Molecular and clinical analyses of PHF6 mutant myeloid neoplasia provide their pathogenesis and therapeutic targeting

PHF6 mutations (PHF6MT) are identified in various myeloid neoplasms (MN). However, little is known about the precise function and consequences of PHF6 in MN. Here we show three main findings in our comprehensive genomic and proteomic study. Firstly, we show a different pattern of genes correlating with PHF6MT in male and female cases. When analyzing male and female cases separately, in only male cases, RUNX1 and U2AF1 are co-mutated with PHF6. In contrast, female cases reveal co-occurrence of ASXL1 mutations and X-chromosome deletions with PHF6MT. Next, proteomics analysis reveals a direct interaction between PHF6 and RUNX1. Both proteins co-localize in active enhancer regions that define the context of lineage differentiation. Finally, we demonstrate a negative prognostic role of PHF6MT, especially in association with RUNX1. The negative effects on survival are additive as PHF6MT cases with RUNX1 mutations have worse outcomes when compared to cases carrying single mutation or wild-type.

1.The interaction in Fig. 4 is potentially important.However, it was only demonstrated between wild-type proteins.What might be the consequences of RUNX1 and/or PHF6 mutations on this interaction?Can the authors show that this interaction is maintained or disrupted by prognostically important mutations of PHF6 and RUNX1 (e.g., those co-mutated in Fig. 3D-F)?Similarly, do mutations of PHF6 affect its chromatin occupancy or that of RUNX1 and vice versa? 2. The authors proposed that PHF6 and RUNX1 co-occupy enhancers.However, the current results cannot exclude the possibility that these two proteins bind to the same loci but do so independently.It would be better if the authors can use Re-ChIP to prove, at least for some candidate genes, that PHF6 and RUNX1 are in the same complex on chromatin.3.In Figs.4B and 4C, there were several bands in the blots.Can the authors clearly label which band(s) belong to RUNX1 (Runx1) and PHF6 (Phf6)? 4. What are the datasets analyzed in Fig. 4F and/or 4G?The methods suggest that these are some microarray datasets, and the figure legends are not very helpful. 5.In addition to overall survivals, can the authors analyze the effect on event-free or relapse-free survivals that better correlate with disease progression?6. Can the authors provide more details about the mutations of PHF6 and RUNX1 in patients showing additive negative outcomes (Figs.3D-F)? 7. The authors have shown that RUNX1 is one of the most mutated genes in patients with PHF6MT.What about the frequency of PHF6 mutations in patients with RUNX1 mutations?This may provide further evidence that these two genes interact in the development of MN.
Reviewer #2 (Remarks to the Author): The study by Kubota et al systematically evaluates the spectrum of PHF6 mutations across a large, multi-institutional cohort of 8443 myeloid neoplasms.This is an interesting study, and potential impactful for the field, but there are a number of technical and experimental concerns.
1.This is primarily a genomics paper yet none of the genomics data is publicly available, thus making it nearly impossible for other investigators to confirm these results.Data should be deposited for data sharing according to NIH standard (especially for those generated by CCF and supported by an R35 award).A statement that it will be available upon request is not appropriate.2. There is no mention in the methods or the data availability statement about the transcriptome data used in Figure 4F.It'll be important to understand the tumor purity of these studies as this could account for the enrichment of lymphoid markers in these studies.3. VAF/tumor fraction should be incorporated into the data presented in Figure 1.Absolute numbers should also be shown for Figure 1 panels B-D.Many of these events are likely rare in these subsets.4. Statistical review is recommended, especially regarding the lack of adjusted for multiple comparisons in Figure 1.Likewise, a thorough statistical review of the outcome data is likely warranted.Notably, a median followup of 13 months is likely too short for OS evaluation. 5.The proteomics data shows that SWI/SNF members seem to be the most significant interactors yet the focus is on RUNX1.Further, the present proteomics studies are intriguing but are overall incomplete.Do the PHF6 mutations observed in patients abolish these interactions?The selection of mouse splenocytes, which will be enriched in B and T cells, is odd for an endogenous validation of an interaction proposed to be important in myeloid cells.What about human CD34 cells?Is there any confirmation of these interactions in primary AML cells, with and without PHF6 mutations?6.The isolated IHC and flow images in Figure 4 only add confusion.The IHC images lack sufficient resolution, and controls, to be meaningful.TdT-positive AMLs can occur outside of this context (e.g.PHF6 mutations) so the overall significance is not clear.
We have extensively revised the manuscript to incorporate responses to the reviewers' critiques, which all were addressed including addition of new or improvement of existing figures.All changes have been indicated in red font in the manuscript and supplemental material and a point-by-point response to the comments is provided in blue font below.
Reviewer #1 Q1.The interaction in Fig. 4 is potentially important.However, it was only demonstrated between wild-type proteins.What might be the consequences of RUNX1 and/or PHF6 mutations on this interaction?Can the authors show that this interaction is maintained or disrupted by prognostically important mutations of PHF6 and RUNX1 (e.g., those co-mutated in Fig. 3D-F)?Similarly, do mutations of PHF6 affect its chromatin occupancy or that of RUNX1 and vice versa?Authors' response: We appreciate the reviewer's scholarly understanding of the importance of investigating the effect of PHF6 mutations on RUNX1 and the overall impact on the interactome.We focused on PHF6 centric prospective.We observed that the genetic alterations in PHF6 predominantly involved stopgains, frameshifts, and splice site mutations resulting in nonsense-mediated decay or produce abnormal proteins due to truncation of abnormal splicing (Supplementary Figure 13).Supplementary Figure 13 The fraction of truncating and non-truncating mutations in PHF6-mutated, RUNX/-mutated, and bothmutated samples.Samples with mutations in both genes have at least one truncating mutation in either This indicates that PHF6 interaction with RUNX1 would be lost in males with PHF6 mutations and in females with PHF6 mutations and chromosome X deletions.Western blotting for AML samples with frameshift or stopgain mutations in PHF6 showed low expression reiterating the above described conclusion (Supplementary Figure 14).
RUNX1-mutated samples with Runt domain nonsynonymous mutations accounted for 98% of nonsynonymous RUNX1 mutations in our cohort.IP-Western analysis demonstrated that Runt domain mutation resulted in a less abundant co-immunoprecipitation of PHF6 than samples with wild-type RUNX1 (Supplementary Figure 15).
This result suggested that RUNX1 interacts with PHF6 via Runt domain also fitting with the conclusion that PHF6 mutant samples have no or weaker interaction between PHF6 and RUNX1.The fact that most of the PHF6 mutated cases are present in cases with RUNX1 truncated after Runt domain also support the notion that the interaction between RUNX1 and PHF6 involves Runt domain (supplementary Figure 16).Of note is that chromatin occupancy of these mutated genes may be also affected because ePHD2 in PHF6 and Runt domain in RUNX1 responsible for chromatin binding are most often affected in nonsynonymous mutations in these genes.
Consequently, the precise clarification of the interaction principles between wild-type and mutant RUNX1/PHF6 may require a separate analysis involving individual mutant types/locations of RUNX1, DNA binding, and analysis of the corepressor interaction and assessment of the wild-type allele effects.
We included now a figure to that end (Supplementary Figure 13-16) and comment about these important issues: Line 105-111: "The ePHD2 domain would be essential for PHF6 function because 88% of nonsynonymous or in-frame mutations concentrated in the ePHD2 domain.This is in accordance with a previous study (PMID 32735658).As the ePHD2 domain is responsible for chromatin binding through recognition of acetylated histones, mutations in ePHD2 domain causing loss of ePHD2 domain would affect chromatin occupancy with PHF6.To support our hypothesis, we have also found that more than 90% of stopgain, frameshift, and splice site mutations are located anteriorly in the ePHD2 domain." Line 252-267: "In this study, we revealed the interaction between wild-type PHF6 and RUNX1.In case of presence of PHF6 MT , such interaction would be lost.Because, in our investigation, we observed that the genetic alterations in PHF6 predominantly involved stopgains, frameshifts, and splice site mutations resulting in nonsense-mediated decay or produce abnormal proteins due to truncation of abnormal splicing (Supplementary Figure 13).This indicates that PHF6 interaction with RUNX1 would be lost in males with PHF6 MT and in females with PHF6 MT and delX.Western blotting for AML samples with frameshift or stopgain mutations in PHF6 showed low expression reiterating the above describe conclusion (Supplementary Figure 14).RUNX1-mutated samples with Runt domain nonsynonymous mutations accounted for 98% of nonsynonymous RUNX1 MT in our cohort.IP-Western analysis demonstrated that Runt domain mutation resulted in a less abundant co-immunoprecipitation of PHF6 than samples with wild-type RUNX1 (Supplementary Figure 15).This result suggested that RUNX1 interacts with PHF6 via Runt domain also fitting with the conclusion that PHF6 mutant samples have no or weaker interaction between PHF6 and RUNX1.The fact that most of the PHF6 MT cases are present in cases with RUNX1 truncated after Runt domain also support the notion that the interaction between RUNX1 and PHF6 involves Runt domain (supplementary Figure 16)." Line 301-311: "By examining the interaction between WT PHF6 and RUNX1, we aimed to establish a baseline understanding of their functional relationship and the mechanistic implications of their interaction in cellular processes.We believe that comprehending the WT interaction is foundational for future investigations to evaluate the consequences of mutations in both PHF6 and RUNX1 on their interaction dynamics, as well as the potential disruption of chromatin occupancy.However, our study has several limitations.First, we were not able to show which domain in PHF6 is important for the interaction with RUNX1 or chromatin.As shown through our mutational analysis, we speculated that ePHD2 domain is essential for PHF6 function.Consequently, the precise clarification of the interaction principles between WT and mutant PHF6-RUNX1 may require a separate analysis involving individual mutant types/locations of RUNX1, DNA binding, and analysis of the corepressor interaction and assessment of the WT allele effects." Q2.The authors proposed that PHF6 and RUNX1 co-occupy enhancers.However, the current results cannot exclude the possibility that these two proteins bind to the same loci but do so independently.It would be better if the authors can use Re-ChIP to prove, at least for some candidate genes, that PHF6 and RUNX1 are in the same complex on chromatin.
Authors' response: We thank the reviewer for the comment.As you pointed out, our results do not entirely clarify at which enhancer site PHF6 and RUNX1 proteins act together.However, the protein complex immunoprecipitated by PHF6 antibody contains RUNX1 and vice versa (as shown in Figure 4B and 4C) confirming the point that PHF6 does interact with RUNX1.Our ChIPseq result showed that PHF6 and RUNX1 bind at multiple regions including active enhancers.Though we were unable to resolve the precise regions where PHF6 and RUNX1 bind, but it is unlikely that PHF6 and RUNX1 proteins bound independently to all the regions occupied by these proteins (please refer to Figure 4D and 4E).
Line 214-217: "Consistent with our hypothesis and co-immunoprecipitation experiments, ChIPseq showed the co-localization of PHF6 and RUNX1 in multiple regions including active enhancers ." Line 311-314: "Second, it remains to be precisely elucidated which genomic regions PHF6 co-occupy with RUNX1.However, our ChIPseq result showed PHF6 and RUNX1 bind at multiple regions including active enhancers.While possible for some sites, it is unlikely that PHF6 and RUNX1 proteins bind independently to all the regions occupied by these proteins." Q3.In Figs.4B and 4C, there were several bands in the blots.Can the authors clearly label which band(s) belong to RUNX1 (Runx1) and PHF6 (Phf6)?
Authors' response: We appreciate your comment regarding the labeling of bands in Figures 4B and 4C.We apologize for any confusion caused by the lack of clear identification of the bands corresponding to RUNX1 and PHF6 proteins.In our revised manuscript, we have addressed this concern by providing clear and specific labels for each band, ensuring accurate interpretation of the data.Also, we have added the sentences in Authors' response: We apologize for the lack of clarity.We used the Beat AML cohort for RNA-expression and DNA methylation analyses.To make it clear, we have rephrased and added the sentences as follows: Line 330-331: "Also we used the Beat AML cohort for RNA-expression and DNA methylation analyses (PMID 30333627 and 33707228)." Line 440: "For RNA-expression analysis, we used the Beat AML cohort." Line 612-616: "(F) Comparisons of mRNA expression levels in genes on X chromosome between male (n = 139) and female (n=113) cases in the Beat AML cohort.The P value was calculated using the two-tailed Student's t-test.The mean, 25th, and 75th percentiles are represented in the box plots by the midline and box edges, respectively.The whiskers extend to 1.5 times the interquartile range.Dots represent each outlier of expression." Line 617-620: "(G) Comparisons of DNA methylation levels in the promoter regions of genes mapping on X chromosome between male (n=103) and female (n=99) cases in the Beat AML cohort.The mean, 25th, and 75th percentiles are represented in the box plots by the midline and box edges, respectively.The whiskers extend to 1.5 times the interquartile range.Dots represent each outlier of DNA methylation." Q5.In addition to overall survivals, can the authors analyze the effect on event-free or relapse-free survivals that better correlate with disease progression?Authors' response: We thank the Reviewer for this important comment.We have analyzed leukemia-free survival in 4889 cases with available information of leukemic relapse and excluded open data cases with no relapse data.In accordance with overall survival data, the PHF6-mutated and the double-mutated cases revealed worse event-free survival.A detailed analysis of the double-mutated cases showed that these cases did not relapse, but did not achieve remission.Thus, conventional therapies would be not effective for these cases.We have incorporated the new supplementary figure (Supplementary Figure 7 and 10) and sentences as follows: Line 172-176: "Using our AML cohorts with available leukemic relapse information, event free survival (EFS) was also shorter in PHF6 MT cases (10% vs. 32% at 3 years, respectively; p=0.00087;Supplementary Figure 7A).Similar to OS, only PHF6 MT male cases showed significantly shorter EFS (8.5% vs. 30% at 3 years, respectively; p=0.0074;Supplementary Figure 7B-C)."Line 190-194: "In accordance with OS, the double-mutated cases revealed worse EFS (p<0.0001 in all; p<0.0001 in male; p=0.004 in females; Supplementary Figure 10A-C).A detailed analysis of the doublemutated cases showed that most of these cases did not relapse, but did not achieve remission.Thus, conventional therapies would be not effective for these double-mutated cases."

Supplementary Figure 7
Event free survival for AML cases with or without PHF6 mutations in the CCF and MLL cohort.
Kaplan-Meier survival curves of event free survival for AML in all (A), male (B), and female (C) cases.

Months Months Months
Supplementary Figure 10 Event free survival for AML cases with or without PHF6 mutations in the CCF and MLL cohort.
Kaplan-Meier survival curves of event free survival for female AML cases with double mutations (PHF6 and RUNX1), single mutations, and negative cases in all (A), male (B), and female (C) cases.
Q6. Can the authors provide more details about the mutations of PHF6 and RUNX1 in patients showing additive negative outcomes (Figs.3D-F)?
Authors' response: We appreciate your helpful suggestion.Please see the attached figure for the rebuttal.Panel A and B shows lollipop plots of PHF6 mutations in cases with double (PHF6 and RUNX1) mutations or only PHF6 mutations, respectively.Also, Panel C and D shows lollipop plots of RUNX1 mutations in cases with double (PHF6 and RUNX1) mutations or only RUMX1 mutations, respectively.There was no difference of mutated domains or mutational types between double-mutated and single-mutated cases.It is similar to RUNX1 mutations that concentrated in the Runt domain in both double-mutated and single-mutated cases.Panel E and F shows the cell fraction with mutated PHF6 or RUNX1 allele in cases with double or single mutation, respectively.As PHF6 is located on X chromosome, we used the cell fraction instead of the variant allele frequency.The cell fraction of each mutated gene also looks similar.Thus, it has meaning that both genes are mutated.RUNX1 is a master regulator for myeloid differentiation.In the 2016 WHO classification, RUNX1 mutations is considered as adverse risk factor with poor prognosis.This is because the loss of RUNX1 function would impair differentiation to myeloid/lymphoid lineage as shown in the expression analysis (Supplementary Figure 12).Also, PHF6 has function related to differentiation to myeloid/lymphoid lineages (Figure 4F and 4G).Thus, PHF6 and RUNX1 regulate differentiation to myeloid/lymphoid lineages either independently or cooperatively by interaction.In addition, PHF6 plays a tumor suppressive role through interacting with DNA repair proteins such as BLM, MSH2, or FANCI (Figure 4A).Therefore, double mutants would show more aggressive phenotype and worse prognosis because mutations in PHF6 and RUNX1 result in loss of not only their individual functions, but also their functions through interaction.The manuscript has been revised as follows: Line 287-290: "Comparing double-mutant (PHF6 and RUNX1) and single-mutant (PHF6 or RUNX1) cases, there was no difference of mutated domains or mutational types between double-mutated and single-mutated cases (Supplementary Figure 17A-D).It is similar to RUNX1 mutations that concentrated in the Runt domain in both double-mutated and single-mutated cases." Line 314-318: "Finally, we were not able to show why PHF6 and RUNX1 mutations have additive effect.This would be due to the differentiation function of PHF6 and RUNX1, and tumor suppressive function of PHF6.To unveil this additive effect, we need to elucidate the functional consequences of these mutations on protein-protein interactions and downstream cellular processes." Q7.The authors have shown that RUNX1 is one of the most mutated genes in patients with PHF6MT.What about the frequency of PHF6 mutations in patients with RUNX1 mutations?This may provide further evidence that these two genes interact in the development of MN.
Authors' response: We acknowledge that this is an important point.PHF6 mutations co-occurred with RUNX1 mutations in AML and MDS.In addition, this combination was more frequent only in male cases.In contrast, PHF6 mutations were not associated with RUNX1 mutations in MDS/MPN and MPN.The combination of PHF6 and RUNX1 may be not important for MDS/MPN and MPN or not permissive, though the number of cases is small.We have added the supplementary figure (Supplementary Figure 4) and sentences in Genomic landscape PHF6 MT of in Results section.
Line 142-146: "PHF6 MT was significantly more prevalent among cases with RUNX1 MT compared to MN cases without RUNX1 MT (4.4 vs. 1.4%; p<0.001) and significantly co-occurred with RUNX1 MT only in male cases (Supplementary Figure 4).In addition, PHF6 MT was significantly more frequent in cases with RUNX1 MT in AML and MDS (p<0.001 and p=0.032, respectively)." Reviewer #2 (Remarks to the Author): Q1.This is primarily a genomics paper yet none of the genomics data is publicly available, thus making it nearly impossible for other investigators to confirm these results.Data should be deposited for data sharing according to NIH standard (especially for those generated by CCF and supported by an R35 award).A statement that it will be available upon request is not appropriate.
ALL suggesting that PHF6 may play an important role as a component of these chromatin remodeling complexes in AML.This reinforces the potential involvement of PHF6 in chromatin remodeling and gene regulation processes.However, interaction with transcription factors including RUNX1, CBFB, and SPI1 has not been reported in any previous study.Thus, in this study, we focused to investigate the novel interaction between PHF6 and the pioneer transcription factor RUNX1.To make it clear, we have rephrased and added the sentences as follows: Line: 202-206: "Interestingly, PHF6 was also associated with several proteins in DNA repair and mRNA splicing such as MSH2, BLM, SRRT, or DDX5.PHF6 plays a role in chromatin remodeling, replication, and DNA repair not only in T-ALL but also in AML.However, in contrast to T-ALL, hematopoietic lineage defining pioneer transcription factors including, RUNX1, CBFB, and SPI1 were among the most noticeable and functionally relevant co-immunoprecipitated proteins." Further, the present proteomics studies are intriguing but are overall incomplete.Do the PHF6 mutations observed in patients abolish these interactions?
Authors' response: We observed that the genetic alterations in PHF6 predominantly involved stopgains, frameshifts, and splice site mutations resulting in nonsense-mediated decay or produce abnormal proteins due to truncation of abnormal splicing (Supplementary Figure 13).Supplementary Figure 13 The fraction of truncating and non-truncating mutations in PHF6-mutated, RUNX/-mutated, and bothmutated samples.Samples with mutations in both genes have at least one truncating mutation in either PHF6 or RUNX1.
This indicates that PHF6 interaction with RUNX1 would be lost in males with PHF6 mutations and in females with PHF6 mutations and chromosome X deletions.Western blotting for AML samples with frameshift or stopgain mutations in PHF6 showed or low expression reiterating the above describe conclusion (Supplementary Figure 14).
Therefore, the manuscript has been revised as follows: Line 252-259: "In this study, we revealed the interaction between wild-type PHF6 and RUNX1.In case of presence of PHF6 MT , such interaction would be lost.Because, in our investigation, we observed that the genetic alterations in PHF6 predominantly involved stopgains, frameshifts, and splice site mutations resulting in nonsense-mediated decay or produce abnormal proteins due to truncation of abnormal splicing (Supplementary Figure 13).This indicates that PHF6 interaction with RUNX1 would be lost in males with PHF6 MT s and in females with PHF6 MT and DelX.Western blotting for AML samples with frameshift or stopgain mutations in PHF6 showed low expression reiterating the above describe conclusion (Supplementary Figure 14)." The selection of mouse splenocytes, which will be enriched in B and T cells, is odd for an endogenous validation of an interaction proposed to be important in myeloid cells.What about human CD34 cells?Is there any confirmation of these interactions in primary AML cells, with and without PHF6 mutations?
Authors' response: It is indeed noteworthy that we have observed consistent PHF6 and RUNX1 protein-protein interactions not only in mouse spleen and bone marrow cells (Figure 4C) but also in AML cell lines, THP1 (Figure 4B), albeit both RUNX1 and PHF6 were wild type in all instances.These findings support the notion that the interaction between PHF6 and RUNX1 is present in various cell types and contexts, including both normal physiological conditions and disease states.The inclusion of AML cells (THP1) in our study adds relevance to the context of myeloid malignancies, as it demonstrates the persistence of the PHF6-RUNX1 interaction in a cell line representative of the disease.Additionally, the consistent observations across multiple cell types further strengthen the significance of the interaction and support its potential importance in physiological processes.Then, we rephrase sentences as follows: Line 208-214: "We confirmed PHF6 and RUNX1 interaction with each other in normal physiological condition with reciprocal IP western-blot using the THP-1, mouse spleens, and bone marrow cells .These results support the notion that the interaction between PHF6 and RUNX1 is present in various cell types and contexts, including both normal physiological conditions and disease states.The inclusion of AML cells (THP-1) in our study adds relevance to the context of myeloid malignancies, as it demonstrates the persistence of the PHF6-RUNX1 interaction in a cell line representative of the disease." RUNX1 stopgain mutations in male AML samples with double (RUNX1 and PHF6) mutations, or only RUNX1 mutations.Mutational groups are shown by colors as indicated.The numbers in circles indicate the number of cases.
Figure legends as follows: Line 658-661: "(B) Western blots of endogenous PHF6 IP of THP-1 nuclear protein extracts.5% input, PHF6 IP, and IgG control IP product were run side by side.The same membrane was probed with an anti-rabbit monoclonal antibody to PHF6 and an anti-mouse monoclonal antibody to RUNX1.PHF6 and RUNX1 bands are indicated by arrows in pink color."Line 662-665: "(C) Reciprocal IPs of endogenous Phf6 and Runx1 with mouse spleen and bone marrow protein extracts.5% input, Runx1 IP, Phf6 IP, and IgG control IP product were run side by side.The same membrane was probed with rabbit monoclonal antibody to Phf6 and mouse monoclonal antibody to Runx1.Phf6 and Runx1 bands are indicated by pink arrows."Q4.What are the datasets analyzed in Fig. 4F and/or 4G?The methods suggest that these are some microarray datasets, and the figure legends are not very helpful.

Q6.
The isolated IHC and flow images in Figure4only add confusion.The IHC images lack sufficient resolution, and controls, to be meaningful.TdT-positive AMLs can occur outside of this context (e.g.PHF6 mutations) so the overall significance is not clear.Authors' response: Thank you for your kind suggestion.We have corrected the IHC figures by adding sufficient resolution, removed the flow cytometry figure, and added the sentences as follows: Line: 674-675: "(H) Positive cytoplasmic immunohistochemistry staining of LY9 in blasts in 2 PHF6-mutated AML cases.Inlet shows positive control of LY9 (x500)."Line: 676-677: "(I) Positive cytoplasmic immunohistochemistry staining of GCSAM in blasts in 2 PHF6mutated AML cases.Inlet shows positive control of GCSAM (x500)."