TRIB3 promotes MYC-associated lymphoma development through suppression of UBE3B-mediated MYC degradation

The transcription factor MYC is deregulated in almost all human cancers, especially in aggressive lymphomas, through chromosomal translocation, amplification, and transcription hyperactivation. Here, we report that high expression of tribbles homologue 3 (TRIB3) positively correlates with elevated MYC expression in lymphoma specimens; TRIB3 deletion attenuates the initiation and progression of MYC-driven lymphoma by reducing MYC expression. Mechanistically, TRIB3 interacts with MYC to suppress E3 ubiquitin ligase UBE3B-mediated MYC ubiquitination and degradation, which enhances MYC transcriptional activity, causing high proliferation and self-renewal of lymphoma cells. Use of a peptide to disturb the TRIB3-MYC interaction together with doxorubicin reduces the tumor burden in MycEμ mice and patient-derived xenografts. The pathophysiological relevance of UBE3B, TRIB3 and MYC is further demonstrated in human lymphoma. Our study highlights a key mechanism for controlling MYC expression and a potential therapeutic option for treating lymphomas with high TRIB3-MYC expression.

1. In the extended data 1 the authors show that in the TRIB3 deletion models there is also deletion of TRIB2. This was not addressed at all in the experimental set up and may constitute an entirely new study. C an the authors comments how they think TRIB2 deletion is occurring in these models? 2. Given the deletion of both TRIB3 and TRIB2, can the authors show the total cell numbers for also the myeloid and B populations as shown for total thymocytes in the extended data 1. 3. It would be better to group the graphs in figure 2 d together and provide a statistical test to show the differences between the groups. 4. Details of the reporter genes in the luciferase assay in figure 2j should be named in the legend and R.L.A abbreviation explained in legend for how the data is graphed. 5. In extended data 2, it appears that while TRIB3 loss affects myc protein in most lymphoma lines, many of the leukaemia lines (eg. U937, malm6, k562, molt4 and jurkat) do not exhibit such affects on myc. The westersn, by eye, do not support the statements on line 158 pg8 and an adjustment on this interpretation may be necessary. The data supporting this mechanism in leukaemia is not as strong as the data and models the authors have for lymphoma. 6. What cells are used in extended data figure 2g and H -it is important to specify given point 5 above. 7. The data is not convincing that K332R ubiquitination is enhanced when UBE3B is overexpressed -a point that does not affect the stated mechanism but is an overinterpretation of the data. 8. What antibody is used in the IP for the tri-molecular complex shown in extended data 6 M. This needs to be in the legend/figure to clarify. 9. The authors discuss treatment with DOX. Page 15 line 297. Please explain and clarify what this is/why it is being used to treat the cells.
Reviewer #4 (Remarks to the Author): In this paper, Li et al describe a molecular mechanism by which TRIB3 promotes lymphoma genesis via preventing proteasome mediated degradation of c-MYC . First, I'd like to commend the authors for their highly detailed mechanistic study, employing a range of mouse models and human/murine in vitro systems to demonstrate mechanistic details for the observed oncogenic role for TRIB3. I have no fundamental concerns with the data presented, but have a number of relatively minor comments that would hopefully improve the presentation of the data and make this work more accessible to readers not directly working in this field.
-The work is very elaborate and a hough amount of data is presented in each figure. As a consequence, it is almost impossible for the outside reader to have a high-level view of the message at present. I would highly recommend the development of a schematic capturing the main points of the molecular model the authors propose and incorporate this is the discussion/final figure.
-There are inconsistencies in abbreviating human/mouse genes and proteins throughout the manuscript. for instance, by convention, protein symbols are not italicized, and all letters are in upper-case, both for human and murine proteins. (see https://www.biosciencewriters.com/Guidelines-for-Formatting-Gene-and-Protein-Names.aspx). This needs to be addressed both in the text and figures.
-Were MycEμC reC D19Trib3F/+ mice homo or heterozygous for the MycEμ allele? Please clarify this in the manuscript/figures -Page 11: what is MAX?? Please ensure that you introduce the various genes you study and describe in this work.
-Ext. Fig. 1 G: Please at least comment on reduced TRIB2 levels you see in the Trib3 KO animals. TRIB2 has been shown to be oncogenic and some speculation at least whether there is a genetic/functional hierarchy between TRIB2 and TRIB3 in this context would be interesting. Better still, it would be a further confirmation of TRIB3-specificity of this mechanism if you had evidence to show that restoration of TRIB2 levels (for instance in Raji cells) has no impact on the system you studying.
-please include an alignment to show how specific C M4 is for TRIB3, compared to the other TRIB proteins in the supplementary material.
-Please comment in the manuscript what is the significance of the statement you make in page 5 ""Targeted sequencing of TRIB3 showed that the allele frequency of nonsynonymous (Q84R) and synonymous (Y111Y and A323A) SNPs was separately 38.2% and 47.2% in lymphoma patients..." -There are a number of places where the manuscript will benefit from a careful proof-reading and correction of English. a few examples are copied below: P4: "none of the inhibitors binding directly to c-Myc has been successfully…" "We recently report that TRIB3…"

Point-by-point response
Reviewer #1 (Remarks to the Author): The authors describe a novel inhibitor/ubiquitin ligase pair and describe significant and interesting biological effects in a MYC-driven lymphoma model. They go on to ascribe these effects to a direct effect of both proteins on MYC ubiquitination and turnover. Finally, they transfect a MYC peptide and claim that the effects of this peptide validate the interaction in vivo.
1. The major and critical weakness of the paper is that the detailed mapping experiment in Extended Data Figure 6 shows that the interaction of MYC broadly maps to the amino terminus and that deletion of the leucine zipper has -relative to input -no effect on the interaction of MYC with either TRIB3 or UBE3B. However, the critical peptide used to validate the relevance of the interaction, CM4, is derived from the leucine zipper of MYC. Hence the experiments with CM4 are completely irrelevant for validating the interaction.
Re: Thank you for your professional criticism. In the revised MS, we thus redesigned and constructed three main truncations of MYC (N-terminal domain, central region and C-terminal domain) and identified that the C-and N-terminal domains, but not the central region (CT) of MYC mediated the interactions of MYC with either TRIB3 or UBE3B. Among them, the C-terminal domain of MYC showed a major but the N-terminal domain showed a minor contribution to its binding with TRIB3 or UBE3B (Revised Fig. 5g left and 5h left). In addition, we performed the reverse Co-IP assay to examine the interactions of MYC and TRIB3 or UBE3B with anti-MYC truncation antibodies (anti-GFP). Similarly, the C-terminal MYC indeed possessed a strong binding affinity to TRIB3 or UBE3B, while the N-terminal MYC weakly bind to TRIB3 or UBE3B, and the CT domain of MYC did not exhibit any detectable level of binding (Revised Fig. 5g and 5h right). These results suggest that the C-terminal domain of MYC binds to TRIB3 or UBE3B with high affinity and that the N-terminal domain of MYC has a weak binding affinity to TRIB3 or UBE3B. Based on these results, it is reasonable to use CM4 derived from the C-terminal domain of MYC to validate the relevance of the interaction of MYC and TRIB3. Following your and the editor's suggestion, we have corrected the corresponding description (Page 14) and toned down the interpretations of peptide CM4 and the interaction of MYC and TRIB3 throughout the revised MS.
In the original MS, the differential expression levels of several MYC mutants in the input samples and the overloading of IP samples caused this reviewer's criticism regarding the Extended Figure 6 of original MS. In the revised MS, we re-conducted Co-IP assays in a reverse way together with adjusted plasmid transfection process and appropriate exposure time of western blotting. The data indicate that deletion of LZ and HLH diminished the binding of MYC and TRIB3 or UBE3B more apparently than that of the N-terminus (Below panels a and b). Deletion of all three domains of MYC almost lost its binding with TRIB3 or UBE3B (Below panels a and b). However, these data did not provide more information for the conclusion of our study, and may bring confusion to reviewers. Hence, we present them for your reviewing and will not display them in the revised MS. 2. The discussion of this central problem in the text is highly misleading. The authors state: "Given that the TAD, HLH and LZ domains of the c-Myc mediated the c-Myc/TRIB3 interaction--(line 267) but this is simply not true based on the Extended Data Figure 6.
Re: Based on the reply to your question 1, we thus deleted this misleading description in the revised MS. New data presented in the revised Fig. 5g suggest that the C-terminal domain of MYC binds to TRIB3 with high affinity. Hence, we modified the description regarding the interacting domains of MYC and TRIB3 in the revised MS (Page 14). Following the editor's suggestion, we have also toned down the interpretations of the MYC-TRIB3 interaction in the revised MS.
3. In the most positive interpretation of the interaction data, the LZ makes a very small contribution to binding to TRIB3 when the TAD is deleted (Extended Data Figure 6g, compare ∆TAD to ∆TAD-∆LZ). But if anything, this difference is larger for UBE3B, so a leucine zipper peptide, which competes with endogenous MYC, if it has any effect at all, would preferentially displace the UBE3B ligase and hence stabilize MYC.
Re: Thank you for your professional criticism. As we replied in question 1, new truncations of MYC were used to conduct Co-IP assays, by which cell lysates were immunoprecipitated with anti-GFP (MYC truncations' tag) antibodies or anti-Flag (TRIB3 and UBE3B's tag) antibodies in the revised MS. The data indicate that the C terminus of MYC mainly mediates the interaction of MYC with UBE3B or TRIB3 ( Revised Fig 5g and 5h). Hence, we modified the description of these data in the revised MS (Page 14).
Because TRIB3 and UBE3B share the same binding domain (mainly C terminus) of MYC, CM4 peptide derived from the C terminus of MYC can displace the UBE3B ligase and stabilizes MYC. However, there exists the possibility that TRIB3 and UBE3B bind to the distinct regions or residues of C terminal MYC, which deserves further exploration in future study. Moreover, distinct peptides derived from the C terminal MYC may differentially bind with TRIB3 or UBE3B. In this study, we have identified via experiments that CM4 preferentially disturbed the binding of TRIB3 and MYC ( Revised Fig 6c and 6d), subsequently restored the interaction of MYC and UBE3B ( Revised Fig 6e and 6f), and destabilized MYC protein ( Revised Fig 6i and 6j). Furthermore, we found that cell-penetrated CM1 (PCM1) peptide derived from HLH domain of MYC impeded the interaction of MYC and UBE3B (Below panel a) and thus reduced the ubiquitylation of MYC (Below panel b). In this study, we emphasize CM4 peptide but not CM1 to validate the mechanism of TRIB3 in the regulation of MYC function. Hence, we present these data for your reviewing and will not display them in the revised MS. clearly shows that the CM4 peptide binds MAX, this has to be the case. This data is also mis-portrayed in the text. Disrupting the MYC/MAX interaction is known to affect all aspects of MYC biology and -importantly-to destabilize the MYC protein, so all reported effects of CM4 are expected based on what is known about the MYC/MAX interaction and the effects of 10058-F4, a soluble inhibitor that disrupts the interaction. This problem and the manner in which the authors deal with it in the text precludes publication of the manuscript.
Re: Thank you for your professional criticism. In the revised MS, we further examined the effects of PCM4 on the co-localization and the interaction of MYC and MAX through confocal and CO-IP assays. The PLA assay showed the reduced foci of MAX/MYC co-localization in PCM4 treated Raji cells ( Supplementary Fig. 7m). Interestingly, in the examination of PCM4's effects on the interaction of MYC and MAX by CO-IP assay, if sufficient MAX antibody (5 μg) was added to the lysates to capture the MAX protein, the quantity of the precipitated MYC protein in control cells was much more than that in PCM4 treated cells ( Supplementary Fig. 7n left). However, if less anti-MYC antibody (1 μg) was added to capture the same quantity of MYC protein in the CO-IP assay, and the quantity of the precipitated MAX protein in control cells was identical with that in PCM4 treated cells ( Supplementary Fig. 7n right). These data show that reduced MYC abundance induced by PCM4 contributes to the observed reduction of the MYC/MAX interaction.
Following your and the editor's suggestion, we have modified the description of the effects of CM4 on the MAX/MYC interaction in the revised MS (Page 16). Especially, we toned down the clarification of peptide CM4 specifically binding with TRIB3, and pointed out that disrupting the MYC/MAX and MYC/TRIB3 interactions contributed to anti-lymphoma effects of CM4 in the revised MS (Page 16).
There are additional issues, several of them major.  Fig. 4h). The corresponding description was indicated on line 207, 208 and 209 of revised MS (page 11).
Following your suggestion, CRISPR/Cas9 technology was used to generate 3. The description of many experiments is cryptic and often panels lack critical controls. Just to give one example: Figure 4B Fig. 4a, 4b and Supplementary Fig. 6c). The corresponding description has been presented in page 12 of revised MS.
It was a mistake for the inaccurate description of Figure 4a legend. We have corrected it in the revised MS (page 54).

4.
A large set of ChIP-sequencing data are based on quantitative comparisons between two sample sets, e.g. wtMYC and a mutant (Extended Data Figure 5a, 5e).
In some cases, they show different individual genes and this is not acceptable: within 20,000 genes you will see any change possible. Where the authors report global changes, they is in line with the expected effects. However, such data need to be spike in-normalized to control for experimental variations: a small percentage of chromatin from another species needs to be added and used to standardize. The experimental methods show that this has not been done, therefore much of this work needs to be repeated.

Re:
Following your suggestion, we re-performed all the ChIP-sequencing experiments using mouse chromatin spike-in for normalization. The corresponding experimental methods have been described in the revised Methods (Page 32 and 33). We also show change of the same individual gene (CCNA2) if two sample-sets have difference. These new data of ChIP-seq have been presented in revised Figure 2k, 2l, 2m, 4a, 4b, 4d, S6c, S6f, 6m and 6n.

Reviewer #3 (Remarks to the Author):
This study is very well executed showing a novel mechanism into Myc driven lymphomas, linked to a TRIB3 interaction, and the possible means of targeting this interaction to abrogate myc driven lymphomas. The experiments are well designed, well executed, and the data fully supports the finding. The paper is fully complete with little room for improvement. I only have minor comments: 1. In the extended data 1 the authors show that in the TRIB3 deletion models there is also deletion of TRIB2. This was not addressed at all in the experimental set up and may constitute an entirely new study. Can the authors comment how they think TRIB2 deletion is occurring in these models? 2. Given the deletion of both TRIB3 and TRIB2, can the authors show the total cell numbers for also the myeloid and B populations as shown for total thymocytes in the extended data 1.
Re: Following your suggestion, we have presented the total numbers of splenic B cell, T cell and myeloid cell populations in 5-week old mice via flow cytometry analysis.
Trib3 ablation in myeloid/thymocyte/B cells did not affect the total number of thymocytes, splenic B cell, T cell and myeloid cell populations (Revised Supplementary Fig. 1f-1i), indicating that specific ablation of Trib3 in these cells does not affect lymphocyte and myeloid cell development.
3. It would be better to group the graphs in figure 2d together and provide a statistical test to show the differences between the groups.
Re: Following your suggestion, we have regrouped the graphs together and provided a statistical data to this panel, which has been presented in Revised Figure 2d  5. In extended data 2, it appears that while TRIB3 loss affects Myc protein in most lymphoma lines, many of the leukaemia lines (eg. U937, malm6, k562, molt4 and jurkat) do not exhibit such affects on Myc. The western, by eye, do not support the statements on line 158 pg8 and an adjustment on this interpretation may be necessary. The data supporting this mechanism in leukaemia is not as strong as the data and models the authors have for lymphoma.
Re: Thank you for your careful observation. Following your suggestion, we have modified the description of these data in the revised MS (Page 8).
6. What cells are used in extended data figure 2g and H -it is important to specify given point 5 above.
Re: Raji cells are used in extended data figure 2g and H. We have added this information in the legend of Supplementary Fig 2f and 2g. 7. The data is not convincing that K332R ubiquitination is enhanced when UBE3B is overexpressed -a point that does not affect the stated mechanism but is an overinterpretation of the data.

Re:
We agree with your point. We have thus deleted the data of K332R ubiquitination regulated by UBE3B from the revised MS. Re: Doxorubicin (DOX) is widely used as a chemotherapeutic agent for the treatment of lymphoma, which was illustrated in the Page 17 of the revised MS. In this study, we found that TRIB3 deletion or PCM4 treatment enhanced the therapeutic efficacy of DOX in lymphoma, suggesting that the combination of DOX with TRIB3 deletion or PCM4 produces a synergistic anti-lymphoma efficacy.
Reviewer #4 (Remarks to the Author): In this paper, Li et al describe a molecular mechanism by which TRIB3 promotes lymphomagenesis via preventing proteasome mediated degradation of c-MYC. First, I'd like to commend the authors for their highly detailed mechanistic study, employing a range of mouse models and human/murine in vitro systems to demonstrate mechanistic details for the observed oncogenic role for TRIB3. I have no fundamental concerns with the data presented, but have a number of relatively minor comments that would hopefully improve the presentation of the data and make this work more accessible to readers not directly working in this field.
-The work is very elaborate and a hough amount of data is presented in each figure.
As a consequence, it is almost impossible for the outside reader to have a high-level view of the message at present. I would highly recommend the development of a schematic capturing the main points of the molecular model the authors propose and incorporate this is the discussion/final figure.
Re: Following your suggestion, we have created a schematic diagram that illustrates the main points of the working model we proposed. The schematic diagram was shown in the Revised Figure 9 (Page 66).
-There are inconsistencies in abbreviating human/mouse genes and proteins throughout the manuscript. for instance, by convention, protein symbols are not italicized, and all letters are in upper-case, both for human and murine proteins. (see https://www.biosciencewriters.com/Guidelines-for-Formatting-Gene-and-Protein-Nam es.aspx). This needs to be addressed both in the text and figures. Re: MAX (MYC-associated factor X), a member of transcription regulators, forms a sequence-specific DNA-binding protein complex with MYC and the MYC/MAX complex functions as a transcriptional activator. We have annotated this in the revised manuscript (Page 12).

Re
-Ext. Fig. 1 G: Please at least comment on reduced TRIB2 levels you see in the Trib3 KO animals. TRIB2 has been shown to be oncogenic and some speculation at least whether there is a genetic/functional hierarchy between TRIB2 and TRIB3 in this context would be interesting. Better still, it would be a further confirmation of TRIB3-specificity of this mechanism if you had evidence to show that restoration of TRIB2 levels (for instance in Raji cells) has no impact on the system you studying. Following your suggestion, we further restored TRIB2 expression in TRIB3-deleted Raji cells and examined the proliferation and self-renewal of these cells. We found that TRIB2 overexpression partially rescued the decreased cell growth of TRIB3deleted Raji cells (Below panel a), but showed no impact on colony-forming capacity of these cells (Below panel b). These data suggest that reduced TRIB2 partially contributes to the inhibition of proliferation but not that of self-renewal induced by TRIB3 deletion in lymphoma cells. Indeed, we will perform an entirely new study to examine whether and how the level of TRIB2 is regulated by TRIB3 in cancer cells. These data are shown for your review (Below panels a and b) but not presented in the revised MS. -please include an alignment to show how specific CM4 is for TRIB3, compared to the other TRIB proteins in the supplementary material.
Re: Homology modeling based on the TRIB1 crystal structure (Murphy et al. Structure 2015, 23 2111-2121) predicted a hydrogen bond between Glu344 (E344) of the TRIB3 protein and the CM4 peptide (Revised Supplementary Fig. 7d), suggesting that the C terminus of TRIB3 is the binding domain of CM4. We have performed sequence alignment within the C termini of TRIB family members and found that E344 of TRIB3 is not conserved at the same positions on TRIB1 and TRIB2 (Revised Supplementary  Fig. 7e), indicating that CM4 is specific for binding to TRIB3.
-Please comment in the manuscript what is the significance of the statement you make in page 5 "Targeted sequencing of TRIB3 showed that the allele frequency of nonsynonymous (Q84R) and synonymous (Y111Y and A323A) SNPs was separately 38.2% and 47.2% in lymphoma patients..." Re: Following your suggestion, we have added described these data and commented this point in the revised MS (Page 5).
-There are a number of places where the manuscript will benefit from a careful proof-reading and correction of English. a few examples are copied below: P4: "none of the inhibitors binding directly to c-Myc has been successfully…" "We recently report that TRIB3…" P10" We thus constructed UBE3B R346Q mutant. This mutant showed a reduced abilities of c-Myc ubiquitination..." Re: Thank you for your pointing out these writing and grammar mistakes, we have carefully corrected and modified the manuscript. A native English speaker has proofread our revised manuscript throughout.
The authors have addressing almost in full my critiques. I would ask for an amendment to figure 2d. The statisitical analysis is not indictated in the legend for each comparison, and please add a comment in the body of text on page 7, line 134, whether the data is statistically significant for the cell doses used in the comparisons.
Reviewer #4 (Remarks to the Author): The manuscript has been throughly revised and addressed all my points of concern.
Reviewer #5 (Remarks to the Author): This manuscript provides a comprehensive and detailed description of the interaction of MYC with TRIB3 that stabilises MYC protein by disrupting UBE3B-mediated ubiquitination and degradation of MYC . The results are novel and interesting and should be published. The manuscript appears to have been substantially revised and improved following the first round of reviews. The data are well-presented with appropriate controls.
I have several minor points that can largely be answered in a modified text (see below). However, there remain two important issues.
First, the cell lines used express elevated levels of MYC whether or not rearrangements of the Myc gene are present. Given the high starting level of MYC and the relatively modest effects on MYC half-life elicited by changes in TRIB3 or UBE3B function there appears to be surprisingly large effect on MYC transcriptional output. Is there a TRIB3-senstive MYC population (phospho-MYC ?) that exhibits quantitatively different transcriptional activity that can account for the reported reduction of MYC -dependent transcription when TRIB3 is inhibited? After all, steady state MYC protein levels are easily detected in cells lacking TRIB3 presumably since MYC expression is constitutive, albeit with a reduced half-life. If only the stability of phospho-Myc (eg at T58) is affected, as shown in Supplementary Fig. 5, then perhaps phspho-T58 MYC should be shown (as in Supplementary Fig. 4l). What proportion of total MYC is phosphorylated at T58 in these cells?
Second, the data regarding the C M4 peptide is confusing and overly interpreted. For example, in Supplementary Fig. 7n the level of input MYC is lower in the presence of PC M4 in the left panel but this does not seem to be true in the right panel? In addition, I'm not sure that I follow the argument behind using "less anti-MYC antibody" for the IP in this figure. Surely 1µg antibody is saturating and the amount of MYC /MAX complex is the same between PC M4 and control. Are the authors suggesting that this amount of MYC /MAX complex is sufficiently low to account for the reduced MYC transcriptional output ( Supplementary Fig 7o)? If so, these data are not convincing. It is also possible the effect of C M4 is mediated by inhibition of the MYC /MAX heterodimer independently of effects on TRIB3 and UBE3B interactions -this has not been adequately addressed in the manuscript.
Minor comments: 1. The efficiency of C re in deleting Trib3 is clearly variable (compare Supplementary Figs. 1c and 1d) but I do not understand why Trib3 is completely absent from Trib3+/F as well as Trib3F/F thymocytes, especially when there appears to be robust expression in wt cells. Moreover, it is incorrect to imply that Trib3 is lost since there is clearly some expression even in Trib3F/F bone marrow and B lymphocytes. 2. I may have overlooked this but I can't find a description of the C reERT2 mice (Fig. 1j). Is C reERT2 expressed in all cells? 3. The title "TRIB3 Promotes Lymphoma Through MYC -Driven Malignant Activity" suggests that TRIB3 affects a peculiar neomorphic "malignant" activity of Myc, rather than promoting MYC 's normal function which, when persistently activated, can lead to cancer. 4. The Ki-67 assay seems to be based on Trib3F/F cells whereas the C FU assay was on Trib3F/+ B cells (Fig 2b and 2c) -why is this the case? Given the effects seen in the C FU assay and in the mice it is important to show whether proliferation is also affected in Trib3F/+ cells. 5. The legend indicates that Fig. 2g shows "relative cell viabilities". I'm not clear whether the authors are referring simply to proliferative capacity or whether cell death was also observed? This is an important point -if Trib3 deletion inhibits the cell cycle (independently of Myc) then constitutive elevated expression of Myc is likely to induce apoptosis. There is also no mention of the possibility that Trib3 deletion induces differentiation (and cell cycle arrest) in, at least, some of the cell lines used. 6. GO term analysis is often misleading and that shown in Fig. 2h contributes very little and, given the density of data presented, could be removed. Likewise the GSEA in Fig. 2i Fig. 2k-m) is circumstantial. The effect of directly reducing Myc expression (eg RNAi) should be included as a control. 9. siTrib3 seems to have no effect on Myc protein expression in U937 cells (Fig. 3a). Was TRIB3 not repressed in U937 cells or is it not required for Myc stability in these cells? 10. Are the graphs shown in Fig. 3b and 3c densitometry scans of the representative westerns displayed in Supplementary Fig. 2g? If so, do the error bars represent technical or biological replicates and how many? This is a common failing throughout presentation of the figures and this information should be include in figure legends. 11. Fig. 3d -is it not slightly surprising that the steady-state level of Myc is not higher in the Flag-TRIB3 input sample, especially considering the apparent reduction in Myc ubiquitination. The same question applies to Supplementary Figs. 4h and 4i. 12. Why, in Supplementary Fig. 3i is the level of Myc protein so much lower in Raji cells with siC trl in the siUBE3B panel compared to the siUBE3A and siUBE3C panels? 13. The authors overstate the result shown in Supplementary Fig. 4k. "The T58A mutation prevented UBE3B from binding to MYC ". I agree that there is a reduction in MYC T58A binding but it is not "prevented". 14. The authors suggest that UBE3B-mediated K427 ubiquitination of MYC interferes with the MYC /MAX interaction ( Supplementary Fig. 6k). Given that interaction of MAX and wt MYC is still observed it is important to determine the biological output of wt and K427R MYC in the absence and presence of UBE3B in Raji cells (eg proliferative assay as shown in Supplementary Fig. 6l). Was Ad-UBE3B present in Supplementary Fig. 6k? It is not mentioned in the figure legend. 15. The authors suggest that UBE3B ablation also induces the ubiquitination and degradation of other tumor suppressor(s) (page 13, line 258). Is the reduction in Raji cell number (Fig. 4f) solely due to reduced proliferation or was increased cell death also noted? 16. The data in Fig. 4 suggests that overexpression of TRIB3 alone does not increase spleen weight, lymph node size or proliferation -is this because endogenous levels of TRIB3 are already sufficient to suppress endogenous UBE3B activity? 17. Page 15, line 298 -I think the authors are referring to Fig. 6a (and not Fig. 5a). 18. In Fig. 6o do the two sets of bars on the graph represent two different isolates of primary DLBC L cells?
The authors have addressing almost in full my critiques. I would ask for an amendment to figure 2d. The statistical analysis is not indicated in the legend for each comparison, and please add a comment in the body of text on page 7, line 134, whether the data is statistically significant for the cell doses used in the comparisons.
Re: Following your suggestion, we have added a comment in the body of text on Page 7, Line 135-137. A significant difference could be observed between the survival rates of recipient mice inoculated from Myc Eμ Cre CD19 Trib3 F/+ and Myc Eμ mice for the transplantation cell dose of 5*10 4 , suggesting that BCL cells from Myc Eμ Cre CD19 Trib3 F/+ mice have a lower capacity for killing the recipient mice (Fig. 2d). We have also indicated the statistical analysis in the legend of Fig 2d for the comparison between the groups of the transplantation cell dose of 5*10 4 .
Reviewer #5 (Remarks to the Author): This manuscript provides a comprehensive and detailed description of the interaction of MYC with TRIB3 that stabilises MYC protein by disrupting UBE3B-mediated ubiquitination and degradation of MYC. The results are novel and interesting and should be published. The manuscript appears to have been substantially revised and improved following the first round of reviews. The data are well-presented with appropriate controls.
I have several minor points that can largely be answered in a modified text (see below). However, there remain two important issues.  Supplementary Fig. 5, then perhaps phspho-T58 MYC should be shown (as in Supplementary Fig. 4l). What proportion of total MYC is phosphorylated at T58 in these cells? Re: Thank you very much for your constructive suggestions. As you proposed in this point of comment, we examined whether there exists a TRIB3-senstive MYC population that exhibits quantitatively different transcriptional activity of MYC  . We used phospho-specific antibodies to accurately quantify the level of Phospho-T58 (P-T58) and Phospho-S62 (P-S62) relative to total MYC in TRIB3 depleted cells. Upon TRIB3 knockdown in Jurkat and Raji cells, we observed a decrease in the P-S62 signal, but no change in the P-T58 signal (Supplementary Fig. 2k and 2l). Quantifying the ratio of P-T58 to total MYC revealed a substantial increase in the amount of T58 phosphorylated MYC with TRIB3 knockdown (Supplementary Fig. 2k and 2l). In addition, the signal for P-S62 relative to total MYC decreased in TRIB3 depleted cells compared with control group (Supplementary Fig. 2k and 2l). Similarly, TRIB3 overexpression increased S62 and decreased T58 phosphorylation relative to total MYC level in Raji cells ( Supplementary Fig.  2m). Furthermore, the analysis of half-life of endogenous MYC protein revealed that TRIB3 deletion reduced the stability of MYC protein, especially accelerated P-T58 MYC protein degradation after cycloheximide treatment, as compared to control gRNA ( Supplementary Fig. 2n), suggesting mainly the stability of T58 MYC protein is affected in TRIB3 deleted cells. These data indicate that the altered S62 and T58 phosphorylation of MYC also contributed to the MYC transcriptional output and regulation of MYC protein stability in TRIB3-depleted cells.
We next explored the mechanism by which TRIB3 regulated MYC phosphorylation at T58 and S62. A number of reports have identified that GSK3B phosphorylates T58, whereas PP2A dephosphorylates S62 (Arnold et al. . We found that endogenous GSK3B and PP2A-C co-immunoprecipitated with TRIB3 in Raji lymphoma cells (Below panel a). TRIB3 overexpression in lymphoma cells decreased the amount of GSK3B and PP2A that co-immunoprecipitated with MYC approximately 60% (Below panel b). Thus, TRIB3 post-translationally regulates MYC T58 and S62 phosphorylation consistent with its effects on the ability of MYC to associate with PP2A-C and GSK3B. Because we focused on investigating the role of TRIB3 in the regulation of MYC ubiquitylation in this study, these data are shown below only for your reviewing butl not displaying in the revised MS.