Ran guanine triphosphatase (GTPase) is a central molecule required for several cellular functions including nucleo- cytoplasmic transport, cell-cycle progression, and nuclear envelope assembly (reviewed by Melchior, 1998; Nakielny, 1999; Sazer, 2000). Ran appears to fulfill at least two distinct functions in the nuclear import process: (1) the energy required to effect nucleo-cytoplasmic transport is provided in the form of GTP hydrolysis by Ran (Melchior, 1993; Moore, 1993) and (2) Ran regulates the interaction between proteins that transport substrates across the nuclear membrane (Rexach, 1995; Gorlich, 1996).

The Ran pathway has been shown to have a role in spindle assembly. However, the extent of the role of the Ran pathway in mitosis in vivo is unclear. Perturbation of the Ran pathway disrupts multiple steps of mitosis in syncytial Drosophila embryos and new mitotic processes have been uncovered that are regulated by Ran. During the onset of mitosis, the Ran pathway is required for the production, organization, and targeting of centrosomally nucleated microtubules to chromosomes. However, the role of Ran is not restricted to microtubule organization, because Ran is also required for the alignment of chromosomes at the metaphase plate. In addition, the Ran pathway is required for postmetaphase events, including chromosome segregation and the assembly of the microtubule midbody. The Ran pathway mediates these mitotic events, in part, by facilitating the correct targeting of the kinase Aurora A and the kinesins KLP61F and KLP3A to spindles (Silverman-Gavrila, 2006).

During the cell cycle, the GTPase Ran regulates multiple cellular functions, including nucleocytoplasmic transport, nuclear envelope formation, and spindle assembly (Hetzer, 2002). The nucleotide-bound state of Ran (GTP or GDP) is spatially regulated by RCC1, the chromatin-bound guanine nucleotide exchange factor for Ran, and by RanGAP, which localizes to the cytoplasm and stimulates the intrinsic GTPase activity of Ran. The localization of these proteins is essential in regulating the activity of Ran throughout the cell cycle (Silverman-Gavrila, 2006 and references therein).

In vitro, RanGTP induces spindle assembly in mitotic Xenopus egg extracts in the absence of centrosomes, kinetochores, and chromatin (Carazo-Salas, 1999; Kalab, 1999; Ohba, 1999; Wilde, 1999) by altering microtubule (MT) dynamics (Carazo-Salas, 2001; Wilde, 2001), changing the balance of motor activity (Wilde, 2001), and increasing centrosomal MT nucleation (Carazo-Salas, 2001). The Ran pathway can regulate spindle assembly by RanGTP binding to nuclear transport receptors (NTRs) and preventing them from interacting with and inhibiting spindle assembly factors (SAFs) that posses a nuclear localization signal (NLS) (Trieselmann, 2003; Tsai, 2003; Ems-McClung, 2004). For this process to drive spindle assembly, RanGTP must be generated at chromosomes where it maintains SAFs in an active form. Indeed, many in vivo and in vitro studies infer that RanGTP exists around mitotic chromatin and spindles, whereas NTRs have a much broader distribution resulting in the creation of an environment around chromosomes where NLS-containing SAFs are active (Kalab, 2002; Trieselmann, 2002; Li, 2004; Caudron, 2005; Silverman-Gavrila, 2006 and references therein).

To define the mitotic role of the Ran pathway in vivo, previous studies injected anti-RanBP1 antibodies or NTRs into mammalian cells (Guarguaglini, 2000; Nachury, 2001). However, these studies did not define the dynamic nature of the observed defects and may have missed the complete series of events that led to the observed phenotype at the time of fixation. In an alternative approach, RNA interference (RNAi) was used in Caenorhabditis elegans to reduce expression levels of different Ran pathway proteins (Askjaer, 2002; Bamba, 2002). These studies defined a role for the Ran pathway in the production of spindle MTs. However, because all spindle MT assembly was inhibited, these studies were unable to define any other mitotic processes regulated by Ran. Furthermore, because spindle assembly was not examined until at least 16-48 h after RNAi application, it is possible that the observed defects could result from the disruption of other cellular processes. Indeed, Bamba (2002) showed that RNAi of Ran pathway components did inhibit nuclear transport. In addition, embryos and pronuclei were smaller after RNAi treatment for Ran pathway components (Askjaer, 2002; Silverman-Gavrila, 2006 and references therein).

To overcome these limitations, a microinjection approach was used to disrupt the Ran pathway just before mitosis and then immediately analyze the consequences of the perturbation on mitotic cellular function by time-lapse microscopy. By manipulating the Ran pathway in multiple ways and by microinjecting different concentrations of perturbants, it was possible to 'turn down' the pathway rather than completely or severely block it. This approach revealed multiple mitotic phenotypes that were not observed when all spindle MT production was inhibited (Askjaer, 2002; Bamba, 2002). By comparing the effects with known phenotypes from different mutants and RNAi screens, an integrative view has been generated of how Ran regulates mitosis in syncytial Drosophila embryos. Indeed, the Ran pathway was found to regulate both pre- and postmetaphase events (Silverman-Gavrila, 2006).

The data suggest that centrosomal MT production is regulated by NTR-sensitive SAFs. Severe disruption of the Ran pathway by injection of high concentrations of RanT24N inhibits MT production at centrosomes. However, less severe disruption of Ran pathway leads to MT production around chromosomes. These MTs begin to organize into a spindle in a manner reminiscent of acentrosomal spindle assembly in higher plants and female meiosis. In these conditions, it is possible that a small amount of RCC1 remains active and capable of producing a small amount of RanGTP. This reduced level of RanGTP may be sufficient to maintain active SAFs close to the chromatin and not around centrosomes some distance from the chromosomes. The MTs fail to resolve into stable spindles, possibly because of the rapid onset of the next interphase (Silverman-Gavrila, 2006).

MTs nucleated from centrosomes when the Ran pathway was only partially inhibited (upon injection of lower concentrations of RanT24N, RanGAP, and NTRs) were short and failed to grow toward the chromosomes. These data suggest that the Ran pathway is required for the stabilization of centrosomal nucleated MTs in vivo, consistent with recent in vitro findings (Caudron, 2005). Thus, embryonic spindle assembly might be driven by both a dominant centrosomal-mediated pathway required for the rapid assembly of spindles and a backup chromosomal pathway as in mammalian cells. The molecular targets of this pathway are unclear because TPX2 and XMAP310, which are required in vertebrates, have no known orthologues in Drosophila (Silverman-Gavrila, 2006 and references therein).

Inhibition of the Ran pathway often causes centrosomally nucleated MTs to randomly probe their environment rather than grow rapidly toward chromosomes. Previous studies have implied that RanGTP complexed with importin ß persists around the spindle throughout mitosis (Kalab, 2002; Li, 2004; Caudron, 2005). Therefore, RanGTP generated at chromosomes could form a spatial cue to guide MTs toward chromosomes. The data support this hypothesis and suggest that MT targeting to chromosomes may not proceed by a random search and capture mechanism. Instead, a 'directed' process operates where chromosomally generated RanGTP creates a molecular environment around chromosomes that facilitates MT growth preferentially toward the chromosomes. Such a scenario has recently been predicted in a mathematical model (Silverman-Gavrila, 2006 and references therein).

This study showed that the Ran pathway has a role in spindle pole/centrosome regulation in vivo as perturbation of the Ran pathway caused defects in spindle pole and centrosome organization. When the Ran pathway is disrupted centrosomes are released, suggesting that the Ran pathway regulates factors required for centrosome attachment to spindle poles. Perturbation of the Ran pathway also leads to the formation of multipolar spindles, sometimes by the splitting of existing poles. The findings are consistent with previous studies in which overexpression of RanBP1 (Di Fiore, 2003) led to centrosome abnormalities (Silverman-Gavrila, 2006).

The mechanism by which Ran regulates the spindle pole/centrosome remains unclear. However, many of the effects have been described upon inhibition of dynein and Aurora A. Aurora A has been shown to be regulated by the Ran pathway in vitro requiring RanGTP for its targeting and activation (Trieselmann, 2003; Tsai, 2003). Whether dynein or a regulator of dynein function is regulated by Ran remains to be tested. However, Ran itself may directly affect centrosomes because it has been localized to centrosomes through an interaction with AKAP450 (Keryer, 2003, Di Fiore, 2004). Interestingly, multipolar spindles form upon injection of RCC1, suggesting that elevated levels of RanGTP also disrupt this process (Silverman-Gavrila, 2006).

In addition, the failure of KLP61F to be recruited to centrosomes correlates with shorter spindles with broad poles. Previous studies that directly inhibit KLP61F have not pointed to a role in these processes. However, because the current experiments inhibit a broad range of processes that are downstream of Ran, it suggests that either a subset of KLP61F regulators, or factors working in coordination with KLP61F, may have a combined role in pole organization (Silverman-Gavrila, 2006).

The Ran pathway is required for the organization of spindle MTs. Strikingly, the disorganization did not result in a single spindle phenotype, suggesting that in Drosophila embryos, Ran regulates multiple factors involved in spindle organization. Many defects were consistent with a disruption of the balance of forces generated by antagonistic motors that are crucial for spindle assembly. Indeed, KLP61F and KLP3A, two motors that operate in balance with other motors, require the Ran pathway for their correct localization to the spindle. The spindle phenotypes observed do not correspond to just the inhibition of KLP61F. This could be due to the motor that operates in balance with KLP61F, Ncd, also being regulated by the Ran pathway. Indeed, the Xenopus homologue of Ncd, XCTK2 has been shown in vitro to be regulated by the Ran pathway (Ems-McClung, 2004). Inhibition of both motors in Drosophila embryos or of homologues of both motors in mammalian system largely cancels out the individual phenotypes. The phenotypes that are seen upon inhibition of these two motors in Drosophila embryos were seen upon perturbing the Ran pathway. The disruption of a balance of forces is unlikely to occur uniformly throughout an embryo due to the gradual diffusion of the injected material and would therefore result in a variety of spindle organization defects (Silverman-Gavrila, 2006).

Injection of RanT24N caused the metaphase plate to be three times wider (when spindle assembly proceeded) than in controls. A widening of the metaphase plate could be due to defects in MT attachment to kinetochores or inhibition of chromokinesins, proteins that generate the 'polar ejection force' that move chromosomes to the metaphase plate. Recently, the Ran pathway was shown to affect MT attachment to kinetochores (Arnaoutov, 2005). This study shows that the Ran pathway can regulate chromosome congression by regulating the MT attachment of KLP3A (a Kinesin-4 family member) reminiscent of regulation by Ran of the chromokinesin Kid (a Kinesin-10 family member) (Trieselmann, 2003). These findings suggest that chromokinesins from different kinesin families may share a common regulatory mechanism. The role of Ran in chromosome congression may not be restricted to the direct regulation of MT kinetochore attachment and chromokinesins because Aurora A, which is also regulated by Ran, is required for chromosome alignment at the metaphase plate in vertebrates (Silverman-Gavrila, 2006).

Anaphase A chromosome movement to the poles was slower in nuclei that otherwise progressed normally through mitosis upon Ran pathway perturbation. The reduced chromosome to pole velocity is identical to chromosome velocity in KLP59C-inhibited embryos. KLP59C, a kinetochore-localized MT depolymerase of the Kinesin-13 family is not an obvious candidate to be regulated by Ran, because it is cytosolic. Therefore, either a regulator of KLP59C or a protein that works in coordination with KLP59C, such as dynein, could be a target of the Ran pathway. Indeed, dynein inhibition causes similar reductions in chromosome movement to the poles. Another potential candidate is KLP3A, which is involved in chromosome segregation and whose targeting is dependent upon the Ran pathway (Silverman-Gavrila, 2006).

Nuclear divisions were also disrupted upon RCC1 injection, resulting in chromosome decondensation without chromosome disjunction. This could be caused by a chromosome decondesation defect or an override of the spindle checkpoint. The latter would be consistent with a study in Xenopus egg extracts, where increased levels of RCC1 circumvent the spindle checkpoint (Arnaoutov, 2003) by disrupting the localization of checkpoint regulators (Silverman-Gavrila, 2006).

Disruption of the Ran pathway prevents the correct assembly of the midbody in morphologically normal spindles. A failure to bundle midbody MTs was the most common defect. Bundling of midbody MTs requires the action of multiple kinesins, including Ncd, KLP3A, KLP61F, and Pavarotti. Indeed, KLP61F fails to localize to midbodies in embryos where the Ran pathway is perturbed. In extreme cases, no midbody MTs formed, suggesting that the Ran pathway, either through MT stabilizing factors or through regulating MT bundling, is required to for midbody stability (Silverman-Gavrila, 2006).

In vitro biochemical assays have shown that Aurora A activation is a downstream event of the mitotic Ran pathway (Trieselmann, 2003; Tsai, 2003). Injection of both RanT24N and DeltaN importin ß cause similar defects in Aurora A localization; suggesting that in vivo the Ran pathway via importins regulates Aurora A by facilitating its targeting to spindle MTs. Therefore, regulating the targeting and subsequent activation of Aurora A could be a major in vivo role for the Ran pathway in coordinating spindle assembly, especially because Aurora A has been implicated in both pre- and post-metaphase mitotic processes (Silverman-Gavrila, 2006).

Together, these data suggest that the Ran pathway has to remain active postmetaphase for chromosomes to successfully segregate. Indeed, proteins expected to increase RanGTP levels (RCC1) and those expected to decrease RanGTP levels (RanT24N and RanGAP) affect mitosis, suggesting that a balance of RanGTP/GDP may be crucial for mitotic progression. The data also suggests that a major mitotic role for Ran in Drosophila embryos is to prevent NTRs inhibiting SAFs during mitosis. Furthermore, the mitotic role of the Ran pathway is not restricted to MT-dependent events; Skeletor organization, which is independent of MTs, is dependent upon Ran. Therefore, the Ran pathway has the potential to regulate mitosis to a far greater extent than previously thought (Silverman-Gavrila, 2006).

GENE STRUCTURE

cDNA clone length - 1639

Bases in 5' UTR - 201

Exons - 3

Bases in 3' UTR - 787

PROTEIN STRUCTURE

Amino Acids - 216

Structural Domains

The Drosophila ran gene was identified in a two-part differential cDNA screen designed to identify CNS neural precursor genes. Whole-mount embryo in situ hybridizations of selected cDNAs were carried out to identify genes that are dynamically expressed during CNS lineage development. Sequence analysis of one of the cDNAs isolated in the above screen revealed that its corresponding poly-adenylated mRNA encodes Ran (Genbank nos. AAF30287, AAF60290 and CG1404). Genomic and cDNA sequence alignment revealed that the Drosophila ran gene contains a single 78 bp intron that separates the ran ORF between codons 145 and 146. Salivary gland polytene chromosome in situ hybridization revealed that ran is located in the 10A7-9 cytological region of the X chromosome (Koizumi, 2001).

The Drosophila ran gene encodes a 216-aa protein that shares an overall sequence identity of 83% with the mammalian (mouse and human) ran and 79% to Sacchromyces cerevisiae GSP2. Compared with vertebrate and yeast proteins, the Drosophila Ran is highly conserved in all but the first 10 N-terminal residues and in amino acids spanning residues 188-203, which are likewise poorly conserved in other Ran proteins. Also conserved are the GTP-binding/hydrolysis domains (Scheffzek, 1995; Koizumi, 2001 and references therein).

A search of the Drosophila genomic database has revealed a second Drosophila ran gene (CG7815), located on the third chromosome (71C3-4), termed ran-like. Ran-like is equally related to the Drosophila and human Ran proteins, sharing 59% identity and 80% similarity. At the nucleotide level, ran-like shows a similar level of homology to its paralog, however, the nucleotide homology is not shared with ran genes of other species. Although the predicted Ran-like primary structure is homologous to all identified Ran proteins it differs from other Rans by 30 aa residues that are scattered throughout the protein. These 30 residues are identical among other plant, yeast, worm, fly and vertebrate Rans. However, 18 of the 30 residues represent conservative substitutions. The greatest divergence between ran-like and other ran proteins is found in three regions: (1) the N-terminal 10 amino acids, which are also divergent in other Ran proteins; (2) the region between residues 127 and 149, coincident with the alpha4 helix, which is poorly conserved between Ras and ran (Scheffzek, 1995), and (3) the C-terminal domain that spans 30 residues. Phylogenetic analysis reveals that the Drosophila ran is closely related to other known metazoan Ran proteins, but ran-like is more divergent (Koizumi, 2001). The Ran-like isoform might be confined to invertebrates, because no homologous counterpart has been identified in other organisms. Given the divergent amino acid sequence of ran-like and the conserved sequence of Ran, it is likely that the evolutionary rate of ran-like change is greater than that of ran . Further analyses of ran and ran-like mutations are required to assess the degree of functional overlap between their encoded isoforms. Currently, no mutant alleles have been reported for either gene and available chromosomal deficiencies that cover ran and ran-like also include many other genes, both known and predicted (Koizumi, 2001).

date revised: 30 October 2006

Home page: The Interactive Fly © 2006 Thomas Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.