NEDD4 and NEDD4L regulate Wnt signalling and intestinal stem cell priming by degrading LGR5 receptor

Abstract The intestinal stem cell (ISC) marker LGR5 is a receptor for R‐spondin (RSPO) that functions to potentiate Wnt signalling in the proliferating crypt. It has been recently shown that Wnt plays a priming role for ISC self‐renewal by inducing RSPO receptor LGR5 expression. Despite its pivotal role in homeostasis, regeneration and cancer, little is known about the post‐translational regulation of LGR5. Here, we show that the HECT‐domain E3 ligases NEDD4 and NEDD4L are expressed in the crypt stem cell regions and regulate ISC priming by degrading LGR receptors. Loss of Nedd4 and Nedd4l enhances ISC proliferation, increases sensitivity to RSPO stimulation and accelerates tumour development in Apcmin mice with increased numbers of high‐grade adenomas. Mechanistically, we find that both NEDD4 and NEDD4L negatively regulate Wnt/β‐catenin signalling by targeting LGR5 receptor and DVL2 for proteasomal and lysosomal degradation. Our findings unveil the previously unreported post‐translational control of LGR receptors via NEDD4/NEDD4L to regulate ISC priming. Inactivation of NEDD4 and NEDD4L increases Wnt activation and ISC numbers, which subsequently enhances tumour predisposition and progression.

In this manuscript, the authors studied the role of Nedd4-1 (Nedd4) and Nedd4-2 (Nedd4L) in WNT signaling in the intestine and in colon cancer, and the role of these E3 ligases in targeting LGR5 (and Dvl2) in the intestinal stem cells. They show that these Nedd4s suppress WNT signaling and suggest that this suppression is mediated by targeting LRG5 for degradation. Evaluation: 1. This MS has some interesting findings, but overall it is not particularly novel. The finding that Nedd4 suppresses WNT signaling in the colon and that intestinal Nedd4 knockout (KO) enhances APC(min)-driven colonic tumor growth was already published several years ago (Lu et al, Oncogene 2016-cited in the current MS), as was the fact that Dvl is a substrate for Nedd4L (Ding et al. JBC 2013-also cited here). The new finding in the current MS is the fact that the Nedd4 close relative Nedd4L also contributes to the regulation of colonic tumor growth (ie has redundant function with Nedd4 in this context, a finding that may be interesting, but not earth-shaking), and that LGR5 may be a target for Nedd4/Nedd4L regulation. The latter observation is potentially important, but the authors have not provided any in vivo (in animal) proof for it, only in transfected cells, as detailed below. 2. Proof that LRG5 is a substrate of Nedd4/Nedd4L in vivo is missing: To correct this, the authors need to show: (i) the accumulation of LRG5 protein in the stem cells of the small intestine and colon of the Nedd4/Nedd4L KO or DKO mice. (ii) that Nedd4/Nedd4L bind LRG5 (including Co-IP of endogenous proteins), if indeed LRG5 is a substrate for Nedd4/Nedd4L; (iii) test Nedd4/Nedd4L-mediated ubiquitination of LGR5 using a proper ubiquitination assay: the ones shown in the paper were not done correctly, since samples have to be boiled in SDS prior to the LRG5 IP and immunoblotting for ubiquitin (to ensure that the ubiquitination observed is of LRG5 itself and not its associated proteins); (iv) all pulse-chase assays to analyze LRG5 stability upon loss of Nedd4 or Nedd4L need to be repeated several times and quantified. The IF studies (Fig 6 & Suppl. 5D) need to be quantified as well. 3. For all intestinal sections immunostained for different markers (eg Fig 1): The sections appear to come from different parts of the small intestine (and from different mice?), so that may affect the results (ie pattern of expression of the same marker may differ between sub-sections of the intestine, such that comparison between expression of the different markers may not be meaningful, as it may represent different parts of the intestine rather than differences between WT vs KO mice). I.e. the real way to perform such comparison of expression pattern of different markers is to use serial sections, and to indicate exactly where they came from, which should be indicated in the legend (it is not enough to just write: intestinal crypt). Also, were siblings used for the comparisons in the study? 4. What is the genetic background for all mice and crosses used in the study? This is very important both for the current study, and for comparison with earlier published studies. 5. Fig 2C (organoids): why is the Edu staining (cell proliferation) lower in the Nedd4-KO or DKO mice relative to WT? This contradicts the statements in the paper. Also, Suppl. Fig 2B (organoids) should be quantified.
Referee #2: In this manuscript, Novellasdemunt et al investigated the role of the E3 ligase Nedd4/Nedd4l in normal intestinal crypt homeostasis and colon cancer driven by APCmin. They find that Nedd4/Nedd4l KO in the intestinal epithelium induces the proliferation of intestinal crypts cells and the number of ISCs. They claim that Wnt signaling is elevated in DKO organoids, which undergo reduced apoptosis upon limited addition of Rspondin. They also show that Nedd4/Nedd4l exacerbates intestinal tumor grade and phenotype by APCmin background. Using 293T cell lines to delineate the mechanism, they show that Nedd4/Nedd4l induces ubiquitination of LGR5, leading to proteasomal and lysosomal degradation as well as degradation of Dvl2, thereby reducing Wnt signaling. The manuscript is well written and the findings are novel (E3 ligase for LGR5) and are well supported. However, there are some critical points need to be addressed to strengthen their study.
Major points 1. A main concern is whether the in vivo effects from DKO can solely be explained by LGR5 regulation. Since the authors also observed Dvl2 degradation by Nedd4/Nedd4l, increased LGR5 or increased Dvl2 or both can cause hyperproliferation of the crypt cells in DKO. To prove that Nedd4/Nedd4l KO increases the proliferation and tumor phenotype via LGR5 regulation, some rescue experiments are needed. For example, if they can rescue the survival rate of APCmin Nedd4/Nedd4l DKO by crossing with LGR5+/-, this would support their hypothesis and also rule out the effect from Dvl2 regulation. However, such a time consuming experiment may be beyond the scope of this study. In its absence, the claim that in vivo effects from DKO are solely via LGR5 regulation should be toned down.
2. In Figure 1, the author only show Olfm4 as ISC marker. Multiple ISC markers should be checked such as LRIG1, Ascl2, LGR5.
3. The authors mainly rely on overexpression to prove their mechanism (LGR4/5, Dvl2). Can they try to detect endogenous LGR4 or LGR5 and Dvl2 to show the effect of Nedd4/Nedd4l overexpression or knockout?
4. The evidence that Nedd4/Nedd4l regulates Wnt signaling only via LGR5, is weak. To further prove their point Nedd4/Nedd4l lof effect in Hek293 cells on Topflash should be compared between Wnt3a only and Wnt3+Rspo. Furthermore, Nedd4/Nedd4l lof Topflash activity should be compared {plus minus} siLGR4,5. Figure 6, the author show that Nedd4/Nedd4l can ubiquitiylate LGR4/5 by pull down of LGR4/5 and detecting HA-UB in cell lysates. To provide more direct evidence for this model, an in vitro ubiquitination assay should to be performed with recombinant proteins. At the minimum, pulled-down sample could be boiled with 1% SDS to remove LGR4/5 bound ubiquitinated proteins; dilute 10 times with lysis buffer and pull down it again with the same antibody to detect ubiquitination of LGR4/5. This way, the authors can at least conclude that ubiquitination is not due to other proteins binding to LGR4/5.

Minor points
1. From RNA scope results, Nedd4 expression is more restricted in the crypts than Nedd4l. However, in many figures, the effect of Nedd4 KO is milder than Nedd4l KO (Figure 2A, C, D. Figure 3C, D. Figure 4A and Supplement 3A and B). Have the author seen the increased Nedd4l level in Nedd4 KO as a compensation? Or is there an explanation for this? 2. Comparing Figure 6A and B for LGR5 level, Nedd4 completely removed the LGR5 band from the western blot while LGR5 level seems not be changed upon Nedd4 overexpression from the immunofluorescence picture. If the authors took the picture where there are still significant LGR5 signaling in the cells, it should be mentioned in the text or legends. Figure  4. Nedd4/Nedd4l overexpression in 293T cell or using Nedd4 or Nedd4l KO 293T cell lines, the authors need to prove mRNA levels of LGR4, 5 are not changed. Figure 5 E-G, the authors showed that Nedd4/Nedd4l overexpression decreased Topflash in HCT116 but not in APC4 and DLD1. However, there is a study (PMID : 23349017; Al-Kharusi at la, 2013, Carcinogenesis) showing that there is no detectable LGR5 protein in HCT116. Thus, it is worth to show that LGR5 is expressed in HCT116, APC4, DLD-1 (qPCR or western blot).

Authors' correspondence 18th Jul 2019
Thank you very much for the timely review. We appreciate very much your effort and the efficiency of the review process (the fastest record I have ever had in the past!), and your suggestion for the transfer.
While the reviewer's comments are relevant and will certainly help improve and strengthen our manuscript, we feel that some of the comments/suggestions from reviewer #1 are not justified or reasonable. First of all, reviewer #2 provided very reasonable advice and comments and we will be able to fully address. For reviewer #1, we will also be able to address most of the concerns and suggestions. There are two major concerns that I believe are the reasons that you recommended against publication in EMBO J. First is the novelty, and second is the in vivo proof of Lgr5 as the target.
Regarding the novelty, what was published before was Nedd4l inhibits Wnt by targeting Dvl2, and Nedd4 increases adenoma growth in full-body KO with poor characterisation. Our current findings uncover the novel role of Nedd4 by targeting Lgr5 receptor, a key stem cell priming factor for degradation. This part was recognised by both reviewer as of importance to the field. I also need to emphasis that the previously reported Nedd4 KO mouse data (Lu at al Oncogene 2016) was poorly characterised and could not be reproduced. It is important to publish a comprehensive analysis of the in vivo data with proper stem cell and tumour analysis to show the importance of both Nedd4 and Nedd4l in intestinal stem cell and tumour development.
Regarding the in vivo proof part, we feel that this is not a reasonable comments. It is well recognised in the field that there is no reliable endogenous antibodies recognising Lgr5 receptor for validation in vivo. This is evident by the pioneer group Hans Clevers who published the Lgr5 as ISC marker, he never published any paper using antibody recognising endogenous Lgr5, simply because the existing ones are not reliable. We have been in close contact with the Clevers group, hoping to obtain reliable endogenous Lgr5 antibodies without success. In that case, it is not justified to request in vivo validation of protein stabilisation when there is no existing reliable antibody available. . This is a widely accepted approach in the field, thus asking for in vivo proof of Lgr5 stabilisation in the absence of endogenous antibody is unreasonable. Importantly, we have provided functional proof by testing the Rspondin response in ex vivo organoids and in vivo Rspo injection (Figure 7), which is the ultimate proof that Lgr receptors are stabilised in vivo. Unfortunately the data was dismissed or under-appreciated by reviewer #1. Thus, we feel that these concerns raised by reviewer #1 was not justified.
On the other hand, we could address these questions using ex vivo organoids. We could generate Lgr5-KI-tag to proof the protein stabilisation. Also, questions 3-5 by reviewer #1 are all based on misunderstanding, which we can address easily. Based on the reasons above, I'd like to kindly ask for reconsideration of our manuscript in EMBO journal. We will be happy and ready to address the rest of the reviewers' concern for a timely resubmission to EMBO J. Thank you for considering it and please let me know if you have any questions. Thank you for contacting me regarding our decision and for your patience with my response, which got delayed due to internal discussions in the team regarding your letter together with re-assessing the manuscript and referee comments.
After looking back into literature mentioned, we appreciate your point regarding the difficulty to employ an antibody on LGR5 endogenous. In addition we realise that you would -judging from the information provided in the your letter -be potentially able to address the issues raised by the referees in a revised version of the manuscript. Overall, we would thus invite you to work towards a re-review. Accordingly, we would -given that you addressed all the experimental issues with compelling data -be prepared to ask the referees for further input.
Please note however, that in particular the proof for an in vivo relevant functional link between Need4-L and Lgr5 as well as the distinction between Lgr5 versus Dvl2-dependent activities of Nedd4 has been major concerns of the referees, thus will in our view be core aspects to be considered.
Given the importance and broad interest of the question addressed, it would be essential for you to provide a definitive and accurately described dataset in the revised version.
Please contact me if you have any questions, need further input on the referee comments or if you consider engaging in a compelling revision, in which case we would not close the file.
However, please note, that since the results of your experiments are entirely open at this stage, we cannot in any way predict the outcome of a re-submission, or make any promises towards publication.

Revision -authors' response 14th Oct 2019
Referee #1: In this manuscript, the authors studied the role of Nedd4-1 (Nedd4) and Nedd4-2 (Nedd4L) in WNT signaling in the intestine and in colon cancer, and the role of these E3 ligases in targeting LGR5 (and Dvl2) in the intestinal stem cells. They show that these Nedd4s suppress WNT signaling and suggest that this suppression is mediated by targeting LRG5 for degradation. Evaluation: 1. This MS has some interesting findings, but overall it is not particularly novel. The finding that Nedd4 suppresses WNT signaling in the colon and that intestinal Nedd4 knockout (KO) enhances APC(min)-driven colonic tumor growth was already published several years ago (Lu et al, Oncogene 2016-cited in the current MS), as was the fact that Dvl is a substrate for Nedd4L (Ding et al. JBC 2013-also cited here). The new finding in the current MS is the fact that the Nedd4 close relative Nedd4L also contributes to the regulation of colonic tumor growth (ie has redundant function with Nedd4 in this context, a finding that may be interesting, but not earth-shaking), and that LGR5 may be a target for Nedd4/Nedd4L regulation. The latter observation is potentially important, but the authors have not provided any in vivo (in animal) proof for it, only in transfected cells, as detailed below.

Response:
We are grateful to the reviewer's comment, and we appreciate the reviewer's concerns and constructive suggestions. These are all relevant to the conclusion of the study, and have now been addressed accordingly in the specific comments below.
2. Proof that LRG5 is a substrate of Nedd4/Nedd4L in vivo is missing: To correct this, the authors need to show: (i) the accumulation of LRG5 protein in the stem cells of the small intestine and colon of the Nedd4/Nedd4L KO or DKO mice.

Response:
We appreciate the reviewer's concern and suggestion about the in vivo proof of Lgr4  Fig. 6A).
The results showed that NEDD4 and NEDD4L indeed bound to LGR4 and DVL2 endogenously.
(iii) test Nedd4/Nedd4L-mediated ubiquitination of LGR5 using a proper ubiquitination assay: the ones shown in the paper were not done correctly, since samples have to be boiled in SDS prior to the LRG5 IP and immunoblotting for ubiquitin (to ensure that the ubiquitination observed is of LRG5 itself and not its associated proteins);

Response:
We thank the reviewer's constructive suggestion to improve our manuscript. To further ensure that the ubiquitination observed is of LGR5 itself rather than its associated proteins, we have now repeated the experiment in boiled SDS condition as suggested by the reviewer. Consistent with our original data, the new results showed that LGR5 ubiquitination was strongly enhanced upon expression of WT NEDD4 or NEDD4L constructs but not the catalytic inactive mutants (new Fig. 6C-6D).
(iv) all pulse-chase assays to analyze LRG5 stability upon loss of Nedd4 or Nedd4L need to be repeated several times and quantified. The IF studies (Fig 6 & Suppl. 5D) need to be quantified as well.

Response:
We have now quantitated the pulse-chase assays and IF data in Fig. 6 and Supplementary Fig. 5. Quantitation of CHX pulse-chase data: Quantitation of SNAP data (new Supplementary Fig. 5E): 3. For all intestinal sections immunostained for different markers (eg Fig 1): The sections appear to come from different parts of the small intestine (and from different mice?), so that may affect the results (ie pattern of expression of the same marker may differ between sub-sections of the intestine, such that comparison between expression of the different markers may not be meaningful, as it may represent different parts of the intestine rather than differences between WT vs KO mice). Ie the real way to perform such comparison of expression pattern of different markers is to use serial sections, and to indicate exactly where they came from, which should be indicated in the legend (it is not enough to just write: intestinal crypt). Also, were siblings used for the comparisons in the study?

Response:
The reviewer was concerned if the immunostaining sections were coming from different parts/regions of the small intestine among different animals. We apologise for the confusion. Indeed, all tissues analysed in the immunostaining in Fig.1 and Suppl Fig.1 (both WT and KO) were from the same proximal regions of the small intestine.
There might be slight crypt-villus length variability due to tissue processing, which might give the wrong impression that they were from different regions. Also, the length of the villi was slightly longer in DKO intestine in general, albeit not significant. We have now replaced some images in Fig.1 with more representative images, and have indicated clearly in the method and legends about the region of the intestine.
Importantly, we have indeed provided quantitation of the staining in Fig. 1 to complement the representative images, which was included in the original Fig. 1O All our animals involved in the study were on C57/BL6J background, which is the same as the previously published ones. We apologise for the missing information in the method, which has now been updated.
5. Fig 2C (organoids): why is the Edu staining (cell proliferation) lower in the Nedd4-KO or DKO mice relative to WT? This contradicts the statements in the paper. Also, Suppl. Fig 2B (organoids) should be quantified.

Response:
The reviewer raised concern about the Edu staining in Fig. 2C. The images chosen in the figures were representative pictures for the individual genotype. We have now provided more images below to give a better idea of the Edu staining patterns for each genotype.
Overall, it is obvious that Edu staining in WT organoids were mostly restricted to the budding crypts, which is consistent with the data reported by many other groups. On the other hand, the number of Edu+ cells were clearly increased in all three mutants as compared to WT, and the staining has extended to the lumen rather than cryptexclusive. Similar to in vivo data, the proliferation increase was more striking in Nedd4l cKO and DKO as compared to Nedd4 cKO. However, to be more quantitative, we have performed organoid formation assay in Supplementary Fig. 2B with quantitation on Fig. 2D, which clearly indicated that Nedd4l cKO and DKO organoids showed increase organoid formation efficiency. This is consistent with the Edu staining data, which together indicate that there are increased proliferative stem cells in the mutant organoids.
Referee #2: In this manuscript, Novellasdemunt et al investigated the role of the E3 ligase Nedd4/Nedd4l in normal intestinal crypt homeostasis and colon cancer driven by APCmin. They find that Nedd4/Nedd4l KO in the intestinal epithelium induces the proliferation of intestinal crypts cells and the number of ISCs. They claim that Wnt signaling is elevated in DKO organoids, which undergo reduced apoptosis upon limited addition of Rspondin. They also show that Nedd4/Nedd4l exacerbates intestinal tumor grade and phenotype by APCmin background. Using 293T cell lines to delineate the mechanism, they show that Nedd4/Nedd4l induces ubiquitination of LGR5, leading to proteasomal and lysosomal degradation as well as degradation of Dvl2, thereby reducing Wnt signaling. The manuscript is well written and the findings are novel (E3 ligase for LGR5) and are well supported. However, there are some critical points need to be addressed to strengthen their study.

Response:
We thank the reviewer's positive comment, and we appreciate the reviewer's concerns and constructive suggestions. These are all relevant to the conclusion of the study, and have now been addressed accordingly in the specific comments below.
Major points 1. A main concern is whether the in vivo effects from DKO can solely be explained by LGR5 regulation. Since the authors also observed Dvl2 degradation by Nedd4/Nedd4l, increased LGR5 or increased Dvl2 or both can cause hyperproliferation of the crypt cells in DKO. To prove that Nedd4/Nedd4l KO increases the proliferation and tumor phenotype via LGR5 regulation, some rescue experiments are needed. For example, if they can rescue the survival rate of APCmin Nedd4/Nedd4l DKO by crossing with LGR5+/-, this would support their hypothesis and also rule out the effect from Dvl2 regulation. However, such a time consuming experiment may be beyond the scope of this study. In its absence, the claim that in vivo effects from DKO are solely via LGR5 regulation should be toned down.

Response:
We appreciate the reviewer's concern about the in vivo phenotype and the complex role of Nedd4  Fig. 6F and Supplementary Fig. 6C).

Although our in vitro data clearly indicate that Nedd4/Nedd4l exhibit Dvl2-independent role for Lgr5 stabilisation and Wnt activation, it is still not sufficient to clarify whether the DKO phenotypes were a consequence of Dvl2 or Lgr5 stabilisation or both. We apologise for not making this point clear in the original manuscript.
We have now improved the discussion to emphasise that the phenotype can be a consequence of both Dvl2 and Lgr5 stabilisation. Figure 1, the author only show Olfm4 as ISC marker. Multiple ISC markers should be checked such as LRIG1, Ascl2, LGR5.

Response:
We thank the reviewer's suggestion. We have now provided the qRT-PCR analysis of WT and DKO intestinal crypts using the ISC markers suggested by the reviewer (new Supplementary Fig. 1C) Supplementary Fig. 6B).  Supplementary Fig. 4B).  Fig. 5D).

The increase in Wnt activity in
In addition, we have now further compared the Nedd4/Nedd4l LOF effect on TOPFlash activity in plus/minue siLGR4/5 condition as suggested by the reviewer (new Supplementary Fig. 6C and Fig. 6F). 5. For the ubiquitination assay in Figure 6, the author show that Nedd4/Nedd4l can ubiquitiylate LGR4/5 by pull down of LGR4/5 and detecting HA-UB in cell lysates. To provide more direct evidence for this model, an in vitro ubiquitination assay should to be performed with recombinant proteins. At the minimum, pulled-down sample could be boiled with 1% SDS to remove LGR4/5 bound ubiquitinated proteins; dilute 10 times with lysis buffer and pull down it again with the same antibody to detect ubiquitination of LGR4/5. This way, the authors can at least conclude that ubiquitination is not due to other proteins binding to LGR4/5.

Response:
Once again, we thank the reviewer's valuable suggestion on the ubiquitination assay.
To further ensure that the ubiquitination observed is of LGR5 itself rather than its associated proteins, we have now repeated the experiment by boiling the pulled-down samples in 1% SDS as suggested by the reviewer. Consistent with our original data, the new results showed that LGR5 ubiquitination was strongly enhanced upon expression of WT NEDD4 or NEDD4L constructs but not the catalytic inactive mutants (new Fig. 6C-6D).
Minor points 1. From RNA scope results, Nedd4 expression is more restricted in the crypts than Nedd4l. However, in many figures, the effect of Nedd4 KO is milder than Nedd4l KO (Figure 2A, C, D. Figure 3C, D. Figure 4A and Supplement 3A and B). Have the author seen the increased Nedd4l level in Nedd4 KO as a compensation? Or is there an explanation for this?

Response:
Indeed, we have noticed the slightly milder phenotype in Nedd4 KO than in Nedd4l KO as indicated by the reviewer. However, we did not see any compensation increase of Nedd4l level in the Nedd4 KO tissues or vice versa (Suppl Fig. 2A 2. Comparing Figure 6A and B for LGR5 level, Nedd4 completely removed the LGR5 band from the western blot while LGR5 level seems not be changed upon Nedd4 overexpression from the immunofluorescence picture. If the authors took the picture where there are still significant LGR5 signaling in the cells, it should be mentioned in the text or legends.

Response:
The reviewer questioned about the discrepancy between the Western blot and SNAP data that NEDD4 appeared to complete remove LGR5 band in the Western Blot while LGR5 level was not affected in the SNAP IF images. First, we want to emphasise that the membrane level of LGR5 (mature form) was completely removed by NEDD4 and NEDD4L in the SNAP IF data, while the total LGR5 level (mostly cytoplasmic as immature form in the biosynthetic pathway e.g. ER) was unaffected (Fig. 6B). This suggest that the NEDD4-mediated LGR5 degradation seems to take place at the cell surface and makes the mature (SNAP-labelled) form disappear. It is indeed interesting to note that LGR5 appears to be completely degraded by NEDD4/4L in the Western Blot (Fig. 6A) . 6A).
The higher band (mature form) is a lot more prominent than the lower band (immature form) in our LGR5-Flag construct. And the degradation of LGR5 happens largely to the higher mature form than the lower immature form, which is consistent to the SNAP IF data (predominant degradation in the mature form at the membrane level of mature LGR5). We thank the reviewer for pointing this out. We have now modified Fig. 6A and have clarified this point in the revised results session. Figure 6B is not matched with Figure 6B. Please check the Figure  legend in the corresponding supplement figure.

Response:
We apologise for the mistake. This has now been corrected in the revised manuscript.
4. Nedd4/Nedd4l overexpression in 293T cell or using Nedd4 or Nedd4l KO 293T cell lines, the authors need to prove mRNA levels of LGR4, 5 are not changed.

Response:
We have now examined the mRNA level of LGR4 and LGR5 in the NEDD4 and NEDD4L KO 293T cells as suggested by the reviewer. Indeed, we did not observe any transcriptional changes of the LGR receptors in NEDD4/4L KO, indicating that the protein changes were caused by post-translational modification. Figure 5 E Thank you for submitting your revised manuscript for consideration by The EMBO Journal. Your amended study was sent back to the referees for re-evaluation, and we have received comments from both of them, which I enclose below.

In
As you will see the referee finds that their concerns have been sufficiently addressed and they are now broadly in favour of publication.
Thus, we are pleased to inform you that your manuscript has been accepted in principle for publication in The EMBO Journal, pending some minor issues related to formatting and data representation as listed below, which need to be adjusted at re-submission. The authors have now addressed most of my previous concerns regarding the original presented results and the new required experiments (or at least explained why they cannot be done, especially the proof that endogenous LGR5 is stabilized in vivo in mice that lack Nedd4/Nedd4L). This is commendable, as they performed and show numerous new experiments. The novelty of the work described in this paper in general is still somewhat in question, however.
Referee #2: The revised manuscript from Novellasdemunt et al is substantially improved. This manuscript aims to study the role of Nedd4/Nedd4l E3 ligases in intestinal stem cell homeostasis and cancer prone situation. They focused on regulation of Dvl2, a previously identified substrate of Nedd4/Nedd4l, as well as a newly discovered substrate, LGR5. They show that loss of Nedd4/Nedd4l increases proliferation of the intestinal epithelium and Wnt signaling. They show convincingly that LRG4/5 levels are regulated by Nedd4/Nedd4l. The authors also showed that Nedd4/Nedd4l induces ubiquitination and proteasomal degradation of LGR5. The main concerns raised in the review and how these have been addressed are summarized below.
1) In vivo effect of Nedd4/Nedd4l DKO solely through LGR5 regulation: They toned down the statement that the effect of DKO could be explained either by LGR5 or Dvl2 or both.
2) Multiple ISC markers should be checked: They addressed this by qPCR showing that Lrig1, Ascl1, and Lgr5 expressions were increased in DKO.
3) Effect of Nedd4/Nedd4l on endogenous LGR4/5 or Dvl2: They tried many antibodies to address this and show the compelling evidence that there is no working antibodies available for LGR5. For LGR4 and Dvl2, they showed that Nedd4/Nedd4l KO increase endogenous LGR4 and Dvl2 protein levels in cells. Moreover, they generated endogenously tagged LGR5-HA Knock-In organoids from WT and DKO and showed that LGR5 and Dvl2 protein levels are increased in DKO organoids. They also provide endogenous interaction between Nedd4/Nedd4l and LGR4 or Dvl2.
4) The evidence that Nedd4/Nedd4l regulates Wnt signaling only via LGR5 is weak: The authors tried to address this by Topflash. They showed that Nedd4/Nedd4l regulate Wnt signaling via LGR5 and Dvl2 and this has been clarified in the discussion. 5) Ubiquitination assay for LGR5: The author addressed this with a modified protocol.
6) Discrepancy between Western blot and SNAP upon Nedd4/Nedd4l overexpression: The author argue that there are two band for LGR5 (Upper band for mature and lower band for immature forms of LGR5) and Nedd4/Nedd4l overexpression predominantly remove mature form of LGR5 from Western blot and SNAP. However, in Figure 6A, there is same lower band in EV transfected control lane. Also, the similar band could be seen in Ctr lane in Figure 6E. Therefore, what the authors claimed as immature LGR5 band could be a nonspecific band unless they perform additional experiments (Surface biotinylation assay or EndoH assay).
7) Quantification, Statistics: They included densitometry quantification for cyclohexamide experiment and showed the quantification for SNAP data.
Overall, the authors have addressed most major and minor points from referees with compelling data or reasonable explanations. With this revised manuscript, I feel now this can be published in EMBO J. The authors performed the requested editorial changes.
3rd Editorial Decision 28th Nov 2019 Thank you for submitting the revised version of your manuscript. I have now evaluated your amended manuscript and concluded that the remaining minor concerns have been sufficiently addressed.
Thus, I am pleased to inform you that your manuscript has been accepted for publication in the EMBO Journal.
! ! ! ! " common tests, such as t--test (please specify whether paired vs. unpaired), simple χ2 tests, Wilcoxon and Mann--Whitney tests, can be unambiguously identified by name only, but more complex techniques should be described in the methods section; " are tests one--sided or two--sided? " are there adjustments for multiple comparisons? " exact statistical test results, e.g., P values = x but not P values < x; " definition of 'center values' as median or average; " definition of error bars as s.d. or s.e.m.
1.a. How was the sample size chosen to ensure adequate power to detect a pre--specified effect size? 1.b. For animal studies, include a statement about sample size estimate even if no statistical methods were used.
2. Describe inclusion/exclusion criteria if samples or animals were excluded from the analysis. Were the criteria pre-established?
3. Were any steps taken to minimize the effects of subjective bias when allocating animals/samples to treatment (e.g. randomization procedure)? If yes, please describe.
For animal studies, include a statement about randomization even if no randomization was used.
4.a. Were any steps taken to minimize the effects of subjective bias during group allocation or/and when assessing results (e.g. blinding of the investigator)? If yes please describe. the assay(s) and method(s) used to carry out the reported observations and measurements an explicit mention of the biological and chemical entity(ies) that are being measured. an explicit mention of the biological and chemical entity(ies) that are altered/varied/perturbed in a controlled manner. a statement of how many times the experiment shown was independently replicated in the laboratory.
Any descriptions too long for the figure legend should be included in the methods section and/or with the source data.
In the pink boxes below, please ensure that the answers to the following questions are reported in the manuscript itself. Every question should be answered. If the question is not relevant to your research, please write NA (non applicable). We encourage you to include a specific subsection in the methods section for statistics, reagents, animal models and human subjects.
definitions of statistical methods and measures: a description of the sample collection allowing the reader to understand whether the samples represent technical or biological replicates (including how many animals, litters, cultures, etc.).

The data shown in figures should satisfy the following conditions:
Source Data should be included to report the data underlying graphs. Please follow the guidelines set out in the author ship guidelines on Data Presentation.
Please fill out these boxes # (Do not worry if you cannot see all your text once you press return) a specification of the experimental system investigated (eg cell line, species name).
Sample size was chosen based on experiment type and on anticipated variation from studies using related methods. The number of independent experiments and technical replicates is indicated in the correspondant figure legends. graphs include clearly labeled error bars for independent experiments and sample sizes. Unless justified, error bars should not be shown for technical replicates. if n< 5, the individual data points from each experiment should be plotted and any statistical test employed should be justified the exact sample size (n) for each experimental group/condition, given as a number, not a range; Each figure caption should contain the following information, for each panel where they are relevant:

Captions
In accordance with the 3Rs, the number of animals used in the experiments was calculated considering the minimum number of animals needed to perform a statistical analyses that could give a significant difference.
No animals or samples were excluded.
We did not use a specific method for randomization.
Yes. We performed all analyses using Prism Software. The statistical methods used were detailed in the corresponding figure legends and in the method session.
We divided animals of the same age ramdonly into different groups based on their genotype, mantaining a balanced gender distribution between treatment groups and time points.
No specific blinding was used. However, when possible, analysis of the experiments was performed without knowing the group allocation or treatment (e.g. during immunostaining and analysis).
For most animal experiments, genotypes were blinded during treatments, and data analysis.

Data
the data were obtained and processed according to the field's best practice and are presented to reflect the results of the experiments in an accurate and unbiased manner. figure panels include only data points, measurements or observations that can be compared to each other in a scientifically meaningful way.