HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Electronic Letters to:

Research: Rachel S. Curwen, Peter D. Ashton, Shobana Sundaralingam, and R. Alan Wilson Identification of Novel Proteases and Immunomodulators in the Secretions of Schistosome Cercariae That Facilitate Host Entry Mol Cell Proteomics 2006; 5: 835-844 [Abstract] [Full text] [PDF]

eLetters: Submit a response to this article

Electronic letters published:

Response to Dr. McKerrow's letter

Rachel C Curwen, Peter D. Ashton and R. Alan Wilson   (31 May 2006)

Schistosoma mansoni cercarial proteomics

James H McKerrow, Giselle M. Knudsen, Katalin F. Medzihradszky, Elizabeth J. Hansell   (30 May 2006)

Response to Dr. McKerrow's letter 31 May 2006
Rachel C Curwen University of York, Peter D. Ashton and R. Alan Wilson Send letter to journal: Re: Response to Dr. McKerrow's letter rsc105{at}york.ac.uk Rachel C Curwen, et al.	We thank Dr. McKerrow for his interest in our paper (Curwen et al. Mol Cell Proteomics. 2006 Feb 9 [Epub ahead of print]), and would like to respond to several serious issues raised by his letter. Firstly, we did not ignore the paper of Knudsen et al. (Mol Cell Proteomics. 2005 Dec; 4(12):1862-75), and referred to it in both our introduction and discussion. However, despite the similar subject matter we felt that the major differences both in our technical approaches and findings made meaningful, detailed comparison very difficult. In our opinion, the 2DE approach was the most suitable for the soluble, secreted proteins under investigation, enabling us to obtain both identities for the major components and their relative abundance within our preparation, whereas Knudsen et al. did not achieve relative quantification. In addition, the 2DE approach allowed us to perform a biosynthetic labelling study that was informative about the potentially separate origins of the main constituents. As Dr McKerrow concedes, it was indeed our contention that the “skin lipid in water” method of transformation was not biologically appropriate because somatic proteins were released due to the mortality engendered by the procedure. We believe our transformation method of tail-shearing by agitation, followed by a three-hour incubation in culture medium is less destructive, and therefore superior to the aforementioned technique, resulting in high viability of the schistosomula produced, as tested both in vitro and in vivo. Knudsen et al. used an analogous method of mechanical transformation, that appears to involve passing the parasites through a syringe needle of unspecified bore, potentially subjecting them to high shearing forces, before a two-hour incubation period in culture medium. We cannot agree that the results in Table 1 of Curwen et al. and Table III of Knudsen et al. are ‘remarkably similar’ given that of the 19 different proteins identified by Curwen et al. less than half were in common, and of the 85 proteins in Knudsen et al. only 10 were also identified in our study as a whole. The proteins in common were either Schistosoma mansoni elastase isoforms or abundant cytosolic constituents. All were previously known schistosome proteins. The presence of nuclear proteins (e.g. Histones H3 and H2B) and substantial numbers of muscle proteins (e.g. myosin heavy chain and paramyosin) in the Knudsen study, but not the Curwen study, implies that proteins are entering the culture medium by means other than holocrine secretion in the former, perhaps as a result of damage by shearing forces. We suggest that the reason Knudsen et al. failed to identify any novel proteins was due to their inadequate use of the available database resources. The approach they took was to search the full-length sequences in the NCBInr database using their MS/MS data. This would enable them to identify only those schistosome proteins whose full-length sequence was present in this database. They might also identify individual peptides from proteins of other species, where these are identical to the schistosome peptides. Only if such a cross-species match occurred, did they consult the >120,000 S. mansoni ESTs that were available at the time. Apparently, these were never directly searched with their MS/MS data. Furthermore, the available genome data (>3,000,000 reads, plus a publicly available preliminary assembly) were not taken into account at any stage. We reject Dr. McKerrow’s assertion that the ‘only unequivocally unique protein’ identified in our study was the metalloproteinase. On the basis of our bioinformatic approaches, we do not believe it is ‘tentatively assigned’ as a protease of the leishmanolysin family, given its BLAST score of 9.00 E-24 to Leishmania donovani leishmanolysin (GP63) and the presence of a high scoring metalloprotease domain containing a perfect active site motif. Dr. McKerrow intimates that this apparently abundant protease was absent from the findings of Knudsen et al. due to the one hour-longer culture period of Curwen et al. Is he seriously suggesting that this protein, clearly visible in the 2DE gel of cercarial homogenate (see Curwen et al. Supplementary material) is retained for the first two hours of transformation and then rapidly released over the next hour? It is absent from secretions collected in the culture period 3h to 3 days post transformation (unpublished data), making it unlikely to be the source of protease activity detected in schistosomula by previous investigators. Again, the more plausible explanation for its absence from the Knudsen study is their failure to search all the available sequence data. The presence of a protease other than elastase in the cercarial secretions is likely to upset current dogma on skin penetration by this parasite. We are equally confident in our identification of six other novel proteins, namely three containing venom allergen-like domains (SmSCP_a, b and c), a potassium channel blocker homologue (SmKK7), a dipeptyl peptidase (SmDPP IV) and a protein of unknown function. All of these are absent from the lists of Knudsen et al. Dr. McKerrow requests the sequence of the peptides for the three novel proteins where identification was on the basis of a single peptide match. We have duly supplied this information in the table below, and would like to emphasise that all identities in our manuscript were only accepted if they met very stringent statistical criteria. The sequence for the novel serpin was not supplied in the manuscript as it is publicly available, simply by typing ‘snapmodel:snap06756’ into the search box for the genome browser at SchistoDB.org, as described in Curwen et al. Spot Number Putative Identity Sequence ID Peptide Sequence Mascot Score Mascot Significance Threshold 14 SCP_a Sm09319 AIYNFHKK 57 37 36 SCP_b snap11344 FEEFSSVAQNIADSPTIEK 98 39 39 SmKK7 Sm12916 VSEHVDTYDEHK 67 41
Schistosoma mansoni cercarial proteomics 30 May 2006
James H McKerrow, Professor of Pathology UCSF Sandler Center for Basic Research in Parasitic diseases, Giselle M. Knudsen, Katalin F. Medzihradszky, Elizabeth J. Hansell Send letter to journal: Re: Schistosoma mansoni cercarial proteomics jmck{at}cgl.ucsf.edu James H McKerrow, et al.	This is a response to the recently published article by Curwen et al. (Mol Cell Proteomics. 2006 Feb 9 [Epub ahead of print]). The authors appear to have overlooked a very similar analysis in a previously published paper in Molecular and Cellular Proteomics (Knudsen et al. Mol Cell Proteomics. 2005 Dec; 4(12):1862-75. Epub 2005 Aug 18). In their discussion Curwen et al. do speculate that the differences they report in the level and complexity of secreted proteins by schistosome cercariae versus Knudsen et al. were likely due to differences in how the secretions were collected. They state “This contrasts with the alternative method involving exposure of cercaria to skin lipids in water that leads to a very high mortality rate with inevitable artefactual (sic) contribution of somatic proteins to the preparation. This may explain the marked differences in the findings reported here with those of the recent shotgun proteomics study by Knudsen et al.” They appear to have overlooked Table III in Knudsen et al. It presents analysis using a similar method of tail shearing and collection in media. In fact Knudsen et al. presented three proteomics analyses; the first being a “natural” induction of secretions using lipid (referred by Curwen et al.), the second non- induced, and the third the tail shearing method into media with specific collection of secreted acetabular gland vesicles (Table III). When the Curwen and Knudsen tail shearing analyses are compared, the results are remarkably similar. The major differences are fewer proteins identified by Curwen et al., probably due to the use of 2D gel electrophoresis rather than 1D SDS-PAGE fractionation followed by LC/mass spectrometry analysis of the proteome. The only unequivocally unique protein identified by Curwen et al. is a tentatively assigned metalloprotease, with 30% sequence identity conservation to leishmanolysin (GP63). Metalloprotease activity was identified by several investigators in schistosomula (Keene, WE et al. (1983) Lab Invest 49(2):201-7). Perhaps the additional three-hour incubation of schistosomula prior to proteomic analysis by Curwen et al. is the reason for the appearance of this protein. In any event, its location could be confirmed by future expression and immunolocalization studies. It would also be useful to have the sequences for single peptide IDs published as done in Knudsen et al. to allow independent verification of protein ID’s. In addition, the sequence of the “novel” serpin should be included.