Arp11 Affects Dynein–Dynactin Interaction and is Essential for Dynein Function in Aspergillus nidulans

Abstract

The dynactin complex contains proteins including p150 that interacts with cytoplasmic dynein and an actin-related protein Arp1 that forms a minifilament. Proteins including Arp11 and p62 locate at the pointed end of the Arp1 filament, but their biochemical functions are unclear (Schroer TA. Dynactin. Annu Rev Cell Dev Biol 2004;20: 759–779). In Aspergillus nidulans, loss of Arp11 or p62 causes the same nuclear distribution (nud) defect displayed by dynein mutants, indicating that these pointed-end proteins are essential for dynein function. We constructed a strain with S-tagged p150 of dynactin that allows us to pull down components of the dynactin and dynein complexes. Surprisingly, while the ratio of pulled-down Arp1 to S-p150 in Arp11-depleted cells is clearly lower than that in wild-type cells, the ratio of pulled-down dynein to S-p150 is significantly higher. We further show that the enhanced dynein–dynactin interaction in Arp11-depleted cells is also present in the soluble fraction and therefore is not dependent upon the affinity of these proteins to the membrane. We suggest that loss of the pointed-end proteins alters the Arp1 filament in a way that affects the conformation of p150 required for its proper interaction with the dynein motor.

The dynactin complex interacts with cytoplasmic dynein and is critical for a variety of dynein functions in vivo (1). Recently, mutations in dynactin have been implicated in motoneuron degenerative diseases (2), further indicating the medical relevance of this complex and the importance of dissecting the functions of individual components.

The structure of the dynactin complex has been revealed by electron microscopy-based studies (3–6). Within the dynactin complex, Arp1 forms an actin-like minifilament (3). The p150^Glued, the p50 dynamitin and p24 subunits form a shoulder/sidearm complex (4). The p150^Glued subunit of dynactin interacts directly with the intermediate chain (IC) of cytoplasmic dynein (7,8). Its N-terminus contains one cytoskeleton-associated protein (CAP)-Gly microtubule-binding domain that allows it to interact with microtubules (9–11) and a downstream microtubule-binding basic domain that increases dynein processivity in vitro (12,13). However, a recent study suggests that the microtubule-binding region may not be important for dynein-mediated vesicle transport in Drosophila S2 cells (14), and this would need to be tested in other systems. Dynactin may bind to membranous cargoes via Arp1’s physical interaction with spectrins (15–17) or an interaction between the Zeste-White 10 (ZW10) and the p50 subunit of dynactin that locates on top of the Arp1 filament (18). However, whether the intact Arp1 filament is important for recruiting dynein to membranous cargoes has also been questioned (19) and would need further studies. One end of the Arp1 filament is capped by the capping proteins, which are barbed-end actin-binding proteins involved in regulating actin polymerization (1,20). The other end (presumably the pointed end) binds to the pointed-end complex that contains p62, p27, p25 subunits and an actin-related protein, Arp11 (4,21–23). Arp11 interacts with Arp1 directly (24,25), but the biochemical functions of Arp11 and other pointed-end proteins in the dynactin complex remain to be investigated.

The pointed-end proteins of dynactin have been studied in filamentous fungus Neurospora crassa and the budding yeast Saccharomyces cerevisiae (23,25,26). The null mutants of Arp11 and p62 in N. crassa (ro-7 and ro-2, respectively) display a similar nuclear distribution phenotype as exhibited by a dynein null mutant (23,26), suggesting that they are essential for dynein function. Furthermore, biochemical analyses suggested that pointed-end proteins negatively regulate the physical interaction between dynein/dynactin and membranes (23). In S. cerevisiae, the Arp11 homologue Arp10 is a component of the dynactin complex, but null mutants of Arp10 do not show any obvious defect in spindle positioning, suggesting that Arp10 is not essential for dynein function (24,25,27). However, Arp10 becomes important for dynein function when Arp1 is defective (25,28). Arp10, Arp1 and Jnm1 (homologue of the p50 dynamitin) may form a ternary complex, and Arp10 seems important for Arp1–Arp1 and Arp1–Jnm1 interactions based on yeast two-hybrid analyses. Finally, a genetic assay for Arp1’s membrane association indicates that Arp10 also negatively regulates the interaction between Arp1 and the plasma membrane (25). However, the mechanism involved in the negative regulation of membrane interaction is not clear.

Here, we found that in the filamentous fungus Aspergillus nidulans, both Arp11 and p62 are important for dynein’s microtubule-plus-end accumulation and nuclear distribution, suggesting that they are essential for dynein function. To further understand their biochemical functions, we constructed a strain with S-tagged p150 of dynactin (NUDM: nuclear distribution protein M) that not only allows us to pull down components of the dynactin complex such as p150 and Arp1 (NUDK) but also allows us to pull down components in the cytoplasmic dynein complex such as dynein heavy chain (HC or NUDA) and IC (or NUDI) under low-salt conditions. Interestingly, depletion of Arp11 decreases the amount of Arp1 pulled down by the S-tagged p150 but significantly increases the amount of dynein HC and IC pulled down. We further show that the enhanced dynein–dynactin interaction in the Arp11 mutant is independent of the interaction between dynein/dynactin and the membrane. These data suggest that loss of pointed-end proteins may alter the Arp1 filament, thereby affecting p150’s conformation involved in its interaction with the dynein motor. Whether the abnormally enhanced dynein–dynactin interaction may cause defects in motor function, for example, a deficient recycling of dynein/dynactin from membranous cargoes, would need to be tested in the future.

Results and Discussion

Arp11 and p62 in A. nidulans are essential for dynein function

The full-length genomic sequence of the A. nidulans Arp11 gene (AN3185) was found from the annotated A. nidulans genome (http://www.broad.mit.edu/annotation/fungi/aspergillus/) (29) by using the N. crassa Arp11 homologue, Ro7, as a query sequence (23). The A. nidulans Arp11 protein is a protein with 554 amino acids, which shows weak sequence similarity to the mammalian Arp11 and the Ro7 protein (Figure S1; 4,23). Arp11 sequences seem quite diverged even among fungi with only about 15% sequence similarity and 7% identity between the A. nidulans Arp11 and Ro7. Further evidence for the identity of this protein came from phenotypic analysis, localization and biochemical studies (see subsequently). To analyze the function of Arp11 in A. nidulans, we constructed a deletion mutant of Arp11 (ΔArp11) in which the coding region is deleted so that it contains only sequences encoding the first six amino acids and the last nine amino acids. The deletion mutant of A. nidulans Arp11 exhibits exactly the same phenotype as previously characterized nud mutants in the dynein pathway (Figure 1). It formed compact colonies on plates, and the nuclei are clustered at the spore end of the germ tube (Figure 1). The nud phenotype was also observed in the alcA promoter-based conditional Arp11 null mutant, alcA-Arp11, on glucose that shuts off the expression of the Arp11 protein (Figure 2A).

Figure 1. Phenotype of the ΔArp11 mutant.

A) Deletion of the Arp11 gene in Aspergillus nidulans causes the same colony growth defect as that exhibited by the null mutant of nudA that encodes dynein HC (65). Cells were grown on YUU medium at 32°C for 3 days. B) DAPI staining showing that the ΔArp11 mutant exhibits a typical nud phenotype after an overnight incubation at 32°C. Bar, approximately 5 μm.

Figure 2. The alcA-based Arp11, p62, p50 and Arp1 mutants all show a typical nud phenotype when the alcA promoter is shut off by glucose.

A) DAPI staining of cells grown on YUU medium at 32°C overnight. Bar, approximately 5 μm. B) Percentage of cells showing the nud phenotype (defined by the presence of a cluster of four or more nuclei). More than 200 cells were counted for each mutant.

The A. nidulans p62 homologue (AN4917) was found from the annotated genome by using the N. crassa p62 homologue, Ro2, as a query sequence (23,26). It shows sequence similarity to the mammalian p62 and the Ro2 protein in N. Crassa (Figure S2; 4,21–23,26). Sequence identity and similarity between A. nidulans p62 and Ro2 are 40 and 52%, respectively. During an effort to characterize several nud mutants whose defective gene products had not been previously identified (30), we found that the DNA fragment containing the coding region of the A. nidulans p62 homologue completely rescued the phenotype of the temperature-sensitive (ts) nudR825 mutant. To confirm that the p62 homologue is indeed the gene for nudR, we made an alcA-based conditional null mutant of p62, which also exhibited a nud phenotype on glucose (Figure 2A). We crossed this mutant to the nudR825 mutant and found no ts+ progeny out of 900 progenies, indicating that the gene encoding the p62 homologue and the nudR gene are at the same locus, and thus, the A. nidulans p62 homologue is the nudR gene product. Together, these results showed that both Arp11 and p62 are essential for dynein functions in A. nidulans.

In this study, we also constructed alcA-based conditional mutants of Arp1 (AN1953) and p50 (AN3589) in the dynactin complex, and they all exhibited a nud phenotype on the repressive glucose medium as expected (Figure 2A). Because the spores are always collected from colonies on the non-repressive glycerol medium and then inoculated on glucose medium, a minor portion of the cells (especially young germlings) does not exhibit a clear nud phenotype after an overnight incubation on glucose. We quantified the percentage of cells that exhibited a nud phenotype (defined by the presence of a cluster of four or more nuclei) and found that more than 70% of the alcA-based Arp11, p62, Arp1 and p50 mutant cells exhibited a nud phenotype after an overnight incubation on glucose (Figure 2B).

In A. nidulans, green fluorescent protein (GFP)-labeled dynein/dynactin components form motile comet-like structures at the hyphal tip representing their accumulation at the dynamic microtubule plus ends (31–35). GFP–Arp11 also formed the same comet-like structures (Movie S1), consistent with it being a dynactin component. Our previous studies using loss-of-function mutants of nudK (Arp1) and nudM (p150) suggested that the dynactin complex is important for dynein’s microtubule-plus-end localization (33,36). In this study, we found that depletion of Arp1 or p50 also dramatically lowered dynein’s plus-end accumulation (Figure 3A and Movies S2, S5 and S6), consistent with our previous conclusion.

Figure 3. Localizations of dynein and dynactin components in various dynactin mutants.

A) Downregulation (or depletion) of Arp11, p62, p50 and Arp1 dramatically decreases the microtubule plus-end accumulation of GFP-NUDA (or GFP–HC), which is represented by the comet-like structures at the hyphal tip in wild-type cells (see Movies S2–S6 for images of multiple cells in the same field). In Arp11- or p62-depleted cells (alcA-Arp11 or alcA-p62), GFP–HC accumulates at dots along filament-like structures (Movies S3 and S4). The GFP-nudA (HC) fusion gene is driven by nudA’s endogenous promoter. B) Besides the dot along filament-like structures, GFP–HC also accumulates at or close to septa in alcA-Arp11 mutant cells on glucose (arrow points to a septum). Cells were grown on minimal glucose medium overnight at room temperature and shifted to 32°C for about 5 h. C) The alcA-driven GFP–HC, GFP–p150 and GFP–p50 all accumulate at or close to septa in the ΔArp11 mutant grown on glycerol overnight at 32°C. But the alcA-driven GFP–p62 is not seen to accumulate at any specific place in the ΔArp11 mutant. Bar, approximately 5 μm.

In Arp11-depleted cells, most of the GFP–dynein signals highlighted dots along filament-like structures instead of accumulating at the microtubule plus end (Figure 3A and Movie S3). These dots were also present in p62-depleted cells although they were less obvious than that in Arp11-depleted cells (Figure 3A and Movie S4). To address whether these dots are associated with microtubules, we treated the cells with benomyl of 2.4 μg/mL for 2 h. The reason behind this long treatment was that a few cytoplasmic microtubules were still present after a 1-h treatment (Figure S3). After a 2-h treatment, the dots were no longer observed along filament-like structures, but rather, they appeared to be associated with the nuclei (Figure S3). Although we do not yet know the nature of these structures, we suspect that they may represent dynein-bound cargoes along microtubules. These signals were not observed in Arp1- or p50-depleted cells, and thus, dynein’s accumulation to these structures may be specifically enhanced by depletion of the pointed-end proteins. In Arp11-depleted cells, GFP–dynein proteins were also observed near septa as dots or short filament-like structures (Figure 3B). Although the septum localization of GFP–dynein was also observed in wild-type cells (37), the signals in the Arp11 mutant were much more striking. Upon benomyl treatment, the septum-associated signals in the alcA-Arp11 mutant were still present (Figure S3). These bright, septum-associated signals were not observed in p62-depleted cells (data not shown), and thus, the accumulation may be specifically caused by the absence of Arp11.

Because the function of p150 dynactin is required for dynein localization (33), we tested the possibility that p150 might be mislocalized, which in turn causes dynein to be mistargeted. Because the GFP-NUDM (p150) fusion was under the control of the alcA promoter (33), we introduced the fusion into the ΔArp11 strain background (instead of the alcA-Arp11 background) by genetic crosses and observed the desired progeny on glycerol medium that allows the fusion to be expressed. To observe dynein localization under similar conditions, we also introduced the alcA-controlled GFP-NUDA fusion (36) into the ΔArp11 background by genetic crosses and observed it on glycerol. In the complete absence of Arp11, the microtubule-plus-end localization of dynein was not observed although the microtubule network appeared normal (data not shown). To our surprise, the dots along filament-like structures observed in the alcA-Arp11 mutant were not seen in the ΔArp11 mutant, suggesting that these dots may only appear at a certain time frame after Arp11 was depleted. However, GFP–dynein in the ΔArp11 mutant did localize to short filament-like structures near the septa as similarly observed in Arp11-depleted cells (Figure 3C). As expected, GFP–p150 also localized to similar structures near septa (Figure 3C), suggesting that loss of Arp11 also causes dynactin to be mislocalized. Although we cannot rule out the possibility that these structures may represent dynein/dynactin aggregates, the presence of these structures allowed us to assess whether other dynactin subunits are also mislocalized to similar structures. To this end, we tested GFP-labeled p62 and p50 subunits that are functional (Movies S7 and S8). We introduced the alcA-controlled GFP–p50 or GFP–p62 fusion into the ΔArp11 mutant by genetic crosses, and the desired progeny was identified by phenotypic and polymerase chain reaction (PCR) analyses. Interestingly, while GFP–p50 also exhibited mislocalization to similar structures near the septa in the ΔArp11 mutant (Figure 3C), GFP–p62 proteins were not found in these structures, but rather, they show only faint background fluorescence in the ΔArp11 cells (Figure 3C). These results suggest that deletion of Arp11 may displace p62 but not p50 from the dynactin complex.

Depletion of A. nidulans Arp11 enhances dynein–dynactin interaction and decreases the amount of Arp1 associated with p150

In Arp1-depleted cells, the protein level of p150 (NUDM) was significantly decreased (Figure 4). This was not a surprise because a similar result was obtained from a previous study on an Arp1-null mutant in N. crassa (38). In addition, a recent study in Drosophila also indicates that Arp1 depletion negatively affects the level of p150^Glued (19). Results from mammalian cells, however, have shown that p150^Glued is stable when it is separated from the Arp1 filament by overexpression of p50 dynamitin (39,40). A possible explanation for this discrepancy is that the initial folding of p150^Glued may depend upon the presence of Arp1, and after it is folded properly, the separation from the Arp1 filament may not affect its stability. Unlike Arp1 depletion, neither Arp11 depletion nor p62 depletion caused any significant decrease in the protein level of p150 (Figure 4). Because the mutants we used for protein depletion were all alcA-based conditional mutants, one possible cause of the difference in the effects of p150 could be the difference in protein stability, for example, Arp11 may be more stable than Arp1 and thus more functional Arp11 proteins were present when the cells were harvested for protein isolation. To address this concern, we included the Arp11 deletion mutant in the analysis and found that in the total absence of Arp11, the protein level of p150 was still close to normal and obviously much higher than that in Arp1-depleted cells (Figure 4).

Figure 4. Arp1 depletion but not Arp11 or p62 depletion significantly decreases the protein level of p150.

Total protein extracts from cells grown on glucose overnight at 32°C were loaded on the gel and probed with an affinity-purified anti-p150 antibody. PONCEAU S staining of the blot shows similar loading of proteins in each lane.

To determine the biochemical effects of Arp11 on the dynactin complex, we first generated a strain, S-p150, in which the C-terminus of NUDM (p150) is tagged with the S-tag that would allow affinity purification of dynactin from total protein extract. In this strain, the S-tagged nudM integrated into the nudM locus and replaced the endogenous nudM gene. This strain grows similarly to a wild-type strain on plates and has a normal nuclear distribution pattern (Figure 5A). Furthermore, GFP–dynein HCs form comet-like structures, indicating that they localize to microtubule plus ends just like in wild-type cells (Figure 5A). Thus, we believe that the S-tagged nudM is functional. After a one-step affinity purification using the S-protein agarose beads, S-tagged p150 was significantly enriched as evidenced by Western blot analysis (Figure 6B).

Figure 5. The S-tagged p150 is functional and is associated with other dynactin components such as Arp11 and Arp1.

A) The S-tagged p150 is functional. The strain with S-p150 forms a wild-type-like colony after a 3-day incubation at 32°C (top). It has a normal nuclear distribution pattern (middle) and exhibits a normal dynein localization pattern (the bottom panel shows images of dynein HC localization) after an overnight incubation at 32°C. Bar, approximately 5 μm. B) S-p150 pulls down Arp11, Arp1 and a low level of dynein HC. Because GFP–Arp11 is under the control of the alcA promoter, glycerol medium was used to allow expression of GFP–Arp11 and glucose medium was used to shut off its expression. We adjusted loading to make the levels of S-p150 appear similar in the two lanes. Cells were grown at 32°C overnight.

Figure 6. Biochemical analyses of the alcA-Arp11 or alcA-p62 mutant grown on glucose.

A) Western blots showing a typical purification result. In this experiment, similar amounts of total proteins from different strains were used for affinity purification. Note that the yield of purified S-tagged p150 is significantly decreased in the alcA-Arp11 mutant. B and C) Western blots showing typical purification results. Because depletion of Arp11 or p62 decreases the yield of purified S-tagged p150, we adjusted loading so that the intensities of purified S-tagged p150 in all lanes appear similar. It was obvious that S-tagged p150 pulled down more dynein HC or IC when either Arp11 or p62 was depleted. This was not because of increased levels of HC and IC in the total protein extracts isolated from Arp11- or p62-depleted cells (B and D). In contrast, the amount of Arp1 pulled down was obviously decreased in the mutants (C). E) A quantitative analysis on the effects of Arp11 depletion. Values were all relative to the wild-type values, which were all set at 1. The mean and standard error values were calculated from three independent experiments. Note that the intensity ratio of HC to S-tagged p150 is significantly increased (p < 0.05), while that of Arp1 to S-tagged p150 is significantly decreased (p < 0.001) in the mutant. Cells were all grown on YUU medium overnight at 32°C.

Thus, the C-terminus of p150 is exposed, as has been previously speculated (1). We then introduced the alcA-Arp11 allele into a strain with S-p150 by genetic crosses followed by selection of a strain with a nud phenotype on glucose and the presence of the S-tag. Because GFP is fused to Arp11 in this strain and the GFP–Arp11 fusion is under the control of alcA, we used an anti-GFP antibody to determine whether Arp11 could be co-purified with S-p150. As expected, GFP–Arp11 was detected in the eluate when the strain was grown on the glycerol-containing non-repressive medium, but its level was dramatically decreased when the strain was grown on the repressive medium with glucose (Figure 5B). Thus, glucose medium can be effectively used for Arp11 deletion in the biochemical experiments. We found that although depletion of Arp11 had no apparent effect on the total protein level of p150, it lowered the yield of purified S-p150 (Figure 6A). This result suggested that without an intact pointed-end complex, the conformation of p150 might be subtly altered, thereby affecting the interaction between the C-terminal S-tag and the S-protein.

Within the dynactin complex, the p150^Glued subunit has been implicated in binding dynein through its direct physical interaction with the dynein IC (1,7,8). Under our experimental conditions, low levels of dynein HC and dynein IC were pulled down by the S-tagged p150 (Figures 5B and 6A–C). Interestingly, the ratio of pulled-down dynein to S-p150 is significantly higher when Arp11 is depleted (p < 0.05; Figure 6A–C,E). This is not because of increased levels of dynein HC and IC in the total protein extracts (Figure 6B,D). Similar results were also obtained in the alcA-p62 mutant (Figure 6B,C). Together, these results indicate that the physical interaction between dynein and dynactin complexes is enhanced in the absence of an intact pointed-end complex.

To further confirm this result, we examined whether Arp11 depletion similarly causes an increased dynein–dynactin interaction in the strain background of an S-tagged dynein IC (S-IC). In wild-type cells, S-IC pulled down dynein HC and a low level of p150 (Figure S4). We introduced the conditional null allele of Arp11 into the S-IC strain by genetic crosses and allowed the strain to grow on glucose-containing repressive medium. Under the same experimental conditions, the amount of p150 pulled down was also increased when Arp11 was depleted (Figure S4), further supporting the conclusion that loss of Arp11 enhances dynein–dynactin interaction.

The physical interaction between dynein and dynactin is most likely regulated because although these two complexes interact, they are not isolated as a single complex. While phosphorylation of the dynein IC negatively regulates this interaction (41), other regulatory mechanisms may exist in vivo. The pointed-end proteins are more likely to modulate dynactin rather than dynein to affect dynein–dynactin interaction. In p150^Glued, there could be multiple sites within the first 811 amino acids that mediate its interaction with the dynein IC (1,7,8,42). However, based on the known structure of the dynactin complex (1), it is unlikely that the pointed-end proteins affect these sites directly. Electron microscopy studies indicated that there is considerable flexibility associated with p150^Glued structure (4), which makes it possible that a conformational change in other regions of the protein may be transmitted to the sites involved in dynein binding.

The C-terminal portion of p150^Glued contains the second coiled-coil motif implicated in binding to the Arp1 filament (9). This is consistent with the result that p150^Glued from the Drosophila Glued mutant missing the C-terminal portion of the protein fails to be assembled into the dynactin complex (43). In N. crassa, a p150 mutant missing the C-terminal portion also exhibits an enhanced dynactin–membrane interaction (44), which is similar to the effect caused by the pointed-end mutations (23). In the dynactin complex, the binding between the p150^Glued and the Arp1 filament may also be enhanced by p50 dynamitin, which is elongated and forms a tetramer in the dynactin complex (4,39,40,45–47). Because Arp11 and its yeast homologue Arp10 interact with Arp1 and p50 dynamitin (or Jnm1 in S. cerevisiae) directly (24,25) and may enhance both Arp1–Arp1 and Arp1–Jnm1 interactions based on yeast two-hybrid analyses (25), we sought to determine whether interaction between Arp1 and p150 is weakened in Arp11-depleted cells.

Under our experimental conditions, depletion of Arp11 indeed decreased the amount of Arp1 pulled down by S-p150, and a similar result was also obtained in p62-depleted cells (Figure 6C). The ratio of pulled-down Arp1 to S-p150 in the Arp11 mutant was significantly lower than that in wild-type cells (p < 0.001, Figure 6E). This result, which has also been recently obtained from the Arp10 mutant in S. cerevisiae (27), is consistent with a weakened interaction between p150 and the Arp1 filament. Alternatively, because multiple Arp1 subunits are present in the Arp1 filament, this result could also be explained by a shortened Arp1 filament with fewer subunits. These possibilities would need to be addressed in the future. It also remains to be tested whether the change in the Arp1 filament may result in a conformational change in p150 to increase its affinity for dynein.

Because an increased dynein/dynactin–membrane interaction in the absence of Arp11 homologues was observed in other fungi (23,25), we thought that the enhanced dynein–dynactin interaction in the Arp11 mutant could simply be because of the possibility that the affinity between dynein and dynactin is higher in the membrane-bound fraction compared with that in the soluble fraction. To address this possibility, we examined dynein–dynactin interaction in a high-speed (100 000× g) supernatant obtained from a protein extract isolated without the addition of any detergent. Dynein and dynactin in this high-speed supernatant are considered to be in the soluble fraction. A similarly prepared high-speed supernatant fraction obtained from a protein extract isolated in the presence of 0.4% Triton-X-100 was used as a control, and in this case, both soluble dynein/dynactin and some membrane-bound dynein/dynactin may be present in the supernatant. We found that in wild-type cells, dynein and dynactin interacted in the high-speed supernatant regardless of whether detergent was present or not (Figure 7). This result is consistent with an early conclusion obtained from a study in N. crassa that both in the soluble fraction and in the membrane-bound fraction, a portion of dynein and dynactin interact with each other (44). In Arp11-depleted cells, a significantly enhanced dynein–dynactin interaction was observed in the soluble fraction (p < 0.01, Figure 7). Thus, this enhanced affinity between the two complexes does not depend on dynein/dynactin’s membrane binding. In addition, in the soluble fraction, the level of Arp1 pulled down was also significantly lowered in the Arp11-depleted cells compared with that in wild-type cells (p < 0.001, Figure 7), suggesting that the effect of Arp11 depletion on Arp1 was also independent of dynactin’s membrane binding. The possible effects of Arp11 depletion on the Arp1 filament and on dynein–dynactin interaction are summarized in Figure 8.

Figure 7. The enhanced dynein–dynactin interaction and the decreased Arp1 association with p150 in Arp11-depleted cells could be detected in the soluble fraction.

A) Western blots showing typical purification results. Cell extracts in the absence (left) and presence (right) of 0.4% Triton-X-100 as detergent were used as starting materials and supernatants from a 100 000× g spin were collected for the S-tag-based purification. B) A quantitative analysis on the effects of Arp11 depletion. Values were all relative to the wild-type values, which were all set at 1. The mean and standard error values were calculated from four independent experiments. Note that in the absence of detergent, the intensity ratio of HC to S-tagged p150 is significantly increased (p < 0.01), while that of Arp1 to S-tagged p150 is significantly decreased (p < 0.001) in the mutant. Similarly, a significantly increased intensity ratio of HC to S-tagged p150 (p < 0.005) and a significantly decreased intensity ratio of Arp1 to S-tagged p150 (p<0.001) in the mutant were also observed in the presence of the detergent (bottom). Cells were grown on YUU medium overnight at 32°C.

Figure 8. A diagram illustrating a model that a loss of pointed-end proteins such as Arp11 and p62 enhances the interaction between dynein and dynactin but weakens the interaction between p150 and the Arp1 filament.

Drawing of the wild-type dynactin complex is mainly based on the dynactin complex depicted in Schroer (1). Our current data support the idea that the extreme C-terminus of p150 is exposed. For simplicity, dynactin subunits such as p24, p25, p27 and capping proteins are not shown (note that these proteins are not included in this study), and only dynein HCs and ICs are shown for the dynein complex. We speculate that in the absence of the pointed-end complex, some Arp1 subunits may get lost from the pointed end and/or the interaction between p150 and the Arp1 filament is weakened. These possibilities are combined in the diagram (note that a shortening of the Arp1 filament may weaken the interaction between Arp1 and p50/p150). Dynein–dynactin interaction is mediated by a direct binding between p150 and dynein IC (1,7,8). This study suggests that the conformation of p150 may be changed upon a loss of the pointed-end complex, which may lead to an increase in the affinity between the two complexes. However, because the exact conformational change of p150 is not revealed by this study, it is simply represented by a gray-to-black color change in the diagram. Similarly, a possible change in the p50 dynamitin multimer associated with p150 is represented by a black-to-gray color change in the diagram.

Whether and how this enhanced dynein–dynactin interaction negatively affects dynein motor function in vivo will be an interesting question for future studies. It is likely that an improperly enhanced dynein–dynactin interaction may prevent the dynein motor from getting recycled from membrane. It is also possible that the membrane association of dynactin is reinforced by associated dynein, a notion consistent with the result that dynein and dynactin’s membrane associations are dependent on each other (44). Thus, if dynein fails to dissociate from dynactin in the pointed-end mutants, both dynactin and dynein will fail to be recycled. This idea is consistent with an enhancement in dynein/dynactin’s membrane binding observed in pointed-end mutants from N. crassa and S. cerevisiae (23,25). Whether dynein with an abnormally high affinity to dynactin is defective in its movement along a microtubule is also a question that deserves investigation. Because motion of the dynein tail (or stem) may be important for force generation (48–57), an abnormally tight binding of dynactin to the dynein tail may negatively affect dynein motility. In addition, the enhanced dynein–dynactin interaction may cause dynein to become excessively processive because dynactin enhances dynein processivity (12). These speculations are consistent with the observation that more GFP–dynein molecules were seen along microtubules after Arp11 had been depleted (Figure 3A).

Finally, it should be pointed out that different organisms or cell types may exhibit subtle differences in dynactin structure and function. For example, while a functional dynactin complex is important for the microtubule-plus-end accumulation of cytoplasmic dynein in filamentous hyphae of A. nidulans and Ustilago maydis (this study, 33,36,58), dynactin in S. cerevisiae most likely plays a role in transferring plus-end dynein to the cell cortex where the motor exerts its pulling force to move the spindle (27,59,60). In addition, while the pointed-end proteins Arp11 and p62 are both essential for dynein function in nuclear distribution in the filamentous fungi A. nidulans (this study) and N. crassa (23), the Arp11 homologue Arp10 only becomes important in S. cerevisiae when Arp1 is defective (24,25,27). Also, although the p62 subunit is reasonably conserved, a yeast p62 homologue could not be found (27,61). Because dynein in S. cerevisiae is mainly required for the positioning of spindle/nuclei, but dynein in filamentous hyphae is required for positioning of multiple nuclei and for vesicle/organelle transport (61–63), the regulatory mechanisms may not be exactly the same in these fungi. However, similarities are also found. For example, our current study and a recent study from S. cerevisiae both revealed a decreased level of Arp1 proteins associated with p150 in the Arp11/Arp10 mutant (27), suggesting a conserved regulatory function of the Arp11 homologues, although this function may be more critical for Aspergillus dynein than for yeast dynein. A comparative analysis on the structure and function of the elaborate dynactin complex in various experimental systems may provide further insights into these issues.

Materials and Methods

A. nidulans strains, growth conditions and techniques

Strains used in this study are listed in Table 1. Aspergillus nidulans growth media such as YAG (yeast extract and glucose), YAG + UU (uridine and uracil) or MM (minimal medium) + glycerol (or glucose) + supplements, growth conditions and A. nidulans molecular genetic methods were as described previously (32,64,65).

Table 1.

Aspergillus nidulans strains used in this work (all the strains have the veA1 marker)

Strain	Genotype	Source
GR5	pyrG89; wA3; pyroA4	G. S. May
R153	wA3; pyroA4	C. F. Roberts
TNO2A3	ΔnkuA-argB; pyrG89; pyroA4	(68)
LH01	ΔArp11-pyrG; pyrG89; wA3; pyroA4	This study
LH02	ΔArp11-pyrG; pabaA1; possibly pyrG89; possibly wA3	This study
LH04	alcA-GFP-Arp11-pyr4; pyrG89; wA3; pyroA4	This study
WX825	nudR825; pyrG89; wA3	(30)
LZ12	GFP-nudA; ΔnkuA-argB; pyroA4; pyrG89	(70)
LZ29	GFP-nudA; alcA-GFP-nudR-pyr4; ΔnkuA-argB; pyroA4; pyrG89	This study
LZ30	GFP-nudA; alcA-GFP-Arp11-pyr4; ΔnkuA-argB; pyroA4; pyrG89	This study
JZ11 or S-IC	S-tagged-nudI; pyrG89, pabaA1, yA1	(70)
JZ14	S-tagged-nudI; alcA-GFP-Arp11-pyr4; pyroA4; possibly pyrG89	This study
JZ15	S-tagged-nudI; alcA-nudK-pyr4; pabaA1; pyrG89; yA1	This study
LZ92	GFP-nudA; alcA-nudK-pyr4; pabaA1; pyrG89; possibly ΔnkuA-argB	This study
LW01	S-tagged-nudM-AfpyrG; ΔnkuA-argB; pyrG89; pyroA4	This study
LW02	S-tagged-nudM-AfpyrG; pabaA1; yA1; possibly ΔnkuA-argB; possibly pyrG89	This study
LW05	S-tagged-nudM-AfpyrG; alcA-GFP-Arp11-pyr4; GFP-nudA; pyroA4; yA1; possibly ΔnkuA-argB; possibly pyrG89	This study
LW07	S-tagged-nudM-AfpyrG; GFP-nudA; pyroA4; possibly ΔnkuA-argB; possibly pyrG89	This study
JZ17	S-tagged-nudM-AfpyrG; alcA-GFP-nudR-pyr4; pabaA1; yA1	This study
JZ148	alcA-GFP-p62-pyr4; pabaA1; yA1; possibly ΔnkuA-argB	This study
JZ170	alcA-GFP-p50-pyr4; ΔnkuA-argB; pyrG89; pyroA4	This study
JZ203	GFP-nudA; alcA-GFP-p50-pyr4; ΔnkuA-argB; pyroA4; possibly pyrG89	This study
GFP-nudA/ΔArp11	ΔArp11-pyrG; alcA-GFP-nudA-pyr4; possibly pyrG89; possibly pyroA4; possibly wA3	This study
GFP-nudM/ΔArp11	ΔArp11-pyrG; alcA-GFP-nudM-pyr4; possibly pyrG89; possibly pyroA4; possibly wA3	This study
GFP–p50/ΔArp11	ΔArp11-pyrG; alcA-GFP-p50-pyr4; possibly pyrG89; possibly pyroA4; possibly ΔnkuA-argB; possibly wA3	This study
GFP–p62/ΔArp11	ΔArp11-pyrG; alcA-GFP-p62-pyr4; possibly pyrG89; possibly pyroA4; possibly ΔnkuA-argB; possibly wA3	This study
GFP-tubA/ΔArp11	ΔArp11-pyrG; alcA-GFP-tubA-pyr4; possibly pyrG89; possibly pyroA4; possibly wA3	This study

Construction of an Arp11 deletion mutant (ΔArp11)

The DNA construct used to make the ΔArp11 strain was constructed as follows. Fragment 1, which is the 2-kb genomic fragment upstream of the Arp11 coding region (ends at the 16th nucleotide after the start codon ATG), was amplified from the genomic DNA with two primers, 5′-AAGGGCCCCGGGCCACAGTTTTC-3′ and 5′-GGGAATTCGACTATGGCGGATCG-3′ (the ApaI and EcoRI sites are underlined, respectively), and was cloned into the ApaI and EcoRI sites of pXX1 (65) containing the selective marker pyrG. Fragment 2, which is the 2-kb genomic fragment downstream of the Arp11 coding region (starts at the 24th nucleotide before the stop codon) was amplified from the genomic DNA with two primers, 5′-GGTCTAGAAGTTGGACGTTGGC-3′ and 5′-AAGCGGCCGCATGGCTTAAATAATC-3′ (the XbaI and NotI sites are underlined, respectively), and was cloned into the XbaI and NotI sites of the pXX1 plasmid. The resulting Arp11 deletion construct contains the pyrG selective marker flanked at each side by fragments 1 and 2. The construct was lineated by NotI digestion and transformed into a wild-type A. nidulans strain GR5. Transformants that showed a nud phenotype on plates were subjected to a Southern blot analysis, and one of them shows a site-specific integration to the Arp11 locus that resulted in the deletion of the Arp11 coding region, and this strain was used for further studies.

Constructions of alcA-GFP-Arp11 strains

The alcA promoter in A. nidulans is shut off by glucose but unrepressed if glycerol is used as a carbon source (66). We made two alcA-GFP-Arp11 strains using the following methods. For making the first strain, the N-terminal 1-kb fragment of the Arp11 gene was obtained from A. nidulans genomic DNA by PCR using the following two primers: 5′-GGGCGGCCGCTGTCCTCAATGTCGATC-3′ and 5′-AACCCGGGCTAGCTCCTCCAGTTTAG-3′ (the NotI and SmaI sites are underlined, respectively). Note that ATG has been changed to CTG in which the C is the last nucleotide of the NotI site (this ATG is six nucleotides before the predicted start codon in the annotated A. nidulans genome). The 1-kb fragment was digested by NotI and SmaI and ligated into the corresponding sites of the LB01 vector (67) that contains GFP downstream of the alcA promoter. The construct was transformed into the wild-type strain GR5 to generate an alcA-GFP-Arp11 strain (LH04). In the expected alcA-GFP-Arp11 strain, the full-length fusion gene is under the control of the regulatable promoter alcA, which can be shut off by glucose but can be induced by glycerol to express a downstream gene at a moderate level. While the full-length GFP–Arp11 is under the alcA promoter, a C-terminal-truncated Arp11 gene (containing 1-kb coding sequence) under its endogenous promoter is still present in the genome. Transformants that show a nud phenotype on plates were further analyzed by 4′-6-diamidino-2-phenylindole (DAPI) staining of nuclei. The nud-like transformants were subjected to a Southern blot analysis, and one of them shows a single site-specific integration to the Arp11 locus, and this strain was used for studies on GFP–Arp11’s intracellular localizations (Movie S1). Although this strain (LH04) can be used as a conditional mutant of Arp11 on glucose medium, we made another conditional mutant of Arp11 with similar strategies except that a larger C-terminal truncation was made. Two oligos were used for amplifying a fragment of about 600 bp from the N-terminus of Arp11: 5′-TTTTTGCGGCCGCTGTCCTCAATGTCGATCC-3′ and 5′-TTTTTCCCGGGCCACTAAAGCCGACCGCAGTC-3′ (truncated after DNA encoding the 199th amino acid of Arp11; the NotI and SmaI sites are underlined, respectively). This PCR product was digested with NotI and SmaI and cloned into the same sites of the LB01 vector. Upon transformation of the resultant plasmid into the LZ12 strain, the alcA-GFP-Arp11 strain (LZ30) was generated, and site-specific integration of the plasmid was confirmed by a Southern blot analysis. This strain was used in our cell biological and biochemical analyses as an Arp11 conditional null mutant (also called alcA-Arp11 for simplicity). The advantage of using this strain versus using the ΔArp11 strain is that sufficient numbers of spores can be harvested from this strain when the cells were grown on the non-repressive glycerol medium. These spores were then inoculated into the glucose-containing repressive medium to deplete Arp11. This strain exhibited a typical nud mutant phenotype on glucose (Figure 2).

Complementation of the nudR825 mutation with genomic DNA encoding the p62 dynactin subunit

Two primers were used to amplify a 2.7-kb genomic fragment of p62 that covers the sequences including the entire reading frame. The sequences of the primers are 5′-GCCTTTGAGACTCGGATG-3′ and 5′-GAGATGACATTTGTGTAG-3′. The 2.7-kb PCR product was transformed into the nud mutants whose genes had not been identified (30). An auto-replicating plasmid pAid that carried the selective marker pyr4 was used as a co-transforming plasmid. Among the tested mutants including nudJ7, nudJ707, nudL43, nudN117, nudP502 and nudR825, the DNA encoding the p62 homologue only rescued the nudR825 mutant.

Construction of an alcA-GFP-p62 strain

We made a conditional null mutant of p62 using a method similar to that used for creating a conditional null mutant of Arp11. The N-terminal nudR (p62) fragment of approximately 0.8 kb was obtained from A. nidulans genomic DNA by PCR using the following two primers: p62-5, 5′-TAGCGGCCGCTGGCGTGTCCATTCCCTTA-3′ and p62-3, 5′-GTCCCGGGCATCTGGTCTTTATAG-3′ (the NotI and SmaI sites are underlined, respectively). The 0.8-kb fragment was digested by NotI and SmaI and ligated into the corresponding sites of the LB01 vector. Upon transformation of the resultant plasmid into the LZ12 strain, the alcA-GFP-p62 strain was generated similarly as the alcA-GFP-Arp11 strains described above, and this strain was used as a conditional null mutant of p62 (also called alcA-p62 for simplicity). On glycerol medium, the cells look healthy, indicating that the GFP-p62 fusion is functional.

Construction of the alcA-Arp1 strain

Two oligos nudK5 (AAGGTACCTCAACAACACCTAGGCC) and nudK3 (AGGATCCAGAGAAACGTAGC) were used for amplifying an approximately 0.8-kb N-terminus fragment of nudK (Arp1), and the fragment was digested and inserted into the Kpn1 and BamH1 sites of the pAL3 plasmid that allows the expression of the alcA promoter. After transforming the resultant plasmid into the JZ11 (S-IC) strain, the transformants with a nud phenotype were selected, and a Southern blot analysis was used to confirm the strain with a site-specific single integration of the plasmid into the genome. A genetic cross was performed, and a strain without S-IC was selected. These strains were used as conditional null mutants of Arp1.

Construction of the alcA-GFP-p50 strain

Two oligos p50/Not1 (5′-TTTTTGCGGCCGCTGGCTTTCAACAAAAAATATGCTGGTC-3′) and p50/SmaI (5′-TTTTTCCCGGGCGACGAGAAGGGTGATGAGTGTTGA-3′) were used for amplifying an approximately 0.8-kb N-terminal fragment of the A. nidulans p50 (the NotI and SmaI sites are underlined, respectively). The fragment was digested by NotI and SmaI and ligated into the corresponding sites of the LB01 vector. Upon transformation of the resultant plasmid into the TNO2A3 strain with ΔnkuA that greatly decreases the frequency of nonhomologous integration (68), the alcA-GFP-p50 strain was generated similarly as the alcA-GFP-Arp11 strains described above, and this strain was used as a conditional null mutant of p50 (also called alcA-p50 for simplicity). On glycerol medium, the cells look healthy, indicating that the GFP-p50 fusion is functional.

Construction of a strain containing an S-tagged nudM gene (also called S-tagged p150 or S-p150)

A 1382-bp C-terminal region of the nudM open reading frame (ORF) and the immediate 3′ untranslated region (possibly also with sequences downstream of the nudM gene; 3′-UTR, 1390 bp) were amplified from genomic DNA, respectively, with corresponding primer pairs of ORF forward (GTTGCTACAGTGAAGATCAACCGTG) and ORF reverse (TAAGGTTGGTTTAATTGCTCGCTCA) and UTR forward (CAGACCTTTTCTATGGGCTGCTTAG) and UTR reverse (TTACCGTCATGAAGGCGACGAC). The DNA containing Aspergillus fumigatus pyrG gene (AfpyrG) and an S-tag sequence at the 5′ of the promoter was amplified from the pAO81 plasmid (obtained from the Fungal Genetics Stock Center (FGSC), deposited by Stephen A. Osmani) (69) with the fusion forward primer (CAGCCTGTTGAGCGAGCAATTAAACCAACCTTAGGAGCTGGTGCAGGCGCTGGAG) and the fusion reverse primer (AGATGGCTAAGCAGCCCATAGAAAAGGTCTGCTGTCTGAGAGGAGGCACTGATG). Note that the S-tag follows an eight amino acid linker GAGAGAGA. Thus, the fusion forward primer contains no S-tag sequence but contains the nucleotide sequence encoding almost the entire linker (GGA GCT GGT GCA GGC GCT GGA G for amino acids GAGAGAGA). The PCR product contains the S-tag and the selective marker AfpyrG and it also contains, at the 5′ and 3′ ends, respectively, 33 and 31 bp sequences derived from the nudM gene. Fusion PCR was performed between the ORF and the S-tag-and-AfpyrG fragments and between the UTR and the S-tag-and-AfpyrG fragments, respectively, to obtain the fused fragments ORF + S-tag-and-AfpyrG and S-tag-and-AfpyrG + UTR. Another round of PCR reactions was performed to introduce SpeI and SacII sites at the beginning of the ORF + S-tag-and-AfpyrG fragment and the end of the S-tag-and-AfpyrG + UTR fragment. Both fragments were digested with NdeI that cuts once in the middle of the S-tag-and-pyrG fragment and ligated, which fused the ORF, the S-tag-and-AfpyrG and the UTR fragments together. This ligated fragment contains a 15-residue S-tag (KETAAAKFERQHMDS) attached to the C-terminal end of NUDM with an eight-residue linker (GAGAGAGA). A stop codon was placed at the end of the S-tag. This final product was ligated into the SpeI and SacII sites of the vector pBluescript II SK(+), and the sequence of the fusion construct was confirmed by DNA sequencing (Molecular Biology Core Facility, Uniformed Services University of the Health Sciences). The fusion fragment was excised from the vector and transformed to the TNO2A3 strain (68). The homologous integration of the S-tagged NUDM fragment into the A. nidulans genome was confirmed by two PCR reactions on genomic DNA with two sets of primers. Within the first set, the forward primer (5′p/SpeI+, GATCTGAACCTCAAACTGCAGTCG) was 700 bp upstream of the 5′ end of the transformed fragment and the reverse primer (pGf+, CAATCACTGGTAACTCCACGGAAC) was within the pyrG sequence. Within the second set, the forward primer (pGf+, TTGGAGCAAAAGTGTAGTGCCAG) was within the pyrG sequence and the reverse primer (3′p/Sac2−, TCGACTGCGTCTGCACTCAC) was 700 bp downstream of the 3′ end of the transformed fragment. The positive PCR results indicate that this fragment has integrated via a double homologous recombination event, and thus, the endogenous nudM has been replaced by an S-tagged nudM. This strain, S-p150, forms healthy colonies, indicating that the S-tag did not disrupt the function of the A. nidulans NUDM (p150; Figure 5A).

S-tag-based purification of dynactin and dynein from A. nidulans

Aspergillus nidulans protein extract was obtained from a 1-L overnight culture using the liquid nitrogen grinding method for breaking the hyphae, as described previously (32), except that the protein isolation buffer contains 25 mM Tris (pH 8.0), 0.4% Triton-X-100, 10 μM ATP, 1 mM DTT and 10 μL/mL of a protease inhibitor cocktail (Sigma). The construction of an S-IC strain and the method for dynein purification were as previously described (70). For purification of A. nidulans dynein and dynactin from S-tagged strains, about 30 mL of a protein extract (about 10 mg/mL) was incubated for half an hour at room temperature with 1 mL of S-protein beads (Novagen, Inc.). The beads were repeatedly washed with the same protein isolation buffer except that no detergent was added. Finally, the S-tagged protein was eluted with 5 mg/mL S-peptide in 0.2 mL of a buffer containing 25 mM Tris–HCl (pH 8.0), 10 μM ATP and 1 mM DTT. About 30 μL of the eluate was loaded onto a 4–15% SDS–PAGE gradient gel (Bio-Rad). For silver staining of protein gels, the Silver Stain Plus Kit from Bio-Rad was used. Concentrations of the dynein HC or p150^Glued in different samples were estimated by comparing the band intensity to differently diluted BSA samples on the same silver-stained gel. In a typical experiment, the concentrations of these proteins are about 1–5 ng/μL. Western blot analyses were performed as previously described (32).

Production of an anti-p150 (NUDM) polyclonal antibody

Two primers 5′-TTTGGATCCGAGCGTAACTTGGCCGAGTTG-3′ (p150+) and 5′-TTTGGTACCTTATTCCCGATACTCGATCTCCTC-3′ (p150−) were used to amplify the DNA corresponding to a NUDM (p150) region from amino acids 278 (Glu) to 501 (Glu). The PCR product was digested with BamHI and KpnI and ligated into the same sites of the pQE30 vector (Qiagen). Fusion protein was expressed in E. coli and partially purified by inclusion body preparation prior to injection to rabbits (Covance Research Products Inc.). The antibody was affinity purified as described previously (64,65).

Production of an anti-Arp1 polyclonal antibody

Two primers 5′-TTTTTGCATGCATGACCGAGGCTACTCTTCAC-3′ (Arp1+) and 5′-TTTTTAAGCTTTGATCTTATGCCCATCCGGC-3′ (Arp1−) were used to amplify the DNA corresponding to a NUDK (Arp1) region from amino acids 1 (Met) to 254 (Ile). The PCR product was digested with SphI and HindIII and ligated into the same sites of the pQE30 vector (Qiagen). Fusion protein was expressed in E. coli and partially purified by inclusion body preparation prior to injection to rabbits (Covance Research Products Inc.). The antibody was affinity purified as described previously (64,65).

Production of an anti-dynein IC polyclonal antibody

Two primers 5′-AAGGATCCATGAACCAGACGGC-3′ (IC+) and 5′-AACCCGGGCTCATTCGGTCCTTATCC-3′ (IC−) were used to amplify the DNA corresponding to a NUDI (IC) region from amino acids 329 (His) to 690 (Ser). The PCR product was digested with BamHI and SmaI and ligated into the same sites of the pQE30 vector (Qiagen). Fusion protein was expressed in E. coli and partially purified by inclusion body preparation prior to injection to rabbits (Covance Research Products Inc.). The antibody was affinity purified as described previously (64,65).

Ultracentrifugation procedures

The following procedure was used for isolating dynein and dynactin in soluble fractions and for subsequent analysis. After being grinded in the presence of liquid nitrogen, hyphal powders were evenly divided into two 2-mL centrifuge tubes. Protein isolation buffer containing 25 mM Tris–HCl (pH 8.0), 1 mM DTT and 10 μL/mL protease inhibitor cocktail (Sigma), with or without 0.4% Triton-X-100, was added into each tube. The protein isolation buffer was mixed with the hyphal powders, followed by a 8 000 × g spin for 30 min. Supernatant from each tube was transferred into a 1.5 mL Beckman ultracentrifuge tube and subjected to a high-speed ultracentrifugation at 100 000× g for 20 min. All centrifugations were performed at 4°C. The high-speed supernatant was transferred into a new 2-mL eppendorf tube, and 150 μL of S-protein agarose beads (Novagen, Inc.) were added. The procedure for S-tag-based purification was as described above. After the binding and washing steps, 45 μL of 2× protein loading buffer was mixed with the beads and boiled for 10 min. The beads were spun down for 1 min, and the supernatants were loaded into the gel wells.

Introducing various GFP fusion proteins and alcA-based mutant alleles into different strains

Standard genetic crosses were performed for introducing the GFP-NUDA fusion into the alcA-based Arp1 and p50-depleted cells. The alcA-Arp11 and alcA-p62 alleles were introduced into the GFP-NUDA background by transformation. Progenies with a nud phenotype and GFP signals on glucose medium were selected for further observations on glucose medium. Standard genetic crosses were performed for introducing the alcA-GFP-NUDA (dynein HC), alcA-GFP-NUDM (p150), alcA-GFP-p62 and alcA-GFP-p50 fusions into the ΔArp11 background. Progenies with a nud phenotype and GFP signals on glycerol medium were selected for further observations on glycerol medium. In the case of alcA-GFP-p62, because the original strain carries not only alcA-GFP-p62 but also GFP-nudA (HC) under its own promoter, progenies without any GFP signal on glucose were selected first, and these progenies should not contain the GFP-nudA (HC) allele. Because the alcA-GFP-p62 fusion only showed faint background fluorescence in the ΔArp11 background on glycerol, we identified the desired progeny with a PCR analysis. The PCR was performed on genomic DNAs isolated from the nud progeny using two primers: GFP forward (5′-AGAGACCACATGGTCCTTC-3′) and p62 (5′-CCTCGCTTGGTGTCAGGGC-3′), and the progeny with an expected 1.0-kb product was selected for further observations. Standard genetic crosses were also performed for introducing the alcA-based mutant alleles into the strain background with S-tagged p150 or S-tagged IC.

Image acquisition

Cells were grown in ΔTC3 culture dishes (Bioptechs) in 1.5 mL of MM + glycerol (or glucose) + supplements. Images were captured using an Olympus IX70 inverted fluorescence microscope (with a 63× objective) as described previously (33,36), except that a PCO/Cooke Corporation Sensicam QE cooled charge-coupled device camera was used. The IPLAB software was used for image acquisition and analysis.

Supplementary Material

Supplemental Captions and Figures

Figure S1: Sequence comparison of the Arp11 proteins from Aspergillus nidulans (called Arp11a in this figure, but called Arp11 in the text for simplicity), Neurospora crassa (Ro7) and mouse (Arp11).

Figure S2: Sequence comparison of the p62 proteins from Aspergillus nidulans (called NUDR in this figure but called p62 in the text for simplicity), Neurospora crassa (Ro2) and rat (p62).

Figure S3: Benomyl treatment abolishes the dots along filament-like structures of GFP-NUDA (or GFP–HC as shown in Figure 3) in the alcA-Arp11 mutant. A) Cytoplasmic microtubules in the alcA-GFP-tubA cells (31) disappeared after a 2-h treatment with 2.4 μg/mL of benomyl, but a few microtubules were still present after a 1-h treatment. Cells were grown on minimal glycerol medium overnight at 32°C. B) The dots along filament-like structures of GFP-NUDA (or GFP–HC as shown in Figure 3) in the alcA-Arp11 mutant disappeared after a 2-h treatment with 2.4 μg/mL of benomyl, and GFP signals were seen to be associated with nuclei (two images at the top). The septum-associated signals were present after a 2-h treatment with benomyl of the same concentration (three images at the bottom). Cells were grown on minimal glucose medium overnight at room temperature and shifted to 32°C for about 5 h.

Figure S4: Western analyses on proteins (dynein HC and p150) pulled down by S-IC. Note that the level of the p150 pulled down was higher in the Arp11-depleted cells (alcA-Arp11) than in wild-type cells. The Arp1-depleted cells (alcA-Arp1) were used as a negative control because there was no detectable p150 from this sample. Cells were grown on YUU overnight at 32°C.

Supplemental Movie 1

Movie S1: GFP–Arp11 proteins (expressed under the alcA promoter) highlight dynamic comet-like structures representing their localization at the microtubule plus ends. Cells were grown on minimal glycerol medium overnight at 32°C. The movie is sped up 10 times.

Supplemental Movie 2

Movie S2: GFP-NUDA (dynein HC) proteins (expressed under the endogenous promoter) highlight dynamic comet-like structures representing their localization at the microtubule plus ends. Cells were grown on minimal glucose medium overnight at room temperature and shifted to 32°C for about 5 h. Multiple cells are shown in this movie. The movie is sped up 10 times.

Supplemental Movie 3

Movie S3: GFP-NUDA (dynein HC) proteins (expressed under the endogenous promoter) in Arp11-depleted cells are mislocalized to dots along filament-like structures. Cells were grown on minimal glucose medium overnight at room temperature and shifted to 32°C for about 5 h. Multiple cells are shown in this movie. The movie is sped up 10 times.

Supplemental Movie 4

Movie S4: GFP-NUDA (dynein HC) proteins (expressed under the endogenous promoter) in p62-depleted cells are mislocalized to dots along filament-like structures. Cells were grown on minimal glucose medium overnight at room temperature and shifted to 32°C for about 5 h. Multiple cells are shown in this movie. The movie is sped up 10 times.

Supplemental Movie 5

Movie S5: The accumulation of GFP-NUDA (dynein HC) proteins (expressed under the endogenous promoter) to comet-like structures is much less obvious in p50-depleted cells than in wild-type cells (compared with Movie S2). Cells were grown on minimal glucose medium overnight at room temperature and shifted to 32°C for about 5 h. Multiple cells are shown in this movie. The movie is sped up 10 times.

Supplemental Movie 6

Movie S6: The accumulation of GFP-NUDA (dynein HC) proteins (expressed under the endogenous promoter) to comet-like structures is much less obvious in Arp1-depleted cells than in wild-type cells (compared with Movie S2). Cells were grown on minimal glucose medium overnight at room temperature and shifted to 32°C for about 5 h. Multiple cells are shown in this movie. The movie is sped up 10 times.

Supplemental Movie 7

Movie S7: GFP–p62 fusion proteins highlight dynamic comet-like structures representing their localization at the microtubule plus ends. Cells were grown on minimal glycerol medium overnight at 32°C. The movie is sped up 10 times.

Supplemental Movie 8

Movie S8: GFP–p50 fusion proteins highlight dynamic comet-like structures representing their localization at the microtubule plus ends. Cells were grown on minimal glycerol medium overnight at 32°C. The movie is sped up 10 times.