1. Yuwei Qian, Werner Ens, Ken G Standing Physics and Astronomy, University of Manitoba, Winnipeg, MB, Canada Marta Izydorczyk, Sharon Bazin, Ken Preston Grain Research Laboratory, Winnipeg, MB, Canada

2. Overview MALDI mass spectrometry provides a fast method of high resolution for identifying and characterizing proteins from barley grain and wort during malting and brewing. Proteins and polypeptides were fractionated by sequential ammonium sulphate (AS) precipitation, HPLC and SELDI techniques. Peptide mapping and amino acid sequencing were conducted by MALDI MS and by MS/MS, respectively. The usefulness of the methods developed in this study was verified by the identification of a wort protein that partially matches alpha-amylase inhibitor of barley and by determination of a sequence of an unknown 19-aa peptide that may play an active role in foam formation and/or stabilization.

3. Introduction Barley polypeptides are known to be actively involved in formation and stabilization of beer foam. A better understanding of the protein-related phenomena associated with beer quality will assist in developing more accurate tests to evaluate barley grain and to predict its end-use quality as well as in developing improved malting barley varieties. We are investigating methods based on MALDI TOF mass spectrometry for identification and characterization of the active barley proteins at various stages of malting and brewing process. Various SELDI chips and RP HPLC were used to select and separate polypeptides.

4. Methods 1.Polypeptides from wort of Copeland and Harrington were sequentially precipitated with 20, 30, and 40% AS. 20%AS fraction was loaded on a CM52 Cation Exchanger column and polypeptides were eluted by NaCl. The eluted polypeptides were further fractionated by RP-HPLC. Peaks were collected for analysis of intact polypeptide masses and for tryptic peptide mapping and MS/MS. 2.Ciphergen WCX chip : Four binding/washing buffers were used 50mM NaAc, pH3; 25mM, 50mM and 100mM NaAc, pH4. 20%AS fraction was mixed with binding buffer and loaded on the WCX spot and incubated for 30 minutes. The spot was then washed with washing buffer, followed by water wash. The spot was used either for intact mass measurement by adding saturated sinapinic acid (SPA) or for on-chip trypsin digestion.

5. Methods - continue 3.Ciphergen H50 chip: 20%AS fraction was loaded on H50 spot, incubated for 20 minutes, and washed with 20 and 40% ACN. The spot was analyzed either for intact mass by adding SPA or on-chip trypsin digestion. 4.Ciphergen NP20 chip: 20%AS fraction was loaded on NP20 spot, and washed with HPLC grade water. The spot was used either for intact mass by adding SPA or on-chip trypsin digestion. 5.Intact masses were obtained from linear MALDI TOF MS (with a Ciphergen PBS-II instument). 6.On-chip trypsin digestion: Trypsin was prepared in ammonium bicarbonate and added to the SELDI spot. The incubation was done at 37°C for 7 hours.

6. Copeland Harrington 10kD 12.5kD 15kD 17.5kD Copeland Harrington 2kD 4kD 6kD 8kD Copeland Harrington 20kD 25kD 30kD 35kD Copeland Harrington 40kD 45kD 50kD 55kD 60kD This shows the pooled polypeptide profiles of different AS fractions Polypeptide Profiles from Wort of Copeland and Harrington Barley by MALDI

7. SELDI Ciphergen Chips for Wort Protein Profiles in 20% AS Copeland WCX pH3 50 mM NaAc WCX pH4 50 mM NaAc WCX pH4 100mM NaAc WCX pH4 25mM NaAc On-chip trypsin digestion MALDI NP20 H50 20%ACN H50 40%ACN

8. MS-WCX MS-H50 Tryptic Peptide Mapping on WCX (pH4 100mM NaAc) and H50 (40%ACN) Chips were useful for isolation of specific fraction/protein for further study. Database search with MS and MS/MS mapping did not match any known protein probably due to lack of DNA database for barley.

9. Amino Acid Sequence determined by de novo Sequencing for the peptide 2185 Dal from WCX MS/MS y19 y18 y16 y15 y14 y13 y12 y11 y10 y9 y8 y7 y5 y4 y3 y6 -17 y14 - 17 Database search for similarity using BLASTp and tBLASTn (NCBI) with this sequence did not produce meaningful matches. The determined sequence is probably part of an unknown barley protein present in wort. QDGNEHQENGEMHEDSSDK

10. 20%AS Copeland Wort->CM column->Elute from CM column->HPLC F6 HPLC of 20AS 0.1M NaCl from CM MALDI MS F2 Trypsin digest Tryptic MS for F6 MALDI MS F4 MALDI MS F6 MALDI MS F8 F1F2 F5F7F8 F3 F4 F6 0.5M NaCl

11. Start – End Mr(expt) Mr(calc) Miss Sequenceconfirmed by our experiment 46 - 54 1111.55 1110.54 0 DYVEQQACR  55 - 65 1185.67 1184.64 0 IETPGPPYLAK  86 - 90 657.32 656.31 0 FFMGR  86 - 91 785.42 784.41 1 FFMGRK ? 91 - 107 1870.94 1869.92 1 KSRPDQSGLMELPGCPR  92 - 107 1742.85 1741.82 0 SRPDQSGLMELPGCPR  108 - 115 1023.51 1022.49 0 EVQMDFVR  Database Search for 20%AS F6 with MS and MS/MS Database search using Mascot for the tryptic peptide MS map revealed a partial match for alpha-amylase inhibitor with 33% sequence coverage. MS/MS peptide mapping further confirmed the results from MS map. Search results from MS: Match to: S78524 Score: 81 Expect: 0.0011 Mass: 16515 dalton alpha-amylase inhibitor, tetrameric, chain CMb precursor - barley Number of mass values matched: 7 Sequence Coverage: 33% Matched peptides shown in Bold Red 1 MASKSSCDLL LAAVLVSIFA AVAAVGSEDC TPWTATPITP LPSCRDYVEQ 51 QACRIETPGP PYLAKQQCCG ELANIPQQCR CQALRFFMGR KSRPDQSGLM 101 ELPGCPREVQ MDFVRILVTP GFCNLTTVHN TPYCLAMDEW QWNRQFCSS

12. Conclusions We developed methods for identifying proteins from barley wort using MALDI TOF. The methods may be used to find proteins that are important for formation and stabilization of beer foam. Wort proteins were first fractionated by sequential AS precipitation. The whole protein profile may be seen by pooling MS patterns from different fractions. The proteins in each fraction were then further separated by HPLC or selected by SELDI in order to obtain individual protein for their identification by MS and MS/MS. The feasibility of the methods was verified by the identification of peptides that partially matched alpha- amylase inhibitor of barley from databases. The feasibility was also confirmed by determination of a peptide sequence by de novo sequencing with MS/MS. ACKNOWLEDGMENT: Financial support was provided partly by BMBRI.