SKP2 attenuates autophagy through Beclin1-ubiquitination and its inhibition reduces MERS-Coronavirus infection

Autophagy is an essential cellular process affecting virus infections and other diseases and Beclin1 (BECN1) is one of its key regulators. Here, we identified S-phase kinase-associated protein 2 (SKP2) as E3 ligase that executes lysine-48-linked poly-ubiquitination of BECN1, thus promoting its proteasomal degradation. SKP2 activity is regulated by phosphorylation in a hetero-complex involving FKBP51, PHLPP, AKT1, and BECN1. Genetic or pharmacological inhibition of SKP2 decreases BECN1 ubiquitination, decreases BECN1 degradation and enhances autophagic flux. Middle East respiratory syndrome coronavirus (MERS-CoV) multiplication results in reduced BECN1 levels and blocks the fusion of autophagosomes and lysosomes. Inhibitors of SKP2 not only enhance autophagy but also reduce the replication of MERS-CoV up to 28,000-fold. The SKP2-BECN1 link constitutes a promising target for host-directed antiviral drugs and possibly other autophagy-sensitive conditions.

4. The role of ubiquitination and the proteins that regulate it are the focus of Figure 3. In Figure 3 the control used in figure 3I is 3-MA treatment. First, how much 3-MA was used? Specificity of its inhibition diminishes at higher concentrations above 10uM. The data presented show that 3-MA is not working in this assay, as there is no decrease in long-term protein degradation observed unlike siSKP2 which shows an increase in long-term protein levels when knocked down. Why should 3-MA not show an increase in this experiment? This is the same result in Figure 4D where 3-MA has no effect on protein levels. This is not discussed in the text. Figure 4, there are several controls missing in the figure. First, there are no toxicity controls showing that the range of each inhibitor used is below significant cell toxicity levels. Secondly, the control in 4D using HBSS as an inducer of autophagy is not shown. Please show experimental evidence that in this experiment HBSS induces autophagy. Third, in 4E, BECN1 levels increase with all drugs; however LC3B increases with only 2 of them, and there doesn't seem to be a change in P62. There should be some correlation between these results but there is no consistency.

For
6. As MERS-CoV infection and proteins are considered for their role in autophagy, there are several conflicting results. First, the figure numbering does not match the text. Secondly, in Figure 5E, for the WCE blot it is not clear how you can observe both SNAP29 and STX17 in the same blot where it does not show up in the IP blots. Shouldn't they be in the same complex since they are crosslinked so they show as 1 band? The order of 5F and G should be reversed. And 5F is not clear. I assume it is a quantitation of 5G, however it does not match the data in 5G. The relationship between ATG14 monomeric and oligomeric forms and autophagy is not made clear. Are there published data showing oliogomeric ATG14 actively inhibits autophagosome to lysosome fusion, or simply that it fails to promote fusion? Also, if MERS is inhibiting oligomerization, I would expect there to be an increase in monomer, which is not seen. Does total ATG14 amount change? Please clarify. figure S3 is also confusing. First, the cropped cells in S3E do not line up, please clean up the cropping. Secondly, there does not seem to be an increase yellow staining and a correlated decrease in red staining in the single cell shown in each panel. This example cell does not correlate with F. Third, the mutant viruses with either 4b or 5 deleted are great, however does the 4b deletion virus still contain the entire 4a coding region or was is 4a ORF effected when 4b was deleted? The oligos are in the methods section but no explanation in the text to the sites of the deletion. There are also no growth curves showing the 2 deletion viruses compared to WT virus. Do they grow equally well in the experiments or are they attenuated in their growth under control conditions? Fourth, the virus output that is examined in K is only looking at mRNA with no statistically significant different noted however it is stated in the text that the SKP2i inhibitor effects MERS growth. The K panel seems to be missing data. And the correlation should be with virus titer not mRNA since the autophagy effect would be expected to be on virus egress not transcription?

The MERS-CoV infections experiments in the supplemental
8. In the model figure and in the discussion, there is no description of data presented to show the direct effect of SKP2 on ATG14 oligomerization. Since this is a key feature of the model, it would be important to show that SKP2 siRNA or inhibition by drugs effects the ATG14 oligo/monomer relationship.
Minor Issues 1. Figure 1B. The whole cell lysates shown in the 2nd and 3rd lane for BECN1 (3rd blot down) are different however they are the same in the IP blot for BECN1 (2nd blot down). Please correct or explain why there are differences.
Reviewer #2: Remarks to the Author: Gassen et al reported here a very complete study of a novel regulation of autophagy by SKP2, a E3 ubiquitin ligase. They nicely demonstrated that SKP2 targets BECN1, an autophagy protein which is involved in autophagosome formation and maturation in two different complexes. SKP2 executes K48-ubiquination at K402 of BECN1, allowing its proteosomal degradation. Activity of SKP2 is negatively controlled by FKBP51, which regulates SKP2 phosphorylation. Pharmacological and genetic inhibition of SKP2 leads to autophagy activation, with increased BECN1 levels, and increased interaction between lysosomal and autophagy SNARE proteins. They identified and studied a pharmacological compound, SMIP004, which strongly inhibits SKP2. The results presented her are novel, technically sound and contribute to a better understanding of autophagy regulation.
A second part of the article is dedicated to a possible therapeutic application of their data on Middle East respiratory syndrome (MERS) coronavirus. MERS-CoV is a human betacoronavirus responsible of a severe viral respiratory disease, first identified in Saudi Arabia in 2012. A third of reported patients with MERS have died and no antiviral treatment is currently available. They first demonstrated that MERS CoV blocks autophagy, in particular the fusion between autophagosome and lysosome, using several reliable autophagy monitoring experiments. They identified three nonstructural viral proteins which can recapitulate the block of the autophagic flux observed during viral infection. They constructed recombinant viruses deleted for these genes but unfortunately poorly characterized their phenotype.
Finally they tested the effect of this SKP2 inhibitor, SMIP004 at 10µM, on MERS-CoV infection and observed an important inhibition of its multiplication. Unfortunately, they only evaluated viral genome production by RT-PCR at two different times post infection. Virus titers were not determined by plaque assay or TCID50 and no growth kinetics are presented, as expected for this kind of studies. Because a similar inhibition was observed with other compounds which interact with BECN1 and stimulate autophagy, they conclude that SMIP004 acts by stabilizing BECN1. They also tested FDA approved compounds reported to inhibit SKP2 and identified for several of them an antiviral activity against MERS-CoV. They conclude from this part of the work that inhibition of SKP2 leads to an important restriction of MERS CoV multiplication and represents an interesting therapeutic approach. This second part is quite interesting but suffers from speculative interpretations and lack of data from the literature. Some important experiments regarding the mechanism of antiviral activity of SMIP004 are missing and should be added to support their interpretation of the data.
In conclusion, this article reports totally new and important data on both autophagy regulation by SKP2 and on interaction between MERS-CoV and autophagy and a possible antiviral role of autophagy, which can be used as an innovative therapeutic approach.

Specific remarks:
Page 4 "CoV propagation and immune evasion is highly dependent on the formation of convoluted membranes and double membrane vesicles (DMVs) that are reminiscent of autophagosomes (ref 17) pointing to a potential role of autophagy during the CoV life cycle." In the introduction the authors improperly and incompletely reported data from the literature regarding impact of autophagy on coronavirus infection. Indeed, the presence of DMVs during Mouse Hepatitis virus infection (another betacoronavirus) has been reported in 2002 in the reference 17 cited by the authors. However, a more recent study, Regiorri et al demonstrated that DMVs are coated with the nonlipidated form of LC3 (LC3-I) and that LC3-II the autophagic marker and autophagy have no role on viral production. This was confirmed in the journal Autophagy by Zhao et al who demonstrated that MHV replication does not require Atg5. Therefore, it is established that autophagy has no role on MHV replication. However, it has been reported that autophagy negatively regulates TGEV replication (Sci Rep 2016). But another study reported a proviral role of autophagy on TGEV (Oncotarget 2016). Indeed, TGEV infection induces mitophagy to promote cell survival and possibly viral infection.
Page 11 "These results points to a block of autophagic flux by MERS-CoV infection at both early (BECN1) and late steps in the pathway" Whereas data clearly demonstrated that MERS-CoV infection blocks autophagic flux and leads to an accumulation of immature autophagosomes, no evidence demonstrates that early steps of autophagy are reduced by MERS-CoV infection. A decrease of BECN1 expression is not sufficient to conclude this, when in parallel they observed an increase of the total number of autophagic vesicles during infection compared to control cells ( Fig S3F). In the article (Cottam et al 2014, ref 41), the authors concluded that the non-structural protein nsp6 of IBV, an avian coronavirus, restricts autophagosome expansion by immunofluorescence by measuring the diameter of the autophagosomes. The authors need therefore to modify this sentence.

Page 14
The results presented in Figure 5 clearly demonstrated that MERS-CoV blocks autophagosome maturation rather that initial steps of autophagosome formation. The titer of Figure 5 is therefore not adapted to the results and should be replaced by "blocks autophagic flux".
Page 15 and Figure 6 panels A and B "In order to confirm the influence of p4b and p5 on virus replication, we performed loss of…for the step-wise construction of an ORF4b-and ORF5-deleted MERS-CoV" "The deletion of p4b or p5 resulted in 10-fold lower replication efficiency 48 hours p.i. (Fig. 6B)" The characterization of the recombinant viruses presented in this study is clearly insufficient. No genome sequence analysis was reported. Growth kinetics of the different viruses showing viral titers (pfu/ml) at different times need to be added instead of measuring an inhibitory effect of the deletion ( fig 6B). More importantly, the authors did not report that in two previous independent studies (ref 44 of this article and Almazan et al, Mbio, 2013), construction of equivalent mutant MERS-CoV (no expression of orf5 or orf4ab) led to different results. Indeed, recombinant viruses lacking orf5 expression replicated to high titers (44). Similarly, Almazan et al observed that virus titers, cpe and plaque morphology of MERS-delta5 are similar to that of parental virus. In the case of MERS-delta4ab the viral titer was only 10 fold lower at 72hpi. How the authors explain this discrepancy? This can be related to an improper way to quantify viral production. Usually viral titers are determined by plaque assay (pfu/ml) in growth kinetics at different times of infection. In this study, they use RT-PCR in supernatant to detect viral genome but this experiment does not measure viral infectivity. However, no quantification of LC3-II/LC3-I is available and the expression level of LC3-I seems important compared to other WB of the article.
Panel B and C: How is the multiplication of these mutant viruses in autophagy deficient cells? If growth kinetics of mutant viruses are similar to that of parental virus, this would allow the authors to confirm the antiviral role of autophagy on MERS-CoV and to reveal a possible requirement of orf4b or orf5 to counteract it.

Figure 6
The titer of the legend does not reflect the results. No data in this figure demonstrates that SKP2 inhibition reduces viral replication.

Page 15
Regarding SKP2 inhibitor, effect on viral replication is impressive, but same remark than above, growth kinetics and determination of viral titers are necessary. Moreover, to prove that the compound is acting via autophagy, effect of SMIP004 should be tested in autophagy deficient cells. This kind of approach was used for antibacterial activity of tat-beclin1 for example, in reference 46, Shoji-Kawata et al Nature 2013.
Page 17 ABT-737 and Tat-Beclin both inhibit MERS-CoV replication and they "leave the levels of BECN1 unchanged". Therefore, the authors cannot conclude just after that "increasing BECN1 is the relevant mechanism". We can make the same remark on page 18, on the mechanism of SKP2 inhibitor by enhancing BECN1 levels. Effects of overexpression of BECN1 on CoV replication should be tested. A siRNA against BECN1 will confirm the involvment of BECN1 and autophagy in CoV replication.

Page 18
The authors need to report that several of their tested drugs have well-described activity on autophagy, such as niclosamide, valinomycin, salinomycin and gossypol. They only confirmed previous studies. Moreover, these compounds are not only SKP2 inhibitors and specific autophagy inhibitors but have other activities since for example Niclosamide increases mitochondrial fission and valinomycin is a potassium ionophore. Once again, viral replication should be explored using plaque assay and growth kinetics. Figure 7 panels B and C The unusable way to determine the antiviral activity does not valorize the impressive results. Even if all these drugs inhibit SKP2, they have other activities. Therefore, the title needs to be modified unless impact of SKP2 is tested by gene silencing.
Discussion page 21 line 5 Same remark than above, unless they are testing SKP2 knock down, the authors cannot conclude on impact of SKP2 inhibition on viral multiplication. The authors should report in the discussion that autophagy has no antiviral effect on MHV, another betacoronavirus.
In conclusion regarding the antiviral activities the authors clearly identified components able to block MERS-CoV replication but the mechanism by SKP2 inhibition is not so far demonstrated.
Minor remarks: Figure 1 "(C-F) FKP51…. Transfected with vector control… or FKP51" add (Co) after control Figure 3 panel I Information regarding $$ is missing in the legend. Unfortunate that only LLP degradation was tested with siRNA against SKP2 and no other readout for autophagy such as LC3 of p62.
Page 10 "Targeting SKP2 INCREASED protein degradation" Page 11 "we aimed to test if blocking the SKP2 dependent autophagy inhibition influences CoV replication". To improve the understanding of this sentence, it is better to say "if stimulation of autophagy by SKP2 inhibitor influences MERS-CoV replication. Add MERS instead of just CoV because, it was tested only for this coronavirus.
Page 13-15 Several mistakes in the labelling of panels for figure S3 Page 13 S3DE is S3EF; S3F is in fact S3G, S3G is S3H page 15 S3H is S3I I is J and J is K Moreover in Figure S3K what is the condition "Co"? sam question for Figure S5 panel I. where are the white histograms? pages 2 4 15 17 20 22 The word "propagation" is used several times in the text, but corresponds in fact to viral cell-to-cell spread and is not tested here. It should be replaced by viral multiplication or production.
Finally the title does not totally reflect data presented in the article, because impact of SKP2 on viral replication has been explored only using compounds which have other activities than just inhibiting SKP2.
Reviewer #3: Remarks to the Author: Autophagy is a cellular homeostatic pathway that can mediate degradation of obsolete organelles, misfolded proteins and other cargo like viral components. Being a destructive process, autophagy has to be tightly controlled. Indeed, intricate signalling pathways regulate both induction and flux rates of autophagy. Recent reports have highlighted the impact of autophagy on many cellular processes and its role as an anti-viral mechanism. Gassen et al had previously identified FKBP51 as a beclin-binding protein, which mediates induction of autophagy (Gassen et al, 2014&2015). Continuing their work on autophagy, they elaborate on the mechanism. FKBP51 is assembling an autophagy-promoting complex, which deactivates/dephosphorylates the E3 ligase SKP2. Inactivated SKP2 no longer mediates K48-linked polyubiquitination of beclin-1 thereby stabilizing its protein levels and promoting autophagy. Furthermore, the authors characterise the interplay between autophagy and MERS-CoV. While infection inhibits autophagic flux via several viral proteins, exogenous activation of autophagy by pharmacological inhibition of SKP2 combats MERS-CoV infection. Experiments characterising autophagy are conducted in a convincing way and the data supports the claims of the authors. The paper by Gassen et al features an elegant exploration of the complicated mechanism of FKBP51 and links SKP2 to beclin-1. They manage to assemble puzzle pieces from previously published studies about the regulation of SKP2 (e.g. Rodier et al, 2008 and others) complemented by their own data to suggest assembly of a multi-protein complex that regulates stability of beclin-1. Finally, this study provides novel data on how to combat viral infections (here MERS-CoV) using insights gained from understanding the molecular mechanism of autophagy induction and virus-autophagy interplay. However, occasionally it feels like there are two (interesting) separate stories existing side-by-side (Mechanism of FKBP51/SKP2, Interplay between autophagy and MERS-CoV), which need to be tied together more tightly. A few specific issues need to be addressed as well: Major: 1. Treatment with drugs can have multiple effects on viral replication. Therefore, even though all used drugs presumably target a similar pathway, it has to be made sure that autophagy is responsible for viral attenuation. The authors should conduct experiments in which autophagy is induced (e.g. by mentioned drugs), but the effects of this induction are blocked. For example, can the growth attenuation be rescued if autophagic turnover is blocked by chloroquine, or LC3B conjugation inhibited by ATG5 KO/KD? Furthermore, the chosen drugs used to inhibit SKP2 (even though it is backed up by literature) need direct evidence that SKP2-dependent beclin-1 ubiquitination is reduced. 2. The interesting data on MERS-CoV prompts the question, how do other viruses, which are known to be sensitive towards autophagy, react towards inhibition of SKP2. Experiments exploring the effect of the drugs on e.g. Sindbis Virus or HIV-1 replication should clarify that issue and would complement the MERS-CoV data and make it less focused on treating a single virus with a system potentially applicable for other viral pathogens. 3. Fig. 2C: To show that SKP2 is the E3 ligase which ubiquitinates beclin-1, the authors should include a SKP2 E3-ligase null mutant and monitor beclin-1 levels/degradation. The effect observed in Fig. 2C-F should be rescued by MG132 treatment, to prove its proteasome dependent degradation. 4. Fig. 3: A few additional details have to be clarified: Is Beclin-1 K402R resistant towards degradation induced by either PHLPPi treatment or SKP2 overexpression? Since AKT1 is involved in the cascade, does pharmacological inhibition of AKT1 lead to stabilization of beclin-1? Minor: 1. The abstract needs rephrasing to emphasise the achievements of this paper in a more cohesive and appealing way. The current abstract undersells the story. 2. The title of the paper could be shortened to sound less convoluted. 3. Writing of the results section (first part) might need improvement to make it more accessible. Furthermore, mechanistic details could be presented more cohesively to focus on the achievements of the paper. 4. Please add size markers to all the western blots. 5. Fig. 1 B: Stabilization of beclin-1 levels by MG132 has to be compared to NH4Cl treatment, to rule out turnover by autophagy. 6. Fig. 2A,B: It would be interesting to see the whole Western Blots (especially for USP18 and USP36, including the input). 7. Fig. 2C,D: Since SKP2 is regulated by phosphorylation. It would be curious to see how phosphomimetic and phospho-null mutants of SKP2 would impact beclin-1 stability. 8. Fig. 3 I: Please show one exemplary western blot of a protein degraded by autophagy. 9. Fig. 4 E: Please add a quantification for the western blots as shown in the figure. 10. Fig. 5D and E: Both figure panels could be moved to the supplements, as the effect observed is only marginal and other data presented is stronger. 11. Fig. 6A: While interesting, the sketch could be moved to the supplements. 12. Fig. 6G: Could be moved to the supplements 13. Fig. 6E: While manual counting is certainly being used a lot, these assays could be alternatively quantified using flow cytometry to get data, which might be less biased by manual observation. 14. Fig. 7 A

Reviewer #1
We thank reviewer #1 for the insightful, critical and constructive comments. We performed new experiments to address these comments, and moderated our conclusions as suggested where necessary, while balancing the modifications keeping comments of reviewer #3 in mind who suggested to better highlight the merits of the study.
Major Issues Comment 1. "Throughout the report, the panels in the figures do not correlate with the references in the text. While the figure # is correct, the letter they relate to in the text is incorrect especially in all of the MERS-CoV figures."

Response:
We apologize for these errors; we double-checked to make sure all the numbers and letters in the text correlate to the right figure and panel.

Response:
We agree with the reviewer. In order to directly show that virus production (on request of reviewer 2 we are avoiding the word "propagation", because this refers to spreading from cell to cell which we do not show directly) is affected we determined the reduction of plaque forming units/ml in parallel to virus RNA copies per ml. Based on this comment and comments from the other reviewers, we performed new experiments and now display the fold difference (log10 scale) of the RNA analysis (genome equivalents, GE) and the plaque forming units, as well as the raw data of the genome equivalents in the new Figures 9B and S9A-C. Figure S9A shows the unprocessed data of the detected genome equivalents per ml, which are in good agreement with the GE data of the first submission. Clearly, the RNA analysis and the determination of the pfus revealed the same pattern of activity, indicating that both assays are valid to investigate the influence of autophagy inducers on MERS-CoV replication/growth. We chose to examine the role of autophagy on MERS-CoV growth in multi-cycle replication assays in order to identify meaningful effects that reflect physiological conditions. Only a few MERS-CoV particles are necessary to establish an infection in the respiratory tract. All assays were, therefore, performed with a low MOI. Importantly, we used two different time points (e.g. 24 and 48 hours post infection) to account for effects during the exponential growth of the virus (24 hours) and the plateau of virus production (48 hours). Comment 3. The control protein for Figure 2 is TRAF6 to show that there is no effect on BECN1 protein levels via an additional ubiquitination protein. However TRAF6 is not included in the pulse chase experiments in Figure 2E and F. It should be shown as a comparison to vector alone.

Response:
We performed the requested experiment and replaced panels E and F of Figure 2. On the request of reviewer #3, these panels now also display conditions including the proteasome inhibitor MG132. Similarly to the cycloheximide assay (panels C and D), TRAF6 did not affect protein stability of Beclin1.

New Figure panels 2E and F (changed to respect the concerns of reviewers 1 and 3):
Legend of Figure 2E,F: "The conditions of C,D were also used in the pulse-chase assay, performed as in Fig. 1E,F, to determine BECN1 stability."  Figure 4D where 3-MA has no effect on protein levels. This is not discussed in the text." Response: 3-MA was used at a concentration of 10 mM, which is the concentration typically used for the inhibition of autophagy (for example Cell 2010; 140: 313-326). We are aware of the reports that 3-MA can have effects beyond blocking autophagy, and even promote autophagy under certain conditions. This is also stated in the third edition of the "Guidelines for the use and interpretation of assays for monitoring autophagy" (AUTOPHAGY 2016; 12: 1-222), page 54: "For example, 3-MA is commonly used to inhibit starvation-or rapamycin-induced autophagy, but it has no effect on BECN1-independent forms of autophagy, and some data indicate that this compound can also have stimulatory effects on autophagy". In our case, the goal was to investigate BECN1-dependent autophagy, for which 3-MA should work according to these guidelines. Moreover, we do see inhibition of induced autophagy, because the enhanced proteolysis of long-lived proteins (induced by si-RNA targeting Skp2 that enhances BECN1 protein levels, Fig We changed the manuscript by providing the concentration of 3-MA in the figure legends as well as in the methods section, by correcting the description of the turnover assay in the results section, and by providing more citations. We noted that there was a wrong wording in our description of the turn-over assay, which may have misled this reviewer, and we apologize for this. The section now reads ( Figure  3I is now Figure 4M): "In order to verify that reduced SKP2 activity drives autophagy in our cellular set-up, we used siRNA and determined the degradation of long-lived proteins 48 . Targeting SKP2 increased proteolysis, but not in the presence of the autophagy inhibitor 3methyladenine (3-MA, Fig. 4M). Furthermore, we observed decreased levels of p62 (Fig. S1G), which typically goes along with enhanced autophagy48. All these data are consistent with an inhibitory effect of SKP2 on autophagy regulated by a phosphorylation cascade involving FKBP51, PHLPP and AKT1.
Comment 5. "For Figure 4, there are several controls missing in the figure. First, there are no toxicity controls showing that the range of each inhibitor used is below significant cell toxicity levels. Secondly, the control in 4D using HBSS as an inducer of autophagy is not shown. Please show experimental evidence that in this experiment HBSS induces autophagy. Third, in 4E, BECN1 levels increase with all drugs; however LC3B increases with only 2 of them, and there doesn't seem to be a change in P62. There should be some correlation between these results but there is no consistency." Response: To 5.1: Figure S2A displays the toxicity controls and shows that each inhibitor is used at a concentration below significant cell toxicity.
To 5.2: We performed flux assays with HBSS and BafA1 which confirmed the induction of autophagic flux by HBSS. These data are now presented as Figure S2D:  Figure 5E, for the WCE blot it is not clear how you can observe both SNAP29 and STX17 in the same blot where it does not show up in the IP blots. Shouldn't they be in the same complex since they are crosslinked so they show as 1 band? The order of 5F and G should be reversed. And 5F is not clear. I assume it is a quantitation of 5G, however it does not match the data in 5G. The relationship between ATG14 monomeric and oligomeric forms and autophagy is not made clear. Are there published data showing oliogomeric ATG14 actively inhibits autophagosome to lysosome fusion, or simply that it fails to promote fusion? Also, if MERS is inhibiting oligomerization, I would expect there to be an increase in monomer, which is not seen. Does total ATG14 amount change? Please clarify." Response: To 6.1: We apologize for the errors in Figure numbering, which we corrected.
To 6.2 Figure 5E: WCE in Figure 5E (now Fig. 6H (page 14): "In addition, it has been shown that ATG14 oligomerizes to dimers and tetramers, which is essential for autophagy by promoting STX17 binding and autophagosomelysosome fusion 7 . Using capillary-based electrophoresis allowing for better resolution at high molecular weight we observed ATG14 oligomerization to 7-8mers, which was enhanced by SMIP004 (Fig. S2H, I)." Comment 7. "The MERS-CoV infections experiments in the supplemental figure S3 is also confusing. First, the cropped cells in S3E do not line up, please clean up the cropping. Secondly, there does not seem to be an increase yellow staining and a correlated decrease in red staining in the single cell shown in each panel. This example cell does not correlate with F. Third, the mutant viruses with either 4b or 5 deleted are great, however does the 4b deletion virus still contain the entire 4a coding region or was is 4a ORF effected when 4b was deleted? The oligos are in the methods section but no explanation in the text to the sites of the deletion. There are also no growth curves showing the 2 deletion viruses compared to WT virus. Do they grow equally well in the experiments or are they attenuated in their growth under control conditions? Fourth, the virus output that is examined in K is only looking at mRNA with no statistically significant different noted however it is stated in the text that the SKP2i inhibitor effects MERS growth. The K panel seems to be missing data. And the correlation should be with virus titer not mRNA since the autophagy effect would be expected to be on virus egress not transcription?" Response: To 7.1 and 7.2: We improved the example images of Figure S3E (now figure 6D) 8A). In the supplement, we also included the unprocessed data showing the GE/m (S5A). In Figure 9B,C we establish that GE/ml and PFU/ml can both be used to obtain an estimate of the virus growth.
Comment 8. "In the model figure and in the discussion, there is no description of data presented to show the direct effect of SKP2 on ATG14 oligomerization. Since this is a key feature of the model, it would be important to show that SKP2 siRNA or inhibition by drugs effects the ATG14 oligo/monomer relationship." Response: Figure 8D,E (former figure 6F) shows the effect of SKP2 inhibition on ATG14 oligomerisation in MERS-CoV infected cells. The inhibitor produced more ATG14 oligomers, but had no effect on the amount of ATG14 monomers. As mentioned in our response to major point 6, we find the absence of effect on ATG14 monomers in the presence of pronounced effects on ATG14 oligomers difficult to explain, but this has been observed by others in different settings as well. The idea is not that SKP2 causes ATG14 oligomerisation directly, but that this has recently been shown as crucial step in functional autophagy (Nature 2015 520: 563-566) and thus was included in our analyses.
We mention this now first in the results to new Figure S2H, as also explained in our response to comment 6.2: "In addition, it has been shown that ATG14 oligomerizes to dimers and tetramers, which is essential for autophagy by promoting STX17 binding and autophagosome-lysosome fusion 7 . Using capillary-based electrophoresis allowing for better resolution at high molecular weight we observed ATG14 oligomerization to 7-8mers, which was enhanced by SMIP004 (Fig. S2H,I)".
Further, the description of panel D in the legend of Figure 9 (former Figure 7) now is: "Summary scheme: SKP2 leads to K48-linked poly-ubiquitination and thus degradation of BECN1. The effect of SKP2 can be diminished in two ways, either by chemical inhibition or by FKBP51, which scaffolds protein interactions ultimately leading to SKP2 inactivation through inhibiting its phosphorylation. Both scenarios enhance autophagy, which involves ATG14 oligomerization (probably 7-8mers) as newly described essential step in functional autophagy 7 . MERS-CoV reduces autophagy through distinct viral proteins leading to blockade of autophagosomelysosome fusion and ATG14 oligomerization. Compounds inhibiting SKP2 reinstate autophagy and efficiently reduce viral production."

Minor Issues
Minor comment 1. " Figure 1B. The whole cell lysates shown in the 2nd and 3rd lane for BECN1 (3rd blot down) are different however they are the same in the IP blot for BECN1 (2nd blot down). Please correct or explain why there are differences."

Response:
We replaced the blot with a more suitable example: "Gassen et al reported here a very complete study of a novel regulation of autophagy by SKP2, a E3 ubiquitin ligase. They nicely demonstrated that SKP2 targets BECN1, an autophagy protein which is involved in autophagosome formation and maturation in two different complexes. SKP2 executes K48-ubiquination at K402 of BECN1, allowing its proteosomal degradation. Activity of SKP2 is negatively controlled by FKBP51, which regulates SKP2 phosphorylation. Pharmacological and genetic inhibition of SKP2 leads to autophagy activation, with increased BECN1 levels, and increased interaction between lysosomal and autophagy SNARE proteins. They identified and studied a pharmacological compound, SMIP004, which strongly inhibits SKP2. The results presented her are novel, technically sound and contribute to a better understanding of autophagy regulation.
A second part of the article is dedicated to a possible therapeutic application of their data on Middle East respiratory syndrome (MERS) coronavirus. MERS-CoV is a human betacoronavirus responsible of a severe viral respiratory disease, first identified in Saudi Arabia in 2012. A third of reported patients with MERS have died and no antiviral treatment is currently available. They first demonstrated that MERS CoV blocks autophagy, in particular the fusion between autophagosome and lysosome, using several reliable autophagy monitoring experiments. They identified three nonstructural viral proteins which can recapitulate the block of the autophagic flux observed during viral infection. They constructed recombinant viruses deleted for these genes but unfortunately poorly characterized their phenotype.
Finally they tested the effect of this SKP2 inhibitor, SMIP004 at 10µM, on MERS-CoV infection and observed an important inhibition of its multiplication. Unfortunately, they only evaluated viral genome production by RT-PCR at two different times post infection. Virus titers were not determined by plaque assay or TCID50 and no growth kinetics are presented, as expected for this kind of studies. Because a similar inhibition was observed with other compounds which interact with BECN1 and stimulate autophagy, they conclude that SMIP004 acts by stabilizing BECN1. They also tested FDA approved compounds reported to inhibit SKP2 and identified for several of them an antiviral activity against MERS-CoV. They conclude from this part of the work that inhibition of SKP2 leads to an important restriction of MERS CoV multiplication and represents an interesting therapeutic approach. This second part is quite interesting but suffers from speculative interpretations and lack of data from the literature. Some important experiments regarding the mechanism of antiviral activity of SMIP004 are missing and should be added to support their interpretation of the data. In conclusion, this article reports totally new and important data on both autophagy regulation by SKP2 and on interaction between MERS-CoV and autophagy and a possible antiviral role of autophagy, which can be used as an innovative therapeutic approach."

Response:
We thank the reviewer for the constructive comments. We believe to have addressed the major concerns of this reviewer by i) characterizing the phenotype of the MERS-  The authors need therefore to modify this sentence."

Response:
We modified and expanded the description as follows (üage 14, former Fig. S3F is now Fig. 6C): "Infection with MERS-CoV further led to an increase of the total number of phagocytic vesicles (sum of AL + AP) per cell (Fig. 6C,D). However, the number of successfully formed AL was reduced significantly, indicating that AP can form but not fuse with lysosomes when cells are infected. A fusion block was also evidenced by the significantly reduced ATG14 oligomerization, essential for AP-lysosome fusion 7 , in infected cells (Fig. 6E,F) and by the increase of the autophagy target P62 (Fig.  S3C)."

Comment Page 14
"The results presented in Figure 5 clearly demonstrated that MERS-CoV blocks autophagosome maturation rather that initial steps of autophagosome formation. The titer of Figure 5 is therefore not adapted to the results and should be replaced by "blocks autophagic flux"."

Response:
We agree and modified the title of ""In order to confirm the influence of p4b and p5 on virus replication, we performed loss of…for the step-wise construction of an ORF4b-and ORF5-deleted MERS-CoV" "The deletion of p4b or p5 resulted in 10-fold lower replication efficiency 48 hours p.i. (Fig. 6B)" The characterization of the recombinant viruses presented in this study is clearly insufficient. No genome sequence analysis was reported. Growth kinetics of the different viruses showing viral titers (pfu/ml) at different times need to be added instead of measuring an inhibitory effect of the deletion (fig 6B)."

Response:
We agree that "inhibitory effect" might have been a less precise wording for Figure  6B. Usually viral titers are determined by plaque assay (pfu/ml) in growth kinetics at different times of infection. In this study, they use RT-PCR in supernatant to detect viral genome but this experiment does not measure viral infectivity."

Response:
As shown above, we reproduced the experiment independently and now include the PFU/ml and the GE/ml data in the manuscript (Fig. S4E,F).

The discrepancy to the previously published data by Almazan and colleagues may be explained by the different ways of introducing the deletions into the infectious MERS-CoV cDNA clone. In addition, we used a different cell line and a very low MOI which allows to determine minor growth differences. We included the following sentences into the manuscript (end of page 24):
…" Deletion of p4b or p5 resulted in reduced MERS-CoV growth. These results differ from previous observations, which may be related to different cell lines and cDNA clone construction strategies 72,73 . In addition, we used a very low MOI and a type I IFN-deficient cell line (VeroB4), which allows to determine minor growth differences."… Comment Figure 6 panel C The authors concluded that mutant viruses do not control autophagy compared to parental virus. However, no quantification of LC3-II/LC3-I is available and the expression level of LC3-I seems important compared to other WB of the article.

Response:
We now provide quantification of LC3BII/I along with the western blot ( Figure  S4C

The reviewer also noted an intriguing point, to which we only can offer a speculation at the moment, namely the apparent effect on the overall levels of LC3B (LC3B-I). We quantified LC3B-I separately, and confirmed a difference, which came out nonsignificant though. We mention this now in the discussion as follows (starting at the end of page 24): "….The absence of p4b and p5 during MERS-CoV replication may also result in lower LC3B levels compared to MERS-CoV wildtype-infected but also to the mock-infected control cells. This points to a more complex interaction between virus replication and autophagy; p4b and p5 may counteract a response of the host to viral infection, but the functional relevance of their link to autophagy requires further investigation….".
Comment Panel B and C: How is the multiplication of these mutant viruses in autophagy deficient cells? If growth kinetics of mutant viruses are similar to that of parental virus, this would allow the authors to confirm the antiviral role of autophagy on MERS-CoV and to reveal a possible requirement of orf4b or orf5 to counteract it. Response:

To address this point, we tried gene silencing to inhibit autophagy. However, this turned out not to be successful in the cells we use for the viral replication assays, as transfection and cell growth in combination with viral infection was insufficient (the flux assays with the tandem-labelled LC3B plasmid do not depend on a high transfection efficiency). However, we eventually were able to produce ATG5-/-Vero4B cells. Viral replication in these cells and in WT cells was monitored by PCR assessing genome equivalents and by determining plaque forming units. We found that deletion of ATG5 increased virus growth up to 50-fold on the one hand, but on
the other hand does not support the conclusion that p4b or p5 are responsible for this effect. Therefore, we moderated our conclusion to this point. The results are presented in the new figure S4E,F which we copied above in response to another issue of the reviewer.

The section in the results is (begins at the end of page 17):
"Deletion of p4b and p5 led to a decreased accumulation of P62 and LC3B-II/I, suggesting that both proteins contribute to the inhibition of the autophagic flux (Fig.  S4C,D). The p4b-and p5-deleted viruses as well as the WT control virus grew to higher levels in ATG5 knockout Vero cells compared to WT cells (Fig. S4E,F). However, the p4b and p5-deleted viruses showed overall an up to 10-fold decreased replication in both WT and ATG5 knockout cells compared to WT virus suggesting a p4b-and p5-dependent attenuation of virus replication that is independent of ATG5directed autophagy." Comment Figure 6 The titer of the legend does not reflect the results. No data in this figure demonstrates that SKP2 inhibition reduces viral replication.

Response:
We agree and took care to make sure the titles better reflect the figures. With the new data, figures were rearranged, so figure 6 now largely is integrated into figure 8, but not entirely.

Comment Page 15
"Regarding SKP2 inhibitor, effect on viral replication is impressive, but same remark than above, growth kinetics and determination of viral titers are necessary."

Response:
In the revised manuscript we included a modified version showing the fold change difference between DMSO vs Skp2i-treated cells (now figure 8A). In the supplement (Fig. S5A), we also include the unprocessed data showing the GE/ml. As outlined above ( Figure S4E,F) and below ( Figures 9B, S9A-C), we showed for several cases that GE/ml can be used as a surrogate for PFU/ml.
New figure 8A and S5A: Comment Page 15, continued: "Moreover, to prove that the compound is acting via autophagy, effect of SMIP004 should be tested in autophagy deficient cells. This kind of approach was used for antibacterial activity of tat-beclin1 for example, in reference 46, Shoji-Kawata et al Nature 2013. "

Response:
We agree that testing the effect of SMIP004 in autophagy-deficient cells would add useful information. We actually considered performing the mentioned experiment; however, due to reconstruction work of the BSL3 laboratory during the revision, we were not able to perform this experiment in a timely manner. We reasoned that other experiments requested by the reviewers (e.g. providing pfus) needed to be given higher priority in using the very limited BSL3 laboratory availability. all these compounds might have additional effects and mention this in the discussion, where we adapted our conclusions (page 25): "From the present results, therapeutic induction of autophagy by inhibition of SKP2 emerges as a novel approach. We should note that all the substances tested here have or may have additional targets. While we cannot exclude the possibility that for each of the compounds it is this additional target that is responsible for the antiviral effect, we consider this scenario as less likely; this assessment is also based on the observation of the antiviral effects of compounds targeting BECN1, which we establish here as client of SKP2." Furthermore, we changed the title of the manuscript to: "SKP2 attenuates autophagy through Beclin1-ubiquitination and its inhibitors combat MERS-Coronavirus infection".

Comment Page 17
"ABT-737 and Tat-Beclin both inhibit MERS-CoV replication and they "leave the levels of BECN1 unchanged". Therefore, the authors cannot conclude just after that "increasing BECN1 is the relevant mechanism". We can make the same remark on page 18, on the mechanism of SKP2 inhibitor by enhancing BECN1 levels. Effects of overexpression of BECN1 on CoV replication should be tested. A siRNA against BECN1 will confirm the involvment of BECN1 and autophagy in CoV replication."

Response:
We agree that the conclusion was not stringent the way it was presented. All we can state is that another and independent mechanism that enhances BECN1's activity in autophagy elicits the same antiviral effect on MERS. This suggests that of the activities potentially elicited by enhancing BECN1 protein levels it is autophagy that is relevant, but it does not prove it. As mentioned above, cotransfection experiments (including overexpression of Beclin1) in combination with viral infection were not successful, unfortunately, in particular using any kind of siRNA (which also would risk to generate off-target effects). We refined the description in the results section (page 20): "Our data strongly suggest that of the potential mechanisms of SKP2i, it is the increase of BECN1-directed autophagy that reduces MERS-CoV multiplication. If this is correct, other ways of enhancing BECN1's effect on autophagy should elicit similar effects. Therefore, we used two recently introduced tools to enhance BECN1 function in autophagy: the BH3 mimetic ABT-737 and a BECN1-derived peptide 58,59 . These compounds, in contrast to SKP2i, do not change the levels of BECN1, but increase autophagy by redirecting BECN1 function towards the autophagy pathway, which we confirmed in VeroB4 cells (Fig. S7B-D,G). Enhanced autophagy was indicated by the increase of LC3B-II/I (Fig. S7E,G) and the reduction of P62 (Fig. S7F-G). This was paralleled by inhibition of MERS-CoV replication by up to 60-fold in the case of the BECN1-peptide (TAT-B) (Fig. S7H) and 10-fold in the case of ABT-737 at 48 h p.i. (Fig. S7I). These results are in accordance with the notion that increasing BECN1 acts through enhancing autophagy as the relevant mechanism for the antiviral effects of SKP2 in vitro."

Comment Page 18:
The authors need to report that several of their tested drugs have well-described activity on autophagy, such as niclosamide, valinomycin, salinomycin and gossypol. They only confirmed previous studies. Moreover, these compounds are not only SKP2 inhibitors and specific autophagy inhibitors but have other activities since for example Niclosamide increases mitochondrial fission and valinomycin is a potassium ionophore. Once again, viral replication should be explored using plaque assay and growth kinetics.

Response:
We agree that autophagy induction of these compounds has been shown before. This actually is the reason why the compounds were chosen, as mentioned in the text. The intention was to begin to explore, whether there might be readily available drugs that tap into the pathway presented in the manuscript. We modified the results section, because it could have been misunderstood (page 22): "These results encouraged us to test other potential inhibitors of SKP2. Autophagyinducing, FDA-approved drugs and drugs from clinical trials were of particular interest as these could become more readily available for treatment in humans. Of note, valinomycin (VAL), which has been shown to target SARS-CoV in vitro 50,61 , is also known to act as an SKP2 inhibitor 49 ." Figures 9B, S9A-C). Raw data (absolute quantification of viral RNA as "genome equivalents, GE" and infectious particles as "plaque forming units, PFU", all presented now in the new Figures S9A,C) of this experiment showed that the difference between GE and PFU is 1000-fold irrespective of the drug used. In other words, whether we used real-time RT-PCR or the pfu assay did not matter, because we obtained the same pattern among all drugs. Therefore, the determination of EC50 values of VAL, NIC and SKP2i was performed only by real-time RT-PCR.  Figure 7 panels B and C "The unusable way to determine the antiviral activity does not valorize the impressive results. Even if all these drugs inhibit SKP2, they have other activities. Therefore, the title needs to be modified unless impact of SKP2 is tested by gene silencing."

Response:
Following the reviewer´s suggestion, we changed the title of Figure 9 (~previous  figure 7) to "Test of potential SKP2 inhibitors for their effects on autophagy and viral replication", because silencing of SKP2 by siRNA did not work and appears to be toxic for the cells, unfortunately. We copied the new figures 9B and S9A-C (the experiments of the previous figure 7B have been repeated and expanded to include pfus) in our response to the comment before.
In addition, we changed the title of the manuscript to: "SKP2 attenuates autophagy through Beclin1-ubiquitination and its inhibitors combat MERS-Coronavirus infection".
Comment Discussion page 21 line 5 "Same remark than above, unless they are testing SKP2 knock down, the authors cannot conclude on impact of SKP2 inhibition on viral multiplication.
The authors should report in the discussion that autophagy has no antiviral effect on MHV, another betacoronavirus."

Response:
We agree that the conclusion only goes as far as the specificity of the compounds. As detailed above, we modified the title and text accordingly. MHV is now mentioned in the introduction (copied in response to an earlier point above).

Concluding major comment
In conclusion regarding the antiviral activities the authors clearly identified components able to block MERS-CoV replication but the mechanism by SKP2 inhibition is not so far demonstrated.

Response:
As detailed in the response to the specific points above, we modified the text to moderate this specific claim.

Response:
We modified the figure panels as requested. Figure 3 panel I "Information regarding $$ is missing in the legend. Unfortunate that only LLP degradation was tested with siRNA against SKP2 and no other readout for autophagy such as LC3 of p62."

Response:
We added the missing information to the figure legend (former panel I of figure 3 is  now panel M of figure 4). Additional effects observed with the siRNA were the decrease of Beclin1 ubiquitination (Fig. 2I), the increase in Beclin1 levels and stability (Figs. 2G, J-M, S1C), the increase in LC3BII/I (Figs. 2G, S1C) the increase of ATG14 oligomerisation (Fig. S2H,I), the restoration of autophagic flux in MERS-CoV infected cells (Fig. S7J), and the decrease of p62 (new Fig. S1G).

Response:
We apologize for this error, which must have been confusing. The sentence is now corrected.

Comment Page 11
""we aimed to test if blocking the SKP2 dependent autophagy inhibition influences CoV replication". To improve the understanding of this sentence, it is better to say "if stimulation of autophagy by SKP2 inhibitor influences MERS-CoV replication. Add MERS instead of just CoV because, it was tested only for this coronavirus."

Response:
We agree with the reviewer and thank for this kind suggestion. This section, marking the transition to the viral investigations, is now completely revised, and this specific sentence disappeared. The transition now reads: "Results up to this point suggested that activity of BECN1 and thereby autophagy is limited by SKP2, which in turn is activated by AKT1 ( 46 and Fig. 3/4). Since MERS-CoV has been shown to enhance phosphorylation of AKT1 25 , it might reduce autophagy." Comment Page 13-15 "Several mistakes in the labelling of panels for figure S3 Page 13 S3DE is S3EF; S3F is in fact S3G, S3G is S3H page 15 S3H is S3I I is J and J is K Figure S3K what is the condition "Co"? same question for Figure  Comment pages 2 4 15 17 20 22 "The word "propagation" is used several times in the text, but corresponds in fact to viral cell-to-cell spread and is not tested here. It should be replaced by viral multiplication or production." Response:

Moreover in
The revised manuscript includes the detection of infectious units using virus plaque assays. As we transferred cell-free virus-containing supernatants to non-infected cells to determine the amount of infectious units, we think that now it would be appropriate to use word virus propagation. Nevertheless, we modified our wording.
Final comment "Finally the title does not totally reflect data presented in the article, because impact of SKP2 on viral replication has been explored only using compounds which have other activities than just inhibiting SKP2."

Response:
We changed the title to "SKP2 attenuates autophagy through Beclin1-ubiquitination and its inhibitors combat MERS-Coronavirus infection".

Reviewer #3 (Remarks to the Author):
Autophagy is a cellular homeostatic pathway that can mediate degradation of obsolete organelles, misfolded proteins and other cargo like viral components. Being a destructive process, autophagy has to be tightly controlled. Indeed, intricate signalling pathways regulate both induction and flux rates of autophagy. Recent reports have highlighted the impact of autophagy on many cellular processes and its role as an anti-viral mechanism.
Gassen et al had previously identified FKBP51 as a beclin-binding protein, which mediates induction of autophagy (Gassen et al, 2014&2015). Continuing their work on autophagy, they elaborate on the mechanism. FKBP51 is assembling an autophagy-promoting complex, which deactivates/dephosphorylates the E3 ligase SKP2. Inactivated SKP2 no longer mediates K48-linked polyubiquitination of beclin-1 thereby stabilizing its protein levels and promoting autophagy. Furthermore, the authors characterise the interplay between autophagy and MERS-CoV. While infection inhibits autophagic flux via several viral proteins, exogenous activation of autophagy by pharmacological inhibition of SKP2 combats MERS-CoV infection. Experiments characterising autophagy are conducted in a convincing way and the data supports the claims of the authors. The paper by Gassen et al features an elegant exploration of the complicated mechanism of FKBP51 and links SKP2 to beclin-1. They manage to assemble puzzle pieces from previously published studies about the regulation of SKP2 (e.g. Rodier et al, 2008 and others) complemented by their own data to suggest assembly of a multi-protein complex that regulates stability of beclin-1. Finally, this study provides novel data on how to combat viral infections (here MERS-CoV) using insights gained from understanding the molecular mechanism of autophagy induction and virus-autophagy interplay. However, occasionally it feels like there are two (interesting) separate stories existing side-byside (Mechanism of FKBP51/SKP2, Interplay between autophagy and MERS-CoV), which need to be tied together more tightly.
A few specific issues need to be addressed as well: Major: "Treatment with drugs can have multiple effects on viral replication. Therefore, even though all used drugs presumably target a similar pathway, it has to be made sure that autophagy is responsible for viral attenuation. The authors should conduct experiments in which autophagy is induced (e.g. by mentioned drugs), but the effects of this induction are blocked. For example, can the growth attenuation be rescued if autophagic turnover is blocked by chloroquine, or LC3B conjugation inhibited by ATG5 KO/KD? Furthermore, the chosen drugs used to inhibit SKP2 (even though it is backed up by literature) need direct evidence that SKP2-dependent beclin-1 ubiquitination is reduced." We tried to use chloroquine, but this turned out to be toxic for the cells. Furthermore, ATG5 KD (or any other transfection experiment, in particular siRNA transfection) was not successful in combination with viral infection. Thus, we constructed ATG5 KO in VeroB4 cells and found that MERS-CoV replicates better in these cells. As mentioned in our response to reviewer 2, we agree that testing the effect of SMIP004 in autophagy-deficient cells would be useful. We actually considered performing the mentioned experiment; however, due to reconstruction work of the BSL3 laboratory during the revision, we were not able to perform this experiment in a timely manner. We reasoned that other experiments requested by the reviewers (e.g. providing pfus) needed to be given higher priority in using the very limited BSL3 laboratory availability. Furthermore, we provide strong evidence that a) SMIP004

induces autophagy, b) MERS-CoV benefits from inhibition of autophagy (new experiments in newly constructed Vero ATG5 KO cells), c) SMIP004 inhibits MERS-CoV replication along with reversing the MERS-CoV induced inhibition of autophagic flux, and d) several autophagy-inducing agents also inhibit MERS-CoV
replication. We agree though, that all these compounds might have additional effects and mention this in the discussion, where we adapted our conclusions (page 25): "From the present results, therapeutic induction of autophagy by inhibition of SKP2 emerges as a novel approach. We should note that all the substances tested here have or may have additional targets. While we cannot exclude the possibility that for each of the compounds it is this additional target that is responsible for the antiviral effect, we consider this scenario as less likely; this assessment is also based on the observation of the antiviral effects of compounds targeting BECN1, which we establish here as client of SKP2." In line with these considerations, we changed the title of the manuscript to "SKP2 attenuates autophagy through Beclin1-ubiquitination and its inhibitors combat MERS-Coronavirus infection". This title leaves open which exact molecular pathway might be responsible for the effect of these inhibitors.

Comment 2.
"The interesting data on MERS-CoV prompts the question, how do other viruses, which are known to be sensitive towards autophagy, react towards inhibition of SKP2. Experiments exploring the effect of the drugs on e.g. Sindbis Virus or HIV-1 replication should clarify that issue and would complement the MERS-CoV data and make it less focused on treating a single virus with a system potentially applicable for other viral pathogens." Sindbis virus also turned out to be somewhat sensitive towards SKP2 inhibition, even though much less pronounced than MERS-CoV. This may be explained by a differential interaction of both viruses with the autophagic flux. The Sindbis data are now included in the manuscript as Figure

Response:
We observed that SKP2 null MEFs grow extremely slow. Therefore, we used siRNA against SKP2 as before (the efficiency is documented in Figs 2G/S1C). We performed all the requested experiments and found that siRNA targeting SKP2 increased the stability of Beclin1 and ectopic overexpression of SKP2 had no effect in the presence of MG132, proving a proteasomal-dependent effect. In addition, to more thoroughly address this question we also assessed ubiquitination of Beclin1 in combination with siRNA against SKP2. These data are now included in Figure  Legend to Figure 2C,D: SKP2 affects BECN1 stability. HEK293 cells were transfected with plasmids expressing SKP2 or TRAF6 and exposed to MG132 (10 µM, 2 h) where indicated and to cycloheximide (CHX, 30 µg/mL) after 72 h for the durations indicated to monitor the decay of BECN1.
Legend to Figure 2E,F: The conditions of C,D were also used in the pulse-chase assay, performed as in Fig. 1E,F, to determine BECN1 stability.
Legend to Figure 2J-M: The stability assays of the panels C-F were performed to determine the effect of SKP2-targeting siRNA on BECN1 stability. All graphs (showing the means ± SEM) and all western blots are representative of three independent experiments. Legend to Figure 2I: Knock-down of SKP2 by siRNA decreases BECN1 ubiquitination (assay as in Fig. 1B). Representative blots are shown. " Fig. 3: A few additional details have to be clarified: Is Beclin-1 K402R resistant towards degradation induced by either PHLPPi treatment or SKP2 overexpression? Since AKT1 is involved in the cascade, does pharmacological inhibition of AKT1 lead to stabilization of beclin-1?"

Response:
We compared the stability of Beclin1 WT and the K402R mutant in the cycloheximide assay, including the PHLPP inhibitor. We found that the K402R exhibited greater stability in comparison to WT Beclin1, and further resisted degradation induced by PHLPPi. We also tested the effect of two different Akt1 inhibitors on Beclin1 stability. Both inhibitors increased the stability of Beclin1. These data are now presented as figure 4G-J: Legend to Figure 4G-J: The cycloheximide protein stability assay was performed (as in Fig. 1C,D) to evaluate the effect of PHLPPi on K402-BECN1 in comparison to wt BECN1 (G,H) and to test the effects of AKT1 inhibitors (AktiX and MK2206) on wt BECN1 (I,J).
The abstract needs rephrasing to emphasise the achievements of this paper in a more cohesive and appealing way. The current abstract undersells the story.

Response:
We thank the reviewer for the kind comment. We tried to emphasize the achievements while avoiding overstatements. Combined with the other reviewers' comments, and in line with this comment, we tried to make sure that the core of our findings comes across, which is a novel pathway for the induction of autophagy. The potential usefulness of compounds addressing this pathway in fighting MERS-CoV infection serves more as an example of possible applications. Whether or not MERS-CoV has evolved mechanisms to tone down autophagy probably is less relevant in this context, but it is a question that naturally comes up and this is why we explored this also. The abstract now focusses more on the novel pathway leading to autophagy induction: "Beclin1 (BECN1) is a key regulator of autophagy. We here identified S-phase kinase-associated protein 2 (SKP2) as E3 ligase that executes lysine-48-linked polyubiquitination of BECN1, thus promoting its proteasomal degradation. FK506 binding protein 51 associates with and stabilizes BECN1 by scaffolding the assembly of a hetero-complex involving PHLPP, AKT1, SKP2 and BECN1 that limits phosphorylation and thus activity of SKP2. Genetic or pharmacological inhibition of SKP2 decreases BECN1 ubiquitination, decreases BECN1 degradation and enhances autophagic flux. Middle East respiratory syndrome coronavirus (MERS-CoV) is an emerging and lethal human respiratory virus for which no specific treatment is available. We found that MERS-CoV multiplication results in reduced BECN1 levels and blocks the fusion of autophagosomes and lysosomes. Inhibitors of SKP2 not only enhanced autophagy but also reduced the replication of MERS-CoV up to 28,000-fold. The SKP2-BECN1 link constitutes a novel target for host-directed antiviral drugs and possibly other autophagy-sensitive conditions." Minor comment 2: The title of the paper could be shortened to sound less convoluted.

Response:
We changed the title to: "SKP2 attenuates autophagy through Beclin1-ubiquitination and its inhibitors combat MERS-Coronavirus infection". We tried to simplify the title by constructing it along the two parts (novel mechanism of autophagy regulation and potential application to treatment). As we describe a novel mechanism regarding the SKP2-dependent control of autophagy and additionally show that the induction of autophagy by SKP2 inhibitors has a pronounced effect on MERS-CoV infection we would like to keep both subjects in the title.
Writing of the results section (first part) might need improvement to make it more accessible. Furthermore, mechanistic details could be presented more cohesively to focus on the achievements of the paper.

Response:
Thank you for the kind comment. We rewrote the results section along the lines laid out in the response to minor comment 2.
Minor comment 4: "Please add size markers to all the western blots."

Response:
We added size markers to all western blots of figure 1. In our opinion, this makes the figure look very busy, so we suggest to refrain from this in the other figures.
Minor comment 5: " Fig. 1 B: Stabilization of beclin-1 levels by MG132 has to be compared to NH4Cl treatment, to rule out turnover by autophagy."

Response:
We included NH 4  Response: We include data on p62 as example of a protein that is degraded by autophagy. These data are now included in Figure S1G.

Response:
The quantification is shown in figure S2E-G (referring to Fig. 5E, former figure 4E).
Minor comment 10: " Fig. 5D and E: Both figure panels could be moved to the supplements, as the effect observed is only marginal and other data presented is stronger." Response: Figure 5D is now figure S3D. We agree the effect shown in former figure 5E is quite small. It is hard to predict though, to what degree SNARE protein associations have to change to produce a significant effect on vesicle fusion; furthermore, but less important, former panel E of figure 5 could nicely be accommodated into the new figure 6. Thus, we kept 5E in the main figures (now figure 6H).
Minor comment 11: " Fig. 6A: While interesting, the sketch could be moved to the supplements."

Response:
We completely agree and moved the panel to the supplement ( Figure S4A).
" Fig. 6G: Could be moved to the supplements"

Response:
Some reasoning as above for Fig. 5E. Former figure 6G is now figure 8F.
Minor comment 13: " Fig. 6E: While manual counting is certainly being used a lot, these assays could be alternatively quantified using flow cytometry to get data, which might be less biased by manual observation."

Response:
We agree that FACS would be a useful method for this type of assay. We would like to point out that counting was performed by a scientist blind to the conditions. This is now mentioned in the methods description. The effect size is considerable, so it appeared to us that the quantification method chosen is useful. In addition, there is no FACS in the BSL3 laboratory we are using. We were not allowed to export samples from the BSL3 to perform FACS analysis in an external core facility.
Minor comment 14: " Fig. 7 A,B: Is there a correlation between autophagy induction and MERS-CoV inhibition?"

Response:
In general, the compounds that are more effective in inducing autophagy are also more effective in MERS-CoV inhibition (now Figure 9A

Reviewer #1
General comment: "The extensive revisions in this manuscript add to the already large datasets that was included in the original version. The amount of data in this new version clearly separates between 2 stories. The authors identify how SKP2 effects BECN1 levels via ubiquitination and how inhibition of SKP2 by either siRNA or inhibitory compounds effect BECN1 and autophagy. The report then presents an entirely separate story, on the role of autophagy and MERS-CoV including live virus and how individual MERS-CoV proteins are shown to effect autophagy flux. It is unclear why these 2 stories are combined into this large of a paper and it diminishes the goal of making a clear and organized manuscript. The first story on SKP2 and Beclin1 are clear and direct while the second story on MERS-CoV and how it effects and is affected by autophagic machinery is still unfinished with data that is discordant with the text descriptions of it."

Response:
We agree that the manuscript might be lengthy and that separating it into two manuscripts might be an option, in principle. In conclusion, we think that the combination of data regarding the MERS-CoVdependent modulation of autophagy and the potential treatability of this virus by addressing the novel SKP2-autophagy pathway is highly beneficial for the manuscript.
Specific comment 1: "The addition of the raw fold changes and PFU/ml are helpful in determining the effects of autophagy on MERS-CoV replication however the RNA and PFU levels do not correlate with each other. In the added figures of Figure 9B and the supplementary figures associated with it, the fold changes and raw PFU and genome numbers do not correlate with increases and decreases going in opposite directions. There is also no quantitation of significance in these figures making it difficult to interpret as described in the text." The comment concerns figures 9B and S9A, which display the fold difference (9B) and the raw data (S9A) of the genome equivalents, as well as figures S9B and S9C, which display the fold difference (S9B) and the raw data (S9C) of the pfu measurements. We realized that there was a mistake in the column coloring for DMSO in Figure 9B which may have caused confusion. We apologize for this and corrected the colors. The DMSO black column (left) and gray column (right) were switched. In addition, we would like to emphasize that one has to compare all gray columns (24 hours post infection) and all black columns (48 hours post infection) separately. In Figure 9B the DMSO control was set to 1 for both time points and cannot directly be related to Figure S9A which shows the raw data in log scale. As growth differences are difficult to identify on a log scale we decided to show fold-change in relation to the DMSO control in Figure 9B. Since the presentation of the fold differences in Figures 9B and S9B are mathematical derivatives of the raw data displayed in Figures S9A and S9C The reviewer is completely right by saying that a direct comparison between GE/ml ( Figure S9A) and PFU/ml (Figures S9C) shows some minor discrepancies for some of the compounds. This may be explained by interference of the respective compounds with the production of infectious viral particles which may subsequently influence the GE/ml to PFU/ml ratio. Mechanistic details of these minor discrepancies were not addressed and would be beyond the scope of this paper.
Overall, however, as explained in our response to the first round of comments, the GE/ml and PFU/ml pattern is highly similar. We are now providing quantification of significance. In addition, we submit all raw data to these figures. The text claims significant differences however that is not evident in the calculations of the figures."

Response:
We are now providing quantification of significance. In addition, we submit all raw data to these figures in a data source file that covers all figures.
Specific comment 3: "The newly added TRAF6 control data requested in Figure 2 does not correlate with the graph presented. An initial concern for this figure is that ectopic TRAF6 was not included in the pulse chase experiments in Figure 2E and F. The authors have now added in the data showing TRAF6 ectopically expressed in this experiment. The text states, as well as the graph of the western blots, that TRAF6 ectopically expressed in the cells does not effect Beclin 1 levels. However the western blots associated with Figure 2E clearly show a stabilizing effect on Beclin1 protein levels, which does not correlate with the graph shown in Figure 2F. There does not seem to be a way to connect the western blot data with that graphed in Figure 2F. The data suggests that the ectopic expression of SKP2 shows beclin1 degradation at the same level as expression of TRAF6. This is not what is shown in the western blots. It is unclear how the graph could be made from the data provided in the figure." Response: It appears to us that the reviewer's impression derives from a combination of our poor selection of the blot example and maybe screen display. We are now submitting less exposed blots to this figure, along with the quantification data of the individual blots in the data source file.
Specific comment 4: "The addition of Figure S2D in response to the question of whether the use of HBSS and BafA1 are inducing autophagic flux as expected is warranted. However again the quantification of the western blot and the images of the western blot are not the same. The graph suggests that the LC3II/I ratio is the same for HBSS and BafA1 treatment however the western blot does not show that same correlation. It is clear in the western blot provided that the 2 sets of data are not equal. While the western blot does show reduction of LC3II in HBSS treatment suggesting autophagic flux, the data do not correlate between the graph and the blot shown."

Response:
We agree that the western blot did not fully reflect the average given in the graph, partly due to, as we suspect, the higher abundance of LC3I in comparison to LC3II which makes it difficult to select blots that visualize differences in the two LC3 species equally well for the eye. As before, we are now submitting less exposed blots (new experiment) along with the raw data of the quantification. II to actin (or any other protein) would not change the numbers. While this is mentioned in the supplementary methods, we realized that we mentioned the Actin antibody, but failed to state that it was used for normalization. We are grateful for the hint and added a respective sentence in the supplement (page 23, last sentence).
Minor point 4: "Regardless of the availability of BSL3 space (which I agree can be a limiting factor), I still think that experiments in ATG5 KO cells need to be done, even though all hints and all evidence points towards autophagy and the MERS CoV phenotype being connected. However, this might be beyond the scope of this paper."

Response:
We agree on the usefulness of this experiment. We further adjusted our wording and mention this experiment for future work, also considering the comments of reviewer 1(discussion, page 29).