund U, Daneholt B. 1991.

Transport of Balbiani ring granules through

nuclear pores in Chironomus tentans. Exp.

Cell Res. 193:72–77

MelchiorF, PaschalB, EvansJ, Gerace L. 1993.

Inhibition of nuclear protein import by non-

hydrolyzable analogues of GTP and iden-

tiﬁcation of the small GTPase Ran/TC4 as

an essential transport factor. J. Cell Biol.

123:1649–59

Melton DA, De Robertis EM, Cortese R. 1980.

Order and intracellular location of the events

involved in thematuration of a spliced tRNA.

Nature 284:143–48

Mermer B, Felber BK, Campbell M, Pavlakis

GN. 1990. Identiﬁcation of trans-dominant

HIV1 rev protein mutants by direct transfer

of bacterially produced proteins into human

cells. Nucleic Acids Res. 18:2037–44

Meyer BE, Malim MH. 1994. The HIV1 Rev

trans-activator shuttles between the nucleus

and the cytoplasm. Genes Dev. 8:1538–

MeyerBE,Meinkoth JL,Malim MH.1996. Nu-

clear transport of human immunodeﬁciency

virus type 1, visna virus, and equine infec-

tious anemia virus Rev proteins: identiﬁca-

tion of a familyof transferable nuclear export

signals. J. Virol. 70:2350–59

Michael WM, Choi M, Dreyfuss G. 1995.

A nuclear export signal in hnRNP A1: a

signal-mediated, temperature-dependent nu-

clear protein exportpathway.Cell 83:415–22

Michael WM, Dreyfuss G. 1996. Distinct do-

mains in ribosomal protein L5 mediate 5S

rRNA binding and nucleolar localization. J.

Biol. Chem. 271:11571–74

Michael WM, Siomi H, Choi M, Pi˜nol-Roma

S, Nakielny S, et al. 1996. Signal sequences

that target nuclear import and nuclear export

of pre-mRNA-binding proteins. Cold Spring

Harbor Symp. Quant. Biol. LX:663– 68

Michaud N, Goldfard DS. 1991. Multiple path-

ways in nuclear transport: The import of U2

snRNP occurs by a novel kinetic pathway. J.

Cell Biol. 112:215–23

Michaud N, Goldfarb D. 1992. Microinjected

U snRNAs are imported to oocyte nuclei via

the nuclear pore complex by three distin-

guishable targeting pathways. J. Cell Biol.

116:851–61

Micklem DR. 1995. mRNA localisation during

development. Dev. Biol. 172:377–95

Moore MJ, Query CC, Sharp PA. 1993. Splic-

ing of precursors to mRNAs by the spliceo-

some. In The RNA World, ed. RF Gesteland,

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 299

JF Atkins, 13:303–57. Plainview, NY: Cold

Spring Harbor Lab.

Moore MS, Blobel G. 1993. The GTP-binding

protein Ran/TC4 is required for protein im-

port into the nucleus. Nature 365:661–63

Munroe SH, Dong X. 1992. Heterogeneous nu-

clear ribonucleoprotein A1 catalyzes RNA-

RNA annealing. Proc. Natl. Acad. Sci. USA

89:895–99

Nagata S, Mantei N, Weissmann C. 1980. The

structure of one of the eight or more distinct

chromosomal genes for human interferon-

alpha. Nature 287:401–8

Nakielny S, Dreyfuss G. 1996. The hnRNP C

proteins contain a nuclearretention sequence

that can override nuclear export signals. J.

Cell Biol. 134:1365–73

Ogert RA, Lee LH, Beemon KL. 1996. Avian

retroviral RNA element promotes unspliced

RNA accumulation inthe cytoplasm. J.Virol.

70:3834–43

Ohno M, Kataoka N, Shimura Y. 1990. A nu-

clear cap binding protein from HeLa cells.

Nucleic Acids Res. 18:6989–95

Okita TW, Li X, Roberts MW. 1994. Targeting

of mRNAs to domains of the endoplasmic

reticulum. Trends Cell Biol. 4:91–96

O’Neill RE, Jaskunas R, Blobel G, Palese P,

Moroianu J. 1995. Nuclear import of in-

ﬂuenza virus RNA can be mediated by viral

nucleoprotein and transport factors required

forproteinimport.J.Biol.Chem.270:22701–

Ornelles DA, Shenk T. 1991. Localization of

the adenovirus early region 1B 55-kilodalton

protein during lytic infection: Association

with nuclear viral inclusions requires the

early region 4 34-kilodalton protein. J. Virol.

65:424–39

Peebles CL, Gegenheimer P, Abelson J. 1983.

Precise excision of intervening sequences

from precursor tRNAs by a membrane-

associated yeast endonuclease. Cell 32:525–

PilderS,MooreM,Logan J,Shenk T.1986. The

adenovirus E1B55K transforming polypep-

tide modulates transport or cytoplasmic sta-

bilization of viral and host cell mRNAs. Mol.

Cell. Biol. 6:470–76

Pi˜nol-Roma S, Choi YD, Matunis MJ, Drey-

fuss G. 1988. Immunopuriﬁcation of hetero-

geneous nuclear ribonucleoprotein particles

reveals an assortment of RNA-binding pro-

teins. Genes Dev. 2:215–27

Pi˜nol-Roma S, Dreyfuss G. 1991. Trans-

cription-dependent and transcription-inde-

pendentnuclear transportof hnRNP proteins.

Science 253:312–14

Pi˜nol-Roma S, Dreyfuss G. 1992. Shuttling of

pre-mRNAbindingproteins betweennucleus

and cytoplasm. Nature 355:730–32

Pi˜nol-Roma S, Dreyfuss G. 1993. hnRNP pro-

teins: localization and transport between the

nucleus and the cytoplasm. Trends Cell Biol.

3:151–55

Pi˜nol-Roma S, Swanson MS, Gall JG, Dreyfuss

G.1989.A novelheterogeneousnuclear RNP

protein with a unique distribution on nascent

transcripts. J. Cell Biol. 109:2575–87

Pokrywka NJ, Goldfarb DS. 1995. Nuclear ex-

port pathways of tRNA and 40S ribosomes

include both common and speciﬁc interme-

diates. J. Biol. Chem. 270:3619–24

PortmanDS, Dreyfuss G.1994. RNAannealing

activities in HeLa nuclei. EMBO J. 13:213–

Qian X-Y, Alonso-Caplen F, Krug RM. 1994.

Twofunctionaldomainsoftheinﬂuenzavirus

NS1 protein are required for regulation of

nuclear export of mRNA. J. Virol. 68:2433–

QiuY, KrugRM.1994. Theinﬂuenza virusNS1

protein is a poly(A)-binding protein that in-

hibits nuclear export of mRNAs containing

poly(A). J. Virol. 68:2425–32

Reddy R, Bush H. 1988. In Structure and

Function of Major and Minor Small Nuclear

Ribonucleoprotein Particles, ed. ML Birn-

stiel, pp. 1–37. Berlin/Heidelberg/New York:

Springer-Verlag

RinkeJ, SteitzJA.1982. Precursormolecules of

both human 5S ribosomal RNA and transfer

RNAs are bound by cellular protein reactive

with anti-La lupus antibodies. Cell 29:711–

Rousseau D, Kaspar R, Rosenwald I, Gehrke

L, Sonenberg N. 1996. Translation initiation

of ornithine decarboxylase and nucleocyto-

plasmic transport of cyclin D1 mRNA are

increased in cells overexpressing eukaryotic

initiation factor 4E. Proc. Natl. Acad. Sci.

USA 93:1065–70

Rout MP, Wente SR. 1994. Pores for thought:

nuclear pore complex proteins. Trends Cell

Biol. 4:357–65

Rudt R, Pieler T. 1996. Cytoplasmic retention

andnuclearimportof5SribosomalRNAcon-

taining RNPs. EMBO J. 15:1383–91

Saavedra C, Tung K-S, Amberg DC, Hopper

AK, Cole CN. 1996. Regulation of mRNA

exportinresponse tostress inSaccharomyces

cerevisiae. Genes Dev. 10:1608–20

St. Johnston D. 1995a. The intracellular local-

ization of messenger RNAs. Cell 81:161–

St. Johnston D. 1995b. New role for tropo-

myosin. Nature 377:483

Santos T, Zasloff M. 1981. Comparative anal-

ysis of human chromosomal segments bear-

ing nonallelic dispersed tRNA

Met

genes. Cell

23:699–709

Sazer S. 1996. The search for the primary func-

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

300 NAKIELNY ET AL

tion of the Ran GTPase continues. Trends

Cell Biol. 6:81– 85

Scheer U, Weisenberger D. 1994. The nucleo-

lus. Curr. Opin. Cell Biol. 6:354–59

Schlenstedt G, Saavedra C, Loeb JDJ, Cole CN,

Silver PA. 1995. The GTP-bound form of

theyeastRan/TC4homologueblocksnuclear

protein import and appearance of poly(A)

RNA in the cytoplasm. Proc. Natl. Acad.Sci.

USA 92:225–29

Schneiter R, Kadowaki T, Tartakoff AM. 1995.

mRNA transport in yeast: time to reinvesti-

gatethe functions of the nucleolus.Mol. Biol.

Cell 6:357–70

Shapiro GI, Gurney T Jr, Krug RM. 1987. In-

ﬂuenza virus gene expression: control mech-

anisms at early and late times of infection

and nuclear-cytoplasmic transport of virus-

speciﬁc RNAs. J. Virol. 61:764–73

Sharma K, Fabre E, Tekotte H, Hurt ED, Toller-

vey D. 1996. Yeast nucleoporin mutants are

defective in pre-tRNA splicing. Mol. Cell.

Biol. 16:294–301

Shaw PJ, Jordan EG. 1995. The nucleolus.

Annu. Rev. Cell Dev. Biol. 11:93–121

ShumanS.1995.Cappingenzyme ineukaryotic

mRNA synthesis. Prog. Nucleic Acid Res.

Mol. Biol. 50:101–29

Singer RH. 1993. RNA zipcodes for cytoplas-

mic addresses. Curr. Biol. 3:719–21

Singh R, Green MR. 1993. Sequence-speciﬁc

bindingoftransfer RNAbyglyceraldehyde3-

phosphate dehydrogenase. Science 259:365–

Siomi H, Dreyfuss G. 1995. A nuclear localiza-

tion domain inthe hnRNP A1 protein. J. Cell

Biol. 129:551–60

SkoglundU,Andersson K,StrandbergB, Dane-

holt B. 1986. Three-dimensional structure of

a speciﬁc pre-messenger RNP particle estab-

lished by electron microscope tomography.

Nature 319:560–64

Stebbings H, Lane JD, Talbot NJ. 1995. mRNA

translocation and microtubules: insect ovary

models. Trends Cell Biol. 5:361–65

SteitzJA,BergC,Hendrick JP, Branche-Chabot

H, Metspalu A, et al. 1988. A 5S rRNA/L5

complex is a precursor to ribosome assembly

in mammalian cells. J. Cell Biol. 106:545–

Stevens BJ, SwiftH. 1966. RNA transport from

nucleus to cytoplasm in Chironomus salivary

glands. J. Cell Biol. 31:55–77

Steward O. 1995. Targeting of mRNAs to sub-

synaptic microdomains in dendrites. Curr.

Opin. Neurobiol. 5:55–61

StutzF,NevilleM,RosbashM.1995.Identiﬁca-

tion of a novel nuclear pore-associated pro-

tein as a functional target of the HIV1 Rev

protein in yeast. Cell 82:495–506

Stutz F, Rosbash M. 1994. A functional inter-

action between Rev and yeast pre-mRNA is

relatedtosplicingcomplexformation.EMBO

J. 13:4096–104

Sun J, Pilch DR, Marzluff WF. 1992. The his-

tonemRNA3

endisrequiredforlocalization

of histone mRNA to polyribosomes. Nucleic

Acids Res. 20:6057–66

Sweet DJ, Gerace L. 1995. Taking from the cy-

toplasmand givingto the pore: soluble trans-

port factors in nuclear protein import. Trends

Cell Biol. 5:444–47

Tachibana T, Imamoto N, Seino H, Nishimoto

T, Yoneda Y. 1994. Loss of RCC1 leads to

suppression of nuclear protein import in liv-

ing cells. J. Biol. Chem. 269:24542–45

Tafuri SR, WolffeAP. 1993. Dual roles for tran-

scription and translation factors in the RNA

storage particles of Xenopus oocytes. Trends

Cell Biol. 3:94–98

Tartakoff AM, Schneiter R. 1995. The nuclear

GTPase cycle: promoting peripheralization?

Trends Cell Biol. 5:5–8

Terns MP, Dahlberg JE, Lund E. 1993. Muli-

ple cis-acting signals for export of pre-

U1 snRNA from the nucleus. Genes Dev.

7:1898–906

Tobian JA, Drinkard L, Zasloff M. 1985. tRNA

nuclear transport: deﬁning the critical re-

gions of human tRNA

Met

by point mutage-

nesis. Cell 43:415–22

Van Doren K, Hirsh D. 1990. mRNAs that ma-

ture through trans-splicing in Caenorhabdi-

tis elegans have a trimethylguanosine cap at

their 5

termini. Mol. Cell. Biol. 10:1769–72

vanZee K, DickmannsA, Fischer U, L¨uhrmann

R, Fanning E. 1993. A cytoplasmically an-

chored nuclear protein interferes speciﬁcally

with the import of nuclear proteins but not

U1 snRNA. J. Cell Biol. 121:229–40

Venkatesh LK, Chinnadurai G. 1990. Mutants

inaconservedregionnearthecarboxy-termi-

nusof HIV1 Revidentify functionally impor-

tant residues and exhibit a dominant negative

phenotype. Virology 178:327–30

Vijayraghavan U, Company M, Abelson J.

1989. Isolation and characterization of pre-

mRNA splicing mutants of Saccharomyces

cerevisiae. Genes Dev. 3:1206–16

Visa N, Alzhanova-Ericsson AT, Sun X, Kise-

leva E, Bj¨orkroth B, et al. 1996a. A

pre-mRNA-bindingprotein accompaniesthe

RNAfrom thegene through thenuclear pores

and into polysomes. Cell 84:253–64

Visa N, Izaurralde E, Ferreira J, Daneholt

B, Mattaj IW. 1996b. A nuclear cap bind-

ing complex binds Balbiani ring pre-mRNA

co-transcriptionally and accompanies the ri-

bonucleoprotein particle during nuclear ex-

port. J. Cell Biol. 133:5–14

Vogel U, Kunerl M, Scholtissek C. 1994. In-

ﬂuenza A virus late mRNAs are speciﬁcally

November 11, 1997 22:21 Annual Reviews AR024-11 AR24-11

RNA TRANSPORT 301

retained in the nucleus in the presence of

a methyltransferase or a protein kinase in-

hibitor. Virology 198:227–33

von Schwedler U, Kornbluth RS, Trono D.

1994. The nuclear localization signal of the

matrix protein of human immunodeﬁciency

virustype1allowstheestablishmentof infec-

tioninmacrophagesandquiescent Tlympho-

cytes.Proc.Natl. Acad.Sci. USA91:6992–96

Warner JR. 1990. The nucleolus and ribosome

formation. Curr. Opin. Cell Biol. 2:521–27

Weichselbraun I, Farrington GK, Rusche JR,

B¨ohnlein E, Hauber J. 1992. Deﬁnition of the

human immunodeﬁciency virus type 1 Rev

and human T-cell leukemia virus type I Rex

protein activation domain by functional ex-

change. J. Virol. 66:2583–87

Wen W, Meinkoth JL, Tsien RY, Taylor SS.

1995. Identiﬁcation of a signal for rapid

export of proteins from the nucleus. Cell

82:463–73

Westaway SK, Abelson J. 1995. Splicing of

tRNA precursors. In tRNA Structure, Biosyn-

thesis, and Function, ed.DS¨oll, UL Raj-

Bhandary, 7:79–92. Washington, DC: ASM

Press

Whittaker G, Bui M, Helenius A. 1996. The

role of nuclear import and export in inﬂuenza

virus infection. Trends Cell Biol. 6:67–71

Whittaker G, Kemler I, Helenius A. 1995. Hy-

perphosphorylationof mutant inﬂuenza virus

matrix protein, M1, causes its retention in the

nucleus. J. Virol. 69:439–45

Wilhelm JE, Vale RD. 1993. RNA on the move:

the mRNA localization pathway. J. Cell Biol.

123:269–74

Wolff B, Cohen G, Hauber J, Meshcheryakova

D, Rabeck C. 1995. Nucleocytoplasmic

transport of the Rev protein of human im-

munodeﬁciency virus type 1 is dependent on

theactivationdomainoftheprotein.Exp.Cell

Res. 217:31–41

Wolffe AP, Brown DD. 1988. Developmental

regulation of two 5S ribosomal RNA genes.

Science 241:1626–32

Yang X, Bani M-R, Lu S-J, Rowan S, Ben-

David Y, Chabot B. 1994. The A1 and A1

proteins of heterogeneous nuclear ribonucle-

oparticles modulate 5

splice site selection in

vivo. Proc. Natl. Acad. Sci. USA 91:6924–

Ye Z, Robinson D, Wagner RR. 1995. Nucleus-

targeting domain of the matrix protein (M

)

of inﬂuenza virus. J. Virol. 69:1964–70

Zapp ML, Green MR. 1989. Sequence-speciﬁc

RNA binding by the HIV1 Rev protein. Na-

ture 342:714–16

Zasloff M. 1983. tRNA transport from the nu-

cleus in a eukaryotic cell: carrier-mediated

translocation process. Proc. Natl. Acad. Sci.

USA 80:6436–40

Zasloff M, Rosenberg M, Santos T. 1982a. Im-

paired nuclear transport of a human variant

tRNA

Met

. Nature 300:81–84

Zasloff M, Santos T, Hamer DH. 1982b. tRNA

precursor transcribed from a mutant human

gene inserted into a SV40 vector is processed

incorrectly. Nature 295:533–35

Transcriptome Analysis of the Marine Nematode Litoditis marina in a Chemically Defined Food Environment with Stearic Acid Supplementation

Article

Full-text available

Mar 2022

Stearic acid represents one of the most abundant fatty acids in the Western diet and profoundly regulates health and diseases of animals and human beings. We previously showed that stearic acid supplementation promoted development of the terrestrial model nematode Caenorhabditis elegans in chemically defined CeMM food environment. However, whether stearic acid regulates development of other nematodes remains unknown. Here, we found that dietary supplementation with stearic acid could promote the development of the marine nematode Litoditis marina, belonging to the same family as C. elegans, indicating the conserved roles of stearic acid in developmental regulation. We further employed transcriptome analysis to analyze genome-wide transcriptional signatures of L. marina with dietary stearic acid supplementation. We found that stearic acid might promote development of L. marina via upregulation of the expression of genes involved in aminoacyl-tRNA biosynthesis, translation initiation and elongation, ribosome biogenesis, and transmembrane transport. In addition, we observed that the expression of neuronal signaling-related genes was decreased. This study provided important insights into how a single fatty acid stearic acid regulates development of marine nematode, and further studies with CRISPR genome editing will facilitate demonstrating the molecular mechanisms underlying how a single metabolite regulates animal development and health.

Dual Promoters Improve the Rescue of Recombinant Measles Virus in Human Cells

Article

Full-text available

Aug 2021

Reverse genetics is a technology that allows the production of a virus from its complementary DNA (cDNA). It is a powerful tool for analyzing viral genes, the development of novel vaccines, and gene delivery vectors. The standard reverse genetics protocols are laborious, time-consuming, and inefficient for negative-strand RNA viruses. A new reverse genetics platform was established, which increases the recovery efficiency of the measles virus (MV) in human 293-3-46 cells. The novel features compared with the standard system involving 293-3-46 cells comprise (a) dual promoters containing the RNA polymerase II promoter (CMV) and the bacteriophage T7 promoter placed in uni-direction on the same plasmid to enhance RNA transcription; (b) three G nucleotides added just after the T7 promoter to increase the T7 RNA polymerase activity; and (c) two ribozymes, the hairpin hammerhead ribozyme (HHRz), and the hepatitis delta virus ribozyme (HDVrz), were used to cleavage the exact termini of the antigenome RNA. Full-length antigenome cDNA of MV of the wild type IC323 strain or the vaccine AIK-C strain was inserted into the plasmid backbone. Both virus strains were easily rescued from their respective cloned cDNA. The rescue efficiency increased up to 80% compared with the use of the standard T7 rescue system. We assume that this system might be helpful in the rescue of other human mononegavirales.

Functional roles of hnRNPA2/B1 regulated by METTL3 in mammalian embryonic development

Article

Full-text available

Jun 2019

Heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNPA2/B1) plays an important role in RNA processing via in m6A modification of pre-mRNA or pre-miRNA. However, the functional role of and relationship between m6A and hnRNPA2/B1 in early embryonic development are unclear. Here, we found that hnRNPA2/B1 is crucial for early embryonic development by virtue of regulating specific gene transcripts. HnRNPA2/B1 was localized to the nucleus and cytoplasm during subsequent embryonic development, starting at fertilization. Knockdown of hnRNPA2/B1 delayed embryonic development after the 4-cell stage and blocked further development. RNA-Seq analysis revealed changes in the global expression patterns of genes involved in transcription, translation, cell cycle, embryonic stem cell differentiation, and RNA methylation in hnRNPA2/B1 KD blastocysts. The levels of the inner cell mass markers OCT4 and SOX2 were decreased in hnRNPA2/B1 KD blastocysts, whereas that of the differentiation marker GATA4 was decreased. N6-Adenosine methyltransferase METTL3 knock-down caused embryonic developmental defects similar to those in hnRNPA2/B1 KD embryos. Moreover, METTL3 KD blastocysts showed increased mis-localization of hnRNPA2/B1 and decreased m6A RNA methylation. Taken together, our results suggest that hnRNPA2/B1 is essential for early embryogenesis through the regulation of transcription-related factors and determination of cell fate transition. Moreover, hnRNPA2/B1 is regulated by METTL3-dependent m6A RNA methylation.

Staufen2 deficiency leads to impaired response to novelty in mice

Article

Feb 2018

Staufen2 (Stau2) is a double-stranded RNA-binding protein (RBP) involved in posttranscriptional gene expression control in neurons. In flies, staufen contributes to learning and long-term memory formation. To study the impact of mammalian Stau2 on behavior, we generated a novel gene-trap mouse model that yields significant constitutive downregulation of Stau2 (Stau2GT). In order to investigate the effect of Stau2 downregulation on hippocampus-dependent behavior, we performed a battery of behavioral assays, i.e. open field, novel object recognition/location (NOR/L) and Barnes maze. Stau2GTmice displayed reduced locomotor activity in the open field and altered novelty preference in the NOR and NOL paradigms. Adult Stau2GTmale mice failed to discriminate between familiar and newly introduced objects but showed enhanced spatial novelty detection. Additionally, we observed deficits in discriminating different spatial contexts in a Barnes maze assay. Together, our data suggest that Stau2 contributes to novelty preference and explorative behavior that is a driver for proper spatial learning in mice.

Comparative transcriptome analysis identified genes involved in testicular development in Takifugu rubripes

Article

Full-text available

Jan 2025
FISH PHYSIOL BIOCHEM

To identify candidate genes and pathways involved in testicular development in Takifugu rubripes, a comparative transcription analysis was conducted across the various developmental stages of the testis (stages II to V). A total of 9520 differentially expressed genes (DEGs) were identified among the different stages, and they were significantly clustered into six clusters (P < 0.05). One thousand four hundred eleven DEGs such as gndf, wnt1, and cyp17b1 were found to be decreased from stage II to V. In contrast, 994 DEGs such as fn1, ift81, and cdc25a were found to be increased from stage II to V. Six thousand three hundred eighteen DEGs (e.g., dmrt1, sdk2, and chrna1) were identified as being expressed at similar levels at stages II and III. However, they were subsequently found to be decreased from stage III to IV. Four hundred one DEGs exhibited a significant upregulation trend from stage II to III. These genes were expressed at similar levels in stages III, IV, and V, including chrnd, wnt4a, and cyp7a1. The highest expression levels of 200 DEGs (e.g., ccnb2, cdk1, and sycp2) were observed in stage IV, while 196 DEGs (e.g., chmp1b, hsd17b3, and zp3) exhibited the highest expression level in stage III. Those DEGs were mainly enriched in the pathways (e.g., neuroactive ligand-receptor interaction, cell adhesion molecules, and calcium signaling pathways) associated with testicular development. Quantitative polymerase chain reaction of eight randomly selected genes validated the RNA sequencing results. This study may provide new insights into the molecular regulatory mechanisms governing testicular development and spermatogenesis in T. rubripes.

Transcriptome analysis reveals hub genes in the hepatopancreas of Exopalaemon carinicauda in response to hypoxia and reoxygenation

Article

Full-text available

Aug 2021
AQUACULT INT

Hypoxia is a common stressor in shrimp, but the molecular mechanisms underlying the adaptation of the ridgetail white prawn, Exopalaemon carinicauda, to hypoxia and reoxygenation remain poorly understood. In the present study, ridgetail white prawns were exposed to gradual changes in hypoxia for 6 h via oxygen consumption in a static respiration chamber and then rapidly aerated to allow reoxygenation for 8 h. The dissolved oxygen concentration after 6 h of hypoxia was 1.04 mg/L, and the survival rate of the shrimp was 47%. A transcriptomic analysis of hepatopancreas tissues after 0 (control), 3 (hypoxic starting point), and 6 (hypoxic lethal half point) h of exposure to hypoxia and after 1 and 8 h of reoxygenation was conducted using the Illumina HiSeqTM 4000 high-throughput sequencing platform. A total of 93,227 genes were obtained, and 4315 of these genes were identified as differentially expressed genes (DEGs) that respond to hypoxia and reoxygenation. A KEGG pathway enrichment analysis suggested that all identified DEGs were mostly enriched in ribosome biogenesis in eukaryotes, apoptosis, the longevity regulating pathway, the MAPK signaling pathway, and protein processing in the endoplasmic reticulum. In addition, 20,203 genes with RPKM values ≥1 were parsed into 20 modules through a weighted gene coexpression network (WGCNA), and three of these modules were related to hypoxia and reoxygenation. GO and KEGG enrichment analyses of these three modules were then performed, and three hub genes were identified based on their connectivity: RREB1 (Ras-responsive element-binding protein 1-like), UBE1 (ubiquitin-activating enzyme E1), and an unknown gene. This study might help further elucidate the hypoxia tolerance mechanism of E. carinicauda.

Dynamically probing ATP‐dependent RNA helicase A‐assisted RNA structure conversion using single molecule fluorescence resonance energy transfer

Article

Full-text available

May 2021
PROTEIN SCI

RNA helicase A (RHA) as a member of DExH‐box subgroup of helicase superfamily II, participates in diverse biological processes involved in RNA metabolism in organisms, and these RNA‐mediated biological processes rely on RNA structure conversion. However, how RHA regulate the RNA structure conversion was still unknown. In order to unveil the mechanism of RNA structure conversion mediated by RHA, single molecule fluorescence resonance energy transfer was adopted to in our assay, and substrates RNA were from internal ribosome entry site of foot‐and‐mouth disease virus genome. We first found that the RNA structure conversion by RHA against thermodynamic equilibrium in vitro, and the process of dsRNA YZ converted to dsRNA XY through a tripartite intermediate state. In addition, the rate of the RNA structure conversion and the distribution of dsRNA YZ and XY were affected by ATP concentrations. Our study provides real‐time insight into ATP‐dependent RHA‐assisted RNA structure conversion at the single molecule level, the mechanism displayed by RHA may help in understand how RHA contributes to many biological functions, and the basic mechanistic features illustrated in our work also underlay more complex protein‐assisted RNA structure conversions.

The Cell Nucleus and Its Compartments

Chapter

Oct 2020

Claudia Tanja Mierke

This chapter presents the basic knowledge about the function of cell divisionCell division and the cell cycle process. In fine detail, the architecture and function of the largest compartment of the eukaryotic cell, the nucleus and its components, such as the nuclear envelopeNuclear envelope, nuclear pore complexNuclear pore complex (NPC), chromosomesChromosomes and nucleoskeleton, are emphasized. Centrioles play an important role in cell divisionCell division and are being discussed, including their replication. It is discussed how forces are transmitted from the cellular microenvironmentMicroenvironment, which are sensed by cell–matrix adhesionCell-matrix adhesion receptors, coupled via focal adhesion proteins to the cell’s cytoskeletal network, and subsequently they have an impact on the mechanical phenotype of the innermost organelle of the cell, the cell nucleusCell nucleus. The biophysical aspects of these main functions of cells under normal physiological conditions and under pathological conditions such as cancer and acute or chronic inflammationChronic inflammation are at the center of attention. Finally, the effect of mechanical force on gene expressionGene expression is addressed, how the cells deal with it.

Global Positioning System: Understanding Long Noncoding RNAs through Subcellular Localization

Article

Mar 2019
MOL CELL

The localization of long noncoding RNAs (lncRNAs) within the cell is the primary determinant of their molecular functions. LncRNAs are often thought of as chromatin-restricted regulators of gene transcription and chromatin structure. However, a rich population of cytoplasmic lncRNAs has come to light, with diverse roles including translational regulation, signaling, and respiration. RNA maps of increasing resolution and scope are revealing a subcellular world of highly specific localization patterns and hint at sequence-based address codes specifying lncRNA fates. We propose a new framework for analyzing sequencing-based data, which suggests that numbers of cytoplasmic lncRNA molecules rival those in the nucleus. New techniques promise to create high-resolution, transcriptome-wide maps associated with all organelles of the mammalian cell. Given its intimate link to molecular roles, subcellular localization provides a means of unlocking the mystery of lncRNA functions.

Identification of lncRNA competitively regulated subpathways in myocardial infarction

Article

Full-text available

Feb 2019

The functions of long non-coding RNAs (lncRNAs) in myocardial infarction (MI) remain largely unknown. Thus, we used the subp athway-LINCE method to characterize the potential roles of lncRNAs in MI. Candidate lncRNA-mRNA interactions were obtained from miRNA-mRNA interactions and lncRNA-miRNA interactions. Then the lncRNA and mRNA co-expression relationship pairs (LncGenePairs) were screened from the lncRNAs and mRNA intersections, which were extracted through candidate lncRNA-mRNA interactions and sample gene expression profiles. The lncRNAs in LncGenePairs were embedded into pathway graphs as nodes through linking to their regulated mRNAs, which resulted in obtaining condition-specific lncRNA competitively regulated signal pathways (csLncRPs). Finally, the csLncRPs were calculated using lenient distance similarity to obtain the lncRNA competitively regulated subpathways. Based on the statistical significance of signal subpathways, lncRNA-mRNA networks were constructed, in which hub lncRNAs were selected. A total of 65 lncRNAs competitively regulated subpathways and 13 hub lncRNAs were obtained, which associated with a risk of MI. Identifying lncRNAs competitively regulated subpathways not only provides potential lncRNA biomarkers for MI, but also helps the understanding of pathogenesis of MI.

Yeast Nucleoporin Mutants Are Defective in pre-tRNA Splicing

Article

Full-text available

Jan 1996

We have screened nucleoporin mutants for the inhibition of tRNA splicing, which has previously been proposed to be coupled to transport. Strains mutant for Nup49p or Nup116p, or genetically depleted of Nup145p, strongly accumulated unspliced pre-tRNAs. Splicing was inhibited for all 10 families of intron-containing pre-tRNA, but no effects on 5' or 3' end processing were detected. Strains mutant for Nup133p or Nsp1p accumulated lower levels of several unspliced pre-tRNAs. In contrast, no accumulation of any pre-tRNA was observed in strains mutant for Nup1p, Nup85p, or Nup100p. Other RNA processing reactions tested, pre-rRNA processing, pre-mRNA splicing, and small nucleolar and small nuclear RNA synthesis, were not clearly affected for any nucleoporin mutant. These data provide evidence for a coupling between pre-tRNA splicing and nuclear-cytoplasmic transport. Mutation of NUP49, NUP116, or NUP145 has previously been shown to lead to nuclear poly(A)+ RNA accumulation, indicating that these nucleoporins play roles in the transport of more than one class of RNA.

UsnRNP Assembly and Transport

Chapter

Jan 1988

Iain W. Mattaj

The purpose of this summary is to present extensively the conclusions of this chapter, and to point out some areas where further research is required. References to the cited work are to be found in the text of the review. Assembly of UsnRNPs and Transport into the Nucleus. The study of UsnRNP assembly began with the definition of protein binding domains within the UsnRNAs. Since the majority of UsnRNP proteins are found in all the major UsnRNPs immunoprecipitable with anti-Sm antisera, the first attempts to define binding sites simply looked for structural domains conserved among the major UsnRNAs. By a combination of nuclease protection experiments and the analysis of the protein binding properties of mutant UsnRNAs one of these domains, a single-stranded region of consensus sequence PuA(U)nGPu, has been shown to be the site through which the common UsnRNP proteins bind. Since these proteins include those recognized by anti-Sm antisera, this region is now generally called the Sm binding site. In the case of U2 snRNA, this domain has been shown to be necessary for nuclear accumulation and for the modification of the co- transcriptionally added monomethyl guanosine cap structure to the UsnRNA- characteristic trimethyl gaunosine cap. This suggests that the Sm binding site, where it is present, has an important role in allowing an RNA to acquire the characteristics of a UsnRNA. This naturally begs the question of how RNAs, such as U3 snRNA, which do not detectably associate with Sm antigens, acquire their trimethyl cap structure and segregate into the nucleus. The importance of the Sm binding site is also supported by the observation that it has been extremely conserved in evolution, snRNAs from humans to fungi are capable of binding to Xenopus laevis Sm antigens. Indeed it has further been demonstrated that fungal RNAs with Sm binding sites are associated in fungi with proteins recognized by human anti-Sm antisera. Thus, at least this epitope has also been conserved on some snRNP proteins throughout eukaryotic evolution. Much less is known about the RNA binding sites of proteins unique to particular snRNPs. In the case of U2 hairpins towards the 3’ end of the molecule are required, in addition to the Sm binding site, for the interaction of the U2 specific A’ and B” proteins with the RNA. The order of assembly which this suggests, first interaction between the RNA and common UsnRNP proteins followed by addition of the unique proteins, is consistent with the in vivo analysis of UsnRNP assembly. These in vivo experiments further suggest that the core proteins D, E, F and G assemble, in the initial step of UsnRNP biogenesis, into a complex, in the absence of RNA. It is worth noting that one of the gaps in our current knowledge concerns the homogeneity of UsnRNPs. The molar ratios of different UsnRNP components is not known, and it is not clear whether all U2 snRNPs, for example, contain one molecule of all the common proteins plus A’ and B″ or whether there is a mixed population, some of which contain only A’ and others only B”. In order that we understand the role of the proteins in the function of snRNPs it is important that an answer is found to this question. A current model of the U2 snRNP is given in Fig. 1. From this model it is evident that we are ignorant of which snRNP proteins interact directly with UsnRNAs and where exactly they bind. A more direct analysis of UsnRNP structure would be extremely useful in this respect. The location of UsnRNP assembly is thought to be the cytoplasm. Evidence for this has come from many different experimental approaches. The fact that UsnRNAs injected into enucleated Xenopus oocytes assemble into anti-Sm immunoprecipitable snRNPs is the most direct demonstration that cytoplasmic assembly can occur. Oocytes are, however, unusual in that they store large amounts of UsnRNP proteins for use in the early stages of embryonic development and therefore the results which suggest that the location of assembly is the same in mammalian culture cells provide important confirmatory evidence. These results are mainly based on the observation that UsnRNAs, at least Ul, U2, and U4, are synthesized as precursors in HeLa cells. These precursors differ from the mature RNAs by having short 3’ extensions. They are short-lived, and are processed to the mature length within an hour of synthesis. The precursors are found mainly in the cytoplasm, as determined by many different cell fractionation methods. In this cytoplasmic phase immunoprecipitation studies have shown that the precursors are already associated with both common and at least some unique snRNP proteins. The short 3’ extensions are apparently removed from the UsnRNA precursors either prior to or during re-entry into the nucleus. This RNA processing event is not the only to occur in the cytoplasm. The conversion of the monomethyl guanosine cap structure to the unique trimethyl cap structure has also been shown to occur in the cytoplasm of Xenopus oocytes. It is currently not known where the other abundant internal modifications which occur on UsnRNAs take place. A final piece of evidence supporting the fact that UsnRNP assembly occurs in the cytoplasm is the observation, again made with Xenopus oocytes, that the common UsnRNP proteins in the free state are cytoplasmic. Migration of these proteins to the nucleus requires their interaction with RNA. This interaction is specific, and depends on the presence on the RNA of an Sm binding site in addition to some other, as yet undefined, structural characteristics. There are two gaps in our knowledge of UsnRNP assembly. The first concerns the details of the order of assembly of the B/B’ proteins and the unique snRNP proteins. The second is that we do not know where in the cell the unique proteins are found in the free state, nor where they assemble onto UsnRNPs, although there is evidence that at least some of these proteins join the snRNPs in the cytoplasm. As an example of assembly the order of events involved in the biogenesis of the U2 snRNP are diagramed in Fig. 2. Intranuclear Location of UsnRNPs. Making use of both light and electron microscopy, immunohistological studies have led to the elucidation of UsnRNP location in the nucleus. UsnRNPs are distributed throught the nucleoplasm, i. e., are not associated with chromatin or the nuclear lamina, in a pattern which reveals local concentrations. These aggregates of staining material are seen both at the light and the electron microscopic level, although the correlation between the two types of analysis is not yet perfect. UsnRNPs are also not associated with chromosomes during mitosis. They simply disperse throughout the cell and re-enter the daughter nuclei during telophase. Detailed analysis with the electron microscope has revealed that UsnRNPs are found in association with nascent hn RNP fibrils, indicating an early association between UsnRNPs and pre-mRNA. It is not clear whether this indicates a very early start to RNA processing events following transcription, or whether it is simply a reflection of an early stage in the assembly of splicing complexes. The major conclusion from these electron microscopic studies is however that the bulk of UsnRNPs are not associated with short-lived hnRNP. This enables me to finish this summary with a question for the future. What are the bulk of the UsnRNPs in the nucleus doing? Are they simply waiting to be used in RNA processing, or do they perform other functions? A much more detailed knowledge of nuclear structure and function will be required before this can be answered satisfactorily.

Small Nuclear RNAs: RNA Sequences, Structure, and Modifications

Chapter

Jan 1988

All eukaryotic cells contain multiple small nuclear RNAs, designated U-sn At present ten U-snRNAs have been identified; they account for about 1% total mammalian cellular RNA. Of these, the U1 to U8 snRNAs have been sively characterized. All the U-snRNAs contain a 5′-cap structure consis a blocked 5′-terminal pyrophosphate linkage; except for U6 RNA, a tri guanosine-containing “cap” is the 5′-end. These U-snRNAs are confined nucleus and are metabolieally stable. They are synthesized by RNA polyII. U6 snRNA, which has a non-nucleotide cap structure, is synthesized by RNA polymerase III. All the U-snRNAs are present in the cells as small ribonucleopro- tein particles (snRNPs), that are metabolically stable.

The Nuclear Pore Complex

Article

Jan 1995

L. I. Davis

Splicing of precursors to messenger RNAs by the spliceosome

Article

Jan 1992

Biosynthesis and function of modified nucleosides in tRNA

Article

Jan 1995

GR Björk

Recherches biochimiques sur l'oogene`se. I. Synthe`se et accumulation du RNA pendant l'oogene`se du crapaud sud-africain Xenopus laevis

Article

Feb 1971

RNA synthesis was studied during oogenesis of the South-African clawed toadXenopus laevis. Seven classes of oocytes were isolated from immature and from mature females. The RNA extracted from these oocytes was analyzed by filtration on Sephadex G-100 and by centrifugation in sucrose gradients. The previtellogenic oocytes (average diameter : 50 to 150 μ) were found to contain very large amounts of D-RNA and of low molecular weight RNA. The latter comprises two classes of molecules whose properties are those of 5 s RNA and of transfer RNA. In the oocytes of 150 μ, D-RNA, 5 s RNA and transfer RNA formed 10%, 40% and 35%, respectively, of the total optical density. The vitellogenic oocytes (average diameter : 375-1100 μ) synthesized mostly D-RNA and 28 s + 18 s RNA. The latter classes of RNA accumulated very rapidly in large oocytes since they made up more than 90% of the RNA present in mature eggs. D-RNA synthesized by the large oocytes sedimented in sucrose gradients as two distinct peaks (40 s and 24 s). The synthesis of the different RNA classes is therefore not at all coordinate during oogenesis. Transfer RNA and 5 s RNA are produced in excess in small oocytes. These classes of RNA are probably not utilized immediately, but stored for further use. RNA synthesis was found to be very different in large oocytes (average diameter: 375-1100 μ) taken from females which have laid eggs since 1 week and in oocytes from females which have not laid eggs since more than 2 months. Differences were noticed in the labeling rate of RNA, and in the sedimentation profile of newly synthesized RNA.

Biochemistry

Article

Oct 1990
TRENDS BIOCHEM SCI

Roger L.P. Adams

Capping Enzyme in Eukaryotic mRNA Synthesis 1 1 Abbreviations: VTF, vaccinia virus termination factor; EpG, enzyme-GMP covalent complex; 5FOA. 5-Fluoroorotic acid

Article

Feb 1995
PROG MOL BIOL TRANSL

Stewart Shuman

Publisher Summary This chapter discusses the enzymatic mechanism of the individual capping reactions and the organization of the functional domains, within the relevant enzymes. Greatest attention is devoted to two model systems, Vaccinia virus and Saccharomyces cerevisim, in which biochemistry, molecular genetics, and protein engineering have fueled considerable progress, since the past review of ribonucleic acid (RNA) capping in this series. It also discusses the genetic studies in yeast that shed light on the physiologic role of the cap in eukaryotic RNA metabolism. Capping may serve to protect the messenger RNA (mRNA) from nucleolytic degradation. Elucidation of the mechanism and potential regulation of cap formation is, thus, pertinent to the understanding of eukaryotic gene expression. Capping occurs, by a series of three enzymatic reactions, in which the 5’ triphosphate terminus of a primary transcript is first cleaved to a diphosphateterminated RNA by RNA triphosphatase, then capped with GMP by RNA guanylyltransferase, and methylated at the N7 position of guanine by RNA (guanine-7) methyltransferase.

mRNA nuclear export

Article

May 1994
CURR OPIN GENET DEV

The export of mRNA from the nucleus to the cytoplasm is an essential step in the expression of genetic information in eukaryotes. It is an energy-dependent process and involves transport across the nuclear pores. It requires both cis-acting ribonucleoprotein particle signals and specific trans-acting factors. Although much remains to be learned, recent information has begun to define this pathway at both the cellular and biochemical levels and indicates that it is used as a key regulatory step by several viruses.

RNA transport

Abstract and Figures

Recommended publications

Global implications of mRNA localization pathways in cellular organization

The trials and travels of tRNA

RNA Processing in a Unicellular Microorganism: Implications for Eukaryotic Cells

Nucleocytoplasmic release of repetitive DNA transcripts in carcinogenesis correlates with a 60 kilod...