Nuclear transporters in a multinucleated organism: functional and localization analyses in Aspergillus nidulans.

Nuclear transporters mediate bidirectional macromolecule traffic through the nuclear pore complex (NPC), thus participating in vital processes of eukaryotic cells. A systematic functional analysis in Aspergillus nidulans permitted the identification of 4 essential nuclear transport pathways of a hypothetical number of 14. The absence of phenotypes for most deletants indicates redundant roles for these nuclear receptors. Subcellular distribution studies of these carriers show three main distributions: nuclear, nucleocytoplasmic, and in association with the nuclear envelope. These locations are not specific to predicted roles as exportins or importins but indicate that bidirectional transport may occur coordinately in all nuclei of a syncytium. Coinciding with mitotic NPC rearrangements, transporters dynamically modified their localizations, suggesting supplementary roles to nucleocytoplasmic transport specifically during mitosis. Loss of transportin-SR and Mex/TAP from the nuclear envelope indicates absence of RNA transport during the partially open mitosis of Aspergillus, whereas nucleolar accumulation of Kap121 and Kap123 homologues suggests a role in nucleolar disassembly. This work provides new insight into the roles of nuclear transporters and opens an avenue for future studies of the molecular mechanisms of transport among nuclei within a common cytoplasm, using A. nidulans as a model organism.

INTRODUCTION

The nucleus is the characteristic organelle of eukaryotic cells. A double membrane, the nuclear envelope (NE), surrounds the nucleus and separates the genetic material from the cytoplasm. The nucleoplasm and cytoplasm communicate through multiprotein nuclear pores inserted at the NE. Nuclear pore complexes (NPCs; Ryan and Wente, 2000) are 60- to 125-MDa structures in vertebrates (Cronshaw ) and 22–44 MDa in Saccharomyces cerevisiae (Rout ).

The NPC allows free diffusion of molecules smaller than 30 kDa (Gorlich and Kutay, 1999). However, macromolecule transport is an energy-requiring process that demands the participation of specific nuclear carriers. The majority of these transporters, also called receptors, belong to the karyopherin-β superfamily and are commonly called karyopherins (KAPs). Fifteen importin-β1–like proteins have been identified in yeast and 22 in mammals (Strom and Weis, 2001; Mosammaparast and Pemberton, 2004). Karyopherins mediate the bidirectional transport of most nuclear proteins between the nucleus and the cytoplasm and the export and import of specific types of RNAs. The repetition of a HEAT repeat within the amino acid chain of KAPs allows the formation of a helical structure that permits the simultaneous interaction with cargoes and constituents of the NPC (Chook and Blobel, 2001; Suntharalingam and Wente, 2003; Madrid and Weis, 2006). Therefore these protein carriers may be considered as transient components of the NPC.

Structural and functional characterization of importin-β1 allowed the generation of a model that has been extrapolated to the rest of the KAPs (Chook and Blobel, 2001). Nuclear transporters are divided into importins—those mediating the transport from the cytoplasm to the nucleus—and exportins, which play the reverse role (Strom and Weis, 2001). Molecular cargoes to be imported into the nucleus may contain, at least, a nuclear localization signal (NLS). Of these, the best characterized are the so-called classic NLS sequences recognized by importin-α and are essentially composed of basic amino acids, Lys and Arg (Makkerh ). Depending on the number of basic residue clusters, classic NLSs are classified as monopartite (Kalderon ) or bipartite (Robbins ). However, importin-α may recognize alternative NLSs (Kosugi ). Cytoplasmic proteins that demand a timely regulation of nuclear exit contain nuclear export signals (NESs; Mattaj and Englmeier, 1998; Macara, 2001; Pemberton and Paschal, 2005). Classic NESs are recognized by exportin 1/Xpo1/CRM1p family transporters and are rich in Leu residues (Fornerod ; Ossareh-Nazari ; Stade ; Neville and Rosbash, 1999). Classic NLSs or NESs are predictable by direct comparison with well-established consensus sequences, and appropriate online search engines have been developed (la Cour ; Nguyen Ba ). However, the nuclear export and import signals recognized by the other KAPs are less predictive, and empirical approaches are usually required. Two additional transport pathways, largely divergent from KAPs, have also been defined. Certain mRNAs are exported by the MEX/NXT heterodimer (Macara, 2001). The import of the small Ras-type GTPase Ran (Ras nuclear), a key element for nuclear transport (see later discussion), is mediated by the nuclear transport factor Ntf2p/NTF2, which is closely related to the Mex and Nxt receptors (Macara, 2001).

The energy required for nuclear transport is supplied by the asymmetric distribution of the Ran GTPase. The cytoplasmic concentration of GTP-bound Ran (Ran-GTP) is low because it rapidly converts into Ran-GDP. In contrast, the nuclear Ran-GTP concentration is elevated. This nucleocytoplasmic gradient is generated by Ran cofactors (Mattaj and Englmeier, 1998; Gorlich and Kutay, 1999; Macara, 2001). Ran guanine nucleotide exchange factor (RanGEF) is a chromatin-associated nuclear protein (Ohtsubo ) that catalyzes Ran GDP-to-GTP exchange. Ran-GTPase activating protein (RanGAP) is a cytoplasmic protein that activates Ran GTPase activity (Bischoff ). Importins bind NLS-containing cargoes in the cytoplasm under low Ran-GTP conditions. In the nucleus the cargo/importin heterocomplex is disassembled by the interaction with Ran-GTP, with the subsequent release of the cargo in the nucleoplasm (Petosa ). Exportins bind NES-containing proteins in the nucleus due to the presence of a higher Ran-GTP concentration. Once in the cytoplasm, the GTP-to-GDP hydrolysis induced by RanGAP causes the release of the cargo (Mattaj and Englmeier, 1998; Gorlich and Kutay, 1999; Macara, 2001; Weis, 2002). Ran-GDP is then imported in the nucleus by Ntf2p/NTF2 action, completing the transport cycle.

The NE remains intact, but NPCs undergo a partial disassembly during mitosis in Aspergillus nidulans (De Souza and Osmani, 2007; Osmani ). Nucleoporins (NUPs) are the elements constituting the basic structure of the NPC and are organized within the pore into three main substructures (Ryan and Wente, 2000): 1) Concentric rings formed by a “core” of structural NUPs that anchor the NPC to the membrane. 2) Radial symmetric modules allowing expansion of the NPC structure toward both the cytoplasm and the nucleoplasm. 3) NUPs organized into flexible filaments that generate the specific environment inside the pore for the translocation of transporter/cargo heterocomplexes in or out of the nucleus. The exact mode of transport through the NPC remains unknown, and three models have been proposed: Brownian movement (Rout ), hydrophobic exclusion (Ribbeck and Gorlich, 2002), and “oily spaghetti” (Macara, 2001). “Core” or structural NUPs remain at the NE during mitosis in A. nidulans, but the more peripheral NUPs generally disperse, allowing free diffusion of macromolecules between nucleus and cytoplasm at this phase (De Souza and Osmani, 2007; Osmani ). Besides these changes in NPC composition, the nucleolus segregates in a unique manner during mitosis in A. nidulans. The nucleolus is first compartmentalized away from DNA during DNA segregation at anaphase. The nucleolar proteins are then disassembled into the cytoplasm and are translocated to daughter nuclei for subsequent regeneration of nucleoli (Ukil ).

In this work we present the first functional and localization analysis of the nuclear transport machinery in a multinucleated cell. The A. nidulans “nuclear-transportome,” defined as the collection of nuclear transporters, has been studied by generating strains carrying single null alleles for each carrier. This reverse genetic study has allowed the identification of essential and nonessential nuclear transport pathways for a coenocytic cell. Cellular distributions were investigated by the generation of endogenous fluorescent fusions. Nuclear transporters are associated with every nucleus of the syncytium during interphase, but distribution changes during mitosis and specific locations for several carriers are observed. Overall the data indicate how nuclear transport might occur and suggest possible roles for some of these transporters outside of nuclear transport in a model filamentous fungal cell.

RESULTS

Nuclear transport receptors in A. nidulans

Blast searches confirmed previous in silico analyses (Mans ; Espeso and Osmani, 2008) in which 17 loci coding for nuclear transporters were identified, as summarized in Table 1 (as indicated in this table, and for easy understanding of the text, we will refer hereafter to each A. nidulans locus with its designation and, in a superscript, the standard gene name in S. cerevisiae). Three genes encode for Ntf2-like family members (Table 1). NtfA (An4942) is a protein highly similar to S. cerevisiae and human nuclear transport factors Ntf2p and NTF2, respectively and is predictably involved in the nuclear import of the Ran-type GTPase RanA (An5482; Kadowaki ). Locus An2737, annotated as mexA, encodes a protein with similarity to S. cerevisiae Mex67p and human nuclear RNA export factor 1, NXF1/TAP (Table 1). MexAMex67 predictably contains all of the characteristic domains described for its orthologues (reviewed in Conti and Izaurralde, 2001; Terry and Wente, 2009; and references therein): the RNA-binding and the leucine-rich repeat domains at the N-terminal part, and an NTF2-like domain and a UBA-like domain in the C-terminal region. The presence of these domains anticipates the interaction of MexAMex67 with NUPs and an NTF2-like export factor (Fribourg ; Grant ). Locus An3864 encodes a putative homologue of the human NTF2-like export factor 1, p15/NXT1, and was designated as nxtA. However, we were unable to identify a putative homologue for the S. cerevisiae essential mRNA transporter Mtr2p (Kadowaki ), which is a structural and functional homologue of NXT1 (Fribourg and Conti, 2003). Thus we predict the formation of a MexAMex67/NxtAp15 heterocomplex in A. nidulans to mediate the export of specific mRNAs as occurs in higher eukaryotes (Zenklusen and Stutz, 2001).

TABLE 1:

Nuclear transporters in Aspergillus nidulans.

A. nidulans locus/systematic namea	S. cerevisiae // H. sapiens homologues % similarity/identity S.c. vs. A.n. // H.s. vs. A.n.	Knockout functional analysis	Preferential cellular localization
			At interphase	During mitosis
Ntf2-like family
ntfA/AN4942ntfANtf2	Ntf2p // NTF2 77/59 // 59/41	Essential	Nuclear	nd
mexA/AN2737mexAMex67	Mex67p // TAP-NXF1 49/30 // 44/23	Nonessential	Perinuclear	Total dispersion
No homologue	Mtr2p // −	—	—	—
nxtA/AN3864ntxAp15b	− // p15-NXT1 − // 47/27	Nonessential	Perinuclear	Total dispersion
Karyopherin-β superfamily
kapA/AN2142kapASrp1	Kap60p(Srp1p) // Imp-α 1-6 75/59 // 73/57	Essential	Nuclear	Partial dispersion (nuclear)
kapB/AN0906kapBKap95	Kap95p(Rsl1p) // Imp β1 60/39 // 57/36	Essential	Perinuclear	Partial dispersion (NE )
kapC/AN0926kapCKap104	Kap104p // Imp 2 51/30 // 56/35	Nonessential	Nuclear	Total dispersion
kapD/AN6006kapDNmd5	Kap108p(Sxm1p) // Imp 7 45/26 // 46/27 Kap119p(Nmd5p) // Imp 8 47/28 // 46/27	Nonessential	Nucleocytoplasmic	Total dispersion
kapE/AN6591kapECse1	Kap109p(Cse1p) // CAS 61/41 // 55/36	Essential	Nuclear	Partial dispersion (nuclear)
kapF/AN6734kapFMtr10	Kap111p(Mtr10p) // Transportin SR 52/31 // 44/26	Essential	Nuclear	Total dispersion
kapG/AN2164kapGKap114	Kap114p(Hrc1004p) // Imp 9 38/22 // 42/24	Nonessential	Nucleocytoplasmic	Total dispersion
kapH/AN4053kapHKap120	Kap120p(Lph2p) // Imp 11 48/27 // 48/28	Nonessential	Nucleocytoplasmic	Nuclear, mitotic spindlec
kapI/AN5717kapIPse1	Kap121p(Pse1p) // Imp 5 59/39 // 57/35	Nonessential	Perinuclear	Nucleolusc
kapJ/AN2120kapJKap123	Kap123p(Yrb4p) // Imp 4 49/28 // 44/24	Nonessential	Nucleocytoplasmic	Nucleolusc
kapK/AN1401kapKCrm1	Kap124p(Crm1p) // Exp1/Xpo1 73/54 // 71/54	Essential	Nuclear	Partial dispersion (nuclear)
kapL/AN3012kapLMsn5	Kap142p(Msn5p) // Exp 5 47/28 // 25/14	Nonessential	Nucleocytoplasmic	Total dispersion
kapM/AN8787kapMLos1	Los1p // Exp-T 45/24 // 45/25	Nonessential	Nuclear	Partial dispersion (nuclear)
kapN/AN7731kapNKap122	Kap122p(Pdr6p) // Imp 13 26/14 // 41/21	Nonessential	Nucleocytoplasmic	Total dispersion

In addition to S. cerevisiae Mtr2p, no putative homologues were found in A. nidulans for human exportins 4 and 7, Ran-binding proteins 17 and 20, and Snurportin.

A.n., A. nidulans; H.s., H. sapiens; S.c., S. cerevisiae. Exp, exportin; Imp, importin, nd, not determined; −, nonexistent.

Name adopted in this article indicating both A. nidulans and, in the superscript, the putative homologue in S. cerevisiae locus designations. For S. cerevisiae loci, the standard names at the Saccharomyces Genome Database (http//:www.yeastgenome.org) are used.

In the absence of a S. cerevisiae NxtA homologue, the human homologue is indicated in the superscript.

Transitory foci during mitosis.

Fourteen loci encode for proteins belonging to the karyopherin-β superfamily and consequently were designated as kap plus locus names from A to N. Among these nuclear receptors, KapD (encoded by locus An6006) was identified as the putative homologue of both S. cerevisiae Kap119p(Nmd5p) and Kap108p(Sxm1p) or human equivalents importins-7 and -8 (Table 1). A number of transporters are absent from fungal genomes, suggesting their specificity in higher eukaryotes (indicated in Table 1; Espeso and Osmani, 2008). Additional in silico searches using Pfam domains related to this superfamily of proteins did not add more candidates to our predictions (Supplemental Figure S1).

The expression of these in silico–identified genes was confirmed by cDNA sequencing and the predicted amino acid sequences compared with those of automatic predictions at the database. Minor errors in intron predictions were found, and database entries were consequently modified. These corrections did not change the classification initially achieved. Thereafter, and with the aim to confirm these predictions, we generated a phylogenetic tree incorporating Neurospora crassa, Schizosaccharomyces pombe, S. cerevisiae, and Homo sapiens homologues (Supplemental Figure S1). In this tree, and in agreement with similarities in their sequences (Table 1), each A. nidulans Kap grouped well with its putative homologues. We then assumed similar functions for these transporters to those initially proposed for their homologues (reviewed in Terry and Wente, 2009). Three genuine exportins are predicted—KapKCrm1, KapECse1, and KapMLos1—and nine importins—KapBKap95, KapCKap104, KapDNmd5, KapFMtr10, KapGKap114, KapHKap120, KapIPse1, KapJKap123, and the importin-α homologue KapASrp1, acting as the importin-β1 adaptor. Finally, there are two candidates for mediators of bidirectional transport: KapNKap122 and KapLMsn5. Msn5p has been shown to have both importin and exportin activities in S. cerevisiae (Yoshida and Blobel, 2001). Although Kap122p is an importin in S. cerevisiae (Titov and Blobel, 1999; Terry and Wente, 2009), KapNKap122 shares similarity with Imp13, a human bidirectional transporter (Mingot ).

Functional analysis of nuclear transporters by generation of null alleles

To assess a role for these nuclear transporters in A. nidulans, we generated single-knockout mutants for each locus by means of a precise gene replacement procedure (see Materials and Methods and Supplemental Figure S2). The deletion of 11 of 17 transporters produced viable homokaryotic colonies in transformation experiments (Table 1, knockout functional analyses), demonstrating that these genes are not essential. kapC, kapD, kapG, kapH, kapJ, kapL, kapM, and kapN deletants showed a wild-type phenotype over a variety of culture conditions (different carbon or nitrogen sources, a range of temperatures, and in saline and/or osmotic stress; Supplemental Figure S3). Of interest, KapIPse1 and MexAMex67 constitute notable exceptions compared with their S. cerevisiae homologues, which are essential (for database see http://www.yeastgenome.org; Seedorf and Silver, 1997; Segref ). kapI- and mexA-null strains grew in standard culture conditions but displayed morphological defects, such as reduced and compact growth and sparse conidiation (Figure 1A) (Etxebeste ). Low (30°C) and elevated (42°C) temperatures affected colonial growth of both kapI Δ and mexAΔ mutants but not the growth of other kap-null strains, such as kapG Δ (Figure 1A; see also Supplemental Figure S3). Considering NxtAp15 as a partner of MexAMex67 activity, it is interesting to note that, in contrast to the morphological defect of mexAΔ mutant, an nxtAnull strain shows wild-type growth at standard culture conditions (37°C) but displays sensitivity to 42°C. All nxtAΔ mexAΔ and mexAΔ strains showed a comparable thermosensitive phenotype and compact morphology at 37°C (Figure 1A), suggesting a major role of MexAMex67 in the putative heteromer. A double-null nxtA strain is viable and demonstrates the nonessentiality of a heteromeric NxtAp15/MexAMex67 transport pathway, but a compact colony morphology of the double-null mutant is observed in different growth conditions (Supplemental Figure S3), showing an additive effect of both null alleles. This additive phenotype suggests independent functions for each carrier (see also later discussion).

FIGURE 1:

Null mutants in nuclear transporters displaying morphological defects or that are essential. (A) Morphological alterations caused by the deletion of nonessential KapIPse1 and MexAMex67/NxtAp15 transporters. Strains MAD2159, MAD2332, MAD3017, and MAD3281 were cultured at different temperatures and colony radial growth compared with that measured at 37°C with the wild-type strain (MAD1425). A kapGΔ-null mutant (MAD2158) was included as a reference because its shows the wild-type phenotype generated after the deletion of most nonessential karyopherins (see text). (B) Heterokaryotic colony morphology caused by the deletion of essential nuclear transporters KapECse1, KapFMtr10, KapKCrm1, and NtfANtf2. The heterokaryon rescue technique allows propagation of kapE-, kapF-, kapK-, and ntfA-null mutant strains as heterokaryons and phenotypic analysis of null conidia from these heterokaryotic colonies, as parental conidia do not grow on selective media lacking pyrimidines (uracil and uridine [UU]). As a representative example, the kapKΔ growth test is shown (inset with colonies inoculated in a radial distribution). The absence of growth onto the selective media (−UU) demonstrates the lethal phenotype of the deletion allele. (C) Microscopic growth analysis of germinating conidia from mutants in essential transport pathways. Conidia from the heterokaryons were incubated in selective media for 24 and 72 h at 37°C. Germ-tube generation was analyzed, and the number of nuclei and their locations and morphology were determined using DAPI staining. Scale bars, 5 μm.

Homokaryotic transformants could not be obtained when deleting kapE, kapF, or ntfA, suggesting that these are essential genes. Consequently, transformants were maintained using the heterokaryon-rescue technique (Osmani ). Conidia from these heterokaryotic colonies were unable to grow on pyrimidine-selective solid medium (without supplementation of uracil and uridine; exemplified for kapK knockout strains; Figure 1B), indicating the nonviability of those conidiospores bearing the null allele of the corresponding transporter. The presence of both recombinant and wild-type nuclei in the heterokaryotic mycelia was confirmed by PCR and Southern blot techniques (unpublished data). Thus we completed the list of indispensable transport pathways in A. nidulans by adding KapECse1, KapFMtr10, KapKCrm1, and NtfANtf2 to the previously identified essential nuclear carriers importin-β1, KapBKap95 (Osmani ), and importin-α, KapASrp1 (Araújo-Bazán ).

We studied the germination process of mutant conidia to determine the effect of the deletion of these essential genes. Conidiospores from the isolated heterokaryons were cultivated in pyrimidine-selective liquid media at 37°C, where those conidia carrying the recombinant nucleus would grow to the extent that lack of the essential gene allows, and visualized after 24 and 72 h of incubation (Figure 1C). We observed isotropic growth and the establishment of polarity in conidia from kapA-, kapE-, kapF and ntfA-null strains, but two main phenotypes were distinguished based on the number of nuclei present in the cell. In the first class, comprising kapFΔ and kapAΔ mutants, a single nucleus per germling was observed after 4′,6-diamidino-2-phenylindole (DAPI) staining (n = 30 cells). This single nucleus was located in the proximity of the germinated conidium (Figure 1C, rows 2 and 3), suggesting a restricted nuclear mobility in these mutant backgrounds. The second class comprised kapEΔ, kapKΔ, and ntfAΔ mutant strains, in which two nuclei were observed and consequently the first mitosis occurred in the cells (n = 30; Figure 1C, rows 4– 6). In common for both classes, polar growth at initial stage of germination seemed not to be impeded, as germ tubes were observed emerging from most of conidia, but hyphal growth arrested, as major differences in length were not found between 24 and 72 h of culture. In addition, after 72 h of incubation, small DAPI-stained masses were found in some cells, indicating the existence of nuclear degradation or DNA missegregation. Overall, these results suggested a major role of KapASrp1 and KapFMtr10 in the onset of the first mitosis, whereas KapECse1, KapKCrm1, and NtfANtf2 are most probably required for completion of the first nuclear division. These transporters seem not to be required for the establishment of polarized growth, and the inhibition in apical extension observed might be triggered as a consequence of this mitotic block or as general failure of a number of cellular processes (e.g., blockage of transcription factor import/export, mRNA cytoplasmic transport). This and previous work (Osmani ) show that the null kapB strain displays the most extreme degree of developmental affection, since conidia germinated, but neither polar growth nor mitosis was observed (Osmani ), supporting, thus, a key role of KapBKap95 in both cellular processes.

Cellular distribution of nuclear transporters during interphase

A noteworthy issue to resolve was the localization of the nuclear transportome in a multinucleated vegetative cell. With this aim we systematically generated strains expressing fluorescent-tagged versions of nuclear carriers (see Materials and Methods for details and Supplemental Figure S2). Strains expressing C-terminally green fluorescent protein (GFP)–tagged versions of KapCKap104, KapDNmd5, KapGKap114, KapHKap120, KapLMsn5, KapMLos1, and KapNKap122 transporters lacked any appreciable phenotype. However, exceptions were those strains expressing NtfANtf2::GFP or KapJKap123::GFP. The former displayed a heterokaryotic development similar to the null ntfA strain, so we concluded that GFP tagging rendered a nonfunctional NtfANtf2 protein. Therefore further characterization of NtfANtf2::GFP also required the heterokaryon-rescue technique. Strains expressing KapJKap123::GFP showed limited growth on minimal compared with complete media, especially when incubation temperature was 30°C (Supplemental Figure S4A). It is worth noting that in S. cerevisiae the kap123Δ allele causes a reduced growth rate in complete medium (Rout ). However, this is not the case for a kapJΔ mutant. In minimal medium, DAPI staining of kapJ cells revealed an anomalous distribution of DNA in nuclei, suggesting that C-terminal GFP tagging of KapJKap123 might cause missegregation of DNA, perhaps causing the restricted-growth phenotype observed (Supplemental Figure S4C). Then we generated a GFP::KapJKap123 strain (MAD3563), which showed a wild-type phenotype, and therefore it was used for subsequent localization analyses.

Expression of full-length chimeras was verified by Western blot assays, in which most of them showed the expected mobility according to their molecular weights (Supplemental Figure S5). However, KapFMtr10::GFP showed a markedly lower mobility, and KapLMsn5::GFP displayed a pattern of four bands, which suggests the existence of either alternative forms or posttranslational modifications in this transporter. Detection of MexAMex67::GFP was always poor, but a band of the expected size was visualized (Supplemental Figure S5).

For localization studies special care was taken to image cells with at least four or five nuclei in either apical or subapical compartments. When fluorescence was low or nuclear fluorescence could not be easily differentiated from the cytoplasmic background, we constructed strains coexpressing the GFP-tagged transporter and the histone H1 (HhoA) fused to mCherry (mCh; Cherry red) as a nuclear marker. Prior to further analyses, we verified that the nuclear transporter chimeras displayed the same localization in wild-type and hhoA::mCh genetic backgrounds.

Three categories were differentiated according to their distribution patterns in interphase (Figure 2). The first group comprised KAPs having a preferential nuclear localization, with a ratio of nuclear/cytoplasmic fluorescence greater than three (Figure 2A). This group included KapASrp1, KapCKap104 KapECse1, KapFMtr10, KapKCrm1, and KapMLos1. Although NtfANtf2::GFP fusion was not functional, we determined its localization in heterokaryotic cells in which at least two mitoses occurred. NtfANtf2 accumulated in the nucleus and at foci close to the nuclear periphery, probably near the NE (visualized in 90% of nuclei; n = 20; Figure 2A, white arrowheads). The seven transporters belonging to this group presented a regular distribution at the nucleoplasm but showed no fluorescence at the nucleolus, as observed in colocalization studies with the mCh-tagged nucleolar protein fibrillarin, An-Fib (AN0745.3; Ukil ). See Supplementary Figure S6, for the KapKCrm1::GFP chimera.

FIGURE 2:

Distribution of nuclear transporters during interphase. nuclear transporters were classified into three groups: (A) those showing an intranuclear localization, (B) those displaying a nucleocytoplasmic localization, and (C) those with preferential perinuclear or NE-associated localization. In the insets, magnifications of nuclei exemplify the absence of fluorescence at nucleoli, which was confirmed by the observation of the nucleolar protein An-Fib. For KapJKap123, a maximal intensity projection of deconvolved z-stacks in an inverted display shows the localization of GFP::KapJKap123 in the proximity of nucleoli. See the text for arrowhead explanation. Homokaryotic strains used here: MAD2067, MAD2913, MAD2161, MAD2162, MAD2163, MAD2164, MAD2165, MAD2175, MAD2325, MAD2914, MAD2652, MAD2328, MAD2331, MAD2330, MAD2333, and MAD3563. Scale bars, 5 μm.

Those KAPs showing a ratio of nuclear versus cytoplasmic fluorescence less than three were designated as “nucleocytoplasmic” karyopherins. The fusion proteins included in this second group were C-terminally GFP-tagged KapDNmd5, KapGKap114, KapHKap120, KapNKap122,and KapLMsn5 and GFP::KapJKap123 (Figure 2B; KapJKap123::GFP is shown in Supplemental Figure S4, B and C). None of the transporters was excluded from nuclei, and nuclear fluorescence from tagged KAPs was always visible (gray arrowheads indicate nuclei positions using HhoA::mCh; Figure 2B). Both GFP/KapJKap123 chimeras presented a similar distribution at the periphery of the nucleolus (Figure 2B and Supplemental Figure S4B) and as demonstrated colocalization with An-Fib-mCh (shown for GFP::KapJKap123 in Supplemental Figure S6). In the cytoplasm, fluorescence distribution along the hypha was homogeneous for all chimeras, with the major exception of KapNKap122::GFP, which showed a speckled distribution (Figure 2B and Supplemental Figure S4D).

The C-terminally GFP-tagged versions of KapBKap95, KapIPse1, MexAMex67, and NxtAp15 constitute the third group, termed “NE-associated” transporters (Figure 2C). These transporters showed an accumulation at the nuclear periphery similar to that observed for nucleoporins (Figure 2C; Osmani ), and KapBKap95 also accumulated in the proximity of the spindle pole body (SPB), the fungal centrosome, as indicated by colocalization with Nud1-mCh (unpublished data; see later discussion for the use of this SPB marker). Of note, localization of NxtAp15 at the NE depends on MexAMex67, since dispersion along the cytoplasm and nuclei was observed for NxtAp15::GFP in a null mexA background (unpublished data; strain MAD3282). As for the rest of transporters, these are also excluded from nucleolus (exemplified for KapBKap95 in Supplemental Figure S6).

In view of the absence of predicted transmembrane domains in these nuclear carriers, it is reasonable to presume direct or mediated contacts with the constituents of the NPC. Although has not been investigated further, the study of the localization of nuclear carriers during mitosis suggests a dependence of the so-called “soluble-transport” machinery on the active transport and full assembly of NPCs.

Distribution of nuclear transporters during mitosis

Mitosis in A. nidulans has been defined as partially opened (De Souza and Osmani, 2007) because partial disassembly of the NPC during mitosis modifies the permeability of the NE, allowing mixing of nuclear and cytoplasmic proteins. Predictably, nuclear transporters may alter their localizations following mitotic FG-nucleoporin reallocations.

We excluded from this study those strains expressing KapJKap123::GFP or NtfANtf2::GFP, due to the functional deficiency caused by the fluorescent tag. A general observation for all tested constructs is that nuclear fluorescence was reduced when nuclei entered mitosis. In this context, we classified the nuclear transporters into three groups. Figure 3 shows those transporters that displayed dispersion of fluorescence between nucleus and cytoplasm during mitosis (see also the associated videos: KapCKap104::GFP (Supplemental Video S3), KapDNmd5::GFP (Supplemental Video S4), KapFMtr10::GFP (Supplemental Video S6), KapGKap114::GFP (Supplemental Video S7), KapLMsn5::GFP (Supplemental Video S11), MexAMex67::GFP (Supplemental Video S13), and NxtAp15::GFP (Supplemental Video S14). For nucleocytoplasmic transporters, mitosis was followed by colocalization with HhoA::mCh (Figure 3).

FIGURE 3:

Nuclear transporters showing total dispersion during mitosis. Seven of the nuclear transporters displayed cytoplasmic dispersion from their interphasic localizations during mitosis. Selected frames and regions from videos (Supplemental Videos S3, S4, S6, S7, S11, S13, and S14) of representative cells at late G2, mitosis (M), and early G1 stages are shown. In the case of carriers with nucleocytoplasmic localization, KapDNmd5, KapGKap114, and KapLMsn5, nuclei were visualized during cell cycle using mCherry-tagged histone H1 (HhoA::mCh) as a nuclear marker (strains MAD2853, MAD2854, and MAD2653, respectively). Scale bars, 5 μm.

The second group included KapASrp1::GFP (Supplemental Video S1), KapECse1 (Supplemental Video S5), and KapMLos1 (Supplemental Video S12), which showed partial dispersion during mitosis (Figure 4A). The intensity of nuclear florescence during mitosis decreased in these cases but was not completely dispersed into the cytoplasm. This indicates that a pool of these transporters may remain inside the nucleus, interacting with or attached to nuclear structures or protein complexes. KapKCrm1 and more dramatically KapBKap95 were partially retained in the nucleoplasm and at the nuclear envelope, respectively (Supplemental Videos S10 and S2, respectively). In particular, KapBKap95::GFP allowed the tracking of the NE during mitosis and the visualization of its rupture to render independent daughter nuclei (Figure 4A, blue arrowheads at KapBKap95 insets). This observation indicated a possible close relationship between KapBKap95 and core nucleoporins of the mitotic NPC. We were able to observe the compartment that contains the nucleolus during mitosis when either KapBKap95 or KapKCrm1 was monitored (Figure 4A, white arrowheads in either KapB or KapK). KapKCrm1 briefly remains at the nucleoplasm bordering the cytoplasmic parental nucleolus before it starts disassembling. An-Fib exits the parental nucleolus after the migration of KapK (Supplemental Video S16 and kymograph in Figure 4B). The maintenance of these two KAPs in close relationship with the nucleus or NE during the mitotic process suggests roles in mitotic progression and/or restart of nuclear transport after NPC rearrangement.

FIGURE 4:

Nuclear transporters displaying partial dispersion during mitosis. (A). As in Figure 3, selected frames and regions from videos (Supplemental Videos S1, S2, S5, S10, and S12) of representative cells at late G2, mitosis (M), and early G1 stages are shown for five KAPs. The fluorescence signal redistributes at the onset of mitosis and in G1 recovers its interphasic localization. Nuclear fluorescence is seen for KapASrp1, KapECse1, and KapMLos1, showing the last two insets in inverted display for best visualization. At early mitosis KapBKap95 and KapKCrm1 disperse but some remains in the mitotic nucleus, being associated with the NE (white arrowheads). (B) KapKCrm1 (Supplemental Video S16; strain MAD3540; KapKCrm1-GFP, green) remains nuclear during mitosis. KapKCrm1 is excluded from the parental nucleolus during the mitotic process. Magnification of a nucleus at early mitosis shows how KapKCrm1::GFP fluorescence is distributed uniformly along the nucleoplasm but excluded from the nucleolus, as shown by colocalization with fibrillarin (An-Fib, magenta). A kymograph of this series of images, along the cell region indicated by the yellow dotted line, illustrates the absence of KapKCrm1 from the nucleolus during the process, but especially when the nucleolus is extruded and starts its disassembly. Time scale, min:s. Scale bars, 5 μm.

The last group includes KapHKap120::GFP, KapIPse1::GFP, and GFP::KapJKap123, which are KAPs showing specific locations during mitosis (Figures 5 and 6; see next sections).

FIGURE 5:

KapHKap120 locates at the mitotic spindle. (A) KapHKap120 disperses from its nuclear localization at G2 and accumulates in a dot at the onset of mitosis, when chromatin starts to condense (visualized using HhoA-mCh; Supplemental Video S8; strain MAD2855), and the signal for KapHKap120 fluorescence elongates (black arrowheads) during mitosis, then concentrating again in daughter nuclei at G1. (B) Selected images from Supplemental Video S17 show the simultaneous visualization of KapHKap120::GFP (green) and the SPB protein Nud1 (Nud1-mCh, magenta) in cells from strain MAD3564. Arrows follow divisions of nuclei n1 and n2, which are shown in more detail on the right. Trajectories of SPBs, labeled with Nud1, are followed with a line. (C) Time-lapse images from a MAD3772 strain cell showing the localization of KapHKap120 during the cytoplasmic dispersion of NLS-DsRed reporter at mitosis. Time scale, min:s. Scale bars, 5 μm.

FIGURE 6:

KapJKap123 and KapIPse1 localizations are linked to the mitotic nucleolus. (A) KapIPse1 nucleolar accumulation coincides with initiation of An-Fib migration to daughter nuclei (right, Supplemental Video S18, strain MAD3539). A kymograph illustrates the colocalization of KapIPse1::GFP (green) and An-Fib-mCh (magenta) in the mitotic nucleolus but not in daughter nucleoli, where KapIPse1 localizes at the NE, as in an interphase parental nucleus (left, bottom). (B) KapJKap123, GFP::KapJKap123 is shown (strain MAD3594), which shows a perinucleolar localization and first disperses but rapidly starts to accumulate in the nucleolus. As previously described for KapIPse1, KapJKap123 (GFP-KapJKap123, green, Supplemental Video S20) localizes to the mitotic nucleolus during the process of fibrillarin segregation to daughter nuclei (An-Fib-mCh, magenta). When An-Fib relocation is completed, KapJKap123 accumulates to the new nucleolar peripheries again. Kymograph at the bottom shows that both KapJKap123 and An-Fib colocalized during most of the mitotic process, but KapJKap123 remains in the parental nucleolus until complete relocation of An-Fib. Time scale, min:s. Kymographs were done along the yellow dotted lines indicated at respective inserts. (C) Accumulation of KapIPse1 and KapJKap123 in the parental nucleolus coincides with recovery of active nuclear transport, as shown by nuclear accumulation of NLS-DsRed reporter at the M/G1 transition (indicated with red arrowheads, strains MAD3771 and MAD3817). Scale bars, 5 μm.

KapHKap120 locates at the mitotic spindle

The start of chromatin condensation in prophase coincided with an aggregation of KapHKap120 fluorescence in a spot that continued stretching out in a shape resembling that of the mitotic spindle (Figure 5A, black arrowhead, and Supplemental Video S8). After mitosis KapHKap120 was redistributed into daughter nuclei. A possible association with the mitotic spindle could not be verified using the available GFP-tagged version of tubulin α, tubA, because of instability of the KapHKap120::mCh protein. Consequently, we visualized the position of the SPB using the Nud1::mCh fusion as a marker. We verified that, at the onset of mitosis, KapHKap120 starts to accumulate at the proximity of the SPBs prior to their separation (see mitosis for selected nucleus 1, n1, in Figure 5B). Then the fluorescence accumulation of KapHKap120::GFP elongated as daughter SPBs moved and mitosis proceeded (Figure 5B, n1 and n2 mitoses, and Supplemental Video S17). After mitotic completion, KapHKap120::GFP fluorescence dispersed and in late G1 recovered the nucleocytoplasmic location described during interphase. The nuclear reporter StuA-NLS-DsRed requires an active nuclear import system to accumulate in the nucleus. The cytoplasmic localization of this reporter when KapHKap120 locates at the mitotic spindle demonstrates that this occurs in the absence of a functional NPC (Figure 5C).

KapIPse1 and KapJKap123 locate at the mitotic nucleolus

KapIPse1 and KapJKap123 display different locations at interphase, but during mitosis both KAPs are localized in the nucleolus. KapIPse1 disperses from the NE, and we noticed a temporary and precise accumulation during mitosis that did not coincide with either of the daughter nuclei but was situated between them after DNA segregation (Supplemental Figure S7 and Supplemental Videos S9 and S19). Colocalization with the nucleolar An-Fib-mCh protein confirmed that such compartment was the nucleolus (Figure 6A and Supplemental Video S18). KapIPse1 accumulates at the parental nucleolus prior to the exit of An-Fib and lasts until the completion of An-Fib incorporation into daughter nucleoli (Figure 6A and Supplemental Video S18).

KapJKap123, initially at the border of the nucleolus, disperses momentarily at the onset of mitosis and then begins to accumulate in the cytoplasmic parental nucleolus (Figure 6B and Supplemental Video S20). KapJKap123 localization, as seen for KapIPse1, coincided with the process of nucleolar dispersion of An-Fib and the de novo formation of daughter nucleoli. Figure 6C shows that active nuclear transport recovered in daughter nuclei when both KapIPse1 and KapJKap123 accumulated in the parental nucleolus (Figure 6C, red arrowheads, M/G1 phase). We did not observe a preferential localization within the daughter nucleoli, suggesting that these KAPs might be specifically directed into the compartments that contain the nucleoli and be involved in their disassembly and the recycling of components to the new nucleoli but not directly in their reassembly at nucleolar organizing regions. Because null kapJ and kapI are viable, a functional redundancy might be proposed in A. nidulans for these two KAPs, as previously described in S. cerevisiae (Rout ).

DISCUSSION

The first systematic characterization of the nuclear transport machinery in a filamentous fungus highlights several interesting issues relevant to the understanding of this key regulatory process in multinucleated cells. Fluorescent tagging and deletion analysis has permitted us to define the cell cycle–specific KAP distribution map and to determine the four essential nuclear transport pathways in A. nidulans. The general import pathway comprises the importin α/β1 heterodimer, KapASrp1/KapBKap95, and KapECse1, which must be the specific importin-α exporter required for cytoplasmic recycling of this adapter. KapKCrm1/CrmA mediates the general nuclear export pathway (Todd ; Bernreiter ). KapFMtr10, a transportin SR, is predicted to export diverse mRNAs and other ribonucleoproteins (Senger ; Murthi ). Finally, NtfANtf2 is the key carrier that might import the Ran GTPase, RanA, an indispensable element for obtaining the energy and transport directionality across the NPC.

Information about the NtfANtf2–RanA relationship is limited due to the inability to obtain functional fluorescent fusions. However, conditional mutations in both general import and export systems previously allowed the identification of different cargoes and demonstrated their conserved mechanisms. The transcription factor (TF) for nitrate assimilation, NirA, and the regulator of nitrogen metabolism, AreA, are cargoes of the exportin KapKCrm1 (Bernreiter ; Todd ). Import of VeA, the developmental regulator in response to light, requires the importin-α/β1 heteromer (Araújo-Bazán ). Importin-α acts as the only known adaptor of importin-β1 in A. nidulans and recognizes monopartite and bipartite NLSs (Fernandez-Martinez ; Stinnett ; Araújo-Bazán ). However, VelB is an example of an alternative import mechanism through the importin-α/β1 pathway, since, lacking an obvious NLS, it is transported into the nucleus via interaction with VeA (Bayram ).

Of interest, deletion of kapI and mexA genes does not cause lethality in A. nidulans, whereas in S. cerevisiae both Kap121p/Pse1p and Mex67p are essential transporters. KapIPse1 activity is likely to mediate the import of key factors during the development of asexual reproductive structures (Etxebeste ). The MexAMex67/NxtAp15 putative heterodimer is predicted to mediate specific mRNA export (Zenklusen and Stutz, 2001). Single- or double-null mutants do not cause lethality, although the morphological defects of the mexAΔ strain and the mislocalization of NxtAp15 in this mutant background demonstrate that MexAMex67 is the principal component of this pathway in Aspergillus.

From a total of 14 predicted transport pathways, we have not found experimental evidence for roles for eight of them. Sensitivity to an elevated salt concentration and sensitivity to alkalinity are phenotypes derived from loss-of-function mutations in the pH regulator PacC (Peñalva ) or the cation stress response TFs SltA and CrzA (Spielvogel ). Although PacC localizes into the nucleus in an importin-α–independent manner (Fernandez-Martinez ), none of the viable kapΔ strains displayed a defect on pH regulation. In S. cerevisiae Crz1p is imported by Nmd5p, being homologues of CrzA and KapDNmd5, respectively. Both Crz1p and Nmd5p nulls show similar phenotypes (Polizotto and Cyert, 2001). However, neither kapDΔ nor any other kapΔ strains showed calcium or cation sensitivity comparable to that shown by crzAΔ or sltAΔ mutants (Spielvogel ). These data strongly support redundant roles of nuclear transporters in a variety of shuttle mechanisms, as previously proposed for other organisms (Fiserova and Goldberg, 2010).

The regular distribution of all nuclear carriers during interphase supports a model of transport organization in which every nucleus within the syncytium possess the required elements to mediate active transport. Proof of this are those cargoes that are coordinately transported in all nuclei, such as CrzA, NirA, NapA, PacC, or VeA (Bernreiter ; Araújo-Bazán , 2009; Peñalva ; Spielvogel ). However, one exception is FlbB, the bZIP-type factor modulating asexual development, which shows preferential accumulation at the most apical nucleus (Etxebeste ). This finding predicts the presence of transport mechanisms that might discriminate among different cargoes and specifically shuttle them into all or selected nuclei.

Changes in the subcellular distribution of nuclear carriers during mitosis offered an avenue for delimiting the processes in which these proteins are involved. The partial disassembly of NPCs changes the localization of several nuclear transporters. Such changes hint at how key processes in the mitotic nucleus might occur. Mitotic dispersion of transportin-SR KapFMtr10 and the NE-associated MexAMex67/NxtAp15 heterocomplex add support to early conclusions that RNA export must be low or practically nonexistent during mitosis (Espeso and Osmani, 2008). However, mechanistically, Mex homologues are expected to interact with the Nup84 subcomplex (Yao ), which in A. nidulans is part of the mitotic NPC core structure (Osmani ); thus the existence of specific mechanisms, that is, posttranslational modifications, acting on MexAMex67 has to be postulated. Interesting regulatory or interaction events might occur during the mitotic localization of KapHKap120 following tubulin assembly at the mitotic spindle. This observation suggests a function for this transporter during or after spindle assembly. It is tempting to suggest a role in the transport of spindle-associated proteins, tubulin itself, or with chromosomes, that is, at the kinetochores, as has been shown for other carriers (Kalab and Heald, 2008, and references therein). A characterization of the spindle composition in S. cerevisiae showed the interaction of Kap120p with Mtw1p, an essential component of the MIND kinetochore complex (Wong ). Finally, the nucleolar localizations of both KapIPse1 and KapJKap123 during mitosis link these transporters with the process of nucleolar division, hypothetically as an extension of their putative interphase roles in the transport of ribosomal components (Sydorskyy ). However, kapH, kapI and kapJ, deletions are not lethal, suggesting that the presence of redundant roles is one open possibility in A. nidulans.

KapBKap95 and KapKCrm1 are essential karyopherins associated with mitotic nuclei. In addition to their predicted and demonstrated roles in general import and export pathways, we postulate that permanence of KapBKap95 and KapKCrm1 at the nucleus during mitosis indicates that they have important postmitotic functions, for instance, in a primary step at the restart of nuclear transport after NPC reassembly, rendering proper distribution of key factors such as Ran-GAP after its unspecific nuclear entry in mitosis (De Souza ). These transporters could also participate in proper distribution of cell-cycle regulatory proteins during interphase and after mitosis because conidium germination and the first mitosis are affected in kapK and, more severely, in kapB deletants (Osmani ; this work). As described for Kap95p or Kap95p/Srp1p heterocomplexes in S. cerevisiae (Taberner and Igual, 2010), KapBKap95 could be involved in the import of transcriptional factors required at Start of which down-regulation causes cell cycle arrest.

This work brings together the localization and functionality of nuclear carriers with previous systematic studies in the organization of NPC at interphase and mitosis. The battery of nuclear carriers in A. nidulans suggests conserved transport pathways between high and low eukaryotes. However, an interesting biological question to address in the future is how these general mechanisms have been adapted to serve a coenocytical lifestyle. Future studies will focus on understanding how cargoes and carriers overcome the long cellular distances in filamentous fungal cells to coordinately respond to ambient and internal stimuli.

MATERIALS AND METHODS

Strains, media, and culture conditions

Media and general techniques for A. nidulans culture and transformation were used as previously described (Araújo-Bazán ). A. nidulans strains used in this study carried markers in standard use (Clutterbuck, 1993) and are listed in Supplemental Table S1. Strain TN02A3 (Nayak ), identified here as MAD1425, was used for the systematic deletion and fluorescent tagging of nuclear transporters. MAD1427 was used for the generation of double-null mexA/nxtA and mexAΔ/gfp-tagged nxtA strains. Strain MAD2484 was used as recipient for colocalization studies of GFP- tagged Kaps and the SPB protein Nud1 and strain AY02 for colocalization studies with the active-nuclear-transport marker NLS-DsRed. Strains carrying deletions in essential genes were handled as heterokaryons (“heterokaryon rescue technique”) and, hence, propagated by transfer of mycelia pads (Osmani ). Phenotypes caused by the deletion of purportedly nonessential genes were studied under a range of ambient stresses. Salt or osmotic stress was induced by addition of CaCl2 (0.1 M), MgCl2 (0.1 M), LiCl (0.1, 0.3M), NaCl (0.1, 1 M), or KCl (0.1 M). Acidic or alkaline stress was induced adjusting media pH to 5.2 or 8.0 with NaH2PO4 (0.1 M) or Na2HPO4 (0.1 M), respectively. Temperature effects were studied by comparing phenotypes at 30 and 42°C to those at 37°C. Radial extension of mutant strains was always compared with that of the parental MAD1425 strain.

Bioinformatic tools

Amino acid sequences of previously identified yeast (S. cerevisiae) and human transporters were used to search for A. nidulans putative homologues using BLASTp and tBLASTn programs at AspGD (http://www.aspergillusgenome.org/) and Broad Institute (http://www.broadinstitute.org/annotation/genome/aspergillus_group/MultiHome.html) databases. Reverse searches were performed to confirm the best putative homologue for each identified gene. Multiple alignments were done with ClustalX (Thompson ) and revised manually using GeneDoc software (http://iubio.bio.indiana.edu/soft/molbio/ibmpc/genedoc-readme.html). Phylogenetic and molecular evolutionary studies were performed using Mega software, version 4.0, neighbor-joining method, with a bootstrap of 50,000 replicates and the amino p-distance substitution model (Tamura ). Similarity, following Blosum62 matrix, and identity percentages were estimated using the complete amino acid sequence for all transporters.

Molecular techniques and generation of null and fluorescent-tagged strains

Strains carrying null alleles, GFP or cherry-mRFP C-terminally tagged fusions of nuclear transporters, and nuclear and nucleolar markers were generated as described (Yang ; Supplemental Figure S2), using Aspergillus fumigatus pyrG, riboB, or pyroA gene as a prototrophic selection marker. To construct an N-terminal GFP-tagged KapJKap123, a null kapJ::pyrG strain was transformed with a 5′UTR::GFP::KapJORF::3′UTR DNA cassette, and the replacement of the null locus was selected by growing the transformants in regeneration plates containing 2 mg/ml 5-fluoro-orotic acid (5-FOA; Apollo Scientific, Stockport, United Kingdom). After 24 h of incubation at 37°C, a layer of Aspergillus minimal medium containing the required nutrients and supplements plus 2 mg/ml FOA was poured onto the protoplast-regeneration plates. Colonies that grew through this layer of medium were purified to homokaryosis and analyzed by Southern blot to verify the correct integration of the cassette at the null kapJ locus. Strains coexpressing tagged karyopherins and histone H1 (HhoA::mCh) or fibrillarin (An-Fib::mCh) were obtained by step-by-step transformation of MAD1425 with each fusion PCR cassette. Amplification of HhoA and An-Fib tagging cassettes follow standard procedures as described, but the selectable marker in this case was A. fumigatus pyroA gene. Strains MAD2653 (kapL; hhoA::mCh) and MAD2654 (kapI; hhoA::mCh) were obtained by transformation of MAD2446 with the karyopherin-tagging PCR cassette. Strains MAD3772 (kapH; nls::Dsred) and MAD3771 (kapI; nls::Dsred) were obtained by transformation of AY02 with the karyopherin-tagging PCR cassette. Strain MAD3817 is a diploid obtained from MAD3773 and MAD3604 haploid strains.

Oligonucleotides used in this study are summarized in Supplementary Table S2. Homologous recombination for each construct was confirmed by both diagnostic PCR and Southern blot analysis.

The expression of tagged proteins was analyzed by Western blotting using standard procedures. Briefly, A. nidulans strains were cultivated for 18 h in fermentation medium (Orejas ), filtered through Miracloth (Calbiochem, La Jolla, CA), squeezed to dry, frozen in dry ice, and lyophilized for 16 h. Protein extraction was carried out as previously described (Araújo-Bazán ), and 50 μg were loaded on an 8% polyacrylamide gel before electrotransfer to nitrocellulose filters. To detect GFP fusions, filters were incubated with anti-GFP mouse monoclonal antibody cocktail (1/5000; Roche, Indianapolis, IN). Actin, used as loading and extract quality control, was detected using mouse anti-actin antibody (1/50,000; ICN Biomedical, Irvine, CA). Peroxidase-conjugated anti-mouse (1/4000; Jackson ImmunoResearch Laboratories, West Grove, PA) was used as secondary antibody.

To characterize respective transporter coding sequence (CDS), cDNA was generated from total RNA extracted using TRIzol reagent (Invitrogen, Carlsbad, CA) and following the manufacturer's instructions. Total RNA was isolated from mycelia of a wild-type strain cultured under standard conditions. A single-strand cDNA library was made using the First-Strand cDNA Kit (Roche) and an oligo-dT primer. cDNA for each nuclear transporter was amplified using gene-specific oligonucleotides, and the CDS was confirmed by sequencing.

Fluorescence microscopy

Germlings were cultured in supplemented watch minimal medium (Peñalva, 2005), using uncoated glass-bottom dishes (MatTek Corporation, Ashland, MA). Fluorescence images were acquired with an upright Eclipse 80i microscope (Nikon, Melville, NY) equipped with Brightline GFP-3035B and TXRED-4040B filter sets (Semrock, Rochester, NY), a 100-W mercury lamp epifluorescence module, a Uniblitz (Rochester, NY) external shutter, a 60× 1.40–numerical aperture (NA) plan apochromat objective, and an ORCA ERG camera (Hamamatsu, Bridgewater, NJ). In vivo imaging was performed at 37°C using a DMI6000B inverted microscope (Leica, Deerfield, IL) equipped with a heating insert P (PeCon, Erbach, Germany), a Hamamatsu ORCA ER-II camera, an EL6000 external light source for epifluorescence excitation, an HCX 63× 1.4 NA objective, and Semrock Brightline GFP-3035B and TXRED-4040B (mRFP) filter sets. Kymographs, contrast adjustment, color combining, and z-stack maximal intensity projections were made using Metamorph (Molecular Devices, Sunnyvale, CA). Deconvolution was made using AutoDeblur software (Media Cybernetics, Bethesda, MD) and a blind deconvolution setup. Time-lapse sequences were converted to QuickTime format using ImageJ 1.37 software (National Institutes of Health, Bethesda, MD).

Nuclei were visualized using mCherry-tagged histone H1 (HhoA; Etxebeste , and references therein) or with DAPI staining. An-Fib (AN0745.3), an rRNA 2′-O-methyltransferase similar to S. cerevisiae Nop1 and vertebrate fibrillarin, was used as a marker for nucleolar positioning and disassembly process (Ukil ). Nud1 tagged with mCh (Xiong and Oakley, 2009) was used for labeling the position of spindle pole bodies. Active nuclear transport during mitosis was followed using the StuA NLS fused to the DsRed fluorescent protein as a reporter whose expression is driven by the constitutive promoter gpdA (Suelmann ; Ukil ).