APEX-based proximity labeling in Plasmodium identifies a membrane protein with dual functions during mosquito infection
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (Review Commons)
Abstract
Transmission of the malaria parasite Plasmodium to mosquitoes necessitates gamete egress from red blood cells to allow zygote formation and ookinete motility to enable penetration of the midgut epithelium. Both processes are dependent on the secretion of proteins from distinct sets of specialized vesicles. Inhibiting some of these proteins has shown potential for blocking parasite transmission to the mosquito. To identify new transmission blocking vaccine candidates, we defined the microneme content from ookinetes of the rodent model organism Plasmodium berghei using APEX2-mediated rapid proximity-dependent biotinylation. Besides known proteins of ookinete micronemes, this identified over 50 novel candidates and sharpened the list of a previous survey based on subcellular fractionation. Functional analysis of a first candidate uncovered a dual role for this membrane protein in male gametogenesis and ookinete midgut traversal. Mutation of a putative trafficking motif in the C-terminus led to its mis-localization in ookinetes and affected ookinete to oocyst transition but not gamete formation. This suggests the existence of distinct functional and transport requirements for Plasmodium proteins in different parasite stages.
Significance
The genome of the malaria parasite Plasmodium contains over 5500 genes, of which over 30% have no assigned function. Transmission of Plasmodium spp . to the mosquito contains several essential steps that can be inhibited by antibodies or chemical compounds. Yet few proteins involved in these processes are characterized, thus limiting our capacity to generate transmission interfering tools. Here, we establish a method to rapidly identify proteins in a specific compartment within the parasite that is essential for establishment of an infection within the mosquito, and identify over 50 novel candidate proteins. Functional analysis of the top candidate identifies a protein with two independent essential functions in subsequent steps along the Plasmodium life cycle within the mosquito.
Highlights
-
first use of APEX based proximity ligation in Apicomplexa
-
identification of >50 putative ookinete surface proteins
-
novel membrane protein essential for microgamete egress and ookinete migration
-
putative trafficking motif essential in ookinetes but not gametes
Article activity feed
-
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
The authors do not wish to provide a response at this time.
-
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
In this manuscript by Kehrer et al., use an elegant Apex2 BioID method to identify novel putative microneme proteins by mass-spectrometry and pick one candidate for further characterization. They identify a novel putative microneme protein they name Akratin which they characterize through targeted gene deletion and a series of complementation experiments. This reveals first that akratin appears to be functioning in male gamete egress, and though complementation using a putative trafficking mutant, also in midgut traversal.
Overall the study is thoroughly performed but some of the conclusions are not fully supported.
1)The newly …
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
In this manuscript by Kehrer et al., use an elegant Apex2 BioID method to identify novel putative microneme proteins by mass-spectrometry and pick one candidate for further characterization. They identify a novel putative microneme protein they name Akratin which they characterize through targeted gene deletion and a series of complementation experiments. This reveals first that akratin appears to be functioning in male gamete egress, and though complementation using a putative trafficking mutant, also in midgut traversal.
Overall the study is thoroughly performed but some of the conclusions are not fully supported.
1)The newly identified microneme protein is still putative in my mind as the authors have not co-localized it with another marker. This is crucial for conclusions about its putative function and crucial for the trafficking experiment as explained below. It is also important given the high number of putative false positives in the BioID experiment.
2)I would consider it essential to also localise the Apex2 tagged SOAP protein as the authors cannot be sure that there is a partial mislocalisation of the protein leading to false positives.
3)I am not convinced by the trafficking defect. This could be because the localsation in the images are not easy to distinguish and it may be much clearer looking down the microscope. I think co-localisation with another microneme marker would go a long way and demonstrating that akratin upon mutation actually localises elsewhere is important. It is even more important since there is no phenotype in male egress, but then later in ookinetes, which is a bit surprising if this is a proper conserved trafficking motif.
4)The candidate selection section is poorly described. A flow chart or clearer inclusion/ exclusion criteria would be useful.
5)I understand the approach to focus on more abundant biotinylated proteins, however, I think it may not be the best approach to use peptide counting. Apex2 labelling as the authors rightfully say, is mainly based on tyrosine labelling of surface exposed areas, so the abundance of proteins in the IP will depend on accessible tyrosines, protein abundance, distance from the bait, size of the protein and how many tryptic peptides can be generated. Reproducible results between 2 conditions are more likely to show true positives and may be the best way to prioritize, or assign confidence. Also: cOuld the authors use mean intensity values for the peptides covering proteins as a metric for abundance using label free quantification? This is not a requirement but may allow quantification in a slightly better way. I am not sure about the Table S1 colour scheme (the legend does not explain green, purple and blue shading). Are all green ones confirmed microneme proteins? Please add a proper descripton of the table and columns.
6)Figure 2C and D are from PlasmoDB and should ideally not be included as figure panels. This is misleading and could either be mentioned in the text, or put into supplementary data with a clear note that the authors have not aquired these data. I would also suggest to move figures 3A-C into figure 2 and present the KO with the complementation data for a direct comparison.
Minor:
1)When the authors say "numbers of peptides identified": is this unique peptides or does it include non-unique peptides?
2)Figures 1 I-K could move into supplementary as they are somewhat non-informative given the nature of BioID described in the main points.
3)Line 253: Whether akratin is involved in membrane lysis directly, or important for microneme secretion so this is a knock-on effect is not known. This could be added to the discussion, but there is no evidence for this statement in the results section.
4)Line 274: Refers to Figure 3F, which does not exist.
5)Line 333: Overall I think this is a bit of an overstatement. The use of Apex2 in these conditions is definitely nice to see but for now the authors have validated none of the microneme proteins by co-localization. So we are still a bit in the dark how well the method works in terms of false positives. The targeting motif in my mind is not yet confirmed in the absence of co-localisations with other markers. An alternative explanation could be that the c-terminus of the protein is important for its function in one stage, but not another but that trafficking is not- or only marginally affected.
Significance
The significance of the manuscript in my mind lies in the application of Apex2 in Plasmodium parasites, which will be an advance for the field. However, we do not learn about labeling times, how short it can be so its potential is not fully looked at.
The list of the putative micronemes will of course be of high interest for the community, but because of the limited validation in this study will require further validation by others.
The identification of the dual function of this protein in transmission in egress and ookinete traversal is interesting and surely leads to further studies. The identification of a putative differential trafficking motif is intruiging, if, as stated in the major concerns, this can be validated.
My expertise lies in Plasmodium biology with good knowledge of mass-spectrometry approaches.
Referees Cross-commenting
I agree with the assessment of the other reviewer, a slightly more detailed discussion of the hits would be desireable (exported proteins, why are they there). This could be a drawback of the system used, and mentioned.
Western blot of the GFP is a very good idea to clarify whether the localization is maybe, in parts, GFP that is not fused to the full lenght protein, either by cleavage, or a breakdown product.
-
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
Proximity labelling using BirA* has emerged as a highly successful approach to identify interaction partners and proteins in compartments of a protein of interest in the living cell. A number of recent studies applied this approach with malaria parasites and demonstrated its usefulness. However, a drawback of using BirA* is the time required to obtain good yields of biotinylated proteins (usually many hours). APEX is a much faster alternative to BirA* that however has so far not been used in apicomplexan parasites. Here Kehrer and colleagues use APEX2 to identify proteins of the micronemes of mature Ookinetes, a task that due to the …
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
Proximity labelling using BirA* has emerged as a highly successful approach to identify interaction partners and proteins in compartments of a protein of interest in the living cell. A number of recent studies applied this approach with malaria parasites and demonstrated its usefulness. However, a drawback of using BirA* is the time required to obtain good yields of biotinylated proteins (usually many hours). APEX is a much faster alternative to BirA* that however has so far not been used in apicomplexan parasites. Here Kehrer and colleagues use APEX2 to identify proteins of the micronemes of mature Ookinetes, a task that due to the short time available for labelling, would have been difficult with BirA*. From the obtained hits the authors chose a protein they named akratin and carried out a detailed functional analysis. They found that akratin is needed for microgametogenesis and for ookinete migration. Complementation with either the Pb or Pf GFP-tagged version of akratin rescued all defects of the mutant, despite the low sequence similarity and differing number of predicted transmembrane segments in the P. falciparum protein. Interestingly a mutant form of this protein with an ablated putative trafficking motif in its C-terminus also caused an ookinete migration phenotype but microgametogenesis was unaffected, hinting at intricacies in its function.
This is a clearly written and interesting paper reporting the first use of APEX-based BioID in malaria parasites and demonstrates that it is possible to take advantage of BioID in short lived stages. The ookinete microneme proteome reported here compares favourably with that derived from organelle purifications and will provide a resource to identify the proteins and understand their involvement in migration and adhesion of this and possibly other parasite stages. The example protein chosen for functional analysis in this work nicely illustrates this. A validation of some more of the unknown hits to be true microneme proteins would have been beneficial, but given the high number of known true positives in the hit list this is not absolutely essential for the paper. Overall this manuscript therefore provides a nice piece of work adding a list of proteins for future study and a new player important for mosquito infection.
Minor points:
1.several proteins taken as true hits in TableS1 are not obvious to me, such as plasmepsin V and several exported proteins. Are they expected to be trafficked via the micronemes in ookinetes?
2.Fig. 1C: '+ Streptavidin', this should be '+ streptavidin beads'
3please insert a referral to Figure S2 (similarity of the Plasmodium akratin homologs) somewhere around line 180, the only reference to this figure I could find in the text was in line 214. Figure S3 (line 184) is mentioned in a context to support the absence of homologues outside Plasmodium spp but shows the generation of the akratin KO. Maybe this should have been a citation to Figure S2, but then something is missing from that figure.
4.line 192, it might be useful to clarify in the text what 10 and 7 blood stage growth means (multiplication rate?).
5.the observation that akratin as a multi TM protein with signal peptide seems to be soluble in the parasite is rather unlikely. Is there a precedent for this? My best guess would be that this is GFP alone due to degradation or processing of the akratin-GFP fusion. A Western blot, if sufficient material is available, would clarify this. In regards to the localisation of akratin in the different stages it should also be taken into account that calling anything 'vesicular' based on fluorescent microscopy is rather speculative.
6.line 199, how can the low number of ookinetes affect their speed? Could this remark have been intended for the next sentence (199-201/Fig. 2I) as the lack of transmission to mosquitoes may have been due to the reduced number of ookinetes rather than a deficiency of individual ookinetes? To exclude the latter, were parasite number used matched to exclude that this was the case in Fig. 2I?
7.line 272-4, is there a way to quantify the phenotype in ookinetes, e.g. using intensity profile plots showing the distribution across the cell? Many cells still seem to show peripheral staining in the trafficking mutant. Could it again be that some of the protein is processed and the increased cytoplasmic staining represents GFP alone (see also comment 5)? In that case the level of processing rather than trafficking (alone) might be affected in the mutant.
8.lines 396 and 397, there is a minuscule -1 before the chemicals in my pdf.
9.line 58. remove 'not' and 'or': neither been identified nor been characterised
10.lines 88 to 94: English could be polished some more in this part. Line 91, replace 'ones', e.g. with 'proteins'
11.lines 245 and 253: add commas after (Figure 4B) and (Figure 4D), respectively
12.line 254; change possibly into possible
13.line 296-71: it might be helpful to mention in the text that this was again the complementation strategy (used for the Pb and Pf akratin)
14.while going beyond the scope of this manuscript, would it be possible to use the APEX introduced here to label structures in EM? A comment on this might be useful for readers interested in using this domain.
Significance
This is the first use of APEX2 in apicomplexans and this will be very useful for the field. The ookinete microneme proteome will provide a resource to study key aspects of this stage. The unknown proteins in the hit list still have to be confirmed to be true positives, but given the high number of true positives among the known proteins, the success rate should be acceptable. Akratin is a new protein among several known to be important for transmission to the mosquito host. As illustrated by this and other studies, many important questions remain how microgametes egress from the cell and how in vitro gliding and mosquito gut wall penetration relate. Akratin, together with the previously known proteins, will be instrumental in solving these questions.
My expertise (to put this assessment into perspective): I am a cell biologist working with P. falciparum blood stages and have no or limited experience with mosquito stages and P. berghei. My grasp of the technical difficulties to work with ookinetes and other mosquito stages as presented in this manuscript are therefore limited. Nevertheless, it appears to me that the work is well done and that the overall outcome from this paper is significant and will be of great interest to the field.
Referees Cross-Commenting
I think the other reviewer's request to carry out a co-loc experiment with a microneme marker to show that akratin and SOAP-APEX are in the correct location is very reasonable and should be done. I also share the view about the trafficking domain. This is reflected in my comment (points 5 and 7) on quantifying the phenotype of the 'trafficking mutant' and the possbility of not looking at the full length protein as it may be a processed/degraded version of the protein that still contains GFP. Western blots, if possible with these parasite stages, might clear this up.
-
