Poster #820 from ASMS 2000 - San Diego CA page last updated July15 2000
Protein Identification by Automated Nanospray Mass Spectrometry - ZoomScan Walking
Matt J. Sweeney1 and Christoph W. Turck2
1-Matt Sweeney-Mass Spec Consulting, San Jose, CA;
mattsweeney@earthlink.net2-UC San Francisco, Howard Hughes Medical Institute, San Francisco, CA 94143
view entire data set here or here
Introduction
Nanospray Mass Spectrometry is an important tool for high sensitivity protein identification. A disadvantage of nanospray infusion is that the operator must search for peptide ions from a protein digest for subsequent collision-induced fragmentation. In order to make the nanospray method less operator-dependent, we have implemented an automated procedure using common, publicly available tools. Protein digests are subjected to unattended nanospray analysis using ZoomScan Walking. The ID of eight proteins from a SDS gel using Automated ZoomScan Walking is shown as an example of the utility of this method. These SDS gel isolates were purified by affinity chromatography prior to being applied to the SDS gel.
A Cautionary note to those attempting to replicate this work: Not all versions of the LCQ software have the "gizmos" feature which allows the use of parent ion-mapping to easily obtain all the MS/MS spectra desired in a fully automated way. We were using the developers package available from Finnigan/Thermo. Once we upgraded to the latest release, the "gizmos" were no longer available. Finnigan states that the next release will re-add this back into the software. Until then, it is possible to use Dynamic Exclusion to simulate the ZoomScan Walking. The dynamic exclusion feature has a 25 element mass list. This then limits the number of parent ions that might be investigated. One option is to build a set of methods each of which only covers say 50-100 amu, each of which has the dynamic exclusion feature enabled.
Methods
Proteins binding to an affinity column were eluted with a linear NaCl gradient and subjected to SDS-PAGE. After staining the gel with Coomassie Brilliant Blue, individual protein bands were cut out and subjected to in-gel digests with Endoproteinase Lys-C. The resultant peptide mixtures were analyzed by nanospray mass spectrometry using automated ZoomScan walking data-dependent scans. A Protana nanospray source is used with an LCQ ion-trap mass spectrometer. An LCQ experiment method allows a large mass range to be automatically scanned and MS/MS and zoom scan data to be obtained automatically. Spectra are converted to dtas, and searched against a NR database using SEQUEST.
The original LCQ automated "Zoomscan Walking" method utilized the parent ion mapping feature of the "gizmos" in the tune view. The method forces the LCQ to scan from the low to the high mass in units of the parent scan window (we utilized 7). We then trigger a ZoomScan if there is anything above the very low threshold we choose (e.g. 4000 counts). The ZoomScan window is 10 amu. Then two or three MS/MS scans are triggered on the most intense ions in these scans. This generates large data sets. A single microliter can generate a 3-4 Mbyte file. The tune is saved with the parent ion mapping feature of the 'gizmos' enabled. This, in combination with a data-dependent triple play function, results in the "ZoomScan Walking" method. The instrument starts at the low mass parent ion mapping settings and proceeds up to high mass in 7 amu increments. In each window, the two or three largest ions used for MS/MS depending upon the number of scan events in the segment. Sample size, signal intensity, and the mass range to be covered, are all factors that play in to the selection of the parameters. We also implemented a ‘dynamic exclusion zoomscan walking’ procedure. This method uses a standard triple play modified by the dynamic exclusion feature. Dynamic exclusion builds a table of masses for which MS/MS spectra have been obtained. The process involves; scanning the full mass range, finding the largest intensity ion which has not been included on the list and doing a ZoomScan and MS/MS on it, adding the parent to the exclusion mass list, and then repeating the cycle. The masses of common contaminants; e.g. trypsin autolysis fragments and plasticizers can be added to a static table of exclusion masses. This method is robust with respect to the commonly observed variations in relative ion intensities seen in spray ionization methods. The repetitive cycling of the dependent scanning combined with dynamic exclusion means that eventually any ion of reasonable intensity will have MS/MS spectra recorded on it. At the present time, the exclusion mass list length is 25 elements. When this list limit is reached, the oldest are rolled off the bottom of the exclusion mass list. The functionality and coverage of the ‘gizmos’ style method can be replicated by running a set of analyses, each covering a limited mass-range to avoid the constraint imposed by the 25 element limit in the exclusion mass list. In both cases we utilized high numbers of microscans(9-16) and long maximum inject times(1000ms) for the MS/MS and ZoomScans. This can result in long scan times, which have little impact on our non-chromatographic system.
Image of SDS Gel of Immunoaffinity Column Fractions
Proteins Identified in Bands by Automated Zoom-Scan Walking
A>Talin
B>Clathrin Heavy Chain
C>Heterogeneous Nuclear Ribonucleoprotein U
D>Ewing Sarcoma Breakpoint Region 1 RNA-Binding Protein
E>Keratin Type II Cytoskeletal
F>Fus-Like Protein
G>Probable RNA-Dependent Helicase P68
H>HSPC117
Example Automated ZoomScan of Band A
NOTE: The data above illustrates a worst case situation. The largest ion in the 7 amu full-scan is the lowest mass isotope of the 803.4 ion from the digest. This is clearly seen in the ZoomScan. This is why we chose the 7amu full scan window and the 10amu zoom scan window. The resulting MS/MS spectra, although produced by an ion of a relatively smaller height, still yields a good MS/MS scan. This scan is identified by SEQUEST, due in part to the 3-4 amu parent ion mass error that we allow during the SEQUEST search.
SEQUEST Result of Scan #131 of Band A Sample (the above data)
Band A.0128.0131.2.out
SEQUEST v.C1, Copyright 1993-97
Molecular Biotechnology, Univ. of Washington, J.Eng/J.Yates
Licensed to Finnigan Corp., A Division of ThermoQuest Corp.
05/30/00, 03:50 PM, 1 min. 2 sec. on ucsf.edu
mass=1607.8(+2), fragment_tol=0.00, mass_tol=3.00, AVG
# amino acids = 151938261, # proteins = 483739, # matched peptides = 31284
immonium (HFYWM) = (00000), total_inten=3569.1, lowest_Sp=65.3
ion series nA nB nY ABCDVWXYZ: 0 1 1 0.0 1.0 0.0 0.0 0.0 0.0 0.0 1.0 0.0
rho=0.200, beta=0.075, top 10, /usr/users/finnigan/database/nr
Enzyme:Trypsin_K
# Rank/Sp (M+H)+ Cn deltCn C*10^4 Sp Ions Reference Peptide
-- ------ ------ ------ ------ ------ ----- ----- ---------------
1. 1 / 1 1605.7 1.0000 0.0000 4.3901 552.1 20/30 gi|6755809|ref|+1(K)NGNLPEFGDAIATASK
2. 2 / 3 1605.9 0.5943 0.4057 2.6089 221.1 13/26 gi|3098148|gb|A (K)SQAIIIKNIDFSQK
3. 2 / 3 1605.9 0.5943 0.4057 2.6089 221.1 13/26 gi|3098150|gb|A (K)SQAILIKNIDFSQK
4. 3 / 39 1605.9 0.5393 0.4607 2.3678 131.3 10/26 gi|1827477|dbj| (K)TKGLCRNRVTDDVK
5. 4 /378 1606.9 0.5135 0.4865 2.2545 71.6 8/30 gi|4928048|gb|A (K)ASQLRAKSGLTGSCVK
6. 5 / 19 1606.9 0.4928 0.5072 2.1634 152.9 11/28 gi|6226324|sp|Q (K)SSIINTICNNKSLAK
7. 6 /323 1606.9 0.4780 0.5220 2.0983 76.0 10/24 gi|2194138|gb|A (K)ERPLMIQVYESLK
8. 7 /116 1606.9 0.4775 0.5225 2.0962 102.4 9/26 gi|2983136|gb|A (K)DLGIKVLNSYENLK
9. 8 / 31 1605.8 0.4739 0.5261 2.0806 136.7 11/26 gi|7291808|gb|A (K)TPEALNEFRLLSSK
10.9 /200 1607.9 0.4706 0.5294 2.0658 88.1 8/26 gi|2496446|sp|Q (K)SSLINLLINKNHLK
1. gi|6755809|ref|NP_035732.1|| talin_gi|135290|sp|P26039|TALI_MOUSE TALIN_gi|111127
|pir||S11661 talin mouse_gi|54258|emb|CAA39588.1| (X56123) talin [Mus musculus]
gi|227256|prf||1617167A talin [Mus musculus]
2. gi|3098148|gb|AAC15522.1| (AF023777) small ribosomal protein 4 [Encalypta rhaptocarpa]
3. gi|3098150|gb|AAC15523.1| (AF023778) small ribosomal protein 4 [Bryobrittonia longipes]
4. gi|1827477|dbj|BAA09445.1| (D50833) stem cell factor [Felis catus]
5. gi|4928048|gb|AAD33394.1|AF128887_1 (AF128887) tailspike protein;
endorhamnosidase [Bacteriophage sf6]
6. gi|6226324|sp|Q9ZE46|Y102_RICPR HYPOTHETICAL GTP-BINDING PROTEIN RP102_gi|
3860670|em b|CAA14571.1| (AJ235270) unknown [Rickettsia prowazekii]
7. gi|2194138|gb|AAB61113.1| (AC002062) Similar to Arabidopsis receptor-like
protein kinase precursor (gb|M84659). [Arabidopsis thaliana]
8. gi|2983136|gb|AAC06739.1| (AE000692) 6-phosphogluconate dehydrogenase [Aquifex aeolicus]
9. gi|7291808|gb|AAF47228.1| (AE003464) CG11414 gene product [Drosophila melanogaster]
10. gi|2496446|sp|Q49435|Y442_MYCGE HYPOTHETICAL PROTEIN MG442_gi|1361600|
pir||H64248 hypothetical protein homolog MG442 - Mycoplasma genitalium
(SGC3)_gi|1046160|gb|AAB016 32.1| (U39731) hypothetical protein (GB:U00021_5)
[Mycoplasma genitalium]_gi|3845035|gb|AAC72462.1| (U39726) conserved
hypothetical protein [Mycoplasma genitalium]
Peptides Identified by SEQUEST in Band "A" digest By Automated ZoomScan Walking
Position Sequence
--------------------------------------------------
2351-2361 SIAAATSALVK
325- 334 LVPRLLGITK
1321-1332 ALSTDPASPNLK
2532-2541 FLPSELRDEH
2044-2063 LAQAAQSSVATITRLADVVK
2362-2375 AASAAQRELVAQGK
1767-1780 TLAESALQLLYTAK
1026-1040 NLGTALAELRTAAQK
1416-1431 NGNLPEFGDAIATASK
2477-2491 AAAFEDQENETVVVK
923- 943 QAAASATQTIAAAQHAASAPK
2044-2063 LAQAAQSSVATITRLADVVK
593- 613 LLAALLEDEGGNGRPLLQAAK
2064-2085 LGAASLGAEDPETQVVLINAVK
307- 316 TYGVSFFLVK
1321-1332 ALSTDPASPNLK
2362-2375 AASAAQRELVAQGK>Band “A” SEQUEST Result- Talin
>gi|6755809|ref|NP_035732.1|| talin_gi|135290|sp| P26039|TALI_MOUSE TALIN_gi|111127|pir||S11661 talin - mouse_gi|54258|emb|CAA39588.1| (X56123) talin [Mus musculus] >averagemass = 269,831 MVALSLKISIGNVVKTMQFEPSTMVYDACRMIRERIPEALAGPPNDFGLFLSDDDPKKGIWLEAG KALDYYMLRNGDTMEYRKKQRPLKIRMLDGTVKTIMVDDSKTVTDMLMTICARIGITNHDEYSLV RELMEEKKDEGTGTLRKDKTLLRDEKKMEKLKQKLHTDDELNWLDHGRTLREQGVEEHETLLLRR KFFYSDQNVDSRDPVQLNLLYVQARDDILNGSHPVSFDKACEFAGFQCQIQFGPHNEQKHKAGFL DLKDFLPKEYVKQKGERKIFQAHKNCGQMSEIEAKVRYVKLARSLKTYGVSFFLVKEKMKGKNKL VPRLLGITKECVMRVDEKTKEVIQEWSLTNIKRWAASPKSFTLDFGDYQDGYYSVQTTEGEQIAQ LIAGYIDIILKKKKSKDHFGLEGDEESTMLEDSVSPKKSTVLQQQYNRVGKVEHGSVALPAIMRS GASGPENFQVGSMPPAQQQITSGQMHRGHMPPLTSAQQALTGTINSSMQAVQAAQATLDDFETLP PLGQDAASKAWRKNKMDESKHEIHSQVDAITAGTASVVNLTAGDPAETDYTAVGCAVTTISSNLT EMSRGVKLLAALLEDEGGNGRPLLQAAKGLAGAVSELLRSAQPASAEPRQNLLQAAGNVGQASGE LLQQIGESDTDPHFQDVLMQLANAVASAAAALVLKAKSVAQRTEDSGLQTQVIAAATQCALSTSQ LVACTKVVAPTISSPVCQEQLVEAGRLVAKAVEGCVSASQAATEDGQLLRGVGAAATAVTQALNE LLQHVKAHATGAGPAGRYDQATDTILTVTENIFSSMGDAGEMVRQARILAQATSDLVNAIKADAE GESDLENSRKLLSAAKILADATAKMVEAAKGAAAHPDSEEQQQRLREAAEGLRMATNAAAQNAIK KKLVQRLEHAAKQAAASATQTIAAAQHAASAPKASAGPQPLLVQSCKAVAEQIPLLVQGVRGSQA QPDSPSAQLALIAASQSFLQPGGKMVAAAKASVPTIQDQASAMQLSQCAKNLGTALAELRTAAQK AQEACGPLEMDSALSVVQNLEKDLQEIKAAARDGKLKPLPGETMEKCTQDLGNSTKAVSSAIAKL LGEIAQGNENYAGIAARDVAGGLRSLAQAARGVAALTSDPAVQAIVLDTASDVLDKASSLIEEAK KASGHPGDPESQQRLAQVAKAVTQALNRCVSCLPGQRDVDNALRAVGDASKRLLSDLLPPSTGTF QEAQSRLNEAAAGLNQAATELVQASRGTPQDLARASGRFGQDFSTFLEAGVEMAGQAPSQEDRAQ VVSNLKGISMSSSKLLLAAKALSTDPASPNLKSQLAAAARAVTDSINQLITMCTQQAPGQKECDN ALRQLETVRELLENPVQPINDMSYFGCLDSVMENSKVLGEAMTGISQNAKNGNLPEFGDAIATAS KALCGFTEAAAQAAYLVGVSDPNSQAGQQGLVEPTQFARANQAIQMACQSLGEPGCTQAQVLSAA TIVAKHTSALCNSCRLASARTANPTAKRQFVQSAKEVANSTANLVKTIKALDGDFTEENRAQCRA ATAPLLEAVDNLSAFASNPEFSSVPAQISPEGRAAMEPIVISAKTMLESAGGLIQTARALAVNPR DPPRWSVLAGHSRTVSDSIKKLITSMRDKAPGQLECETAIAALNSCLRDLDQASLAAVSQQLAPR EGISQEALHTQMLTAVQEISHLIEPLASAARAEASQLGHKVSQMAQYFEPLTLAAVGAASKTLSH PQQMALLDQTKTLAESALQLLYTAKEAGGNPKQAAHTQEALEEAVQMMTEAVEDLTTTLNEAASA AGVVGGMVDSITQAINQLDEGPMGDPEGSFVDYQTTMVRTAKAIAVTVQEMVTKSNTSPEELGPL ANQLTSDYGRLASQAKPAAVAAENEEIGAHIKHRVQELGHGCSALVTKAGALQCSPSDVYTKKEL IECARRVSEKVSHVLAALQAGNRGTQACITAASAVSGIIADLDTTIMFATAGTLNREGAETFADH REGILKTAKVLVEDTKVLVQNAAGSQEKLAQAAQSSVATITRLADVVKLGAASLGAEDPETQVVL INAVKDVAKALGDLISATKAAAGKVGDDPAVWQLKNSAKVMVTNVTSLLKTVKAVEDEATKGTRA LEATTEHIRQELAVFCSPEPPAKTSTPEDFIRMTKGITMATAKAVAAGNSCRQEDVIATANLSRR AIADMLRACKEAAFHPEVAPDVRLRALHYGRECANGYLELLDHVLLTLQKPNPDLKQQLTGHSKR VAGSVTELIQAAEAMKGTEWVDPEDPTVIAENELLGAAAAIEAAAKKLEQLKPRAKPKEADESLN FEEQILEAAKSIAAATSALVKAASAAQRELVAQGKVGAIPANALDDGQWSQGLISAARMVAAATN NLCEAANAAVQGHASQEKLISSAKQVAASTAQLLVACKVKADQDSEAMKRLQAAGNAVKRASDNL VKAAQKAAAFEDQENETVVVKEKMVGGIAQIIAAQEEMLRKERELEEARKKLAQIRQQQYKFLPS ELRDEH
Protein Coverage for Talin:
211 / 2541 =8.3% by amino acid count 21608 / 269831 =8.0% by mass
Conclusions
Automated Zoomscan Walking is an easy and time saving way to reduce the tedious work of MS/MS on nanospray samples. It avoids the requirement of an HPLC and can identify proteins at below 50fg/ul while freeing lab personnel for more critical tasks. About 10% of the amino acid sequence of even a very large 270kDa protein is easy and fast as shown in the data above. The method may be implemented in a variety of ways, using the parent-ion mapping feature or the dynamic exclusion feature on the trap. We prefer the parent ion mapping method because ions of low intensity experience less discrimination and are better represented in the final data set. Parent ion mapping methods are slightly quicker to set up and run and can yield more MS/MS spectra in a single data file without complex methods or sequence(lists of files to run) set-ups
Feel free to write with any comments or requests for further information to mattsweeney@earthlink.net