Comparison of CID, ETD, and HCD for Top-down Characterization of Histones

Zhixin(Michael) Tian Nikola Tolić, Rui Zhao, Shawna, Hengel, Si Wu, Ronald J. Moore, Errol W. Robinson, Richard D. Smith, Ljiljana Paša-Tolić Environmental Molecular Sciences Laboratory at Pacific Northwest National Laboratory

Core histones: H4, H2B, H2A, H3

DNA Repair1

DNA Replication2

DNA Transcription3

1. Nature Reviews Mol. Cell Bio. 6, 757 (2005); 2. Molecules and Cells 28, 149 (2009); 3. Science 272, 371 (1996)

Histones and post-translational modifications (PTMs)

Histones are heavily modified and very complex

400 600 800 1000 1200 1400 1600 1800 2000m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rela

tive

Abun

danc

e

1022.92z=15

959.05z=?

1096.91z=14

903.46z=?

1181.29z=13852.65

z=?

808.57z=?

1704.20z=9766.74

z=?1278.31

z=12

697.86z=?

1535.37z=10

621.35z=? 1393.25

z=?

1920.86z=?

1676.08z=?

539.32z=?

1786.29z=?

RPLC-MS/MS of HeLa core histones

0 20 40 60 80 100 120Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Relat

ive A

bund

ance

33.87

36.93

48.91

49.5251.39 63.69

103.3731.60 83.5386.0967.52 107.21

87.4068.84 111.176.95 114.5321.88

1000 1005 1010 1015 1020 1025 1030 1035 1040m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Relat

ive Ab

unda

nce

1022.92z=15

1023.92z=15

1024.78z=15

1022.12z=15

1025.65z=15

1020.92z=15

1026.65z=151019.05

z=?1009.85z=?

1028.51z=?1010.51

z=?1017.91

z=?1005.71z=?

1011.51z=? 1031.84

z=? 1037.98z=?

MARTKQTARK STGGKAPRKQ LATKAARKSA PATGGVKKPH RYRPGTVALR EIRRYQKSTE LLIRKLPFQR LVREIAQDFK TDLRFQSSAV MALQEACEAY LVGLFEDTNL CAIHAKRVTI MPKDIQLARR IRGERA

M

A

M2

P

P P

M

P

M2

MM

M

A

M3

P

P

A

M2

M

A

M

A

M2

M

A

M3

P

M2

M

A

M3

M

P P

MM2

M

M3

P

P

P

Human H3.1 (UniProt)

A=Acetylation, M=Methylation, M2=asymmetrical dimethylationMM=symmetrical dimethylation, M3=trimethylation, P=Phosphorylation

Online RPLC/WCX-MS/MS HeLa core histones

Methods IDs RPLC-MS/MS 127

WCX-MS/MS 135

RPLC-WCX-MS/MS 708

0 40 80 120 160 2000.0

0.2

0.4

0.6

0.8

1.0

H3_

2

H3_

1

H2A

H2B

Inte

nsi

ty (a.

u.)

Elution time (min)

H4

60 80 100 120 140 160 180 2000

1x106

2x106

3x106

0ac5ac 4ac

K5acK16acR3me2K12acK16acK20me3S1acK16acK20me2K12acK16acK20me2K5acK16acK20me2S1acK16acK20meS1acK12acK20me2S1acR3meK16acS1acK12acK5acK16acK20meS1acK5acK20meS1acK8acK20me3K5acK8acK12acK16acR3me2K5acK8acK20meS1acK8acK5acK8acK20meS1acK12acS1acK12acK20me3S1acK12acK20meK5acK12acK20me2

3ac

1ac

Inte

nsity

(a.u

.)

Elution time (min)

2ac

S1acK5acK8acK12acK16acS1acK5acK8acK12acK16acK20me2S1acK5acK8acK12acK16acK20meS1acK5acK8acK12acK16acK20me3S1acR3me2K5acK8acK12acK16ac

S1acK5acK12acK16acS1acK8acK12acK16acK20meS1acK8acK12acK16acK20me2R3meK5acK8acK12acK16acS1acK5acK12acK16acK20meS1acK5acK12acK16acK20me2S1acK5acK12acK16ac

K5acK5acK20me3S1acR3meK20meS1acK20me2K5acK20me2S1acR3meS1acR3me2K16acK20meK16acK20me2K16acK20me3K5acK20me3K8acK20meS1acS1acK20me2R3me2K16acK20me3S1acR3meK20me2S1acK20me3K12acS1acK20meS1pK5acK20me2

K20meR3me2R3meK20meK20me2R3me2K20me2R3meK20meR3meK20meR3meK20me3K20me2K20me3R3meK20me2S1pR3me2S1pK20me2S1pR3me2K20me3S1pR3me2K20me2

S1acK5acK12acR3me2K8acK12acK16acS1acK5acK8acS1acK12acK16acS1acK12acK16acK20meS1acK12acK16acK20me2S1acK12acK16acK20me3S1acR3meK12acK16acK20me2S1acK12acK16acK20me2S1acK5acK16acK20meS1acK8acK16acK20me2S1acK8acK16acK20me3K8acK12acK16acK20me3S1acK5acK16acS1acK5acK16acK20me2S1acK8acK16acS1acK8acK12acS1acK8acK12acK20me2S1acK5acK8acK20me3R3me2K8acK12acK16ac

RPLC separation of histones

WCX-MS/MS of H4

0 2500 5000 7500 10000 125000

20000

110000112000114000116000118000120000

Inten

sity (

a.u.)

Monoisotopic Mass (Da)

Ambiguous identifications: H4_S1acK{5, 8, 12}acK16acK20me2

Unambiguous localization of K12ac in histone H4

Unique identification: H4_S1acK12acK16acK20me2

0 2000 4000 6000 8000 10000 120000.0

5.0x103

1.0x104

1.5x104

2.0x104

Inten

sity (

au)

Monoisotopic Mass (Da)

CID

ETD

Optimizing fragmentation for histone characterization

HeLa core histones (3.75 µg)

.RAW

IDs

CID ETD HCD

NCE (%) AT (ms) HCD_45_30 45 30 HCD_45_60 45 60 HCD_60_60 60 60 CID_35_30 35 30 CID_45_60 45 60 CID_60_60 60 60 ETD_10 / 10 ETD_15 / 15 ETD_20 / 20

ProSightPC 2.0

LTQ Orbitrap Velos

RPLC Column (200 µid*70cm, C5(J), 5µ, 300Å) A: 20%ACN, 5%IPA, 0.6%FA B: 45%ACN, 45%IPA, 0.6%FA

NCE (%) RT (ms) IDs Avg [–log(P_Score)]

HCD_45_30 45 30 123 16.9

HCD_45_60 45 60 136 15.3

HCD_60_60 60 60 144 20.4

CID_35_30 35 30 62 10.3

CID_45_60 45 60 56 16.5

CID_60_60 60 60 65 21.2

ETD_10 / 10 83 23.3

ETD_15 / 15 160 34.5

ETD_20 / 20 138 28.9

Different fragmentation conditions

±8 ±0.3

HCD conditions identification of H4_S1acK16ack20me2

HCD_60_60: Matching fragments=42, P_Score=1.3E-50

HCD_45_30: Matching fragments=27, P_Score=6.5E-31

HCD_45_60: Matching fragments=38, P_Score=5.9E-39

Optimal CID, ETD, and HCD

CID_60_60 (65)

30 22

5 98

35

79

8

EDT_15 (160)

HCD_60_60 (144)

NCE (%) RT (ms) IDs Avg [–log(P_Score)] HCD_60_60 60 60 144 20.4

CID_60_60 60 60 65 21.2

ETD_15 / 15 160 34.5

ETD, HCD, and CID: H4_S1acK16ack20me2

CID: Matching fragments=31, P_Score=1.1E-40

ETD: Matching fragments=56, P_Score=1.1E-76

HCD: Matching fragments=46, P_Score=3.3E-56

H2AX_S1acS139p uniquely identified only by ETD

793 794 795 796 797 798 799m/z

z15(2+)

934 935 936 937 938 939 940m/z

z18 (2+)

Top-down only! Not with bottom-up or middle-down!

Advantage of top-down fragmentation

HCD_60_60: H31_K9me2K27me2K36meK79me

CID_60_60: H2B1O_K46meK57me2

Fragmentation efficiency of CID, ETD, HCD

CID ETD HCD0

20

40

60

80

100

% o

f uni

que

IDs

_% _ _ _ 100_ _

unique IDsof unique IDsunique IDs ambiguous IDs

= ×+

Conclusions

• An online 2D LC-FTMS platform dramatically improved throughput and sensitivity compared to more traditional platforms, and resulted in identification of hundreds of histone isoforms from microgram levels of protein, including phosphorylation.

• For a single fragmentation method, ETD provides the most confident identiifcations on average.

• Total number of identifications more than doubled when CID was switched to ETD or HCD.

• Combinatorial modifications across the entire protein sequence could only be identified using top-down proteomics.

Acknowledgements

Richard D. Smith and other group members

Bioinformatics Gordon A. Anderson Matthew E. Monroe

Samuel O. Purvine Nikola Tolić

Instrumentation

Heather M. Mottaz-Brewer

Therese R. W. Clauss

Boyd Champion

Thomas Fillmore

Robbie Heegel

Daniel J. Orton Carrie D. Nicora

Ronald J. Moore

Karl Weitz

Rui Zhao

Anil K. Shukla

Ryan T. Kelly

Randy Norheim Michael Russcher

Ljiljana Paša-Tolić

Shawna Hengel Robby Robinson

Si Wu

Joshua Aldrich

Support and Funding: EMSL at PNNL, DOE, NIH NCRR (RR018522)