KLF7 is a general inducer of human pluripotency

  1. Mattia Arboit
  2. Irene Zorzan
  3. Marco Pellegrini
  4. Paolo Martini
  5. Elena Carbognin
  6. Graziano Martello

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Pluripotency is the capacity to give rise to all differentiated cells of the body and the germ line and is governed by a self-reinforcing network of transcription factors. The forced expression of only some of these factors enables the reprogramming of somatic cells to pluripotency. In murine cells, several kruppel-like factors (KLFs) have been identified as stabilisers and inducers of pluripotency. Human somatic cells are routinely reprogrammed by expression of KLF4 in combination with OCT4, SOX2 and cMYC (OSKM). An extensive transcriptome analysis revealed, however, that KLF4 is barely expressed in conventional human pluripotent stem cells (PSCs). Here we show that KLF7 is robustly expressed in conventional human PSCs and it allows transcription factor-mediated somatic reprogramming. Moreover, we find that KLF7 is highly expressed in naive PSCs and its forced expression in conventional hPSCs induces upregulation of naive markers and boosts efficiency of chemical resetting to naive PSCs, overall suggesting that KLF7 is a general human pluripotency factor and an inducer of pluripotency.

Article activity feed

  1. Review Commons

    Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    Reviewer #1 (Evidence, reproducibility and clarity (Required)): Evidence reproducibility and clarity The authors report that the human-specific KLF factor KLF7 can induce pluripotency in humans and can improve the reset toward naïve pluripotency when cells are cultured the PXGL medium. KLF7 falls behind KLF4 in reprogramming efficiency but might have a unique role in naïve reset (10-20 fold less efficient in iPSC colony yield). The topic of the study is interesting and adds important insights into the roles of KLF factors along the pluripotency continuum and pinpoints differences between mice and human. There are implications for stem cell engineering and boosting the developmental potency of stem cells (blastoid formation potential, interspecies chimera formation). However, some of the claims as to the unique role of KLF7 are unconvincing in the absence of comparison with other KLF factors, especially the Yamanaka factor KLF4. The flow and coherence of the text can be improved - at times reasoning and motivation of experiments are hard to follow

    Major comments

    • Why would a pan-pluripotency factor KLF7 which is expressed in both primed and naïve cells more potently trigger the naïve reset than the naïve specific factors KLF4/5/17? Such a comparison could widen the scope and interest of their work.. I would find it interesting if authors would compare the ability of key KLF factors to induce naivety. This is of particular interest as the overexpression of engineered Sox along with KLF4 was reported to improve the quality and developmental potential of PSC in multiple species (MacCarthy et al bioarxiv). Such an analysis could reveal unique features of KLF family members and lead to advanced stem cell models. They actually claim the SK naïve reset does not require naïve medium but the expression of SK alone is sufficient to induce this state. What do the authors think about this claim? Overall I feel the potential role of KLF7 in naïve reset is interesting but underdeveloped.

    We thank the reviewer for the useful comments.

    It has been shown in murine PSCs, that the pluripotency factors Nanog and Oct4 are expressed both at the naive and primed state and their forced expression, in combination with a medium supporting naive pluripotency, efficiently resets primed murine PSCs to naive (Radzisheuskaya A. et al., 2013; Theunissen T. et al., 2011). It is therefore not surprising that a similar regulation might also be conserved in human and that the general pluripotency factor KLF7 is expressed in both states and drives efficient resetting.

    Moreover, we agree that a direct comparison with another KLF factor could improve our work, so thanks to the reviewer’s suggestions, we will generate conventional/primed hPSCs with exogenous KLF4 expression in order to assess the efficiency of chemical resetting compared to hPSCs with overexpression of KLF7.

    Minor comments

    • P3, line 52: "Surprisingly, however, KLF4 is also routinely used to generate conventional human iPSCs." Why is this surprising? KLF4 (and SOX2) are the most potent iPSC factors whilst MYC and OCT4 can be omitted (at least in mouse).

    Thank you for pointing this out. We have rephrased the text accordingly (line 51).

    • It would be nice if the demonstration of pluripotency and quality of KLF7 iPSC go beyond transcriptome profiling and included some further assays common in the field.

    We assessed the quality of our OSK7M iPSCs by performing an EBs differentiation assay (Fig. 3d). We rephrased the text to further highlight this experiment (line 106). Of note, *in vivo *assays like teratoma formation are not allowed in Italy due to official regulations on animal testing.

    • Fig 1A-B: color coding (of dots) is very confusing- which ones are PSCs and which ones are iPSCs? Another colour palette might fix. What is meant by "interrogating previously published data" (line 67)? Are these public RNA-seq data that were re-analyzed? I

    We will highlight in the figures which cells are PSCs or iPSCs using different colours and shapes.

    We rephrase the text to clarify that available RNA-seq data were reanalysed (line 67).

    • Fig 2b: how were the colony numbers obtained? By morphology, or using live cell staining? So form of staining is recommended colony counting (i.e. TRA-1-60).

    We scored colonies both based on their morphology and after OCT4/NANOG staining. Actually, we observed that the counting based on morphology underestimated the number of iPSC colonies, so it is a more stringent method to score reprogrammed cells.

    • Fig 2e: Also, they say that "[t]hree technical replicates were carried out for all quantitative PCR". Unless I'm mistaken, it seems that only two technical replicates were performed for these qPCR reactions (two dots visible per bar).

    In figure 2e dots refer to two independent experiments. In each experiment we carried out three technical replicates for each sample.

    • Fig 3c: "colture"; change to "culture" (and the title: "bone fide" should be "bona fide")

    Thank you. We amended the typos in the figure and in the text (line 111).

    • For Fig 2/3: since the paper is on KLF4/7, I'm surprised that expression levels of OCT4 and SOX2 were analysed but not KLF4. Given that the main finding was that KLF4 was not upregulated in PSCs, I would be interested to see what the KLF4 levels are like in the iPSCs. RNA-seq analysis/qPCR would be best; but if the authors would like to use other methods, that's fine too.

    This is a good suggestion, we will add to Figure 3b the KLF4 expression levels.

    • Fig 4: The explanatory text is too sparse. Readers should be reminded of the differences between of naïve and primed PSCs and the known roles of KLF4 (this could also be improved in the introduction). List names of naïve media used on top of author names (5iLA, PXGL, EPSCM etc). Why was HENSM by Hanna excluded?

    We will amend the text explaining the main differences between naive and primed PSCs and the role of KLF4.

    We will add PSCs derived in the HENSM medium in the analyses shown in figure 4.

    • Fig 5: KLF7 is classified as a general pluripotency marker, but KLF4/KLF17 are classified as naïve markers. In that case, wouldn't it make more sense to overexpress a naïve specific marker in order to achieve naïve iPSCs at least as a control? What was the motivation here? I think the authors need to provide a more compelling reasoning why only KLF7 was studied or add more data for other KLFs (especially since it seems that the reprogramming efficiency of KLF4 is higher than that of KLF7 for conventional reprogramming (see Fig 2B)...)

    We will perform resetting experiments using KLF4, as suggested, in order to compare the efficiency of KLF7 to a known naive factor.

    o Fig 5B: the text currently says that the cells on the left side of Fig 5B are from Day7; but it says the cells are from Day0 in the actual figure. Which one is it? Also, based on how the text is written, do the cells on the left also contain EOS, or are they the wild-type variety?

    We agree that the text was confusing. Colonies appeared at day 7, but we showed them at day 12, when they were larger and easier to see. We amended the text accordingly. Moreover, the images at day 0 are simply the cell lines at the beginning of the resetting, which also contain EOS, as quantified on the right panels of Fig. 5b.

    o Fig 5c: not all markers in this figure are naïve markers (as stated in the text); would suggest separating the markers and labelling them accordingly AND rewriting the text to reflect that.

    We labelled the markers in the Fig. 5c as suggested by the reviewer and rephrased the text (line 136-137).

    o Life cell reporters for naivety (CD75,SUSD2) could enrich this study.

    We believe that the combination of bulk RNAseq and immunostaining for functional regulators of naive pluripotency (i.e. KLF17 and OCT4 (Lea et al., 2021 Development; Theunissen et al., 2014 Cell Stem Cell) are sufficient to described the acquisition of naive pluripotency.

    • Schemes in 5A/6A could indicate when transgenes were added

    For our chemical resetting experiments we used conventional hiPSCs (KiPS) with stable expression of KLF7 or an EMPTY vector (lines 126-127). We have also added this detail in the figure legends (line 291).

    • Fig 7: the claim regard mouse pluripotency is a little outside of the scope of this paper; would recommend de-emphasizing the claim .

    We will streamline the discussion and put less emphasis on murine PSCs.

    We thank the reviewer for the good suggestion that will be included in the revised manuscript.

    • Similarly, are there features outside the DBD that might suggest a unique activity (IDR, TAD,PTM)? It seems KLF7 generates iPSCs much less efficiently than KLF4. Given the high similarity between their DBDs I wonder why this is so.

    As above, this is an excellent point for discussion that will be added to revised manuscript.

    Reviewer #1 (Significance (Required)): Significance • General assessment: The strength of the study is that the authors provide a potentially new way for the naïve reset in humans. This could improve human stem cell and embryo models. A limitation is that evidence is solely based on molecular (not functional) profiling and the uniqueness of KLF7 versus other KLF's (first and foremost KLF4) was not established. • Advance: Findings on the human-specific role of KLF7 are novel and interesting especially the ability to facilitate the naïve reset. Yet, in the absence of a more systematic comparison with other methods (and KLF factors), the claim that KLF7 is essential for this feat is unconvincing. • Audience: It's of interest to basic researchers in the broader stem cell community and those interested in early embryo development. I work on cellular reprogramming, sequence-structure-function analysis of reprogramming factors and pluripotency.

    Reviewer #2 (Evidence, reproducibility and clarity (Required)): The naïve pluripotency is established in the inner cell mass (ICM) of blastocysts. After implantation, the naïve epiblast becomes primed for lineage specification. Pluripotent stem cells (PSCs) have been successfully derived from early embryos at different stages. In mice, stem cell derivations from ICM yield naïve ESCs. Primed PSCs derived from E5.5-7.5 epiblast are epiblast stem cells (EpiSCs). In humans, stem cell derivations from human embryos have yielded PSCs with features distinct from mouse ESCs and more like EpiSCs. Recently, naïve human PSCs have been directly isolated from pre-implantation epiblast or transformed from primed PSCs. Derivation of naïve hPSCs contributes to studying the molecular events of early lineage specification and accelerates the development of the generation of humanized organs in animal models from naïve hPSCs, opening an exciting avenue for regenerative medicine.

    In this manuscript, the authors found that OSK7M could enable the reprogramming of human primary somatic cells. KLF7 is highly expressed in naive PSCs and its forced expression in conventional hPSCs induces upregulation of naive markers and boosts the efficiency of chemical resetting to naive PSCs, suggesting that KLF7 is a general human pluripotency factor and an inducer of pluripotency. The new findings extend KLF7 function in naïve PSC generation and also provide references for the efficient generation of naive PSCs. The people who focus on studying pluripotency and early embryo development might be interested in and influenced by the findings. The data are in general convincing. However, there are some issues that need to be resolved and improved.

    Major comments:

    1. Line 90: The authors showed that colonies derived from OSKM and OSK7M cocktails could be readily propagated for at least 10 passages. How many passages can OSK7M-iPSCs maintain in vitro prolonged culture?

    And how about the pluripotency and developmental potential of OSK7M-iPSCs for a long-time culture? For example, pluripotency gene expression and teratoma formation.

    We culture OSK7M-iPSCs up to 10 passages without noticing any abnormalities in the morphologies and duplication rate. However, we could extend such cultures for 5-10 more passages (i.e. a total of 2 months from iPSC generation) and perform staining for pluripotency markers or molecular analyses (by qPCR) and EBs differentiation assay to assess their developmental potential.

    *In vivo *assays like teratoma formation are not allowed in Italy due to official regulations on animal testing.

    1. Overexpression of KLF7 promotes the derivation of naïve PSCs. Are they different from naïve PSCs derived only by chemical resetting? For example, the pluripotency, the in vitro or in vivo developmental potential, and the efficiency of human blastoid generation.

    A key feature of naive PSCs is the potential to differentiate towards the trophoblast lineage in addition to the 3 germ layers. We will perform in vitro differentiation and EB formation assay to gauge the effect of KLF7 on differentiation potential.

    However, establishing a human blastoid generation protocol would be beyond the scope of the current study.

    As the manuscript mentioned, KLF7 is a general human pluripotency factor and an inducer of pluripotency. How does KLF7 knock-out affect the biological characteristics of hESCs? And whether KLF17 KO affects the derivation of naïve PSCs?

    We agree that it would be informative to study the requirement of KLF7 for the maintenance of primed pluripotency and during resetting. We plan to do so either by knockdown or CRISPRi, depending on which technique allows efficient and controllable depletion of KLF7. It might be the case that a straight KO of KLF7 induces the collapse of primed PSCs, making resetting experiments not feasible.

    1. Can naïve PSCs be directly reprogrammed from somatic cells with OSK7M under the PXGL medium? If so, how is the efficiency?

    We believe that studying the role of KLF7 in the context of direct reprogramming of somatic cells to naive pluripotency would go beyond the scope of this manuscript, as it would require substantial work for optimisation and generation of reagents.

    Moreover, we think that both by over-expression and inhibition of KLF7 during resetting, we will be able to investigate its involvement in naive pluripotency acquisition.

    1. Figure 6d: The data showed that in PXGL medium, KiPS (EMPTY) contained about 66% of KLF17+ cells on day 7 and declined to 30% of KLF17+ cells on day 12. Why do KLF17+ cells (naïve PSCs) decline in PXGL medium? Cells overexpressing KLF7 contained about 62% of KLF17+ cells on day 7 and increased to 89% of KLF17+ cells on day 12. Whether KLF7 function at this stage?

    The reviewer raised an intriguing point, concerning the maintenance of naive markers during resetting. Chemical resetting seems to induce transiently >60% of KLF17+/OCT4+ positive cells by day 7, however only a fraction of these cells is stabilised until day 12 (30%). In the presence of KLF7 overexpression, we observed a similar induction at day 7, which is maintained, or increased, up to day 12.

    This would indicate that KLF7 is important for the maintenance of a population of naive cells, rather than only for their induction.

    We will add this important point to the discussion.

    1. Figure 6e: The authors showed transcriptome analysis of KiPS KLF7 cells compared to KiPS16 EMPTY cells in standard culture conditions and found that trophoblast markers were not significantly changed. How is the gene expression during primed to naive transition or TSC differentiation?

    We have already investigated this aspect, showing that at day 12 during primed to naive transition there is a strong induction of TSC markers, which is ablated by KLF7 expression (Fig. 5d). Quantitative immunostaining for GATA3 (TSC marker) confirmed this lack of activation in the presence of KLF7 (Fig. 6c).

    Minor comments:

    1. KLF7 is expressed in both primed and naive PSCs and when overexpressed in conventional PSCs, it enhances chemical resetting to naive PSCs. During primed to naïve transition, how does the KLF7 gene expression pattern change?

    This is a good suggestion, we will analyse the expression patter of KLF7 during resetting.

    1. Line 52: The reference should be added.

    Thank you, we will add the reference.

    1. Line 210-212: The reference should be added.

    Thank you, we will add the reference.

    Reviewer #2 (Significance (Required)): The people who focus on studying pluripotency and early embryo development might be interested in and influenced by the findings. The data are in general convincing. However, there are some issues that need to be resolved and improved.

    Reviewer #3 (Evidence, reproducibility and clarity (Required)): Summary: In this manuscript, the authors found that KLF7 is generally expressed in both prime and naïve human pluripotent stem cells. They showed that KLF7 could replace KLF4 to induce human iPS cells in the microfluidic reprogramming system. The authors then found that overexpression of KLF7 in human prime iPSCs can facilitate the generation of naïve iPS cells. They also showed that KLF7 is a repressor of trophoblast markers. Collectively, these findings indicated that KLF7 is a general pluripotency inducer for human iPS and naïve iPS induction.

    Major comments:

    1. In Figure 2, as the reprogramming efficiency of OSK7M is much lower than that of OSKM, the authors should provide an OSM control to show whether the cells can be reprogrammed without KLF4 and KLF7.

    We have performed the requested experiment (reprogramming with OSM only) as part of a manuscript in preparation. We observed an efficiency of reprogramming significantly lower than OSK7M, yet primary iPS colonies could be obtained.

    We believe that this is due to the expression of KLF4 and KLF7 in human fibroblasts, as shown in Figure 4a.

    1. It will be more convincing to perform a teratoma assay of OSK7M-iPSCs to demonstrate their multilineage differentiation potential.

    *In vivo *assays like teratoma formation cannot be performed in Italy due to official regulations on animal testing.

    However, we could extend such cultures for 5-10 more passages (i.e. a total of 2 months from iPSC generation) and perform staining for pluripotency markers or molecular analyses (by qPCR) and EBs differentiation assay to assess their multilineage differentiation potential.

    1. Since KLF7 is also expressed in primed human iPS cells, the authors should show the expression level of KLF7 in the established KLF7-iPSC and EMPTY-iPS.

    Good suggestion, we will add it to Figure 3b.

    Minor comments:

    The author claimed that KLF7 is a direct repressor of trophoblast markers, but the data in the manuscript cannot support this conclusion. The author can only claim that KLF7 can inhibit the expression of trophoblast markers.

    We agree with the reviewer, and we believe that there was a misunderstanding. On pages 8-9 line 182-190 we also concluded that KLF7 regulates naive pluripotency markers, rather than trophoblast markers. We will rephrase the text to make it clearer.

    Reviewer #3 (Significance (Required)): KLF family proteins such as KLF4 and KLF17 have been identified as pluripotent inducers. In this study, the authors demonstrated that KLF7 is a novel pluripotent inducer of human IPS and naïve iPS cells, providing new insights into the functions of KLF family proteins in human pluripotency induction.

  2. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    Summary:

    In this manuscript, the authors found that KLF7 is generally expressed in both prime and naïve human pluripotent stem cells. They showed that KLF7 could replace KLF4 to induce human iPS cells in the microfluidic reprogramming system. The authors then found that overexpression of KLF7 in human prime iPSCs can facilitate the generation of naïve iPS cells. They also showed that KLF7 is a repressor of trophoblast markers. Collectively, these findings indicated that KLF7 is a general pluripotency inducer for human iPS and naïve iPS induction.

    Major comments:

    1. In Figure 2, as the reprogramming efficiency of OSK7M is much lower than that of OSKM, the authors should provide an OSM control to show whether the cells can be reprogrammed without KLF4 and KLF7.
    2. It will be more convincing to perform a teratoma assay of OSK7M-iPSCs to demonstrate their multilineage differentiation potential.
    3. Since KLF7 is also expressed in primed human iPS cells, the authors should show the expression level of KLF7 in the established KLF7-iPSC and EMPTY-iPS.

    Minor comments:

    The author claimed that KLF7 is a direct repressor of trophoblast markers, but the data in the manuscript cannot support this conclusion. The author can only claim that KLF7 can inhibit the expression of trophoblast markers.

    Significance

    KLF family proteins such as KLF4 and KLF17 have been identified as pluripotent inducers. In this study, the authors demonstrated that KLF7 is a novel pluripotent inducer of human IPS and naïve iPS cells, providing new insights into the functions of KLF family proteins in human pluripotency induction.

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    The naïve pluripotency is established in the inner cell mass (ICM) of blastocysts. After implantation, the naïve epiblast becomes primed for lineage specification. Pluripotent stem cells (PSCs) have been successfully derived from early embryos at different stages. In mice, stem cell derivations from ICM yield naïve ESCs. Primed PSCs derived from E5.5-7.5 epiblast are epiblast stem cells (EpiSCs). In humans, stem cell derivations from human embryos have yielded PSCs with features distinct from mouse ESCs and more like EpiSCs. Recently, naïve human PSCs have been directly isolated from pre-implantation epiblast or transformed from primed PSCs. Derivation of naïve hPSCs contributes to studying the molecular events of early lineage specification and accelerates the development of the generation of humanized organs in animal models from naïve hPSCs, opening an exciting avenue for regenerative medicine.

    In this manuscript, the authors found that OSK7M could enable the reprogramming of human primary somatic cells. KLF7 is highly expressed in naive PSCs and its forced expression in conventional hPSCs induces upregulation of naive markers and boosts the efficiency of chemical resetting to naive PSCs, suggesting that KLF7 is a general human pluripotency factor and an inducer of pluripotency. The new findings extend KLF7 function in naïve PSC generation and also provide references for the efficient generation of naive PSCs. The people who focus on studying pluripotency and early embryo development might be interested in and influenced by the findings. The data are in general convincing. However, there are some issues that need to be resolved and improved.

    Major comments:

    1. Line 90: The authors showed that colonies derived from OSKM and OSK7M cocktails could be readily propagated for at least 10 passages. How many passages can OSK7M-iPSCs maintain in vitro prolonged culture? And how about the pluripotency and developmental potential of OSK7M-iPSCs for a long-time culture? For example, pluripotency gene expression and teratoma formation.
    2. Overexpression of KLF7 promotes the derivation of naïve PSCs. Are they different from naïve PSCs derived only by chemical resetting? For example, the pluripotency, the in vitro or in vivo developmental potential, and the efficiency of human blastoid generation. As the manuscript mentioned, KLF7 is a general human pluripotency factor and an inducer of pluripotency. How does KLF7 knock-out affect the biological characteristics of hESCs? And whether KLF17 KO affects the derivation of naïve PSCs?
    3. Can naïve PSCs be directly reprogrammed from somatic cells with OSK7M under the PXGL medium? If so, how is the efficiency?
    4. Figure 6d: The data showed that in PXGL medium, KiPS (EMPTY) contained about 66% of KLF17+ cells on day 7 and declined to 30% of KLF17+ cells on day 12. Why do KLF17+ cells (naïve PSCs) decline in PXGL medium? Cells overexpressing KLF7 contained about 62% of KLF17+ cells on day 7 and increased to 89% of KLF17+ cells on day 12. Whether KLF7 function at this stage?
    5. Figure 6e: The authors showed transcriptome analysis of KiPS KLF7 cells compared to KiPS16 EMPTY cells in standard culture conditions and found that trophoblast markers were not significantly changed. How is the gene expression during primed to naive transition or TSC differentiation?

    Minor comments:

    1. KLF7 is expressed in both primed and naive PSCs and when overexpressed in conventional PSCs, it enhances chemical resetting to naive PSCs. During primed to naïve transition, how does the KLF7 gene expression pattern change?
    2. Line 52: The reference should be added.
    3. Line 210-212: The reference should be added.

    Significance

    The people who focus on studying pluripotency and early embryo development might be interested in and influenced by the findings.

    The data are in general convincing. However, there are some issues that need to be resolved and improved.

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    The authors report that the human-specific KLF factor KLF7 can induce pluripotency in humans and can improve the reset toward naïve pluripotency when cells are cultured the PXGL medium. KLF7 falls behind KLF4 in reprogramming efficiency but might have a unique role in naïve reset (10-20 fold less efficient in iPSC colony yield). The topic of the study is interesting and adds important insights into the roles of KLF factors along the pluripotency continuum and pinpoints differences between mice and human. There are implications for stem cell engineering and boosting the developmental potency of stem cells (blastoid formation potential, interspecies chimera formation). However, some of the claims as to the unique role of KLF7 are unconvincing in the absence of comparison with other KLF factors, especially the Yamanaka factor KLF4. The flow and coherence of the text can be improved - at times reasoning and motivation of experiments are hard to follow

    Major comments

    • Why would a pan-pluripotency factor KLF7 which is expressed in both primed and naïve cells more potently trigger the naïve reset than the naïve specific factors KLF4/5/17? Such a comparison could widen the scope and interest of their work.. I would find it interesting if authors would compare the ability of key KLF factors to induce naivety. This is of particular interest as the overexpression of engineered Sox along with KLF4 was reported to improve the quality and developmental potential of PSC in multiple species (MacCarthy et al bioarxiv). Such an analysis could reveal unique features of KLF family members and lead to advanced stem cell models. They actually claim the SK naïve reset does not require naïve medium but the expression of SK alone is sufficient to induce this state. What do the authors think about this claim? Overall I feel the potential role of KLF7 in naïve reset is interesting but underdeveloped.

    Minor comments

    • P3, line 52: "Surprisingly, however, KLF4 is also routinely used to generate conventional human iPSCs." Why is this surprising? KLF4 (and SOX2) are the most potent iPSC factors whilst MYC and OCT4 can be omitted (at least in mouse).
    • It would be nice if the demonstration of pluripotency and quality of KLF7 iPSC go beyond transcriptome profiling and included some further assays common in the field.
    • Fig 1A-B: color coding (of dots) is very confusing- which ones are PSCs and which ones are iPSCs? Another colour palette might fix What is meant by "interrogating previously published data" (line 67)? Are these public RNA-seq data that were re-analyzed? I
    • Fig 2b: how were the colony numbers obtained? By morphology, or using live cell staining? So form of staining is recommended colony counting (i.e. TRA-1-60).
    • Fig 2e: Also, they say that "[t]hree technical replicates were carried out for all quantitative PCR". Unless I'm mistaken, it seems that only two technical replicates were performed for these qPCR reactions (two dots visible per bar).
    • Fig 3c: "colture"; change to "culture" (and the title: "bone fide" should be "bona fide")
    • For Fig 2/3: since the paper is on KLF4/7, I'm surprised that expression levels of OCT4 and SOX2 were analysed but not KLF4. Given that the main finding was that KLF4 was not upregulated in PSCs, I would be interested to see what the KLF4 levels are like in the iPSCs. RNA-seq analysis/qPCR would be best; but if the authors would like to use other methods, that's fine too.
    • Fig 4: The explanatory text is too sparse. Readers should be reminded of the differences between of naïve and primed PSCs and the known roles of KLF4 (this could also be improved in the introduction). List names of naïve media used on top of author names (5iLA, PXGL, EPSCM etc). Why was HENSM by Hanna excluded?
    • Fig 5: KLF7 is classified as a general pluripotency marker, but KLF4/KLF17 are classified as naïve markers. In that case, wouldn't it make more sense to overexpress a naïve specific marker in order to achieve naïve iPSCs at least as a control? What was the motivation here? I think the authors need to provide a more compelling reasoning why only KLF7 was studied or add more data for other KLFs (especially since it seems that the reprogramming efficiency of KLF4 is higher than that of KLF7 for conventional reprogramming (see Fig 2B)...)
      • Fig 5B: the text currently says that the cells on the left side of Fig 5B are from Day7; but it says the cells are from Day0 in the actual figure. Which one is it? Also, based on how the text is written, do the cells on the left also contain EOS, or are they the wild-type variety?
      • Fig 5c: not all markers in this figure are naïve markers (as stated in the text); would suggest separating the markers and labelling them accordingly AND rewriting the text to reflect that.
      • Life cell reporters for naivety (CD75,SUSD2) could enrich this study.
    • Schemes in 5A/6A could indicate when transgenes were added
    • Fig 7: the claim regard mouse pluripotency is a little outside of the scope of this paper; would recommend de-emphasizing the claim .
    • Could authors comment on the molecular features and whether there might be any non-redundant biochemical of KLF7 compared to other stemness-related KLFs? Looking at the conservation of the amino acids mediating base readout (-1,2,3,6) I expect specificity for DNA to be identical between KLF7 and KLF4 i.e. Figure S1A as reference for the C2H2numbering convention: https://www.cell.com/cms/10.1016/j.stemcr.2018.07.002/attachment/51171b7f-e644-4b0e-93c9-837632fd5d10/mmc1.pdf
    • Similarly, are there features outside the DBD that might suggest a unique activity (IDR, TAD,PTM)? It seems KLF7 generates iPSCs much less efficiently than KLF4. Given the high similarity between their DBDs I wonder why this is so.

    Significance

    • General assessment: The strength of the study is that the authors provide a potentially new way for the naïve reset in humans. This could improve human stem cell and embryo models. A limitation is that evidence is solely based on molecular (not functional) profiling and the uniqueness of KLF7 versus other KLF's (first and foremost KLF4) was not established.
    • Advance: Findings on the human-specific role of KLF7 are novel and interesting especially the ability to facilitate the naïve reset. Yet, in the absence of a more systematic comparison with other methods (and KLF factors), the claim that KLF7 is essential for this feat is unconvincing.
    • Audience: It's of interest to basic researchers in the broader stem cell community and those interested in early embryo development.

    I work on cellular reprogramming, sequence-structure-function analysis of reprogramming factors and pluripotency.

  5. Version published to 10.1101/2023.09.06.556189v1 on bioRxiv