A redox-active crosslinker reveals an essential and inhibitable oxidative folding network in the endoplasmic reticulum of malaria parasites

Article activity feed

1. Version published to 10.1371/journal.ppat.1009293

   Feb 3, 2021
2. 

   eLife

   Jul 21, 2020

   ##Author Response

   ###Reviewer #1:

   In this manuscript, Cobb and colleagues report on the biochemical and functional characterization of redox active ER proteins in the malaria parasite Plasmodium falciparum. They studied a protein called PfJ2, which contains HSP40 J and Trx domains and is homologous to human ERdj5. Using the TetR-PfDOZI aptamer system to tag PfJ2 and conditionally regulate its expression, they show that PfJ2 is localized in the parasite ER and is essential for parasite growth during the asexual blood stages. Using co-immunoprecipitation combined with mass spectrometry, they identify partner proteins of PfJ2 including other ER proteins such as PDI and BIP. Using a chemical biology approach based on DVSF crosslinker, they document the redox activity of PfJ2 and identify redox substrates of PfJ2, which include PDI8 and …

   ##Author Response

   ###Reviewer #1:

   In this manuscript, Cobb and colleagues report on the biochemical and functional characterization of redox active ER proteins in the malaria parasite Plasmodium falciparum. They studied a protein called PfJ2, which contains HSP40 J and Trx domains and is homologous to human ERdj5. Using the TetR-PfDOZI aptamer system to tag PfJ2 and conditionally regulate its expression, they show that PfJ2 is localized in the parasite ER and is essential for parasite growth during the asexual blood stages. Using co-immunoprecipitation combined with mass spectrometry, they identify partner proteins of PfJ2 including other ER proteins such as PDI and BIP. Using a chemical biology approach based on DVSF crosslinker, they document the redox activity of PfJ2 and identify redox substrates of PfJ2, which include PDI8 and PDI11 protein disulfide isomerases. They further functionally characterized PDI8 and PDI11 using the glmS ribozyme for conditional knockdown. These experiments confirm that PDI8 and PDI11 are partners of PfJ2 and show that knockdown of PDI8 impairs parasite blood stage growth. Finally, the authors show that inhibitors of human PDI inhibit parasite growth (at best in the micromolar range) and block the redox activity of PfJ2 and parasite PDI.

   This is an interesting study combining genetic and chemical biology approaches to investigate an understudied compartment of the malaria parasite. The manuscript is clearly written and the work technically sound. In summary, this study illustrates that ER redox proteins in the malaria parasite perform similar functions as in other organisms. The main limitation of this study is that evidence showing that redox ER parasite proteins are druggable is rather weak. PfJ2 is very similar to human ERdj5 in terms of active redox site and function, and the authors used inhibitors that are active on human PDI. It thus remains uncertain whether an antimalarial strategy targeting such conserved pathways is achievable.

   RESPONSE: We thank the reviewer for their appreciation of our work. While PfJ2 shares some similarity to human ERdJ5, we disagree that they are functionally similar. Our data show that, unlike ERdJ5, PfJ2 substrates are primarily other redox chaperones. In terms of the redox active site, our data clearly identifies a pathway that is targeted by a small molecule inhibitor. There is a lot of precedence for targeting conserved pathways as an antimalarial strategy. For example, anti-translational and anti-proteasomal inhibitors are being widely studied for their potency as antimalarials (Baragana et al 2015 Nature; Li et al 2016 Nature; Wong et al 2017 Nat. Microbiol.; Kirkman et al 2018 PNAS; Stokes et al 2019 PLoS Path.), several proteases (with conserved active sites) are well known antimalarial targets (Sleebs et al 2014 PLoS Biol.; Nasamu et al 2017 Science; Pino et al 2017 Science; Favuzza et al 2020 Cell Host Microbe), and effective inhibitors targeting a parasite chaperone has been repurposed for antimalarial drug development (Lu et al 2020 PNAS). We thank the reviewer for recognizing that there is a long road ahead of us to develop a more specific inhibitor for PfJ2, however, that is beyond the scope of this study.

   In addition, a number of specific points should be addressed to improve the quality of the manuscript:

   Although PDI8 and PDI11 gene edition were performed in the PfJ2apt line, the authors did not attempt to knockdown both PfJ2 and PDI8/11 simultaneously (because PfJ2 is essential). Therefore, referring to "double conditional mutants" is misleading.

   RESPONSE: We are open to alternative ways to refer to these mutants. Since we have orthogonal systems for knockdown of two proteins, we refer to these as double conditional mutants.

   The authors should provide details on the parasite lysis conditions used for the co-IP experiments to identify interacting proteins (Table 1) and redox partners (figure 3). In their proteomic analysis, the authors considered proteins with a 5-fold increase in the specific versus control conditions. A more stringent analysis would retain only proteins identified exclusively in the modified J2apt line.

   RESPONSE: We will include this in a new version. We agree that a more stringent analysis would lead to fewer proteins being identified, however, it also runs the risk of missing real interactors. We chose to use a 5-fold cutoff based on previously published work (Boucher et al 2018 PLoS Biol; Florentin et al 2020 PNAS).

   In figure 6, the authors should probe the blots for a control protein that is not co-immunoprecipitated with PfJ2 or PDI8. In Supplementary fig 4, control untreated parasites should be analyzed in parallel to GlcN-treated parasites.

   RESPONSE: We will do this once our labs reopen after the pandemic.

   The partial reduction of protein levels (Fig S4) shows that the glmS system is not very efficient here, which might explain why there is no phenotype in the PDI11 mutant (Fig5B). This questions the conclusion that PDI11 is dispensable.

   RESPONSE: We agree and we state that “These results...suggest that PfPDI11 may be dispensable... conclusions are supported by a genome-wide essentiality screen performed in P. falciparum” (Lines 319-322). We will add more discussion to explain this result.

   ###Reviewer #2:

   The claim here is of having discovered a druggable cellular process in P. falciparum, one that opens the door to therapeutic intervention in the most deadly form of malaria.

   The study commences with a focus on what appears to be the Pf homologue of a eukaryotic protein disulphide isomerase, known to many as ERdJ5 and referred to here as PfJ2. Its cellular contingents were identified by cross-linking and pull down, it’s (predicted) thiol reactivity explored with agents that react with reduced thiols and it’s functional importance to parasite fitness (in the lab) explored by gene knockout. These experiments provide evidence that PfJ2 and it’s associated Pf PDIs engage in thiol redox chemistry in the ER of the parasite and that integrity of this biochemical process is important to viability of the parasite.

   Lacking all expertise in molecular parasitology, this reader is unable to judge the specific significance of these findings to the field nor indeed the extent to which these are hard-won discoveries.

   RESPONSE: We are gratified to note that the reviewer is cognizant of their limitations and their ability to judge the significance of this work.

   However, from the perspective of the fundamentals of ER redox chemistry the findings represent a modest advance, showing that what is true of yeast and mammals is also true of Apicomplexa. The important mystery related to the juxtaposition of a J-domain and thioredoxin domains in PfJ2, remains.

   The most important claim however is the one with translational potential, namely that one might be able to discover (electrophilic) compounds that, despite the monotony of shared chemical features of thiol chemistry, will nonetheless possess sufficient specificity towards this or that malarial protein to be converted one day to a useful drug. However, in regards to this important point the authors offer very little in the way of evidence how and if this might be achieved.

   RESPONSE: We disagree. The work does not reconfirm the ‘fundamentals of ER redox’ chemistry. There is no work, in any system, that has shown that PfJ2-like proteins act as reductases for PDIs. In fact, as we state in the paper, in other model systems, there is a lot of redundancy built in the ER redox systems and PfJ2-like proteins work with specific clients like SERCA pumps or LDL receptor. Thioredoxin domain proteins in the ER of other eukaryotes have not been shown to work with each other or other chaperones. Furthermore, our data actually does suggest a reason why the J-domain is juxtaposed to thioredoxin domains. It recruits BiP to the mixed disulfides formed by PDIs. This insight would not have been possible in other systems because of the redundant redox mechanisms. In terms of the translational aspect, this work identifies an essential, pathway and a starting point for developing better inhibitors. As the reviewer may be aware, once a starting drug-like molecule has been identified, one has to embark on a medicinal chemistry program to develop more potent inhibitor. However, this is beyond the scope of this manuscript.

   Therefore, the main conclusions to draw from this paper are that ER-localised thiol chemistry is also important in malaria parasites and that, assuming one were able to explore localised context-specific features of thiol reactivity in malarial proteins, it may one day be possible to develop anti-malarial drugs that exploit this as a mechanism of action. The generic nature of these considerations limits the significance of the conclusions one might draw from this paper.

   RESPONSE: We are disappointed that we were unable to satisfy the reviewer’s need for ‘a giant leap for mankind’ insights.

   ###Reviewer #3:

   This paper describes redox-active proteins in the ER of malaria parasites. The authors start out with PfJ2, a J- and Trx-domain containing protein. They find that it is an essential ER protein that interacts with other chaperone and Trx domain proteins. Using a crosslinker with specificity for redox-active cysteines they identify PfPDI8 and PfPDI11 as redox-partners that together may aid folding of other proteins in the secretory pathway. Finally the authors use inhibitors that act on human PDIs and show that they inhibit parasite growth, albeit at rather high concentrations. This may be fortunate as this suggests different specificities for host and parasite PDIs. However, it also means that from this work it is difficult to judge if the parasite PDIs can be specifically targeted.

   RESPONSE: We thank the reviewer for recognizing the important insights gained from this work. We agree that the specific inhibitor identified is not an ideal antimalarial. There is a lot of precedence in the field for antimalarial inhibitors that target conserved mechanisms such as protein translation (Baragana et al 2015 Nature; Wong et al 2017 Nat. Microbiol.), aspartic proteases (Sleebs et al 2014 PLoS Biol.; Nasamu et al 2017 Science; Pino et al 2017 Science; Favuzza et al 2020 Cell Host Microbe), the proteasome (Li et al 2016 Nature; Kirkman et al 2018 PNAS; Stokes et al 2019 PLoS Path.), the TRiC chaperone complex (Lu et al 2020 PNAS) etc. We are starting a medicinal chemistry program to identify more potent inhibitors of these redox chaperones. However, that is beyond the scope of this paper.

   This is an interesting paper and rightly emphasises that it addresses a much understudied process and organelle in the parasite. The DVSF-crosslinking and the knockdown cell lines are highlights (although the knockdown cell lines were not fully exploited). The paper covers a lot of ground. However, this comes at the cost of depth. The actual function of the studied proteins on folding of other proteins and on the state of the ER was not evaluated and it is also not clear if the human PDI inhibitors indeed target the parasite enzymes. The high concentrations of inhibitors needed to show an effect on DVSF-crosslinking might indicate a secondary effect due to loss of parasite viability. As a result it is not fully clear if the studied proteins are indeed critical for folding of relevant substrates and if this process is druggable. More work is needed to support the main conclusions of the paper.

   RESPONSE: We thank the reviewer for appreciating the diverse toolsets used here to gain important insights into the ER of malaria-causing parasites. Due to the short time-frame of the DVSF-crosslinking experiment (30 mins vs 48h life cycle), we are able to conclude that the effect of the drug is not secondary. A new version will clarify this.

   Major points:

   1. The authors describe conditional knockdown lines and find that PfJ2 and PfPDI8 are essential but these lines are not further exploited for functional studies. Did the knock downs have any effect on proteins they mention as potential substrates (Table 1)? Did it affect the state/morphology of the ER? Did knock down of PfPDI8 remove/shift one of the PfJ2 bands after DVSF-crosslinking, as would be expected? Is there an effect on BiP? A general folding problem in the ER with such a lethal phenotype might have profound effects on the morphology of the organelles receiving protein from the ER. What happens to other cellular markers after knock down of these proteins? Were the knock down cells analysed by EM? Was there an effect on protein export? As it stands the knock down data does not show a role of the complex in the folding of any type of substrate and the function in oxidative folding, as indicated in the title, remains tentative.

   RESPONSE: The morphology of the ER is difficult to address due the fact that in these lifecycle stages the ER is quite condensed. Further, the ER is not clearly identifiable via EM. The knockdown of PDI8 is not complete, therefore, it is not possible to perform the suggested experiment as we will always see the residual PDI8 crosslinked with PfJ2. We are not sure what or if there’s any effect on BiP upon knockdown of PfJ2. BiP does not crosslink with PfJ2 and its expression levels do not change. We are not sure what other effect the reviewer expects on BiP. The co-IP data show that BiP is part of a complex with PfJ2 and PDI8, this complex has not been previously observed in the ER of any organism. Since the parasites die during the trophozoite/early schizont stages, several of these organelles such as Rhoptries, micronemes etc probably do not form. Once the lab reopens after the pandemic, we will test for the presence of these organelles via immunofluorescence microscopy as well as EM. Similar experiments could show an effect on protein export. However, since we didn’t identify any exported proteins to be putative substrates of PfJ2 (despite the expectation that chaperones are sticky and bind everything), and therefore, any effect we observe is likely to be indirect. Given the published data establishing the function of PDIs as oxidative folding chaperones, their high degree of conservation, and in vitro characterization, we conclude that they function in oxidative folding. Furthermore, we show that PfJ2 regulates the function of Plasmodium PDIs as well as recruits BiP to the mixed disulfide complex. BiP is a highly conserved chaperone that has clear function in protein folding. Based on this and the data presented here, we conclude that PfJ2 functions as a regulator of oxidative folding in P. falciparum.

   1. While I like the idea to use established commercial drugs as novel potential antimalarials, those used here are specific for non-infectious human diseases and target the host which is not a desirable property. Considering this, their rather low activity against the parasite can be taken as a positive result. However, the low activity is less convincing to establish the folding pathway in the parasite ER as a drug target. Beside the issue that it is unclear if indeed oxidative folding is the essential function of the PfJ2 complex (see previous point), the data in Fig. 7 does not clearly establish that this function is targeted by the inhibitors used. The effect is only seen at concentrations of 5xIC50. It is possible that this severely reduced viability which could be a non-specific reason for the lack of DVSF-crosslinked products. This needs to be examined in more depth. For instance, is the crosslink still seen after equivalent treatment of cultures with 5xIC50 of other unrelated drugs? Were other, unrelated processes unaffected? What was the effect of exposure to the drug on the ER and parasite morphology? Was the appropriate parasite stage affected? Can it be tested how fast exposure to 5xIC50 of the drug kills the parasites (at least morphologically, but preferably also by more specific means)?

   RESPONSE: We agree that the drugs identified here are not ideal antimalarials but rather they are starting molecules for a larger medicinal chemistry program, that is beyond the scope of this manuscript. While we see significant loss of DVSF crosslinking (for PfJ2) even at the IC50, the relationship between protein activity inhibition and parasite death isn’t always linear. We are testing analogs of 16F16 to identify more potent inhibitors of these proteins. We thank the reviewer for the suggested experiments, and when the pandemic is no long limiting access to the lab, we will perform some of these.

   1. While generally sound, a few experiments would have benefitted from more controls. A reducing sample from the same parasites for Fig. S7 (loaded a couple of lanes away to avoid interference of the reducing agent) would have been nice for comparison to show specificity of the higher molecular weight adducts. Detection of a control protein not expected to co-purify (for instance a cytosolic protein or a membrane-bound protein to control for residual parasite material) would have been appropriate for the co-immunoprecipitations (e.g. Fig. 6A,D, Fig. S9).

   RESPONSE: We show that there are no non-specific bands for PDI11, because when we mutate the cysteines, we do not observe any cross-linking. We will include the control proteins for the co-IPs, they were not included for the sake of clarity.

   Read the original source
3. 

   eLife

   Jul 21, 2020

   ###Reviewer #3:

   This paper describes redox-active proteins in the ER of malaria parasites. The authors start out with PfJ2, a J- and Trx-domain containing protein. They find that it is an essential ER protein that interacts with other chaperone and Trx domain proteins. Using a crosslinker with specificity for redox-active cysteines they identify PfPDI8 and PfPDI11 as redox-partners that together may aid folding of other proteins in the secretory pathway. Finally the authors use inhibitors that act on human PDIs and show that they inhibit parasite growth, albeit at rather high concentrations. This may be fortunate as this suggests different specificities for host and parasite PDIs. However, it also means that from this work it is difficult to judge if the parasite PDIs can be specifically targeted.

   This is an interesting paper and …

   ###Reviewer #3:

   This paper describes redox-active proteins in the ER of malaria parasites. The authors start out with PfJ2, a J- and Trx-domain containing protein. They find that it is an essential ER protein that interacts with other chaperone and Trx domain proteins. Using a crosslinker with specificity for redox-active cysteines they identify PfPDI8 and PfPDI11 as redox-partners that together may aid folding of other proteins in the secretory pathway. Finally the authors use inhibitors that act on human PDIs and show that they inhibit parasite growth, albeit at rather high concentrations. This may be fortunate as this suggests different specificities for host and parasite PDIs. However, it also means that from this work it is difficult to judge if the parasite PDIs can be specifically targeted.

   This is an interesting paper and rightly emphasises that it addresses a much understudied process and organelle in the parasite. The DVSF-crosslinking and the knockdown cell lines are highlights (although the knockdown cell lines were not fully exploited). The paper covers a lot of ground. However, this comes at the cost of depth. The actual function of the studied proteins on folding of other proteins and on the state of the ER was not evaluated and it is also not clear if the human PDI inhibitors indeed target the parasite enzymes. The high concentrations of inhibitors needed to show an effect on DVSF-crosslinking might indicate a secondary effect due to loss of parasite viability. As a result it is not fully clear if the studied proteins are indeed critical for folding of relevant substrates and if this process is druggable. More work is needed to support the main conclusions of the paper.

   Major points:

   1. The authors describe conditional knockdown lines and find that PfJ2 and PfPDI8 are essential but these lines are not further exploited for functional studies. Did the knock downs have any effect on proteins they mention as potential substrates (Table 1)? Did it affect the state/morphology of the ER? Did knock down of PfPDI8 remove/shift one of the PfJ2 bands after DVSF-crosslinking, as would be expected? Is there an effect on BiP? A general folding problem in the ER with such a lethal phenotype might have profound effects on the morphology of the organelles receiving protein from the ER. What happens to other cellular markers after knock down of these proteins? Were the knock down cells analysed by EM? Was there an effect on protein export? As it stands the knock down data does not show a role of the complex in the folding of any type of substrate and the function in oxidative folding, as indicated in the title, remains tentative.

   2. While I like the idea to use established commercial drugs as novel potential antimalarials, those used here are specific for non-infectious human diseases and target the host which is not a desirable property. Considering this, their rather low activity against the parasite can be taken as a positive result. However, the low activity is less convincing to establish the folding pathway in the parasite ER as a drug target. Beside the issue that it is unclear if indeed oxidative folding is the essential function of the PfJ2 complex (see previous point), the data in Fig. 7 does not clearly establish that this function is targeted by the inhibitors used. The effect is only seen at concentrations of 5xIC50. It is possible that this severely reduced viability which could be a non-specific reason for the lack of DVSF-crosslinked products. This needs to be examined in more depth. For instance, is the crosslink still seen after equivalent treatment of cultures with 5xIC50 of other unrelated drugs? Were other, unrelated processes unaffected? What was the effect of exposure to the drug on the ER and parasite morphology? Was the appropriate parasite stage affected? Can it be tested how fast exposure to 5xIC50 of the drug kills the parasites (at least morphologically, but preferably also by more specific means)?

   3. While generally sound, a few experiments would have benefitted from more controls. A reducing sample from the same parasites for Fig. S7 (loaded a couple of lanes away to avoid interference of the reducing agent) would have been nice for comparison to show specificity of the higher molecular weight adducts. Detection of a control protein not expected to co-purify (for instance a cytosolic protein or a membrane-bound protein to control for residual parasite material) would have been appropriate for the co-immunoprecipitations (e.g. Fig. 6A,D, Fig. S9).

   Read the original source
4. 

   eLife

   Jul 21, 2020

   ###Reviewer #2:

   The claim here is of having discovered a druggable cellular process in P. falciparum, one that opens the door to therapeutic intervention in the most deadly form of malaria.

   The study commences with a focus on what appears to be the Pf homologue of a eukaryotic protein disulphide isomerase, known to many as ERdJ5 and referred to here as PfJ2. Its cellular contingents were identified by cross-linking and pull down, it’s (predicted) thiol reactivity explored with agents that react with reduced thiols and it’s functional importance to parasite fitness (in the lab) explored by gene knockout. These experiments provide evidence that PfJ2 and it’s associated Pf PDIs engage in thiol redox chemistry in the ER of the parasite and that integrity of this biochemical process is important to viability of the parasite.

   Lacking all …

   ###Reviewer #2:

   The claim here is of having discovered a druggable cellular process in P. falciparum, one that opens the door to therapeutic intervention in the most deadly form of malaria.

   The study commences with a focus on what appears to be the Pf homologue of a eukaryotic protein disulphide isomerase, known to many as ERdJ5 and referred to here as PfJ2. Its cellular contingents were identified by cross-linking and pull down, it’s (predicted) thiol reactivity explored with agents that react with reduced thiols and it’s functional importance to parasite fitness (in the lab) explored by gene knockout. These experiments provide evidence that PfJ2 and it’s associated Pf PDIs engage in thiol redox chemistry in the ER of the parasite and that integrity of this biochemical process is important to viability of the parasite.

   Lacking all expertise in molecular parasitology, this reader is unable to judge the specific significance of these findings to the field nor indeed the extent to which these are hard-won discoveries. However, from the perspective of the fundamentals of ER redox chemistry the findings represent a modest advance, showing that what is true of yeast and mammals is also true of Apicomplexa. The important mystery related to the juxtaposition of a J-domain and thioredoxin domains in PfJ2, remains.

   The most important claim however is the one with translational potential, namely that one might be able to discover (electrophilic) compounds that, despite the monotony of shared chemical features of thiol chemistry, will nonetheless possess sufficient specificity towards this or that malarial protein to be converted one day to a useful drug. However, in regards to this important point the authors offer very little in the way of evidence how and if this might be achieved.

   Therefore, the main conclusions to draw from this paper are that ER-localised thiol chemistry is also important in malaria parasites and that, assuming one were able to explore localised context-specific features of thiol reactivity in malarial proteins, it may one day be possible to develop anti-malarial drugs that exploit this as a mechanism of action. The generic nature of these considerations limits the significance of the conclusions one might draw from this paper.

   Read the original source
5. 

   eLife

   Jul 21, 2020

   ###Reviewer #1:

   In this manuscript, Cobb and colleagues report on the biochemical and functional characterization of redox active ER proteins in the malaria parasite Plasmodium falciparum. They studied a protein called PfJ2, which contains HSP40 J and Trx domains and is homologous to human ERdj5. Using the TetR-PfDOZI aptamer system to tag PfJ2 and conditionally regulate its expression, they show that PfJ2 is localized in the parasite ER and is essential for parasite growth during the asexual blood stages. Using co-immunoprecipitation combined with mass spectrometry, they identify partner proteins of PfJ2 including other ER proteins such as PDI and BIP. Using a chemical biology approach based on DVSF crosslinker, they document the redox activity of PfJ2 and identify redox substrates of PfJ2, which include PDI8 and PDI11 protein …

   ###Reviewer #1:

   In this manuscript, Cobb and colleagues report on the biochemical and functional characterization of redox active ER proteins in the malaria parasite Plasmodium falciparum. They studied a protein called PfJ2, which contains HSP40 J and Trx domains and is homologous to human ERdj5. Using the TetR-PfDOZI aptamer system to tag PfJ2 and conditionally regulate its expression, they show that PfJ2 is localized in the parasite ER and is essential for parasite growth during the asexual blood stages. Using co-immunoprecipitation combined with mass spectrometry, they identify partner proteins of PfJ2 including other ER proteins such as PDI and BIP. Using a chemical biology approach based on DVSF crosslinker, they document the redox activity of PfJ2 and identify redox substrates of PfJ2, which include PDI8 and PDI11 protein disulfide isomerases. They further functionally characterized PDI8 and PDI11 using the glmS ribozyme for conditional knockdown. These experiments confirm that PDI8 and PDI11 are partners of PfJ2 and show that knockdown of PDI8 impairs parasite blood stage growth. Finally, the authors show that inhibitors of human PDI inhibit parasite growth (at best in the micromolar range) and block the redox activity of PfJ2 and parasite PDI.

   This is an interesting study combining genetic and chemical biology approaches to investigate an understudied compartment of the malaria parasite. The manuscript is clearly written and the work technically sound. In summary, this study illustrates that ER redox proteins in the malaria parasite perform similar functions as in other organisms. The main limitation of this study is that evidence showing that redox ER parasite proteins are druggable is rather weak. PfJ2 is very similar to human ERdj5 in terms of active redox site and function, and the authors used inhibitors that are active on human PDI. It thus remains uncertain whether an antimalarial strategy targeting such conserved pathways is achievable.

   In addition, a number of specific points should be addressed to improve the quality of the manuscript:

   Although PDI8 and PDI11 gene edition were performed in the PfJ2apt line, the authors did not attempt to knockdown both PfJ2 and PDI8/11 simultaneously (because PfJ2 is essential). Therefore, referring to "double conditional mutants" is misleading.

   The authors should provide details on the parasite lysis conditions used for the co-IP experiments to identify interacting proteins (Table 1) and redox partners (figure 3). In their proteomic analysis, the authors considered proteins with a 5-fold increase in the specific versus control conditions. A more stringent analysis would retain only proteins identified exclusively in the modified J2apt line.

   In figure 6, the authors should probe the blots for a control protein that is not co-immunoprecipitated with PfJ2 or PDI8.

   In Supplementary fig 4, control untreated parasites should be analyzed in parallel to GlcN-treated parasites.

   The partial reduction of protein levels (Fig S4) shows that the glmS system is not very efficient here, which might explain why there is no phenotype in the PDI11 mutant (Fig5B). This questions the conclusion that PDI11 is dispensable.

   Read the original source
6. 

   eLife

   Jul 21, 2020

   ##Preprint Review

   This preprint was reviewed using eLife’s Preprint Review service, which provides public peer reviews of manuscripts posted on bioRxiv for the benefit of the authors, readers, potential readers, and others interested in our assessment of the work. This review applies only to version 1 of the manuscript.

   Read the original source
7. 

   Version published to 10.1101/2020.05.13.093591 on bioRxiv

   May 15, 2020