AccessMyLibrary provides FREE access to millions of articles from top publications available through your library.
KEY WORDS: nuclear transport; Agrobacterium tumefaciens T-DNA, snRNA, and hnRNA; plasmodesmata; cell-to-cell movement of plant viruses
INTRODUCTION 168 NUCLEAR TRANSPORT 168 Nuclear Import of Agrobacterium T-DNA by the Host-Plant Cells 170 Nuclear Import of Viral Genomes 174 Nuclear Traffic of snRNA and mRNA 176 CELL-TO-CELL MOVEMENT OF PLANT VIRUSES 179 Plasmodesmata 180 TMV-Type Cell-to-Cell Movement 181 CPMV-Type Cell-to-Cell Movement 187 ARE THERE GENERAL RULES FOR NUCLEIC ACID TRANSPORT? 188 FUTURE PERSPECTIVES 191
Transport of nucleic acids through cell membranes is an essential biological process that occurs in all living organisms. This review focuses on two plant systems in which nucleic acid molecules are transported through membrane channels: transport of Agrobacterium T-DNA through nuclear pores and movement of plant viruses through intercellular connections. To provide a broader perspective, nuclear uptake of animal viruses and nuclear import/export of small nuclear (sn) RNA and messenger (m) RNA are described. By comparing the examined cases of nucleic acid transport, the review proposes a set of general rules for this type of transport through membrane channels.
Multicellular organisms as well as individual cells are highly compartmentalized. To maintain normal cellular functions, these compartments must constantly communicate. Intra- and intercellular communication occurs in a wide variety of biological processes ranging from the directional spread of electric impulses along neuronal axons in animals to transport of macromolecules through the intercellular connections between companion and sieve cells in plant vascular tissue. Although transport of electrolytes, hormones, and proteins is well studied, movement of nucleic acids across biological membranes is only now being addressed.
Transport of nucleic acids through cell membranes is a biological process basic to all living organisms. Nucleic acid molecules are transported through membrane channels during host-pathogen interactions (e.g. transport of viral genomes into the host cell) as well as during normal cellular processes [e.g. nuclear export/import of messenger (m) RNA]. Furthermore, transfer of genetic material has been shown to occur between evolutionarily distant organisms (9)(58)(109). A few common mechanisms potentially underlie these different events.
In this review, we focus mainly on two plant systems in which nucleic acid molecules are transported through membrane channels: (a) transport of Agrobacterium T-DNA through nuclear pores and (b) movement of plant viruses through intercellular connections. To provide a broader context, nuclear uptake of animal viruses and nuclear import/export of small nuclear (sn) RNA and heterogeneous nuclear (hn) RNA is also described. Then, the common structural and biochemical requirements of these transport processes are discussed.
Molecular transport across the nuclear envelope involves many different proteins and nucleic acids. This transport is bidirectional and occurs exclusively through the nuclear pore complex (NPC), a 125,000-kDa structure composed of a central transporter mounted within spoke and ring protein assemblies that are integrated into the two membranes of the nuclear envelope (1) (Figure 1a). In the passive state, the NPC allows diffusion of small molecules (up to 40 kDa) (reviewed in 1, 52, 89). Transport of larger molecules occurs by an active mechanism mediated by specific nuclear localization signal (NLS) sequences contained in the transported molecule (reviewed in 46). The size limit for diffusion through the NPC was determined using microinjected foreign molecules such as dextrans (96). However, transport of small endogenous nuclear proteins such as H1 (21 kDa) also occurs by an active process (7). Recently, a model for active nuclear transport was suggested in which two symmetrical rings of protein subunits of the NPC transporter form a double iris similar to the airlock of a spacecraft (1)(2) (36); both of the irises must open sequentially to allow passage of a transported molecule. In the open conformation, the maximal size exclusion limit of the NPC is 23 nm in diameter (39).
As mentioned above, nuclear targeting of the transported molecule requires a specific NLS. With few exceptions (e.g. influenza virus nucleoprotein NLS, yeast GAL4 protein NLS), all NLSs can be classified into two groups: the SV40 large T antigen NLS (PKKKRKV) paradigm, and the bipartite motif consisting of two basic regions separated by a variable number (not less then 4) of spacer amino acids exemplified by the nucleoplasmin NLS (KR-[X.sub.10]-KKKL) (reviewed in 37).
Nuclear transport of karyophilic molecules occurs in discrete steps. Recent data suggest that nuclear import initiates in the cytoplasm with binding of NLSs to the first cytoplasmic receptor, hsp70 (or its cognate hsc70), which may present the NLS in a locally unfolded form to the second type of cytoplasmic receptors, the NLS binding proteins (NBPs) (reviewed in 36). The NBPs then direct the transported molecule to the NPC where actual translocation across the nuclear envelope occurs (reviewed in 89).
Nuclear Import of agrobacterium T-DNA by the Host-Plant Cells
The interaction of Agrobacterium spp. with plant cells is the only known natural example of interkingdom DNA transfer. In nature this process results in crown gall tumors, an agronomically important disease that affects most dicotyledonous plants. Most functions for Agrobacterium--plant cell DNA transfer are carried on a large (200 kb) Ti (tumor inducing) plasmid contained in the bacterial cell. The Ti plasmid has two important genetic components. One, the T-DNA, is copied and transferred to the plant cell. The T-DNA is delimited by two 25-bp direct repeats at its ends, the T-DNA borders. Any DNA between these borders is transported to the plant cell. Although the T-DNA is the mobile element, it does not itself encode the products that mediate its movement. Instead, a second component of the Ti plasmid, the virulence (vir) region, provides most of the trans-acting products for T-DNA transfer. Following induction of vir gene expression by small phenolic signal molecules excreted from wounded susceptible plant cells, a transferable copy of the T-DNA is generated. This molecule, designated the T-strand, is a single-stranded copy of the bottom strand of the T-DNA region (reviewed in 24, 126, 127).
Evidence to date suggests that the T-strand directly associates with two different protein products of the vir region. During T-strand synthesis, the VirD2 protein tightly (probably covalently) attaches to the 5[feet] end of the T-strand molecule (60)(66)(118)(125), whereas VirE2, a single-stranded (ss) DNA binding protein (SSB), is proposed to coat the T-strand along its entire length (15)(16)(28)(49). Binding of VirE2 to ssDNA is cooperative (21)(103) and results in formation of long unfolded and very thin (less than 2 nm in diameter) protein-ssDNA complexes (21). The T-strand with its associated proteins, VirD2 and VirE2, comprise the Agrobacterium spp. T-DNA transfer complex, designated the T-complex (63)(65).
To genetically transform the host-plant cell, the T-complex must travel from Agrobacterium spp. across three different biological barriers, i.e. the bacterial cell wall and cell membranes, the plant cell wall and cell membrane, and the plant nuclear envelope. Ultimately, the T-DNA is integrated into the plant genome. The relatively simple three-component structure of the T-complex (VirD2 and VirE2 proteins and the T-strand) presents a useful model system to study nuclear import of nucleic acids.
Nuclear transport of the T-complex likely occurs in a polar fashion from its 5[feet] end (reviewed in 126). Potentially, the VirD2 protein bound to the 5[feet] end of the T-strand provides a piloting function. Indeed, several recent studies demonstrate nuclear localization of VirD2 in plant cells (61)(64)(112). The amino acid sequence relevant to VirD2 mediated nuclear transport was localized to a bipartite NLS at the carboxyl terminus of the protein. Deletion of this sequence in the VirD2 protein results in reduction of Agrobacterium spp. tumorigenicity (106). These results together suggest that the VirD2 protein, attached to the 5[feet] end of the T-strand, acts to direct the T-complex to the host-cell nucleus.
However, the T-complex is a very large structure; a 20-kb T-strand (the approximate size of the nopaline-type T-DNA) would contain more than 600 molecules of VirE2 (16). This T-complex has a predicted length of 3.6 [micro]m (21) and a combined molecular mass of about 50 x [10.sup.6] Daltons. Thus, the molecular mass of the T-complex is almost 20 times larger than that of the 60S ribosomal subunit (2.8 x [10.sup.6] Daltons). In addition, because the nuclear pore is approximately 60 nm thick (100), the T-complex is about 60 times longer than the dimensions of the nuclear pore. Can such a large DNA-protein complex be transported through the nuclear pore by a single molecule of VirD2? That deletion of VirD2 NLS decreased but did not completely abolish tumorigenicity suggests that VirD2 is not the sole mediator of the T-complex nuclear uptake. Since VirE2 is a major structural protein component of the T-complex, it may assist in nuclear transport.
In fact, VirE2 was identified recently as a nuclear localizing protein (23). The VirE2 nuclear localization function is mediated by two bipartite NLS sequences designated NSE 1 and 2. As opposed to the typical bipartite NLS of VirD2, the VirE2 homology to bipartite consensus NLS is not perfect. Generally, the first domain of a bipartite NLS has two adjacent basic residues, and the second domain contains at least three out of five basic amino acid residues (37). Both NSE sequences of VirE2, however, have a modified first domain in which the two basic residues are separated by one amino acid; the consensus structure of the second domain, on the other hand, is preserved in the VirE2 NLSs (23).
In addition to functioning in nuclear targeting, VirE2 acts as an SSB (15, 16, 28, 49). Mutational analysis showed that deletion of the VirE2 major ssDNA binding domain (at the carboxyl-terminus of the protein) does not alter nuclear uptake. However, site-specific mutations in NSE1 and NSE2 alter ssDNA binding (23). These latter mutations may cause conformational changes in VirE2 that inactivate its putative ssDNA binding domain(s). Conformational changes caused by deletion of short stretches of amino acid residues have been shown to inactivate single-stranded nucleic acid binding of the P30 protein of tobacco mosaic virus (20). Alternatively, both NSE sequences may reside within another VirE2 region required for ssDNA binding. For example, karyophilic signals of the progesterone and glucocorticoid receptors may overlap their DNA binding domains (53)(57). If VirE2 NLSs and the DNA binding domain(s) indeed overlap, the NLS sequences must be accessible to the nuclear transport machinery when bound to the T-strand.
That VirE2 is involved in nuclear uptake of T-DNA is further strengthened by the observation that plants transgenic for VirE2 complement the virulence of an A. tumefaciens strain with an inactivated virE locus (23). This result indicates that the VirE2 function, essential for tumor formation, is required inside the plant cell. Nuclear localization of the T-complex is one such cellular function. Also, VirE2 expressed in transgenic plants may help to protect the T-strand from cellular nucleases. These in planta functions of VirE2 also help to explain the observation that virE mutants can be complemented to wild-type virulence if coinoculated on plants with A. tumefaciens carrying a wild-type vir region (95). Such complementation by coinoculation is possible only if the virE product functions outside the bacterial cell, for example, inside the host-plant …