Gle1 is required for tRNA to stimulate Dbp5 ATPase activity in vitro and to promote Dbp5 mediated tRNA export in vivo

Cells must maintain a pool of processed and charged transfer RNAs (tRNA) to sustain translation capacity and efficiency. Numerous parallel pathways support the processing and directional movement of tRNA in and out of the nucleus to meet this cellular demand. Recently, several proteins known to control messenger RNA (mRNA) transport were implicated in tRNA export. The DEAD-box Protein 5, Dbp5, is one such example. In this study, genetic and molecular evidence demonstrates that Dbp5 functions parallel to the canonical tRNA export factor Los1. In vivo co-immunoprecipitation data further shows Dbp5 is recruited to tRNA independent of Los1, Msn5 (another tRNA export factor), or Mex67 (mRNA export adaptor), which contrasts with Dbp5 recruitment to mRNA that is abolished upon loss of Mex67 function. However, as with mRNA export, overexpression of Dbp5 dominant-negative mutants indicates a functional ATPase cycle and that binding of Dbp5 to Gle1 is required by Dbp5 to direct tRNA export. Biochemical characterization of the Dbp5 catalytic cycle demonstrates the direct interaction of Dbp5 with tRNA (or double stranded RNA) does not activate Dbp5 ATPase activity, rather tRNA acts synergistically with Gle1 to fully activate Dbp5. These data suggest a model where Dbp5 directly binds tRNA to mediate export, which is spatially regulated via Dbp5 ATPase activation at nuclear pore complexes by Gle1.


27
Key to the production of proteins is the delivery of amino acids to ribosomes by transfer RNAs 28 (tRNAs). Precursor tRNAs (pre-tRNAs) are first transcribed by RNA Polymerase III with a 5´ leader and

32
Importantly, it has come to be appreciated that these steps in tRNA processing are not necessarily

74
Similarly, in addition to Mex67 and Crm1, the DEAD-box Protein 5 (Dbp5) has also been implicated in 75 each of these RNA export pathways, including tRNA export (   identified (e.g., dbp5 L12A ) that disrupt a nuclear export sequence (NES) recognized by Crm1 to promote 97 Dbp5 transport out of the nucleus. In contrast, another mutant, dbp5 R423A , was found to be impaired in 98 nuclear import from the cytoplasm. Importantly, neither of these mutations disrupt the essential mRNA 99 export functions of Dbp5; however, limited nuclear access of dbp5 R423A induced tRNA export defects 100 suggesting a potentially novel role for nuclear Dbp5 in tRNA export. In support of this hypothesis, a 101 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made

103
In this study, genetic and biochemical characterization of Dbp5 mediated tRNA export was 104 performed. The data show that Dbp5 is recruited to pre-tRNA independent of Los1, Msn5, and Mex67, 105 positioning the DEAD-box protein function parallel to the primary Los1 mediated pre-tRNA export 106 pathway. In contrast to single stranded RNA substrates (e.g., mRNA), this study further demonstrates an

126
. CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made

151
. CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted June 29, 2023. ; https://doi.org/10.1101/2023.06.29.547072 doi: bioRxiv preprint Arvind Arul Nambi Rajan 8 experiments were performed with protein-A (prA) tagged Dbp5, integrated at its endogenous locus and 152 present as the sole copy of the gene, in strains where tRNA export factors were deleted. Co-153 immunoprecipitated RNAs were analyzed by RT-qPCR with primers specific to un-spliced intron-154 containing pre-tRNA Ile UAU. The abundance of the pre-tRNA Ile UAU target in each IP was normalized to the 155 abundance of the target in the corresponding input sample to control for changes in gene expression.

156
Relative enrichment of the target RNA was then compared to the background signal obtained from RNA

157
IPs in a common untagged control. In a los1D strain, a ~2-fold reduction in the relative amount of pre-158 tRNA co-immunoprecipitated with Dbp5 (~9.5-fold enrichment in WT to ~4.5-fold in los1D) was observed,  and support the hypothesis that Dbp5 functions in tRNA export parallel to Los1.

164
In mRNA export, Mex67 is a proposed target of Dbp5 activity to promote directional nuclear export Mex67-AA resulted in strongly reduced binding of Dbp5 to the FBA1 mRNA (from a median of ~8 fold 180 enrichment above background before addition of rapamycin to no enrichment after, Figure 2C). Together, 181 these RNA IP experiments suggest that Dbp5 binds to tRNAs independent of known tRNA export proteins 182 and does so in a manner that is distinct from mRNA.

197
Based on the observation that Gle1 functions to support tRNA export (Lari, Arul Nambi Rajan et 198 al. 2019), these Dbp5 mutants deficient in ATPase activity or altered Gle1 binding were tested for their 199 influence on tRNA export to infer if tRNA export is more similar to mRNA or pre-ribosomal subunit export.

200
To individually express these lethal mutants and the wildtype control, DBP5, dbp5 R426Q , dbp5 R369G , or 201 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted June 29, 2023. ; https://doi.org/10.1101/2023.06.29.547072 doi: bioRxiv preprint Arvind Arul Nambi Rajan 10 dbp5 E240Q were integrated at the URA3 locus under regulation of the inducible pGAL promoter. Protein 202 expression was induced for 6 hours by shifting cells from raffinose to galactose containing media.

203
Importantly, induction was followed by 1 hour of growth in glucose to halt expression and relieve potential 204 changes in tRNA export induced by a change in the primary carbon source. Using this approach, all 205 proteins were expressed as indicated by western blotting ( Figure 3A) and showed the previously reported    (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted June 29, 2023. ; https://doi.org/10.1101/2023.06.29.547072 doi: bioRxiv preprint Arvind Arul Nambi Rajan 12 substrate (i.e., pA) that can maximally stimulate ATP turnover to ~0.5 ATP/sec, neither tRNA nor poly(I:C)

252
(used as an orthogonal dsRNA substrate) stimulated Dbp5 ATPase activity over a range of RNA 253 concentrations ( Figure 5A). These results with the above binding data suggest Dbp5 can engage 254 structured and dsRNA substrates, but this does not lead to productive ATP hydrolysis.

255
The nucleoporin Gle1 with the small molecule inositol hexakisphosphate (InsP6) has been shown 256 to synergistically activate Dbp5 ATPase activity at low RNA concentrations (Weirich, Erzberger et al.

266
To confirm that this enhanced RNA activation by Gle1/InsP6 was not the result of contaminating ssRNA,

267
ATPase assays were performed after treatment of tRNA or poly (I:C) with RNase T1 for 2 hours at 37°C.

325
and then be fully activated by Gle1/InsP6, presents the potential for a novel mechanism of function in 326 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted June 29, 2023. ; https://doi.org/10.1101/2023.06.29.547072 doi: bioRxiv preprint Arvind Arul Nambi Rajan 15 regulating tRNA export. Namely, that Dbp5 may enter the nucleus to stably engage tRNA and direct 327 events leading to export, following which Dbp5 is removed from the tRNA by Gle1/InsP6 upon nuclear 328 exit ( Figure 6). Importantly, in both mRNA and tRNA export, Gle1/InsP6 activation is required to stimulate 329 ATPase activity and recycle the enzyme. In contrast, for pre-ribosomal subunit export Dbp5 ATPase

334
In mRNA export, it has been proposed that Dbp5 displaces mRNA export factors such as Mex67

336
Observations that Mex67 and Dbp5 appear to both function parallel to Los1 to promote tRNA export may

376
. CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted June 29, 2023.

386
. CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted June 29, 2023. ; https://doi.org/10.1101/2023.06.29.547072 doi: bioRxiv preprint Arvind Arul Nambi Rajan 18

389
A list of all yeast strains used in this study is provided in Table S1. Deletion mutants and C-terminal

433
. CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made

473
were used for disrupting cells by vortexing 5 times for 30 seconds followed by 1min on ice between each 474 pulse. Lysate was pre-cleared by centrifugation at 4000xg 5min followed by 20min at 20000xg at 4°C.

475
1% pre-IP lysate was preserved as input RNA sample and additional pre-IP sample was reserved for 476 western blotting. Remaining lysates were then diluted to 10ml with TN150 and incubated with IgG-477 conjugated magnetic dynabeads at 4°C for 30min with constant rotation. Immunoprecipitate was washed 478 once with 1ml TN150, once with 1ml TN1000 (50mM Tris-HCl pH7.8, 1M NaCl 0.1% IGEPAL, 5mM beta-479 mercaptoethanol), and again once with 1ml TN150 each for 5min at 4°C. Beads from RIPs were then 480 resuspended in 500ul proteinase K elution mix (50mM Tris-HCl pH 7.8, 50mM NaCl, 1mM EDTA, 0.5% 481 SDS, 100ug proteinase K) in parallel with input samples and incubated at 50°C for 2 hours followed by 482 65°C for 1 hour to allow crosslink reversal. RNA was isolated by extraction with acidic 483 Phenol:Chloroform:Isoamylalcohol (pH 4.3-4.7) and ethanol precipitation. Half volume of RNA was 484 . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made   (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted June 29, 2023. ; https://doi.org/10.1101/2023.06.29.547072 doi: bioRxiv preprint . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint this version posted June 29, 2023. ; https://doi.org/10.1101/2023.06.29.547072 doi: bioRxiv preprint