VezA/vezatin facilitates proper assembly of the dynactin complex in vivo.

In both wild-type and Δ vezA strains, we could observe GFP-labeled dynein comets near the hyphal tip, which represent the microtubule plus-end accumulation of dynein. 67 The maximal comet signal intensity is not significantly reduced in the Δ vezA mutant ( Figures 1A and 1B ). We then examined the effect of VezA on plus-end dynein accumulation in the absence of the cargo adapter HookA. 13 Loss of HookA (Δ hookA ) significantly increases dynein comet intensity ( Figures 1A and 1B ), because dynein can no longer leave the plus end with its early-endosome cargo. In the Δ hookA , Δ vezA double mutant, the GFP-dynein comets are present, but the signal intensity is significantly lower compared to that in the Δ hookA single mutant ( Figures 1A and 1B ). Thus, VezA enhances plus-end dynein accumulation before cargo binding. The reason we could not detect this effect in the Δ vezA single mutant could be because the plus-end accumulation of dynein and its HookA-mediated departure are both decreased upon loss of VezA. Overexpression of the cytosolic ΔC-HookA (missing the cargo-binding C terminus) drives dynein relocation from the microtubule plus ends near the hyphal tip to septa or the spindle-pole bodies, 80 where microtubule-organizing centers (MTOCs) are located. 81 , 82 This is consistent with cargo-adapter-mediated dynein activation in vitro as well as dynein activation caused by the cortical adapter Num1 in yeasts. 28 , 29 , 83 – 86 In A. nidulans , this plus-end-to-minus-end relocation diminishes the signal intensity of plus-end dynein and causes dynein to be largely accumulated at the septa (note that the accumulation at the spindle-pole bodies can be seen only at G1 87 ). 80 To determine if VezA affects the ΔC-HookA-mediated dynein activation, we introduced into the Δ vezA mutant the gpdA -ΔC- hookA -S allele, which causes ΔC-HookA overexpression. In the gpdA -ΔC- hookA -S, Δ vezA double mutant, plus-end dynein comet intensity is obviously lower than that in the Δ vezA control but significantly higher than those in the gpdA -ΔC- hookA -S single mutant ( Figures 1C and 1D ). In the same double mutant, the dynein signal intensity at the septal minus ends is significantly decreased compared to that in the gpdA -ΔC- hookA -S single mutant although still higher than that in the Δ vezA single mutant ( Figures 1C and 1E ). Thus, while cargo-adapter-mediated dynein activation still occurs in the Δ vezA mutant, the presence of VezA significantly enhances this process. This also explains the observation that, while loss of VezA deceases the plus-end accumulation of dynein in the Δ hookA background, this effect is not obvious in the wild-type background, because the defect in cargo-mediated dynein departure in turn increases the number of plus-end-located dynein molecules. Since VezA is involved in the microtubule-plus-end accumulation of dynein before cargo binding and cargo-adapter-mediated dynein activation, two processes that both need dynactin, 28 , 29 , 39 we further examined the relationship between VezA and the dynactin complex. VezA-GFP signals are quite faint, but by incubating the cells at a lower temperature, we detected its actin-dependent localization at the hyphal tip. 67 Using a myosin-V-null mutant, Δ myoV , 88 we showed that myosin V is important for this localization ( Figures S1A and S1B ). However, myosin V is not required for dynein-mediated early-endosome transport ( Figure S1C ), indicating that the hyphal-tip localization of VezA per se is not important for dynein function. VezA contains two predicted transmembrane (TM) domains, and ΔTM-VezA-GFP (without the TM domains and the region between them) is cytosolic. 67 While VezA-GFP or ΔTM-VezA-GFP pulls down very little dynactin in cell extract ( Figure S2A ), overexpressed ΔTM-VezA-GFP (driven by the gpdA promoter) pulls down dynactin reproducibly ( Figures 2A , S2A , and S2B ). This pull-down is specific because dynactin was not pulled down with GFP overexpressed under the same gpdA promoter ( Figures 2A , S2A , and S2B ). Dynein and its binding proteins are pulled down as well, as expected ( Table S1 ; Data S1 ). To determine which dynactin component is involved in the VezA-dynactin interaction, we used several conditional-null mutants, including alcA -Arp11, alcA -p50, and alcA -Arp1, in which the alcA promoter can be repressed by glucose. 89 Loss of Arp11 did not obviously affect p150 protein level but decreased the amount of Arp1 associated with p150. 47 Here, we found that loss of Arp11 ( alcA -Arp11) caused a significant decrease in the amount of p150 pulled down with ΔTM-VezA-GFP ( Figures 2B , 2C , and S3A ). The amount of Arp1 pulled down is also decreased significantly, but this may be more directly due to the effect of Arp11 loss on the Arp1 mini-filament. 47 The significant decrease in the pull-down of p150 upon loss of Arp11 was confirmed by a mass spectrometry analysis, which further shows a significant decrease in the amount of p50 and other dynactin components pulled down with ΔTM-VezA-GFP ( Table S1 ; Data S1 ). These results suggest that the pointed end enhances the VezA-dynactin interaction ( Figure 2D ). Loss of p50 ( alcA -p50) diminished the protein level of p150 in total extract ( Figures 2B and S3A ) and consequently the amount of p150 pulled down with ΔTM-VezA-GFP ( Figures 2B and 2E ). However, it did not decrease the amounts of Arp1 and Arp11 pulled down with ΔTM-VezA-GFP ( Figures 2B and 2E ). Consistently, a mass spectrometry analysis on ΔTM-VezA-GFP pull-down in the alcA -p50 extract detected no peptides of p150 but detected about normal numbers of peptides from the pointed-end proteins Arp11, p62, and p25 ( Table S1 ; Data S1 ). Thus, p50 and p150 are not important for the VezA-dynactin interaction ( Figure 2F ). Previously, loss of Arp1 decreased the protein level of p150 in total extract. 47 Here, we confirmed this result in the gpdA -ΔTM- vezA -GFP background ( Figures 2G and S3B ) and found that loss of Arp1 ( alcA -Arp1) abolished the pull-down of p150 ( Figures 2G and 2H ). Importantly, although Arp11 appears stable in the alcA -Arp1 strain, the amounts of Arp11 and S-tagged p25 (p25-S) pulled down with ΔTM-VezA-GFP are dramatically decreased ( Figures 2G , 2H , and S3B ). Consistently, the mass spectrometry analysis of ΔTM-VezA-GFP pulled-down proteins from the alcA -Arp1 extract detected no peptides of p150 and p50 but did detect peptides of Arp11, p62, and p25, whose numbers were lower compared to those in the control with Arp1 ( Table S1 ; Data S1 ). Thus, the Arp1 mini-filament is critical for the VezA-dynactin interaction, although the pointed-end sub-complex may weakly interact with VezA ( Figure 2I ). So far, we have not been able to purify dynactin and VezA to test whether the VezA-dynactin interaction is direct, but we used AlphaFold2 to test whether a direct interaction is possible. VezA was predicted to form a homodimer (a prediction made by Andrew Carter at MRC Laboratory of Molecular Biology). Thus, we used two copies of VezA in an AlphaFold2-based analysis, including the pointed-end proteins (Arp11, p62, p25, and p27) and four copies of Arp1 ( Figures 3A and 3B ). Four models were obtained, and one of them is shown ( Figures 3B and 3C ). In this model (as well as in two other models), the N terminus of VezA (aa1–20) is close to the pointed end via p62 and p25 ( Figure 3D ). The C-terminal a helix (563–615) is docked in a pocket formed by two Arp1 subunits ( Figure 3E ). To test the importance of the N- and C-terminal regions, we made the vezA Δ1–20 -GFP and vezA Δ563–615 -GFP mutants. Like the functional full-length VezA-GFP, the VezA Δ1–20 -GFP and VezA Δ563–615 -GFP proteins can be detected on western blots after pull-down but not in total extract ( Figures 3F and S3C ). VezA-GFP is not as effective as the overexpressed ΔTM-VezA-GFP in dynactin pull-downs ( Figure S2A ), and only a faint Arp11 signal was detected in the pull-down sample ( Figures 3F and S3C ). Although the faint Arp11 signal was not seen in the mutant samples, future work using overexpressed proteins will be needed to confirm the roles of these regions in the VezA-dynactin interaction. Using live-cell imaging, we found that, while VezA Δ1–20 -GFP localizes normally to the hyphal tip, VezA Δ563–615 -GFP protein is more diffused in the cytoplasm ( Figure 3G ), but most importantly, both mutants exhibited a clear defect in dynein-mediated early-endosome transport ( Figures 3G and 3H ). Thus, the N and C termini predicted to be important for the VezA-dynactin interaction are indeed critical for the function of VezA in dynein-mediated cargo transport. Consistent with earlier observations in mammalian cells, 90 – 92 dynactin p150-GFP in A. nidulans forms plus-end comets, which depends on its microtubule-binding domain. 40 , 42 Similarly, Arp11-GFP, 93 p25-GFP, 7 and p62-GFP all form plus-end comets near the hyphal tip ( Figures 4A and S4 ), although their signal intensities appeared lower than that of p150-GFP, most likely because they are monomeric, whereas p150 is dimeric within dynactin. In addition, p50-GFP forms bright plus-end comets ( Figure 4A ). In the Δ vezA mutant, p50-GFP comet intensity is significantly decreased, and Arp11-GFP comets could hardly be seen ( Figures 4A and 4B ). The comet intensity of p150 is also decreased in the mutant, although the decrease was less significant compared to the other components ( Figure 4B ). Similar to Arp11-GFP, p62-GFP and p25-GFP comets could hardly be seen in the Δ vezA mutant ( Figure S4A ). Previously, we found that the plus-end accumulation of p150 is increased in the Δ hookA mutant, likely due to a lack of cargo-mediated dynactin-dynein departure from the plus end 10 ( Figure 4C ). Consistently, the plus-end accumulation of Arp11-GFP, p62-GFP, or p25-GFP is also more visible in the Δ hookA background ( Figures 4C and S4B ). We found that the plus-end accumulation of all examined dynactin components is reduced in the Δ vezA , Δ hookA double mutant compared to that in the Δ hookA single mutant ( Figures 4C , 4D , S4B , and S4C ). Among the dynactin components examined, the pointed-end proteins were most severely affected by the loss of VezA (note that the mean of Arp11-GFP comet intensity in the Δ vezA , Δ hookA background is ~36% of that in the Δ hookA background) ( Figure 4D ); p50-GFP was affected more severely (mean in Δ vezA , Δ hookA is ~51% of that in Δ hookA ) compared to p150-GFP (mean in Δ vezA , Δ hookA is ~72% of that in Δ hookA ) ( Figure 4D ). We next examined the protein levels of p150-GFP, p50-GFP, and Arp11-GFP in total extracts using western blot analysis. Interestingly, the protein levels of p150-GFP and p50-GFP but not Arp11-GFP are obviously decreased in the Δ vezA mutant ( Figure 5A ). To quantify the result, we probed p150-GFP or p50-GFP together with Arp11 in the same samples. While Arp11 levels are similar in wild-type and Δ vezA samples, the ratio of p150-GFP or p50-GFP to Arp11 is significantly lowered in the Δ vezA mutant ( Figures 5B – 5D , S5A , and S5B ). Furthermore, in strains containing Arp11-GFP, the protein levels of S-tagged p50 (p50-S) and p150 (without any tag) show the same reduction relative to Arp11 in the Δ vezA mutant ( Figures 5E , 5F , and S5C ). Since the protein level of Arp11 is not obviously lower in the Δ vezA mutant, the significant reduction in the microtubule plus-end accumulation of Arp11 is most likely caused by a decrease in its association with p150, whose microtubule-binding domain is critical for its microtubule plus-end accumulation. 42 , 92 , 94 – 96 To further confirm this notion and to examine other dynactin components that we are unable to detect in total extract (for example, Arp1), we did pull-downs using extracts containing Arp11-GFP. Loss of VezA (Δ vezA ) caused a significant reduction in the amounts of p150, p50-S, and Arp1 pulled down with Arp11-GFP ( Figures 5G , 5H , S5D , and S5E ). Furthermore, our mass spectrometry analysis showed that the amounts of p150, p50, Arp1, and capping protein pulled down with Arp11-GFP were all decreased in Δ vezA samples ( Figure 5I ; Table S2 ; Data S2 ), consistent with a defect in dynactin assembly. Arp11-GFP also pulled down a much lower amount of dynein in the Δ vezA extract ( Figure 5I ; Table S2 ), which is expected, since p150 and Arp1 are critical for binding dynein. 9 , 12 , 19 , 22 , 23 However, the amounts of pointed-end proteins p62 and p25 pulled down with Arp11-GFP were not significantly affected ( Figure 5I ; Table S2 ), suggesting that the pointed-end sub-complex is stable upon loss of VezA. In the Δ vezA mutant, some functional dynactin complexes must still exist, because dynein-mediated transport of early endosomes still occurs, albeit at a significantly lowered frequency, 67 and the Δ vezA colony is healthier than dynactin mutants. 47 , 67 Dynein-mediated nuclear distribution appears normal in some Δ vezA germ tubes, 67 but our quantitative analyses revealed a partial defect in nuclear distribution in the Δ vezA mutant ( Figures S6A – S6C ). However, the defect is less severe compared to the nudA 1 dynein mutant examined under the same conditions. 97 Thus, VezA is not essential for dynactin assembly but facilitates this process to ensure optimal dynein function. This is consistent with the results that vezatin homologs in higher eukaryotes are not essential for dynein function but are important for the transport of some dynein cargoes. 79 To better understand the effect of VezA on dynactin assembly, we further examined how other dynactin components affect the integrity of the complex in vivo . In this set of experiments, we used the alcA -promoter-based conditional-null dynactin mutants to perform western analyses on total extracts and on proteins pulled down with a GFP-labeled dynactin component. Loss of p50 ( alcA -p50) diminished the protein level of p150 in total extract and in p25-GFP pull-down ( Figures 6A and S7A ). Interestingly, the amount of Arp1 but not Arp11 pulled down with p25-GFP is also lowered upon loss of p50 ( Figures 6A , 6B , and S7A ). Thus, loss of p50 affects the assembly of the Arp1 mini-filament ( Figure 6C ), supporting the proposal that multiple p50 proteins control mini-filament length. 9 We next examined dynactin components upon loss of p150 ( alcA -p150) ( Figures 6D – 6G ). Interestingly, p50 is still stable without p150 ( Figures 6D , 6E , and S7B ), but its association with p25-GFP is almost non-detectable in pull-downs ( Figures 6F , 6G , and S7B ). Thus, p150 is required for p50 to associate with the Arp1 mini-filament. Consistent with a role of p50 in Arp1 mini-filament assembly, loss of p150 also lowered the amount of Arp1 pulled down with p25-GFP ( Figures 6F , 6G , and S7B ). Thus, p150 and p50 require each other for the assembly of a shoulder sub-complex that interacts with the Arp1 mini-filament. Without the shoulder, Arp1 mini-filament assembly is defective ( Figure 6H ). Previous data suggested that Arp1 is required for p150 stability 47 , 98 , 99 and Arp11 is required for the integrity of the dynactin complex. 8 , 47 , 100 Here, we confirmed that upon loss of Arp1, p150 is not detectable in total extract ( Figures 6I and S7C ). The level of p50 is also significantly lowered ( Figures 6I , 6J , and S7C ). In comparison, loss of Arp11 does not significantly affect the levels of p150 and p50 ( Figure 6K ). We performed p50-GFP pull-downs to further test the effects of alcA -Arp1 and alcA -Arp11 in dynactin integrity ( Figures 6L – 6N ). Loss of Arp1 ( alcA -Arp1) further reduced the amount of p150 pulled down relative to p50-GFP ( Figure 6M ) (note that the observed reduction in protein amount is most likely due to a lower overall protein amount in the total extract). Thus, loss of Arp1 diminishes the assembly of the shoulder complex that includes p150 and p50. No Arp11 was detected in p50-GFP pull-down upon loss of Arp1 ( Figure 6L ) in three pull-down experiments. Thus, although an Arp11-p50 direct interaction was detected in a yeast two-hybrid assay, 100 the Arp11-p50 connection is mainly mediated by the Arp1 mini-filament in vivo , consistent with a vertebrate dynactin structure. 9 Interestingly, loss of Arp11 ( alcA -Arp11) significantly reduced the amount of Arp1 pulled down with p50-GFP ( Figures 6L , 6N , and S7C ), while it had no significant effect on the levels of p150 and p50 in total extract ( Figure 6K ) or the amount of p150 pulled down with p50-GFP ( Figures 6L and 6N ). Thus, while Arp11 affects the Arp1 mini-filament significantly, a defective Arp1 mini-filament or simply the presence of the Arp1 protein is sufficient for supporting shoulder assembly ( Figure 6O ), which is in contrast to a much more significant defect in shoulder assembly if no Arp1 protein is present ( Figure 6O ).

Requests for further information and resources and reagents should be directed to and will be fulfilled by the lead contact, Xin Xiang ( [email protected] ). A. nidulans strains generated in this study are available upon request. All data reported in this paper will be shared by the lead contact upon request. This paper does not report original code. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

VezA most likely binds Arp1 and its pointed-end sub-complex during the dynactin assembly process ( Figure 7 ). Specifically, both Arp1 and the pointed-end protein Arp11, but not the shoulder proteins p50 or p150, are important for the VezA-dynactin interaction. Either the pointed-end sub-complex or Arp1 itself is able to interact weakly with VezA, but the interaction is significantly enhanced with both Arp1 and the pointed end being present. Although we have not obtained purified dynactin or the Arp1 mini-filament with its pointed-end sub-complex to test if they bind VezA directly, our AlphaFold2-based prediction suggests that a direct VezA interaction with the mini-filament is possible ( Figure 3 ). The interaction may be transient, since VezA may leave dynactin after its role as an assembly factor is accomplished. Loss of VezA does not affect the assembly of the pointed-end sub-complex, as Arp11-GFP pulls down normal amounts of p25 and p62 without VezA, but it significantly decreases the amounts of Arp1, p50, p150, and capping protein associated with Arp11 ( Figure 5 ). Thus, some pointed-end sub-complexes may be bound with a reduced number of Arp1s in the absence of VezA. Interestingly, we also found that the protein levels of p150 and p50 are decreased upon loss of VezA ( Figure 5 ). Since VezA interacts with Arp1 and the pointed end rather than p150 or p50, the effect of VezA on p50-p150 is likely due to a defect in the Arp1 mini-filament, which in turn prevents the attachment of p50-p150, thereby causing them to be unstable. Combined with previous data, our current study further indicates that shoulder integrity depends on Arp1. This dependency most likely happens during dynactin assembly. While the p150-p50-p24 shoulder complex can be separated from the rest of the purified dynactin complex, 3 the p150 protein level is significantly decreased upon loss of Arp1 in cells ( Figure 6 ). 47 , 98 , 99 Loss of Arp1 also decreases the protein level of p50 and its association with p150 ( Figure 6 ). Thus, Arp1 is required for dynactin shoulder assembly. The effect of Arp1 on p150 stability is likely mediated via p50, as loss of p50 diminishes p150 protein level ( Figure 6 ). This is consistent with previous data that loss of p24 disrupts the p150-p50 interaction in budding yeast 101 and lowers p150 stability in N. crassa . 99 Interestingly, although p50 and p150 are both in the shoulder complex, the p50 protein level is not affected by a loss of p150, while it is significantly affected by a loss of Arp1 ( Figure 6 ). It is possible that Arp1 is required for assembly of the two p50-p24 parts of the shoulder that bind the p150 C terminus, 6 , 9 thereby affecting p50 stability. Upon its Arp1-dependent assembly onto dynactin, the shoulder complex controls the integrity of the Arp1 mini-filament. Specifically, we found that the amount of Arp1 pulled down with p25-GFP is significantly reduced upon loss of either p150 or p50 ( Figure 6 ). The effect of p150 loss on mini-filament assembly is likely due to p50 not being able to attach to the Arp1 mini-filament (although p50 protein is stable without p150) ( Figure 6 ). There are four copies of p50 within dynactin, 3 which extend out tentacle-like sites that touch the Arp1 mini-filament. 9 , 27 It was proposed that the four p50 sub-units and their extended regions may control the Arp1 mini-filament assembly, 9 and our current result provides support for this idea. Although purified Arp1 can self-assemble to form a filament in vitro , 102 the pointed end plays a significant role in Arp1-mini-filament integrity in vivo . In A. nidulans , loss of Arp11 or p62 significantly decreases the amount of Arp1 pulled down with p150, although p150 is stable. 47 In mammalian cells, loss of the pointed end severely affects the assembly of Arp1 into the dynactin complex although p150 is stable. 8 Consistent with a role of the pointed-end sub-complex in Arp1 mini-filament assembly in vivo , loss of Arp11 significantly reduced the amount of Arp1 pulled down with p50-GFP ( Figure 6 ). However, loss of Arp11 affects neither the protein levels of p150 and p50 nor the p50-p150 association ( Figure 6 ). Thus, in the absence of the pointed-end sub-complex, Arp1 must be present, and the residual Arp1-p50 interaction, possibly provided by a defective mini-filament, is sufficient for assembly of the dynactin shoulder. This is very different from the effect of VezA loss, because the protein levels of both p50 and p150 are reduced in the absence of VezA. Thus, VezA on the Arp1 mini-filament with its pointed end may not only affect mini-filament assembly but also facilitate the Arp1-p50 interaction to keep p50 stable. Based on current and previous data, we propose a model of dynactin assembly in the presence of VezA ( Figure 7 ). In this model, the pointed-end sub-complex is assembled first. The idea that this sub-complex is stable is also consistent with its purification from cells expressing recombinant proteins. 6 , 11 The assembly of the pointed-end sub-complex is followed by addition of a few Arp1 sub-units, which may happen before VezA binding or be facilitated by VezA. This is then followed by assembly of the shoulder onto the filament, which is facilitated by VezA. Shoulder assembly onto the filament further facilitates assembly of the filament to reach a correct length before being capped at the barbed end. In the absence of VezA, the pointed-end sub-complex can still be assembled normally. However, the initial assembly of the Arp1 mini-filament may be defective in a way that further reduces the chance of shoulder assembly (model A). Or, the initial assembly of the Arp1 mini-filament is normal, but the shoulder cannot be assembled onto the filament, which in turn makes the filament shorter than normal (model B). In either model, the cellular concentrations of p150 and p50 proteins would be decreased because they may be unstable if they cannot be assembled onto the filament, and the cellular concentration of the intact dynactin complex is also decreased. Dynactin and dynein are both multiple-protein complexes, 1 , 103 and how multiple-protein complexes are assembled in a coordinated manner is a general question in cell biology. Vezatin and its homologs, including VezA, contain predicted TM domains as well as disordered regions, and they may associate with vesicles 67 , 79 or the ER membrane ( https://humancellmap.org/explore/reports/prey?id=VEZT ). 104 Given that vesicle/organelle-bound mRNAs/ribosomes may support local protein translation, function, and/or assembly of multi-protein complexes, 53 , 105 – 108 a possibility is that dynactin components are translated near VezA/vezatin to be assembled into a complex. VezA-GFP is enriched at the hyphal tip, but this enrichment is unnecessary for dynein-dynactin function. Since VezA-GFP signals are weak, VezA-associated vesicles in other locations may be present but undetected. While the ΔTM- vezA mutant exhibited a clear defect in dynein-mediated early-endosome transport 67 (~88% of hyphal tips show an abnormal accumulation of early endosomes, n = 172), this defect is much less obvious in a strain where ΔTM-VezA is overexpressed by the gpdA promoter (~28% of hyphal tips show an accumulation, n = 248). Thus, increasing the concentration of cytosolic VezA may compensate for the loss of its membrane association, suggesting that the membrane association may serve to increase the local concentration of VezA. Mammalian vezatin also interacts with Arf6 (or ADP-ribosylation factor 6), 71 a membrane-associated protein important for dynein-mediated cargo transport. 109 , 110 However, whether VezA interacts with Arf6 (ArfB in A. nidulans ) 111 is unclear, as only one peptide of ArfB (An5020) was detected in the proteomic data from a ΔTM-VezA pull-down experiment, while five peptides of ArfA (An1126) 112 were detected ( Data S1 ). Future work will be needed to study the cellular localization of VezA for its function in dynactin assembly. Could the function of VezA in dynactin assembly be evolutionarily conserved? Recent results from Drosophila and zebrafish suggest that vezatin-like proteins play a conserved role in dynein function. 79 In addition, two independent lines of evidence suggest that mammalian vezatin also interacts with Arp1: first, vezatin (gene name: VEZT) associates with Arp1 in a human interactome study 113 (see Table 4 of our previous publication 93 ). Second, vezatin is in close proximity to Arp1 as revealed by a human BioID study ( https://humancellmap.org/explore/reports/prey?id=VEZT ). 104 Thus, it is possible that vezatin also plays a role in dynactin assembly, but future work will be needed to test this possibility. One limitation is that the direct VezA-dynactin interaction is predicted by AlphaFold2 but not shown by using purified proteins. In addition, as our Arp1 antibody does not specifically detect Arp1 in total protein extract, we could not test whether the Arp1 protein level is low in the absence of VezA. A positive result could also explain the reduction in the p50-p150 protein levels because their levels depend on Arp1.

Cytoplasmic dynein distributes organelles/vesicles/proteins/mRNAs in eukaryotic cells, and all in vivo functions of dynein need a multi-protein complex called dynactin. 1 Dynactin contains a mini-filament of the actin-related protein Arp1, whose two ends are capped by capping protein and a pointed-end sub-complex containing Arp11, p62, p25, and p27. 2 – 6 The Arp1 mini-filament and its pointed-end proteins are important for the dynein-cargo interaction via binding to cargo adapters. 6 – 16 The largest sub-unit of dynactin is p150 Glued (called p150 for simplicity), 17 , 18 which contains an N-terminal microtubule-binding domain and also interacts with the dynein intermediate chain. 19 – 24 Another key component is p50 (also called dynamitin), 25 and within dynactin, four copies of p50, two copies of p24, and two copies of the p150 C terminus form the dynactin shoulder. 3 , 6 , 9 , 26 , 27 Dynactin and cargo adapters activate dynein in vitro . 28 , 29 In the activated dynein complex, the dynein tails bind the Arp1 mini-filament, 29 – 31 and dynein motor domains are parallel to each other, allowing dynein to move along the microtubule. 32 In several cell types, dynein accumulates at the microtubule plus ends, which facilitates cargo binding or cortical interaction. 33 – 39 In filamentous fungi and mammalian neurons, the plus-end accumulation of dynein needs kinesin-1 and the dynactin complex. 37 , 40 – 48 Despite the importance of dynactin in dynein localization before cargo binding, cargo-dynein interaction, and dynein activation, we know very little about how the dynactin complex is assembled in vivo . Filamentous fungi such as Aspergillus nidulans and Neurospora crassa contain almost all the dynactin components, including all four pointed-end proteins. 10 , 47 , 49 , 50 In elongated fungal hyphae, dynein transports early endosomes and other cargos, including those that hitchhike on the motile early endosomes. 37 , 48 , 51 – 59 The dynactin pointed end (especially the p25 protein) is critical for the dynein-early-endosome interaction, 7 a function conserved in mammalian cells. 8 Genetic screens have identified the endosomal dynein adapter FHF (FTS-Hook-FHIP) complex, 60 including HookA in A . nidulans and Hok1 in Ustilago maydis . 13 , 61 In A. nidulans , HookA binds dynein-dynactin, 13 and FhipA is required for HookA to bind early endosomes. 62 The function of the FHF complex in early-endosome transport is conserved in mammalian cells, with Hook directly interacting with dynein-dynactin and FHIP directly interacting with Rab5 on the early endosome. 6 , 31 , 63 – 66 A genetic screen also identified VezA, a vezatin homolog in A. nidulans , as an important protein for dynein-mediated early-endosome transport. 67 The mammalian vezatin was discovered as a myosin-VIIA-binding protein involved in stabilizing adherens junctions. 68 It is also involved in neuronal functions and/or pathological conditions such as endometriosis and cancer. 69 – 77 Its budding yeast homolog Inp2p links myosin V to peroxisomes. 78 In A. nidulans , the Δ vezA mutant exhibits an abnormal accumulation of early endosomes at the hyphal tip, 67 due to a significant reduction in the frequency of dynein-mediated transport. 67 Vezatin homologs in Drosophila and zebrafish are also important for dynein-mediated axonal transport, suggesting an evolutionarily conserved role of vezatin in dynein function. 79 Here, we used A. nidulans to reveal the mechanism of VezA action, and our results suggest that VezA is important for dynactin assembly.

The model organism we used for this study is Aspergillus nidulans , a filamentous fungus. A. nidulans strains grow on solid rich medium made of either YAG (0.5% yeast extract and 2% glucose with 2% agar) or YAG+UU (YAG plus 0.12% uridine and 0.11% uracil) 120 . Solid minimal medium containing 1% glucose was used for selecting progeny from a genetic cross. Note that the minimal medium (pH 6.5) also contains 0.6% NaNO 3 , 0.052% KCl, 0.0152% KH 2 PO 4 and 0.051% MgSO 4 . Asexual spores of the strains can be stored in 70% glycerol at −80°C for several years. For most live-cell imaging experiments, cells were cultured overnight at 32°C in a liquid minimal medium containing 1% glycerol. For the live-cell imaging analysis on VezA-GFP in wild type and the Δ myoV mutant, cells were grown for 24 h at room temperature in minimal medium containing 0.1% glucose. For experiments involving the alcA- Arp1, alcA -Arp11, alcA -p150 and alcA -p50 conditional-null mutants, we harvested spores from the solid minimal medium containing 1% glycerol (non-repressive for the alcA promoter) and cultured them in liquid YG medium (same as YAG but without agar) containing glucose (repressive for the alcA -promoter) for protein analysis. Strains containing multiple mutant alleles were constructed by genetic crosses. Progeny with desired genotypes were selected based on colony phenotype, imaging analysis, western analysis, diagnostic PCR, and/or sequencing of specific regions of the genomic DNA. Microscopic images used in all figures (except for Figures S4 and S6 ) were generated using a Nikon Ti2-E inverted microscope with Ti2-LAPP motorized TIRF module and a CFI apochromat TIRF 100 × 1.49 N.A. (numerical aperture) objective lens (oil). The microscope was controlled by NIS-Elements software using 488 nm and 561 nm lines of LUN-F laser engine and ORCA-Fusion BT cameras (Hamamatsu). All the images were taken with a 0.1-s exposure time (binning: 2×2). Images used in Figure S4 were captured using an Olympus IX73 inverted fluorescence microscope linked to a PCO/Cooke Corporation Sensicam QE cooled CCD camera. This system also includes a UPlanSApo 100x objective lens (oil) with a 1.40 N.A., a filter-wheel system with GFP/mCherry-ET Sputtered series with high transmission (Biovision Technologies), and the IPLab software used for image acquisition and analysis. Images in Figure S6 were captured using an Olympus IX73 inverted fluorescence microscope linked to ORCA-FLASH4.0LT + sCMOS camera. An UPlanSApo 100× objective lens (oil) with a 1.40 N.A. was used. An ET-EGFP/mCherry Dual Multiband Filter Set w/BX3 cube (purchased from Olympus Inc.) was used. The cellSens Dimension Version 3 software (Olympus Inc.) was used for image acquisition and analysis. Image labeling was done using Adobe Photoshop. For all images, cells were grown at 32°C for ~18 h in the LabTek Chambered #1.0 borosilicate coverglass system (Nalge Nunc International, Rochester, NY). Images were taken at room temperature. For quantitation of signal intensity, a region of interest (ROI) was selected and maximal intensity within the ROI was measured. Then the ROI box was dragged outside of the cell to take the background value, which was then subtracted from the intensity value. Hyphae were chosen randomly from images acquired under the same experimental conditions. For measuring the signal intensity of microtubule plus-end comets formed by GFP-labeled dynein or dynactin components, only the comet closest to hyphal tip was measured. For measuring GFP-dynein signal intensity at septa, usually only the septum most proximal to the hyphal tip was measured. For line scans presented in Figure S1B , we draw a line starting from the hyphal tip in the middle of the hypha and get the average intensity value for the width of 2 μm, which is normally the hyphal width. Strains were constructed by using standard procedures used in A. nidulans . 117 , 121 , 122 For constructing the gpdA -ΔTM- vezA -GFP fusion, we first performed PCRs using genomic DNA from XY167 67 and fours oligoes: VTMNN (5′-TTCAGGGACACGCATTTGTG-3′), VTMNC (5′-TATTTGCGTGAGACCAGAGC-3′), VTMATGF (5′-ATGGAATCCCTGGTTTACGAGAA-3′) and VTMCC (5′-CGGTCTTCTTCGTTGAGGAC-3′). These PCRs generated two ~1 kb genomic fragments upstream and downstream of vezA translation start site. Using RQ247 80 genomic DNA and two oligoes, GPDFVTM (5′-GCTCTGGTCTCACGCAAATAGACTCGAGTACCATTTAATTCTATTTGTG-3′) and GPDRVTM (5′-CTCGTAAACCAGGGATTCCATTGTGATGTCTGCTCAAGCGG-3′), we amplified the ~1.23-kb fragment containing the gpdA promoter. We then used VTMNN1 (5′-GCCCTTGCTAGTCGAAGCA-3′) and VTMNC as primers to perform a fusion PCR of the three fragments and generated a ~3.3 kb fragment containing the gpdA promoter upstream of the vezA gene. By using VGORFFi (5′-CATACAGGAGGTGGAACTGGTA-3′) and VGUTRRi (5′-AACGACATCAGGAGAGTCGTC-3′) as primers, a ~4.7 kb fragment containing the C terminus of VezA linked to GFP and the selectable marker AfpyrG ( Aspergillus fumigatus pyrG ) was amplified by PCR from the genomic DNA of XY163. 67 The ~3.3 kb and ~4.7 kb fragments were co-transformed into the XY42 strain 10 containing Δ nkuA 121 and mCherry-RabA. 123 , 124 For A. nidulans transformation, 120 spores from the XY42 strain were cultured in a flask containing 50 mL YG + UU liquid medium, which was shaken overnight at 80 rpm at room temperature and then at 180 rpm at 32°C for about 1.5 h. The medium was poured off, and hyphae were then treated with about 20 mL solution containing cell-wall-lysing enzymes. This solution contains 10 mL of solution 1 (52.8 g of ammonium sulfate and 9.6 g of citric acid in 500 mL water, pH adjusted to 6.0 with 5 M KOH), 10 mL of solution 2 (5 g of yeast extract and 10 g of sucrose in 500 mL water), 0.25 mL of 1 M MgSO4, 200 mg of fraction V bovine serum albumin, 0.8 g of Extralyse (Laffort CS61611), and 0.05 mL of β-glucuronidase (Sigma, G8885). This mixture was made and filter-sterilized within 1 h before being used to treat the hyphae. After about 3 h of treatment with this solution at 32°C with shaking at 180 rpm, protoplasts were generated. The protoplasts were collected by centrifugation at 1700 rpm for 1 min using a swing-bucket rotor (Eppendorf S-4–72), washed with 15 mL of ice-cold solution 3 (26.4 g of ammonium sulfate, 5 g of sucrose and 4.8 g of citric acid in 500 mL water, pH adjusted to 6.0 with 5 M KOH), and finally suspended in 0.5 mL of ice-cold solution 5 (4.48 g of KCl, 0.75 g of CaCl 2 and 0.195 g of MES in 100 mL water, pH adjusted to 6.0 with 5 M KOH). In a 15-mL tube, 100 μL protoplast was mixed with 20 μL DNA (1–2 μg total) and 50 μL ice-cold solution 4 (25 g of PEG 6000 or 8000 (Sigma, P2139), 1.47 g of CaCl 2 .2H 2 O, 4.48 g of KCl, and 1.0 mL of 1 M Tris-HCl pH 7.5 in 100 mL water). This mixture was kept on ice for 20 min, followed by addition of 1 mL solution 4 with gentle mixing. The tube was kept at room temperature for 20 min. 10 mL of 50°C pre-melted solid medium (YAG +0.6 M KCl) was added into the tube and the mixture was poured into a Petri dish with a thin layer of the same solid medium (YAG +0.6 M KCl). After the plates were incubated at 37°C for 2–3 days, colonies of transformants appeared. Autoclaved toothpicks were used to touch the top of the individual colony and transfer the asexual spores onto a YAG plate, which was incubated at 37°C for 2 days. The transformants were then screened by microscopically observing the GFP signals, and the presence of ΔTM-VezA-GFP was confirmed by western blotting analysis with a mouse monoclonal anti-GFP antibody from Clontech. In addition, we also performed a diagnostic PCR to verify the homologous integration of the ~3.3-kb fragment using VTMNN and GPDRVTM. We also verified the homologous integration of the ~4.7-kb fragment using AfpyrG5 (5′-AGCAAAGTGGACTGATAGC-3′) and VGUTRR (5′-AGTGCTCCTGGTCAATGTCCA-3′). In the alcA -p150 strain, the alcA promoter is inserted in front of the nudM gene encoding dynactin p150 to conditionally shut off its expression. This was done by first making a ~2.4 kb fragment containing the AfpyrG selectable marker linked upstream to the alcA -promoter (AfpyrG-alcA). Specifically, we used ALCAR (5′-TTTGAGGCGAGGTGATAGGA-3′) and ALCAF (5′-AGACCGAGTGAACGTATACC-3′) as primers to amplify a ~0.5 kb alcA fragment from GR5 genomic DNA. We then used APYRGR (5′-GGTATACGTTCACTCGGTCTCTGTCTGAGAGGAGGCACTGA-3′) and APYRGF (5′-TGCTCTTCACCCTCTTCGCG-3′) to amplify a ~1.9 kb AfpyrG fragment from the plasmid pAO81. We then fused these two fragments by using APYRGF and ALCAR as primers to obtain the ~2.4 kb AfpyrG-alcA fragment. To insert this ~2.4 kb AfpyrG-alcA fragment upstream of the p150 gene, we used the following six oligoes: p150NN (5′-ATCTGTAAGGGTCGCACGG-3′), p150NN1 (5′-TTGTCGCCAGGAGAGCCTG-3′), p150NC (5′-GGTATACGTTCACTCGGTCTTGTTGTTATTGACCGCGCC-3′), p150CN (5′-TCCTATCACCTCGCCTCAAATTCCAAACACAACGGCCGT-3′), p150CC (5′-CTCAATTCGTCAAGCTCCCTG-3′) and p150CC1 (5′-ATTGCCATGTTCTGAGACTGTCG-3′). Specifically, p150NN, p150NC, p150CN and p150CC were used to amplify two ~1 kb fragments upstream and downstream of the p150 translational start site from GR5 genomic DNA. These fragments were fused to the ~2.4 kb AfpyrG-alcA fragment by a fusion PCR using p150NN1 and p150CC1 as primers. This 4.4 kb fusion fragment was transformed into the XY42 strain to get the alcA -p150 strain. Transformants that exhibited a compact-colony phenotype on the glucose-containing YAG medium were selected for PCR verification of the correct genotype. Specifically, genomic DNA was extracted and PCR reactions were performed to confirm the homologous integration event using alcA5 (5′-AGCACTTTCTGGTACTGTCC-3′) and p150CC as primers. In the alcA -p50 strain, the alcA promoter is inserted in front of the p50 gene to conditionally shut off its expression. This was done by first making a ~2.4 kb fragment containing the AfpyrG selectable marker linked upstream to the alcA -promoter (AfpyrG-alcA), as described above. To insert this ~2.4 kb AfpyrG-alcA fragment upstream of the p50 gene, we used the following six oligoes: p50NN (5′-AGGGAGGTTTGAACCATGG-3′), p50NC (5′-CGCGAAGAGGGTGAAGAGCAAGCTAGAATATTGAAGGATCTTAGTTGTC-3′), p50CN (5′-TCCTATCACCTCGCCTCAAAATGGCTTTCAACAAAAAATATGCTGGTC-3′), p50CC (5′-AAGTGCTTCAGCGTCTGCTG-3′), p50NN1 (5′-TCGGAGATGGTTCGATCCTG-3′) and p50CC1 (5′-TTCCTGACGCGGGTAGAAAG-3′). Specifically, p50NN, p50NC, p50CN and p50CC were used to amplify two ~1 kb fragments upstream and downstream of the p50 translational start site from GR5 genomic DNA. These fragments were fused to the ~2.4 kb AfpyrG-alcA fragment by a fusion PCR using p50NN1 and p50CC1 as primers. This 4.4 kb fusion fragment was transformed into the XY42 strain to get the alcA -p50 strain. Transformants that exhibited a compact-colony phenotype on the glucose-containing YAG medium were selected for PCR verification of the correct genotype. Specifically, genomic DNA was extracted and PCR reactions were performed to confirm the homologous integration event using two pairs of primers: (1) alcA5 and p50CC; (2) AfpyrG3 (5′-GTTGCCAGGTGAGGGTATTT-3′) and p50NN. In the alcA -Arp1 strain, the alcA promoter is inserted in front of the nudK (Arp1) gene to conditionally shut off its expression. This was done by first making a ~2.4 kb fragment containing the AfpyrG selectable marker linked upstream to the alcA -promoter (AfpyrG-alcA), as described above. To insert this ~2.4 kb AfpyrG-alcA fragment upstream of the Arp1 gene, we used the following six oligoes: Arp1NN (5′-TGGCAAGGACGGACAGCAG −3′), Arp1NC (5′-CGCGAAGAGGGTGAAGAGCATGCAGGGAATTGGTTGCGAG-3′), Arp1CN (5′-TCCTATCACCTCGCCTCAAAATGACCGAGGCTACTCTTCAC-3′), Arp1CC3 (5′-AATGTTGAGGTAAAGCGACTTGCG −3′), Arp1NN1 (5′-TTTGACGACTTTGTCGCAAAC-3′) and Arp1CC2 (5′-CAAGTCGAGATCCGTAGGAT-3′). Specifically, Arp1NN, Arp1NC, Arp1CN and Arp1CC3 were used to amplify two ~1 kb fragments upstream and downstream of the Arp1 translational start site from GR5 genomic DNA. These fragments were fused to the ~2.4 kb AfpyrG-alcA fragment by a fusion PCR using Arp1NN1 and Arp1CC2 as primers. This 4.35 kb fusion fragment was transformed into XY42 and RQ54 strains to get alcA -Arp1 strains. Transformants that exhibited a compact-colony phenotype on the glucose-containing YAG medium were selected for PCR verification of the correct genotype. Specifically, genomic DNA was extracted and PCR reactions were performed to confirm the homologous integration event using two pairs of primers: (1) alcA5 and Arp1CC3; (2) AfpyrG3 and Arp1NN. In the alcA -Arp11 strain, the alcA promoter is inserted in front of the Arp11 gene to conditionally shut off its expression. This was done by first making a ~2.4 kb fragment containing the AfpyrG selectable marker linked upstream to the alcA -promoter (AfpyrG-alcA), as described above. To insert this ~2.4 kb AfpyrG-alcA fragment upstream of the Arp11 gene, we used the following six oligoes: Arp11NN (5′-CAGCTTCTTTCGAGAAGTTCATG-3′), Arp11NC (5′-CGCGAAGAGGGTGAAGAGCATCTCGTCAATAAATTTCTGTGGTTGG-3′); Arp11CN (5′-TCCTATCACCTCGCCTCAAAATGTCCTCAATGTCGATCCGC-3′), Arp11CC (5′-GTATACCAGTAGAGAGATCGGC-3′), Arp11NN1 (5′-CAGTAAATCCCGATCCAAACATCAG-3′) and Arp11CC1 (5′-TTCTTGGTCGTCCAGGTCG-3′). Specifically, Arp11NN, Arp11NC, Arp11CN and Arp11CC were used to amplify two ~1 kb fragments upstream and downstream of the Arp11 translational start site from GR5 genomic DNA. These fragments were fused to the ~2.4 kb AfpyrG-alcA fragment by a fusion PCR using Arp11NN1 and Arp11CC1 as primers. This 4.35 kb fusion fragment was transformed into the XY42 strain to get the alcA -Arp11 strain. Transformants that exhibited a compact-colony phenotype on the glucose-containing YAG medium were selected for PCR verification of the correct genotype. Specifically, genomic DNA was extracted and PCR reactions were performed to confirm the homologous integration event using two pairs of primers: (1) AfpyrG3 and Arp11NN; (2) alcA5 and Arp11CC. We would like to point out that the conditional-null mutants including alcA -p150, alcA -p50, alcA -Arp1 and alcA -Arp11 constructed in this study all contain the alcA -driven gene integrated as a linear-fragment replacing the endogenous gene, which is more stable than the circular plasmid-based integration in previously constructed alcA mutants. 47 In the vezA Δ1−20 -GFP strain, the sequence encoding the first 20 amino acids of VezA is deleted, and GFP is linked to the C terminus of VezA. This was done by first making a ~2.0 kb fragment of vezA Δ1–20 . Specifically, by using GR5 genomic DNA as template, a 1.1 kb and 1.0 kb PCR products were obtained using two pairs of primers: (1) VTMNN (5′-TTCAGGGACACGCATTTGTG-3′) and VNDNC (5′-GCCAGTCTGAACTATGTTCACCCATAATTTATGCTCATTCCGTAAGCG-3′), (2) VNDCN (5′-GGTGAACATAGTTCAGACTGGC-3′) and VTMCC (5′-CGGTCTTCTTCGTTGAGGAC-3′). These two fragments were used to make a fusion PCR product of ~2.0 kb by using primers VTMNN1 (5′-GCCCTTGCTAGTCGAAGCA-3′) and VTMCC. By using VGORFFi (5′-CATACAGGAGGTGGAACTGGTA-3′) and VGUTRRi (5′-AACGACATCAGGAGAGTCGTC-3′) as primers, a ~4.7 kb fragment containing the C terminus of VezA linked to GFP and the selectable marker AfpyrG was amplified by PCR from the genomic DNA of XY163 (containing vezA -GFP). 67 The ~2.0 kb and ~4.7 kb fragments were co-transformed into the RQ54 strain. Transformants were examined microscopically, and those with an abnormal early-endosome accumulation at the hyphal tip were selected, and the correct genotype was confirmed by PCR with the primers AfpyrG5 (5′-AGCAAAGTGGACTGATAGC-3′) and VGUTRR (5′-AGTGCTCCTGGTCAATGTCCA-3′) followed by sequencing of the PCR product using primers VNDSq5 (5′ CTC ACGTCACGACTTGTCGA-3′), VTMATGF (5′-ATGGAATCCCTGGTTTACGAGAA-3′) and GFP5R (5′-CCAGTGAAAAGTTCTTCT CCTTTAC-3′). In the vezA Δ563–615 -GFP strain, the sequence encoding the last 53 amino acids of VezA is deleted, and GFP is linked to the C terminus of VezA. This was done by first making a ~3.8 kb GFP- AfpyrG -containing fragment from XY163 genomic DNA using GAGAF (5′-GGAGCTGGTGCAGGCGCTG-3′) and VGUTRR (5′-AGTGCTCCTGGTCAATGTCCA-3′) as primers. We then used TMTNC3 (5′-GGCACCGGCTCCAGCGCCTGCACCAGCTCCAGAAGCTCGCTTGTTATTTCGC-3′) and VGORFF (5′-TCGATGCTGCTGTGCTGTTGA-3′) to obtain a 0.9 kb fragment. The 3.8 kb and the 0.9 kb fragments were fused to obtain a 4.7 kb fragment in a fusion PCR reaction using primers VGORFFi (5′-CATACAGGAGGTGGAACTGGTA-3′) and VGUTRRi (5′-AACGACATCAGGAGAGTCGTC-3′). The 4.7 kb fragment was transformed into the strain XY42. Transformants were examined microscopically, and those with an abnormal early-endosome accumulation at the hyphal tip were selected, and the correct genotype was confirmed by PCR with the primers AfpyrG5 and VGUTRR followed by sequencing of the PCR product using primers VTMATGF and GFP5R. Eight primers were used to make the p50-GFP strain: P50GNN (5′-AAGATGAGATGGCGGCGTC-3′), P50GNN1 (5′-CGAGGCGAAGGACACATCA-3′), P50GNC (5′-GCTCCAGCGCCTGCACCAGCTCCCTTCCCACTCTCCAACTTCTCC-3′), P50GCC (5′-TCCAACCACACAAGGAATG-3′), P50GCC1 (5′-CATCTTTAACAGCTGCCGCC-3′), P50GCN (5′-CATCAGTGCCTCCTCTCAGACAGAGTACTTAATATAGTGTAAGGTGAGATG-3′), new pyrG3 (5′-CTGTCTGAGAGGAGGCACTGATGCG-3′) and GAGAF (5′-GGAGCTGGTGCAGGCGCTG-3′). Specifically, P50GNN and P50GNC were used to obtain a ~1 kb p50 open reading frame fragment using GR5 wild-type strain genomic DNA as template. P50GCC and P50GCN were used to obtain a ~1 kb fragment covering the 3′ untranslated region of the gene using GR5 genomic DNA as template. GAGAF and new pyrG3 were used to obtain a 3.6 kb GFP-AfpyrG fragment using the pFNO3 plasmid as template. Primers P50GNN1 and P50GCC1 were used to fuse the three fragments together by fusion PCR to obtain a 4.6 kb fragment, which was transformed into the RQ54 strain. Transformants were screened for GFP signals under microscope, and homologous integration was confirmed by PCR using AfpyrG5 and P50GCC as primers. The same eight primers were also used for making the p50-S strain. Both franking fragments were the same as above, but GAGAF and new pyrG3 were used to obtain a 3.0 kb S-AfpyrG fragment using the pAO81 plasmid as template. Primers P50GNN1 and P50GCC1 were used to fuse the three fragments together by fusion PCR to obtain a 4.0 kb fragment, which was transformed into the XY42 strain. The selected transformant was confirmed by PCR using AfpyrG5 and P50GCC as primers and by western blot analysis. For constructing the p62-GFP fusion, we used the following six primers to amplify wild type genomic DNA and the GFP-AfpyrG fusion from the plasmid pFNO3: P62GNC (5′-GGCTCCAGCGCCTGCACCAGCTCCTGAAGAACTGGAACTGCCAGC-3′), P62GNN (5′-CGAGTCTGAAGTTGCCATCATC-3′), P62GCN (5′-ATCAGTGCCTCCTCTCAGACAGTTCTCCTTCACCCTTATCTACTATATTC-3′), P62GCC (5′-GCATTGTTGTTAGGCAGTGGC-3′), GAGAF (5′-GGAGCTGGTGCAGGCGCTG-3′) and pyrG3 (5′-CTGTCTGAGAGGAGGCACTGAT-3′). A fusion PCR was performed using P62GNN and P62GCC as primers to generate the 4.7 kb P62-GFP-AfpyrG fragment that we used to transform into XY42. Transformants were screened for GFP signals under microscope, and homologous integration was confirmed by PCR using primers AfpyrG5 (5′-AGCAAAGTGGACTGATAGC-3′) and P62GCC2 (5′-TGGCAGCGAATGGAGGCATT-3′). For constructing the p25-S fusion, we used the following six primers to amplify p25 open reading frame from the XY41 strain and the S-AfpyrG fusion from the plasmid pAO81: ORFF (5′-TATGAGCTTAGCCTGCCCCAC-3′), ORFR (5′-TCGGATACTTCGATATCTCTCCCG-3′), FUSF (5′-TCGGGAGAGATATCGAAGTATCCGAGGAGCTGGTGCAGGCGCTGGAG-3′), FUSR (5′-CCGAGGCCGACTCCAAGTACAGTACCTGTCTGAGAGGAGGCACTGATG-3′), UTRF (5′-GTACTGTACTTGGAGTCGGCCTCG-3′) and Q6 (5′-CGA ATCTTCAACTCCTGGGTGCG-3′). A fusion PCR was performed using ORFF2 (5′-AGTAAGTGTACTGTGGCTTCACCG-3′) and UTRR2 (5′-GTCATTGACTATCCTGCAGGTGAG-3′) as primers to generate the 3.9 kb p25-S-AfpyrG fragment that we used to transform into RQ54. Transformants were screened for homologous integration by PCR using Q2 (5′-TGGTAATAGACGGCAGTGGG-3′) and STAG3 (5′-GCTGGCGTTCGAATTTAGC-3′) as primers. For creating a strain as a negative control for pulldown assays using anti-GFP MicroBeads, we fused the gpdA promoter with the GFP gene and used the genomic DNA from a spore-color gene wA locus as flanking regions to allow the replacement of the wA coding region by gpdA -GFP via homologous recombination (Note that the wA gene product An8209 is required for producing the yellow pigment of asexual spores but not needed for hyphal growth). To obtain the DNA fragment for transformation, the following PCRs were performed using the genomic DNA from RQ11 as template. First, wAUF1 (5′-CGCATTTGCTCAGGTCAAG-3′) and gpdAR3 (5′-GTACTCGAGTCGATCAGGAGAAGGAGAGTCAAGT-3′) were used as primers to produce a 772 bp fragment gpdG1, and gpdAF3 (5′-CTCCTGATCGACTCGAGTACCATTTAATTCTAT-3′) and gpdAR4 (5′-CTCCTTTACT CATTGTGATGTCTGCTCAAGC-3′) were used as primers to produce a ~1.25 kb fragment gpdG2. Then, gpdAF4 (5′-GCAGACATCACAATG AGTAAA GGAGAAGAACTTTTCACTGG −3′) and GFPgR (5′-AGTCCATTCCCAACTGTCTGAGAGGAGGCACT-3′) were used as primers to produce a ~2.6 kb fragment gpdG3, and GFPgF (5′-CAGACAGTTGGGAATGGACTCGCTTC-3′) and wAUR1 (5′-CGACTATTTCGCCCTGCT-3′) were used to produce a 575 bp fragment gpdG4. The gpdG1 and gpdG2 fragments were fused to make gpdG5 using fusion PCR with wAUF1 and gpdGR as primers, and the gpdG3 and gpdG4 fragments were fused to make gpdG6 using fusion PCR with gpdGF and wAUR1 as primers. Finally, gpdG5 and gpdG6 were fused using wAUF1 and wAUR1 to make the gpdA-GFP-AfpyrG fragment that we transformed to XY42. Homologous integration to the wA locus is expected to change the colony color from yellow to white. Multiple white transformants were obtained that showed very bright GFP signals under microscope, and two of them were further tested using western analysis to confirm that GFP is indeed overexpressed under the control of the gpdA promoter. To harvest cells, overnight grown fungal mycelia (or hyphal mass) were filtered through miracloth (EMD Millipore Corp.) and washed with ~200 mL distilled water. Excess liquid was removed by pressing the folded miracloth (containing the hyphal mass inside) between paper towels. After the liquid has been removed, the weight of the hyphal mass (without the miracloth) was measured on a balance, and about 0.6 g hyphal mass was collected and wrapped with a foil paper, labeled and stored in a −80°C freezer. To break hyphal cells, a mortar and pestle were used to grind the frozen 0.6 g hyphal mass to a fine powder in liquid nitrogen (Basically, liquid nitrogen was added into the mortar a little bit at a time during grinding just to keep the sample frozen). Cell extracts were prepared adding 1.5 mL of lysis buffer containing 50 mM Tris-HCl (pH 8.0), 0.01% Triton X-100 and 10 μg/mL of a protease inhibitor cocktail (Sigma-Aldrich). Cell extracts were centrifuged at 20,000 g for 60 min at 4°C, and supernatant was used for the pull-down experiment. Before the pull-down experiment, 90 μL of the ~1.5 mL total lysate was saved in −80°C and 30 μL of it (~2% of the total extract) would be used later as a total lysate sample on western blots. The μMACS GFP-tagged protein isolation kit (Miltenyi Biotec) was used to pull down proteins associated with the GFP-tagged protein, which was done as described previously. 13 To pull down GFP-tagged proteins, 35 μL anti-GFP MicroBeads were added into the cell extracts for each sample and incubated at 4°C for 60 min. The MicroBeads/cell extracts mixture was then applied to the μColumn followed by gentle wash with 200 μL of the same lysis buffer used above for protein extraction. After washing three times, 80 μL of pre-heated (95°C) 2 × SDS-PAGE sample buffer was used to elute the proteins. The eluted proteins were applied to premade 4–15% Criterion TGX Stain-Free Protein Gels (Bio-Rad), and after finishing running, proteins in the gels were transferred overnight at 4°C to nitrocellulose membrane (0.45 μm) (Bio-Rad). Western blot analyses were performed using the alkaline phosphatase system and blots were developed using the AP color development reagents from Bio-Rad. Quantitation of the protein band intensity was done using the Image Studio Lite software (version 5.2). Specifically, an area containing the whole band was selected as a region of interest, and the intensity sum within the region of interest was measured. Then, the region of interest box was dragged to the equivalent region of the negative control lane or a blank region without any band on the same blot to take the background value, which was then subtracted from the intensity sum. The rabbit polyclonal antibody against GFP was from Takara Bio Inc. (Catalog number: 632592). Monoclonal antibody against GFP was from Takara Bio Inc. (Catalog number: 632381). The rabbit monoclonal antibody against the S-tag was from Cell Signaling Technology (Catalog number: 12774S). Polyclonal antibodies against dynactin p150 and dynein heavy chain were previously generated by injecting proteins produced in bacteria into rabbits followed by affinity purification of the antibody. 47 , 116 More recently, we have generated a new polyclonal antibody against p150 by using the service of Pacific Immunology ( www.pacificimmunology.com ). Specifically, an immunograde peptide of Cys-PDRKANAVQPVEPAIEPTF from the C terminus of p150 was synthesized and conjugated to the KLH carrier protein, and this conjugated form was used as an antigen for rabbit injection. The antibody was affinity-purified using the same peptide. Similarly, a polyclonal antibody against Arp1 was generated using the service of Pacific Immunology ( www.pacificimmunology.com ). Specifically, an immunograde peptide of Cys-SADEWHEDPEIIHRKFA from the C terminus of Arp1 was synthesized and conjugated to the KLH carrier protein, and this conjugated form was used as an antigen for rabbit injection. The antibody was affinity-purified using the same peptide. Polyclonal antibody against Arp11 was generated using the service of Boster Antibody and Elisa experts ( www.bosterbio.com ). A recombinant protein of 217 amino acids (RSALVVDIGWAETVVSGIYEYREVTTKRSTRAMRSLIQETGRMFTRLLGGDSQPDTISVEFEFCEEVVSRFAWCQPSRSGYYKETAENSLADILDKTISIPSPSNPGSSDIELPFSKLEELVEKVLLAQGMADSDLDDQEKPISLLVYNTLLSLPPDVRGICMSRIVFVGGGANIAGIRSRILDEVAHLIELYGWSPVRGRLIEQQIQKLQSLKLSQ) was used for rabbit injection and purification of serum. For genomic DNA isolation, overnight grown fungal mycelia (or hyphal mass) were filtered through miracloth (EMD Millipore Corp.) and washed with distilled water. Excess liquid was removed by pressing the folded miracloth (containing the hyphal mass inside) between paper towels. About 0.25 g of hyphal mass was pulverized with liquid nitrogen by using mortars and pestles. The DNeasy Plant Mini Kit (QIAGEN) was used for genomic DNA isolation. PCRs were done using Invitrogen AccuPrime Taq DNA Polymerase, High Fidelity (12346–094) or AccuPrime Pfx DNA polymerase (12344–032). Normally, we use the AccuPrime Taq DNA Polymerase, High Fidelity (12346–094) to generate the fragments and then use AccuPrime Pfx DNA polymerase (12344–032) to do fusion PCRs. 117 , 121 , 122 The QIAquick Gel Extraction Kit (QIAGEN) was used for purifying PCR products from agarose gels. Sequencing was done using the service of Quintarabio ( https://www.quintarabio.com/service/ngs_services ). Protein structure prediction was done using ColabFold version 1.5.2 (alphaFold2_multimer_V3) in Google Cloud Platform with NVIDIA A100 80GB GPU. 119 For each protein-complex prediction, we set parameters of num_models to 5, num_recycles to 25, recycle_early_stop_tolerance to 0.5, and use_templates to false. The final models were analyzed with UCSF ChimeraX. For all the quantitative analyses of the pulldown data, the values were generated from three independent pull-down experiments ( n = 3 for all). We are mainly interested in the difference between a mutant and a wild-type control regarding how well a dynactin component (for example, p150 or Arp1) is pulled down with a tagged protein (for example, Arp11-GFP or ΔTM-VezA-GFP). The negative control (no GFP or GFP alone) is not included in the quantitative analysis as it does not show any dynactin signals in the pulldown assays. Previously, 13 a ratio of the pulled-down protein to the tagged protein was presented, which was done separately for wild type and mutants. This way of presentation is effective when multiple mutants are included. However, in this current work, we are interested in comparing multiple dynactin components in the same quantitative analysis and presenting the mutant-to-control ratio (for example, Δ vezA / vezA + or alcA- Arp1/Arp1+) for all dynactin components in the same figure. The ratio for the tagged protein is set to 1 and the ratios for the other dynactin components are normalized to the ratio of the tagged protein. This way of presentation allows a direct visualization of how a mutation affects different dynactin components differently. All statistical analyses were done using GraphPad Prism 10 for macOS (version 10.1.1, 2023). The D’Agostino & Pearson normality test was performed on all datasets except for datasets with small n ( n = 3). For western blot data, mass spectrometry data and the percentage of hyphal tips with abnormal accumulation of early endosomes, data distribution was assumed to be normal but this was not formally tested because the n number is 3 for these datasets. For these data, student t -test (unpaired, two-tailed) was used for analyzing any two datasets and ordinary one-way ANOVA (unpaired) was used for analyzing multiple datasets. Note that adjusted p values were generated from the ordinary one-way ANOVA test with Dunnett’s, Šídák’s or Tukey’s multiple comparisons test. For all other datasets, the D’Agostino & Pearson normality test was performed. The datasets that did not pass the D’Agostino & Pearson normality test (alpha = 0.05) were analyzed using the Mann-Whitney test (unpaired, two tailed), or Kruskal-Wallis ANOVA test (with Dunn’s multiple comparisons test, unpaired), non-parametric tests without assuming Gaussian distribution. The dataset that passed the D’Agostino & Pearson normality test were analyzed using the student t -test (unpaired, two tailed). For analysis of line scans, two-way ANOVA with Bonferroni’s multiple comparisons test was used. For Figures 1 , 2 , 3 , 4 , 5 , and 6 , which contain quantitative analyses using scatterplots, mean and S.D. values generated by Prism 10 were presented. In Figure 1B , the p values were generated by Kruskal-Wallis ANOVA test (with Dunn’s multiple comparisons test, unpaired). In Figure 1D , the p values were generated by Mann-Whitney test (unpaired). In Figure 1E , the datasets of Δ vezA and the gpdA -ΔC- hookA -S, Δ vezA double mutant both passed the D’Agostino & Pearson normality test, and thus, the p value was generated by the student t -test (unpaired, two tailed). The p value for the gpdA -ΔC- hookA -S, Δ vezA and gpdA -ΔC- hookA -S pair was generated by the Mann-Whitney test (unpaired). In Figure 2C , the p values were generated by ordinary one-way ANOVA with Dunnett’s multiple comparisons test. In Figure 2E , the p vales were generated by ordinary one-way ANOVA with Dunnett’s multiple comparisons test. In Figure 2H , the p values were generated by ordinary one-way ANOVA test with Dunnett’s multiple comparisons test. In Figure 3H , the p values were generated by ordinary one-way ANOVA with Tukey’s multiple comparisons test. In Figure 4B , the p values were generated by Mann-Whitney test (unpaired) except for the p150-GFP pair that was generated by the student t -test (unpaired, two tailed) as both datasets passed the D’Agostino & Pearson normality test. In Figure 4D , the p values were generated by Mann-Whitney test (unpaired). In Figure 5D , the p values were generated by Student’s t test (two tailed, unpaired). In Figure 5F , the p values were generated by ordinary one-way ANOVA with Dunnett’s multiple comparisons test. In Figure 5H , the p values were generated by ordinary one-way ANOVA with Dunnett’s multiple comparisons test. In Figure 5I , the p values were generated by ordinary one-way ANOVA with Dunnett’s multiple comparisons test. In Figure 6B , the p values were generated by ordinary one-way ANOVA with Dunnett’s multiple comparisons test. In Figure 6E , the p value was generated by Student’s t test (two tailed, unpaired). In Figure 6G , the p values were generated by ordinary one-way ANOVA with Dunnett’s multiple comparisons test. In Figure 6J , the p values were generated by ordinary one-way ANOVA with Šídák’s multiple comparisons test. In Figure 6K , the p values were generated by ordinary one-way ANOVA with Dunnett’s multiple comparisons test. In Figure 6M , the p value was generated by Student’s t test (two tailed, unpaired). In Figure 6N , the p values were generated by ordinary one-way ANOVA with Dunnett’s multiple comparisons test.