Isobaric Mass Tagging Quanti� cation Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey,1 Xiaoyue Jiang,2 Eugen Damoc,1 RosaViner,2 Yue Xuan,1 Zeller Martin,1 Michaela Scigelova,1 Thomas Moehring,1 Markus Kellmann1

1Thermo Fisher Scienti� c (GmbH) Bremen, Bremen, Germany2Thermo Fisher Scienti� c, 355 River Oaks Parkway, San Jose, CA, USA

Po

ster No

te 64

412

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

2 Isobaric Mass Tagging Quanti� cation Using Q Exactive Instruments – Approach and Expectations

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

PN64412-EN 0615S

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

Isobaric Mass Tagging Quantification Using Q Exactive Instruments – Approach and Expectations Tabiwang N. Arrey1, Xiaoyue Jiang2, Eugen Damoc1, RosaViner2, Yue Xuan1, Zeller Martin1, Michaela Scigelova1, Thomas Moehring1, Markus Kellmann1. 1Thermo Fisher Scientific (GmbH) Bremen, Hanna-Kunath-Str 11, 28199 Bremen 2Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134

Conclusion • Small isolation width can be used to address the issue of co-isolation, however,

the isolation width should be set in relation with the ion accumulation time

• NCE plays an important role in accuracy of quantitation. The best compromised between identification and quantification was achieved at NCE of 32. However, the use of step collision energy, where the lower energy is used for identification and higher energy for quantification is another approach to improve quantitation.

• High resolution > 30000 is a necessity for multiplexed quantitation, especially with TMT-10 plex reagents.

• Upto 86 % of the number of Ecoli proteins identified/quantified in a neat sample was identified/quantified

References 1. Navin Rauniyar and John R. Yates, III, J. Proteome Res., 2014, 13 (12), pp 5293-

5309. 2. Mikhail M. Savitski et al. Science 3 October 2014: Vol. 346 no. 6205. 3. Ting L, Rad R, Gygi SP, Haas W., Nat Methods, 2011 Oct 2;8(11):937-40. 4. Kall, L. Canterbury, J, Weston, J., Noble, W.S., MacCoss, M. Nature Meth. 2007,

4: 923-925.

Overview Purpose: Establish optimal parameters for relative quantification with Thermo ScientificTM Tandem Mass TagTM Reagents (TMT) on the Thermo ScientificTM Q ExactiveTM Mass Spectrometer platforms.

Methods: Data dependent MS/MS bottom-up proteomics.

Results: Improved accuracy and precision of quantitation for TMT labeled proteins.

Introduction Isobaric mass tagging (i.e., TMT) has become a common technique in mass spectrometry for relative quantification of proteins [1]. The tags are designed in a way that the same peptides from samples with different experimental conditions will have an identical precursor ion m/z. Upon fragmenting in mass spectrometer, however, diagnostic fragments - reporter ions - are generated. The intensities of these reporter ions are used to determine the relative amount of a particular peptide in individual samples. Nevertheless, there are a couple of factors that impede reliable quantification in complex proteomic samples. The co-isolation/co-fragmentation of peptides leading to the systematic compression of quantitative ratios presents a major problem [2]. Co-isolation interference has been addressed using various kinds of approaches, such as relying on MS3 experiments [3] or employing a narrow isolation width.

The Q Exactive Plus and Q Exactive HF mass spectrometers are characterized by a fast acquisition speed and an efficient quadrupole isolation. In this study we demonstrate the influence of certain parameter settings, i.e. collision energy (NCE), isolation width, and MS2 resolving power settings, on the protein quantification results from the point of protein quantification accuracy and precision.

Methods Sample Preparation/Liquid Chromatography

Thermo ScientificTM PierceTM HeLa Protein Digest Standard or MassPREP E.coli Digestion Standard (Waters Corporation) were labeled according to manufacturer's instructions with selected channels of the Thermo ScientificTM TMT-10plexTM reagents. Aliquots from 8 or all 10 labeled HeLa digest channels were mixed in equimolar ratios. The labeled E.coli channels (127C, 128N, 128C, 129N, 129C, 130N) were mixed in the ratio 20:10:1:10:20. Furthermore, for selected experiments 3 channels from the HeLa labeled sample (127N, 128N and 128C) mixed in equimolar portions were spiked into the equal amount of E.coli digest mixture (as above). The gradient is summarized in Table 1

*The total run time (including washing and equilibration steps) was 146 min for the 120 min.

Mass Spectrometry

Eluting peptides were analyzed on the Q Exactive Plus and Q Exactive HF mass spectrometers. The instruments were operated in the data-dependent acquisition mode selecting the top most intense 15 or 20 precursors from each scan. A summary of the MS parameters is shown in Table 2.

Results One of the several factors that affect relative quantification using TMT is the applied collision energy. HCD fragmentation of TMT labeled peptides requires higher NCE as shown in Figure 2. With lower collision energies (Fig.2 A) the precursor ion dominates the spectrum. Increasing NCE to 32 (B) significantly improves fragmentation. Such a well-balanced spectrum serves well for both identification and quantification. Further increasing NCE to 40 (C) improves quantification but decreases identification score.

© 2015 Thermo Fisher Scientific Inc. All rights reserved. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Presented at the 63nd ASMS Conference on Mass Spectrometry and Allied Topics St Louis , MO May 31 - June 4, 2015.

FIGURE 1. Consensus workflow and parameter settings for the quantifier node used in the study.

FIGURE 4. Number of peptide and protein groups identified (green) and quantified (blue) using different isolation width settings. The results represent an average of triplicate experiments.

FIGURE 7. Number of identified (green) and quantified (blue) peptide (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da. The results represent an average of duplicate experiments.

With recent advances in multiplexed quantitative approaches based on isobaric tagging, high resolving mass spectrometers are indispensable. For instance, in the case of TMT10plex, the difference between 13C and 15N isotopologues is only 6.32 mDa. Figure 5 shows the reporter ion region of a peptide labeled with 6 channels (127C, 128N, 128C, 129N, 129C, 130N) in different ratios. The MS2 spectra were acquired with resolving power setting 30,000, 35000 and 60,000 at m/z 200. All other parameters were kept constant. With increasing fold-change in concentration of the samples, resolving the reporter ions employing “just” the 30,000 resolving power setting, becomes a challenge; quantitative precision will be affected. Using 35000 or 60,000 resolving power setting, both N and C isotopologues are fully resolved. The applicability of the optimized parameters for TMT quantitation was evaluated using both a neat E.coli sample (6plex mixed in the ratio 20:10:1:1:10:20) and a more complex mixture where the neat E.coli sample was mixed with the same amount of a HeLa digest (channels 127N, 127C and 128N), total peptide load on column was 1 mg. Since the Q Exactive HF, operating with an MS2 resolution setting of 60000 performs the same as the Q Exactive Plus MS, operating at 35000 and to demonstrate the versatility of our method, both samples were analyzed on the Q Exactive Plus MS using 2 isolation widths 0.7 and 1.2 amu. The results for the neat E.coli sample are shown in figure 6. Generally, there are only slight difference in the number of identified/quantified E.coli protein groups for both isolation widths. In both cases more than 90 % of the identified proteins were quantified. The standard deviation is, however, larger for 0.7 amu isolation width compared to 1.2 amu for the neat sample, however, quantification precision of more complex sample (HeLa+E.coli ) as expected is better with 0.7 amu isolation width (Figure 7, C).

The isolation width is another important setting that directly influences the TMT quantification accuracy due to possible precursor co-isolation3. However, the extent of co-isolation depends to a large degree on sample complexity (the extent of pre-separation). Narrowing the isolation width results in a lower co-isolation interference. However, this requires longer ion fill times to accumulate sufficient number of ions. Depending on the sample complexity and other parameters of the acquisition method, there would be an optimum representing an acceptable compromise between the quantitative accuracy and identification success. Figure 4 shows the results from TMT-6plex HeLa digest (in the ratio of 1:1:1:1:1:1) obtained for different isolation width settings. The number of identified and quantified peptide and protein groups increased with increasing isolation width. However, using the isolation width of 2 Da, the number of identified and quantified protein groups starts to drop. It is worth mentioning though, that larger isolation width might improve the identification should appropiate software algorithms be used, conducting, for example, a further round of a search for unassigned fragements. Nevertheless, an adverse effect on quantification would be expected. The best overall outcome was achieved with isolation width of 1.2 Da on Q Exactive HF MS.

FIGURE 2. Spectra of an E.coli peptide acquired at different normalized collision energy settings. (A) NCE 28; (B) NCE 32; (C) NCE 35.

Parameter Q Exactive HF MS Q Exactive Plus MS

Full MS parameters

Resolution settings (FWHM at m/z 200) 120000 70000

Full MS mass range (m/z) 350-1400 350-1400

Target value 3e6 3e6

Max. injection time (ms) 50 50

MS2 parameters

Resolution settings (FWHM at m/z 200) 30000; 60000 35000

Target value 1e5 1e5

Max. injection time (ms) 120; 200; 250 120; 200; 250

Isolation width 0.7; 1.0; 1.2; 2.0 Da 0.7, 1.2 Da

Collision energy (HCD) 30; 32; 35 32

Loop count 15; 20 15

Charge state recognition 2-6 2-6

Peptide match Preferred Preferred

Dynamic exclusion (s) 30 s 30 s

Intensity threshold 2 e4 2 e4

TABLE 1. Liquid Chromatography

Chromatography Settings

LC Thermo ScientificTM UltimateTM 3000RSLCnano equipped with nano pump NCS-3000 and autosampler WPS-3000TPL

Mobile Phases A: 0.1 % FA in water; B: 0.1 % FA in Acetonitrile (Fisher Chemicals)

Gradients* 10–25 % B in 120 min; 5min to 40 % B; 5 min to 90 % B; 8 min at 90 % B back to 5 % B in 2 min.

Flow Rate 250 nL/min

Trapping Column Thermo ScientificTM AcclaimTM PepMapTM100 µCartridge Column C18, 300 μm x 0.5 cm, 5 μm, 100 Å (back flush mode).

Separation Column Acclaim PepMap C18, 75 μm x 50 cm, 2 μm, 100 Å

TABLE 2. Acquisition method parameters (Q Exactive Plus MS and Q Exactive HF MS)

Data Analysis

Thermo Scientific™ Proteome Discoverer™ software version 2.1 was used to search MS/MS spectra against the IPI-human database or Swiss-Prot® E. coli using Sequest HT™ search engine. Static modifications included carbamidomethylation (C), and Thermo Scientific™ TMT-6plex™ reagents (peptide N-terminus; K). Dynamic modifications included methionine oxidation and deamidation (N; Q).

Peptide groups were filtered for maximum 1% FDR using Percolator with Qvality [4]. Protein groups were filtered to 1% FDR based on the number of hits obtained for searches against the forward and decoy database.

The TMT reporter ion quantification method within Proteome Discoverer software was used to calculate the reporter ratios with mass tolerance ±10 ppm. Isotopic correction factors were applied according to the pertaining CoA. Only confidently identified peptides containing all reporter ions were designated as “quantifiable spectra”. Protein ratio was expressed as a median value of the ratios for all quantifiable spectra of the unique peptides pertaining to that protein group. The Consensus workflow and the parameter settings for “Peptide and Protein Quantifier “ node are shown in Figure 1.

A B

C

In the second set of experiments, we further optimized the NCE values. Figure 3 shows the results from TMT6plex HeLa digest (ratio of 1:1:1:1:1:1) for NCE 30, 32, and 35. Slightly more unique peptide and protein groups were identified and quantified using NCE of 32% compared to the other two settings.

FIGURE 2. Effects of different NCE on the number of unique peptide and protein groups identified (green) and quantified (blue) from TMT6plex HeLa digest. The results represent an average of triplicate experiments.

30 32 35Identified proteins 2829 3115 2740Quantified proteins 2795 3085 2707

0

500

1000

1500

2000

2500

3000

3500

no o

f pr

otei

n gr

oups

Protein groups

99 %

99 %

99 %

30 32 35Unique peptides 18874 21730 17829Peptides used for quan 17580 20414 16597

0

5000

10000

15000

20000

25000

no o

f uni

que

pept

ides

Unique peptides

93 %

94 %

93 %

A B

0.7 1 1.2 2Unique peptides 12097 15644 21730 19432peptides used for quan 11054 14742 20414 18031

0

5000

10000

15000

20000

25000

No

of u

niqu

e pe

ptid

es

Unique peptides

91 %

94 %

94 %93 %

0.7 1 1.2 2Identified proteins 2131 2511 3115 2923Quantified proteins 2087 2480 3085 2892

0

1000

2000

3000

No

prot

ein

grou

ps

Protein groups

98 %

99 %

99 %99 %

A B

FIGURE 5. Detail of the reporter ion region for an E.coli peptide acquired with resolving power setting 30,000 (30k), 35000 (35k) and 60,000 (60k). The isobaric channels are only partially resolved when 30,000 resolving power setting was used.

1.2 0.7Peptides identified 8574 8387Peptides used for quan 7096 6393

0

2000

4000

6000

8000

no o

f pep

tides

iden

tifie

d

Unique peptides

1.2 0.7Proteins identified 1330 1331Proteins quantified 1218 1177

0

500

1000

1500

No

of p

eptid

es id

entif

ied

Protein groupsA B

FIGURE 6. Number of identified (blue) and quantified (green) peptides (A) and protein (B) groups obtained from the analysis of E.coli digest employing isolation with settings 0.7 and 1.2 Da.

Results from the E.coli and HeLa mix is summarized in Figure 7. The goal of the this experiment was to determine approximately the number of E.coli proteins that can be quantified in the complex background and the effect HeLa peptides may have on the precision and accuracy of quantification. In total, approximately 4000 protein groups (HeLa + E.coli) were identified, about a third of these were E.coli proteins. This represents about 86 % of the number of proteins that were quantified in the neat E.coli sample (Figure 6). However, both precision and accuracy were significantly affected by presence of the HeLa peptides. Both accuracy and precision were improved by using narrow isolation width without penalty for identified or quantified proteins and peptides (Figure 7)

1.2 0.7Ecoli peptides identified 5822 5671Ecoli peptides quantified 4070 4138

0

1000

2000

3000

4000

5000

6000

No

of u

niqu

e pe

ptid

es

Unique peptides

1.2 0.7Proteins identified 1019 1011Proteins quantified 918 852

0

200

400

600

800

1000

1200

No

of p

rote

in g

roup

s

Protein groups

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

127C 128N 128C 129N 129C 130N

Accuracy and precision Isolation width: 1.2

Isolation width: 0.7

A

C

B

30k60k

35k

Africa +43 1 333 50 34 0Australia +61 3 9757 4300Austria +43 810 282 206Belgium +32 53 73 42 41Canada +1 800 530 8447China 800 810 5118 (free call domestic)

400 650 5118

Denmark +45 70 23 62 60Europe-Other +43 1 333 50 34 0Finland +358 10 3292 200France +33 1 60 92 48 00Germany +49 6103 408 1014India +91 22 6742 9494Italy +39 02 950 591

Japan +81 45 453 9100Korea +82 2 3420 8600Latin America +1 561 688 8700Middle East +43 1 333 50 34 0Netherlands +31 76 579 55 55New Zealand +64 9 980 6700Norway +46 8 556 468 00

Russia/CIS +43 1 333 50 34 0Singapore +65 6289 1190Spain +34 914 845 965Sweden +46 8 556 468 00Switzerland +41 61 716 77 00UK +44 1442 233555USA +1 800 532 4752

www.thermoscientific.com©2015 Thermo Fisher Scienti� c Inc. All rights reserved. ISO is a trademark of the International Standards Organization. MassPREP E.coli Digestion Standard is a registered product of waters Corporation. All other trademarks are the property of Thermo Fisher Scienti� c and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scienti� c products. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Speci� cations, terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representative for details.

Thermo Fisher Scienti� c, San Jose, CA USA is ISO 13485 Certi� ed.

ISO 13485