Pretaporter, a Drosophila protein serving as a ligand for Draper in the phagocytosis of apoptotic cells

Transcription

1 Manuscript EMBO Pretaporter, a Drosophila protein serving as a ligand for Draper in the phagocytosis of apoptotic cells Takayuki Kuraishi, Yukiko Nakagawa, Kaz Nagaosa, Yumi Hashimoto, Takashi Ishimoto, Takeshi Moki, Yu Fujita, Hiroshi Nakayama, Naoshi Dohmae, Akiko Shiratsuchi, Naoko Yamamoto, Koichi Ueda, Masamitsu Yamaguchi, Takeshi Awasaki, Yoshinobu Nakanishi Corresponding author: Yoshinobu Nakanishi, Kanazawa University Review timeline: Submission date: 02 June 2009 Editorial Decision: 08 July 2009 Revision received: 29 September 2009 Editorial Decision: 14 October 2009 Revision received: 15 October 2009 Accepted: 15 October 2009 Transaction Report: (Note: With the exception of the correction of typographical or spelling errors that could be a source of ambiguity, letters and reports are not edited. The original formatting of letters and referee reports may not be reflected in this compilation.) 1st Editorial Decision 08 July 2009 Thank you for submitting your manuscript for consideration by The EMBO Journal, two referees, whose reports I enclose below, have evaluated your manuscript. As you will see from their comments the referees express potential interest in the identification of Pretaporter as a ligand for Draper during phagocytosis of apoptotic cells and request some further experimental analysis to make the study suitable for publication in the EMBO Journal. Both referees find that overall the experiments are well performed, however, referee #2 would like to see some further genetic evidence for the pathway in which Pretaporter functions and also further evidence that Pretporter has a function in the dying cell and is itself not involved in apoptosis but is a novel "eat-me" signal. Should you be able to address these criticisms, we would be happy to consider a revised manuscript. I should remind you that it is EMBO Journal policy to allow a single round of revision only and that, therefore, acceptance or rejection of the manuscript will depend on the completeness of your responses included in the next, final version of the manuscript. When you submit a revised version to the EMBO Journal, please make sure you upload a letter of response to the referees' comments. Thank you for the opportunity to consider your work for publication. I look forward to your revision. Yours sincerely, European Molecular Biology Organization 1

2 Editor The EMBO Journal REFEREE COMMENTS Referee #1 (Remarks to the Author): The manuscript prepared by Kuraishi et al entitled "Pretaporter, a Drosophila protein serving as a ligand for Draper in the phagocytosis of apoptotic cells" reports the identification of Pretaporter, an ER-localized protein, as a ligand for Drosophila phagocytic receptor Draper. The authors further characterized the properties and function of Pretaporter in the engulfment of apoptotic cells using both in vitro and in vivo assays and made a series of novel findings. These findings include: Pretaporter resides in the ER in living cells yet is translocated to the outer surface of cells during apoptosis, the coating of Pretaporter on the surface of living cells or latex beads is sufficient for attracting Draper-mediated engulfment, null mutations of the gene encoding Pretaporter reduces the rate of apoptotic-cell engulfment in Drosophila embryos, and, Pretaporter acts in the same genetic pathway as Draper and Ced6 in controlling apoptotic-cell removal. These findings together support the model that Pretaporter is a novel "eat me" signal presented on the surface of apoptotic cells, which attracts and activates engulfing cells through binding to Draper. The clearance of apoptotic cells from animal bodies is a conserved biological process that plays important roles in tissue remodeling and the regulation of immune responses. The findings reported here make important contribution to the understanding of apoptotic-cell clearance, particularly in understanding how apoptotic cells are recognized. Researchers in the field of apoptosis and signal transduction will be highly interested in this manuscript. This manuscript is concise and well written. The experiments are well planned and thoroughly conducted with proper controls, and the results are convincing. The authors interpret their data carefully and the model presented is solid and well supported. I have only minor comments that should strengthen the manuscript. Minor comments: 1. Draper aa was expressed as a recombinant protein for affinity purification of Draper ligand. Is this part of Draper known to be sufficient for binding to apoptotic cells? 2. Figure 4C, images of higher magnification are needed for evaluating the degree of overlap between Pretaporter and apoptotic cells. 3. Previously, the authors group reported the identification of calreticulin, also an ER protein, as a marker for apoptotic-cell engulfment in Drosophila. It should be helpful to discuss the functional relationship between Pretaporter and calreticulin in apoptotic-cell signaling. Do these two ER proteins act in parallel to attract engulfing cells? Does Draper mediate calreticulin signaling as well? I would like to see these and related topics discussed. Referee #2 (Remarks to the Author): In the reviewed manuscript by Kuraishi et al they describe the identification and physiological relevance of a novel protein Pretaporter which they show is exposed on apoptotic cells and facilitates their clearance in the model organism Drosophila melanogaster. This work furthers our understanding of the molecular mechanisms that govern the removal of dying and effete cells during development and complement previous work identifying the molecule Draper as a phagocytic receptor in flies. The work is generally of a high quality and well controlled. It is also nicely written. I have a few issues that i think should be addressed prior to puplication. European Molecular Biology Organization 2

3 Major issue 1) In a set of nice experiments shown in figure 3 the authors conclude that pretaporter is genetically in the Draper pathway using both double mutants and a tranheterozygous startegy. However, they also conclude that it is not in the Ced2/5 pathway based on no observable increased phenotype in the prtp/mbc transheterozygotes. It would be nice if they could also perform those experiemnts using double null flies to show i) decreaed uptake in the the mbc and elmo null and ii) demonstrate that this is increased further when crossed with the prtp null as would be expected if they are in different pathways. 2) they state using conditional rescue of Prtp in hemocytes did not rescue the phenotype suggesting that this was therefore a factor required in the dying cell. However whether the protein has a role in the dying cell is not clear. As an example is the rate of apoptosis as measured by DNA fragmentation, PS exposure and morphological changes altered in the absence of Prtp (eg in S2 cells in which it has been silenced or in null embryos). I think it is important to demonstarte that it itself is not playing a role in the death process that might alter the "eatability" of the target cells. 3) I was left with little undrstanding of how this protein worked in the context of what is known about SIMU. Are they mutually exclusive? ie if Simu is present does Prtp no longer play a role in Draper mediated uptak and vs vs. Do they somehow compete for Draper? I think this could be addressed experiemntally or genetically and would significantly add to the paper. Otherwise it certainly should be better discussed in the paper Minor issue 1) On page 7 the authors comment in passing that degenrated axons was not different in the Prtp null flies. Perhaps i missed it but is Prtp exposed in this situation but simply redundant or does it not participate at all. This observation raises some interesting questions about tissue specificty that could be better expanded in the discussion 2) the cartoon showing the model of Prtp function has Simu but no molecule and does not place Prtp in the context of this alternative bridging molecule for Draper. As discussed above i think it would be good if this could be addressed and otherwise SIMU should be removed form the cartoon as it is meaningless as currently presented. 1st Revision - authors' response 29 September 2009 Point-by-point reply to referees' comments Referee #1 Comment 1 Draper aa was expressed as a recombinant protein for affinity purification of Draper ligand. Is this part of Draper known to be sufficient for binding to apoptotic cells? Yes, we have data regarding this issue: Draper-GST bound to the surface of S2 cells undergoing apoptosis (induced by the treatment with cycloheximide) but not to viable S2 cells. These results have been added to the manuscript as Supplementary Figure S1, and the text (page 5) has been changed accordingly. Comment 2 Figure 4C, images of higher magnification are needed for evaluating the degree of overlap between Pretaporter and apoptotic cells. European Molecular Biology Organization 3

4 We have included magnified views (as insets) of portions of the micrographs in the top row of Figure 4C. These views show that most Pretaporter-positive cells contain apoptotic nuclei. The figure and text (page 11) have been changed accordingly. Comment 3 Previously, the authors group reported the identification of calreticulin, also an ER protein, as a marker for apoptotic-cell engulfment in Drosophila. It should be helpful to discuss the functional relationship between Pretaporter and calreticulin in apoptotic-cell signaling. Do these two ER proteins act in parallel to attract engulfing cells? Does Draper mediate calreticulin signaling as well? I would like to see these and related topics discussed. In response to the reviewer's comment, we genetically examined the relationship between pretaporter and calreticulin in terms of the phagocytosis of apoptotic cells. The level of phagocytosis by embryonic hemocytes of a double mutant for pretaporter (prtp[δ1]) and calreticulin (Crc[S114307]) was about 70% of that by hemocytes of the pretaporter single mutant. Furthermore, embryos of a trans-heterozygote for the null alleles of pretaporter and calreticulin showed the level of phagocytosis equivalent to that observed with embryos of a heterozygote for each null allele. These results suggested that Pretaporter and Calreticulin act independently in the phagocytosis of apoptotic cells by embryonic hemocytes. We speculate that Calreticulin is recognized by a receptor other than Draper presumably present at the surface of hemocytes. These results have been incorporated into the revised manuscript as Supplementary Figure S9, and the text (pages 10-11) has been changed accordingly. Referee #2 Major issue Comment 1 In a set of nice experiments shown in figure 3 the authors conclude that pretaporter is genetically in the Draper pathway using both double mutants and a transheterozygous strategy. However, they also conclude that it is not in the Ced2/5 pathway based on no observable increased phenotype in the prtp/mbc transheterozygotes. It would be nice if they could also perform those experiments using double null flies to show i) decreased uptake in the mbc and elmo null and ii) demonstrate that this is increased further when crossed with the prtp null as would be expected if they are in different pathways. In the original manuscript, we took a transheterozygous strategy to examine roles of Rac1, Rac2, and Mbc because null mutants for both Rac1 and Rac2 (Nature 416:438) and for mbc (Genes Dev. 12:3337) reportedly have developmental defects in embryos. To confirm the defect caused by a lack of mbc expression, we morphologically examined embryos of the mbc null mutant compared to control flies. We found that the mutant embryos show abnormal structure in the midgut (referee-only supplementary Figure 1A). Furthermore, the number of embryonic hemocytes as identified by the expression of Croquemort was significantly smaller in the mutant than in the control (referee-only supplementary Figure 1B). These results indicated that embryos of the mbc null mutant indeed do not develop normally. Consequently, we were unable to analyze a double null mutant for mbc and draper or pretaporter. We would thus like to leave the data for mbc in the transheterozygous study as they are in the original manuscript. We have added in the revised manuscript the explanation why we adopted a transheterozygous strategy rather than analyzing null mutants (page 10). To conduct the experiment that the reviewer has suggested for elmo, we first analyzed the level of apoptotic cell clearance in embryos of the elmo null mutant. To our surprise, the analysis of cells dispersed from embryos showed no change of the level of European Molecular Biology Organization 4

5 phagocytosis between the elmo mutant and control flies. There was no difference in the numbers of Croquemort-expressing hemocytes and TUNEL-positive apoptotic cells. These results suggested that ELMO does not play an important role in the phagocytosis of apoptotic cells in embryos, and therefore, in the engulfment pathway activated by Pretaporter and Draper. The above described results have been incorporated into the revised manuscript as Figure 3D, and at the same time, we eliminated the data for elmo in the transheterozygous study in Figure 3C. The text has been changed accordingly (page 10). Micrographs of embryos of the elmo null mutant as well as the result of Western blotting for the examination of ELMO protein expression in the mutant have been submitted as referee-only supplementary Figure 2. Comment 2 They state using conditional rescue of Prtp in hemocytes did not rescue the phenotype suggesting that this was therefore a factor required in the dying cell. However whether the protein has a role in the dying cell is not clear. As an example is the rate of apoptosis as measured by DNA fragmentation, PS exposure and morphological changes altered in the absence of Prtp (eg in S2 cells in which it has been silenced or in null embryos). I think it is important to demonstrate that it itself is not playing a role in the death process that might alter the "eatability" of the target cells. In response to the reviewer's comment, we compared the timing and extent of apoptosis (in terms of caspase activation and DNA fragmentation) during embryogenesis between the pretaporter null mutants (prtp[δ1] and prtp[δ2]and control (prtp[g19]) flies. Apoptosis was not evident at stage 10 and began to be detectable at stage 11 in embryos of the control flies, as reported previously (Development 117:29), and this was the same for embryos of the mutant flies. These results indicate that Pretaporter does not play a role in the process of apoptosis itself, and suggest that the overexpression of pretaporter in the null mutant rescued a defect not in the process of apoptosis but in the "eatability" of apoptotic cells. The data have been incorporated into the revised manuscript as Supplementary Figure S4, and the text (pages 6-7) has been changed accordingly. Comment 3 I was left with little understanding of how this protein worked in the context of what is known about SIMU. Are they mutually exclusive? ie if Simu is present does Prtp no longer play a role in Draper mediated uptake and vs vs. Do they somehow compete for Draper? I think this could be addressed experimentally or genetically and would significantly add to the paper. Otherwise it certainly should be better discussed in the paper First of all, we have an opinion, as well as a result, different from that of authors in the paper describing Six-microns-under (Simu) (Cell 133:498). They claim that Draper is involved not in the engulfment but in the degradation of internalized apoptotic cells. However, our data (JBC 279:48466; this study) and also those of other investigators (Neuron 38:567; Cell 135:524; J Neurosci 29:4768) clearly show a role for Draper in the engulfment of apoptotic cells by hemocytes and glia. For this reason, we do not want to deeply argue the relationship between Pretaporter and Simu in this manuscript. In response to the reviewer's comment, we carried out an additional experiment, in which the phagocytosis of apoptotic cells by glia was examined in embryos of the pretaporter null mutant and control flies. We did so because glia express Simu (Cell 133:498). The results in the analysis of whole embryos and dispersed embryonic cells showed that the level of phagocytosis by Pretaporter-lacking glia was about two thirds of that by glia in control flies. This indicated that Pretaporter is required for apoptotic cells to be efficiently phagocytosed by Simu-expressing glia, and suggested that Pretaporter participates in the Draper-mediated phagocytosis of apoptotic cells irrespective of the presence of Simu in phagocytic cells. However, how Simu relates to Pretaporter, and also to Draper, in apoptotic cell clearance remains unsolved. The data have been incorporated into the revised manuscript as a part of Figure 2, and the text (pages 8 and 14) has been European Molecular Biology Organization 5

6 changed accordingly. Minor issue Comment 1 On page 7 the authors comment in passing that degenerated axons was not different in the Prtp null flies. Perhaps i missed it but is Prtp exposed in this situation but simply redundant or does it not participate at all. This observation raises some interesting questions about tissue specificity that could be better expanded in the discussion Although we have not examined the exposure of Pretaporter at the surface of degenerated axons (from a technical difficulty), we would expect that it is not the case. Because, the pruning of degenerated axons occurs independently of the action of caspases (Neuron 50:855) while Pretaporter exposure is restricted to cells undergoing apoptosis (this study). We speculate that a molecule other than Pretaporter serves as a ligand for Draper in the phagocytic removal of degenerated axons. This issue has been discussed in the revised manuscript (page 15). Comment 2 The cartoon showing the model of Prtp function has Simu but no molecule and does not place Prtp in the context of this alternative bridging molecule for Draper. As discussed above i think it would be good if this could be addressed and otherwise SIMU should be removed form the cartoon as it is meaningless as currently presented. As we responded to Major Comment 3, we obtained data suggesting that Pretaporter and Six-microns-under function independently. We would like to leave Six-microns-under included in Figure 7 because the relationship between these two proteins still remains to be solved. 2nd Editorial Decision 14 October 2009 Your manuscript has been reviewed once more by one of the original referees. Pending satisfactory minor revision, this referee recommends publication of your manuscript in the EMBO Journal. The referee finds that the model is a little speculative at this stage and it should be removed from the manuscript. When you send us your revision, please include a cover letter with an itemised list of all changes made, or your rebuttal, in response to comments from review. Thank you for the opportunity to consider your work for publication. I look forward to reading the revised manuscript. Yours sincerely, Editor The EMBO Journal Referee #2 (Remarks to the Author): the authors have addressed most of my concerns. however, i do not think the cartoon in figure 7 should be included. they present conclusive evidence for pretoporter, draper, ced6 and rac being in the same pathway. however in their own words "Further investigation is necessary before we obtain European Molecular Biology Organization 6

7 a complete picture of the Prtp/Drpr-initiated signaling pathway for the phagocytosis of apoptotic cells." it is in my mind premature to include this model and as some aspects of it are conjecture it does not add to the manuscript and should be removed 2nd Revision - authors' response 15 October 2009 Authors' reply to referee's comments Referee #2 Comment The authors have addressed most of my concerns. However, I do not think the cartoon in figure 7 should be included. They present conclusive evidence for pretoporter, draper, ced6 and rac being in the same pathway. However in their own words "Further investigation is necessary before we obtain a complete picture of the Prtp/Drpr-initiated signaling pathway for the phagocytosis of apoptotic cells." It is in my mind premature to include this model and as some aspects of it are conjecture it does not add to the manuscript and should be removed. We agree with the referee. We have removed the figure from the manuscript and modified the relevant description in Discussion (page 14). European Molecular Biology Organization 7