This is a preview of subscription content, access via your institution
Access options
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
References
Kirchhofer, D. et al. Proc. Natl. Acad. Sci. USA 104, 5306–5311 (2007).
Bandeira, N., Pham, V., Pevzner, P., Arnott, D. & Lill, J.R. Nat. Biotechnol. 26, 1336–1338 (2008).
Edman, P. Arch. Biochem. 22, 475 (1949).
Suckau, D. & Resemann, A. Anal. Chem. 75, 5817–5824 (2003).
Geer, L.Y. et al. J. Proteome Res. 3, 958–964 (2004).
Henzel, W.J., Tropea, J. & Dupont, D. Anal. Biochem. 267, 148–160 (1999).
Nicolardi, S., Switzar, L., Deelder, A.M., Palmblad, M. & van der Burgt, Y.E.M. Anal. Chem. 87, 3429–3437 (2015).
Polevoda, B. & Sherman, F. J. Mol. Biol. 325, 595–622 (2003).
Nomura, D.K., Dix, M.M. & Cravatt, B.F. Nat. Rev. Cancer 10, 630–638 (2010).
Borodovsky, A. et al. Chem. Biol. 9, 1149–1159 (2002).
Henzel, W.J., Rodriguez, H. & Watanabe, C. J. Chromatogr. A 404, 41–52 (1987).
Acknowledgements
The authors thank S. Hymowitz and K. Dong (Department of Structural Biology, Genentech, Inc. South San Francisco, California) for providing the purified BAP1–ASXL1 protein complexes used in this study, as well as D. Kirkpatrick and G. Manning for helpful discussions.
Author information
Authors and Affiliations
Contributions
C.E.B. conceived of and implemented the ISDetect algorithm, analyzed data, and wrote the manuscript. Y.G. carried out MALDI-ISD and ISDetect experiments to generate the data used in the study. I.W. designed the DUB probes and analyzed DUB results. J.R.L. participated in the design of the study. W.S. conceived of the study and wrote the manuscript. All authors read, revised, and approved the final version of the manuscript.
Corresponding author
Ethics declarations
Competing interests
All authors are employees of Genentech, Inc., a biotech company.
Additional information
Editor's note: This article has been peer-reviewed.
Integrated supplementary information
Supplementary Figure 1 Distributions of ISDetect scores from scrambled spectra for false positive rate determination.
The overall distribution of ISDetect scores from random matches was estimated through 100 randomized reassignments of peak intensities within 250 actual MALDI-ISD spectra (see methods). Results were tallied either across all results for all trials (blue; n=20,061,341 out of 25,000 trials) or for only the best score reported per trial (red). Histograms (A,B) illustrate the frequencies of scores in the overall distributions; the cumulative fraction of results at or above the given scores (C,D). These values are equivalent to the frequency of occurrence of a given score due to random matching (i.e. a false positive rate). Panels E,F display the same data as C and D, respectively, on a logarithmic scale to better visualize higher score thresholds.
Supplementary Figure 2 Effects of MALDI matrix choice on spectral quality.
The intensities of ion peaks and their relative signal-to-noise vary depending on the MALDI matrix employed. Shown here are representative spectra from MALDI-ISD on a purified human growth hormone protein sample analyzed upon a MALDI TOF/TOF mass spectrometer (4800; AB Sciex, Foster City, CA). (A) 2,5-diaminonapthalene (DAN); (B) sinapinic acid (SA); (C) dihydrobenzoic acid (DHB); (D) hydroxycinaminic acid (CHCA).
Supplementary Figure 3 Venn diagram depicting success rates of the ISDetect algorithm across 1,787 distinct protein samples.
Matches include analyses where at least one candidate match achieved a score of ≥100 as defined in the text. Poor spectral quality indicates that no satisfactory in source decay spectrum could be obtained for the sample and the sample was not submitted to the algorithm.
Supplementary Figure 4 Differences in ISDetect performance as a function of purification tag type and location.
Differences in the ability of in source decay and ISDetect to successfully verify terminal information varies according not only to composition of the protein purification tag used but also to its location at the N- or C-terminus of the protein.
Supplementary Figure 5 Western blot assessment of binding of DUB probes to a panel of deubiquitinases.
Labeled lanes correspond to those depicted in Figure 2d; other lanes contain probes not a part of this study.
Supplementary Figure 6 Confirmation of N-terminal start residue using a T3 sequencing approach.
As matrix adduct ions can obscure lower-mass fragment ions, MALDI-ISD spectra (A; human growth hormone SOMA_HUMAN) are typically acquired at an m/z range from 900 to 5000 and the missing ions are extrapolated based upon the putative sequence match. Confirmation of this extrapolated match can be obtained through collision-induced fragmentation of c-type ions (e.g. 1189.8 m/z) from in source decay5, resulting in an MS/MS spectrum (B) containing b- and y- type ions assignable by a typical database searching algorithm.
Supplementary Figure 7 ISDetect results for purified interacting proteins.
ISDetect coupled with chromatography was applied to the structural characterization of the deubiquitinase BRCA1 Associated Protein 1 (Bap1), which suppresses tumorigenesis and is frequently mutated in malignancies. Bap1 interacts with polycomb repressive complex members Asxl1 and Asxl2 in vivo, and in efforts to obtain structural insights into the Bap1-Asxl complex, N-terminal His-tagged Bap1 was co-expressed with Asxl1 (fragment 1-365). Chromatographic fractionation followed by ISDetect was subsequently performed. By combining ISDetect results with intact mass determination, N-terminal truncations of Asxl1 were identified at multiple positions. (A) N-terminal ISDetect output of purified Asxl1 reveals the presence of multiple protein forms including an N-terminal truncation (blue) in addition to the full length protein (green). (B) ISDetect results for purified Bap1 reveal N-terminal acetylation of the full protein (blue), which prevented the application of Edman degradation to elucidate the protein sequence.
Supplementary information
Supplementary Figures and Tables
Supplementary Figures 1–7, Supplementary Tables 2 and 3 (PDF 1549 kb)
Supplementary Table 1
ISDetect results for approximately 3,000 analyses of 1,787 proteins. (XLSX 744 kb)
Supplementary Table 4
Comparison results between Mascot and ISDetect for a set of 250 randomly selected spectra. (XLS 141 kb)
Rights and permissions
About this article
Cite this article
Bakalarski, C., Gan, Y., Wertz, I. et al. Rapid, semi-automated protein terminal characterization using ISDetect. Nat Biotechnol 34, 811–813 (2016). https://doi.org/10.1038/nbt.3621
Published:
Issue Date:
DOI: https://doi.org/10.1038/nbt.3621
