RUFY3 and RUFY4 are ARL8 effectors that couple lysosomes to dynein-dynactin


 The small GTPase ARL8 associates with lysosomes and recruits several effectors that mediate coupling to kinesins for anterograde transport, as well as tethering for eventual fusion with other organelles. Herein we report the identification of the “RUN- and FYVE-domain-containing” proteins RUFY3 and RUFY4 as novel ARL8 effectors that couple lysosomes to dynein-dynactin for retrograde transport. Using various biochemical approaches, we find that RUFY3/4 interact with both GTP-bound ARL8 and dynein-dynactin. In addition, we show that RUFY3/4 are both necessary and sufficient for concentration of lysosomes in the juxtanuclear area of the cell. RUFY3/4 also promote retrograde transport of lysosomes in the axon of hippocampal neurons. The function of RUFY3/4 in retrograde transport is required for juxtanuclear redistribution of lysosomes upon serum starvation or cytoplasmic alkalinization, and may underlie the reported roles of RUFY3/4 in axon development/degeneration, cancer and immunity. These studies thus establish RUFY3/4 as novel ARL8-dependent, dynein-dynactin adaptors, and highlight the role of ARL8 in the regulation of both anterograde and retrograde lysosome transport.  


Introduction 36 37
The ADP-ribosylation factor (ARF) family of small GTPases comprises ~30 members that 38 regulate various aspects of cell physiology (Sztul et al., 2019). Among these members, the 39 mammalian ARL8A and ARL8B paralogs (referred to indistinctly as "ARL8" unles otherwise 40 specified) are unique in their ability to associate with lysosomes and to regulate multiple 41 lysosomal functions (Khatter et al., 2015b) (for simplicity, herein we use the term "lysosomes" to 123 124 Interestingly, another top hit for both the ARL8A (Fig. 1b) and ARL8B (Fig. 1c)

138
(e.g., https://www.proteinatlas.org/) indicate that RUFY3 is expressed in all cells and tissues, 6 although with higher expression levels in the brain. Our mass spectrometric analyses identified 140 6 peptides derived from the longer RUFY3.1 spliceform ( Supplementary Fig. 1b), demonstrating 141 that this species is expressed in HEK293T cells.

143
To confirm the identification of RUFY3 as an ARL8 effector and to determine whether other 144 members of the RUFY family also interact with ARL8, we examined the intracellular 145 localization of GFP-tagged forms of the RUFY proteins co-expressed with MTS-BioID2-ARL8B-146 T34N and Mito-BioID2-ARL8B-Q75L in HeLa cells (Fig. 1e,f). We observed that, in the presence

154
To further corroborate these findings, we performed pull-down assays using recombinant GST-155 ARL8B proteins and FLAG-tagged RUFY proteins expressed by transient transfection in

167
Taken together, these assays demonstrated that both RUFY3.1 and RUFY4 have the ability to 168 interact with GTP-bound, but not GDP-bound, ARL8, suggesting that they behave as bona fide 169 ARL8 effectors. RUFY1 and RUFY2, on the other hand, did not bind to any form of ARL8, ruling 170 out their function as ARL8 effectors. Because RUFY3.2 was cytosolic and did not interact with 171 ARL8, this spliceform was omitted from subsequent experiments, and RUFY3.1 was simply 172 referred to as RUFY3. RUFY4 (also known as ZFYVE31) is expressed at low levels in most 173 tissues and cells, with the exception of the brain, lung and lymphatic organs 174 (https://www.proteinatlas.org/search/rufy4), probably explaining why it was not identified in the MitoID experiments using HEK293T cells. Nevertheless, because it also behaves as an ARL8 176 effector, we performed some experiments with this protein.

178
The CC2 domain of RUFY3 is required for binding to ARL8

179
We next sought to identify the region of RUFY3 that mediates interaction with ARL8. To this 180 end, we generated deletion constructs of RUFY3-GFP (Fig. 2c) and co-expressed them with 181 MTS-BioID2-ARL8B-Q75L in HeLa cells (Fig. 2d). Interaction with ARL8 was inferred from re-182 localization of the RUFY3 constructs to mitochondria. By analogy with SKIP, which interacts    (Fig. 3a). We observed that both RUFY3-GFP and RUFY4-206 GFP co-localized with ARL8B-mCherry to a cluster of vesicles adjacent to the nucleus (

212
To further dissect the requirement of ARL8 binding for RUFY3 recruitment to vesicles, and the 213 domains of RUFY3 required for this recruitment, we examined the intracellular localization of 214 the RUFY3 deletion mutants depicted in Fig. 2c. We observed that RUFY3 constructs lacking the 215 RUN and/or CC1 domains were largely associated with the juxtanuclear cluster, whereas those 216 lacking the CC2 or FYVE domains were more cytosolic (Fig. 3d). Combined deletion of the CC2 217 and FYVE domains resulted in a protein that was largely cytosolic (Fig. 3d)

238
Conversely, siRNA-mediated knock down of RUFY3 in HeLa cells (Fig. 4g) caused dispersal of 239 LAMP1 toward the cell periphery ( Fig. 4h-j). RUFY4 mRNA could not be detected by qRT-PCR 240 of HeLa cells (Fig. 4g), consistent with the low expression levels of this mRNA in most cell lines 241 (https://www.proteinatlas.org/ENSG00000188282-RUFY4/celltype). For this reason, the effect 242 of RUFY4 knock down in these cells was not tested.

244
To determine if the effects of RUFY3 and RUFY4 on the distribution of lysosomes resulted from 245 changes in lysosome transport, we examined the co-localization and movement of vesicles 246 labeled with fluorescently-tagged RUFY3, RUFY4, ARL8B and LAMP1 in the axon and 247 dendrites of rat hippocampal neurons in primary culture, where vesicle movement can be more 248 readily tracked (Fig. 5). We observed that both RUFY3-FLAG and RUFY4-FLAG co-localized 249 with a subpopulation of vesicles containing ARL8B-mCherry and LAMP1-GFP (Fig. 5a,b), as 250 well as the endogenous lysosomal marker LAMTOR4 (Fig. 4c,d) (Fig. 6a). In addition, purified, recombinant 6His-GFP-RUFY3 pulled 270 down both endogenous DIC and p150 Glued from an extract of HEK293T cells (Fig. 6b).

271
Recombinant 6His-GFP-RUFY4 was degraded and could not be analyzed using this latter assay.

272
Finally, we found that purified, recombinant 6His-GFP-RUFY3 could be pulled down with the

279
To test the functional relevance of interactions of RUFY3 and RUFY4 with dynein-dynactin in 280 cells, we compared the distribution of RUFY3-mCherry and RUFY4-mCherry in the absence or 281 presence of overexpressed GFP-tagged CC1 domain of p150 Glued , a construct that functions as a 282 dominant-negative inhibitor of dynein-dynactin (Quintyne et al., 1999) (Fig. 6d). We observed 283 that, in the absence of GFP-p150 Glued -CC1, RUFY3-mCherry and RUFY4-mCherry localized to a 284 juxtanuclear cluster, whereas in the presence of GFP-p150 Glued -CC1, RUFY3-mCherry and 285 RUFY4-mCherry were associated with peripheral clusters, often found at cell tips (Fig. 6d).

286
These observations demonstrated that interference with dynein-dynactin does not prevent 287 association of RUFY3 and RUFY4 with lysosomes, but precludes their ability to move lysosomes 288 toward the cell center.

292
To determine whether RUFY3 and RUFY4 are sufficient for organelle coupling to dynein-293 dynactin, we next used a peroxisome re-localization assay (Kapitein et al., 2010). Peroxisomes 294 are particularly suited for this assay because they are not very motile. The assay consisted of co-295 expressing i) a peroxisomal targeting signal from PEX3 (amino acids 1-42) fused to FKBP and 296 RFP, together with ii) RUFY3 or RUFY4 fused to FRB and GFP (Fig. 7a). As a positive control for 297 a known dynein-dynactin adaptor, we used a BICD2 25-400 -FRB-GFP construct (Fig. 7a). Addition 298 of rapalog brings together the FRB and FKBP domains, leading to the targeting of RUFY3 or 299 RUFY4 to peroxisomes (Fig. 7b). We observed that, in the absence of rapalog, peroxisomes

364
The 571-amino-acid RUFY4 protein had been previously shown to be expressed mainly in lung

898
We used One-way ANOVA when multiple groups were compared and unpaired Student's t-899 test when two groups were compared. Data in Figs. 4c,e,h,I were normalized to GFP. This was 900 done to account for experiment-to-experiment variability (the trends within each experiment 901 was always consistent).