Prefabrication of a ribosomal protein subcomplex essential for eukaryotic ribosome formation

[…]

Essential revisions:

1) It is important to show by Western analysis that the failure of overexpressed catalytically dead fab1-7 to confer suppression of the Tsr2 depletion is not a trivial result of its failure to accumulate in cells.

This is an important control to exclude the possibility that a catalytically inactive Fap7 is unstable in vivo, and therefore unable to rescue the observed phenotypes associated with Tsr2-depletion. To this end, Western analyses of whole cell lysates derived from Tsr2-depleted cells overexpressing Fap7 and the catalytically inactive fap7-2 was performed. These studies show that overexpressed catalytically inactive Fap7-2 protein accumulates to levels comparable to that upon WT Fap7 overexpression (Figure 4F, left panel).

2) It is important to attempt to demonstrate, e.g. by coIP analysis, that eS26 accumulates in a complex with its importin in Tsr2-depleted cells, which is reversed by co-overexpression of Fap7 and uS11.

As requested, we have attempted to address by coIP analyses the specific concern raised by reviewer 3, point 1.

First, we analyzed eS26-FLAG protein levels in WT, Tsr2-depleted cells and Tsr2-depleted cells co-overexpressing FAP7 and uS11. As expected, in whole cell extracts (WCE) eS26-FLAG levels were strongly reduced upon Tsr2-deleption, which were restored upon FAP7 and uS11 co-overexpression (Author response image 1A).

Author response image 1

Download asset Open asset

Next, we immuno-precipitated S26-FLAG from these cells (Author response image 1B). We loaded equal amounts of eS26-FLAG isolated from each these cells on a SDS-gel and performed Western analyses to monitor importin (Pse1 and Kap123) co-enrichment (Author response image 1B). However, we were unable to co-enrich detectable levels of Pse1 and Kap123 in these IPs. One possible explanation could be that at steady state only a small amount of eS26, below the detection levels of our Western analyses, is bound to its importin during transfer to the nucleus. Alternatively, it is possible that the importin:eS26 complexes fall apart upon cell lysis.

In response to the specific sentence in the Discussion that was criticized by reviewer 3 (point 1), we would like to clarify our idea regarding Tsr2 function. Our model presented in Schütz et al., 2014 eLife was based on a combination of in vitro and in vivo findings. In that study, we showed that importin:eS26 complexes, unlike typical importin:cargo complexes, were inefficiently dissociated by RanGTP in vitro. Instead, we found that Tsr2, without the assistance from RanGTP, efficiently releases eS26 from its importin, and prevents its aggregation in vitro. in vivo, Tsr2-depletion renders eS26 susceptible to proteolysis, and precludes its incorporation into the 90S pre-ribosome. Based on these and other data, we proposed a model in which Tsr2 triggers the release of eS26 from its importin in the nucleus and therefore directly couples the process of eS26 nuclear import and transfer to the 90S.

In the current study, we revealed how Fap7 and uS11 can compensate Tsr2-requirement, both in vitro and in vivo. Specifically:

A) Fap7:uS11 recruitment to the Pse1:eS26 complex rendered the resulting Pse1:eS26:uS11:Fap7 complex sensitive to RanGTP-dependent disassembly in vitro.

B) Fap:uS11 overexpression rescued eS26 levels in vivo, and ensured that eS26 is safely transferred to the 90S in the Tsr2-depletion mutant.

We agree with reviewer 3 that our sentence in the Discussion section "Co-overexpression of FAP7 and uS11 bypassed Tsr2-requirement in extracting eS26 from its importin and safely transferring the r-protein to the 90S." is not precisely phrased. Therefore, we have corrected this to: “The Fap:uS11 complex with the assistance of RanGTP bypasses Tsr2-requirement in extracting eS26 from its importin in vitro. Co-overexpression of FAP7 and uS11 bypasses the functional requirement of Tsr2 to safely transfer the r-protein to the 90S in vivo.”.

3) The relative ratios of H23 to eS26 in lanes 8-13 of Figure 5B must be quantified.

4) It is important to investigate whether a tagged from of eS26 is capable of interacting with H23 in the absence of S11, a possibility strongly suggested by the data in Figure 5B.

We have repeated the binding assays presented in Figure 5B (see Figure 5C). In particular, we have included controls to directly compare and quantify H23 rRNA levels bound to GST-uS11 and GST-uS11:eS26 complexes. These studies revealed that, in the presence of Fap7 and ATP, nearly identical levels of H23 rRNA are recruited to GST-uS11 and GST-uS11:eS26 complexes (Figure 5C, compare lanes 5 and 7 and quantification). Further, as requested, we tested whether a tagged version of eS26 (GST-eS26) interacts with H23 rRNA in absence of uS11. We could not detect any interaction between GST-eS26 and H23 RNA (Figure 5C, lane 1-2).

Together, these data strongly support the idea that H23 rRNA specifically interacts with the GST-uS11:eS26 complex through uS11. These experiments have been included in the Results subsection “Fap7 ATPase activity organizes and recruits the uS11:eS26 subcomplex to helix 23 of 18S rRNA”.

Reviewer #1:

[…]General critique: The results are a strong combination of genetics and biochemistry, and of in vivo and in vitro analyses, which mainly support the main conclusions of the paper, and the paper is largely well written. While pertaining to only a single aspect of ribosome biogenesis, the findings will likely have broader implications for other ribosome assembly steps. There are two experimental shortcomings that should be redressed and several unclear passages that should be revised.

Subsection “Co-overexpression of Fap7 and the r-protein uS11 bypasses in vivo requirement for Tsr2”, second paragraph: It wasn't stated whether the high-copy plasmids were subcloned and re-tested to confirm that the FAP7 and uS11 genes were responsible for the suppression observed. This would be required.

This was indeed performed. We have now included a statement that FAP7 and uS11 genes were subcloned into high-copy plasmids, and their ability to rescue phenotypes associated with Tsr2-depletion was re-tested and confirmed (Results subsection “Co-overexpression of Fap7 and the r-protein uS11 bypasses in vivo requirement for Tsr2”, second paragraph).

Subsection “Fap7 ATPase activity is required to bypass Tsr2 requirement”, last paragraph: It is necessary to show that fap7-2 protein accumulates to levels comparable to that of WT Fap7 to justify the claim that loss of catalytic activity underlies the loss of suppression.

We thank the reviewer for pointing out this important control. Please see our response above to the Essential revisions section (point 1).

Reviewer #2:

[…]In my opinion the experiments support the authors' model. The manuscript could be improved by the following:

1) Introduction, second paragraph: I do not think that 200 assembly factors are required for large subunit assembly; rather, closer to 80. Can the authors double-check this number?

We apologize for this mistake, and have corrected the number. Indeed, the 60S assembly pathway requires ~80 assembly factors (Introduction, second paragraph).

2) Introduction, last paragraph, and Discussion, first paragraph: the statements imply that all r proteins assemble before nuclear export. The authors state that particles escape nuclear proofreading in the absence of Fap7. Perhaps a detail for this work, but they should clarify that a few other r proteins assemble in the cytoplasm in wild type cells, but that S11 and S26 are among the majority that assemble in the nucleus.

We agree that our statements appear to imply that all r-proteins are assembled in the nucleus. This is clearly not the case. We have now rephrased these sentences (Introduction, fourth paragraph and Discussion, first paragraph), and clearly state that we refer to only those r-proteins that need to enter the nucleus for incorporation into the developing pre-ribosome.

3) The authors might add to the Discussion:

How does their work extend and/or contradict previous work describing how Fap7 helps S11 assemble with rRNA, e.g., the work of Loc'h et al.

In addition maybe briefly compare and contrast their ideas re S11 and S26 with published models for incorporation of r protein L4 by its chaperone, e.g., where or how Fap7 recognizes and binds S11.

We thank the reviewer for this comment that has helped to improve our Discussion. We have now included a comparison for our study with those to the studies by Loc’h et al. and added a section in which our findings are put into context of how uL4 is incorporated (subsection “Fap7 prefabricates an uS11:eS26 subcomplex for ribosome assembly”, last paragraph).

4) Subsection “Two distinct mechanisms extract eS26 from the import machinery”, last paragraph: to clarify the discussion, can the authors state upfront that the existence of major and minor pools of the Fap7-S11 complex and their respective roles in assembly are a model based on results in the paper, in particular based on the suppression of Tsr2 depletion phenotypes by extra copies of both and the S11 and FAP7 genes?

We have now clearly stated that our model is based on our genetic suppression data (Discussion subsection “Two distinct mechanisms extract eS26 from the import machinery”, last paragraph).

Reviewer #3:

[…]1) The authors write "Co-overexpression of FAP7 and uS11 bypassed Tsr2-requirement in extracting eS26 from its importin and safely transferring the r-protein to the 90S." This is not actually shown but is based on extrapolation from in vitro results.

Our model regarding the function of Tsr2 presented in Schütz et al., 2014 eLife is based on a combination of in vitro and in vivo findings. In that study, we showed that importin:eS26 complexes, unlike typical importin:cargo complexes, were inefficiently dissociated by RanGTP in vitro. Instead, we found that Tsr2 without the assistance from RanGTP, efficiently releases eS26 from its importin, and prevents its aggregation in vitro. in vivo Tsr2-depletion renders eS26 susceptible to proteolysis, and precludes its incorporation into the 90S pre-ribosome. Based on these and other data, we proposed that Tsr2 triggers the release of eS26 from its importin in the nucleus and therefore directly couples the process of eS26 nuclear import and transfer to the 90S.

Here, we reveal how Fap7 and uS11 can compensate Tsr2-requirement both in vitro and in vivo. Specifically:

A) Fap7:uS11 recruitment to the Pse1:eS26 complex rendered the resultant Pse1:eS26:uS11:Fap7 complex sensitive to RanGTP-dependent disassembly in vitro.

B) Fap:uS11 overexpression rescued eS26 levels in vivo, and ensured that eS26 is safely transferred to the 90S in the Tsr2-depletion mutant.

We agree with the reviewer that our sentence in the Discussion section "Co-overexpression of FAP7 and uS11 bypassed Tsr2-requirement in extracting eS26 from its importin and safely transferring the r-protein to the 90S." does not precisely reflect these findings. Therefore, we have corrected the sentence to: “The Fap:uS11 complex with the assistance of RanGTP bypasses Tsr2-requirement in extracting eS26 from its importin in vitro. Co-overexpression of FAP7 and uS11 bypasses the functional requirement of Tsr2 to safely transfer the r-protein to the 90S in vivo.”

Does eS26 accumulate in a complex with its importin in Tsr2 depleted cells and is this relived by overexpression of Fap7 and uS11? This should be tested in vivo by co-IP.

Please see our response above to the Essential revisions section (point 2).

This point is important not just to justify this statement but also because this is a major conclusion of the work that is based almost entirely on in vitro analysis.

We politely disagree with this reviewer’s opinion regarding the major conclusion of this work. The main finding of this study is that the ATPase Fap7 brings together uS11 and eS26 interactions through native-like tertiary contacts and its ATPase activity is required to integrate a pre-formed uS11:eS26 complex into the 90S. This ensures stoichiometric integration of uS11 and eS26 into the 90S pre-ribosome. Different lines of genetic, cell-biological and in vitro /in vivo biochemical evidences support this conclusion. Specifically:

A) Tsr2-depletion compromises eS26 stability and precludes uS11 integration into the 90S. Fap7-depletion compromises uS11 stability and precludes eS26 integration into the 90S. These in vivo data demonstrate an interdependence of eS26 and uS11 for their incorporation into the 90S.

B) Together with the above data and the specific genetic interaction between Tsr2 and Fap7:uS11 (suppression data) led us to show that Fap7 organizes a uS11:eS26 complex through tertiary contacts and that its ATPase activity is required to load this complex onto H23 rRNA. The importance of native-like contacts between uS11:eS26, and the importance of Fap7 ATPase activity to load H23 rRNA were validated by genetic, cell-biological and in vivo biochemical studies.

2) Figure 5B, the authors claim that the loading of uS11 onto H23 is dependent on Fap7 and recruitment of eS26. The dependence on Fap7 is quite nice but the retention of H23 seems to correlate closely with eS26 levels. Is H23 binding to uS11 as claimed or to eS26? The relative ratios of H23 to eS26 in lanes 8-13 should be quantified. The authors must also include controls of eS26 alone with H23.

We thank the reviewer for pointing out this important control. Please see our response to points 3 and 4 in the Essential revisions above.

3) If uS11 needs Fap7 for its stability, as suggested by Figure 2F, the model for independent import of uS11 seems unlikely. When uS11 accumulates in the cytoplasm in pse1-1 cells, is it associated with Fap7? Do they co-IP? If not, how is it imagined that uS11 is imported without Fap7 to stabilize it?

As requested, we have immuno-precipitated uS11-GFP from WT and pse1-1 cells (Author response image 2A). These biochemical studies show that uS11-GFP can co-enrich Fap7. It is unlikely that all uS11-GFP in the cytoplasm is bound to Fap7, since nuclear targeting of GFP-Fap7 is not impaired in a pse1-1 mutant.

Author response image 2

Download asset Open asset

Similarly, when Fap7 accumulates in the cytoplasm in RanGTP cycle mutants, is it associated with uS11?

As requested, we have immuno-precipitated Fap7-TAP from WT and prp20-1 cells (Author response image 2B). These biochemical studies show that Fap7 can co- enrich uS11.

Newly synthesized uS11, after emerging from the translating ribosome, is likely to be stabilized by cytoplasmic factors before being captured by importins. Yeast cells deficient for the ribosome associated NAC and SSB-RAC chaperone systems that associate with newly synthesized polypeptides accumulate several ribosomal protein aggregates (Koplin et al., 2010). NAC and SSB-RAC chaperone systems might bind uS11 before their transport to the nucleus. Upon arrival in the nuclear compartment interaction with Fap7 stabilizes uS11 and prepares the r-protein together with eS26 for assembly.

4) Figure 6. It's not clear what the significance is of uS11 interacting with many different importins. These proteins are typically highly negatively charged. Is it possible these simply represent nonspecific electrostatic interactions driven by the high in vitro protein concentrations?

We could not test interactions between uS11 and the different importins, since uS11 is aggregation-prone when not bound to Fap7. All interaction studies between importins and Fap7 and Fap7:uS11 complexes are well-established binding assays in the nuclear transport field. These assays are always carried out in the presence of competing E. coli lysates to minimize non-specific interactions.

Budding yeast employs 11 importins to target diverse cargos to the nucleus. In this study, we have tested interactions between Fap7:uS11 complex and Fap7 with all 11 importins. Amongst the 11 importins, we observe differences in binding. For e.g. Sxm1, Kap120, Mtr10 and Nmd5 hardly interact with Fap7 as well as the Fap7:uS11 complex. In contrast, Msn5 efficiently binds to Fap7, but rather weakly with the Fap7:uS11 complex. Finally, only Pse1 and Kap104 binds to the Fap7:uS11 complex. Therefore, we do not think that these interactions are non-specific electrostatic interactions.