The “8-kD” Cytoplasmic Dynein Light Chain Is Required for Nuclear Migration and for Dynein Heavy Chain Localization in Aspergillus nidulans

Strain ts289 (nudG8, pabaA6, biA1) carrying the nudG8 mutation was identified by fluorescence microscopic inspection of nuclear distribution in 4′,6-diamidino-2-phenylindone (DAPI)-stained germlings from a collection of 1,164 temperature sensitive mutants generated by 4-nitroquinoline oxide mutagenesis of strain FGSC (Fungal Genetics Stock Center) A28 (pabaA6, biA1). This collection of ts mutants was made by S. Harris, M. Momany, and J. Hamer of the Department of Biological Sciences of Purdue University (Lafayette, IN), who generously shared their collection with us. Crosses between ts289 and XX19 (nudA2, chaA1, pyrG89, nicA2, and/or nicB8), AO1 (nudC3, wA2, pabaA1, nicA2, and pyrG89), and XX21 (nudF7, yA2, and pyrG89) identified nudG8 as a mutation in a new nud gene. ts289 was outcrossed to GR5 (wA2, pyroA4, and pyrG89) to introduce pyrG89, a selective marker suitable for transformation selection, into the strain. The resultant strain, named SB09, (nudG8, wA2, pyroA4, pyrG89, and biA1) was crossed to A391 (chaA1 and biA1) to produce SB10 (nudG8, chaA1, pyroA4, pyrG89, and biA1). Media and growth conditions were as previously described for other nud mutations (Xiang et al., 1994; Xiang et al., 1995a,b). Progeny from a cross between SB09 (nudG8, wA2, pyroA4, pyrG89, and biA1) and A391 (chaA1 and biA1) were used for self-crosses to test for fertility defects. A double mutant strain was generated by crossing SB10 with XX60 (ΔnudA::pyrG; pyrG89; Xiang et al., 1995a).

To observe nuclei of A. nidulans germlings, spores were inoculated onto coverslips overlaid with medium on the bottom of a Petri dish and grown 8–12 h at either permissive or restrictive temperatures. Cells were fixed in 100 μl of 5% glutaraldehyde in 50 mM K₂HPO₄, pH 6.5, 0.25 μg/ml DAPI and 0.2% Triton X-100 for 20 min at room temperature. The coverslips were then rinsed with water and air dried. Coverslips were mounted on glass slides with a drop of Citifluor (UKC Chemlab, Canterbury, UK) and observed on an epifluorescence microscope (Carl Zeiss, Inc., Thornwood, NY). CDHC indirect immunofluorescence was performed as described previously (Xiang et al., 1995a). The nudA CDHC antiserum has been described previously (Xiang et al., 1995a). Tubulin was stained with DM1A antibody (Sigma Chemical Co., St. Louis, MO) against alpha tubulin. Antibody against actin was a gift from Andersland et al. (1994).

We mapped the nudG8 mutation to chromosome VIII by conventional parasexual genetic methods (reviewed by Clutterbuck, 1992). We used a chromosome VIII–specific cosmid library (Brody et al., 1991) obtained from the Fungal Genetics Stock Center and sib selection as described in Xiang et al. (1994) to clone the wild-type nudG gene. The library was divided into seven pools, each containing 50 cosmids. The SB10 (nudG8, chaA1, pyroA4, pyrG89, and biA1) strain was transformed with each pool, and we looked for DNA-mediated complementation of the mutant phenotype. The cosmid was subcloned to the smallest fragment, 2 kb, that would complement nudG mutation in the SB10 strain.

A. nidulans was transformed according to the protocol of Yelton et al. (1984) with the following modifications. We initially used ∼3 μg of cosmid DNA plus 13 μg of pyrG DNA. The final protoplast pellet was resuspended in STC buffer containing 1.2 M sucrose instead of sorbitol. An equal volume of 60% PEG solution (polyethylene glycol 4000, #81240; Fluka Chemical Corp., Milwaukee, WI) was added to the protoplast/DNA mixture, and after 20 min incubation, 1.0 ml of YGS (0.5% yeast extract, 2% glucose, and 1% sucrose) medium containing sucrose in place of sorbitol was added and the eppendorf tubes were shaken at 150 rpm for 1–2 h at room temperature. The protoplasts were plated on YAG-KCL (0.5% yeast extract, 2% agar, 2% glucose, 1.6% KCl) and incubated at 32°C for 12–20 h, after which YAG top agar was added to the plates and incubation continued at 44°C until colonies grew through the top agar. These were regridded and retested at 44°C on YAG. Rescue of the nuclear migration defect phenotype produced wild-type colony growth and abundant production of asexual spores at restrictive temperature.

We screened a lambda gt10 A. nidulans cDNA library (Osmani et al., 1988) using the 2-kb genomic fragment that rescued the nudG8 phenotype. The library had been transformed into Escherichia coli C600 strain using the modified procedure described in the Cloning Kit Instruction Manual (Stratagene Inc., La Jolla, CA). Prehybridization and hybridization of the filters and preparation of the probe was done according to the nonradioactive Genius kit (Boehringer Mannheim Co., Indianapolis, IN). E. coli (DH5α) were made competent and transformed according to Sambrook et al. (1989). The phage DNA was isolated as described in Sambrook et al. (1989) and subcloned into pBluescript II LS+ (Stratagene Inc.), purified through CsCl and sequenced using the 4000L Sequencer (LI-COR, Inc., Lincoln, NB).

Spores were inoculated into YAG medium containing uridine and uracil and grown overnight at restrictive temperature (44°C). The mycelia were harvested by filtration through Miracloth (Calbiochem-Novabiochem, La Jolla, CA), pressed dry, and frozen in liquid nitrogen. The frozen mycelia were ground to a powder with a mortar and pestle in liquid nitrogen. Proteins were extracted by vortexing the powder in a 30-ml capped tube containing 2 ml of acid-washed glass beads (425–600-μm diameter; Sigma Chemical Co.) and Tris/KCl buffer (20 mM Tris HCl, pH 7.6, 50 mM KCl, 5 mM MgCl₂, and 0.5 mM EDTA) (Collins and Vallee, 1989). After a 4°C centrifugation at 3,000 g in an HB-4 Sorvall rotor for 5 min to remove cell wall debris, the supernatant was recentrifuged at 17,000 g in a Sorvall centrifuge (Newtown, CT) for 15 min at 4°C. The protein concentration ranged from 10 to 30 mg/ml as measured by the Bradford assay (Bradford, 1976).

NUDG antiserum was generated against the first 14 amino acids of the CDLC sequence, H2N-MASEKKDKLEPQIK-COOH, conjugated to keyhole limpet hemagglutinin and injected into rabbits by Alpha Diagnostic Int. (San Antonio, TX). Each of two rabbits produced antiserum that recognized the 8-kD CDLC band by Western blotting of protein extracts at 1:15,000 dilution. Because only a single band was detected, we did not find it necessary to affinity purify the NUDG antiserum. The preparation of affinity purified anti–NUDA antibodies was described previously (Xiang et al., 1995a). For immunoprecipitation experiments, we incubated 4 mg of protein extract in 1 ml of Tris/KCl buffer for 2 h with 40 μl of A. nidulans NUDG (CDLC) undiluted antiserum at 4°C on a rotating wheel. 100 μl of 50% Protein A Sepharose (Pharmacia LKB Biotechnology Inc., Piscataway, NJ) was added, and incubation continued for an additional 2 h at 4°C. The mixture was then washed six times with 1 ml Tris/KCl buffer by centrifugation for 4 s in a microfuge. The immunoprecipitate was resuspended in Laemmli sample buffer, subjected to electrophoresis on a 4–20% gradient SDS-PAGE gel (Bio-Rad Laboratories, Richmond, CA) and transferred to an Immobilon-P membrane (Millipore Corp., Bedford, MA). The transfer efficiency was monitored by Ponceau S (Sigma Chemical Co.) staining. After primary antibody and goat anti–rabbit alkaline phosphatase-conjugated secondary antibody incubations, the blots were developed colorimetrically with nitro blue tetrazolium chloride and 5-bromo-4-chloro-3-indolyl-phosphate from Kirkegaard & Perry Laboratories, Inc. (Gaithersburg, MD).

Wild-type and nudG3 mutant cells were grown in YG medium 15 h at 42°C and protein extracts prepared as described above. 1 ml of extract was sedimented through a 12 ml linear 17–51% sucrose density gradient made with the same buffer in which the extract was prepared. The gradient was centrifuged in a SW41 rotor for 4 h at 270,000 g at 4°C. 600-μl fractions were collected from the top of the gradient and 20 μl of each fraction was subjected to gradient (4–20%) SDS-PAGE and transferred to membrane for Western blotting as described above.

A new collection of temperature-sensitive mutants of A. nidulans (see Materials and Methods) was screened by DAPI staining of nuclei and fluorescence microscopy at restrictive temperature (42°C) to identify mutants with a defect in nuclear migration (nud mutants). Several mutants with this phenotype were identified and crossed with previously characterized nud mutants (nudA, nudC, and nudF) to determine whether they represented new genes. One such mutant was identified as a new nud gene because it produced 25% wild-type progeny when crossed with strains carrying either a nudA, nudC, or nudF mutation. This new mutation was designated as nudG8. Colonies of the nudG8 mutant strain grew slowly at restrictive temperature with a growth rate 10–15% that of wild type and a tight colony morphology indistinguishable from that of other previously described Aspergillus nud mutations (Fig. 1) (see also Osmani et al., 1990). DAPI staining of hyphae germinated at restrictive temperature revealed the typical nuclear migration defect characteristic of nud mutants. The nuclei divided normally, but failed to migrate from the spore end of the germling into the germ tube, leaving a cluster of nuclei at the spore end. When crossed with a wild-type strain, the temperature sensitivity of nudG8 segregated 1:1, confirming it was a single gene mutation. Diploids heterozygous for nudG8 grew as well as their wild-type parental strains at restrictive temperature (44°C), indicating that the nudG8 mutation was recessive and therefore presumably a loss of function mutation. NudG8, like other previously characterized nud mutants of A. nidulans, was also defective in the production of asexual spores (conidia). Because the asexual spores are responsible for colony color (Clutterbuck, 1990), this was easily seen by inspection of colonies growing on YAG agar, which were gray-brown rather than brightly colored. Because other previously characterized nud mutants of A. nidulans exhibited low fertility in matings, we compared the ascospores (sexual spore) production of self-crosses of wild-type strains and nudG8 strains. The nudG8 mutation was linked to a reduction in fertility, as demonstrated by the low numbers of ascospores (sexual spores) produced in self-crosses in contrast to wild-type ascospore production. Among eleven ascospores examined, the nudG mutant produced ∼3–10% the number of conidia as the wild type.

To determine whether the nuclear migration defect caused by nudG was mediated by an effect on cytoplasmic dynein or by an effect on some other motility system involved in nuclear migration, we constructed a doubly mutant strain carrying both nudG8 and a ΔnudA CDHC deletion mutation (Xiang et al., 1995a). If the function of the nudG gene product was similar to that of the nudA CDHC, the growth defect of the double mutant should be no more severe than that caused by the more deleterious of the single mutations. Alternatively, if the mutations affected different processes, the effects of the mutations would be expected to be additive and the phenotype of the double mutant more extreme than either single mutant. The result of this experiment was that the nudG8, ΔnudA double mutant grew no more slowly than the worse of the singly mutant parental strains, suggesting that nudG was involved in the same motility function as the nudA-encoded CDHC (Fig. 1).

The nudG8 mutation was mapped to chromosome VIII. A cosmid that rescued the mutant phenotype was identified by DNA-mediated complementation of nudG8 temperature sensitivity using sib selection from pools of chromosome VIII cosmids. The rescuing cosmid, SW27C04, is located on the right arm of chromosome VIII near bimG, abaA, and can67 (Brody et al., 1991; Wang et al., 1994). The 40-kb cosmid clone was subcloned to the smallest fragment, 2 kb, that was able to rescue the nudG8 mutant phenotype. Southern blot analysis using the 2-kb fragment as a probe showed a single band when genomic DNA was cut with BamHI and EcoRI, and two bands when cut with HindIII (data not shown). We later determined that the nudG sequence contains an internal HindIII site, which explains the presence of two bands. Thus, nudG appears to be represented by a single gene in the A. nidulans genome. The 2-kb genomic fragment was used to isolate two cDNA clones of 572 and 619 bp from an A. nidulans lambda gt10 phage cDNA library. The two clones encoded sequences that differed by a 54-bp sequence, possibly the result of a splicing abnormality. Both clones were able to completely rescue the growth and nuclear migration defects of nudG8 by DNA-mediated complementation (data not shown). The sequence data for the 572-bp cDNA clone is available from GenBank/EMBL/DDBJ under accession number U81827.

The nudG 572-bp cDNA encodes a putative protein of 94 amino acids with a predicted molecular weight of 11,031 D and an isoelectric point of 6.49. It shares 67% identity with the 8-kD outer arm CDLC from Chlamydomonas reinhardii flagella (King and Patel, 1995), 66.3% with the Caenorhabditis elegans cosmid T26A5 sequence (GenBank accession number U00043), the CDLC1 of Drosophila melanogaster (Dick et al., 1996b), the Rattus norvegicus protein inhibitor of neuronal nitric oxide synthase (Jaffrey and Snyder, 1996) and the Homo sapiens CDLC1 (Dick et al., 1996b), 53.9% with the Schistosoma mansoni CDLC (Hoffmann and Strand, 1996), and 43.5% identity with the SLC1 of Saccharomyces cerevisiae (Dick et al., 1996a) (Fig. 2). An antibody was generated against the first 14 amino acids of the wild-type nudG CDLC sequence. Western blotting of extracts from wild-type strains identified a NUDG protein band migrating at 8 kD on SDS-PAGE gels even though the molecular weight predicted by the sequence is 11 kD. This is similar to the anomalous migration of CDLC protein from other species. The 8-kD CDLC band was absent from Western blots of nudG8 protein extracts grown at restrictive temperature (Fig. 3 A). Thus, the effect of the nudG8 mutation is either to increase the rate of degradation of the NUDG protein, to decrease its rate of synthesis, or both. We attempted to use our antibody to localize the nudG light chain within the A. nidulans germling; however, this antibody was not useful for immunocytochemistry.

To determine whether the nudG CDLC interacts with the CDHC in A. nidulans, we used antibodies against the nudG CDLC and the nudA CDHC to immunoprecipitate proteins from cell-free extracts and analyzed the immunoprecipitates by Western blotting (Fig. 4). The NUDG antiserum precipitated both the nudG 8-kD CDLC and the nudA CDHC from wild-type extracts. Similarly, antibody against the nudA CDHC also precipitated both the CDHC and the 8-kD CDLC from wild-type extracts. Control experiments without antiserum or with protein extracts prepared from a nudA1 mutant grown at restrictive temperature, which lacks CDHC protein, or the nudG8 mutant grown at restrictive temperature, which lacks the 8-kD CDLC protein, demonstrated the specificity of the interaction. In the absence of NUDA protein (the nudA1 strain), antibody against NUDA failed to precipitate NUDG protein and, in the absence of NUDG protein (the nudG8 strain), antibody against NUDG failed to precipitate NUDA protein. These coimmunoprecipitation data show that nudG CDLC and nudA CDHC are associated in a protein complex in A. nidulans. They do not, however, necessarily indicate that the CDHC is bound directly to the light chain. The association between the nudG 8-kD CDLC and the nudA CDHC could be mediated by another chain of cytoplasmic dynein or by some as yet uncharacterized protein. Nor does this result necessarily mean that all of the CDLC is bound to the CDHC or vice versa. In fact, sedimentation analysis (see below) indicates that most of the 8-kD light chains are not associated with the CDHC.

Sedimentation of cell free extracts of wild-type A. nidulans showed that the NUDA cytoplasmic heavy chain protein sediments as a large molecular weight complex at ∼20 S. The NUDG CDLC for the most part sedimented faster than expected from its calculated molecular weight and exhibited a very broad distribution, suggesting that it was involved in multiple protein complexes. Although there was an overlap between the sedimentation profiles of the NUDA and NUDG proteins, most of the NUDG protein did not cosediment with the 20-S cytoplasmic dynein complex (Fig. 5), indicating that the major fraction of NUDG is not associated with the cytoplasmic dynein complex under the conditions of this experiment. Sedimentation of extracts from nudG8 mutant cells grown at 42°C produced NUDA and NUDG sedimentation profiles that were indistinguishable from that of wild-type extract grown under the same conditions (data not shown). Thus, the nudG8 mutation has no observable effect on the structure of the cytoplasmic dynein complex as detectable by this assay.

We have shown that cytoplasmic dynein is concentrated at the growing tips of wild-type A. nidulans germlings by immunocytochemistry, using an affinity purified rabbit polyclonal antibody directed against the CDHC (Xiang et al., 1995a), and more recently by observation of a green fluorescent protein tagged–CDHC fusion protein in living cells (Xiang et al., 1997). To determine whether the nudG8 mutation affected the localization of cytoplasmic dynein, we stained nudG8 germlings with affinity purified anti– NUDA (anti–CDHC) antibody. In wild-type germlings, antibody against the CDHC stains a dot at the hyphal tips at both permissive and restrictive temperatures. A similar dot of stain is also seen at the hyphal tips of nudG8 germlings at permissive temperature. At restrictive temperature, however, the dot of CDHC at the tip disappears (Fig. 6), and the anti–NUDA antibody produces only a faint, diffuse staining throughout the cytoplasm. Because the loss of CDHC staining from the tip might represent a decrease in the total amount of heavy chain in the cells, we compared the amount of CDHC in wild-type and nudG8 cells by Western blotting. Equal amounts of protein were loaded on the gels. Western blot analysis of protein extracts from wild-type and mutant cells revealed that the CDHC protein level was similar in the wild-type and nudG8 mutant strains (Fig. 3,B). Another potential source of artifact was the possibility that the nudG8 mutation might affect the accessibility of the tip to antibody. To control for this possibility, we stained hyphae with antibodies against actin and tubulin. Both antibodies stained structures at the hyphal tip. We found no discernible difference in actin or tubulin staining at the tips of the nudG8 mutant as compared with wild-type cells (data not shown). Thus, the nudG8 mutation did not cause a difference in CDHC tip staining either by altering the amount of CDHC in the cell or by altering tip permeability to antibody, but caused a loss of CDHC protein from the hyphal tip. Because the 8-kD CDLC band was absent from protein extracts of nudG8 grown at restrictive temperature (Fig. 3 A), these data suggest that the nudG 8-kD CDLC is required for targeting of the nudA CDHC to the hyphal tip in A. nidulans. We were unable to determine whether the 8-kD CDLC was located at the hyphal tip in wild-type cells because our anti–light chain antibody did not stain the mycelium.

Here we report on the cloning and characterization of a new gene, nudG, required for nuclear migration in A. nidulans. The wild-type gene was cloned by DNA-mediated complementation of the temperature-sensitive nudG8 mutation and encodes a homologue of the 8-kD cytoplasmic dynein light chain protein. Experiments to determine how the nudG8 mutation affects cytoplasmic dynein showed that the nudG8 mutation caused a large decrease in the intracellular concentration of the 8-kD CDLC when the mutant fungus was grown at restrictive temperature. This decrease had two observable effects on the cell. It inhibited nuclear migration through the mycelium, and it caused CDHC staining to disappear from the hyphal tip. Western blotting showed that the loss of tip staining was not caused by a decrease in the intracellular concentration of CDHC, as the total amount of CDHC protein was not reduced in the mutant at restrictive temperature. Nor was the failure of anti–CDHC antibody to stain the tip caused by decreased antibody permeability, since anti–actin and anti– tubulin antibodies stained the tip of the nudG8 mutant as well as they stained the wild-type tip. Thus, altered CDHC tip staining caused by the nudG8 mutation apparently reflects a real change in the distribution of CDHC within the cell.

The effect of the nudG8 mutation at restrictive temperature is to cause a decrease in the amount of 8-kD CDLC in the cell. Whether this is the result of decreased synthesis or increased degradation is unknown. Precisely how the deficiency of 8-kD CDLC affects tip localization and nuclear migration is also unknown. Among the possibilities, the 8-kD CDLC could be required for assembly of the cytoplasmic dynein complex, it could be required for the interaction of dynein with dynactin, or it could be required for dynein motor function per se. Any of the aforementioned defects might prevent dynein from reaching and becoming localized at the hyphal tip. The fact that the nudG8 mutation does not affect the sedimentation of the CDHC, however, indicates that the lack of NUDG protein does not cause a gross disruption of the cytoplasmic dynein complex. Alternatively, the 8-kD CDLC might be involved in anchoring the dynein/dynactin complex to the hyphal tip. Whatever the mechanism, our results are consistent with a model in which cortically anchored dynein moves nuclei by pulling on astral MTs attached to nuclei (Palmer et al., 1992; Plamann et al., 1994; Morris et al., 1995; Carminati and Stearns, 1998; Efimov and Morris, 1998). However, it is important to recognize that they by no means prove the model.

In higher eukaryotes, cytoplasmic dynein is bound to vesicles (Sheetz, 1996; Vallee and Sheetz, 1996; Hirokawa, 1998). The tips of fungal hyphae are packed with vesicles, and the CDHC tip staining in the wild type could simply represent vesicle-associated dynein. Were this the case, decreased tip staining in the nudG8 mutation could result from any CDLC requirement for dynein translocation of vesicles to the tip. Interpretation of the role of the nudG CDLC is further complicated by the fact that the 8-kD CDLC has recently been shown to be associated with myosin-V (Espindola et al., 1996), with I kappaB alpha (Crepieux et al., 1997), and with the tegument of the blood fluke, Shistosoma (Hoffman and Strand, 1996). It has also been shown to act as an inhibitor of the neuronal form of nitric oxide synthase (Jaffrey and Snyder, 1996). The difficulty of definitively understanding the molecular function of the CDLC protein in higher eukaryotes is exemplified by phenotypic analysis of mutations in genes for the 8-kD CDLC in Drosophila melanogaster and Chlamydomonas reinhardii. Mutations in the Drosophila gene (known as DLC1 or cut up) cause axon projections to follow abnormal pathways in the central nervous system (Phillis et al., 1996). Other partial loss of function mutations in DLC1 cause disorganization of the ovaries and female sterility. Total loss of function mutations in D. melanogaster are lethal. Because the 8-kD protein has been shown to interact with myosin-V, I kappaB alpha, and nitric oxide synthase, these interesting phenotypes do not necessarily implicate the 8-kD light chain in cytoplasmic dynein function in D. melanogaster. In C. reinhardii, the 8-kD light chain has been shown to be essential for the retrograde intraflagellar transport of migrating “rafts” by a mechanism unrelated to its flagellar dynein function and therefore is likely to be specific for cytoplasmic dynein (Pazour et al., 1998), but an interaction with one of the other molecules cannot rigorously be excluded. In principle, the effect of nudG8 on tip staining and nuclear migration in A. nidulans could similarly be mediated through myosin-V or the nuclear factor– κB transcription factor rather than via cytoplasmic dynein. An effect mediated via nitric oxide synthase is unlikely, as no NO synthase activity has been detected in A. nidulans (Beckwith, S.M., N.R. Morris, and D. Wolff, unpublished observations). Because the nudG8 CDLC mutant accurately phenocopies the nuclear migration, colony growth, and sporulation defects of nudA CDHC mutants, it seems probable that nudG CDLC function is mediated via cytoplasmic dynein. The fact that the phenotype of the nudA, nudG double mutant is the same as that of the single mutations provides additional evidence that nudG functions via dynein.

In summary, the characterization of the nudG 8-kD CDLC of A. nidulans has extended the evolutionary range of this dynein light chain to the filamentous fungi, has demonstrated that it is an essential component of the main machinery responsible for nuclear migration, and has provided strong evidence that its effect is mediated through a physical interaction with the cytoplasmic dynein complex.

We thank S. Harris and J. Hamer for sharing their collection of temperature sensitive A. nidulans mutants with us. The nudG mutation was isolated from this collection. We also thank Peter D'Arpa and Xin Xiang for constructive comments on this manuscript.

This work was supported by grants to N.R. Morris, S.M. Beckwith, and B. Liu from the National Institutes of Health and to N.R. Morris from the American Cancer Society.

CDHC

cytoplasmic dynein heavy chain

CDLC

cytoplasmic dynein light chain

DAPI

4′,6-diamidino-2-phenylindone

MT

microtubule

SPB

spindle pole bodies

ts

temperature sensitive